Manual on

MUTATION BREEDING
THIRD EDITION

Print j-5109 E5 Cover Page 1.indd 1 22.05.18 10:13

 Manual on Mutation Breeding

Third Edition

Edited by

Spencer-Lopes, M.M.
Forster, B.P.
and
Jankuloski, L.

Plant Breeding and Genetics Subprogramme

Joint FAO/IAEA Division of Nuclear Techniques in Food and
Agriculture
Vienna, Austria

Food and Agriculture Organization of the United Nations

International Atomic Energy Agency
Vienna, 2018
 DISCLAIMER

Recommended citation

FAO/IAEA. 2018. Manual on Mutation Breeding - Third edition. Spencer-Lopes,

M.M., Forster, B.P. and Jankuloski, L. (eds.), Food and Agriculture Organization of
the United Nations. Rome, Italy. 301 pp.

The designations employed and the presentation of material in this information

product do not imply the expression of any opinion whatsoever on the part of the Food
and Agriculture Organization of the United Nations (FAO) concerning the legal or
development status of any country, territory, city or area or of its authorities, or
concerning the delimitation of its frontiers or boundaries. The mention of specific
companies or products of manufacturers, whether or not these have been patented,
does not imply that these have been endorsed or recommended by FAO in preference
to others of a similar nature that are not mentioned.

The views expressed in this information product are those of the author(s) and do not
necessarily reflect the views or policies of FAO.

ISBN 978-92-5-130526-3

© FAO, 2018

FAO encourages the use, reproduction and dissemination of material in this

information product. Except where otherwise indicated, material may be copied,
downloaded and printed for private study, research and teaching purposes, or for use
in non-commercial products or services, provided that appropriate acknowledgement
of FAO as the source and copyright holder is given and that FAO’s endorsement of
users’ views, products or services is not implied in any way.

All requests for translation and adaptation rights, and for resale and other commercial
use rights should be made via www.fao.org/contact-us/licence-request or addressed
to copyright@fao.org.

FAO information products are available on the FAO website

(www.fao.org/publications) and can be purchased through publications-
sales@fao.org.

Cover photos credits:

Shutterstock.com/Tefi, Khakimullin Aleksandr, Zketch, Vitalii Nesterchuk
Joint FAO/IAEA Division
 CONTACT INFORMATION OF THE AUTHORS

A. BADIGANNAVAR J.B. JANKOWICZ-CIESLAK

Nuclear Agriculture and Biotechnology Plant Breeding and Genetics Laboratory
Division FAO/IAEA Agriculture and Biotechnologies
Bhabha Atomic Research Center Laboratories
Trombay Friedenstrasse 1
400 085, Mumbai, India A-2444 Seibersdorf, Austria
Email: Ashokmb1@gmail.com Email: J.Jankowicz@iaea.org

B.P. FORSTER L. JANKULOSKI

BioHybrids International Limited Plant Breeding and Genetics Section
P.O. Box 2411, Early Joint FAO/IAEA Division
Reading RG6 5FY, United Kingdom P.O. Box 100, Wagramer Strasse 5
Email: Brianforster@biohybrids.co.uk; A-1400 Vienna, Austria
Brianpforster@gmail.com Email: L.Jankuloski@iaea.org

J.S. (Pat) HESLOP-HARRISON P. JANOS

University of Leicester Deputy Manager
Department of Biology Cereal Research Nonprofit Ltd.
Adrian building Alsokikotosor 9.
University Road Szeged 6726, Hungary
Lei 7RH, Leicester, United Kingdom Email: janos.pauk@gabonakutato.hu
Email: Phh4@le.ac.uk
A. KODYM
R. IBRAHIM University of Vienna
Elite Scientific Instruments Sdn. Bhd Core Facility Botanical Garden &
A-LG-03, Block A, Serdang Perdana Selatan Department of Pharmacognosy
Section 1, 43300 Seri Kembangan Vienna, Austria
Selangor, Malaysia Email: Andrea.Kodym@icloud.com
Email: Rusli.anm@gmail.com
B. KUNTER
I. INGELBRECHT Turkish Atomic Energy Authority (TAEK)
Plant Breeding and Genetics Laboratory Sarayköy Nuclear Research and Training
FAO/IAEA Agriculture and Biotechnologies Centre
Laboratories Division of Agriculture, Ankara Turkey
Friedenstrasse 1 Email: Burak.Kunter@taek.gov.tr
A-2444 Seibersdorf, Austria
Email: I.Ingelbrecht@iaea.org

iii
M. MATIJEVIC R. SANGWAN
Plant Breeding and Genetics Laboratory Université de Picardie Jules Verne UFR
FAO/IAEA Agriculture and Biotechnologies Sciences
Laboratories Laboratoire EDYSAN-AEB, UMR 7058
Friedenstrasse 1 CNRS
A-2444 Seibersdorf, Austria 33, rue Saint Leu, 80000 Amiens France
Email: M.Matijevic@iaea.org Email: Rajbir.Sangwan@u-picardie.fr

C. MBA F. SARSU
Food and Agriculture Organization of the Plant Breeding and Genetics Section
United Nations (FAO) Joint FAO/IAEA Division
Viale delle Terme di Caracalla P.O. Box 100, Wagramer Strasse 5
I-00153 Rome, Italy A-1400 Vienna, Austria
Email: Chikelu.Mba@fao.org Email: F.Sarsu@iaea.org

A. MUKHTAR ALI GHANIM M. SPENCER-LOPES

Plant Breeding and Genetics Laboratory University CAD
FAO/IAEA Agriculture and Biotechnologies Faculté des sciences et technologies
Laboratories Dept. de biologie végétale
Friedenstrasse 1 Dakar, Sénégal
A-2444 Seibersdorf, Austria Email: Manmymadou2@gmail.com
Email: A.Mukhtar-Ali-Ghanim@iaea.org
I. SZAREJKO
H. NAKAGAWA Department of Genetics
Hamamatsu Photonics K. K. University of Silesia
Central Research Laboratory Jagiellonska 28
5000, Hirakuchi, Hamakita-ku 40-032 Katowice, Poland
Hamamatsu, Shizuoka, 434-8601 Japan Email: Iwona.Szarejko@us.edu.pl
Email: Hitoshi.Nakagawa@crl.hpk.co.jp
M. SZURMAN
S. NIELEN Faculty of Biology and Environmental
Plant Breeding and Genetics Section Protection
Joint FAO/IAEA Division University of Silesia
P.O. Box 100, Wagramer Strasse 5 Jagiellonska 28
A-1400 Vienna, Austria 40-032 Katowice, Poland
Email:S.Nielen@iaea.org Email: Miriam.Szurman@us.edu.pl

S. PENNA W. THOMAS
Nuclear Agriculture and Biotechnology The James Hutton Institute
Division Invergowrie
Bhabha Atomic Research Institute Dundee DD2 5DA
Trombay, Mumbai 400 085, India Scotland, United Kingdom
Email: Penna888@yahoo.com Email: Bill.Thomas@hutton.ac.uk

iv
B.J. TILL A. YASSIER
University of Vienna Plant Genomics
Department of Chromosome Biology Verdant Bioscience
Vienna, Austria Galang
Email: Till.Brad@gmail.com North Sumatra, 20585, Indonesia
Email: Yassier.Anwar@verdantbioscience.com

v
vi
 CONTENTS

CHAPTER 1. PHYSICAL MUTAGENESIS 5

Spencer-Lopes, M.M., Jankuloski, L., Mukhtar Ali Ghanim, A., Matijevic,
M., and Kodym, A.

RADIATION TYPES AND SERVICES 5

X-rays 5
Gamma-rays 5
Ultraviolet light 11
Alpha particles 11
Beta particles 12
Particles from accelerators 12
Neutrons 12
Ion beam irradiation and ion beam implantation 13
Cosmic irradiation 14
Laser beam irradiation 14

RADIO-BIOLOGY 14
Absorption of ionizing radiation 14
Chemical effects of ionizing radiation: DNA damage and DNA repair 15
Lethal effects of ionizing radiation: DNA damage and repair 16

DOSIMETRY 16
Exposure and dose determination 17
Absorbed dose in irradiated targets 18
Dosimeters 19

OBJECTIVES AND TREATMENTS 23

Target plant materials 23
Irradiation treatments and conditions 25

RADIATION SENSITIVITY AND MODIFYING FACTORS 28

Environmental factors 28
Biological factors 30

PRE- AND POST-IRRADIATION TREATMENTS 31

Pre-treatment 31

vii
Adjustment of seed water content 31
Post-irradiation storage 31

PROTOCOL FOR RADIO-SENSITIVITY TESTING 32

List of equipment, supplies and facilities needed 32
Radio-sensitivity testing procedure 34

STANDARD PROTOCOLS FOR SEED MUTAGENESIS USING X- 44

RAY IRRADIATORS
Pre-treatment preparations of seed samples 44
Radiosensitivity tests 44
Irradiation treatment of the seeds 45
Post irradiation treatment and handling 47

PRACTICAL EXAMPLE OF APPLICATIONS USING PHYSICAL 48

MUTAGENS

SEED IRRADIATION SERVICES AT THE JOINT FAO/IAEA PLANT 49

BREEDING AND GENETICS LABORATORY

OTHER MUTAGENS 49

CHAPTER 2. CHEMICAL MUTAGENESIS 51

Ingelbrecht, I., Jankowicz-Cieslak, J., Szurman, M., Till, B.J., and
Szarejko, I.

MAIN CHEMICAL MUTAGENS 51

A1kylating agents 52
Sodium azide 53
Other chemical mutagens 54

MODE OF ACTION AND MUTATION SPECTRA 56

Alkylating agents 56
Sodium azide 57

GUIDELINES FOR CHEMICAL MUTAGENESIS 59

Target plant materials 59
Dose, dose determination and mutational load 61
Status of plant materials 62
Physico-chemical properties of chemical mutagens and mutagen solutions 63
Pre- and post-treatment procedures 64
Advantages and limitations of chemical mutagenesis 65

viii
STORAGE, HANDLING AND DECONTAMINATION PROCEDURES 66
OF CHEMICAL MUTAGENS
Alkyl alkanesulphonates and alkyl sulphates 67
Nitroso compounds 68
Azides 69

EXAMPLES OF TREATMENT PROCEDURES 69

Ethyl methanesulphonate (EMS) mutagenesis of in vitro banana meristem 70
explants
Ethyl methanesulphonate (EMS) mutagenesis treatment of barley seeds 74
Combined treatment of barley seeds with sodium azide and N-nitroso N- 77
methylurea
Conclusion 81

CHAPTER 3. TYPES OF MUTATIONS 83

Nielen, S., Forster, B.P., and Badigannavar, A.

PHENOTYPIC MUTANTS 83

GENOTYPIC MUTATIONS 84
Genome mutations 85
Gene mutations 94
Genes mutations expression at the trait level 95

PRACTICAL EXAMPLES 99
Example 1 - Selection of a single gene mutation: Ug99 rust resistance in 99
wheat
Example 2 - Induced mutations for genetic improvement of quantitative 101
traits in sorghum

CHAPTER 4. MUTAGEN EFFECTS IN THE FIRST 105

GENERATION AFTER SEED TREATMENT: BIOLOGICAL
EFFECTS OF MUTATION TREATMENTS
Nielen, S., Forster, B.P., and Heslop-Harrison, J.S. (Pat)

PLANT INJURY AND LETHALITY 105

CYTOLOGICAL EFFECTS 107

Chromosome observations 109
The comet assay 109
Low dose stimulation 110
Effects on meiosis 111

ix
STERILITY 111

CHIMERAS 112

SECONDARY EFFECTS: TRANSPOSON ACTIVATION 114

Examples of TE-induced mutations in plant breeding 115

CHAPTER 5. MUTATION BREEDING IN SEED PROPAGATED 119

CROPS: PARENTAL SELECTION, MUTANT GENERATION
DEVELOPMENT, MUTATION DETECTION, MUTANT
EVALUATION AND FACTORS INFLUENCING SUCCESS
Mukhtar Ali Ghanim, A., Spencer-Lopes, M.M., and
Thomas, W.

SELECTING PARENTS AND HANDLING M1 TO M3 GENERATIONS 119

FOR MUTANT SELECTION
Selection criteria for parent genotype 119
Planning for the M1 generation 120
Sowing of M1 seeds 122
Isolation of M1 material 124
Care during cultivation and data recording 125
Harvest of M1 128
Management of M2 population 129
Management of the M3 generation 138
Contamination in self-pollinated mutant crops 139

THE DETECTION OF INDUCED MUTATIONS 143

Inter- and intra-somatic selection 144
Genetic structure 146
Function of locus 147
Mutability 147

IDENTIFICATION, EVALUATION AND DOCUMENTATION OF 147

MUTANTS
Identification of mutants 147
Propagation and evaluation of useful mutants 148
Recording of experiments and documentation of mutants 149

FACTORS INFLUENCING THE SUCCESS OF MUTATION 150

BREEDING
Differences due to the genotype 151
Type of mutagen and dose 152
Pleiotropy and linkage 153

x
CHAPTER 6. MUTATION BREEDING FOR VEGETATIVELY 157
PROPAGATED CROPS
Sarsu, F., Penna, S., Kunter, B., and Ibrahim, R.

APPLICATION OF MUTATION TECHNIQUES 157

SELECTION OF PARENTAL LINES AND MUTAGENIC 158

TREATMENTS
Selection of parental lines 158
Population size 159
Mutagenic treatments 161
Selection of optimal mutagenic dose 161

CHIMERAS 164

HANDLING MUTANT POPULATIONS AND RELEASE OF NEW 165

VARIETIES

SCREENING TECHNIQUES FOR ABIOTIC STRESS 167

SCREENING TECHNIQUES FOR BIOTIC STRESS 171

PRACTICAL EXAMPLES OF THE APPLICATIONS OF MUTATION 172

BREEDING IN VPCS
Practical example of mutation breeding in Cherry 172
Practical example of mutation breeding in Potato 174

CHAPTER 7. MAJOR TRAITS TO BE IMPROVED BY 177

MUTATION BREEDING
Jankuloski, L., Forster, B.P., and Nakagawa, H.

YIELD IMPROVEMENT 177

Yield and yield components 177
Examples of mutants with improved yield 180

TOLERANCE TO ABIOTIC STRESS 180

Drought 181
Salinity 181
Temperature 181

xi
TOLERANCE/RESISTANCE TO BIOTIC STRESS 183
Resistance to disease 183
Resistance to pests 185
Examples of mutants with improved biotic tolerance/resistance 187

QUALITY 187
Quality, nutrition and functionality 187
Starch 189
Protein 190
Fats, oils and fatty acids 191
Toxins and anti-nutritional factors 191
Examples of mutants with improved quality 193

AGRONOMIC TRAITS 194

Flowering and ripening time 194
Adaptability 195
Plant type, growth habit and architecture 195
Resistance to lodging and stem breakage 196
Shattering and shedding resistance 198
Other agronomic traits 200
Examples of mutants with improved agronomic traits 202

MUTANTS TO FACILITATE PLANT BREEDING 203

CHAPTER 8. SPECIFIC TECHNIQUES FOR INCREASING 205

EFFICIENCY OF MUTATION BREEDING
A) IN VITRO METHODS IN PLANT MUTATION BREEDING 205
Sarsu, F., Spencer-Lopes, M.M., Sangwan, R., and Penna, S.

BRIEF REVIEW OF PLANT TISSUE CULTURE 205

PLANT REGENERATION SYSTEMS 206

Micro-propagation 206
Meristem culture 207
In vitro morphogenesis 207
Haploid and doubled haploid production 208

PLANT TISSUE CULTURE FOR MUTATION BREEDING 209

Micro-propagation of mutated plants 209
In vitro mutation induction 209
Types of explants for in vitro mutagenesis 211
Mutagens used in vitro 211

xii
In vitro radio-sensitivity testing 211
Chimeras 214

HANDLING MUTATED PLANT POPULATIONS IN VITRO 214

IN VITRO MUTATION SCREENING METHODS 219

In vitro screening techniques for abiotic stress 219
In vitro screening for biotic stress 219

SOMACLONAL VARIATION 221

PROCEDURE FOR MUTATION INDUCTION IN MUSA SPP. 222

PRACTICAL EXAMPLES OF IN VITRO MUTAGENESIS 226

B) HAPLOIDS AND DOUBLED HAPLOIDS IN MUTATION 230

BREEDING
Jankuloski, L., Szarejko, I., and Pauk, J.

INTRODUCTION 230

METHODS FOR HAPLOID/DOUBLED HAPLOID PRODUCTION 231

Haploids via androgenesis 231
Haploids via gynogenesis 236
Haploids via wide crosses followed by chromosome elimination 236
Spontaneous haploids 236
Haploids via aberrant pollination 237
Haploid inducer genes 238

MAIN APPROACHES IN HAPLOID AND DOUBLED HAPLOID 238

MUTAGENESIS
Doubled haploid production from mutant plants 238
Mutagenesis of haploid cells 239

HAPLOID CULTURE AND MUTATION BREEDING 240

PROTOCOLS IN MUTAGENESIS USING HAPLOID/DOUBLED 241

HAPLOID SYSTEMS
Barley: mutant haploid/doubled haploid production from M1 populations 241
Barley: in vitro haploid/doubled haploid mutant production 242
Doubled haploids in pedigree mutation breeding 245

xiii
SCREENING FOR HAPLOID/DOUBLED HAPLOID MUTANTS 246
Phenotyping 246
Genotyping 247

C) DNA MARKER AND GENOTYPING APPLICATIONS FOR 249

MUTATION BREEDING
Ingelbrecht, I., Jankowicz-Cieslak, J., Mukhtar Ali Ghanim, A., Anwar Y.,
Forster, B.P., and Till, B.J.

INTRODUCTION 249

ADVANTAGES AND USE OF MOLECULAR TECHNIQUES FOR 249

PLANT MUTATION BREEDING

MARKER-ASSISTED BACKCROSSING 253

GENOTYPIC SELECTION 254

ADDITIONAL APPLICATIONS OF DNA MARKERS FOR 256

MUTATION BREEDING
Excluding off-types using DNA markers 256
Background selection for clean-up of mutational load 257

EXAMPLE PROTOCOLS 257

Identification of SNPs and indels in large mutant populations using high- 257
throughput amplicon sequencing applicable to diploid plant species
Genotyping for shell thickness selection in oil palm 264

xiv
 FOREWORD

FAO estimates that global food production must increase by 70% to feed the projected
growth in the world’s population from about seven to nine billion people by 2050.
This cannot be done without increasing land productivity and conserving natural
resources in the face of the multitude of severe challenges posed, including climate
change, drought and flooding, soil erosion and salinization – all of which contribute
to hampering crop productivity and making it an increasingly high-risk venture.
Global cutting-edge R&D efforts are accelerating to both develop and put into practice
sustainable, climate-smart agricultural practices that are of distinct benefits not only
to agricultural productivity, farming communities and food security but also to the
environment and the natural resource base that must necessarily sustain these
agricultural practices.

As demonstrated during the Green Revolution of the 1960s and 1970s, which resulted
in numerous new high-yielding varieties, particularly of cereals, plant mutation
breeding can indeed play a crucial and most valuable role in meeting challenges
relating to food security. Whereas fifty years ago, mutation breeding focussed
predominantly on increasing yields, especially dwarf wheats and rice, the challenges
today are primarily related to increasing the tolerance of crops to environmental and
weather associated hazards and to driving opportunities for climate-smart agriculture.

The 2nd edition of the Manual on Mutation Breeding was published in 1977. After
nearly 40 years, and considering the numerous technological advances in this field, a
completely revised and updated version has now been long overdue. The third edition,
the Manual on Mutation Breeding that you now peruse, describes advances in plant
mutation breeding, in irradiation techniques as well as in the use of chemical
mutagenesis, in seed-propagated and vegetatively propagated crops, and in the types
of traits that we believe warrant urgent attention to achieve the set target of global and
nutritious food security for all. It also provides a comprehensive overview and
guidelines for new high-throughput screening methods – both phenotypic and
genotypic – that are currently available to enable the detection of rare and valuable
mutant traits and reviews techniques for increasing the efficiency of crop mutation
breeding.

Since its establishment in 1964, when the Food and Agriculture Organization of the
United Nations (FAO) and the International Atomic Energy Agency (IAEA) took the
visionary step of merging their mandates to create the Joint FAO/IAEA Division of
Nuclear Techniques in Food and Agriculture, the Joint FAO/IAEA Division has
remained a global leader in the application of irradiation for plant mutation breeding

xv
and crop improvement. Over 3275 mutant varieties in more than 220 plant species
have to-date been officially released worldwide (see http://mvd.iaea.org/). Their value
is measured in billions of dollars of additional revenue, in millions of cultivated
hectares and – most importantly – in innumerable people leading happy and healthy
lives. It is our sincere hope that this third edition will help the global community in
its endeavour to provide sustenance also for the 10% of the global population that
currently suffer from hunger and malnutrition.

Qu Liang
Director
Joint FAO/IAEA Division of Nuclear Techniques in Food and Agriculture

xvi
 ACKNOWLEDGEMENTS

This manual is largely based on the previous edition coordinated by Alexander Micke
and published as a Technical Series N°120 in 1977 by the Division of Publications,
IAEA Vienna. This edition resulted of the work of numerous pioneers in mutation
induction and mutation breeding including technical officers of the Plant Breeding
and Genetics section and laboratory and their collaborators from various parts of the
world. So, we wish here to gratefully thank these forerunners for their work but also
for having paved the way to the work of all the technical officers who followed in
their foot-steps. All the scientific topics covered in the 2nd edition were revisited,
updated and enhanced with the addition of new technologies and new scientific
results. More importantly the new technologies, such as in vitro and molecular
techniques, which appear to clearly and notably speed-up and widen the outcomes of
mutation breeding have been added.

We also wish to thank our counterparts and collaborators under the coordinated
research and the technical projects for their invaluable contributions to mutation
breeding. In fact, if the IAEA mutant variety database can now include more that 3200
new priceless mutant varieties, which cover thousands, if not millions of Ha and bring
additional income to farmers and agricultural systems all over the world, it is thank to
each, and every one of these passionate scientists and breeders.

Among the major additions in this edition are the practical examples and photographs
provided by some of the counterparts and collaborators and by active staff members.
Their contributions are sincerely appreciated and gratefully acknowledged. Thus, the
section would like to express their appreciations and indebtedness to the following
persons: Ms Luz Gomez-Pando; Ms Joanna Jankowicz-Cieslak; Mr Burak Kunter;
Ms Süreyya Sekerci; Ms .DGUL\H <DSUDN .DQWR÷OX, and Mr Penna Suprasanna and
their collaborators for providing some inspiring practical examples of successful
mutation breeding programmes.

The completion of this undertaking could not have been possible without the
assistance of Mr Indra Giri who assembled and so artfully crafted the schemes and
figures to better showcase the examples and/or methodologies provided, and
Ms Katayoun Allaf for her kind assistance in formatting the document according to
the FAO and IAEA Publication’s standards.

xvii
 INTRODUCTION
Spencer-Lopes, M.M., Forster, B.P., Mba, C., and Jankuloski, L.

Mutation, i.e. the heritable change to an individual’s genetic makeup, results in new
traits which are passed on from parent to offspring and thereby, drives evolution. In
nature, mutations are caused by errors in the replication of deoxyribonucleic acid
(DNA). This hereditary material could also be changed due to exposure to
surroundings’ natural radiations. A resulting modified individual is then known as a
spontaneous mutant.

Mutation is the underlying cause of evolution as an individual with a novel trait may
be preferentially selected for in nature – because its superior fitness arising from novel
(mutant) adaptive features – or artificially by man – because of the desirability of the
novelty.

Following the discoveries of X-rays by Roentgen in 1895; radioactivity by Becquerel

in 1896; and radioactive elements by Marie and Pierre Curie in 1898, it was shortly
afterwards demonstrated that radiation caused mutations in fruit flies (Muller, 1927)
and in the crop plants – maize and barley (Staedler, 1928). The subsequent rapid and
widespread adoption of induced mutations as a crop improvement tool derives directly
from these pioneering discoveries. It became evident that man did not have to wait for
chance discoveries of desirable off-type plants as was the case for our forebear hunter-
gatherers. Man could, in fact, induce mutations at will!

Mutation breeding has witnessed spectacular successes since the release of the first
induced mutant variety – a light green mutant of tobacco released in Indonesia in the
mid-1930s. Easy targets for plant mutation breeding are annual, inbred, seed
propagated crops: seeds are ideal for mutation induction and short life-cycles mean
mutant generations can be produced quickly and desirable mutant lines can be
developed into varieties rapidly. Thus, early successes were made in crops such as
rice, barley and tobacco and these have been sustained ever since. Next up are the
annual, outbred seed crops, these have a slightly more complex breeding system, but
nevertheless early successes in developing mutant varieties were reported and there
has been a continual production of new mutant varieties in crops such as maize by the
late 60s in various countries. More problematic are the vegetatively propagated crops
which have lagged, behind seed propagated crops in mutation breeding. This group
was targeted by Frantisek Novak circa 1980s. Working at the FAO/IAEA’s Plant
Breeding and Genetics Laboratory, Novak and his team pioneered tissue culture
methods needed for banana micro-propagation. Micro-propagation is essential as it
allows large numbers of cuttings to be produced for both mutation induction and
subsequent mutant lines development. The vegetatively propagated crops are

1
currently undergoing a renaissance with respect to plant mutation breeding as
numerous biotechnologies can be applied for efficient mutagenesis and mutant
screening, particularly tissue culture techniques. And last, but not least, are the
perennial crops, these naturally have long juvenile stages and have been neglected as
it takes many years for them to bear fruit. A further hindrance in plant mutation
breeding has been the reliance on phenotypic screening which is normally applied in
the second generation after mutation induction at the earliest. This modus operandi is
changing with the emergence of DNA analytical tools. Genotypic screening has the
potential to accelerate both mutation detection and mutant line development. This can
be applied to all crops, but has special relevance to (orphaned) perennial and
plantation crops such as oil palm, cocoa, rubber, tea and coffee.

According to the International Grains Council (www.igc.int) world estimates for

Maize production, for 2017/2018 stands above 1050 million tonnes. This grain cereal
together with wheat, barley and rice covers most of the food supply for the current
world population of 7.6 billion people. However, not everyone knows that this
incredible cereal found in so many various forms from pop-corn to porridge has been
manufactured by civilisations all over the world for about 10 000 years. In fact, it was
through the accumulations of several spontaneous mutations that Zea mexicana
(Schrad.) Kuntze or the Mexican teosinte evolved to become Zea mays (maize or
corn), the most widely produced cereal crop (FAOSTAT, 2014).

However, the greatest impetus for induced mutation breeding would come from the
establishment, under the aegis of the United Nations, in 1957 of the International
Atomic Energy Agency (IAEA) with its headquarters in Vienna, Austria in keeping
with the notion of “Atoms for Peace and Development”. In 1964, the Food and
Agriculture Organisation (FAO), another UN specialized agency – with a mandate to
eradicate hunger and malnutrition, and the IAEA established the Joint FAO/IAEA
Division of Nuclear Techniques in Food and Agriculture. The aim was to pool the
resources of both organizations to more effectively leverage peaceful uses of atomic
energy in assisting their member countries produce more, better and safer food. One
major activity of the Joint Division was the provision of support to Member States of
both organizations in establishing gamma-ray irradiators for plant mutation breeding.

The Plant Breeding and Genetics Section (PBGS), one of the 5 sections under the
Joint FAO/IAEA Subprogramme has, since then, devoted its effort towards promoting
the use of mutation induction for improving both productivity and crop nutritional
values using physical mutagens such as: X- and gamma-UD\VQHXWURQVĮȕDQGRWKHU
accelerators particles, ion or laser beam, etc.

The PBGS provides technical assistance to Members States in activities dealing with
mutation induction for crop improvement, organising trainings and expert missions

2
and coordinating meetings all relating to mutation breeding for crop improvement.
This support is provided through Coordinated Research Programmes (CRPs) and
Technical Cooperation Programmes (TCPs), in the form of research projects and
capacity building (e.g. laboratory equipment and training). The outputs of these
activities are generally published under different information formats from research
articles, to newsletters, protocols and manuals. Considering that there is an upsurge in
the use of induced mutations – for crop improvement and genomics – and at the same
time, there is the urgent need to generate novel useful heritable variations for
developing the nutritious, resilient, input use-efficient and productive crop varieties,
the Joint FAO/IAEA Division made the decision to update the existing manual. The
Manual on Mutation Breeding is founded on the experience and knowledge of the
PBGS staff, but has been written in collaboration with external experts. The eight
chapters of this 3rd edition provide up-to-date insights on the use of physical and
chemical mutagenesis on seed and vegetatively propagated crops. It provides
guidance on practical ways for conducting mutation induction to take a full advantage
of the power of this technique for unmasking novel alleles that could be incorporated
into plant breeding programmes or for studying the functions and patterns of
inheritance of genes and the traits that they control.

Among the most notable examples of mutation breeding is the development of barley
(Hordeum vulgare L.) semi-dwarf mutant varieties: Golden Promise and Diamant in
the early 1970s. These founding mutant varieties revolutionised the crop as they were
better adapted to mechanical (combine) harvesting, had higher yields (especially
under adverse weather conditions) and set a new standard for quality. The first
chapters describe physical and chemical mutagenesis and cover the action modes of
the different mutagenic agents with details of application procedures, the expected
outcomes as well as precautionary measures necessary for practical use. Chapters 3
and 4 explore the various types of mutations, which serve practical crop improvement
(mutant descriptors) and the more fundamental aspects of the exploration,
identification and understanding of mutations in DNA sequences. Chapters 5 and 6
deal with the practicalities of mutation breeding in seed and vegetatively propagated
crops, including the new processes in developing mutant populations, screening and
selection. The main factors affecting the success of mutation breeding such as
population size, propagation and isolation of mutant plants and examples of screening
for desired mutants are described in detail and illustrated with schematic diagrams
and photographs to assist practical mutation breeding. Chapter 7 attempts to provide
the most accurate and extensive examples of development of new mutant varieties
displaying superior traits and wider adaptability to ever-changing environmental
conditions. Finally, Chapter 8 considers new and emerging concepts and technologies
that are expected to impact on future plant mutation breeding. These focus on more
efficient methods in accelerating the mutation breeding process by embracing new
developments in pertinent biotechnologies, particularly in the areas of DNA analysis

3
and doubled haploidy. This 3rd edition demonstrates the practicalities of applied
mutation breeding, now more than ever before, to all crop species.

4
 1. PHYSICAL MUTAGENESIS

1.1.RADIATION TYPES AND SERVICES

This chapter is an update, largely based on the chapter on Mutagenic Radiation

presented in the second edition of the Manual on Mutation Breeding published in
1977. Physical mutagens comprise all nuclear radiations and sources of radio-activity
including ultraviolet light (a non-ionising radiation), several types of ionizing
radiations, namely X- and gamma-rays, alpha and beta particles, protons and neutrons.
An overview of the main physical mutagens used in plant mutation breeding is
presented here, covering their physical characteristics, their mode of action and all
general principles and consideration on how they may be applied for mutation
induction in plants.

Several types of ionizing radiation are available for plant mutation induction. Each of
these has the common feature of releasing ionizing energy. However, there are several
differences among ionizing radiations regarding the energy deployed the penetrating
capability and the level of hazard involved for operators (Table 1.1 and Figure 1.1).

1.1.1. X-rays

X-rays are known to originate from electrons and not from nuclear energy. Like
gamma-rays and ultraviolet (UV) light, X-rays are electro-magnetic radiations emitted
as quanta, their difference is based on the wavelengths; 0.001 – 10 nm for gamma-
and X-rays compared to 2000 – 3000 nm for UV light. In an X-ray machine, electrons
are electrically accelerated in a high vacuum and then stopped abruptly by striking a
target, e.g. a tungsten, gold or molybdenum barrier resulting in the emission of
radiation (Figures 1.2a,b). For mutation induction hard X-rays (short wavelength) are
usually preferred since their penetration is greater than soft X-rays (which have a
longer wavelength). The shortest wavelength emitted (except for constant potential
machines) is related to the peak operating voltage (kVp) of the X-ray tube, the higher
the kVp, the shorter the wavelength. Specific filters, e.g. aluminium filter: 0.5nm, are
often used in hard X-ray production to absorb unwanted soft radiation. The kVp,
milliamperes (mA), thickness and type of filter, distance of tube to target, dose and
dose rate affect the results and should always be recorded (Mehta and Parker, 2011).

1.1.2. Gamma-rays

In general, gamma-rays emitted by decay of an unstable nucleus of an atom, have a

shorter wavelength and therefore possess more energy per photon than X-rays. Mono
energetic gamma radiation is usually obtained from radio-isotopes, in contrast to X-

5
rays. A gamma irradiation facility can be used in a similar manner as an X-ray
machine for acute or semi–acute exposures. Gamma cells are the most commonly
used emitters for plant mutation induction, as of 2004 there were about 200 gamma-
cells in use world–wide (IAEA, 2004). However, the gamma radiation source has a
distinct advantage for prolonged treatments in that it may be placed in a controlled
environment chamber (Figures 1.3a,b), in a greenhouse (Figures 1.4a,b), or in a field
so that plants may be exposed at various times and at various developmental stages.

Mutagens

Physical Mutagens Chemical Mutagens

Non-ionising Akylating agents

Ionising radiations radiations e.g. mustard gas, EMS,
e.g. UV rays MMS

Particulate radiation Base analogues

e.g. alpha rays, etc. e.g. 5-chlorouracil

Acradine dye
Non-particulate radiations
e.g. acradine orange,
e.g. X-rays, gamma rays
proflavine

Figure 1.1. Common mutagens used in plant mutation induction.

6
Figure 1.2a. RS-2400 self-shielded X-ray irradiator (Produced by RAD Source Technologies Inc.,
USA) in the Entomology Unit - Seibersdorf Laboratories - Joint FAO/IAEA - Programme of Nuclear
Techniques in Food and Agriculture, Vienna, Austria (Mehta and Parker, 2011).

vacuum
housing
e-

gold target

X-ray tube principle

Canister

Canister Canister

X-rays
source

Canister Canister

X-ray tube (centre) surrounded by orbiting and rotating sample canisters.

Figure 1.2b. X-ray tube principle and structure

7
 TABLE 1.1. PROPERTIES OF RADIATIONS AND THEIR APPLICATIONS IN MUTATION BREEDING

Types of radiation Source Description Energy Hazard

X-rays X-ray tube with cylindrical anode Electro-magnetic radiation potent 1 to 500 KeV. Dangerous,
(Kirk and Gorzen, 2008) Mutagen agent for mutation breeding: penetrating
50-300 keV
Gamma rays Radio-isotopes and nuclear Electro-magnetic radiation like X-rays Up to several MeV Dangerous, highly
reactions penetrating
(60Co and 137Cs)
Neutrons (fast, slow, and Nuclear reactors /atomic piles Uncharged particles slightly heavier than From less than 1eV to Very hazardous
thermal) or accelerator a proton, not observable except through several million
[Californium (252Cf) and interaction with nuclei in the material it
Curium (248Cm)] traverses.
Beta particles Radioactive isotopes or Emitted by the nucleus of radio-nuclide Up to several MeV May be dangerous
(ß- - negatron; accelerators (7Be) with neutron/proton imbalance efficient

8
ß+ - positron and EC -electron only when incorporated directly into cells
capture)
Alpha particles Radio-isotopes from radio- Two protons and two neutrons emitted as 2 – 9 MeV Very dangerous
ĮHPLWWHGE\QXFOHXVZLWK nuclide (32P) nucleus decay products, not efficient in internally
atomic number >81 mutation due to very low tissue
penetration
Particles from accelerators Nuclear reactors or particle Beam of fast-moving, electrically 600 MeV to 2.75 GeV Very hazardous
accelerators charged atomic or sub-atomic particles
(i.e., Quarks)
Application ion implantation and/or
mutation
Figure 1.3a. Gamma-rays machine in the Joint FAO/IAEA Plant Breeding and Genetics Laboratory,
Seibersdorf – Austria. Courtesy of M. Matijevic.

Figure 1.3b. Loading of seed samples. Courtesy of M. Matijevic.

9
Figure 1.4a. External view of the greenhouse field in Kuala Lumpur, Malaysia. Courtesy of L.
Jankuloski.

137
Figure 1.4b. View of the plants arrangement in the greenhouse with the Cs source. Courtesy of S.
Nielen.

The isotopes Cobalt-60 (60Co) and Caesium-137 (137Cs) are the main sources of
gamma rays. In addition to naturally occurring radioactive isotopes, artificial gamma-
rays may be produced using cyclotrons (IAEA, 2004). Caesium-137 has a half–life of
30.17 years and is used in many installations as this is much longer than the half–life
of Cobalt-60, which is of 5.26 years. It should be noted that, both these radio-isotopes

10
must, at all time, be shielded in lead containers, for safety and security purposes. The
International Safety Standards for Protection against Ionizing Sources or Basic Safety
Standard Handbook, published by the IAEA in 2016 provides details for the safe
handling of these gamma sources.

1.1.3. Ultraviolet light

Ultraviolet light or UV light is a non-ionizing radiation at the wavelength commonly

employed (e.g. the 2537 nm line of mercury germicidal lamps), but it will be included
in this discussion because it has frequently been used in plant mutation induction
especially in pollen grains, cell and/or plant tissue cultures. UV radiations are
generally divided into three classes: UV-A, UV-B and UV-C. The UV-C region of
the UV spectrum includes wavelengths below 280 nm; UV radiation in the UV-B
region those from 280 to 320 nm, and UV wavelengths from 320 to 390 nm make up
the UV-A region of the spectrum.

Ultraviolet light has limited tissue penetration and its use is restricted to treating
sensitive materials, often single cells or single layer tissues, such as spores, suspension
cell cultures and pollen grains. However, the increasing use of cell and tissue culture
for mutation breeding of plants has led to increased use of UV light as a mutagenic
agent, especially when single mutant genes are sought (see Chapter 8). In order to
make quantitative assessments of experimental results, it proved necessary to use
monochromatic (or near monochromatic) UV-C light because it has confirmed
biological effects on photosynthesis, dark respiration and transpiration (Castronuovo
et al., 2014).

Initial research on UV light focussed on DNA damage, DNA repair, and pollen
irradiation. The latter showed the reactivation of transposable elements and thereby
indirect gene mutation, for example in maize (Jardim et al., 2015). Descriptions of
equipment and procedures for treating plant materials with UV light could be found
in Mba, (2013) and Mba et al., (2012). UV-B light has a strong effect on surface or
near-to-surface zones of plant cells including plastid structure (mostly thylakoid
membranes) and thus on photosynthesis (Kovacs and Keresztes, 2002).

1.1.4. Alpha particles

The alpha particles, are structurally equivalent to the nucleus of a helium atom, and
are emitted from radionuclides with atomic numbers larger than 82 such as Radium
and Plutonium (L’Annunziata, 2016). They are considered potential health hazards
when ingested or inhaled, but their low tissue penetration power, e.g. through
epidermis, renders them quite inefficient for mutation induction in plants (van Harten,
1998).

11
 1.1.5. Beta particles

Beta particles are emitted from the nucleus of an atom during radioactive decay
(L’Annunziata, 2016), and are effective in inducing mutations. Beta particles, such as
those from 3H, 32P and 35S produce in target tissue similar effects to those of X- or
gamma-rays, although the penetration of beta particles is lower than that of X- or
gamma-rays. However, this difficulty may be resolved by placing the radioisotopes in
a solution administered directly to the target plant material. Thus, 32P or 35S may be
incorporated directly into cell nuclei and induce mutations as observed in rice and
cotton (Mba et al., 2012). Due to variability from tissue to tissue and cell to cell, it is
difficult to determine the exact internal beta particle dose, and thus its use in mutation
breeding has been limited. A successful example is the mutation induction in rice
seeds using a solution of 32P as reported by Kharkwal, Pandey and Pawar, (2004).

1.1.6. Particles from accelerators

Amaldi, (2000) listed around 15 000 different types of particles accelerators world-
wide, e.g. the Cockroft-Walton and Van de Gruff accelerators, the betatrons, the
cyclotrons, the synchrocyclotrons, the synchrotrons and the linear accelerators. In
general practice, they are used to accelerate protons, deuterons and electrons. Particle
accelerators produce beams of energetic ions and electrons which may be deployed
for various purposes, including mutation induction in plants. Particle induced X-ray
emissions (micro–PIXE) with focused ion beams have recently been applied in plant
mutation investigations (see also section below on ion beam irradiation).

1.1.7. Neutrons

According to Byrne (2013), Pauli (1930) was the first to propose the idea that, to have
a better conception of the relations in the nucleus, in addition to protons and electrons,
there should be neutral particles, which he then called “neutrons”. The neutron is
stable only in the confines of atom nucleus, once separated from the nucleus the
neutron decays with a mean lifetime of about 15 minutes releasing various kinetic
energies. Table 1.2, below presents the different neutrons as per the energy released;
thermal (0.4 – 100 eV) and fast neutrons (200keV – 10MeV) are the most currently
used for plant mutation induction.

12
TABLE 1.2. CATEGORIZATION OF NEUTRONS ACCORDING TO THE ENERGY RELEASED
(L’Annunziata, 2016)

Cold neutrons < 0.003 eV

Slow (thermal) thermal neutrons 0.003 – 0.4 eV
Slow (epithermal) neutrons 0.4 – 00 eV
Intermediate neutrons 100 – 200 keV
Fast neutrons 200 KeV – 10 MeV
High energy (relativistic) neutrons >10 MeV

The radio-nuclide Californium-252 (252Cf) is currently the most commonly used as a

spontaneous neutron source (Karelin et al., 1997). Neutrons have been shown to be
highly effective for the induction of mutations in plants. Even though, their practical
use has been hampered by a lack of adequate dosimetry techniques, by a lack of
uniform dosimetry among various reactor facilities, and by the range of procedures to
report the dose and spectrum of neutrons used. Hence, a certain degree of confusion
exists concerning the results of early neutron experiments. In the past decade,
however, situation has improved considerably and recommended procedures are now
available for mutation induction, e.g. in the report on Neutron irradiation of seeds
(IAEA, 1973).

1.1.8. Ion beam irradiation and ion beam implantation

The demonstration of the effectiveness of ion beam irradiation for mutation induction
in tobacco embryos during fertilization without any damage to other plant tissue led
to the widespread use of this technology, for instance in Japan (Abe et al., 2007).

Ion beam implantation is a method where atoms are injected into the surface layers; a
process mostly used in industry, but it can also be applied on plant tissues and this
turned out to be an effective tool in mutation induction. Feng and Yu reported in
Shu et al., (2012) that the effects of ion beam implantation on plants were first
revealed in the late-80s by Ziegler and Manoyan (1988). The latter author extensively
described the process of ion beam implantation and the resulting effects. It has been
shown that this method has many outstanding advantages including: lower damage
rate; higher mutation rate and wider and novel mutational spectrum. Many new mutant
lines of rice with higher yield, broader disease resistance, shorter growing period but
higher grain quality have been bred; several new lines of cotton, wheat and other crops
have also been developed (Zengquan et al., 2003; Shu et al., 2012).

13
 1.1.9. Cosmic irradiation

Cosmic rays were discovered by Victor Hess in 1911–1913, as reported by

L’Annunziata, (2016). These so called “air showers”, which constantly strike the top
of the atmosphere, consist of cascades of sub-atomic particles and electro-magnetic
radiations resulting from the various collisions occurring in space (L’Annunziata,
2016).

Cosmic radiations have been largely investigated in plant biology and mutation
induction. For example, in maize, somatic mutations, including white-yellow stripes
on leaves, dwarfing, change of leaf sheath or seedling colour were observed in plants
developed from seeds flown into space. In China, sHYHUDOQHZPXWDQWV§LQULFH
wheat, cotton, rapeseed, sesame, pepper, tomato and alfalfa were obtained and
released as new varieties from seeds subjected to space travel (Liu et al., 2009) . This
has led to the establishment of new techniques and methods of mutation induction by
simulating the space environment in laboratories.

1.1.10. Laser beam irradiation

In recent years several research groups have investigated the efficiency of laser beams
in radiation induced mutation. Interesting findings have been presented mainly on the
changes in cells, organelles and genomes caused by laser beam. Rybianski (2000)
demonstrated the effectiveness of a helium-neon laser (He-Ne) with a wavelength of
632.8 nm and a power density of 1mWcm–2 in inducing phenotypical mutations in
barley; some of the mutants showed good agronomic traits such as high yield.

1.2.RADIO-BIOLOGY

1.2.1. Absorption of ionizing radiation

As described earlier, there are many types of radiations, but the two most common are
electro-magnetic radiations and ionizing radiations. Electromagnetic radiations are
described as waves of photons. Ionizing radiations refer to radioactive particles, such
as alpha and beta particles, but include also some electro-magnetic waves, such as X-
or gamma-rays, which have sufficient energy to detach electrons from atoms and
create ions, hence the name “ionizing radiation.” Absorbed energy from ionizing
radiation induces changes in plants at the molecular level, i.e. the macro-molecules
such as DNA or enzymes or even some smaller molecules such as ATP and co–
enzymes (Harrison, 2013). The effect of radiation involves two mechanisms: i) the
direct (physical) action, which reflects on the molecule damage and ii) the indirect
(chemical) action from the highly reactive free radicals derived from ionized water
molecules (Lagoda et al., 2012). The process of irradiation involves physical,
physico-chemical, chemical and biochemical, and biological effects and the ionizing

14
radiations can thus, affect plant growth and development, the severity or magnitude
of the observable modifications is strongly dependent upon several factors including
species, genotype, plant age, physiology and morphology as well as the plant genome
size and organization (Lagoda, 2009).

The ionization events caused by X-rays, gamma-rays and beta particles occur sparsely
along the track of the ionizing units. When the ionizing particle consists of an atomic
nucleus, alpha-rays or recoil nuclei knock-out by fast neutrons, the ionization events
are relatively dense. Each radiation is characterized by its ion density, which is
expressed in terms of Linear Energy Transfer (LET); the energy dissipated per unit
length along the tracks of the ionizing particles. Gamma- and X-rays are examples of
low LET radiations, alpha–rays and fast neutrons are high LET radiations. De Micco
et al., (2011) provided a comprehensive review on the different effects of irradiation
in plants focusing on genetic alterations, modifications of growth and reproduction
and changes in biochemical pathways especially photosynthetic behaviour.

1.2.2. Chemical effects of ionizing radiation: DNA damage and DNA

repair

Mutation induced by physical mutagens is brought about by the chemical effects in

the living cell. In the ionization process positive radical ions and free electrons are
produced. When penetrating the biological systems, the electron is trapped in the polar
surroundings and then the radical ion, which is both unstable and reactive, may either
react with other molecules or pass through internal rearrangements. In water solutions
the free electron can polarize several water molecules and become a “hydrated
electron” (eí aq). The free radicals generated in solution will sooner or later recombine
with each other to form stable products. There are several definitions for the terms:
free radical. Halliwell and Gutteridge, (2015) in a review on free radicals proposed a
simple definition – a free radical is any reactive species capable of independent
existence that contains one or more impaired electrons – and a superscript dot is used
to denote a free radical, e.g. H• and O2•. A free radical can be formed by losing a single
electron from a molecule, leaving behind an impaired electron and a positive charge:

x X – Ǩ– X•+ (radical cation)

or by gaining a single electron

x <Ǩ– Y• – (radical anion)

Radicals can also be formed if a covalent bond is broken to leave one electron from
the bonding pair on each atom; a process called “homolytic fission” (Halliwell and
Gutteridge, 2015). If molecular oxygen (a bi-radical) is present, it reacts readily with
radiation induced free radicals to form peroxy-radicals. In the solid state, where

15
molecular movements are restricted, the radiation-induced radicals are stable. This is
the case in plant seeds of low water content. The higher the water content of the
irradiated target (e.g. seed) the greater the related damage.

1.2.3. Lethal effects of ionizing radiation: DNA damage and repair

The lethal action of ionizing radiations on cells is often measured as the loss of cell
division, i.e. mitotic activity. Indirect evidence that this is the result of chromosomal
aberrations has been obtained in several systems. Muller and then Staedler in the
1920s were the first to demonstrate that applying mutagenic agent, such as X-rays, on
living cells resulted in phenotypic mutations. These observations were rapidly
connected with the DNA described in the 50s, as resulting from DNA damages.
Chromosomal aberrations were soon detected in the dividing cells at anaphase or as
micro-nuclei, but non-dividing cells also died after irradiation, this is termed
“interphase death”, although this generally requires larger doses of radiation than for
reproductive cells (especially those undergoing meiosis). DNA damages can be
broadly classified into three types of lesions: mismatched bases, double-strand breaks,
and chemically modified bases.

When the effects of gamma-rays and, for example, fast neutrons are compared, it is
found that gamma-ray treatments have a greater biological effect, also called Relative
Biological Effectiveness (RBE). Several investigations have shown that the RBE is a
function of the linear energy transfer (LET). Later, it became largely recognized that
all organisms, which very often recover after irradiation and/or show DNA damages,
also possess a range of conserved biochemical activities responsible in restoring the
DNA to its undamaged state, this is called: “DNA repair” (Croteau and Bohr, 2013).
DNA repair studies often focus on chemical lesions, but the repair pathways for the
other two classes of damage are particularly interesting to the plant geneticist and have
been investigated thoroughly. In both bacteria and animal cells, this repair is known
as efficient and fast, similar findings have been reported in higher plants (Gill et al.,
2015). It is, however, interesting to note that a fraction of the breaks remains
unrepaired. This fraction of unrepaired breaks is significantly higher after fast neutron
irradiation than after gamma-rays (Lagoda et al., 2012); Croteau and Bohr, 2013).
One reason for this could be the high ion density after fast neutron irradiation, which
leads to a locally more extensive or “clustered” damage, and is therefore more difficult
to repair.

1.3.DOSIMETRY

Radiation dosimetry deals with methods for the quantitative determination of

absorbed dose (of the target material) and its physical interpretation. Absorbed dose
is the energy which is deposited in a unit mass of matter, because of interaction of

16
radiation with this matter. From the symposium co-organized by the IAEA and the
World Health Organization (WHO) in the early 70s, some classical measurements
systems were thoroughly evaluated and presented and regularly revisited and the
results published by the IAEA in a series of Technical Reports (IAEA,1973). Since
then, a substantial development has been observed due, both to the various radiations
sources developed, the multitude of uses and applications of these radiations, and
expected outcomes.

1.3.1. Exposure and dose determination

In many practical cases absorbed dose is not measured directly but calculated from
the measured number of ions produced in air by the ionizing radiation. Measurements
of this kind are done with ionization chambers and the quantity measured in this way
is called 'exposure'. Exposure is symbolized by “X” and has been defined by the ICRU
(1998)DV³WKHTXRWLHQWRIWKHHOHFWULFFKDUJH¨4RIHLWher sign) produced in dry air
in standard conditions by X- or gamma-radiation to the masV¨PRIWKHDLU´ZKHQDOO
the secondary electrons liberated by the photons are completely stopped in the
corresponding air volume).
οொ
X=
ο௠ೌ೔ೝ

The SI unit is: C kg-1

Another important measure is the exposure rate, which expresses the variation of
exposure over time:

The exposure rate ‫ ܩ‬is defined as:

௱௑
‫=ܩ‬
஽௱௧

Usual unit of exposure rate is: C Kg-1s-1

The exposure rate in air, ‫ܩ‬, is inversely proportional to the squared distance (‫ ݎ‬ଶ ) from
the point source of activity, ǻ(t):
୻ο(௧)
‫=ܩ‬ ௥మ

īLVWKHVSHFLILFH[SRVXUHJDPPD-ray constant at 1 m distance from the source and is

defined for each radioactive nuclide (ICRU, 1998; AdlieQơDQG$GO\Wơ.

17
 1.3.2. Absorbed dose in irradiated targets

In quantitative radiation biology it is desirable to relate the observed biological effect

to a well-defined and easily measurable physical quantity characterising the amount
of radiation responsible for the effect. It was generally accepted that the "radiation
absorbed dose or rad” best meets this demand. Therefore, it is common and convenient
to describe radiation effects in biological targets in terms of dose-effect relationships.
The absorbed dose or, briefly, dose: D for any ionizing radiation is defined as the
amount of energy absorbed per mass of irradiated matter at the point of interest. In
1925, the First International Congress for Radiology in London saw the Foundation
of the "International Commission on Radiation Units and Measurement" this
organisation regularly publishes reports and provides updates on dosimetry for
various types of radiations (ICRU, 1998). As reported by Podgorsak (2005) the
DEVRUEHGGRVHFRUUHVSRQGVWRWKHTXRWLHQWRI¨Eab E\¨PZKHUH¨(ab is the mean
absorbed energy imparted by the ionising radiation to the mass of the medium used.
οாೌ್
D=
ο௠

However, the rad is not any more in use, the current SI unit of absorbed dose is the
gray (Gy), which corresponds to 1 Joule per kg (1Gy = 1Jkg-1), and 1gray = 100 rads.

It should be noted that, although the concept of absorbed dose is independent of any
specified material, irradiation of two different materials under exactly equal
conditions will generally lead to different absorbed doses in the two materials; this is
because of the different absorption coefficients of the two materials.

Podgorsak (2005) defined a dosimeter as: “any device that is capable of providing a
reading that is a measure of the average absorbed dose deposited in a sensitive volume
by an ionizing radiation”. A dosimeter can generally be considered as consisting of a
sensitive volume filled with a given medium lying in a container of another medium
to ensure accuracy. The dosimeter and the associated reader are generally referred to
as: dosimeter system.

Following thorough reviews of the various existing dosimetry systems applied in

radio-biology and mutation breeding; some requirements of dosimetry systems as well
as practical sequence of steps required for efficient plant mutation induction and have
been established and have been largely followed by breeders over the years as refer
to by Legrand, Hartmann and Karger (2012). These studies also recognized the work
by 'UDJDQLüand Gupta, 1973) who specified that to serve as a measure of absorbed
dose, any dosimetry systems must satisfy a number of criteria.

18
 a. The amount of chemical changes should be linearly proportional to the dose
absorbed, or at least, the calculation of the corresponding dose must be
simple.

b. Any side events during the irradiation, i.e., accumulation of radiolysis

products, should not interfere with the calculation of the absorbed dose which
must remain simple.

c. The amount of chemical changes must be independent over a wide range of

LET of radiation, of dose rates and of temperatures, when it is not the case
there must be a sure way to integrate those factors.

d. Chemical dosimeters require chemically pure substances, not needing any

additional purification prior to be used in the dosimeter and the sample
should allow a normal handling under laboratory conditions.

e. A method that is accurate, but also simple should be used for determining the
chemical change which serves as measure of the absorbed dose of energy.

1.3.3. Dosimeters

A major objective in plant mutation breeding programme is to obtain the highest

mutagenic efficiency, i.e. a maximum number of desired mutations, while preserving
the viability of the plants and a low background mutation rate. Thus, the plant breeder
must assess the most appropriate mutagen dose, because the damages involved may
create an additional cost in terms of manpower, time and field area, and the rounds of
crossing/backcrossing needed to clean up unwanted background mutations. These
considerations lead to careful consideration of dosimetry, meaning practical
measurements of the mutagen quantities applied as well as that received by the target
plant material. Brunner, 1995 proposed a set of steps that should be considered both
at the level of the mutagen and at the level of the plant itself (Table 1.3).

Different dosimetry systems are used for different purposes in industrial and research
irradiation facilities. They may have different requirements for dose determinations.
Radiation safety standards and issues involving the radiation protection of humans
against radiation exposure have their own dosimetry metrology $GOLHQơDQG$GO\Wơ
2017). Some dosimeters commonly used in mutation induction are briefly described
below.

19
TABLE 1.3. STEPS INVOLVED IN DOSE ASSESSMENT FOR CROPS SPECIES IN A
MUTATION BREEDING EXPERIMENT (Brunner, 1995)

A. Radiation source characteristics

High or low LET radiation
Energy distribution
Degree of contamination with other radiations
Dose gradients, requirements of dose homogeneity
Methods to monitor radiation dose and/or dose rate:
(i) Physical Ionization chambers, threshold detectors, etc.
(ii) Chemical Determination of ionic (chemical) yield, e.g. Fricke
(iii) Biological Determination of an index of primary damage, e.g. seedling
height, epicotyl length in comparison with non-irradiated control
materials
B. Characteristics of the biological target
Seeds Whole plants
Pollen grains Vegetative organs
Gametophytes and zygotes Cells and/or tissue in culture
Criteria of radio-sensitivity
Biological factors, environmental factors, etc.
C. Prediction of dose effects
Early assessable criteria of primary damage in e.g. seedling height of the first leaf, epicotyl
length, etc. and their correlation to mutation frequency in M2, e.g. Usually to chlorophyll
indicator mutations.

1.3.3.1.The Fricke dosimeter

The Fricke dosimeter is an aqueous solution containing ferrous ions (Fe2+), which are
oxidized to ferric ions (Fe3+) under ionizing radiation proportionally to the absorbed
dose (Fricke and Hart, 1966). The Fricke or ferrous sulphate dosimeter is a very useful
chemical system for gamma irradiation dose measurements. With some modifications
it can also be used for neutrons and mixed gamma/neutron fields, but with a reduced
accuracy. Combined use of ionization chambers and the Fricke dosimeter is
recommended whenever it is desirable to have two independent dose measurements
(Boudou et al., 2004).

20
The standard Fricke solution for measurement of gamma doses consists of

x Ferrous ammonium sulphate: Fe (NH4)2, 6 (H20) at 10–3 mol/l

x Sodium chloride: NaCl at 10–3 mol/l
x Sulphuric acid: H2S04 at 0.4 mol/l concentration in aqueous solution.

The standard Fricke dosimeter can be used in the dose range 40 – 400 Gy for gamma
radiation and for doses rates up to 106 Gy/s (deAlmeida et al., 2014). Since it provides
an accurate and direct dose determination, it can be used for calibration of other
systems, e.g. ionization chambers. However, it such cases, one has to take into account
the factor G, which measures the radiation chemical yield and may vary with the types
of radiation (Klassen et al., 1999).

1.3.3.2.The FeCu dosimeter – (ferrous sulphate–copper sulphate dosimeter)

As stated above, the Fricke dosimetry is applicable in the dose range between 40 Gy
and 0.4 kGy. For higher doses, the system goes into saturation. With a modification
of the solution it is possible to shift this measuring range to higher doses; raising the
Fe2+ concentration delays the saturation effect up to 1.5kGy. The addition of Cu2+
enables dose measurements up to 25kGy. This is used mainly when the gamma and
fast neutron doses from a mixed field are to be determined separately (Haninger and
Henniger, 2016).

Composition of the solution:

x Ferrous sulphate: FeSO4 10–3 mol/l

x Copper sulphate: CuSO4 10–2 mol/l
x Sulphuric acid: H2S04 5x10–3 mol/l

The following G values have been reported for cobalt 60 and neutrons (deAlmeida
et al., 2014) :

G (FeCu) = 0.66 for 60Co radiation

G (FeCu) = 2 for fission neutrons.

1.3.3.3.The Fricke-gel dosimeters

Even though the above chemical dosimeters have been used routinely, modern users
have found several limitations, for example in terms of stability of the air after
irradiation, moreover because these were mostly one- and two- dimensional

21
measuring devices. Thus, there has been much interest in the development of a three-
dimensional (3D) radiation measuring device. The most promising development in
3D dosimetry has been the addition of polymers and gels, which is then measured by
magnetic resonance imaging. The first developments in this field were carried out
with gels infused with Fricke–gel solution (Schreiner, 2004).

1.3.3.4.Neutron dosimetry

Neutrons, as well as photons ionize indirectly, so the methods of dosimetry used are,
in general, similar. However, there will be a wide range of variation depending on the
target; live tissue or physical material. The challenge in using ionization chamber is
to have a phantom material as close as possible to the target tissue. The usual phantom
material is water and the calculation is made using the ICRU (1993)
recommendations.

1.3.3.5.Personal dosimetry

The applications of X-rays, gamma-rays, neutrons and electron beams have proven to
be of great benefit in medicine, biology, industries, etc. However, these applications
also mean substantial acute or recurrent exposure for the operator/personnel involved.
The latest IAEA recommendations for radiation protection are provided in the Basic
Safety Standards series (IAEA, 2010, 2014 and 2016). They are based on the
knowledge of radiation effects and on established principles of radiation protection
recommended by the International Commission on Radiological Protection $GOLHQơ
DQG$GO\Wơ.

Different methods are used to assess the occupational dose of an individual staff
depending on the different exposure situations in the radiation fields (external
exposure) and/or on the different radionuclides (intake), which might contribute to the
dose. The ICRU (1993) defined several units for assessing the exposure of workers to
radiation, among which; the dose equivalent (H) measuring the dose absorbed at a
point by an organ or tissue multiplied by the relevant weighing or quality factor (Q)
depending on the type of radiation, which is pre-established for each one of them:
alpha particles, electrons, photons or neutrons. The dose equivalent unit is the Sievert
(Sv), which is equal to 1 joule of energy deposited in a kilogram of human tissue
$GOLHQơDQG$GO\Wơ.

Moiseenko et al., (2016) in their study on the effects of radiation on workers and
population 30 years after the Chernobyl disaster (1986), highlighted the main
requirements for biological dosimetry methods for a reliable and useful assessment:

22
 x Low detection threshold.
x Low person– to– person variation in dose–response for healthy individuals.
x Ability to obtain calibration curves in laboratory conditions, e.g. in vitro.
x Stability of the biological effect so that dose can be reconstructed at long
time periods, years or decades, after being exposed.

The authors mostly referred to the evaluation of the different types of biological
dosimetry methods established by the IAEA (2011).

1.4.OBJECTIVES AND TREATMENTS

All plant parts can be irradiated by one method or another, but some are easier to treat
than others. Besides the commonly treated seeds and pollen, whole plants, cuttings,
tubers, corms, bulbs, stolons and organs tissues, or cells cultured in vitro may also be
irradiated.

There are wide differences in the radio-sensitivity of the various plant parts; the
reaction of a given type of cell to radiation depends on its physiological conditions at
the time of irradiation as well as on the pre- and post-irradiation conditions. The
decision to be made by the investigator concerning the most appropriate plant part or
stage to expose requires a thorough knowledge of the organism and clear objectives
for the experiment.

1.4.1. Target plant materials

1.4.1.1.Whole plants

Large plants have often been irradiated in a gamma field, a gamma greenhouse or a
gamma room. Seedlings or small plants, on the other hand, can easily be irradiated by
most X-ray machines or by gamma sources in a greenhouse or shielded rooms. Today,
the use of (open) gamma fields is extremely reduced considering the restrictions
associated with environmental and human health concerns; other more contained
methods of treatments are available and have proven to be more economic for plant
breeding, at least for seed propagated plant species.

1.4.1.2.Seeds

Seeds are the favoured material for irradiation in many mutation induction
experiments and in practical mutation breeding. Seeds can be irradiated in many
physical environments and they can be desiccated, soaked, heated or frozen prior to
the treatments. They can be stored for extended periods of time in air-tight, vacuum

23
and refrigerated conditions. When dry, the seeds are almost inert biologically
(quiescent), they are also easiest to handle and can be shipped over large distances.
However, larger radiation doses are required to produce sufficient genetic mutations
than when other plant materials are irradiated. On the other hand, soaking of seeds
before irradiation can reduce the dose level required, but might also introduce some
complicating factors as it will promote germination.

1.4.1.3.Pollen

A great advantage of pollen irradiation as opposed to the irradiation of seeds or

growing plants is the fact that the former rarely produces chimeras, i.e. the M1 plants
resulting from fertilization by irradiated pollen of a non-irradiated egg cells will be
fully homozygous for any induced mutation (see Chapter 8.2). The disadvantages of
pollen irradiation include the difficulty of obtaining sufficient material from some
species and the short viability of pollen grain in many plant species. By using proper
techniques, however, pollen of some species can be kept alive for several months (or
years for species that have bi-nucleate pollen, e.g. oil palm) and can be used for
germplasm conservation. Large quantities of pollen are obtainable from most
naturally cross-pollinated plants, for example a maize plant can produce 14 to 50
million of pollen grains. Pollen grains are generally considered to be the most suitable
plant part for UV irradiation treatment.

1.4.1.4.Meristems

Seed irradiation is essentially a treatment of the embryo meristems. The anatomy and
pattern of embryo meristems is important for mutagenic treatments of seeds (as well
as other plant material) since it determines whether a mutated cell will be lost during
differentiation or produce sufficient cell progenies to be found throughout much of
the plant including germ cells.

For most vegetatively propagated crops (VPCs), seeds are not available and therefore
other plant parts are needed as targets for mutation induction. The structure of
meristematic regions and the development of new meristems from differentiated
tissue are particularly important when investigating radiation induced mutation of
VPCs. In most cases, the new shoots originate from a single epidermal cell from a
tissue and this could directly lead to homohistant mutant plants whose genetics may
be investigated further. For more information (see Chapter 4. 6 and 8).

1.4.1.5.Plant cells and in vitro tissue culture

The use of plant cells and tissues in culture offers exciting applications in crop
mutation breeding. The in vitro methods for plants micro-propagation launched in the

24
early 60s, rapidly became a powerful tool for scientists working on plant mutation
induction, especially in VPCs. Plant organ, tissue and cell culture provided ways for:

x rapid and mass propagation of target M0 plants;

x rapid and mass propagation of any mutant populations;
x thorough and easy methods for analysing mutations morphogenetic and
physiologic changes associated with mutation induction;
x easy and rapid ways of dissociating chimeras;
x in vitro mutagenesis; and
x in vitro screening.

1.4.2. Irradiation treatments and conditions

There are several experimental variables to be considered for irradiation treatments.

1.4.2.1.Irradiation dose and dose rate

The dose percentage at which a given dose of radiation is administered – dose rate –
frequently has a significant effect, both qualitatively and quantitatively, on the results
obtained. For this reason, the dose rate should be carefully chosen and recorded in all
experiments. For most species/targets recommended dose treatments are available
from the literature (Table 1.4) and may be applied in every specific condition on the
same crop with some precautionary measures, such as using the given value X and
two additional ones: X ± 20%.

1.4.2.2.Acute versus chronic irradiation

The frequency of mutations induced by ionizing radiation in a genome appears to be

directly proportional to the dose. Different results might be expected depending on
whether the induced mutations are caused as a-one-hit event, such as simple
chromosome deficiency or deletion, or if they appear as point mutations resulting from
several one-hit events distributed throughout the genome. There is, therefore, a need
to weighing the need for high dose during a short period versus low dose during a
longer time. Kovalchuk et al., (2000) reported observing a strong and significant
correlation between the frequency of homologous recombination (HR) in plants, the
radio-activity of the soil samples and the doses of radiation absorbed by plants.

Exposures that are continued over long periods of time (usually weeks, months or
years) are referred to as “chronic”, exposures delivered in minutes or a few hours are
referred to as “acute”. Almost any source of radiation can be used for acute exposures.

25
A comparison of acute versus chronic irradiation usually implies a comparison of high
versus low dose rates and/or high versus low radioactivity for gamma emitters.

Comparative studies on the effects of acute and chronic doses of X- and gamma-rays
have been made in terms of growth, survival, fertility, and yield and mutation
induction after exposure of both seeds and plants. Acute irradiation of seeds might be
more effective in inhibiting growth and decreasing survival and fertility because of
the recovery phenomena at low irradiation intensities with chronic irradiation.

1.4.2.3.Recurrent irradiation treatments

The procedure of recurrent irradiation – irradiating plant material that had already
been irradiated in one or more subsequent generations – has been proposed as a
method of accumulating and expanding genetic variability to be utilized in plant
breeding. Appropriate strategies in mutation induction as application of fractionated
doses and recurrent irradiation combined with in vitro culture techniques have been
chosen in assessing their cost-effective applicability in ornamental plants.

Since various physical and chemical agents are known to cause different types of
mutations; recurrent treatments using various chemical mutagens such as ethyl
methanesulphonate (EMS) or hydroxylamine (HA) separately in a combination or
alternatively associated with physical mutagen such as X-rays and also the
combination of UV light and X-rays have also been investigated in several plants
(Chopra, 2005).

26
TABLE 1.4. RADIO-SENSITIVITY OF SOME CROP SPECIES TO GAMMA AND FAST
NEUTRON RADIATION (Brunner, 1985)
Genus or Family Species Range of GR 50* Typical doses
(common name) (dose in Gy) ܵ (dose in Gy) ܵ
POACEAE
Oat Avena sativa 300-450 100-250
Barley Hordeum vulgare 300-450 100-250
Rice Oryza sativa
japonica 250-400 100-280
indica 350-500 150-350
Rice Oryza glaberrima 300-400 150-300
Bread wheat Triticum aestivum 450-600 150-350
Durum wheat Triticum durum 350-500 150-300
FABACEAE
Peanut Arachis hypogaea 300-450 100-350
Pigeon pea Cajanus cajan 150-240 80-150
Chickpea Cicer arietinum 180-300 100-200
POLYGONACEAE
Buckwheat Fagopyrum esculentum 300-500 150-300
BRASSICACEAE
White mustard Sinapis alba 900-1500 500-1000
Wild turnip Brassica campestris 800-1600 500-1000
ollifera
Indian mustard Brassica juncea 1600-2000 1000-1500
SOLANACEAE
Red pepper (chilli) Capsicum annuum 250-500 100-350
Tomato Lycopersicon esculentum 450-600 200-400
LILIACEAE
Onion Allium cepa 160-280 80-180
Leek Allium scorodoprasum 200-250 80-140
Asparagus Asparagus officinalis 300-400 150-250
UMBELLIFERAE
Carrot Daucus carota 550-700 250-400
CHENOPODIACEAE
Spinach Spinacea oleracea 300-500 150-300
Quinoa Chenopodium quinoa 300-500 150-300
MALVACEAE
Cotton Gossypium arboretrum 140-250 80-150
Cotton Gossypium hirsutum 300-500 150-300
Okra Hibiscus esculentus 600-850 300-500
ASTERACEAE
Sunflower Helianthus annuus 250-500 100-300
Safflower Carthamus tinctorius 600-700 200-450
Niger Guizota abyssinica 200-260 80-160
TILIACEAE
Jute Corchorus Olitorius 2n 700-850 300-550
Corchorus Olitorius 4n 550-700 250-450

27
 Genus or Family Species Range of GR 50* Typical doses
(common name) (dose in Gy) ܵ (dose in Gy) ܵ
CUCURBITACEAE
Squash Cucurbita maxima 500-700 250-450
Cucumber Cucumis sativus 450-600 200-400
Melon Cucumis melo 350-500 200-350
PEDALIACEAE
Sesame Sesamum indicum 700-900 400-700
LINACEAE
Flax Linum usitatissimum 600-1000 300-600
*GR 50 = 50% seedling height (or epicotyl height) reduction after irradiation of quiescent seeds equilibrated to
12-14% moisture with 60Co ܵ radiation (dose rate varied between 60 and 7 Gy min-1) or with fast neutrons derived
from the SNIF (Standard Neutron Irradiation Facility). Precision of the applied doses: ± 5% and irradiation of dry,
quiescent seeds with 12 14% moisture.

1.5.RADIATION SENSITIVITY AND MODIFYING FACTORS

The response of cells of higher plants to physical and chemical mutagens is influenced
to a varying degree by numerous biological, environmental and chemical factors.
These factors modify the effectiveness and efficiency of mutagens in the cells of
higher plants. The mechanisms involved are poorly understood, but it is very
important to monitor these factors closely, as they may interfere with the process of
radiation.

The two most important modifying factors for seed irradiation are oxygen and water
content, whereas for active tissues, factors such as stage of development including the
relation to DNA synthesis and dose rate are more important. Factors such as nuclear
and interphase chromosome volumes are also important for, both, resting and active
tissues. Specific problems related to the mutation breeding of VPCs are discussed in
Chapter 6.

A protocol for radio-sensitivity testing is given in Section 1.7. below. Here we discuss
the factors that modify the response of seeds to ionizing radiations, they may be
grouped into two major categories: (1) environmental factors such as atmosphere
(oxic versus anoxic), seed water content, post-irradiation storage, and temperature;
and (2) biological factors such as genetic differences, nuclear and interphase
chromosome volumes, etc.

1.5.1. Environmental factors

1.5.1.1.Oxygen

Oxygen is one of the best-known modifiers of radiation sensitivity and the biological
effects of irradiation are usually greater in the presence of oxygen. Other factors, such
as water content, temperature and post-irradiation storage conditions appear to be

28
secondary. Environmental factors are less important with densely ionizing radiations
such as fast neutrons. Using yeast (Saccharomyces cerivisae), Nairy et al., (2014)
conducted a comprehensive study of Oxygen Enhancement Ratio (OER) and their
variation as a function of the radiation doses.

Oxygen enhancement effect of gamma radiation induced damage is greatest in very

GU\VHHGV” percent water content). The degree of enhancement, however, is not the
same for all species. In general, higher mutagenic efficiency (less damage in terms of
seedling injury and chromosome aberrations in relation to mutation frequency) can be
obtained if oxygen effects are minimized. Data from various experiments show that
this can be accomplished by irradiating seeds in an anoxic atmosphere – in nitrogen
filled medium or under partial vacuum – or by adjusting the seed water content to
12 – 14 percent.

However, the control of oxygen is often impractical for general mutation induction in
plant breeding and is often ignored. Additionally, the oxygen applied under pressure
can, by itself, act as a mutagen.

1.5.1.2.Water content

In all cases of mutation induction, including seed irradiation, water content is an

important and easily regulated factor (van Harten, 1998). Various methods have been
used for determining the water content of seeds: weight difference between fresh and
dehydrated seeds using oven or desiccator systems; use of saturated solutions of
various salt; and the use of the glycerol-water solutions (Forney and Brandl, 1992).

Even minor differences in water content can have a very pronounced influence on the
end biological effect. Seeds stored under normal laboratory conditions are often in the
range of 10.0 and 11.5 percent water and a difference of only 0.2 to 0.3 percent might
greatly alter radio-sensitivity of some species. This should be weighed in, when
choosing an effective mutagenic dose; a dose too low might not induce any mutations,
and a dose too high might result in excess sterility or no surviving plants. Seeds of
different species equilibrate water content at different rates and may differ not only in
water content but also in radio-sensitivity when equilibrated at the same relative
humidity (IAEA, 1977). Thus, pre-irradiation desiccation is generally considered a
routine treatment for seeds.

1.5.1.3.Temperature

The temperature of plant cells before, during and after irradiation may affect the total
amount of genetic damage induced by X- or gamma-rays. However, the effect of
temperature as a modifying factor of radiation damage is not clearly understood and
appears unimportant to the plant breeder. A combination of heat shock and oxygen-

29
free hydration was found to be most protective against the post-irradiation oxygen-
dependent damage (scored as seedling injury and chromosomal aberrations).

1.5.2. Biological factors

1.5.2.1.Nuclear and interphase chromosome volume

Since the cell nucleus is generally considered to be the primary site of radiation
damage, it seems logical to look for the factors influencing radio-sensitivity in the
nuclei of different species. The previous edition of the IAEA Manual on Mutation
Breeding (1977), thoroughly explored the correlation between the nuclear volume of
a species and its radio-sensitivity. Several studies have shown that neither DNA
content, chromosome number, nor chromosome arm number could be responsible for
the differences in radio-sensitivity (Leonard, Jacquet and Lauwerys, 1983; Bakri et
al., 2005) , but there exists a relationship between average interphase nuclear volume
(INV) and cell sensitivity to radiation. It was concluded, that the higher the
chromosome number, the higher the resistance to radiation, resulting from the fact
that other chromosomes or parts of chromosomes might compensate for the mutations,
this is particularly true for polyploid species (Datta, 2014). Chromosomal
organisation, number and position of centromeres and chromosome size are also
associated with radio-sensitivity (large chromosomes are generally more radio-
sensitive than small chromosomes).

1.5.2.2.Cell cycle

The sensitivity of the cell or more precisely of the nucleus is dependant of the length
of the cell cycle, and the stage of the cell division: mitosis or meiosis. A thorough and
very informative review of the effects of radiation on cell division and plant growth
could be found in Lagoda et al., (2012).

It appears that tissues which are growing rapidly, and have a high mitotic index are
much more sensitive to irradiation than sessile or dormant targets, such as seeds. In
addition the more complex mechanisms of meiosis make this process more sensitive
to irradiation (van Harten, 1998; Datta, 2014).

1.5.2.3.Genetics

Differences in mutagen sensitivity among genotypes are common. Amongst the more
recent studies, the investigation of radio-sensitivity of the wild grass Roegneria spp.
seeds, confirmed genotypic effects on: seed germination, seedlings growth, plant
height, and plant survival (Luo et al., 2013). However, differences in radio-sensitivity
among genotypes within a species are usually much less than between species.
Therefore, for plant breeders wishing to induce mutations, the genotype factors can

30
be ignored, and a routine radio-sensitivity test will determine an appropriate dose rate
for mutation induction.

1.6.PRE- AND POST-IRRADIATION TREATMENTS

1.6.1. Pre-treatment

As stated in the previous section, it is important that the plant breeder ascertains
practical conditions for materials in preparation for X- and gamma- irradiation of
seeds to obtain optimal mutagenic efficiency and repeatability.

1.6.2. Adjustment of seed water content

Water content is a crucial factor to take into consideration when initiating radiation
induced mutagenesis. Thus, before the irradiation treatment, seeds should be brought
to standard water content for that specific species as described in relating literature,
using various physical and/or chemical technologies. An easy, less damaging, stable
and reliable method is the use of glycerol/water solutions (Figure 1.5).

1 2 3

A
B

Seeds

Figure 1.5. Humidity (water content) adjustment in seeds. The system for adjusting seed moisture
(water content) uses flow-through chambers. The gas (or air) to be humidified) flows through tube 1
into the glycerol-water solution in jar A. The humidified gas then flows through tube 2 into the
controlled-environment chamber (jar B) and exits into the atmosphere through tube 3. (Modified from
Forney and Brandl, 1992).

1.6.3. Post-irradiation storage

Undesirable post-irradiation storage effects may be avoided by proper adjustment of

seed water content, for example to 12 – 14 percent for small grain cereals such as
wheat, rice, barley, etc., before irradiation and by storing the seeds for no more than
2 to 4 weeks at temperatures not exceeding 18 – 23°C (Table 1.5). However, it is

31
usually possible to store seeds that have been desiccated at less than 14 percent water
content in a freezer (-20 or -80°C) for several months. A post-irradiation storage
period in a vacuum for about one to two weeks at room temperature will extend
storage life. Seeds of more than 14 percent water content may not be shipped
successfully, but may be shipped frozen.

TABLE 1.5. WATER CONTENT (%) OF BARLEY SEEDS STORED OVER DIFFERENT
CONCENTRATIONS OF GLYCEROL AT 22°C

Storage
Glycerol in Calculated vapour pressure Relative Seed water
period
solution (vol. %) of solution (mm Hg) humidity (%) content (%)
(days)
100 0.0 0.0 8 7
95 3.5 17.6 9 7
85 8.3 41.7 10 4
75 11.4 57.5 11 4
70 12.6 63.5 12 4
65 13.6 68.6 13 4
60 14.5 73.0 14 4

1.7.PROTOCOL FOR RADIO-SENSITIVITY TESTING

This protocol is intended for scientists who want to determine effective doses of
gamma irradiation to generate induced mutants in seed propagated crops. The
information given is based on protocols established at the Plant Breeding and Genetics
Laboratory of the Joint FAO/IAEA Agriculture and Biotechnologies Laboratories,
Seibersdorf, Austria.

1.7.1. List of equipment, supplies and facilities needed

x Seeds
x Water permeable bags
x Paper clips or stapler
x Desiccator
x 60 percent glycerol: distilled water mixture (v/v)
x Gamma cell machine available on-site or easily contacted

32
 x Ruler (starting with zero) or measuring board (millimeter graph paper on a
piece of plywood, best sealed inside transparent plastic film)
x Greenhouse and/or laboratory facility
x Trays and heat /steam sterilized potting mix or germination containers with
filter paper, bleach, sterile water and Tween 20
x Pots and potting mix, if cultivation to maturity of the plants is planned

1.7.1.1.Seeds

The number of seeds will depend mainly on the level of information available on the
radio-sensitivity of the crop genotype of interest.

Seeds should be:

x genetically uniform;
x representative of the genotype;
x dry;
x quiescent1;
x with a high germination rate;
x and have a water permeable seed coat2.
1
If seeds are dormant, any procedures required for breaking dormancy should precede the radiation
treatment.
2
If the seed coat is not water permeable it must be removed or chemically/mechanically modified
(scarified). For this purpose, seed coats can be rubbed with sandpaper, nicked with a knife or filed with
a metal file.

1.7.1.2.Water permeable bag

Bags can be made of mesh, and can be made from simple materials available at local
markets. The number and size of the bags will depend on the amount of seeds. The
bags must have a size that fits into the irradiation compartment of the gamma cell (see
Figure 1.3).

1.7.1.3.Desiccator (vacuum tight) with 60 percent glycerol: water mixture (v/v)

The desiccator should be large enough to hold 1000 ml of liquid in the base. To
prepare a 1000 ml of the mixture use 600 ml glycerol and 400 ml distilled water.
Inside the desiccator the relative humidity should be about 73 percent which can be

33
monitored with a hygrometer. In a desiccator with 1000 ml glycerol/water mixture,
up to 500 g of small seeds can be treated.

1.7.1.4.Trays

Trays with a size of 400 × 600 × 120mm can accommodate 4 – 7 rows of seeds of
cereal and bean species (Figure 1.6). Alternatively, compartmentalized seed trays
(commercially available) can be used. Trays need to have holes in their base for
drainage.

1.7.1.5.Germination containers

Choose a germination container (Petri dish, clear box, etc.) which is large enough to
contain all the seeds without over-crowding.

1.7.2. Radio-sensitivity testing procedure

Keep detailed records of material, treatment dates, radiation source, dose, dose rate,
the treatment conditions, growing season, seed storage and growing conditions, in
order to be able to eventually repeat successfully the experiment.

1.7.2.1.Project planning

1. Check for recommended doses in the literature or refer to Shu et al., (2012)

2. If reliable data are available then move on to point 4 of this protocol. If no

information is known for your target plant, run a preliminary test.

3. For a preliminary test use 20 seeds per dose, one replication only. Use several
doses at a wide range at even intervals e.g. 0, 150, 300, 450, 600 Gy; or 0,
200, 400, 600, 800 Gy. Most plants will fall into the range between 100 Gy
to 700 Gy.

4. For the radio-sensitivity test use 20 – 25 seeds, with 3 replications (a total of

60 – 75 seeds per dose). Increase the number of seeds per treatment if the
germination rate in the control is low to compensate for the probable loss.
Use appropriate doses cited in literature on the crop of interest as a point of
reference and extend treatments 25 to 50 percent higher and lower at a narrow
dose range (intervals of e.g. 50 or 100 Gy). For example: if the literature
consulted states for 50 percent seedling height reduction at a range of
250 – 400 Gy, you may use 100, 200, 300, 400, 500, 600 Gy plus a control
(0 Gy).

34
5. Always include a same-sized control (non-radiated) population to gauge
treatment effects and to assess phenotypic variability of the parent stock. The
control is treated like the material to be mutagenized except for the radiation
exposure.

6. Identify a gamma facility that offers radiation service (see Section 1.9) and
inquire about their specific requirements and procedures. Discuss with
specialists your required doses and the amount of the material to be treated
to ensure the facility is suited for your needs. Check if a phytosanitary
certificate and quarantine procedures are needed for shipment.

1.7.2.2. Pre-conditioning of seeds and gamma radiation treatment

1. Discard any injured, atypical or diseased seed.

2. Count out the number of seeds required per dose and pack them loosely
in water-permeable bags.

3. Label the bags with information on species, variety/genotype, date and the
treatment dose.

4. Fold the tops of the bags over and close with a paper clip or staples to
avoid spillage.

5. Place the bags on the plate of the desiccator above a 60% glycerol-water
mixture at room temperature for a minimum of seven days for small grain
cereals such as rice and wheat. This is extended for up to 14 days for
greater quantities, larger size seeds (e.g. beans) and/or seed with thick
seed coats. The bags should not come in contact with the liquid.

6. Remove the seeds from the desiccator just before radiation treatment. If
the moisture equilibration cannot be followed immediately by radiation
treatment or seeds need to be shipped to the irradiation facility, pack the
seeds in air-tight containers or sealed plastic bags to maintain the desired
moisture content of 12 – 14%.

7. Hand moisture equilibrated seeds in bags or containers over to the

operator of the gamma cell for irradiation. The gamma source must be
operated by experienced and trained personnel who will calculate the
required exposure time based on the present dose rate.
8. Receive treated seeds which can be handled without precautions as
gamma irradiation does not leave any radioactivity behind.

35
 9. Record information on the present dose rate, the type of gamma source
and the exposure time.

1.7.2.3. Post irradiation storage

Plant seeds as soon as possible after irradiation to avoid an increase of damage arising
from prolonged post-irradiation storage. If seeds are shipped or not planted
immediately, store them at room temperature only for up to four weeks. If longer
storage is required, store them dry in sealed airtight bags or vials, in the dark at a
temperature range of –5oC to 2oC to minimize metabolic activity. Most of all choose
storage conditions that suit the seed material.

1.7.2.4. Planting

Choose one of the two planting techniques bellow according to the requirements of
the crop or your preference. Field experiments are not recommended at this stage,
mainly because they are more difficult to control and interpret, due to additional
biotic and abiotic stresses and other varying environmental conditions.

x Flat method: Ideal for beans, cereals and larger sized seeds. This planting
technique warrants organized measurement conditions. The effect of the
dose is easily ascertained by visual assessment of the sowing trays. (Figure
1.6) Continue under point 1 in Flat method below of the protocol.
1. Sow seeds in trays containing potting mix in the greenhouse. Keep
the sowing depth uniform.
2. Plant seeds in rows keeping in mind the requirements of the species.
3. The seeds are planted in order of increasing dose with replications
sown in different trays.
4. Keep the moisture adequate to ensure germination and all
environmental conditions uniform for all treatments. Move on to data
collection (1.7.2.5).
x Petri dish method: Ideal for cereals or smaller sized seeds and for seeds that
require light for germination. Fungal contamination can become a problem.
Continue under point 1 below in Petri dish method of the protocol.
1. Disinfect seeds e.g. by placing them in 10 percent bleach for 10
minutes. Prepare 200 ml of bleach by using 20 ml of sodium
hypochlorite (NaOCl - approx. 5 percent w/v active ingredient), a few

36
 drops of Tween 20 and 180 ml of distilled water. Rinse the seeds
three times with sterilized water to remove disinfectant solution.
2. Wet filter paper in a germination container with sterilized water and
place seeds on filter paper in a grid using forceps. Use a separate
container for each replication.
3. Wrap the container in foil if darkness is required for germination.
4. Tightly seal containers inside a plastic bag to prevent moisture loss.
5. Remove any foil and lift off the lid of the container when seedlings
reach the top.
6. Ensure the paper remains always moist as germination is strongly
influenced by water uptake.

1.7.2.5. Data collection

Collect data on the germination percentage, seedling height, survival, fertility,

sterility in M1 and M2 populations. For optimal results all these measurements
should be performed on the same day for all treatments and replications to reduce
bias.

Germination percentage

Count the number of seedlings when germination is completed. Be aware that

delayed germination may be observed in irradiated seeds as compared to the control.
Count all germinated seeds, also the ones that have died. When using the flat
method, seedling emergence rather than germination is taken. Calculate Germination
(seedling emergence) percentage for each replication (Figure 1.7).

Number of germinated seeds × 100

Germination percentage =
Number of seeds planted

Seedling height variation

Determine the time at which the first true leaves in the control plants cease to expand.
When the first true leaves in control plants have stopped growing, determine seedling
height for the control and for the radiated material. It may be useful to take two full
sets of measurements a few days apart to determine afterwards what the optimal day
was. Use a ruler or measuring board and collect data to the nearest millimeter.

37
In monocotyledons, measures are taken from soil level to the tip of the first or
secondary leaf. In the Petri dish method start measuring from the root-shoot-junction
instead of the soil level. In cereals, take measurements when shoots of control
seedlings are between 11 and 20cm, usually after 10 to 14 days.

Make sure to identify the first leaf correctly and not the cotyledons, which may
resemble leaves in some plant species. In cereals, the first true leaf emerges through
the coleoptile (leaf sheath). Do not measure seedlings where only the coleoptile has
developed.

In dicotyledons, distinguish between epigeous and hypogeous germination. In

epigeous germination the cotyledons are pulled upward through the soil during
germination, e.g. common bean; Phaseolus vulgaris. In hypogeous germination the
cotyledons remain below the soil surface, e.g. pea; Pisum sativum.

Measure the seedling height from soil level to the tip of the primary leaf or to the stem
apex manually by straightening out the plant, or in the petri dish from the root-shoot
junction. Include in the measurements and calculations only the plants with a fully
developed primary leaf. With dicotyledons and epigeous germination, alternatively
you can measure the epicotyl between the points of attachment of the cotyledons to
the tip of the primary leaves or to the stem apex. The parameter would then be referred
to as epicotyl height rather than seedling height. However, make sure you measure all
plants with the same method and keep a record of it, see example in Table 1.6.
Calculate average seedling height (epicotyl height) for each replication.

Sum of seedling height in mm

Average seedling height =
Number of plants measured

38
Figure 1.6. Radiosensitivity test in rice (Oryza sativa) using 60Co gamma-rays at doses: 0,150, 200,
300, 400, and 500 Gy. Note the seedling height decrease, the poor germination rate and/or lethality at
the higher doses: 400 and 500 Gy. Courtesy of A. Kodym.

110

100

90

80
Emergence
70
% of control

60 Seedling height

50

40

30

20

10

0
0 100 200 300 400 500 600
Dose (Gy)

Figure 1.7. Schematic presentation of emergence and seedling height data after gamma radiation
(control=100%).

39
Survival

Count all healthy seedlings. Record the time and stage of plant development at which
data are taken, generally at the time that seedling height is measured. However, it
should be noted that actual plant death might occur at any time between the onset of
germination and maturity. If plants are grown to full maturity it is best to score plant
survival again at the time of harvest, as a plant that successfully produced seeds.
Calculate average plant survival for each replication using the equation below.

Number of surviving seedlings

Survival percentage = × 100
Number of seeds planted

Check leaves for visible leaf spotting or leaf streaks, caused by arrested pigment
development in some cells. It is frequently generated in plants of the Fabaceae family.
Record the degree (number and/or size of the leaf spots), where feasible, as that can
be difficult to quantify.

Decide if you have the time, facilities and the need to carry on with the experiment.
Most data collections could stop here. If you are running a preliminary test, always
stop here and go to data analysis (Section 1.7.2.6). If you decide to carry on, replant
plants into pots while meeting all their specific requirements after the determination
of germination/emergence, seedling height and leaf spotting.

Fertility

Record the number of flowers/spikes/inflorescences, fruits and seeds produced to

assess fertility at the time of maturity of seeds/fruits. These parameters are determined
according to the development of the plants. Calculate the average for each
characteristic per replication: (example here it is given for seeds harvested).

Number of seeds harvested

Average number of seeds per plant =
Total number of plants (including plants
that did not produce seed)

Sterility in the M2 population

Germinate about 50 seeds, according to the availability, of the harvested seeds using
the method described under ‘Planting’ to evaluate sterility in the M2 population.
Calculate the germination percentage in the M2 population:

40
 Number of germinated seeds
Germination percentage in the M2 = × 100
Number of seeds planted

1.7.2.6. Data analysis

Compare the germination percentages of the replications. If values are within a

reasonable range, calculate the average per treatment from the three replications. If
not, look for off-types and try to see where the variation might come from
(e.g. problems with watering) and critically decide what values are representative and
should be included for further analyzing. Express the reduction/increase of the
radiated materials as a percentage of the non-irradiated control in percentages
(Table 1.7).

Average of irradiated treatment

Percent of control = × 100
Average of control treatment

Draw the calculated values as a dose-response curve using the dose as x-axis and the
percentages on the y-axis (Figure 1.7). The control treatment is set as 100 percent.

Repeat these steps with the other parameters (seedling height, survival, fertility,
sterility in the M2). Note that at low dose seedling height of irradiated material may
exceed control values.

Compare the graphs from the different parameters (reduction in germination

percentage, seedling height, survival and possibly fertility and sterility in the M2).
They may show different percent reductions at the same dose, but an overall trend can
be derived by combining the data.

In breeding programs, fertility and sterility are very important criteria as they will
determine the size of the M2 population, which will be available for evaluation and
screening of the mutations. Seeds may form after radiation treatment but fail to
germinate properly or die after germination.

Keep in mind when planning later field trials that emergence and survival obtained
under glasshouse testing conditions may differ considerably from that under field
conditions due to environmental stresses.

41
 TABLE 1.6. SEEDLING HEIGHT MEASUREMENTS IN MM AND SURVIVAL DATA OF RICE VARIETY XY, RAW DATA: 25 SEEDS
PER REPLICATION AND SEVEN DIFFERENT DOSES
No. Ctrl Ctrl Ctrl 100Gy 100Gy 100Gy 200Gy 200Gy 200Gy 300Gy 300Gy 300Gy 400Gy 400Gy 400Gy 500Gy 500Gy 500Gy 600Gy 600Gy 600Gy
I II III I II III I II III I II III I II III I II III I II III
1 195 230 210 225 210 190 225 180 175 170 150 335 160 90 120 30 60
2 175 210 195 195 200 210 225 145 200 150 155 170 95 125 110 40
3 130 245 200 185 230 200 215 170 200 155 160 150 40 115 70 40
4 165 240 235 210 240 220 210 165 205 185 180 45 130 140 70
5 200 230 230 125 235 220 190 165 210 100 140 190 100 145 2
6 115 210 200 190 220 190 175 165 115 125 140 205 145 50 70
7 225 205 225 215 165 220 220 195 190 175 140 165 120 95 85
8 200 215 200 235 190 225 210 175 180 165 155 185 105 35 115
9 190 210 200 215 230 220 240 190 195 190 150 180 100 95 110
10 240 215 215 200 90 225 240 190 165 155 145 190 70 90 80
11 195 205 200 190 210 215 220 180 150 155 175 165 95 130 95

42
12 170 225 225 220 220 100 240 195 175 95 180 145 80 30 115
13 200 220 160 240 175 180 240 165 200 195 140 140 100 90 120
14 200 190 185 220 225 230 220 210 155 165 170 165 150 75 60
15 215 230 210 220 230 200 260 175 185 195 185 185 90 45 130
16 120 220 215 215 230 210 165 185 200 160 180 195 110 90 115
17 170 200 225 225 195 215 190 235 190 185 170 140 85 115 2
18 140 165 220 225 200 215 175 215 185 170 165 125 15 130 120
19 120 200 225 215 210 165 195 205 120 140 170 140 130 100 120
20 180 210 185 190 185 200 175 210 170 160 60 150 90 140 5
21 130 225 210 175 160 180 155 205 210 175 160 160 130 140 40
22 195 240 210 165 180 230 165 215 175 165 180 140 115 125
23 220 230 215 195 220 185 185 200 220 160 185 120
24 210 225 205 160 215 160 205 210 180 140 140
25 200 225 210 215 190 155 155
Data in gray cells were included for the evaluation of survival but not considered for seedling height. The primary leaf did not develop, only the coleoptile was measured.
 TABLE 1.7. EMERGENCE PERCENTAGE (RAW DATA NOT PRESENTED), SURVIVAL PERCENTAGE AND SEEDLING HEIGHT OF
RICE VARIETY XY BASED ON TABLE 1.6 AVERAGE FROM THREE REPLICATIONS (I; II; III) AT SEVEN DIFFERENT DOSES AND
EXPRESSED AS A PERCENTAGE OF RADIATED MATERIAL OVER THE NON-IRRADIATED CONTROL (=100%)

Replication/Dose Control 100 Gy 200 Gy 300 Gy 400 Gy 500 Gy 600 Gy

Emergence I 100 100 100 96 88 12 0
Percentage II 100 96 96 100 84 4 0
III 100 96 100 100 88 12 0
Average 100 97.3 98.7 98.7 86.7 9.3 0
% of control 100 97 99 99 87 9 0

Survival I 100 100 96 96 88 0 0

Percentage II 100 96 96 100 84 4 0

43
III 100 92 100 100 88 12 0
Average 100 96 97.3 98.7 86.7 5.3 0
% of control 100 96 97 99 87 5 0

Seedling height I 180 203 204 161 103 0 0

mm II 217 203 189 157 98 30 0
III 208 202 183 163 98 47 0
Average 201.7 202.7 192 160.3 99.7 25.7 0
% of control 100 100 95 79 49 12 0
1.8. STANDARD PROTOCOLS FOR SEED MUTAGENESIS USING X-RAY
IRRADIATORS

The protocol given below is for small grain cereals, or plants with seeds of similar
size and may be adapted for other types of seeds.

1.8.1. Pre-treatment preparations of seed samples

x Screen and decide on the specific variety/genotype to be treated (preferably

fresh breeder or foundation seeds).
x Clean the fresh seeds from shaft, remove broken, shrivelled or under-size
ones, and if needed, disinfect to eliminate any contaminant.
x Test the viability of the seeds, i.e. germination rate on a small sample
(10 – 20 seeds) according to the species specificities (dormancy,
temperature/light and time) and enter, if possible, only seeds with more than
85 percent germination rate should be preferred.
x For inbreeding species, it is preferable to select homozygous material;
doubled-haploids are ideal, when available.
x Seeds should come from a seed increase field experiment with vigilant
growing conditions to prevent outcrossing (minimal heterozygosity) and
mixing (heterogeneity) with other seed sources of the same species.

1.8.2. Radiosensitivity tests

x Assess and equilibrate the seed moisture, if necessary using 60 percent

glycerol solution (Figure 1.5) to 12 – 14 percent, which is considered optimal
for irradiation treatment.
x Pack 15 – 20 seeds in an envelope for each specific dose, the dose range
should be between 6 – 10 depending on the literature information for the
target species, the growing media (trays, Petri dishes, pots or fields) and the
available space for testing.
x Carefully label each sample with the variety/genotype name, the replication
number, the dose amount, the source of irradiation, and the date. Note that
the amount of seeds per treatment and the number of treatments vary with
the crop, i.e. type of seeds, growing conditions, etc. At the PBGL, planning
for 6 – 7 treatments including the control (untreated) are standard procedure.

44
 1.8.3. Irradiation treatment of the seeds

Below is the description of treatment procedures for two types of irradiators available
for mutation induction (vertical and horizontal rotating machines); the RS2400 Bio-
Rad X-ray machine and the Faxitron and Hitachi Irradiators (Figure 1.8). For any of
these machines, the treatments may be replicated 2 – 3 times depending on the seed
availability and the planting space. As for any manipulation of hazardous tools, e.g.
irradiators, the steps given in this section should be carried out by a qualified irradiator
operator.

1.8.3.1.Vertical rotating RS2400 X-rays irradiator

x Place the seed samples into the appropriate containers, and depending on the
size and the amount of seeds, fill the remaining space firmly with instant rice,
then use appropriate fixing device to ensure stability of the samples during
the rotating motion.

Figure 1.8. Representative X-ray machines; a) vertical rotating Rad Source RS2400 showing the
machine (upper) and the sample chamber with the upper most Canister removed for ease view, b)
Faxitron 650 (Faxitron bioptics LLC, Tuscon, AZ, USA) showing horizontally rotating tray with seed
bag samples placed in the centre, and c) Hitachi Standard MBR-1520R-3 machine showing the sample
chamber at the lower most position with green wheat spikes ready for irradiation. Courtesy of A.
Mukhtar Ali Ghanim.

45
 x Place samples into the canister and fix the containers with the specific
brackets, fill the containers with instant rice again to avoid vacuum condition
during the rotation and firmly close the canisters with the lid.
x Switch on the machine, open the shielding window of the irradiation
chamber, place canister in each one of the 5 canister holders, and close the
shielding window of the irradiator chamber (Figure 1.8. a).
x Set the exposure time, by imputing the amount of kW needed to produce the
required dose and run until the countdown monitor displays: 0 kW. The
machine may then, be turned off.
x Open the shielding window take out and open each canister.
x Gently remove the rice filling and extract the samples and adaptors.
x Mark the envelopes/bags contained the treatment M1 seeds together with the
untreated control samples to be entered in the next steps of the experiment
(laboratory, greenhouse or field) with appropriate information.
1.8.3.2.Horizontal rotating Hitachi and Faxitron X-ray irradiators

In both machines Hitachi (MBR-1520R-3) and Faxitron bioptics (LLC, Tucson, AZ

– USA) the loading and rotation are horizontal and the irradiation source emits
vertically, usually from top to bottom. (Figure 1.8 b, c)
x Turn on the X-ray cabinet (power button) and follow the warm up procedure
as described by the manufacturer.
x Set the energy (display “Tube Voltage”; kVp) by turning the “kVp Control”
button if not using pre-stored settings. Tube current (mA) is usually pre-
adjusted. If using filtration by Aluminium filter (typically 0.5mm), make sure
it is placed into the filter holder below the X-ray tube to cut off the soft X-
ray and leave only hard X-rays.
x Choose the tray position according to the planned dose rate and space needed
to accommodate the samples. These positions vary according to the source
(X-ray tube), the sample distance, and may be adjusted to the dose rate.
x Place the samples onto the sample tray within the marked area for the chosen
position to ensure desired irradiation. Make sure to properly turn on the
rotation of the sample tray located below the tray.
x Adjust the timer setting (Time set; minutes and seconds) to the desired
irradiation duration, which is calculated with the following equation:

46
 Dose (Gy)
o Time (min) =
Dose rate (Gy/min)
x Appropriately close the X-ray cabinet and start the irradiation process. The
lamp “X-ray on” turns automatically on and the elapsing count down time
starts. The machine will automatically turn off when the irradiation process
is completed.
x Mark the envelops/bags with the proper information: date, irradiation dose
and duration, operator name, etc.

1.8.4.Post irradiation treatments and handling

These procedures are identical to those described earlier for gamma-ray irradiation
(section 1.7).

47
1.8.PRACTICAL EXAMPLE OF APPLICATIONS USING PHYSICAL
MUTAGENS

Example on mutation breeding for barley (Hordeum vulgare L.); cultivar UNALM96,
(Gómez-Pando et al., 2009) from the Cereal and Native Grain Research Program,
Universidad Agraria La Molina, Lima, Peru (Figure 1.9).

2
3

7
4

6 5
Figure 1.9. Steps in barley (Hordeum vulgare L.) mutation breeding: 1. 1997 - Preliminary
radiosensitivity test were conducted to determine the LD50 and select the most efficient dose; 2. Same
year: two batches of 250g of M0 barley seeds (cv. UNALM96) were irradiated using a gamma ray
machine at 200 and 300 Gy; 3. 1998 - M1 seeds were sown in the field along with non-irradiated
control seeds and grown to maturity; 4. 1999 - M2 Seeds from individual spikes were planted in 342958
rows and inspected for signs of mutation: off-types i.e. albino, viridis, altered spikes, etc., which were
carefully recorded; 5. 2000 - All M3 seeds were separately harvested and sown as M3 family/row;
6. 2001 to 2005 - From M4 to M8 generations of putative mutant lines selected on the basis of their
agronomical performances were grown at two locations under standard agronomic conditions [in
irrigated fields (Coast) and rain-fed (Highland)]; 7. 2006 - Release of the new barley mutant cultivar
‘Centenario’ from the 12 mutant lines outperforming the parental checks. Courtesy of L. Gomez-
Pando.

48
1.9.SEED IRRADIATION SERVICES AT THE JOINT FAO/IAEA PLANT
BREEDING AND GENETICS LABORATORY

The Joint FAO/IAEA Plant Breeding and Genetics Laboratory, Seibersdorf, Austria
provides an irradiation service for plant mutation induction for Member States. The
requestor must follow the following procedures:

1. Select high quality seeds. The seeds should be disease-free, uniform and
representative of the variety/line/genotype. The seeds should have a high
germination rate (90 percent or more).

2. The samples should be bagged and clearly labelled.

3. The size of the seed sample should be determined prior to sending.

4. If the irradiation dose is not known, about 100 seeds are needed for a radio-
sensitivity test.

5. A Standard Material Transfer Agreement (SMTA) should be signed.

6. Seed samples should be inspected by local Quarantine Officers prior to

sending and sent with a Phytosanitary Certificate.

7. Import Quarantine regulations may also apply.

For a Laboratory Mutagenesis Service Request form and further details contact:

Head, Plant Breeding and Genetics Laboratory

FAO/IAEA Agriculture and Biotechnologies Laboratories
Friedenstrasse 1
A-2444 SEIBERSDORF
AUSTRIA

1.10. OTHER MUTAGENS

In addition to physical mutagens, there are also chemical mutagens (Chapter 2) and
number of biological mutagens that can induce mutations in plants either
experimentally or naturally. A close analysis of the so-called “spontaneous mutations”
frequently observed in nature, show that they result in fact, from intrinsic and/or
extrinsic mutagenic factors affecting the organism. The genetic constitution of the
organism appears as the main factor, comprising variation in chromosomes number
(polyploidy, aneuploidy, etc.) and structural changes in the chromosomes such as
inversions and translocations resulting from chromosomes cross-overs during cell

49
division. It is also known that some genes present in the genome, might induce these
mutations (see also Chapter 3 on transposons).

The physiological conditions, such as age and sex, have been shown to have strong
effects on chromosomes in living cells, which often increase with age. As to the origin
of mutations in ageing seeds various evidence indicate that they are the result of
chemical action of metabolites and /or waste products that accumulate in seeds over
time.

50
 2. CHEMICAL MUTAGENESIS

This chapter reviews commonly used plant chemical mutagens with particular
attention to alkylating agents and sodium azide, the main chemical mutagens in
current use for practical crop improvement or experimental plant mutagenesis. This
chapter also provides guidelines on methods of application and the various parameters
that can influence the outcome of a plant chemical mutagenesis experiment. Basic
information on health and safety considerations to ensure the safe use of chemical
mutagenesis is provided. Chemical mutagenesis in reverse genetics has enjoyed a
renaissance since the early 2000s due to technological innovations including
Targeting Induced Local Lesions in Genomes (TILLING) and, more recently, Next
Generation DNA Sequencing (NGS) technologies. These advances have yielded
important new insights into the mechanism and spectrum of chemically-induced
mutations. Example spectra for key mutagens and crops are included to help guide
plant breeders and/or researchers in designing plant mutagenesis experiments.
Likewise, advances in in vitro plant tissue culture created new opportunities in
expanding chemical mutagenesis to in vitro tissues. This is particularly important for
the vegetatively propagated crops (VPCs) which have lagged behind annual seed
propagated crops for mutation breeding. The chapter includes detailed protocols for
the use of ethyl methanesulphonate (EMS) mutagenesis of banana (Musa acuminata)
in vitro shoot tips and barley (Hordeum vulgare) seeds. Both protocols may be
adapted for various vegetatively and seed propagated crops. In addition, an efficient
protocol for combined mutagenesis treatment of barley seeds using sodium azide (SA)
and N-methyl N-nitrosourea (MNU) is provided as an example of increasing the
mutation spectrum.

2.1.MAIN CHEMICAL MUTAGENS

The use of chemicals as mutagens dates back to the 1940’s with the treatment of
Drosophila melanogaster with mustard gas (Auerbach, 1946; Auerbach and Robson,
1946). To date, the number of chemicals known to have mutagenic effects on living
organisms - animals, plants or microorganisms - is enormous. By contrast, only a few
number of different chemicals are routinely used for experimental plant mutagenesis
or crop mutation breeding. Figure 2.1. summarizes the number of varieties that have
been released following chemical mutagenesis along with the type of mutagen that
was used (data extracted from the IAEA/FAO Mutant Varieties Database;
http://mvd.iaea.org, November 2017). As shown in Figure 2.1a, the top seven most
frequently used chemical mutagens all belong to the class of alkylating agents, except

51
for colchicine. Figure 2.1.b shows that rice, barley, wheat and maize constitute nearly
half of the released mutant varieties.

2.1.1. Alkylating agents

A1kylating agents are known to be mutagenic in plants for many decades (Ehrenberg,
Lundqvist and Ström, 1958). They are by far the most successful from the perspective
of producing new mutant cultivars because of their effectiveness and ease of handling,
and most importantly the convenient detoxification process through simple hydrolysis
for disposal. Alkylating agents are electron deficient (i.e. electrophilic) compounds
with one or more alkyl groups, which can be transferred to biological molecules such
as DNA that contain nucleophilic groups. Most alkylating agents are pro-mutagens,
i.e. they undergo transformation to produce reactive intermediates. These
intermediates can react with DNA by alkylating the phosphate groups in the
phosphodiester backbone as well as the various imino- or carbonyl- groups present on
the purine (adenine, guanine) or pyrimidine (cytosine, thymine) bases.

Figure a Figure b

Figure 2.1a. Chemical mutagens most frequently applied in generating mutant varieties. Among top
agents are EMS (ethyl methanesulphonate), with 106 officially registered mutant varieties, NEU
(nitrosoethyl urea) (57), MNU (N-methyl N-nitrosourea) (53), colchicine (46), and EI (ethylenimine)
(36). Figure 2.1b. Officially released mutant crop varieties registered in the MVD produced via
chemical mutagenesis.

52
Figure 2.2. Molecular structure of alkylating agents commonly used in chemical mutagenesis of plants.

Alkylating agents can be classified as mono-, bi- or poly-functional types, depending

on the number of alkyl groups present in the compound. The alkylating agents most
commonly used in plant mutation breeding are mono-functional. Among these, EMS,
MNU and diepoxybutane are frequently used (Figure 2.2) Bifunctional alkylating
agents can induce inter- and intra-strand DNA-DNA cross-links causing the inhibition
of DNA replication.

2.1.2. Sodium azide

Sodium azide (NaN3; SA) is an inorganic, highly toxic compound and the only other
chemical mutagen besides the alkylating agents that has been used frequently for
practical crop improvement (Figure 2.1). SA is a well-known inhibitor of cellular
respiratory processes in living cells (Tsubaki et al., 1993), and has proven to be an
effective mutagen in many crop species such as barley, rice, soybean and maize but
not in other plants such as Arabidopsis thaliana (Kleinhofs et al., 1975).

53
Sodium azide is also considered a pro-mutagen as it is metabolized in vivo to a
powerful chemical mutagen through an organic intermediate, identified in barley as
L-azido-alanine. It appears that L-azido-alanine itself does not directly interact with
DNA, but that mutagenesis is mediated by the host-plant cellular processes involved
in DNA excision-repair (Owais and Kleinhofs, 1988; Sadiq and Owais, 2000). These
findings may also explain the lack of SA mutagenic effects in some plant species such
as Arabidopsis. Hence, initial experiments are required to assess the effectiveness of
SA in a new plant species prior to conducting large-scale mutagenesis. In plants, SA
affects multiple metabolic pathways explaining its cytotoxic and physiological effects
in addition to its mutagenic effects (Gruszka, Szarejko and Maluszynski, 2012).

The mutagenic effects of SA have been extensively studied in barley (Olsen, Wang
and von Wettstein, 1993; Maluszynski and Szarejko, 2003; Lababidi et al., 2009) as
well as in other crops, such as tomato (Abdulrazaq and Ammar, 2015), the oat species
Avena longiglumis (Khan, Al-Qurainy and Anwar, 2009), and rice where an advanced
mutant line with enhanced amylose content was developed (Suzuki et al., 2008).

The mutagenic effect of SA depends greatly on the acidity of the treatment solution
(Nilan et al., 1973). The mutagen should be applied at low pH (< 4), e.g. for barley
mutagenesis it is being dissolved in a phosphate buffer at pH 3 (see also Section 2.5.
for examples of treatment procedures).

2.1.3. Other chemical mutagens

In addition to the alkylating agents and azides, the previous (second) edition of the
IAEA Manual on Mutation Breeding (IAEA, 1977) described the following groups of
chemical mutagens: (i) base analogues; (ii) antibiotics; (iii) acridines; (iv) nitrous
acid; and, (v) hydroxylamine. Leitão, (2012) described the acridines under a more
general category of intercalating agents together with topoisomerase inhibitors and
poisons. The exploitation of these chemical mutagens for plant genetic improvement
is much more restricted compared to the alkylating agents and azide, because they are
either less effective, less well studied or more challenging to handle from a health or
safety perspective. Hence, selected groups are reviewed below. Colchicine is
considered a chemical mutagen senso latu, because its main effect is on ploidy and
not on genes (see Chapter 3) and is briefly described below. For additional
information on the types, properties, and mutagenic effects of these additional
chemical mutagens, the reader is referred to (Leitão, 2012).

2.1.3.1.Base analogues and related compounds

True base analogues are closely related to the DNA bases: adenine, guanine, cytosine
or thymine and can be incorporated into the DNA molecule without hindering its
replication. However, since these analogues differ subtly, occasional base pairing

54
errors can occur during DNA synthesis or DNA replication. The most frequently used
analogues are 5-bromo-uracil (BU) and 5-bromo-deoxyuridine (BUdR), which are
analogues of thymine and adenine, respectively. BU is able to induce mutations in
higher plants but the mutation frequency remains low (Handro, 2014; Gautam, Saxena
and Kumar, 2016). Overall, base analogues have not been extensively tested as agents
for mutation induction in plants.

2.1.3.2.Antibiotics

Antibiotics are functionally defined as having anti-microbial effects, but structurally

they comprise a very diverse range of compounds. Antibiotics also differ with regards
to their cytotoxic or mutagenic properties, which have been primarily studied in
microbial and animal systems. For example, streptozoocin (STZ) is a potent mutagen
and carcinogenic and is used as an anti-neoplastic agent. STZ causes primarily point
mutations while other antibiotics have chromosome-breaking properties. Mitomycin
C (MMC) is a naturally occurring antibiotic isolated from Streptomyces caespitosus.
MMC is a bi-functional alkylating agent that reacts with guanosine residues to form
cross-links between DNA strands (see also 2.1.1. and Palom et al., 2002).

While antibiotics have had limited use in plant mutation breeding, streptomycin has
been successfully used to induce male sterility in several plant species including rice,
sorghum, pearl millet, sugar beet, and sunflower, the latter resulting in the official
release of several mutants (Hu and Rutger, 1991; Jan and Vick, 2006; Elkonin and
Tsvetova, 2008).

2.1.3.3.Intercalating agents

Intercalating agents can reversibly intercalate with double-stranded DNA, but do not
covalently bind to it. Classical intercalating agents include ethidium bromide, 4', 6-
diamidino-2-phenylindole (DAPI) and acridines which are widely used as dyes in
biological or biochemical studies. Acridines and their derivatives have light absorbing
properties, and display photo-enhanced cytotoxic and mutagenic effects. The
mutagenic effects induced by acridines can range from base-pair substitutions to
frameshift mutations to chromosome-breaks depending on the type of acridine used
and are well demonstrated in prokaryotic and mammalian systems.

However, these compounds have rarely been studied in plants. Recent experiments
include wild ginger species (family Zingiberaceae) with ornamental potential,
(Prabhukumar et al., 2015) and flax (Linum usitatissimum L.), (Bhat et al., 2017).
These authors have compared the mutagenic effects of acridine to other chemical or
physical mutagens such as EMS, colchicine, gamma- or X-rays. The studies clearly
demonstrated the mutagenic effects of acridines on plant growth and development.
For example, treatment of the wild ginger species Larsenianthus careyanus with

55
1 percent acridine produced white variegation on the leaves. The strong mutagenic
capabilities of these compounds as demonstrated in prokaryotic and mammalian cells,
warrant further investigation in plants in view of their potential to induce new and
unique mutations for plant scientists and breeders.

2.1.3.4.Colchicine

Colchicine is a toxic alkaloid derived from the meadow plant: Colchicum autumnale
(Autumn crocus). Colchicine is widely used in plant breeding work to produce
changes in ploidy. The increased number of chromosomes usually brings about
changes in plant morphology and functions. Colchicine treatments of meristem-
containing propagules or tissues can be performed in many ways using concentrations
ranging from 0.005 percent to 1.5 percent (van Harten, 1998). Common methods for
chromosome doubling involve soaking the seeds in a colchicine solution, applying
colchicine using a brush on growing shoot apices, or culturing (in vitro) plantlets in
colchicine-containing medium (Hamill, Smith and Dodd, 1992). A major use for
colchicine is the treatment of haploids to produce doubled haploids which are
completely homozygous, i.e. genetically pure. Practical methods in doubled haploid
production in a wide range of plant species are available in Maluszynski et al., (2003);
see also Chapter 8-B in this manual.

2.2.MODE OF ACTION AND MUTATION SPECTRA

The mutagenic effect of a chemical depends on the lesion initially induced in the DNA
as well as on any DNA repair mechanisms present in the host plant cells. Therefore,
both the characteristics of the chemical and the host DNA repair processes play an
important role for determining the ultimate mutagenicity of a chemical mutagen.
Details on plant host repair mechanisms can be found in Chapters 1, 3 and 4.

2.2.1. Alkylating agents

The DNA breakage and clastogenic effects, i.e. disruption or breakage of

chromosomes, induced by alkylating agents have been documented for over 80 years
(Auerbach and Robson, 1946). To date, alkylating agents are widely applied to induce
single base pair changes to alter protein function or structure.

Alkylation is defined as the transfer of an alkyl group from one molecule to another.
The alkyl group may be transferred as an alkyl carbocation, a free radical, a carbanion
or a carbene (or their equivalents). The dialkyl nitroso amines (e.g. diethyl nitroso
amine) are stable compounds, which apparently act on DNA only after enzymatic
activation (removal of one alkyl group). The major DNA alkylation mechanisms
present in host plants have been reviewed by Leitão, (2012).

56
Lee et al., (2014) reported that EMS induces alkylation on guanine resulting in
GC>AT transitions, which can lead to single nucleotide mutations. Resulting data sets
from Targeting Induced Local Lesions in Genomes experiments (TILLING) in over
15 plant species have shown that EMS primarily causes GC to AT transitions as
expected for alkylation of guanine at the 6O position (Jankowicz-Cieslak and Till,
2015).

Several studies have shown that EMS mutations are distributed randomly across the
genome (Greene et al., 2003; Till et al., 2003). According to these authors, a bulk
mutagenesis experiment producing a population of 3 000 to 6 000 lines is typically
sufficient to recover multiple alleles in any gene in case of diploids. An extensive
characterization of EMS-induced rice mutants demonstrated that EMS-induced
mutagenesis has a strong local sequence context bias specifically targeting guanine
residues in the context RGCG (R is A or G; the mutated guanine indicated in bold),
(Henry et al., 2014b).

While the high mutation densities achieved in these studies enables the recovery of
allelic variants in any gene, this high mutational load may present a challenge for
functional genomics studies, and for practical crop improvement. This is especially
the case when attempting to improve elite materials for one or a few traits as the
presence of the numerous undesirable mutations in the background may disrupt the
finely tuned genetic architecture of the elite variety (see Section 2.3.2).

2.2.2. Sodium azide

Sodium azide induces chromosome aberrations at a very low rate. The type and
number of mutations induced by SA has been studied in barley (Talamè et al., 2008;
Kurowska et al., 2011) and more recently also in rice (Tai et al., 2016). These studies
showed that SA is a powerful mutagen for inducing point mutations. Both in barley
and rice, GC to AT transitions were the predominant mutation type. It appears that
sodium azide-induced mutagenesis has a different local sequence context bias (GGR)
compared to EMS (Tai et al., 2016). Therefore, combining different mutagenic
compounds may expand the spectrum of induced mutations and the resulting mutant
phenotypes.

An overview of the types and density of mutations induced by different chemical

mutagens based on DNA sequence analysis is provided in Table 2.1.

57
TABLE 2.1. SPECTRUM OF CHEMICALLY INDUCED POINT MUTATIONS IN DIFFERENT
SEED-PROPAGATED SPECIES AND BANANA, A VPC

Mutation Transitions Transitions Trans-

Species (common
Mutagen density G/C > A/T A/T > G/C versions Reference
name), ploidy level
(kb) (%) (%) (%)

Arabidopsis thaliana,
EMS 1/200 100 0 0 Greene et al., 2003
2x

Avena sativa (oat), 6x EMS 1/24 94.4 0 5.6 Chawade et al., 2010
Brassica rapa (field 1/56 and
EMS - - - Stephenson et al., 2010
mustard), 2x 1/67
Cucumis melo Dahmani-Mardas et al.,
EMS 1/573 97.8 0 2.2
(melon), 2x 2010
Glycine max EMS
1/74 84.3 - 23 to 47 Tsuda et al., 2015
(soybean), 4x (repeated)

Hordeum vulgare
EMS 1/500 n. a n. a n. a Gottwald et al., 2009
(barley), 2x

Hordeum vulgare
EMS 1/1,000 70 10 20 (Caldwell et al., 2004)
(barley), 2x

Musa acuminata Jankowicz-Cieslak et al.,

EMS 1/57 100 0 0
(banana), 3x 2012
Oryza sativa japonica
EMS 1/147 88 - - Henry et al., 2014
(rice), 2x
Solanum lycopersicum
EMS - - - 55 Minoia et al., 2010
(tomato), 2x
Triticum aestivum 1/23.3 to
EMS 99.2 0 0.8 Dong et al., 2009
(bread wheat), 6x 1/37.5

Triticum durum
EMS 1/51 - - - Uauy et al., 2009
(durum wheat), 4x
Triticum durum
EMS 1/50 - - - Henry et al., 2014
(durum wheat), 4x
Glycine max EMS or 1/140 to
90 - - Cooper et al., 2008
(soybean), 4x MNU 1/550
Oryza sativa japonica EMS and 1/265 to 29.6 to
70.4 to 66.7 0 Till et al., 2007
(rice), 2x SA-MNU 1/294 33.3
Oryza sativa (rice), 2x MNU 1/135 92 - - Suzuki et al., 2008a
Hordeum vulgare
MNU 1/504 23 33 37 Kurowska et al., 2011
(barley), 2x
Hordeum vulgare Olsen, Wang and von
SA - 86 14 -
(barley), 2x Wettstein, 1993
Hordeum vulgare
SA 1/374 95.5 - 4.5 Talamè et al., 2008
(barley), 2x
Hordeum vulgare
SA-MNU 1/477 88 4.5 7.5 Szarejko et al., 2017
(barley), 2x

58
2.3.GUIDELINES FOR CHEMICAL MUTAGENESIS

This section provides general guidelines for chemical mutagenesis of both seeds and
vegetative propagules, including in vitro cell cultures. Many factors can influence the
outcome of chemical mutagenesis including the characteristics of the target plant
material, the dose of the chemical applied, the physico-chemical properties of the
chemical mutagen, the nature of the mutagenic solution (e.g. pH), the environmental
conditions of the laboratory (e.g. temperature) as well as the growing conditions
(greenhouse, nursery, field, in vitro, etc.) of the plant seeds and/or propagules before
and after the mutagenic treatment.

2.3.1. Target plant materials

The choice of the most suitable material depends on the objectives of the mutagenic
treatment and the plant species. Different types of plant propagules are described
further below, including seeds, so called ‘in vivo’ vegetative propagules and in vitro
explants or tissues. The genetic constitution of the target material such as
heterozygosity or ploidy level is also a critical consideration for applied mutagenesis
studies. This aspect is covered in Chapters 1, 2, 3, and 4 of this manual.

Seeds are the most commonly used target tissue for chemical mutagenesis. For
example, seeds of cereals or legumes can be easily stored and shipped and treated in
large quantities. Over the past few decades, standardized protocols for EMS
mutagenesis of many seed crops have been established and routinely used in various
laboratories.

Soaking the seeds in the mutagen solution is the most convenient and most widely
used method, thus small grain cereals and other seeds that imbibe rapidly are easy
targets. Plant species and varieties may respond differently to a certain chemical
mutagenic treatment. Likewise, the actual experimental conditions may vary from
laboratory to laboratory. Therefore, it is highly recommended to always perform dose-
response experiments for seed mutagenesis of new species or varieties prior to
conducting a large-scale mutagenesis experiment. Two examples of protocols for
EMS mutagenesis and one protocol for combined MNU and sodium azide (SA, NaN3)
mutagenesis of barley seeds are described below (see examples of treatment
procedures section 2.5.3).

It should be noted that one can also treat in vivo vegetative propagules and explants
such as tubers, bulbs or corms, ramets, cuttings or scion woods, rooted cuttings or
growing plants, bud woods, or stolons (see Chapter 6). The treatment procedures for
such vegetative propagules, however, are less well established compared to seed
mutagenesis protocols mainly because of technical challenges regarding the uptake

59
and penetration of the chemical in the plant tissue resulting in an uneven distribution
of the chemical mutagen within the target meristem. Consequently such in vivo
treatments may lack reproducibility. When the target material is small such as small
cuttings or shoot tips, the challenges faced may be fewer.

In vitro explants are currently becoming a useful target for chemical mutagenesis.
In vitro systems may offer several advantages such as the availability of more
standardized conditions and the possibility to prevent or restrict the formation of
chimeras (see Chapter 8). For example, successful EMS mutagenesis of banana shoot
tip explants (Jankowicz-Cieslak et al., 2012), and callus tissue of rice (Serrat et al.,
2014), wheat (Simonson, Baenziger et al., 1991) and sugarcane (Purnamaningsih and
Hutami, 2016) have been reported. Callus tissue was induced from bahiagrass seeds
treated with sodium azide (Kannan et al., 2015). A large mutant population of 19 630
plants was regenerated from these calli via somatic embryogenesis and a superior
mutant line with improved traits was later identified in multi-locational field trials.
Collectively, these studies demonstrate that EMS mutagenesis can be like in the case
of seeds applied to in vitro explants for the recovery of superior mutant crops with
improved traits. A protocol for EMS mutagenesis of in vitro banana shoot tips is
described in section 2.5.

To our knowledge, the frequency and types of mutations induced using in vitro cell
cultures has only been determined in a few cases, e.g. banana (Jankowicz-Cieslak
et al., 2012) and rice. The molecular mutation spectra in populations derived from in
vitro explants in these studies were consistent with results obtained from EMS
mutagenized seeds.

A wide range of plants can be regenerated from single cells via in vitro tissue culture
(see Chapter 8-A). This provides an excellent opportunity for combining tissue culture
protocols with mutagenesis techniques. As is the case with radiation treatment of
VPCs, appropriate tissue culture techniques will greatly facilitate the regeneration of
a whole homohistont plant from a single cell to avoid the development of chimeras.

The ideal plant material for chemical mutagenesis would be haploid cells, especially
those that can be manipulated to produce doubled haploids (this is discussed further
in Chapter 8-B). Chemical mutagenesis of pollen has been extensively described for
maize (Neuffer, 1994). In fact, this has become the method of choice for chemically
induced mutations in maize for several decades, primarily because it avoids the
creation of chimeric plants that may or may not transmit induced mutations to
progeny.

60
 2.3.2. Dose, dose determination and mutational load

The two most commonly used experimental variables to describe dose in the context
of chemical mutagenesis are the concentration of the chemical in the mutagenic
solution and the duration of the treatment (dose = concentration × duration).

In practice, when the optimal dose for a specific crop (or cultivar), target tissue, or
chemical mutagen cannot be found in the literature, a dose-response curve needs to
be established. This is equivalent to conducting a radio-sensitivity experiment for
physical mutagenesis. The dose-response curve, also called ‘kill curve’ or ‘chemical
toxicity test’, establishes the relationship between the survival rate or growth
reduction of the propagules after treatment with increasing concentrations of the
mutagenic chemical(s) during specific periods of time.

In the case of seeds, a seedling test is conducted whereby seed germination and
seedling survival or growth can be measured after the mutagenic treatment. For
vegetative propagules such as in vivo cuttings or in vitro cell cultures, similar methods
to measure growth or survival of the propagules can be followed. Example of dose-
response curves for in vitro banana shoot tips and barley seeds are described in
Section 2.5.

It should be noted that the dose-response curve for chemical mutagenesis may
considerably differ from the one from radio-sensitivity tests, this is due to the
specificity of chemical toxicity on cells. According to van Harten, (1998), a 20 – 30
percent growth reduction (which may correspond to a survival rate of 70 – 80 percent)
may produce an optimal mutation yield in cereal crops. The duration of the treatment
is also relevant, it should enable proper uptake of the chemical mutagen by the plant
tissues. In case of seeds, the duration may be shortened when using pre-soaked seeds
(see also section 2.3.5.). Typically, in a dose-response experiment, the seeds or
explants will be exposed to increasing concentrations of the mutagen over different
durations.

The volume of the treatment solution may also play a role: the volume should be
sufficiently large to provide each seed (or propagule) the opportunity to absorb the
same amount of mutagen. For example, 0.5 – 1ml is recommended per seed in the
case of small grain cereals. To ensure a uniform concentration of the mutagen
throughout the treatment, the solution should be gently shaken.

The temperature of the mutagenic solution greatly influences the treatment, mainly
because of the effect of temperature on the reactivity of the chemical
(see section 2.3.4.). Some authors recommend short time treatments of 0.5 – 2h at
temperatures of about 20 – 25°C to seeds that have been pre-soaked for different times
at room temperature (= pulse treatment). These conditions facilitate the absorption of

61
the mutagen, increase the metabolic activity of the seed and enhance the reaction
between the chemical and the genetic target. The optimal dose ultimately depends
upon the desired goal of the mutagenesis experiment or breeding programme. As
illustrated in Table 2.1, mutant populations with a high density of mutations have now
been produced for reverse genetic studies of the major cereals and legumes.

In fact, in reverse genetics one identifies an altered sequence and then proceeds to
determine its effect, if any, on the phenotype. To increase the efficiency of this
process, it becomes imperative to induce a high mutational load per line to reduce the
size of the mutant population needed. For example, in the case of wheat, single mutant
lines can carry, on average, several hundred thousand mutations.

This high mutational load can have significant implications for practical plant
breeding because these background mutations may disrupt the finely tuned genetic
constitution of the elite parental line. From a practical perspective, plant breeders may
therefore, consider applying lower doses and increase the size of the mutant
population similarly to the mutant population development methods described when
using physical mutagenesis (see Chapters 1, 4 and 6). When attempting to modify just
one or two characteristics, doses inducing less than 30 percent growth reduction have
been recommended for plant breeding projects (Maluszynski et al., 2009).

Different chemical mutagens can be combined to broaden the mutation spectrum. An

example of a combined treatment of SA with MNU for barley seeds is given in section
2.5.3. Similarly, chemical mutagenesis can be combined with physical mutagenesis
treatments to broaden the spectrum of mutations.

2.3.3. Status of plant materials

It is usually preferable for seeds as well as for vegetative propagules to apply the
treatments while they are in an actively growing stage. Various methods exist to
stimulate or enhance the efficiency of chemical mutagenesis in case soaking the seeds
or plant propagules is not feasible or effective, as illustrated below.

a. Seeds of bahiagrass were scarified, surface sterilized and treated with SA.
Thereafter, callus tissue was induced in vitro and plants were regenerated via
somatic embryogenesis to produce M2 mutant progeny (Kannan et al., 2015).

b. Treatment of an inflorescence (or bud) can be performed by covering these

plant parts with a wad of cotton wool soaked in the chemical (van Harten,
1998).

c. A mutagen may be applied in low concentrations to the growth medium and

allowed to enter the plant through the roots. This simple method offers

62
 advantages when studying chronic mutagen exposures, or for determining the
sensitivity of different stages of growth and development to the chemical
mutagen.

d. A frequently used protocol for EMS mutagenesis of pollen in maize involves

the use of paraffin oil to prepare an emulsion of the mutagen and the pollen
thereby avoiding lysis of the pollen in the aqueous solutions (Weil and
Monde, 2009).

Plant materials less suited to chemical mutagenesis include those that do not readily
imbibe the chemical solution; this includes woody tissues and seeds with thick shells
(e.g. nuts) and plant parts that are dormant. Still, various pre-treatments can be utilized
to break dormancy or to increase cell permeability and adsorption of the chemical
mutagen such as scarification or similar methods.

2.3.4. Physico-chemical properties of chemical mutagens and mutagen

solutions

Properties of mutagens that limit their effectiveness are their (a) solubility,
(b) toxicity, and (c) chemical reactivity. The useful range of concentrations is
restricted by the solubility of the mutagen in the treatment solution as well as by its
toxic effects on the plant propagule.

Mutagens vary widely in their toxicity. In general, the methylating agents, e.g. MMS,
are more toxic than their corresponding ethylating agents, e.g. EMS. Even though
methylating agents are more mutagenic than ethylating agents, the efficiency of e.g.
MMS is lower than EMS because of the higher toxicity of the MMS resulting in a
higher level of damage to the plant propagule, which in turn results in a lower survival
rate after the mutagenic treatment.

The alkylating agents are very reactive agents and will hydrolyse in a water solution.
This implies that solutions must be prepared fresh and never stored. The reaction with
water usually gives rise to compounds that are no longer mutagenic but can still be
harmful or toxic for the operator. For EMS, the hydrolysis reaction is as follows:

H3SO2OC2H5 + H2O CH3SO2OH + C2H5OH

EMS Methane Ethanol
sulphonic acid

The rate of hydrolysis of a chemical mutagen is usually measured by its half-life. For
a given compound, the half-life is a function of the temperature and sometimes of the
pH. For example, in case of the alkylating agents, the hydrolysis rate decreases with
decreasing temperature thus, a mutagen will be stable for a longer time at lower

63
temperature, ensuring its reactivity with the nucleophilic centres in the target. The pH
is especially important for ethyleneimine derivatives, sulphur and nitrogen mustards
and some nitroso-compounds, which should always be dissolved in buffers of well-
defined pH, usually below 7.

In terms of chemical reactivity, the alkyl alkane sulphonates and alkyl sulphates
produce strong acidic products upon hydrolysis in the mutagenic solution, as well as,
inside the cell. Therefore, significant physiological damage may occur in unbuffered
solutions. This can lower the mutagenic efficiency through reduced M1 plant survival.
The negative effects of hydrolysis can be greatly reduced by properly balancing the
solutions with buffers. The pH of the solution should hence be monitored before and
after the treatment.

DMSO (dimethyl sulphoxide) is known to increase cell permeability and enhance

absorption through biological membranes and thus, it has been tested as a carrier in
chemical mutagenesis. At the PBGL in the Joint FAO/IAEA Agriculture and
Biotechnologies Laboratories in Seibersdorf, Austria, the mutagenesis is conducted in
a 2 percent DMSO solution to ensure solubilization of the EMS. Amin, Laskar and
Khan, (2015) showed that the action of the chemical mutagen MMS alone and in
combination with DMSO induces physiological, biochemical, metabolic, and genetic
disturbances which results in significant bio-morphology and quantitative variations
in lentil (Lens culinaris) but they also demonstrated enhancing effects of DMSO on
MMS mutagenicity.

2.3.5. Pre- and post-treatment procedures

In general pre-soaking the seeds prior to mutagen treatment increases the efficiency
of mutation induction by activating metabolic processes and DNA synthesis in the
cells (IAEA, 1977). Thus, pre-soaking seeds or buds triggers the transition from a
dormant status to an actively metabolizing and synthesizing stage. Pre-soaking may
also speed up the uptake of the mutagen by increasing the cell membrane
permeability. Several important changes take place in seeds when they are soaked.
These depend to some extent on the conditions of soaking (duration, temperature,
soaking solution) and on the type of seed or plant propagule. The duration of pre-
soaking of seeds in water can be estimated experimentally. Seeds should stay in the
solution, as long as, they actively absorb it. To optimize the duration of pre-soaking a
pilot experiment can be carried out whereby the incubated seeds are weighed every
hour to determine when their increasing weight reaches a plateau. The pre-soaking
should not be shorter than the time estimated experimentally.

Post-treatment handling procedures up to initiation of growth can also affect the

mutagenic efficiency of a treatment. Important factors are the duration and
temperature of storage of the treated seeds. Storing of mutagen-treated (M1) seeds

64
mostly enhances injury. However, post-washed and rapidly re-dried seeds can be
stored at 0°C to 4°C for long periods without seriously altering the mutagenic effects,
since the post-washing rapidly removes both non-reacted chemicals and their
hydrolytic by-products from the seeds.

Various methods of post-treatment washing and/or drying have been used. Treated
and post-washed seeds may be simply air-dried by laying them out on a blotting paper.
Drying time may be reduced using an electric fan blowing over the seeds or under a
ventilated fume hood. Elevated temperature drying may also be convenient, but in this
case the temperature used should not exceed 35°C and uncontrolled heating should
not be used. Post-treatment drying is especially desirable for convenient handling and
shipping of mutagen-treated M1 seeds. With most alkylating agents increased damage
may occur on re-drying and storage of seeds. Several factors seem responsible for the
phenomena observed including: (1) the hydrolysis rate of the mutagenic agent; (2) the
enzymatic actions in the biological system; (3) and the uptake of hydrolysis by-
products by the treated seeds or explants. Scientists and breeders must therefore
consider the specific requirements of the specific crop and carefully plan any further
laboratory, greenhouse or field activities before initiating a mutation breeding
programme using chemical mutagens.

2.3.6. Advantages and limitations of chemical mutagenesis

The advantages and limitations of chemical mutagenesis for experimental plant

mutagenesis or plant breeding have been previously summarised (van Harten, 1998).
Taking into consideration the more recent findings pertaining to plant chemical
mutagenesis, these can be updated as stated below.

Advantages

x Well characterized mutation spectrum producing mainly point mutations.

x Less chromosomal damage when compared to physical mutagens.
x High mutation frequency allows to create allelic variation at any target gene.
x Mutations appear to be evenly spread across the entire genome.
x Standardized protocols for seed treatment of the major seed propagated food
crops.
x Can be equally applied to in vitro tissues or explants.
x EMS mutagenesis can be applied in a standard laboratory setting.

65
Limitations

x Dense mutation rate, this may require several rounds of backcrossing to

remove undesirable mutations.
x Penetration in multi-cellular or woody plant tissues is often difficult or has
low reproducibility.
x Materials or seeds that are dormant or have long germination times, e.g. nuts,
may require special pre-treatments or manipulations.
x Limited repertoire of well-characterized chemical mutagens for plant
mutagenesis.
x May not be effective to induce large chromosomal variations that are
heritable.
x Health and safety concerns due to toxic or carcinogenic properties.

2.4.STORAGE, HANDLING AND DECONTAMINATION PROCEDURES OF

CHEMICAL MUTAGENS

Most chemical mutagens are potential carcinogens and therefore, appropriate health
and safety issues need to be fully understood and complied with. The purpose of this
section is to provide basic information relative to the storage, handling and clean-up
of the commonly used chemical mutagens described in this chapter, i.e. EMS, MNU
and SA.

It is highly recommended that chemical mutagenesis is carried out by trained

personnel in specialised facilities.

One should always practise the following laboratory safety procedures.

x Wear the correct personal protective equipment such as gloves, safety

goggles, and long-sleeved lab coats.
x Perform chemical mutagenesis under a functional fume hood to ensure
disposal of chemical vapours.
x Store the chemicals in a designated area with the appropriate hazards sign
and ventilation if required.
x Consult the Material Safety Data sheet (MSDS) which is an important
component of occupational safety and health and product stewardship.

66
Additional information on the safety, biological activity and properties of other
chemical mutagens can be retrieved from the Pubchem database
(https://pubchem.ncbi.nlm.nih.gov/).

2.4.1. Alkyl alkanesulphonates and alkyl sulphates

Common example

x Ethyl methanesulphonate (EMS)

Physico-chemical properties

x Generally, liquids, highly soluble in organic solvents, slightly soluble in

water; undergo hydrolysis forming strong acid with the rate of hydrolysis
largely dependent on the alkyl group. The reactivity of alkylating agents
varies greatly and this is influenced by the nature of the mutagen solution or
reaction medium.

Storage

x Store in a small, airtight bottle in a refrigerator, inside a sealed chamber

containing a desiccant and protected from the light.

Clean-up

x At the Joint FAO/IAEA Plant Breeding and Genetics Laboratory,

decontamination of the working surface, laboratory equipment or glassware
that has come into contact with EMS is done using a freshly prepared 1 M
sodium thiosulphate stock solution (Na2S2O3.10H20) diluted to 100mM.
Special care should be taken when gram amounts or more of EMS or MMS
need to be destroyed as violent reactions can occur with sodium thiosulphate
solutions. In this case, large volumes of aqueous bicarbonate solutions are
recommended.

Health hazards

x If the mutagen is accidently swallowed, induce vomiting. Drink saline or

other alkaline solution. Get a physician to carefully examine the functions of
the liver and kidney. Person with diseases of the central nervous system,
kidney and liver should not work with these compounds.

67
 2.4.2. Nitroso compounds

Common example

x N-methyl N-nitrosourea (MNU)

Physico-chemical properties

x Generally, exists in solid state, highly soluble in organic solvents, and

reactivity dependent on pH of the solution.

Storage

x Store in small units (50-100 g) in a refrigerator. Avoid exposure to heat,

friction, or impact. Hazardous above room temperature and therefore must
always be kept cool.

Clean-up

x If the powder is spilled, dampen the powder and carefully brush onto a pan,
empty it in a plastic bag. If the liquid is spilled, absorb with paper or
vermiculite and scoop into a plastic bag. Sponge up the spillage with water
followed by decontamination with a 10 percent ceric ammonium nitrate
solution.

Dangerous reactions

x Nitroso compounds can undergo violent thermal decomposition. Above

200°C the vapour is known to explode.
x In the presence of alkali nitroso guanidine evolves diazomethane, which can
explode even at low temperatures if traces of organic matter are present.

Health hazards

x Exposure to nitroso compounds can cause corneal ulcers, asthma, contact

dermatitis, etc. Some of the diagnostic tests include prominent hilar shadows
in chest X-rays and non-specific changes in electrocardiograms.

68
 2.4.3. Azides

Common examples

x Mainly sodium and potassium azide

Physico-chemical properties

x Mainly exists in the form of crystalline salts. The alkali metals salts are
relatively stable, but when in contact with water or acids they are readily
converted to hydrazoic acid (HN3). The acid form is volatile, boiling at 36°C.

Storage

x Store as alkali metal salts in small quantities in glass containers inside a

refrigerator at 4°C. Do not store where accidental spillage or breakage of
containers could cause inadvertent mixing with acids.

Clean-up

x Mop any spillage with excess water and soap or detergent. If spill is under
acidic conditions, also ventilate the area. Wear self-contained protective
breathing apparatus.

Health hazards

x Sodium azide is highly and acutely toxic by all routes of exposure, e.g. oral
LD50 for rat was determined at 27 mg/kg. Hydrazoic acid is a toxic gas with
a pungent odour.

Precautions

x For seed mutagenesis, sodium azide is most effective at acid pH and during
the treatment the solutions are bubbled with oxygen or air. Under these
conditions, HN3 is readily volatilized. Therefore, all the treatments should be
exclusively performed in a well-ventilated fume hood.

2.5.EXAMPLES OF TREATMENT PROCEDURES

As stated previously, chemical mutagenesis can be performed in both sexually

propagated plants as well as in VPCs. Treatment procedures need to be optimised
based on the plant material chosen, the type of mutagen as well as on the objectives
of the mutagenesis experiment or breeding programme. For better understanding,

69
three detailed treatment procedures for seed propagated crops and VPCs chemical
mutagenesis, which may be adapted to other experimental situations are described
here.

1. A protocol for EMS mutagenesis of shoot apical meristems of banana (Musa

acuminata).

2. A treatment procedure for mutagenesis of barley (Hordeum vulgare) seeds

using EMS.

3. An efficient combined mutagenesis treatment procedure for barley seeds

using sodium azide and N-nitroso-N-methylurea (NMU).

2.5.1. Ethyl methanesulphonate (EMS) mutagenesis of in vitro banana

meristem explants

Below is presented the description of the treatment procedures, as modified and

adapted from Jankowicz-Cieslak and Till, (2016).

2.5.1.1.Preparation

Prepare a sufficiently large number of in vitro shoot tip explants of banana (e.g. 1000
for bulk mutagenesis or 50 for establishing the dose-response curve) for each EMS
dosage (concentration + duration). Select uniformly sized, healthy looking explants
and distribute into autoclaved bottles (Figure 2.3). It is important to note that all items
coming into contact with the in vitro tissues, including the mutagen, must have been
adequately sterilised prior to the initiation of the experiment. Consider from the onset
that three to four cycles of in vitro micro-propagation subcultures will be required to
dissolve any chimeras. As a result, the original population size will increase
significantly during this process. This may, however, be balanced by the loss of
propagules due to injuries resulting from the mutagenic treatment. Thus, space, time
and labour resources should be considered accordingly before initiating any chemical
mutagenesis experiment.

2.5.1.2.Establishing the dose-response curve

In the absence of reliable information on optimal dose for conducting a mutagenesis

experiment, a dose-response curve is typically established prior to conducting bulk
mutagenesis. It is important to note that the frequency of induced mutations may be
different in different genotypes and due to variations in experimental procedures. In
the case of banana shoot tips, the experimental variable measured for determining the
optimal dose is the reduction in fresh weight in relation to the different EMS
concentrations and different incubation periods.

70
Figure 2.4 illustrates the growth reduction and lethality of in vitro banana shoot tips
with increasing EMS concentrations. Based on these results, the optimal doses for
bulk mutagenic treatment were selected. This protocol has been successfully applied
to other vegetative propagules from Jatropha, potato and cassava and could be further
adapted to other in vitro explants such as embryogenic callus, nodal cuttings, in vivo
cuttings, etc.

2.5.1.3.Mutagenesis treatment

Prepare fresh 1M sodium thiosulphate stock solution and dilute to 100mM. This
solution will be used for deactivation of EMS as well as for the decontamination of
working surface and laboratory equipment that has come into contact with EMS.

Calculate the volumes of EMS and DMSO needed, dispense required volumes of
distilled water and autoclave. Let the liquid cool to room temperature; add DMSO
using a sterile pipette tip and required volume of EMS using a sterile syringe and filter
membrane. Shake the EMS/DMSO solution vigorously for 15 seconds for optimal
solubilization.

Pour EMS mixture into each bottle containing the in vitro plant material ensuring that
the tissue is fully immersed in the liquid.

Incubate at room temperature on a rotary shaker at 150 rpm for the pre-determined
length of time. If needed, adjust the rotation speed so that tissues are gently and
regularly moving.

2.5.1.4.Post-treatment

After incubation, fill the bottles with sterile water, mix gently and immediately decant
carefully into an empty beaker using a sterile sieve to capture any material that may
accidentally fall out of the bottle (Figure 2.5.).

It is important to dilute and remove as much EMS solution as possible while

maintaining a sterile environment. It is however advisable to leave a small amount of
liquid in the bottle rather than risk having material drop into the sieve and become
contaminated. Repeat this washing step four times. After the final wash, pour tissue
into a sterile sieve over a beaker. Note that although the tissue is washed, a strong
smell of DMSO may remain. Using a sterile forceps transfer tissue to a Petri plate
containing sterile water and then, carefully transfer rinsed tissue culture material into
the growth medium. Incubate mutagenized material following standard procedures
established for the investigated crop. The next day transfer all treated plantlets into
fresh growth medium to remove any residual DMSO. Tissue growth should be
monitored regularly.

71
Wipe the laminar flow hood with a damp paper towel soaked in sodium thiosulphate
followed by a water rinse to ensure there is no remaining trace of EMS contamination
in the work area. Decontaminate also all laboratory equipment that has come into
contact with EMS.

2.5.1.5.Conclusions

The treatment procedures described here are based on several studies at the Joint
FAO/IAEA Plant Breeding and Genetics Laboratory, Seibersdorf, Austria. Figure 2.6
shows a stable mutant phenotype induced with EMS mutagenesis using this protocol.
With modifications these treatments can be successfully adapted to other VPCs or
other in vitro plant materials.

Because the target meristematic tissues are multi-cellular in origin, tissues with a
genetic mosaic structure may arise immediately post-mutagenesis. Repeated sub-
culturing is performed to reduce genotypic heterogeneity. Eventually, sub-culturing
will result in clonal propagation of fixed mutant alleles (see Chapter 6). This should
be considered when developing a mutant population.

Figure 2.3. Preparation of banana in vitro materials for chemical mutagenesis. Steps include tissue
multiplication under aseptic conditions and transfer of explants into autoclaved bottles, here 200
meristems/bottle, for transfer to the chemical mutagenesis laboratory.

72
Figure 2.4. Establishing EMS dose-response curve for in vitro banana shoot tip explants. Explants
below: different EMS concentrations, from left to right: 0.25%; 0.5%; 1% and 1.5% EMS. Explants on
top: different types of controls, from left to right: water, DMSO and untreated explants. Plate left: 2 hr
incubation; Plate right: 4 hr incubation. Figure adapted from (Jankowicz-Cieslak and Till, 2016).

Figure 2.5. Post-treatment washing of banana in vitro explants. Treated meristems need to be carefully
washed after EMS treatment to remove the remaining EMS solution. After a minimum of 3 washes, the
explants are placed on Petri plates, sealed with parafilm and moved back into the in vitro laboratory.
Mutated banana explants should be immediately transferred to fresh liquid growth medium. Figure
adapted from Jankowicz-Cieslak et al., 2012.

73
Figure 2.6. A mutant banana plant exhibiting a stable rolled leaf phenotype induced via EMS
mutagenesis. Courtesy of J. Jankowicz-Cieslak.

2.5.2. Ethyl methanesulphonate (EMS) mutagenesis treatment of barley

seeds

This is a three-day protocol consisting of three major steps as modified and adapted
from Konzak and Mikaelsen, (1977) and Jankowicz-Cieslak and Till, (2016). The first
step involves pre-soaking the seeds overnight in water. On day 2, EMS is diluted to
the desired concentration and added to the seeds for an overnight incubation. The
EMS solution is removed on day 3. Seeds are then washed with a deactivating solution
(100mM sodium thiosulphate) and rinsed with water before planting.

All steps involving EMS treatment and the post-treatment washes on day 2 and 3 are
performed under a ventilated fume hood.

74
 2.5.2.1.Preparation

Select uniformly sized seeds with 95 – 100 percent germination rate. If the
information on seed viability is not available, it is advised to determine the
germination percentage of the seed stock prior to the treatment with EMS.

The total number of seeds treated depends on the scale of the experiment, and whether
a chemical toxicity test is being performed or a bulk mutagenesis. For chemical
toxicity testing, approximately 200 seeds per treatment are sufficient. As mentioned
previously, for a bulk mutagenesis for a reverse-genetic screen of diploid plants, 3000
to 6000 lines are typically sufficient for the recovery of mutant alleles in any gene.
When attempting to improve one or two traits in an elite germplasm in a mutation
breeding programme, this number may need to be adjusted as well as the optimal dose
to reduce the high mutational load observed in reverse-genetic screens, as discussed
in section 2.3.2. Either way, an excess of seeds should be treated considering that a
percentage of the M1 seeds will not germinate, and, in addition, a percentage of the
M1 plants produced will be sterile following the EMS treatment.

2.5.2.2.Pre-soaking

Estimate the best ratio of seeds to liquid for the EMS incubation. Add seeds to
approximately 1/5th of the total beaker’s volume. Add distilled (or deionized) water
to approximately 1/3rd of the total volume and place on the orbital shaker. Adjust the
rotation speed of the shaker so that all seeds can freely move in the water. Split seeds
into multiple beakers with reduced volume to avoid spillage. Soak the seeds for
12 – 20 hours at 20 – 22°C (~room temperature).

After this period of pre-soaking the uptake or diffusion of the mutagen has reached its
optimal rate or speed, which means that the maximum amount of mutagen can
penetrate the embryo in the shortest possible time. At this stage the coleoptile and
radicle have begun to emerge in case of barley or other small grain cereals.

Drain the water carefully measuring the volume poured-off to estimate the volume of
the EMS solution as well as the wash solution to be added to the seeds during and
post-treatment, respectively.

2.5.2.3.Mutagenesis treatment

All steps are to be carried out under a well-ventilated Biohazard fume hood.

Prepare fresh 1M sodium thiosulphate stock solution and dilute to 100mM which will
be used for deactivation of EMS.

75
The concentrations of EMS to be used can be estimated based on previously published
studies of the species you work on. However, it is important to note that the frequency
of accumulated mutations may be different in different genotypes or due to variations
in experimental procedures. It is therefore advisable to perform a kill curve
experiment using different doses of mutagen. An example calculation for EMS
mixtures of different concentrations containing 2 percent DMSO (volume/volume) in
a final volume of 1 L is shown in Table 2.2. Some publications use molarity of EMS
rather than percentage. Conversion between percentage and molarity is accomplished
using the formula weight for EMS (124.16 g/mol). EMS is not easily soluble in water
so DMSO is added to 2 percent to improve solubility.

The concentrations of EMS to use can be estimated based on previously published

studies of the species you work on. However, it is important to note that the frequency
of accumulated mutations may be different in different genotypes or due to variations
in experimental procedures. It is therefore advisable to always perform a kill-curve
experiment using different doses of the mutagen. EMS is not easily soluble in water
so a 2 percent (volume/volume) DMSO is added to improve EMS solubility. An
example calculation for EMS mixtures of different concentrations (in mM) containing
2 percent DMSO in a final volume of 1 L is shown in Table 2.2. Some publications
use percentage of EMS rather than molarity. Conversion between percentage and
molarity is accomplished using the formula weight for EMS (124.16 g/mol). The
EMS/DMSO solution should be thoroughly mixed. The mixture is prepared in a bottle
sealed with a screw cap, and then shaken vigorously before adding to the seeds. Test
the bottle first by mimicking the shaking procedure using water in the ventilated fume
hood to ensure that the bottle does not leak. Carefully add the required volume of
EMS solution to the beaker containing the seeds. Avoid adding excess EMS/DMSO
solution to the beaker with seeds as this may result in spills during orbital rotation.
Set the orbital shaker to the appropriate speed and incubate for the set length of time.

TABLE 2.2. DIFFERENT CONCENTRATIONS OF EMS MIXTURE CONTAINING DMSO

Final EMS 0 20 30 40 50 60
concentration (mM)
Volume EMS (ml) 0 2.1 3.1 4.1 5.2 6.2
Volume DMSO (ml) 20 20 20 20 20 20
Volume Water (ml) 980.0 977.9 976.9 975.9 974.8 973.8

76
 2.5.2.4.Post-wash

Decant the EMS solution and pour into the toxic waste bottle. Be very careful when
pouring off liquid to avoid splashes. A mesh screen can be placed into a funnel to
capture seeds that may be unintentionally poured out of the beaker. Add 100mM
sodium thiosulphate to the mutagenized seeds and incubate for 15 minutes on the
orbital shaker. Repeat this step for a total of 2 washes with sodium thiosulphate. Add
deionized water to the beaker and incubate for 10 minutes under orbital rotation,
repeat this step for a total of two rinses.

Pour all liquid into the toxic waste bottle. Decontaminate the entire working area as
well as all the tools and the glassware that came into contact with EMS using a100mM
sodium thiosulphate solution.

After post-washing the seeds should either be given a short surface drying or be
planted in the field as soon as possible. This is called the wet treatment. If the seeds
cannot be planted soon after post-washing, they should be readily dried to a moisture
content of approximately 13 percent to prevent any further physiological damage. A
simple practical procedure is to let the seeds dry on filter paper on a laboratory bench
at room temperature (20 – 25°C). This procedure is called the dry-back treatment.
Under these conditions the seeds will remain dormant and can maintain a good
germination capability for several weeks. If longer storage time is required, storage at
very low temperature is advisable.

The treatment procedures described here are based on several studies in the Joint
FAO/IAEA Plant Breeding and Genetics Laboratory using barley and can be readily
adapted to other small grains.

2.5.3. Combined treatment of barley seeds with sodium azide and N-

nitroso N-methylurea

This example describes the application of a combined mutagenic treatment of barley

seeds using sodium azide (SA) and N-nitroso N-methylurea (MNU) with a period of
germination in between the two mutagenic treatments. This protocol results in a high
frequency of point mutations in barley and was utilized to create a TILLING
population of barley cv. ‘Sebastian’ (Szarejko et al., 2017) and rice (Till et al., 2007).
Both mutagens induce mostly GC to AT transitions but in a different local sequence
context (Kurowska et al., 2011; Tai et al., 2016). The objective of using two different
mutagenic compounds in a combined treatment is to broaden the spectrum of induced
mutations.

77
 2.5.3.1.Preparation

Calculate the amount of seeds needed for the treatment. For a large-scale treatment
use the number of seeds estimated on the basis of the evaluation of the mutagen effects
on somatic cells. It is crucial to select well-filled and uniform barley seeds from a
batch with a high germination rate (~100 percent).

Remember that in addition to DNA lesions in the nucleus and cytoplasmic organelles,
mutagens can generate damage in all components of the cytosol and disturbances of
the cell cycle. Therefore, mutagenic treatment can impair metabolism of cells in
various tissues and organs and influence the growth and development of the M1 plants.
These effects, called ‘somatic effects’ are manifested by a delay in seed germination,
reduction of plant emergence, growth reduction, appearance of chlorophyll defects,
reduction of fertility and plant survival. The size of the M1 population should thus, be
calculated bearing in mind the lethality and sterility of the M1 plants to guarantee
sufficient seeds for the subsequent M2 generation. It is worthwhile to organize a pilot
experiment to compare the somatic and genetic effects induced by a range of doses.
Such a pilot experiment will expand the procedure but will surely help in a proper
selection of the optimal doses for the large-scale treatment.

If a complete pilot experiment is not feasible before performing a large-scale

mutagenesis, it is essential to conduct a preliminary mutagenic treatment with
different doses of mutagens to create a kill-curve and evaluate an optimal mutagen
dose. For barley use the standard seedling assay of measuring seedling emergence and
growth reduction. To perform such assay plant the seeds treated with a range of
mutagen doses into pots filled with soil and covered with a 3-cm layer of sand. Seven
to ten days after the mutagenic treatment cut-off all the seedlings near the surface of
the sand, count their number and measure their length. Calculate the growth reduction
separately for each variety, dose and replication (see Chapter 1). If you cannot plant
M1 seeds immediately after the treatment, completely dry the seeds on a filter paper
and store them in plastic bags at 4°C until sowing time. The response to mutagens
may differ between barley genotypes, therefore it is recommended to evaluate the
optimal dose of mutagens separately for each genotype (Figure 2.7.). As mentioned
earlier, the optimal doses may be different depending on the objectives of the
mutagenesis experiment or breeding programme.

2.5.3.2.Pre-soaking

The seeds should be pre-soaked in distilled water before treatment with the mutagen
for physiological activation. The amount of distilled water used in pre-soaking should
be at least two to three times the volume of the dry seeds. Eight hours of pre-soaking
at room temperature (20 – 24°C) is optimal for barley, but for convenience the seeds

78
can be pre-soaked overnight. The pre-soaking reduces the somatic effects of the
chemical mutagen.

2.5.3.3.Mutagenic treatment

It should be stressed that most chemical mutagens are also strong carcinogens. For
this reason, all steps of mutagenic treatment should be carried out under a well-
ventilated Biohazard fume hood. Disposable gloves and a laboratory coat should be
worn at all times when performing the treatments and dealing with treated seeds.
Taking these precautions is especially important during treatment with MNU –
a strong mutagen and carcinogen.

The mutagenic effect of SA depends on the acidic pH of the treatment solution

(Nilan et al., 1973). The doses of SA that are routinely used for the mutagenic
treatment of barley seeds are within the range of 0.5 – 4mM for 3 – 5 hours (Nilan et
al., 1973; Maluszynski et al., 2003), note that a dose as high as 10mM for 2 hours has
been applied for the creation of a TILLmore population of cv. ‘Morex’ (Talamè et al.,
2008).

When a combined treatment with two mutagens is performed, the first regular protocol
of mutagenic treatment is followed with the addition of a 5 – 6 hours inter-incubation
germination (iig) period between treatments during which the seeds are incubated on
a wet filter paper at room temperature.

Calculate the amount of solutions needed for the treatments (for all testing doses). For
small grain seeds such as barley, prepare a volume ensuring a 0.5ml solution per one
seed.

Prepare appropriate amount of fresh solutions of sodium azide and of MNU. MNU
should be dissolved in dH2O, whereas SA should be dissolved in a phosphate buffer
with pH=3.0. To prepare the phosphate buffer at pH=3.0, use 54.436 g KH2PO4 to
which 3.67 ml H3PO4 is added per 1L buffer.

When evaluating an optimal dose prepare the mutagenic solutions starting from the
basic solution (the highest concentration used for treatment). Leave part of this
solution for the treatment and dilute the rest to the other required concentrations. You
can use a formula: C1 × V1 = C2 × V2, where: C1 is the concentration of the basic
solution, V1 – volume of the basic solution, C2 – concentration of the solution
required, V2 – volume of the solution required. Use a ventilated fume hood for the
preparation of the mutagenic solutions. Before the treatment, pour out the dH2O from
the beakers containing thee seeds and rinse twice with tap water. Be cautious to
completely remove the water after rinsing.

79
Perform the treatment, i.e. pour the mutagen solutions into the beakers with the pre-
soaked and rinsed seeds. Keep the same order of combinations during the whole
procedure, i.e. pre-soaking, rinsing, treatment, and rinsing after treatment. Perform
the mutagenic treatment at room temperature.

After 3 hours of treatment with SA (the first applied mutagen), pour off the mutagen
solution and thoroughly rinse the seeds (3 – 4 times) in tap water. Then put the seeds
into trays containing a few layers of a filter paper, cover them with a wet sheet of the
filter paper and keep for 6 hours at room temperature. Next, transfer the seeds into the
labelled beakers and add the second mutagenic solution to be applied, i.e. MNU. Treat
the seeds for 3 hours and then pour off the mutagen and rinse the seeds again 3 – 4
times in running tap water. The mutagen solutions should always be poured off into
the toxic waste bottles and appropriately handled.

Figure 2.7. Differential sensitivity of barley cultivars ‘Class’ (A, B) and ‘Mauritio’ (C, D) to mutagenic
treatment with MNU (A, C) and NaN3 (B, D) based on seedling emergence and growth reduction.

80
For the barley cultivar ‘Sebastian’ that was used to create a TILLING population, we
applied two different treatment combinations:

1. 1.5mM NaN3/3 h – 6 h iig – 0.75mM MNU/3 h

2. 1.5mM NaN3/3 h – 6 h iig – 0.5mM MNU/3 h

In both combinations the same dose of SA was used (1.5mM/3 h) while the dose of
MNU was different. The treatment with the higher dose of MNU (0.75mM/3 h) caused
a higher mutation frequency than the 0.5mM, but also resulted in a much higher
sterility of M1 plants (Szurman-Zubrzycka, pers. comm).

2.5.3.4.Post-treatment

An extensive post-treatment of rinsing in tap water is necessary to terminate the action

of the mutagen and to remove any mutagenic residues from the surface of the seeds.
To facilitate sowing, the treated seeds can be allowed to dry on a filter paper under a
ventilated fume hood. Not, however, that a too intensive drying, especially at
increased air temperature, can enhance the somatic damage effects of the mutagen.

2.5.4. Conclusion

Chemical mutagenesis has proven extremely useful to create new allelic variants that
then can be used in functional genomics studies and/or plant breeding. Advantages
include low-cost, a high density of variation and the technique can be applied to many
species. Point mutations induced by EMS treatment can have varying effects on gene
expression ranging from knockouts to (subtle) changes in protein function. Therefore,
EMS can produce a range of phenotypes and provide an in-depth characterization of
gene function. Mutagenesis using two different mutagens such as SA with MNU may
lead to a broader spectrum and different types of mutations. Furthermore, relatively
small populations are required to recover desired traits. However, the accumulation
of a high density of induced mutations does imply that each plant line will harbour
numerous mutations. Therefore, additional steps need to be taken, such as
backcrossing, for the unambiguous assignment of the mutant gene causing the altered
trait and to reduce or eliminate undesired mutations in the genetic background.

81
82
 3. TYPES OF MUTATIONS

Mutations are heritable changes that occur in the genetic material of living organisms.
These changes occur for various reasons and may be natural or induced. They can be
recognized as phenotypic variants at different stages in the life cycle, but the primary
effects are genetic (Lundqvist, Franckowiak and Forster, 2012).

3.1.PHENOTYPIC MUTANTS

All mutations occur in the DNA and various classes are listed in Section 3.2. However,
for practical purposes including plant breeding the expressed mutant phenotype is
usually the first description of a mutant. Phenotypic selection is a fundamental part of
plant breeding and therefore, there has been a natural link between mutant phenotypes
and their selection by plant breeders, which traditionally has been based on phenotypic
selection. The exploitation of mutants via phenotypic selection often occurs long
before the underlying genetic change is understood. Thus, for example, induced semi-
dwarf mutants in rice in the USA were exploited in breeding new short stature
varieties before the direct cause, a mutation in the gibberellin biosynthesis, gene was
discovered (see Section 3.2.2).

Forster et al., (2012) using barley as an example described a wide range of induced
phenotypic mutants and how these may be classed and catalogued in various
developmental stages.

x Seed
x Seedling development
x Vegetative growth
x Reproductive growth
x Inflorescence formation
x Spike development
x Meiosis and flowering
x Adult plant ripening

In the hands of plant breeders novel and desirable mutant traits are quickly detected
and selected from field observations, they include agronomic traits, e.g. semi-dwarf
stature and flowering time, pest and disease resistance and yield variation. Selected
lines are then incorporated into breeding programmes using phenotypic mutant

83
descriptors. The classification of mutants at the phenotypic level also includes quality
and nutritional traits (composition and content of proteins, oil, minerals, vitamins
etc.).

Phenotypic mutants should not be confused with physiological disorders, which are
non-heritable; these can often mimic mutations (Lundqvist, Franckowiak and Forster,
2012). This is particularly important in mutation induction and detection at the first
mutant generation (M1) is often weak and suffers from physiological disorders as a
result of mutagen treatments. Thus, the M1 cannot be used for mutant phenotypic
selection. The M2 generation is the first opportunity to apply phenotypic mutant
screens (though, this is limited to single plant selection, see Chapters 4 and 5).

3.2.GENOTYPIC MUTATIONS

Genotypic mutations describe the primary mutant event in the genome, i.e. changes
to the DNA sequences and are described in the sections below, but may also include
epigenetic modifications, which often do not result in heritable mutations
(Text Box 3.1).

Text Box 3.1.

Mutations can also be mimicked through epigenetic modifications. The most common
modification is DNA methylation/demethylation, which, often triggered through
environmental effects, can interfere with gene expression. Numerous examples are
reported on DNA methylation changes in response to abiotic stresses, such as cold,
salinity, drought, osmolality, or imbalance of mineral nutrition. One example is a
methylated gene (Asr2) that in plants, such as tomato, is linked to alleviation of water
stress response. Induction of artificial drought has led to demethylation in the regulatory
region of this gene and thus, to its expression and adaption of the plant to the water-deficit
stress condition (González, Ricardi and Iusem, 2013). Epigenetic changes in response to
biotic stress are documented as well, for instance, the demethylation of resistance genes
in rice after infection with the bacterial blight disease pathogen Xanthomonas oryzae (Li
et al., 2012). Epigenetic modifications can be stably transmitted through mitosis or
meiosis in the absence of the original inducing signal. In contrast to animals, the
methylome in plants is maintained during sexual reproduction (Eichten, Schmitz and
Springer, 2014; Quadrana and Colot, 2016). In a mutation breeding programme, it is
important not to confound epigenetic changes with the nucleotide or chromosome changes
that have been induced through the use of physical or chemical mutagens. In some cases,
however, the borders between both may be unclear. Mutation of the ddm1 gene that is
required to maintain normal cytosine methylation pattern in Arabidopsis has led to
demethylation of mainly repetitive sequences and, subsequently, to activation of
transposable elements (Jeddeloh, Stokes and Richards, 1999). In the next chapter the
effects of transposable elements, which can induce new mutations in the genome, are
described (see section 4.5).

84
The genome of the cell comprises DNA embedded in chromosomes, which are in the
nucleus, but also in the organelles. In plants, both mitochondria and plastids also carry
DNA and these are in the cytoplasm of the cell.

Mutations may thus happen in any DNA, those occurring in the nucleus will be
transmitted to male and female germlines (sperm and eggs), whereas mutations in
cytoplasmic (organelles) DNA may only be transmitted via the cytoplasm of the egg
(though, in some rare cases, e.g. banana the cytoplasm of off-spring is donated via the
sperm).

3.2.1. Genome mutations

The cell karyotype encompasses the entirety of the chromosomes in the nucleus, thus
defining the organism with a number (n) of chromosome sets. Ploidy mutations
comprise changes in the genome number, either in the form of a subtraction or an
addition of a complete set of chromosomes, and/or a reduction of diploid (2n) to
haploid (n) genomes which, can, for example, be spontaneous or experimentally
induced (see Chapter 8.2.). In polyploid species, the haploids have more than one set
of chromosomes and are called “polyhaploids”. Spontaneous or colchicine induced
chromosome doubling in developing doubled haploid (DH) plants results in the
generation of stable homozygous lines. These are genetically pure and fully fertile
(having a balanced set of paired chromosomes at meiosis) and are thus invaluable in
plant breeding and genetics. They are of outmost relevance in mutation breeding for
fixing any recessive mutant allele (see Chapter 8-B). Detailed protocols for DH
production in more than 20 crop species including cereals, vegetables and fruits have
been described in Maluszynski et al., (2003). One method of producing haploid
embryos is via aberrant pollination using pollen killed or made impotent by
irradiation. Here the pollen may stimulate the egg to develop as a haploid embryo
without fertilisation (parthenogenesis). The method is fairly universal and can be
applied to a wide range of species (see Germana, (2012) for practical examples).

Polyploidy induced either by genome duplication (autoploidy) or genome addition

(alloploidy) has occurred naturally during the evolution of many plant species
including crops, and has been used to produce new species, e.g. triticale, which
combines genomes from wheat and rye. One effect of polyploidy is to increase the
volume of the nucleus; this in turn increases the size of the cells, tissues, and organs
and consequently of the entire plant, which at that stage may out-perform their diploid
relatives. Polyploids can originate through doubling of somatic chromosome numbers
from diploids, but it is suggested that in the wild, the most likely origin of polyploidy
are 2n gametes exhibiting a modified gametogenesis due to errors in meiosis (Heslop-
Harrison and Schwarzacher, 2007; Ortiz and Peloquin, 1992). Alloploids have
additional advantages as the different genomes contain dissimilar sets of genes and

85
thereby enrich the gene diversity – by adding new genes – promote heterosis
(hybrid vigour) and buffer deleterious mutations. Polyploids, spontaneous or induced
have therefore been attractive for domestication and crop improvement. Examples of
polyploid crops include:

x triploids: banana, watermelon and apple;

x tetraploids: cotton, groundnut, brassicas, durum wheat, leek, potato and
tobacco;
x hexaploids: bread wheat, oat, triticale and chrysanthemum;
x octaploids: dahlia, strawberry, triticale and pansies.

In some species the polyploidy event is so ancient that it cannot be recognised by

cytological techniques and can only be detected at the molecular level, e.g. by
discovering gene duplications. Such paleopolyploids include: oil palm, maize, rice,
rubber and soybean.

Deliberate induction of polyploidy began in the mid-20th century and was successful
in vegetable crops where organ size is of particular importance (Akerberg and
Hagberg, 1963). A more recent example is the development of the wheat/rye hybrid;
triticale, which combines productivity traits of wheat with abiotic stress tolerance of
rye. Triticale has been produced at the hexaploid and octoploid levels. The hexaploids
are now established as a crop for marginal lands, usually not suitable for wheat
cultivation (Hao et al., 2013).

3.2.1.1.Chromosome mutations

The euploid (normal) number of chromosomes, their arrangement and their structure
are generally established for each species. However, apart from the variation in the
number of chromosome sets, i.e. n = × 1, × 2, × 3, × 4, etc., described above, several
other types of alterations may appear as a result of mutations.

Aneuploidy

Aneuploidy is a category of chromosome mutation in which the chromosome number

is abnormal (non-euploid). Generally, the aneuploid chromosome set differs from
wild type by only one or a small number of chromosomes. Aneuploids can have a
chromosome number either greater or smaller than the euploid chromosome
complement. Aneuploid nomenclature is based on the number of copies of specific
chromosomes. For examplHWKHDQHXSORLGFRQGLWLRQQí1 is called a monosomic
(meaning “one chromosome”) because only one copy of some specific chromosome
is present instead of the usual two (a pair) found in the euploid. The aneuploid 2n + 1

86
is called trisomic, 2n í 2 is nullisomic, and n + 1 is disomic, which is an aberration of
a haploid organism. (Griffiths et al., 2000). Such aneuploidy sets are common in
polyploid species such as bread wheat.

Aneuploids can occur naturally, but are also produced in abundance in progenies of
crosses where parents contribute unequal numbers of genomes and chromosomes. For
a detailed discussion on the use of aneuploids in wheat see Law, Snape and Worland,
(1987). Aneuploids such as monosomics can be generated by the application of low
doses of X-rays, gamma-rays and fast neutrons to seeds, flower organs or pollen grains
and have proven to be of great value in a number of plant species, particularly crop
species, in developing genetic stocks and locating genes (Sanamyan et al., 2011)

In triploid Musa cultivars, treatment of cell cultures with gamma-rays has been shown
to result in a significant reduction of the chromosome number (Shepherd and Bakry,
2000). This demonstrates the need for determining the karyotype of mutagenized plant
cells, particularly when they are derived from in vitro mutagenesis experiments.
Roux et al., (2003) demonstrated that flow cytometry can be applied to facilitate
screening of mutagenized plant material. They irradiated shoot tips of the triploid
(2n = 3x = 33) Musa clone ‘Grande Naine’ with gamma rays at 35 Gy from a 60Co
source and proceeded to ensuing sub-cultures up to M1V4 and identified aneuploid
plants with chromosome numbers of 2n = 31 or 32.

Chromosomal rearrangements

Hermann Josef Muller in his Nobel Lecture entitled “The Production of Mutations”
(1946) pointed out the effect of ionizing radiation on rearrangements of parts of
chromosomes caused by “breakages of the chromosomes, followed afterwards by
attachments occurring between the adhesive broken ends that joined in a different
order than before.” The Nobel Peace prize winner also reported on a clear dose-effect
relationship that he and his co-workers detected.

DNA double strand breaks (DSB) are subjected to the cell’s own repair mechanisms,
which preserve the genetic stability/integrity. The main DSB repair pathways are:
1) homologous recombination (HR), and 2) non-homologous end joining (NHEJ)
(Puchta, Dujon and Hohn, 1996; Waterworth et al., 2011). For a comprehensive
review on repair processes in plants see Manova and Gruszka (2015). The most
prominent pathway for DSB repair is NHEJ. HR functions mainly during S and G2
phases of the cell cycle. However, both pathways are error prone and it is ultimately
these “mistakes” that lead to chromosomal rearrangements; these can be deletions,
duplications, inversions, insertions, or translocations (Figure 3.1). An example of
chromosomal rearrangements analysed by genomic in situ hybridization (GISH) is
shown in Figure 3.2.

87
Translocations

Translocations are of two main types: intra- or inter-chromosomal rearrangements. In

intra-chromosomal translocations a transposition or shift of an interstitial segment
within an arm or from one arm to the opposite occurs. Inter-chromosomal
translocations can be terminal, whole arm, reciprocal or intercalary. Wang et al.,
(2012) explained these four types of translocations as described below.

x Terminal translocations (Figure 3.1e) are the most common type, these
involve the distal segment of a chromosome replacement by a segment of
another (e.g. an alien) chromosome.
x When breaks occur within the centromeric region, it may result in a whole
arm translocation, in which a whole new arm of a chromosome replaces that
of another chromosome.
x A reciprocal translocation (Figure 3.1e-i) is a type of chromosome
rearrangement involving the exchange of chromosome segments between
two chromosomes that do not belong to the same pair of homologous
chromosomes.
x A compensating translocation, in which a desired alien segment replaces an
equivalent segment of a homoeologous (related) chromosome, is more likely
to be beneficial. In an intercalary translocation, a chromosome segment is
inserted into another; however, this seldom occurs as this requires several
simultaneous breakage and reunion events. This type of translocation will be
desirable when an alien segment containing desirable genes is inserted into a
host chromosome without loss of host genes. This requires one break in the
host chromosome and two breaks in the alien chromosome, with the desirable
genes between the breaks. The excised alien segment has unstable ends,
which can unite with the unstable ends of the host chromosome at the break
point. If the inserted alien segment is quite short, it should not interfere with
the pairing of the homologous host chromosomes, particularly if the
chromosome with the insertion is made homozygous, e.g. by selfing.

Irradiation is used efficiently for inducing translocations in amphiploids or

chromosome addition lines in order to introgress alien chromosome segments with
useful genes, such as genes for disease resistance. In particular for wheat, numerous
examples for the transfer of genes from alien chromosome using 60Co-gamma
irradiation of seeds or gametes are available (Friebe et al., 1991; Mukai et al., 1993;
Liu, Chen and Liu, 1998; Liu, Chen and Liu, 2000; Chen et al., 2013).

88
Inversions

In the case of inversions, a fragment of chromosome is re-ligated in the original place

but after a 180° rotation (Figure 3.1c). Thus, the linear order of the genes is opposite
of that of wild type. If the inversion involves the centromere it is known as a
pericentric inversion (Figure 3.1c-ii), if not it is called a paracentric inversion (Figure
3.1c-i). Meiotic recombination events involving inversions can result in recombinants
either carrying duplications or displaying deficiencies for parts of that chromosome.
Pericentric inversions often give rise to new karyotypes. Paracentric heterozygotes
usually form reverse loop pairing at prophase of meiosis. As a result of crossing-over
within the loop, dicentric bridges and acentric fragments may appear at anaphase 1,
leading to the formation of aberrant gametes that carry deletions, insertions, and either
zero or two centromeres. Paracentrics are further characterized by a reduced
recombination frequency due to abortion of cross-over chromatids and imperfect
pairing of homologues. The paracentric heterozygotes give rise to pollen abortion, as
shown in maize (Morgan Jr, 1950), bean (Sjödin, 1971), or barley (Ekberg, 1974) .

Chromosomal inversions have greatly contributed to species differentiation in nature.

They are found as fixed differences between species and as polymorphisms within
species in many groups of animals and plants. In some groups, speciation is associated
with inversions and other changes in the karyotype (White, 1978). The evolutionary
role of chromosomal inversions in plants has been extensively studied by Hoffmann
and Rieseberg, (2008) and Lowry and Willis, (2010).

Segmental duplications

As described earlier, duplication of DNA sequences is one common form of mutation

that occurred during the evolution of all organisms. Generally, if the size of the
duplication is bigger than 1 kb in length it is called a segmental duplication (Figure
3.1b). Such duplications are found in maps of diploid and polyploid species, which
for example could provide evidence for the hypothetical development of a diploid
Brassica spp. from an ancestor with a lower than expected original basic chromosome
number. Likewise, through comparative genomics, duplications of chromosomal
regions in sorghum (Paterson et al., 2009) and oil palm (Singh et al., 2013b) were
proven.

89
 Deleted
D e
elete
ed
e

Before:
E F

A B C D E F G H
Breakage
After:

A B C D G H

Figure 3.1a. Deletion: An interstitial part of a chromosome is lost after double breaks. After the end
re-joined, the chromosome is shortened. The deleted fragment is acentric and will be lost if not inserted
into another chromosome (see Insertion 3.1d. below).

Before:

A B C D E F G H
After: Duplicated
p

A B C D E F G H G H

Figure 3.1b. Duplication: An extra copy of a chromosome region is produced. The duplicated region
can be located adjacent to each other or can be unlinked.

90
 Before: 180o
Inverted

A B C D E F G H
Breakage
After:

A B C F E D G H

(i) Paracentric inversion

Before: 180o
Inverted

A B C D E F G H
Breakage
After:

A B C D E H G F

(ii) Pericentric inversion

Figure 3.1c. Inversion: After two breaks in a chromosome the region between the breaks rotates 180
degrees before re-joining with the two end fragments. Since no genetic material has been deleted or
duplicated, inversions generally do not cause phenotypic changes. Two types of inversions are
differentiated: paracentric and pericentric. In paracentric inversions (i) the centromere is outside the
inversion, whereas pericentric inversions (ii) include the centromere.

91
 Before: After:

Chromosome 1
A B C D E F G H A B C D G H
Breakage

Moved and inserted into

another chromosome
Inserted
Chromosome 2
G H A B C G E F H A B C
Breakage

Figure 3.1d. Insertion: A deleted chromosome fragment resulting from a double break is inserted into
another chromosome, which requires a single break in that chromosome and subsequent re-joining of
the ends.

Before:
Chromosome 1 Chromosome 2

A B C D E F G H G H A B C
Breakage Breakage
After:
Translocated Translocated

G C D E F G H A B H A B C

(i) Reciprocal translocation

Before:
Chromosome 1 Chromosome 2

A B C D E F G H G H A B C
Breakage Breakage
After:
Translocated

C D E F G H A B H A B C

(ii) Non-reciprocal translocation

Figure 3.1e. Translocations: Two non-homologous chromosomes exchange segments after breaks have
been induced in both chromosomes; here they result in a terminal translocation. This exchange can be
either reciprocal (i) or non-reciprocal (ii), if the segment of one chromosome is lost.

Figure 3.1. Major types of chromosomal mutations: (a) deletion, (b) duplication, (c) inversions, (d)
insertion, and (e) translocation.

92
Figure 3.2. Example of chromosomal rearrangements, as a consequence of X-ray irradiation in a
somatic hybridization experiment between Brassica napus and irradiated B. nigra protoplasts.
Metaphase chromosome spreads are analysed by genomic in situ hybridization (GISH) (left: DAPI
counterstaining; middle: GISH, B. napus orange, B. nigra yellow-green; right: detail with “zebra”
chromosome resulting from multiple insertions of B. nigra fragments, arrow showing small B. nigra
insertion in B. napus background), (Nielen et al., 1998).

3.2.1.2.Extra-nuclear mutations

As stated earlier in the chapter, apart from the nuclear genome of eukaryotes, genes
are also located in cytoplasmic organelles: mitochondria and plastids. These
organelles are called semi-autonomous because only a part of the proteins they need
for their function is encoded by their own genome, the rest is encoded by the nuclear
genome. In plants we differentiate between the chloroplast genome (plastome) and
the mitochondria genome (chondriome). In most higher plants they encode for
100 – 150 genes, among them the large sub-unit of ribulose: 1.5-bisphosphate
carboxylase/oxygenase (RuBisCO). The chondriome is more complex and has
circular and linear DNA molecules and its size ranges from 200 – 2000 kbp,
depending on the species.

Organelles characteristically display a non-Mendelian inheritance, which is

predominantly uni-parental, usually maternal. Another characteristic is that organelle
inheritance includes somatic segregation (sorting-out) of genetically distinct
organelles. For more details on organelle inheritance, see Greiner et al., (2015). In
general, mutation induction of organelle genes is not well documented and it is
assumed that it is far more difficult than for nuclear genes. Prina, Pacheco and Landau,
(2012) reported on such mutated genomes and their function. A main character of
interest, due to its importance in hybrid production, is cytoplasmic male sterility
(CMS), encoded by mitochondrial genes (Wang et al., 2006). This can be exploited
to produce female only (seed bearing) plants that must be cross pollinated to yield F1
hybrid seeds. Cytoplasmic male sterility for F1 hybrid production was first exploited
in maize, which resulted in spectacular yield increases. F1 hybrid breeding is currently
of great interest and several major crops are being converted to F1 hybrids, notably
rice, but also rye, and with active development programmes in many other crops.

93
 3.2.2. Gene mutations

Gene mutations can be divided in structural variations that encompass copy number
mutations, and nucleotide point mutations.

As in chromosomal mutations, where the number of chromosomes of a genome or of

single chromosomes is abnormal, the copy number of DNA sequences can also change
due to mutation. This refers to regions typically larger than 1 kb, but also to short
insertions and deletions (indels) with a size of 20 – 50 bp. A typical example of an
increase in copy number is active retro-elements, which multiply through a “copy and
paste” mechanism (see Chapter 4). Whereas retro-elements are well-defined entities
that encode the complete machinery for reverse transcription and re-integration of the
new copy into the genome, other copy number variations (CNVs) are thought to result
from non-allelic homologous recombination between DNA segments of high
similarity that are not alleles. ĩPLHĔNRet al., (2014) reviewed the current knowledge
on copy number polymorphism in plant genomes, including associations of CNVs
with plant phenotypes. One example given for such association is the diversity of
flowering time and plant height in wheat (Triticum aestivum), where CNVs
association resulted in the alteration of two major genes; the gene for photoperiod
response (Ppd-B1) and the gene for vernalisation requirement (Vrn-A1) creating new
alleles for each (Díaz et al., 2012).

In point mutations, which comprise all DNA mutations where only one base pair is
affected, the differentiation is between base substitution and base additions or
deletions. A base substitution can be either a transition or a transversion. Whereas in
a transition, a purine (A or G) is changed to another purine, or a pyrimidine (C or T)
to another pyrimidine; in a transversion a purine is exchanged by a pyrimidine and
vice versa. If these substitutions are within the coding region of a gene, they can result
in:

x a missense mutation, where a changed triplet codes for a different amino

acid, or
x a nonsense mutation, the new triplet creates a stop codon inducing a
premature stop in translation, or
x a silent mutation in which the change has no consequence to the amino acid
sequence of the synthetized peptides/proteins (Table 3.1).

The effect of point mutations on a phenotypic character can be wide ranging from null
(no effect) to abolishing all function or inducing partial malfunction of an enzyme, or
even producing an entirely new trait that was not part of the prior plant portfolio. As
explained earlier (Chapter 2) point mutations, i.e. changes in one nucleotide, are

94
primarily induced by chemical mutagens. However, the example of the semi-dwarf
rice mutant Calrose 76 that has been developed through gamma irradiation
(Rutger et al., 1976; Rutger, Peterson and Hu, 1977) has shown, that the phenotype
was the result of a point mutation, namely a C to T transition in a gene coding for the
gibberellin synthesis pathway, forming the Sd1 locus (Spielmeyer, Ellis and Chandler,
2002).

Base additions and deletions engender frameshift mutations that can add up to a
completely new protein being synthesised, and this may lead to the recovery of one
or several new traits. Frameshift mutations can be suppressed by another insertion or
deletion that reconstitutes the original frame (see Figure 3.3).

3.2.3. Gene(s) mutations expression at the trait level

Traits may be classed into two types: qualitative traits (single gene traits) and
quantitative (multi-gene traits). Selection for each of the two groups of traits should
be considered carefully prior to the initiation of a mutation breeding programme. Any
proposal to use induced mutations in plant improvement must first consider the
likelihood of success when compared with conventional techniques and the effort
required to obtain the desired genotype. This likelihood of success can be measured
in relation to the breeding system of the species and the genetic control of the character
to be improved. The selection for a specific trait is usually the first option, e.g. yield,
height, flowering time, disease resistance, etc., and this is often done without any
consideration of the genes involved. Single genes, especially developmental genes
such as those controlling flowering time (photoperiod and vernalisation genes) and
height (semi-dwarfing genes) can have major pleiotropic effects on other traits,
especially yield, as they are major genetic factors that adapt a cultivar to its growing
environment. However, these major genes are often fixed in elite germplasm and if a
breeder wants to alter such traits subtly he may be forced to develop mutations in
minor quantitative genes that influence the trait and this is more challenging for the
breeder.

3.2.3.1.Qualitative traits

Qualitative traits are controlled by one or two genes and are characterized by their
simple Mendelian inheritance. Typical examples for qualitative traits include plant
stature, flowering time, kernel size, flower and leaf colours, shell thickness, seed
amino acid content, etc. The segregants from crosses typically separate into distinct
classes – non-continuous variation – such as green or yellow seeds in peas, thin or
thick shell in oil palm.

Disease resistance is often considered as a qualitative trait when it is controlled by

race specific major R-genes. The plant becomes vulnerable if a new race of the

95
pathogen arises. Also, if a particular cultivar with a qualitative resistance is cultivated
continuously, the pressure on the pathogen becomes high and new races evolve that
break down plant resistance. It is therefore desirable, for a sustainable breeding
programme to combine several major resistance genes into new cultivars in order to
achieve broad resistance. This strategy has been successfully applied to control wheat
stem rust (Singh et al., 2011). The resistance, however, may break down when a new
aggressive race of the pathogen appears, e.g. the race of Puccinia graminis that
appeared in Uganda in 1999, Ug99 (Pretorius et al., 2000). Ug99 is spread via wind
borne spores and quickly moved out of East Africa, to the south into Yemen, and as
far east as Iran. It is a threat to worldwide wheat production and wheat production in
affected areas can only be achieved through the high use of expensive fungicides.
New resistant mutant varieties were developed under an IAEA Interregional TC
Project INT/5/150 (see examples in Section 3.3.1).

TABLE 3.1. TYPES OF MUTATIONS WITHIN THE CODING REGION OF DNA

Type of Mutation in Example

Change the DNA

None None 5’ - A - T - G - A - C - C - G - A - C - C - C - G - A - A - A - G - G - G - A - C - C - 3’ *
Met Thr Asp Pro Lys Gly Thr

Silent Base 5’ - A - T - G - A - C - C - G - A - C - C - C - C - A - A - A - G - G - G - A - C - C - 3’
substitution Met Thr Asp Pro Lys Gly Thr

Missense Base 5’ - A - T - G - C - C - C - G - A - C - C - C - G - A - A - A - G - G - G - A - C - C - 3’
substitution Met Pro Asp Pro Lys Gly Thr

Nonsense Base 5’ - A - T - G - A - C - C - G - A - C - C - C - G - T - A - A - G - G - G - A - C - C - 3’
substitution Met Thr Asp Pro STOP!

Frameshift Addition/ 5’ - A - T - G - A - C - C - G - A - C - G - C - C - G - A - A - A - G - G - G - A - C - C - 3’
deletion Met Thr Asp Ala Glu Arg Asp

Frameshift mutations: Insertions or deletions of nucleotides that cause a shift in the translational
reading frame.
Suppressor mutations: Mutations that counteract or suppress the effects of another mutations.

96
 Figure 3.3. Frameshift mutations and their possible suppression.

3.2.3.2.Quantitative traits

Quantitative traits are characterised by continuous variation such as normal frequency

distributions in progenies. Theoretically it takes only a small number of genes (4) with
equal effects to produce a normal distribution; therefore, quantitative traits can be
controlled by a small number (4 or 5) or more genes (polygenes). Segregation of genes
for a quantitative trait, each contributing a portion of the total variation would modify
their expression by interacting with other genes and with the environment (Paran and
Zamir, 2003). The net effects of each gene affecting the trait can be partitioned into
components attributable to additive, dominance and epistatic genetic variance,
variance of G×E interactions and other environmental variances and are highly
specific to the population under study.

In the context of increased complexities of the underlying gene actions involving

several loci with unknown effects and interactions, many quantitative trait loci (QTLs)
remain unknown. Many traits of interest such as grain/ fodder yield, seed weight,
panicle weight, stem diameter, plant height and seed quality parameters are
quantitative in nature and are important to the plant breeders. In order to enhance the
genetic variation for quantitative traits, mutagenic agents can be used, as it is practiced
for qualitative traits.

The frequency and effects of segregating genes for a QTL decides the quantum of
variances and heritability of a population and environment (Kharkwal, 2012). Such
genetic variances induced by mutations generally show narrow ranges over many

97
traits and crop species (Keightley and Halligan, 2009). Under the stabilizing selection,
segregating genes with large effects would show low expected heterozygosity
(Turelli, 1984), such that the predicted variance maintained is proportional to the total
mutation rate of genes controlling a quantitative trait and inverse of the selection
strength.

Text Box 3.2.

Among studies on efficient methods for the selection of mutants for QTLs, Ukai
and Nakagawa, (2012) demonstrated a one-spike-one-grain selection in cereals.
The reasoning is that assuming a quantitative trait has a phenotypic value that
follows a normal distribution (A) with a mean N and a standard deviation s,
where: N is the genotypic value of the normal plant and s2 is the environmental
variance.

If one supposes that a targeted mutation (Aoa’) is induced at a locus with a high
genetic effect with a mutation rate per cell p1, then, assuming also that the
phenotypic value of the homozygote a’a’ follows another normal distribution (B)
of mean M (>N) with the same magnitude of standard deviation as a normal plant
(V). In other words, the mutant has a genotypic value of M and an environmental
variance of V2. Finally, the authors also assume that the mutated allele is
completely recessive to the original allele, and that the phenotypic value of the
heterozygote (Aa’) follows the distribution A. Then, the phenotypic values of
plants in M2 population is thought to follow a distribution which is a combination
of two normal distributions A and B with a ratio of 1–0.25 p1 and 0.25 p1. In such
cases, all plants with phenotypic values above a threshold point (Th) may be
selected. In this type of selection, the value of Th in relation to the mutant
genotype and environmental variation is therefore, important.

Many genetical theories have been proposed to dissect the complex traits governed
by polygenes and statistical methodologies were developed for improving our ability
to understand the effect of these genes (Text Box 3.2). Statistical genetics has focused
on methods leading to partitioning of the components of variation, parameter
estimation and prediction of breeding value for phenotypic data.

Phenotypic manifestation of quantitative characters altered by the mutagenic

treatment can be detected by measures of central tendency (mean) and dispersion
(variance) instead of ratios or inheritance patterns. The mean values for quantitative
traits in populations obtained from irradiated gametes (pollen grains) or embryos
(dormant seeds) are in most instances lower in treated than in untreated populations.
Khan, Wani and Parveen, (2004) in a study on induced genetic variability for

98
quantitative traits in mung bean (Vigna radiata) demonstrated that the assessment of
variance has been the most dependable statistical measure to find the mutagenic effect
on the QTLs. Estimation of various genetic parameters viz., genotypic coefficient of
variation (GCV), heritability (h2) and genetic advance (GA) for three quantitative
characters of the two varieties of mung bean provided ample evidence that mutagenic
treatments could alter mean values and create additional genetic variability for
quantitative traits.

With the advent of genomic tools, QTLs are being identified using a diverse array of
molecular markers and sequencing facilities. Molecular markers are used to know the
inheritance of the phenotype for a quantitative trait, which are useful in locating the
specific genes in the genome and the magnitude of their effects on these traits. High
density genetic maps with DNA marker based techniques will help in understanding
the nature of these QTLs, such as knowing the effects and chromosomal location of
gene/s affecting a trait, nature and effect of multiple copies of a single gene,
interaction between and among genes governing a trait, the pleiotropic nature and the
stability of gene action under different environmental conditions (Paterson et al.,
1988). QTL mapping has been very efficient for identifying the genetic regions linked
to quantitative agronomic traits, in conjunction with high-throughput genotyping and
phenotyping techniques. In addition, two or more QTLs identified in different
varieties or compatible species and responsible for different traits are introgressed into
the same elite line (Ashikari and Matsuoka, 2006) through marker assisted selection
(MAS). In this way, the molecular basis of a quantitative trait can be inferred.

3.3.PRACTICAL EXAMPLES

3.3.1. Example 1 - Selection of a single gene mutation: Ug99 rust

resistance in wheat

x In 1999 a new virulent race, Ug99 (known as race TTKSK), of the black stem
rust disease caused by the fungus Puccinia graminis appeared on wheat crops
in Uganda. The disease can cause complete failure of wheat crops.
x No wheat varieties carried resistance to Ug99 and the disease increased in
severity and since the spores are wind-borne began to spread rapidly.
x Ug99 accounts for 8.3M tonnes of lost wheat grain per year (US$ 1.23
billion), thus threatening the global multi-billion-dollars wheat production.
x In 2009 the growing concern of Ug99 led to the establishment of the IAEA
Technical Cooperation project INT/5/150 entitled: ‘Responding to the
Transboundary Threat of Wheat Black Stem Rust (Ug99), which involved
over 20 countries.

99
x This project has involved over 18 countries and 5 national and international
institutions and examined possible mutation induction treatments to deal
with the challenges posed by Ug99. Meetings and workshops to facilitate the
project efforts were held in Kenya and Turkey.
x Mutation induction treatments were carried out at the Joint FAO/IAEA Plant
Breeding and Genetics Laboratory (PBGL) in Seibersdorf (Austria) in 2009
by determining radio-sensitivity of each wheat variety and then irradiating
seeds at optimum dose levels for mutation induction. The treatment provided
enhanced biodiversity and treated seeds were sent for testing at Eldoret,
Kenya, a hot spot for the disease. IAEA support to Kenya also included the
establishment of irrigation systems, which allowed two generations of wheat
to be grown and tested per year.
x In 2013, 13 resistant advanced mutant lines were developed in wheat
varieties from 6 Member States (Algeria, Iraq, Kenya, Syria, Uganda and
Yemen).
x In February 2014 the first mutant wheat variety resistant to Ug99 was
officially released to farmers. It was named, “Eldo Ngano1” (Swahili for
“Eldoret Story1”) and combines disease resistance with high yield. The short
time taken from mutation induction to variety release (less than 5 years) is
unprecedented. Six tonnes of seed were raised in 2013 and multiplied further
at Eldoret, Kenya. In 2014 local farmers received sufficient seeds of “Eldo
Ngano1” for 400 – 500 hectares in the first year of release. “Eldo Ngano1”
has been followed swiftly by the mutant variety “Eldo Mavuno” (“Eldoret
Harvest”) which received Kenyan Government approval in May 2014.

100
TABLE 3.2. SUMMARY OF THE DEVELOPMENT OF “ELDO NGANO1” RESISTANT
MUTANT VARIETY: FROM MUTATION INDUCTION TO OFFICIAL VARIETY RELEASE

Date Development of variety

March 2009 Selection of parental variety: Chozi and Njoro II (Kenya)
April 2009 Mutation induction – gamma ray treatment at the PBGL, Austria
June 2009 – September 2009 First mutant generation (M1 bulk grown in Eldoret, Kenya)
October 2009 – February 2010 M2 population (about 10 000 head rows) for each variety grown at
Eldoret, Kenya (hotspot for the disease)
16 resistant plants selected from 5 head rows
May 2010 – September 2010 M3 selections grown as head rows at Eldoret, Kenya
5 resistant lines selected
October 2010 – February 2011 M4 selections re-tested by artificial infection of seedlings in glasshouse
tests, Eldoret, Kenya
Selected lines grown as field plots and confirmed as resistant
June – September 2011 M5 selected lines (head rows) grown in infected fields, Eldoret, Kenya as
plots
5 lines confirmed as resistant
October 2011 – February 2012 M6 selected lines grown in infected fields, Eldoret, Kenya
March 2012 – October 2012 M7 selected lines entered into official National Performance Trials
4 of 5 selected mutant lines (3 from Chozi and 1 from Njoro II)
2012 – 2013 Multiplication of seeds of 4 selected mutant lines, at Eldoret, Kenya
March 2013 Start 2nd year National Performance Trials
1 line dropped due to low yield, 3 continued, 1 new line included
August 2013 Official release of first approved variety: Eldo Ngano1

3.3.2. Example 2 - Induced mutations for genetic improvement of

quantitative traits in sorghum

Sorghum is the fifth most important cereal crop worldwide and grown on marginal
soils of arid and semi-arid regions. African and Asian countries are major sorghum
producing countries (83 percent of the area and 57 percent of the world production)
with a predominance of local landraces and varieties used mainly for food purposes
(Rakshit et al., 2014). In India, Southern and Central States account for 70 percent of
the total sorghum production. Most of the area is under rainfed conditions and local
landraces of the durra type are predominantly grown. They are tall, late maturing,
photosensitive and possess low harvest indices compared to irrigated cultivars. In
order to improve the locally adapted landraces for food and fodder qualities, biotic
and abiotic stress tolerance, attempts have been made to improve these traits using
mutation breeding methods. Various physical (X-rays, gamma rays, electron beams
and fast neutrons) and chemical mutagens (EMS, MMS, SA and NEU) and various

101
combinations of both types have been used to induce mutations in sorghum (Reddy
and Rao, 1981). Recently, carbon ion beam mutagenesis has been used to improve
stem juiciness in sweet sorghum (Dong et al., 2017). Gamma rays have been used to
induce mutations in quantitative traits in sorghum such as increased panicle size, grain
yield and fodder quality (Soeranto, et al., 2001); grain numbers/panicle, seed weight
and grain yield in Co-S-28 (Jayaramachandran et al., 2010), and Yezin -7 genotypes
(Htun, Min and Win, 2015).

Popular sorghum landraces such as Maldandi, TC-2, TH-11-10 and TJP-1-5 have
been grown widely for food and fodder purpose in southern Indian states. They
produce marginal yields with lustrous pearl yellow seeds and are susceptible to shoot
fly and charcoal rot. In 2012, a research programme was initiated to improve the local
landraces for quantitative traits using gamma rays and EMS. Briefly, 3000 self-
fertilized seeds of each landraces having 10-12 percent moisture content were selected
for induced mutagenesis. They were subjected to gamma irradiation of 300 Gy (dose
rate of 38 Gy/min.). Half of the irradiated seeds were used for additional EMS
treatment. 1 500 seeds were soaked with agitation in 250 ml water for 16 hrs and then
treated with 0.1 percent (v/v) of EMS for 8 hrs at room temperature. The treated seeds
were washed with tap water for four hours at room temperature and air dried. These
air-dried seeds were used for direct planting in the field, one seed per Hill plot with a
spacing of 45cm × 10cm. Gamma ray and EMS treated seeds were separately sown
in 15m × 3m blocks along with the untreated parents as control. Local cultural and
management practices were followed to raise a healthy crop in M1 generation.
Individual panicles were bagged before anthesis to ensure selfing, and harvested
separately. Each fertile panicle was planted as a head to row progeny in the M2
generation and harvested separately.

All morphological and other variants (albino, xantha, viridis and chlorina) were
recorded as the plants developed. In subsequent generations, 15 – 20 percent of the
elite variants with the most outstanding features were selected and forwarded as plant-
to-row progenies until M5 in comparison with control plants. In the M6 generation,
promising mutants showing ten days earlier flowering time were identified in TC-2
and TJP-1-5 populations with high grain yield (2600kg/ha against check 1600kg/ha),
large seeds (4.1 g/100 seeds against check, 3.5 g/100 seeds, see Table 3.3). Wide
variability was also observed for other quantitative traits such as plant height
(115 – 338cm), stem diameter (0.8 – 2.2cm), panicle length (8 – 34cm) and width
(7 – 25cm) in these mutants (Badigannavar et al., 2017). Bold lustrous pearl yellow
seeds are preferred by consumers and fetch premium prices in the market. Further,
induced mutagenesis of two parents TC-2 and TJP1-5 resulted in the identification of
five mutants for enhanced iron content (17 – 21 mg/100g) and four for zinc content
(2.5 – 4.1 mg/100g) compared to their parents. Currently, these mutant lines, TC-2
and TJP1-5 are being tested in official National Performance Trials of ICAR (Indian

102
Council for Agriculture Research) for post rainy season conditions. For more details
on chemical mutagenesis see Chapter 2.

TABLE 3.3. MEAN PERFORMANCES OF FOUR ELITE SORGHUM MUTANTS AT TWO

LOCATIONS OVER THREE SEASONS

Quantitative traits TC-2 TJP-1-5 TH-11-10 M-35-1 (check)

Days to flower 53 52 57 62
Panicle length (cm) 23.5 20.5 19.2 14.5
Panicle width (cm) 6.5 5.8 7.2 4.5
Seed weight (g/100 seeds) 4.0 3.95 4.0 3.6
Grain yield (kg/ha) 2700 2500 2450 1550

103
104
 4. MUTAGEN EFFECTS IN THE FIRST GENERATION AFTER SEED
TREATMENT: BIOLOGICAL EFFECTS OF MUTATION TREATMENTS

4.1.PLANT INJURY AND LETHALITY

As discussed in Chapter 3, the first mutation population (M1) suffers from

physiological disorders as a result of the mutagen treatment. This is a major reason
why phenotypic selection for mutation cannot be done in the M1 generation. In
addition, most induced mutations are recessive and therefore the mutant phenotype
cannot be observed until the mutation is homozygous. Moreover, the mutation
induced is originally a one-cell event and is not present in every cell of the plant. Thus,
M1 plants must be regarded as chimeric plants (see Section 4.4.). For practical
purposes the most important effects are growth retardation, sterility and death of the
M1 plants. Physiological disorders may be linked to chromosomal and/or extra-
chromosomal damage, but a separation of the two causes is usually not possible.
Regardless of these effects, the general weakened state of M1 plants usually means
that the M1 population should be grown in benign (stress-free) environments to
maximise growth, fertility and the production of the next (M2) generation. All effects
of the mutation treatment depend on the mutagen doses applied. For a review on the
effects of gamma irradiation on morphological, physiological, and biochemical
aspects in plants see Jan, Parween and Siddiqi, (2012).

In mutagenesis experiments, the resulting physiological damage sets practical limits

to dose increase; in fact, an end point is reached with 100 percent lethality of the
irradiated material and/or no germination of the M1 seeds. In the case of seeds, the
disruption and disorganisation of the tunica, or seed layers, or operculum (germination
pore) or impaired mitosis or even complete elimination of cell division have been
described as causes of failed germination after gamma irradiation (Lokesha et al.,
1992). Gamma irradiation of seeds also affects the survival rates of plants at maturity
in a dose dependent relationship, this is also true to a lesser extent for EMS
mutagenesis (Mahamune and Kothekar, 2012). Plant death might occur at any time
between the onset of germination and maturity; however, there are critical phases
during plant development at which lethal effects are more prominent, and M1 plants
are frequently sterile because of aberrations at meiosis, pollen, embryo sac or seed
development (Micke and Wöhrmann, 1960). Therefore, when reporting lethality rates
or survival percentages it is essential to note the time and stage of plant development
at which data are collected. Survival rates obtained under laboratory conditions may
differ considerably from those observed under field conditions, due to the possible
occurrence of environmental stress at critical phases of the plant development.

105
At the chemical level, the generation of reactive oxygen species (ROSs) is a major
factor responsible for physiological disorders following mutation induction
treatments. ROSs have toxic properties and cause oxidations of poly-unsaturated fatty
acids in lipids (lipid peroxidation) or oxidative deactivation of specific enzymes and
may also cause damage on DNA or RNA (Figure 4.1). In addition, ROSs may trigger
transcription of specific genes as a response to the stress caused by the treatment. This
is currently discussed as one of the reasons for the often-observed stimulated
germination rate at lower irradiation doses (in the range of about 10 to 50 Gy,
depending on the species). An increase of assimilatory pigments, chlorophyll a,
chlorophyll b, and carotenoids, in response to increasing irradiation dose has also been
described (Marcu, Cristea and Daraban, 2013). The stimulating effect of gamma rays
on cell division, growth and development is called “hormesis”. It should be noted,
however, that hormetic effects (those that surpass normal performance of non-
irradiated controls) are usually minor and short-lived, and do not cause significant
increases in yield (Miller and Miller, 1987). With increasing doses, the inhibitory and
toxic effects of ROSs as well as the direct impact of ionizing radiation on DNA prevail
over hormesis stimulation (Marcu, Cristea and Daraban, 2013).

Generation of ROSs is dependent on the water content of the irradiated tissue; this is
one reason in vitro explants or tissue cultured materials are irradiated at much lower
dosages as compared to (dry) seeds. The water content of seeds at the time of
irradiation plays an important role as regards to the effect of the mutagen, and it is
normal practice to standardise this by placing seeds in a desiccator. The higher the
water content in the seeds, the greater the susceptibility to ionizing radiation. This can
clearly be perceived in radio-sensitivity tests using seeds adjusted to various water
content levels. Therefore, it is imperative to adjust the water content of seeds before
mutation induction, normally to about 12 – 15 percent (for further details Chapter 1).

For any mutation induction in plant breeding a mutagen at a given dose should be
applied that results in low plant injury, but induces a workable frequency of mutation
induction for target traits with a low mutational (background) load. A balance is
therefore needed, which is often determined by the size of the M1 and subsequent
mutant populations that can be handled by the breeder. For any given mutagenic
treatment there is a correlation, for example, in cereals between M1 seedling height
and survival on one hand and M1 mutation frequency on the other Gaul, (1959) as
cited by (Suresh et al., 2017).

The quantitative determination of M1 injury and radio-sensitivity testing should be

routine procedures in mutation breeding programmes. The easiest and most routinely
applied methods to assess M1 injury are: 1) the determination of germination
frequency, and 2) shoot and/or root growth rates at a particular plant developmental
stage (usually in seedlings). These tests are typically performed in the laboratory or a

106
greenhouse. Parameters that can be measured and compared to the control (untreated)
under field and/or laboratory conditions include, germination, seedling establishment,
height, root growth, leaf colour, flowering time, flower production, seed set and yield
per plant (these are discussed in detail in Chapter 5).

4.2.CYTOLOGICAL EFFECTS

Some mutagenic treatment effects can be observed at the cytological level and the
frequency of chromosomal aberrations in the M1 can be determined. After treatment
of seeds, the analysis of the first mitotic cycle in shoots or root cells offers a quick test
to determine the effect of the mutagen (Figure 4.2). This test is more laborious than
measuring seedling traits, e.g. height reduction; however, it adds additional
information and should be applied whenever a new mutagenic treatment is introduced
in a breeding or research programme. In many crops seedling roots are routinely used
for chromosome studies and protocols are well established to examine metaphase
chromosome spreads (see e.g. Maluszynska, 2003).

107
Figure 4.1. Direct and indirect action of gamma and X-rays on DNA. Top: reactive oxygen species
(ROSs, here: hydoxyl radical) generated through ionizing radiation causing DNA strand breaks.
Bottom: strand break is a direct effect of ionizing radiation on DNA.

Figure 4.2. Chromosomal aberrations in onion Allium cepa L. induced by gamma irradiation. (A) Non-
congression onto metaphase plate. (B) Laggards (C) Chromosome bridges. (D) Interphase with micro-
nucleus. (Images are taken from Kumar et al., (2011) ; copyright © University of Florence, reprinted
by permission of Taylor & Francis Ltd, www.tandfonline.com on behalf of University of Florence,
Italy).

108
 4.2.1. Chromosome observations

Chromosomal aberrations can be visualized in mitotically active tissues (root tips,

leaf) of young seedlings derived from mutagenized seeds (Figure 4.2). In anaphase
cells chromosome bridges and fragments are observed and interphases often show
micronuclei as an effect of the treatment. For examples see Maluszynski et al., (2009)
and Kumar et al., (2011).

Mutagens often delay the germination of treated seeds; they may also delay the onset
of cell divisions and slow down the mitotic cycle. This fact needs to be taken into
consideration when root or shoot tips are to be treated and fixed for cytological
examination. In barley the nuclear divisions start in the seminal roots; they are
followed somewhat later in the coleoptile, leaves and apical bud of the shoot (Wertz,
1940).

4.2.2. The comet assay

The comet assay is an elegant method of measuring the degree of DNA strand breaks
and thereby chromosomal aberrations after mutagenic treatment (Figure 4.3). In
practice, nuclei are released from young leaves by chopping with a razor blade in cold
Tris-HCl buffer. The nuclei are then embedded on agarose coated slides. By treatment
with high-salt lysis solution the membranes and nucleoplasm are removed, the
nucleosomes disrupted and the histones solubilized. The remaining nucleoids with
negatively super-coiled DNA are subjected to electrophoresis. After electrophoresis
and staining with a fluorescent dye, such as ethidium bromide (EB) or 4.6-diamidino-
2-phenylindole (DAPI) the slides are observed under a florescence microscope using
UV excitation. DNA breakage leads to a loss of super-coiling and enables the debris
to migrate towards the anode thereby forming a structure resembling the tail of a
comet. Unaffected super-coiled DNA in contrast does not move and forms the head
of the comet. There is a clear correlation between the length of the comet tail, the
DNA damage and the dose applied for mutation induction. For a critical review on
the comet assay, its principle, applications and limitations, see Collins, 2004.
Interestingly, the comet assay allows monitoring of the kinetics of DNA repair by
analysing samples taken at certain times after treatment. Gichner et al., (2000)
irradiated tobacco seedlings with low dosages of gamma rays (20 or 40 Gy) and
determined complete DNA repair after 24h. An interesting modification of the assay
is the application of fluorescent in situ hybridization (comet-FISH). This allows, in
addition to quantitative measurements, a qualitative evaluation on the chromosomal

109
regions that have been affected by mutagenic treatment. For example, using comet-
FISH, Kwasniewska and Kwasniewski (2013) showed that telomeric DNA sequences
are more frequently present in the comet tail than centromeric ones, which is
explained by the different number of breaks needed for the chromosomal aberrations:
one break for a deletion at the end of the chromosome, two breaks for a deletion in a
centromeric region.

Cells with damaged (relaxed) DNA The damaged DNA is separated

having single-strand / double- from the intact DNA (the “head”)
strand breaks and
an generates a comet “tail”
il”
- +
DNA Damage
(Chemical / UV /
Ȗ-irradiation)

Single cells are embedded on After electrophoresis and

A
agarose-coated slide and lysed
- fluorescent staining
+
Living cells
from culture
media or plant
tissue

Cells without mutagenic treatment Control cells not forming a comet

(control) “tail”

Figure 4.3. Comet assay – principle (modified for plant tissue, http://www.sigmaaldrich.com).

4.2.3. Low dose stimulation

The low dose stimulatory effect of gamma irradiation causes an increase of mitotic
activity, but increasing the dose also reduces cell division. For example, in cowpea
chromosomal abnormalities analysis (Badr, El-Shazly and Halawa, 2014; Kozgar
et al., 2014) at metaphase, four types of induced abnormalities were observed:

x c-metaphase configurations, with complete inhibition of spindle fibre

formation;
x chromosomal stickiness where the chromosomes appear clumped together;

110
 x disturbed metaphase configurations and;
x polyploid cells, however less frequently.

At anaphase-telophase chromosome bridges and lagging chromosomes were

observed. In chromosome preparations of irradiated ginger rhizomes, Kamaruddin
et al. (2016) observed clumping of chromosomes at prophase, metaphase and
telophase; and sticky chromosomes were detected at anaphase and metaphase.

4.2.4. Effects on meiosis

Meiosis is a sensitive stage in plant development and is influenced by both genetic

and environmental factors (Wijnker and de Jong, 2008). The control of meiosis is a
“holy grail” for plant breeding but has rarely been achieved in practical conditions
with the exceptions of the use of apomixis for seed production (e.g. Brachiaria forage
grasses), through bridge crosses, and polyploidization. Mutagenic treatments
(physical, chemical and biological) affect meiotic chromosome pairing and
recombination, and can induce chromosome translocations in plants, sometimes with
desirable results (Puchta, Dujon and Hohn, 1996; Lagoda et al., 2012). Mutagen
treatments targeting meiosis have been used to induce alien chromosome segment
introgression into crop species (Wang et al. in Shu et al., 2012). For example,
irradiation at meiosis of aneuploid stocks of wheat carrying additional chromosomes
of: 1) T. umbellulatum carrying rust resistance – X-ray irradiation (Sears et al., 1956)
and 2) Leymus racemosus carrying scab head blight resistance – gamma irradiation
(Chen et al., 2005), were used to induce chromosome breaks and provide
opportunities for unions with other chromosomes in creating wheat/alien translocation
lines. The irradiation induced double-strand breaks and the repair mechanism
produced non-homologous chromosome recombination. Sadly, there are few
examples of irradiation treatments to manipulated meiosis and recombination; this
may be due to the ultra-sensitivity of the meiotic processes. More research is needed
in this area.

4.3.STERILITY

Mutagen-induced reduction of reproductive capacity encompasses various

phenomena and various sources. The phenomena include: (1) severe stunting or
growth inhibition which prevents flowering; (2) flowers are formed but lack
reproductive structures; (3) reproductive structures are present but pollen and/or
ovules abort; (4) fertilization occurs but embryos abort before maturity; and/or (5)
seeds form but fail to germinate properly or die after germination. Most common is
the occurrence of non-functional gametes as a result of failed meiosis. Mutagen-
induced sterility may be caused by (1) chromosome mutations, (2) gene mutations (3)
cytoplasmic mutations, and (4) physiological effects. Chromosome mutations are

111
probably the major origin of all mutagen-induced sterility. As with induced
physiological disorders, sterility issues set practical limits on increasing dose
treatments.

Following chemical (EMS) treatment in Arabidopsis, it was observed that the

mutation frequency was limited by sterility in M1 plants, and not by the increased M1
lethality. Several mutants were studied in detail with defects that included the
establishment of anther morphology, microspore production, pollen differentiation,
and anther dehiscence, leading to male sterility (Sanders et al., 1999). Several reports
have presented data on genes involved in mutation induced sterility for both pollen
(male) and ovule (female), (Sanders et al., 1999; Robinson-Beers, Pruitt and Gasser,
1992). This is an issue for breeding as, although, plant breeders want to produce a
reasonable number of lines carrying a desired mutation, they do not want a high
background mutational load as the target material is usually an elite line with a finely
tuned and valuable genetic background, and the mutational load requires lengthy
backcrossing procedures to clean up.

While, sterility may be an issue in mutant line development, it can also be exploited,
especially in the production of F1 hybrids. In the context of hybrid breeding the
development of male sterile plants (female lines) is highly desirable in commercial
production of hybrid seeds. Chaudhury, (1993) in a review on male fertility in plants
has described various male-sterile mutants and concluded that several genes are
involved in controlling male fertility.

4.4.CHIMERAS

In a simple definition a chimeric plant is a plant that is composed of cells harbouring

more than one genotype. In Chapters 2, 3, 5, 6 and 8 the generation and dissolution of
chimeras are discussed in detail. The occurrence of chimeras after mutagen treatment
is of great importance for the implementation of a mutation breeding programme, in
particular with regards to the handling of the mutated populations.

L1 - Tunica
Central Zone
L2

L3 - Corpus

Figure 4.4. Organization of cell layers in the apical dome of a shoot meristem.

112
Chimera types are broadly classed as either somatic or reproductive (see Chapter 5).
The most visible indication of somatic chimerism is chlorophyll variegation in leaves
with chlorophyll deficient sectors forming longitudinal streaks in monocotyledons
and irregular patches in dicotyledons. Chimeras can persist in vegetatively propagated
crops (VPCs) but may be dissolved quickly in seed crops. However, even in seed
crops chimeras can be transmitted to the next generation at a low frequency. The
segregation of chimeras in M2 plants depends on the number of initial cells giving rise
to the germline (genetically effective cells) and thus to the formation of the gametes,
as well as on the mutations induced or not in each individual cell. For more
information see text box 4.1.

Text box 4.1.

For barley it has been established that each of the main 4-5 spikes originate
from 2-4 initial cells (Gaul, 1964). Ukai and Nakagawa in (Shu et al., 2012)
presented an example of mutant segregation using the M1 spike method that was
previously developed by Stadler (1928) and is based on separate harvests of M2
seeds from each spike of M1 plants. The segregation frequency of mutants in the
M2 then appeared to depend on the number of initial cells that gave rise to a
spike. In case of only one initial diploid mutated cell the segregation frequency
in the M2 is equal to the segregation ratio of a recessive mutant (a’a’), namely
0.25. If in addition to the mutated cell, more cells are involved in the
development of the spike, it will have a chimeric sector and the segregation
frequency in the M2 will be lower than 0.25. The relationship between initial
cells and segregation frequency can be described by the formula 0.25/k, with k
being the number of initial cells. This means, in practical terms that with a higher
number of initial mutant cells the chances of finding a homozygous recessive
mutant in the M2 becomes lower. It should be noted that mutagenic treatments
can kill initial cells and that more than one initial cell can be mutated differently.
In the latter case the breeder will obtain a larger number of M2 mutant plants
with different characters. An additional influence on the segregation frequency
is given by so-called diplontic- or haplontic selection, which refers to the
competition among cells with different genotypes, specifically between mutated
and non-mutated cells, either in the diplophase between somatic cells or the
haplophase during gamete production and forming the zygotes (see Chapters 3,
5, 6 and 8 for more information).

In vegetatively propagated crops (VPCs), the issue of chimeras can be more

problematic as it takes considerable effort to eliminate chimeras and obtain
homohistont plant material (see Chapter 6).

Chimeras in VPCs result from mutations generated in the shoot apical meristem. The
specific nature of chimeric results from mutations in the shoot apical meristem known
to be organized in layered structures (Figure 4.4). Shoot apical meristems of

113
angiosperms typically have three distinct layers of cells, L1, L2 and L3. Layer L1 is
the single layer of cells making up the epidermis, layer L2 the single cell sub-
epidermal layer that gives rise to the sub-epidermal mesophyll cells in the leaves and
to the gametes and layer L3 constitutes the rest of the internal cells including the
vascular system. The type of chimera depends from the layer in which the mutation
was induced: sectorial-, mericlinal- and periclinal chimeras (see Chapter 8-A). Plants
that have one layer of cells that is genetically distinct from another layer are called
periclinal chimeras. Periclinal chimeras are the most stable type of chimeras and are
derived from a mutation in a single cell of the layer that through anticlinal division
forms a uniform layer bearing this mutation. In a mericlinal chimera a mutation in one
layer does not spread over the entire layer, these are unstable and often loose the
mutation or develop into periclinal chimeras. Sectorial chimeras are characterized by
a mutated sector being spread over more than one layer. This type of chimeras is
unstable and often develops into shoots which are not chimeras and may give rise to
genetically homogenous types. An art in mutation breeding of VPCs, such as the
triploid Cavendish banana, is the dissociation of chimeras after mutation induction,
with the aim to develop homohistont clones that can be screened for the trait of interest
as soon as possible. (Roux et al., 2001a) tested different tissue culture methods for
chimera dissociation in diploid and triploid Musa acuminata and used polyploid
cytochimeras as a model system for mutations, which can be easily followed using
flow cytometry. Three different techniques were applied for three subcultures each
(shoot tip culture- ST, multi-apexing culture - MA, and corm slice culture- CS). The
average percentage of cytochimeras was reduced from 100 percent to 36 percent using
ST, from 100 percent to 24 percent when propagating by CS technique and from 100
percent to 8 percent using the multi-apexing technique.

In the case of chimeric sectors in ornamental horticultural plants with attractive

mutations such as new flower colour or shape, the mutant tissue can be isolated and
subjected to tissue culture techniques in order to regenerate plants that carry the
favourable mutation. This has been shown for example by (Mandal, Chakrabarty and
Datta, 2000) who regenerated mutated sectorial floret tissue from chrysanthemum
through organogenesis thereby retaining the mutation.

4.5.SECONDARY EFFECTS: TRANSPOSON ACTIVATION

The vast majority of variation induced by all forms of mutagens is a loss of gene
function, and hence most mutations are recessive. In contrast, transposable element
(TE) insertions can produce a much broader range of phenotypes (Lisch, 2013). TEs
also called “jumping genes” are endogenous mobile elements able to move and insert
themselves at different positions in the genome. In doing so, they often produce
mutations when they re-insert into genes. In contrast to other sources of genetic
variation TE insertions have the potential to induce homeotic transformations,

114
dominant ectopic expression mutations and environmentally induced changes in gene
expression (Naito et al., 2009; Lisch, 2013). There are several well documented
examples of a direct involvement of these endogenous mutagens in plant
domestication and improvement. Importantly, several of these mutations would not
have arisen via simple point mutations or insertions and deletions (INDELs). This is
due to the regulatory sequences carried by TE insertions, as well as epigenetic
regulation of TEs, that can alter the expression of neighbouring genes. Thus, although
they can certainly create null mutations, TEs have the additional capacity to create a
new spectrum of regulatory mutations.

Stress, including radiation exposure can activate otherwise silent TEs (Bui and
Grandbastien, 2012; Bradshaw and McEntee, 1989; Sacerdot et al., 2005;
Farkash et al., 2006; Qüesta, Fina and Casati, 2013), see also: (https://www.cospar-
assembly.org/abstractcd/COSPAR-12/abstracts/F4.6-0009-12.pdf).

The involvement of TEs activated by ion beam treatments in the induction of flower
mutants has been documented by Okamura et al. (2006). At the Joint FAO/IAEA Plant
Breeding and Genetics Laboratory, Seibersdorf, Austria, molecular analysis of rice
plants derived from radio-sensitivity tests indicated that the retroelement Tos17 was
activated by gamma irradiation of rice seeds (Nielen, Guzman and Zapata-Arias,
2000). These results indicate that TEs have a major role in mutation induction, the full
potential of exploiting TEs in mutation breeding has not yet been realised.

4.5.1. Examples of TE-induced mutations in plant breeding

There are a number of success stories in TE-induced mutation in crop improvement.

For example, the reduced branching, associated with maize domestication from
Teosintes, is due to the introduction of an enhancer element carried by a
retrotransposon that inserted upstream of the tb1 gene (Studer et al., 2011). The
temperature-sensitive expression of red pigment in the flesh of blood oranges is due
to regulatory elements encoded by a retrotransposon inserted upstream of a colour
gene (Butelli et al., 2012); this provides a useful dominant gain-of-function mutation
in a species propagated by cuttings or by apomictic seeds. Interestingly, a secondary
rearrangement leaving behind a small portion of the transposable element, leads to an
intensification of that red flesh colour.

The propensity for TEs to mobilize coding sequences via retro-transpositions lead to
a change in tissue specificity in the IQD12 gene in tomatoes (Solanum lycopersicum),
which is responsible for the oval shape of most modern tomatoes (Xiao et al., 2008).
The white colour of grape berry (Vitis vinifera) is also due to an insertion of a
retrotransposon upstream of a Myb-related gene that regulates anthocyanin
biosynthesis (Kobayashi, Goto-Yamamoto and Hirochika, 2004) and the phenotype

115
is only partially reversed to a pink berry colour after internal recombination leaving a
solo-LTR at the insertion site (Pelsy, 2010).

Thus, TE insertions can exert various subtle regulatory changes that can produce
beneficial traits for plant breeding purposes. Adaptive changes mediated by TE
insertions have also been reported. A retrotransposon insertion in the soybean
phytochrome A photoreceptor gene GmphyA2 conferred photoperiod insensitivity,
which has contributed to the expansion of soybean cultivation to higher latitudes of
East Asia (Kanazawa et al., 2009). Also, the Pit disease resistance gene has been
transcriptionally reactivated in a resistant rice cultivar because of an insertion of a
new promoter provided by a retrotransposon (Hayashi and Yoshida, 2009).

Several examples of similar adaptive mutations caused by recurrent insertions have

also been reported including the peculiar ‘hose-in-hose’ primrose mutant (Text Box
4.2). These are important because they suggest that without the presence of TEs these
mutations may not have been recovered. In a line of hexaploid bread wheat, the Vrn3
vernalization gene is upregulated in correlation with the upstream insertion of a
retrotransposon, allowing early flowering after a vernalization treatment, thus
conferring Spring growth habit (Yan et al., 2006). In tetraploid durum wheat, the exact
same phenotype is controlled by the upstream insertion of a different retrotransposon
in the homoeologous Vrn1 gene (Chu et al., 2011). Aluminium tolerance, in wheat,
sorghum and barley involve TEs. Aluminium tolerance is associated with upstream
insertions of different TEs that mediate up-regulation and redirection to the root apex,
of the expression of citrate transporter genes (Delhaize, Ma and Ryan, 2012).

There is evidence that TEs can not only respond to stress, but also confer stress
tolerance. For instance, the retro-element insertion that causes the colour change in
blood orange also confers responsiveness to cold temperatures. Similarly, insertion of
the miniature inverted-repeat transposable element (MITE) in rice can confer
tolerance to cold and salt stress (Naito et al., 2009). The same is likely true of a wide
range of other biotic and abiotic stresses, which is of interest in adapting crops to
environment change.

TEs are normally kept in check via a sophisticated system of epigenetic silencing. One
of the most interesting aspects of the responsiveness of TEs to stress is that it can
result in a reversal of that silencing, which makes the TEs available as a source of new
variation. Tos17, for instance, is a retro-element in rice that is transiently reactivated
in tissue culture and has been used to generate tens of thousands of new mutant alleles
(Piffanelli et al., 2007). Similarly, the En/Spm element in maize was released from
silencing due to exposure to radiation in the 1950s (Peterson, 1991). Recent evidence
suggests that UV-B and perhaps ion beam irradiation also trigger TE activation
(Yan et al., 2006; Huiru et al., 2009; Ya et al., 2011; Qüesta, Fina and Casati, 2013).

116
 Text Box 4.2.

Other examples of TE-induced mutation include the much prized "hose-in-

hose" (Figure 4.5) primrose flower phenotype, which shows a conversion of
sepals to petals, resulting from changes in tissue-specificity of a MADS box
gene that is associated with a gypsy retrotransposon insertion in its promoter
(Li et al., 2010), and the hose continuous flowering phenotype (blooming in all
seasons) is caused by an intronic insertion that results in splicing failure of a
gene controlling flower transition, a characteristic under photoperiodic and
thermal control (Iwata et al., 2012). Recombination of the retrotransposon
leaving a solo-LTR, restores normal splicing, but the resulting phenotype is
not the wild-type phenotype (Spring blooming), but a climbing phenotype
(occasionally re-blooming in Autumn).

Hose-in-hose Semi-revertant
nt Revertant

Figure 4.5 Hose-in-hose mutant of primrose, Primula vulgaris

TE activity is far less deleterious, on average, than random mutagenesis because TEs
have been selected through evolution to minimize negative effects (Naito et al., 2009).
Furthermore, if a specific TE family is known to be activated in a particular species,
new mutations caused by this TE family are thus tagged, which could make molecular
identification of the cause of a new phenotype relatively simple (compared to DNA
walking to a point mutation). An additional potential advantage of TE-enhanced
radiation-induced mutagenesis is that the level of radiation needed to activate TEs
may be far lower than that required to produce large numbers of mutations directly
from irradiation exposure, thus minimising mutational load. Presumably, this would
lead to fewer gross chromosomal rearrangements that could hinder, for instance,
subsequent introgression experiments. Finally, TEs can be exploited as polymorphic
DNA markers in measuring biodiversity studies and genotypic constitution in plant
breeding germplasm.

Acknowledgement: this section was drafted as an outcome of the FAO/IAEA

consultants meeting on “Enhancement of the efficiency of mutation induction by
physical and combined mutagenic treatments”, held 2 – 6 June 2014 in Vienna,
Austria, for which we are grateful.

117
118
5. MUTATION BREEDING IN SEED PROPAGATED CROPS: PARENTAL
SELECTION, MUTANT GENERATION DEVELOPMENT, MUTATION
DETECTION, MUTANT EVALUATION AND FACTORS INFLUENCING
SUCCESS

This chapter is an adaptation and up-date of the Chapter 7: “Induced mutation

techniques in breeding seed propagated species” by Konzak et al. 1977 in the 2nd
edition of the Manual on Mutation Breeding. It deals with the practicalities of
mutation breeding in seed propagated crops and uses annual cereal and legume species
as examples. The processes of developing mutant populations, screening and selection
of mutant plants and populations are described starting with the selection of the
target/parent genotype to be mutated, i.e. the production of M0 seeds for treatment,
irradiation to produce M1 seeds and the handling of mutant generations, with emphasis
on M1 to M3 generations. The chapter introduces the main factors affecting the success
of mutation breeding such as population size, propagation and isolation of mutant
plants and examples of screening for desired mutants. Schematic diagrams and
photographs illustrate the various steps and assist in the understanding of the practical
application for mutation breeding.

5.1.SELECTING PARENTS AND HANDLING M1 TO M3 GENERATIONS FOR

MUTANT SELECTION

A successful mutation breeding programme starts with well-defined objectives for

improvement of a defined plant phenotype/genotype. Common targets are:

a) to improve one or a few specific traits of a preferred variety or an elite line

(see Chapter 7);

b) to induce a morphological marker (colour, awns, bracts, hairiness, etc.) in

order to establish distinctness in a promising line to make it easy to identify
and meet the requirement for variety registration;

c) and/or to induce male sterility or fertility restoration making a line useful as

a component for hybrid variety production.

5.1.1. Selection criteria for parent genotype

The parent genotype in which mutation is to be induced should be either: 1) a recently

released cultivar, 2) an advanced promising line about to be released, or 3) a promising
advanced line or introduced variety restricted from release by specific limitations, e.g.
susceptibility to lodging, to a specific disease or pest, and to shattering, etc.

119
A primary consideration in selecting the parental variety seed source is its lack of
natural variability for the trait(s) to be introduced by mutation. The chosen variety
should have sufficient uniformity in most of the important agronomic traits. Usually,
a stock of breeder or foundation seed is available from a recently released cultivar,
and in accordance with most pure seed programmes; the genetic variability for
agronomically important features is restricted to the levels outlined in local
registration schemes, e.g. > 98 percent true to type. Commonly, however, to fully
exploit an advance made by cross breeding, the mutation breeder may need to initiate
his experiments bearing in mind that the parental material may carry some
contamination via admixtures or outcrossing. However, for plant-breeding purposes
this is of minor consequence and the non-availability of a conclusively uniform and
pure stock need not deter the use of mutation induction. In such cases, to increase the
chances for success it is especially important that the breeder: 1) limits his/her
objectives, 2) restricts further introgression or genetic variability, 3) grows sufficient
untreated M0 (control) material to establish a concept of the genetic variation present
in the stock and for increasing the purity of the parent stock selected for back-up
experiments and, 4) develops a detailed information sheet on the characteristics of an
individual or line as improved by the type of mutation sought. For some crops, genetic
pure lines may be available in the form of doubled haploid populations.

5.1.2. Planning for the M1 generation

5.1.2.1.Radio-Sensitivity

Once the decision on the genotype to be mutated is made and a homogenous seed
stock is available, the next step is mutagenesis. The seeds before treatment are known
as the M0 generation. As explained in Chapter 1, radio-sensitivity testing should
precede the bulk treatment to determine the optimum dose levels for mutation
induction. Radio-sensitivity tests are normally performed in the greenhouse, but fields
can also be used. Responses to irradiation vary between species and among varieties
of the same species. Figure 5.1 shows typical radio-sensitivity test settings in a bean
(cowpea) and a cereal (maize) after seed irradiation.

5.1.2.2.Control population

A control (untreated) population should always be grown to serve three purposes:

a. provide a comparison of the treatment effects on germination, growth,

survival, M1 injury and sterility;

b. assess the phenotypic variability of the parent genotype stock used to produce
M1 and;

120
 c. provide a 're-purified' parent genotype as a back-up for initiating a new M1
generation to be grown during the same season with the M2 generated from
the first M1 if needed.

5.1.2.3.Mutagen and dose treatment

It is advisable to use three doses of the chosen mutagens, which should be ± 20 per
cent of the optimal dose found through the radio-sensitivity tests. Normally the
selected doses to be applied on cereals cause between 30 to 50 per cent reduction in
seedling growth in laboratory tests. In practice, at least two replicates should be made,
each with half the quantity of seeds selected for use with each dose, as an insurance
measure against failure and against errors. When rather large numbers of seeds are
treated, i.e. 5000 to 10 000 seeds per treatment, subdivision into several treatment
replicates will improve the uniformity of the treatments and this is usually convenient
because of the physical limitations of the irradiation chamber size.

5.1.2.4.Population size of Ml

Assuming a 90 per cent probability of success in recovering a mutant occurring at a

frequency of 1 × 10-3 per test unit (e.g. per spike, fruit, or branch), and that each plant
grown is expected to produce three units, the amount of seeds to be treated, if the Ml
has 80 per cent survival rate, would be about 600 seeds. However, as described below,
because of the errors in estimating the frequency of the desired mutation, uncertainties
in predicting Ml survival, variation in treatment severity, etc., it is recommended that
as much as ten times the estimated required

quantities of seeds be treated to ensure that an adequate population will be available

for screening. Thus, in the above example treating about 6000 seeds might yield the
breeder as many as 10 mutations in the desired direction.

While calculations to estimate the size of a treated population needed for isolating a
desired mutant type are recommendable, such calculations may be of limited value in
practice for several reasons, among which:

a) even the best estimates of mutation frequencies may involve considerable

errors;

b) the optimum mutagen treatment for obtaining a desired frequency of

mutations cannot yet be ensured since many unforeseeable physical and
biological factors may influence the mutagenic effectiveness of a treatment;

121
 c) and the breeder can hardly achieve his goal with a single desired mutant, thus
it is preferable to have several mutants with similar phenotype which will
provide option for further evaluation and selection of the best mutant.

In gamma irradiation the activity of the source affects mutagenesis and this declines
over time. The Ml generation normally requires the least space and effort. Therefore,
growing extra Ml material has little effect on cost, but adds to the assurance that a
sufficiently large population of M2 will be available for screening to obtain the desired
mutant(s).

A1 A2

1 2 3 4 5 6 1 2 3 45 6

B1 B2

1 2 3 4 5 6 1 2 3 45 6

Figure 5.1. Photographs of radio-sensitivity: A) in cowpea (beans) and B) maize (cereals) showing
survival rates and reduced growth in two-week-old seedlings after gamma irradiation of seed at: 1) 0
(control), 2) 75, 3) 150, 4) 300, 5) 450 and 6) 600 Gy. The source of gamma rays was a cobalt 60
gamma cell producing150 Gy/min. Courtesy of A. Mukhtar Ali Ghanim.

5.1.3. Sowing of M1 seeds

Considering the detrimental effects of mutagens on seed viability (see Chapter 4), the
M1 must be handled with more care than untreated controls. The M1 should therefore
be grown in benign conditions.

122
 5.1.3.1.Greenhouse conditions

If possible the M1 should be grown in a greenhouse where careful attention can be

paid with respect to watering, fertiliser provision, lighting and temperature, weed, pest
and disease control to maximise plant survival and production of the next generation.
Isolation of the M1 population is also easier in the greenhouse, thus limiting
pollination from external sources, which could bring in unwanted (non-mutant)
variation. Normally only a few (M2) seeds are needed per M1 plant and multiple
flowering units (e.g. tillers) may be reduced by growing individual plants in small
pots or collectively in trays, which also maximises greenhouse space. However, it
should be noted that the use of greenhouses is relatively expensive compared to the
field.

5.1.3.2.Field conditions

If greenhouse conditions are not available or are not affordable, then field conditions
can be used. It is particularly important to ascertain that the moisture and the physical
conditions of the seed bed prepared for sowing the M1 are optimal for seedling growth
and development. Nitrogen fertility of the soil should be normal or slightly lower to
limit excessive tillering (as normally only one seed head is harvested), but other
nutrients should be at optimum levels. A restriction in field cultivation is that it is
restricted to the selected variety’s cropping season.

5.1.3.3.Time of sowing M1

The M1 material will develop optimally if sown during the season when the climate
is best for early seedling and plant development and weed control is less of a problem.
However, slightly later sowing (2 or 3 weeks) may help reduce tillering and may
improve the conditions for isolation against cross-pollination. The delay should not
be so much as to encourage weed types that are more difficult to control, alter the
maturity in response to day length or temperature factors, or increase susceptibility of
the crop to other hazards. Because of the induced injuries, crops requiring
vernalization should be vernalized in the laboratory and transplanted or sown in the
very early season when vernalization would normally occur from low temperatures.
Where possible isolated facilities (greenhouse, protected nursery, etc.) can be used to
propagate the M1 population and minimize the losses.

5.1.3.4.Condition of the treated M1 seeds

Dry seeds are easier to plant with machinery or by hand, and a more uniform growth
can be obtained without extra care. If adequately low in moisture, dry seeds can be
stored for some time prior to sowing. Vacuum packing is the best option, if available,

123
to store treated seeds for longer time in order to match with the normal growing
season.

5.1.3.5.Density of sowing

In general, the spacing of M1 seeds within and between rows should be such as to
restrict development to primary tillers; to 2 – 3 in cereals, and the primary branching
in grain legumes and other dicotyledonous species. This can also be adjusted based
on the space available, the number of treated M1 seeds, the expected survival rate
based on the dose effect and the expected M2 population size.

5.1.3.6.Weed control

Normally, a relatively weed-free seed bed should be prepared just prior to planting of
the M1 seeds. Pre-emergence herbicides recommended for the area and crop may be
used successfully. Contact types of post-emergence herbicides may be useful for weed
control if the M1 plot is too large or the weeds are too numerous for manual control.
Systemic herbicides, such as 2, 4-D should normally not be used with cereals because
they tend to cause side-effects and may more severely affect one portion of the
population relative to another, often resulting in sterility, distortion of plant
phenotypes and reduced production of seed on the primary tillers of the M1 plants.
Other measures such as mulching can also be followed based on the crop and the area
of plantation.

5.1.4. Isolation of M1 material

Generally, it can be assumed that some level of genetic heterogeneity is always

present even in populations of parent material from self-pollinating plants. Cultivars
often encompass a group of related lines and advanced lines in a bulk population
derived from F5 or later generation. In addition, the probability is high that there will
be some limited contamination through mechanical mixing or by outcrossing.
Therefore, several disrupting hazards may be expected, especially in field plantings,
which may affect the certainty of the origin of the variability observed in a mutation
breeding programme as presented below.

a. Outcrossing – distant pollen grains may be transmitted by wind or insects

from varieties of the same species growing nearby; the extent of the
contamination varies with crop variety, the treatment, the mating system of
the plant species and the distance from the mutated crop field in relation to
wind directions and potential sources of contamination. Appropriate selfing
method, such as bagging to shield flowers before anthesis, should be applied
at the right time to prevent outcrossing (Figure 5.2). In cross pollinated
species, where male and female flowers are separated, selfing can be

124
 accomplished by hand pollination of the female flower. The female flower
should be covered before opening to present cross pollination.

b. ‘Volunteer’ crop – M1 should not be sown on land previously used for the
same species this is particularly important for hand-seeded legumes.

c. Bird damage – the risk of loss of M1 material to birds is often greater than
with non-mutated plant materials because the range of maturity variation of
in treated materials may be greater. In practice, the M1 plots should be
planted at some distance from bird populations. In some cases, an extra plot
area might be sown on different dates to divert birds away from the mutagen-
treated plots.

d. When the M1 population must be planted where bird damage is possible, the
plants might be covered with bird-proof nylon or metal netting, which is
relatively inexpensive or the selfing bag should be left until maturity to serve
as protection from bird damage (Figure 5.2).

e. Soil borne toxicity, disease or in some cases, parasitic weeds, such as Striga
spp. may cause complete loss of the M1 population and therefore, extreme
care should be exercised to avoid planting the M1 seeds on soil having such
problematic history.

5.1.5. Care during cultivation and data recording

As stated previously, the M1 population should receive optimum cultivation practices

in either greenhouse or field cultivation for the selected crop, including supplemental
irrigation, weed control by herbicide or by mechanical means, prevention of severe
disease levels if necessary. In addition, records on the condition of the M1 at various
developmental stages are useful and need to be recorded.

a. Emergence – because the application of a mutagen commonly induces some

delay in emergence for treated seeds, recording the estimates of emergence
percentage at a time when the control population can be considered 50 – 90
percent emerged is useful. If emergence of treated populations is poor, either
the treatments are too severe, or the cultural conditions are poor, and
adjustments must be made for the next M1 planting.

b. Seedling survival – estimates of seedling survival recorded at the tillering or

branching stages provide data on delayed effects of the treatments. If
seedling survival of treated population is low relative to emergence, the
mutagen treatments could be considered too severe for the cultural
conditions.

125
c. M1 chimera induction – even crude data on the occurrence of sectors of
chlorophyll deficiency or other morphological changes in the appearance of
M1 plants may be useful data for estimating the effectiveness and intensity
of the treatments.

d. Delayed development – the retarded growth of some mutagen-treated

materials can often be estimated in relation to seedling establishment, the
time of flowering or maturation of the plants as well as by the variability in
plant development within treatments.

e. Survival to maturity – estimates of the number of surviving plants in each

treatment at the time of maturity provide information on the severity of the
injury induced by the mutagen when compared with the number of seeds
planted.

f. Sterility in M1 – useful estimates of M1 sterility can be obtained in various

ways, or detailed counts may be made on appropriate samples of the
population. These estimates may sometimes be made by visual inspection
(Figure 5.3), or via M1 seed yield (weight) corrected for differences in
survival relative to the control population.

126
Figure 5.2. Isolation of sorghum M1 plants by protecting heads before flowering with paper bags to
prevent outcrossing and ensure self-pollination. Bags can also serve to protect the seeds from bird
damage if maintained until harvest. Courtesy of A. Mukhtar Ali Ghanim.

0 Gy 100 Gy 200 Gy 300 Gy

Figure 5.3. Example of reduced fertility (seed set) in M1 plants of barley with increased gamma
irradiation dose rate from 0 (control) to 300 Gy. Courtesy of A. Mukhtar Ali Ghanim.

127
 5.1.6. Harvest of M1

Methods of harvesting the M1 populations will depend on the pattern of ontogenetic

development in the species, the methods of screening and the foreseen generation to
be screened for desired mutants. In most instances, the genetic changes induced by
mutagen treatments occur as chimeras in the somatic tissue of the M1 plant (see
Chapter 4), and the ontogenetic pattern is a prime factor affecting the expression of
observed mutant-tissue chimeras in generative (reproductive) tissue. However, both
the ontogenetic pattern of development and the seed yield of each M1 inflorescence
may have a bearing on the efficiency of the methods for analysing the Ml generation.
As a rule, monocots have a different developmental pattern from dicot species, but
genera, species and even varieties may also differ in developmental pattern even
though, the latter may be minor. In cross pollinated plants; male and female flowers,
located at different position on the same plant (allogamous), might originate from
different initial cells of the treated seeds and hence the M2 progenies are likely to be
heterozygous for mutant genes and might then require further controlled mating
(selfing) to segregate for homozygous mutant plants in M3 progenies.

A consideration of the relation of these factors to the methods of managing mutagen-

treated populations of different plant forms is presented below.

5.1.6.1.Tiller, branch or plant progeny methods

With monocot species like cereals and grasses the maximum potential for induced
genetic variability is in the primary tillers, which arise from the already differentiated
primordial meristem present in the treated seed embryos. Some secondary tillers may
individually yield higher frequencies of M2 mutants but the same mutations would,
generally, be also present in the progeny from primary tillers. The primary tillers are
often the first ones to show signs of maturity, which is a useful guide especially if M1
survival is low and the density of planting has not been effective in reducing tillers of
all M1 plants. With dicots largely self-pollinated seed plants, like beans, peas, tomato,
etc., the methods applicable to Ml analyses may be similar to those used for cereals
except that each 'primary tiller' is equivalent here to a main branch on the M1 plant,
but in some studies even secondary branches might be analysed. However, as already
stated, the approach to be followed with any particular species should be based on the
knowledge of the ontogenetic pattern of that species development, since the nature of
the primordia, including the number of preformed buds, degree of apical dominance,
as well as other factors may influence the pattern of chimera formation.

When the seed yield from each branch or from the whole plant is relatively low, as it
might be with crops like lentil (Lens culinary), pea (Pisum sativum) and chickpea
(Cicer arietinum) whole branches, if not whole plant progeny bulks, have been widely
used. In some other species there may be many branches, many inflorescences and

128
many seeds per fruit or aggregates of fruits on each plant. Thus, it would seem logical
that a sample taken from each primary branch of each M1 plant would be adequate.

The number of Ml seeds sown, must be adjusted to render the expected levels of
germination and survival. Where the number of seeds per fruit is not large enough and
not many fruits are produced per branch, it is reasonable to suggest that each main
branch be sampled, – in perhaps a somewhat large sample depending on the space
requirements of each plant and the availability of resources.

5.1.6.2.Single or multiple seeds bulk methods

The single-seed bulk method, in some form, is applicable to both monocots and dicots
and to all experimental situations. Its usefulness is based on the fact that the
probability of occurrence of a single mutant offspring within the progeny of a fruit
developed from mutated tissue is higher than the frequency of the particular mutation
in the total population of plants (or branches) sampled. However, the method requires
that the same number of seeds (one, or more in modified form) is taken from each
fruiting structure (spike, panicle, etc.).

A modified single-seed bulk method has been successfully used for mutation studies
with Medicago polymorpha (Brock et al., 1971). The authors harvested pods from M1
plants in bulks and in the following generation grew one M2 plant from each of a
randomly selected sample of pods.

5.1.6.3.Mass bulk methods

Mass population management also is applicable if the land and resources for
mechanization are less costly than the labour required for other operations. However,
in this case, with both monocot and dicot species some mechanism to limit the seed
yield on each M1 plant should be devised, even if only the sample for bulk handling
is taken by hand at harvest.

5.1.7. Management of M2 population

In practical mutation breeding, the nature of the trait sought, the availability of space
in the field, greenhouse or laboratory, the labour needed, the possibility of
mechanization, and other resources will have an important bearing on the harvest
methods to be chosen and the precision and selection efficiency. A scheme for the
development of mutant generations from M0 to M3 is given in Figure 5.4.

Selection of mutant traits is usually practiced for qualitative traits, in self-pollinated

crop plants, in the M2 generation as most of the mutants are – by then – recessive, the
mutant phenotype can thus, only be seen in the M2 generation at the earliest

129
(Figure 5.5). However, in cross-pollinated plants mutant genes are likely to be
heterozygous in M2 where further selfing should be practiced for producing M3
progenies in which homozygous individuals for the mutant genes will segregate and
selection can be applied. However, a useful strategy on outcrossing species is to knock
out the dominant allele at heterozygous loci to unveil the recessive phenotype.

5.1.7.1.Systems of handling M2 populations

All methods for the isolation of mutant genotypes in sexually reproduced plants are
based on the pedigree method, modified to account for the chimeric structure of the
M1 plants. Furthermore, the applicable methods are based on population genetics
procedures since the induced frequency of any specific mutant gene or desired mutant
phenotype is appreciably lower in the M1 population than that of a specific gene
introduced into an F1 population by hybridization. In addition, because a mutated
tissue in an M1 plant may appear only in part of the spike, pod or fruit, the segregation
ratio of mutants in the progeny of seed units (pods, fruits, spikes, etc.) will usually be
lower than in normally monogenic heterozygous material. The mutation breeder in
such case must choose the method of screening most adaptable to his or her own
circumstances. He/she should consider the merits, requirements and other aspects of
alternative methods for managing M2 populations in screening for mutants – the
methods described below are primarily used for self-pollinated cereals, but with the
appropriate minor modifications they could be applicable to other self-or largely self-
pollinated plants.

130
 About 6000 homogenous • Initial parental germplasm: highly
M0 homogenous seeds before treatment.
seeds for mutagenic
• 1st mutant generation, treated seeds
grown with narrow spacing to maintain
few tillers/branches.
M1
• Ensure protection from out crossing.
• Harvest each plant separately or as
appropriate bulking.

nd
• 2 mutant generation, apply the
screening method: plant in head to row,
and select individual mutant plant
separately (in case of qualitative traits).
M2
• Control of untreated seeds should be
planted simultaneously to verify the
identity of the mutant and exclude
contaminations.
• For cross-pollinated plant self M2 as
appropriate to produce M3.

rd
• 3 mutant generation, seeds of selected
mutants are planted for further selection
and verification.
M3 • Selection for quantitative trait on row
bases may start at M3.
• Ensure homogeneity of the mutant seeds
by appropriate measures of isolation.
• Start selection in cross-pollinated plants
in M3.

Figure 5.4. Scheme for mutant population development, identification, selection and advancement of
mutants from M0 to M3 generation.

131
Figure 5.5. Segregation of M2 tomato plants for chlorotic mutant seedlings (yellow arrows) in tomato
after seed (M0) treatment with gamma irradiation at 300 Gy. Courtesy of A. Mukhtar Ali Ghanim.

1. M1 population bulk - If the parental material is quite homogeneous, the

tillering of M1 well controlled by close spacing, and the M1 grown in
isolation, the bulk method of M2 population management can be very
efficient, especially for the selection of relatively divergent mutant
phenotypes. Here, the whole M1 population is harvested as one bulk and the
M2 population is grown out as a population of single plants that are screened
for mutant phenotypes. The method is adaptable to mechanization at virtually
all phases including the selection of variants, e.g. mechanical screening for
seed size, weight, shape, etc. In some instances, particularly when mutants
are better identifiable in M3 than in M2 a combination of methods may be
desirable:

a. M1 population bulk to M2 single seed bulk to M3 ear to row progenies;

b. M1 population bulk to M2 ear to row progenies.

2. M1 ear to row bulk - This method, based on randomly harvested M1 ears, is

similar to method 5 below, but differs in that the relation of the ears,
branches, fruit, etc., to one another is not maintained, permitting a type of
bulk processing comparable to that obtained with method 4 but requiring
smaller M2 progenies (perhaps 25 – 30) and adaptable to semi-bulk
harvesting of units from M1 plants.

As with methods 1 and 5, planting of the M1 in a way to limit the production

of primary tillers remains especially important. The method is intermediate
in cost of operation and about as precise as method 4 but less precise than
method 5.

132
3. M1 single-seed or multiple-seeds bulk - This method involves selecting a
single seed at random from each M1 spike (or fruit, branch, etc.) of M1 plant
to constitute an M2 population of single plants from the resultant bulk. M2
single plants can be selected for mutant phenotypes that can be further
progeny-tested in the M3.

Alternatively, single M2 ears can be harvested for selecting within M3 ear-

row progenies for new mutant phenotypes. It is perhaps the most efficient
method in terms of cost and space utilization, but its effectiveness depends
on the ability to identify a single individual mutant in M2 as is the case with
method 1. It should be noted, especially for the single-seed method, that the
greater efficiency of the method is achievable only by growing the same
quantity of M2 (in terms of number of M2 plants) as with other methods. This
requires treating a larger amount of parent material (M0) for production of a
larger M1 population. If the method is followed through M3 or a later
generation, it has the further advantage that there is no increase in population
with each generation. However, a drawback is the greater labour
requirement.

4. M1 plant to row - In this method, all seeds or a sample of the seeds produced
from a given M1 plant are grown to produce the M2 generation, which is then
screened for mutant phenotypes. The success of its use will depend to a large
extent on how well the secondary tilling or branching has been controlled
because the secondary tillers tend to dilute the yield of M1 mutants. This
method is preferred when seeds produced per plant are relatively low as in
common bean (Vicia faba), pea (Pisum sativum), lentil (Lens esculenta), etc.
Successful use in cereals will depend on the screening efficiency, targeted
trait and land availability since generally a somewhat larger number of M2 is
needed. The overall cost for land, labour, etc., is intermediate between the
spike progeny and the bulk methods.

5. M1 ear, branch, pod, and fruit (within plant) to row. Here, each ear taken
from the M1 is processed as a separate entity and sown out as an ear-row
progeny, which is then screened for mutant phenotypes. This method offers
the greatest precision with regard to the origin of a mutant when the material
treated is genetically homogeneous as regards the non-mutant allele, and
when outcrossing is controlled. This is because the progeny of several ears
from a given plant will rarely all carry the same mutant phenotype, and
virtually never in the same ratio. The method is, however, the costliest in
terms of space, labour, equipment and materials (Figure 5.4).

133
All these management methods assume some control over outcrossing, and the ability
of the breeder to distinguish features of the parent genotype in any mutant type
induced.

5.1.7.2.Size of M2 population

The M2 population size will, to some extent, be a function of the available space and
the screening methods to be used. The size of the population may be assessed by either
sampling a few seeds from many M1 plants or more seeds from fewer M1 plants. If
the number of M1 plants is low, but with relative high fertility then 20 – 25 seeds may
be harvested from 2 – 3 spikes per M1 plant.

However, if the population of the M1 is large and with low fertility then 1 – 5 seeds
may be sampled per spike. In principle, the greater the number of individuals, the
higher the chances for selecting the desired mutant. With the M1 bulk progeny method
(1) of analysis, the estimated M2 population size should be about twice that of the M1
spike or M1 plant progeny populations (M2) as an offset to the usually lower selection
efficiency.

With the M1 single-seed bulk method (3), management efficiency is achieved only
when taking as large an M2 as for the M1 spike progeny method, i.e. if you have 5000
M1 plants each represented by 3 spikes and each spike is represented by 30 seeds, then
the total seeds will be 30 seeds × (3 × 5000 spikes), that is about 450 000 seeds. In
practice this number of seeds each from a single main branch or spike might be
somewhat difficult to obtain since a very large M1 (450 000 plants) would need to be
grown for that. A modified bulk approach might involve making several (2 or 3)
different single-seed bulks from the same M1 population. The prime consideration is
that each test unit would be equally represented in the M2 population.

The successful use of the mutation breeding method depends on the choices the
breeder can make among the selected phenotypes, since many of them will carry
mutations in other traits as well. Therefore, the population grown should be large
enough to ensure an opportunity to select more than one instance of a mutant with the
desired phenotype. There is also evidence that certain kinds of mutations appear more
rarely than others as discussed later in this chapter.

Mutations influencing, e.g. fertility, flowering time, flower morphology, plant height
and pale green, are rather common; some other types of mutations are much more
infrequent. Dominant mutations are rare, but can still be obtained and might in fact be
sought for specific purposes. Disease and insect resistance mutations are relatively
rare, but major differences in the frequencies of particular kinds of mutations may
also depend on the parent genotype (especially in polyploid plants) as well as on the
ontogenetic development pattern of meristematic tissues following treatment. With

134
those plants in which the M1 plant develops from a single cell or adventitious bud, i.e.
for vegetatively propagated crops (VPCs), the frequency of mutations could be higher
than in plants with multi-cellular meristem initials. However, in general, mutagen
treatments of pollen have so far yielded much lower frequencies of mutations
compared to mutagen treatment of seeds.

0 5 10 15 dS/m

Figure 5.6. Screening for salinity tolerance, in hydroponic nutrient solution, of rice mutants at four
levels of salt (NaCl2) (0, 5, 10 and 15 dS/m) showing variation in shoot and root growth after 14 days
of applied stress (The Joint FAO/IAEA Plant Breeding and Genetics (PBGL), Seibersdorf, protocols
2014). Top row depicting show and lower row for roots. Courtesy of A. Mukhtar Ali Ghanim.

D A

C B

Figure 5.7. Screening lentil mutants, in hydroponic solution, for drought tolerance using PEG6000 at
four levels of concentrations (A, B, C and D), respectively: 0, 10, 15 and 20%. Photos were taken 6
weeks applying stress pressure protocol optimization experiment at PBGL, Seibersdorf, Austria in
2014. Courtesy of A. Mukhtar Ali Ghanim.

135
 5.1.7.3.Screening methods and mutant selection techniques

Various screening methods to identify and select desired mutants have been proposed
and tested. The efficiency of each method is dependent on factors more or less under
the control of the breeder.

1. Visual methods of selection for identifying mutant phenotypes are common

and can be very efficient. However, when breeding for specific traits it is
important that the breeder ignores all other deviants. Unusual genetic variants
may however, be of value for fundamental studies and in developing a mutant
germplasm base for future breeding purpose. As a precautionary measure
seeds from variants should be preserved. Visual selection is often the prime
basis for selecting for disease resistance, earliness plant height, colour
changes, non-shattering, adaptation to soil, climate, growing period, etc. The
procedures involved are essentially the same as for variation introduced by
cross breeding. However, with mutation breeding, ancillary techniques can
be used as aids to visual selection and must be stringent and efficient for
concentrating the breeder's attention on a few specific individuals from
relatively large populations. Thus, mass-screening techniques (phenotypic
and genotypic) are particularly suited to mutation breeding. Advances in
screening methods such as laboratory and greenhouses techniques (Figures
5.6 and 5.7) enhance efficiency of visual selection of mutants.

2. Mechanical or physical methods of selection can also be used very efficiently

in screening for seed size, shape, weight, density. etc., using appropriate
sieving machinery as they are readily adaptable for processing of large
quantities of seeds. These would normally be applied to seed produced from
M2 plants and are perhaps most adapted to the M1 bulk population method of
managing M2 material. Processing individual M2 lines or plants would also
be feasible, but costlier.

3. Other selection methods, such as chemical, biochemical, physiological,

physio-chemical, and various specific methods may be needed for selecting
certain types of mutants. However, virtually all employ visual parameter to
expedite detection. Low alkaloid content mutants, for example, maybe
sought using colorimetric tests on M2 seeds or plants or even M3 seeds;
protein analyses by colorimetric, chromatographic or electrophoresis
techniques may be conducted on individual seeds from M1 plants, on seeds
from M2 plants or on the bulk from M2 plant progenies, the efficiency
depending on the degree to which such techniques can be mechanized. When
searching for resistance to herbicides or fungicides in a susceptible variety
the chemical products might be applied to M2 seedlings in the laboratory or

136
 even in the field using slightly higher concentrations. Repeated application
of the chemical to plants showing evidence of tolerance or resistance should
then be made to confirm the response, but progeny tests of selected
individuals are also necessary to confirm the genetic basis. For example,
screening for herbicide tolerance can easily be conducted by thinly sowing a
large M2 population and then applying an herbicide to the entire population
and retaining any survivors for further testing. Whilst this provides a very
efficient screen, it does mean that the rest of the population is sacrificed in
order to identify the desired mutants. It has been successfully used to produce
the imidazolinone tolerant barley ‘Scope’ in Australia (Moody, 2015). In
germination and seed viability testing, changes in phenol reaction of seed
coat tissue may simply involve exposure of seeds from M2 plants (or an M3
seeds population bulk) to a given concentration of phenol, then visually
selecting the individual seeds for appropriate changes. Insensitivity to
gibberellin (which can induce short stature) can be identified by spraying
gibberellin solution onto seedlings or soaking seeds in a gibberellin solution,
then searching for seedlings with little or no response. When an enzyme
involved in a particular pathway is known to be altered by mutation, methods
to identify presence or absence of this particular enzyme, its precursors or its
derivatives might be devised for use in rapid screening for desired mutants.

4. Screening for abiotic stresses such as drought, salinity, heat etc., requires
setting up of the selection pressure and maintaining a uniform stress over the
M2 and M3 populations. Recent advances in hydroponics and laboratory
techniques have led to the development of different screening methods for
abiotic stresses for conventional breeding programmes that can readily be
adapted to handle mutant population of larger size with greater efficiency to
identify phenotypic mutants in M2 or M3 generations (Sarsu, et al., 2017 in
press; Bado et al., 2016). The choice of the mutant generation on which to
apply the selection protocols will depend largely on the nature of the trait –
qualitative or quantitative. For qualitative traits the protocol should be further
expanded to provide for quantification of the effect of the trait (e.g. yield)
and testing in some multiple environments to ensure stability and heritability
of the trait and ensure its wide use among breeders. In recent years there have
been huge advances in high-throughput phenotyping, now known as
phenomics. Phenomics screening often involves optical imagining, including
red-green-blue (RGB) colour model and multi-spectral cameras, coupled
with bespoke soft-ware imaging analysis programmes (Tardieu et al., 2017).
Advanced phenomics can involve automated, robotised greenhouse, nursery
and field platforms. Field trials can be imaged remotely at multiple levels (by
hand, by drone or from outer-space). These methods offer unprecedented
throughputs which have huge potential in increasing the efficiency of

137
 mutation detection both in terms of accuracy and the increased numbers of
mutants that can be screened. The downside is that these are very expensive,
especially the more sophisticated systems, however, simple, cheap systems
such as hand-held imaging cameras are also available.

5.1.8. Management of the M3 generation

Progeny tests are essential for the identification of all mutant lines useful for plant
improvement and re-selection from M3; this is done to establish that the trait is
heritable. Further progeny tests, may be necessary to stabilize a potentially useful
variant. Furthermore, it is not uncommon that a mutant may be homozygous for the
desired character but segregate for other undesirable ones, they still can be chosen
when their selection might be useful for improving the genetic background of the
desirable mutant. In rare cases a mutant may be the result of a modification in epistatic
relations of more than one modified locus; in such a case the M2 mutant phenotype
may not re-appear among M3 progeny. If the phenotype is the result of an interaction
involving a heterozygous locus then it cannot be fixed in an inbred line, which may
explain the disappearance of the phenotype between the M2 and M3 generations. If
this is the result of interactions between independent loci that was lost due to
independent assortment of gametes, then going back to the M2 and sampling a larger
M3 should facilitate the detection of the phenotype and eventual fixation.

In several situations, M3 progeny tests may be essential for the detection of mutants
particularly of those not easily discernible from single (M2) plants. This may be
particularly true for traits that are influenced by environment, e.g. pigmentation and
some biochemical or physiological mechanisms. When the number of seeds per plant,
pod, fruit, spike, etc., in the M1 is low, it is desirable to grow out an M3 population
from all M2 plants and screen those plants, since in some practical cases as much as
60 percent of the total mutants are observed first in the M3. The frequency of mutant
individuals in the unselected population is normally higher in M2 than in M3 but space
requirements and other considerations make it generally more cost-effective to screen
for mutants only in M2. Genotypic screening (e.g. for a mutation in a specific gene)
in the M2 is more efficient than phenotypic screening which may be influenced by the
environment (see Chapter 8-C), but obviously requires knowledge of the gene in
question and an efficient system to detect mutations that are likely to result in
phenotypic changes.

In some other cases, especially in polyploids like durum wheat, Triticum durum and
bread wheat, Triticum aestivum it may be necessary to reselect even an M3 line and
grow the M4 progeny of few selected individuals with the mutant phenotype to be
reasonably sure that the trait in question is due to mutation. In cases where one wants
to test the multi-variate phenotype of a specific mutation seeds harvested from M3

138
families can be submitted for field trialling in comparison with the parent line and
suitable local controls. In some rare cases, new varieties can be developed from
especially useful M3 mutants, but it is more likely that further crossing and selection
is necessary to produce a marketable product. The breeding of Kebari, the ultra-low
gluten barley, is a good illustration of this effect (Tanner et al., 2016). Here 3 null
mutants at hordein loci were combined to create a triple null, which was then selected
for acceptable phenotypes whilst maintaining the mutants to result in the release of a
line that is well below the maximum gluten levels that coeliac sufferers can tolerate.

5.1.9. Contamination in self-pollinated mutant crops

Sometime, authenticating the genetic origin of the variation in mutagen-treated plant

material may be of interest and concern not only to geneticists but also to breeders.
Contamination is a common problem in any plant-breeding programme and causes
even greater problems in mutation breeding. This is potentially a bigger problem for
a non-breeding organisation that must purchase seeds of a commercially available
cultivar rather than a breeding company utilising its own pure stock seeds.
Commercially available seed stock must reach strict levels of purity but even if this is
99 percent, it means that 1 in every 100 seeds could be a contaminant. Contamination
is manageable when it affects only a very small proportion of the treated population,
and when the characteristics of the contaminants can be recognized easily
(Figure 5.8), but the differences may be much subtler in practice as obvious
differences should have been eliminated in the stock production process. However, it
should be stressed that, neither the risk of contamination nor the fear of uncertainty
over the origin of the variation isolated from mutagen-treated populations should deter
the breeder from using the mutation approach when that approach can be more
efficient and expedient to attain a specific objective. Contamination can be reduced to
manageable levels by taking certain simple precaution measures. Moreover,
contaminations (depending on the source) are often distinguishable by characteristic
phenomena, and a number of analyses can be made that will usually lead to a
conclusion about the origin of genetic variants isolated from mutagen-treated
populations. A whole genome profile can now be generated for many crops at a
relatively low cost and a pool of potential mutants can then be checked against the
parent genotype to rapidly distinguish between genuine mutations (identical profile to
parent) and contaminants (different profile).

5.1.9.1.Basis for and sources of contamination

The prime sources of contamination in mutant populations are described below.

1. Mechanical mixtures sometimes occur during the harvest of the immediate

or previous generations of the parent genotype seeds from: inadequately
cleaned harvesting machinery or inter-mingling of plants from different plots

139
 during harvest, or ratoon of the same species from seeds of the previous
plantings, or movement of seeds (manually, by machine, water, or animals,
etc.). Ideally, the mutants should be grown in isolation to prevent
uncontrolled outcrossing with other varieties of the same crop grown nearby.
Contamination can also occur in seed stocks in any generation prior to or
following the mutagen treatment but is less important after selections and
isolation into progeny lines because deviants can be removed easily. The
chances for outcrossing are increased with increasing mutagen dose, because
of increased male sterility in the mutated population.

2. It is advisable to identify, purify, ensure homogeneity through selfing and

progeny test of the targeted parent genotype material. Rouging-out of off-
types prior to flowering will reduce cross contamination. Reducing genetic
variability in the parent material before treatment is the most efficient
precautions against contamination.

5.1.9.2.Criteria for distinguishing contaminants from mutants

It is impossible to prevent all and every condition that could lead to contaminations,
due to costs, expediency and availability of proper facilities. However, there are tests
by which the breeder can categorize or classify variations into contaminant or possible
contaminant; mutant or possible mutant origin.

Figure 5.8. Outcrossing in the pale green mutant rice fields. The dark tall green plants are likely a
result from outcrossing of the wild type with the mutant. Outcrossing rates are often higher than
mutation rates. Courtesy of Q. Shu.

140
Contamination

1. Genetic variation from contamination due to admixture

The variation is likely to be of contaminant origin as a mechanical mixture present in

the parent variety population, in any of the following situations.

i. Variant individuals having essentially the same phenotype(s) appear

repeatedly in both M2 and control populations.

ii. When no segregation is observed for a certain phenotype in the M2, i.e.
it is uniform.

iii. Specific variant individuals segregate in an approximately 3:1 ratio in

an M2 plant progeny line.

iv. Variant individuals are found in an M2 branch, spike or plant progeny

line that also carries a wide range of variation in other traits.

2. Genetic variation from contamination due to outcrossing

Although not always conclusive, the variation may be suspected of contaminant origin
as an out-cross between the M1 and an outside source – or mechanical mixture – in
the following situations.

i. An M2 variant individual has been changed in many ways and/or

segregation in M3 of the selected variant shows the variation involving
several independently varying traits; conversely a variant with only one
or only a few characters changed is likely to be a mutant.

ii. Variant individuals of the same phenotypes appear unusually

frequently in M2 populations.

A similar variant does not appear in the progeny of one or more sibling
M2 plants from an M1 branch, spike, and panicle or plant progeny. This
test is not critical, however, since mutants can originate even in rather
small chimeras of a single plant, and possibly also as recombinants
from an altered base sequence, which would be not likely to occur in
siblings.

iii. A possible 'donor' parent for the trait(s) in question can be identified;

141
 iv. Partial pollen sterility occurs in some M2 or M3 plants of a questioned
progeny line, suggesting that the M1 was partially sterile and more
receptive to cross-pollination.

Mutation

1. Genetic variation of mutation origin

Due to chimerical nature of the induced variation in seed primordia, several types of
variations may be directly associated with the mutation event, when the following
observations can be made.

i. Variants of the same phenotype will commonly appear at a very low

frequency even in a bulk M1.

ii. Variants will appear as one to a few individuals in a branch, a spike or

plant progeny, virtually never appearing in a 3:1 ratio in more than one
or a few of several branches or spike progenies, but usually at a far
lower ratio in one or two of a group of sub-lines and be absent from the
remaining sub-lines derived from branches or spikes of a single M1
plant. The ratio of mutant to normal should be appreciably less than
1:3 or 3:1, especially if an M1 plant progeny consists of a sufficient
number of individuals (about 40 – 50).

iii. Genetic variation of induced mutation origin.

A mutant is most likely to have been induced if in addition to meeting the criteria of
mutation origin there have been similar variants isolated from several different plant
progenies in a treated population at a statistically significant higher frequency than
what could be observed in the M0 control population. The number of spontaneous
mutations occurring at the same time is extremely low. Spontaneous mutations can,
of course, occur at any time during the life cycle, but can be determined to some extent
by the procedures mentioned above.

5.1.9.3.Genotypic and phenotypic analyses

In general, for practical purposes, the criteria described below are the most critical,
even though they can seldom be performed for large numbers of variants. However,
tests of large numbers of variants are rarely necessary as, in most instances; the
breeder finally uses relatively few of the variants isolated. The following analyses will
usually permit a conclusion relative to the origin of any specific genetic variant.

142
 1. Genotypic

i. Progeny testing: selected variants should breed true in progeny grown

from selected M3, but might not do so in M2 progeny tests. If
segregation does occur in progenies from M2 single plants, the
variation should be limited to the selected trait although it may
sometimes occur for a few (1 or 2) others.

ii. Backcross and other crossings: selected variants should show a

relatively simple genetic segregation in backcrosses to the parental
genotype and to some other strains (1 or at most 2 gene segregations).

2. Phenotypic

i. Morphological characteristics: in general mutants should show a

measure of similarity to the parent genotype except for modifications
related to the mutation involved. Often a whole complex of changes in
phenotype could be caused by a simply inherited mutant, but a
similarity in many other features should still remain.

ii. Physiological, phyto-pathological and biochemical characteristics: the

variant should show similarity to the parent genotype in a large number
of measured traits, especially in traits governed by several different
genes or complexes, such as quality characteristics, disease and pest
resistance, and biochemical traits.

5.2.THE DETECTION OF INDUCED MUTATIONS

Mutation induction and mutation detection are two independent processes. The extent
to which an induced mutation may survive and give rise to a mutant at the level of the
organism is controlled by many factors as listed below.

x Uni-cellular or multi-cellular initial structures (bud and embryo meristems).

x Mode of reproduction: vegetative or seed propagated; self-pollinated or
cross-pollinated.
x Stage of differentiation of the primordial cells from which inflorescences
arise (whether they are already present in the dormant embryo or arise later
after the mutation event).
x Number of primordial cells involved in the origin of each inflorescence.
x Time in the life cycle of the plant when such primordia arise.

143
 x Genetic architecture of the organism (whether primarily diploid or
polyploid).
x Characteristics of the locus involved in the mutation process – single or
multiple gene(s).

Other genetic factors, such as polygenic inheritance, linkage, gene interaction and
previous selection history of the character under study, may reduce the rates of
phenotypic detection of an induced mutation. Variables relating to the mutagen used,
treatment conditions, and pre-and post-treatment modifying factors also influence the
manifestation, transmission and recovery of an induced mutation (van Harten, 1998;
Toker, Yadav and Solanki, 2007, see also Chapters 1, 2 and 3).

Induced mutation can be accomplished easily, but such mutations will be of no use to
the breeder unless they are manifested at the level of the organism and transmitted to
subsequent generations, and this will depend on several principles. Thus, great
attention should be given to the adoption of suitable screening procedures as described
earlier.

5.2.1. Inter- and intra-somatic selection

Unicellular organisms are quite easy to screen for mutations and do not present the
same problems for mutant recovery as described in the multicellular organisms
(Brunner, 1995;). On the other hand, crop plants are all multicellular; hence the cells
carrying any mutation will have to compete with normal cells regarding growth and
survival. While, in asexually propagated plants, such competition can be overcome in
vitro, for example, by raising whole plants from single cells as in Saintpaulia spp. or
other ornamentals (Broertjes C, & van Harten, 2013), in sexually propagated plants,
this competition leads to two types of endogenous selection, which will intervene
before any mutation induced in a treated seed may be expressed in the M2 generation.

The first process occurs in the MI somatic tissue and is termed ‘diplontic selection’
defined as a “competition between cells within a meristem” (Klekowski, 2011 and
Rajarajan et al., 2014). The second process of selection happens in the gametes of the
M1 plants and is hence referred to as ‘haplontic selection’, which can be defined as
‘the competition occurring during haplophase, i.e. between gametes’ for the
production and transmission of the mutation to the zygote (Figure 5.9). Haplontic
selection operates more rigorously in the pollen than in the ovules. Only a mutation
that passes through both the somatic and gametic sieves will show phenotypic
expression in the M2 and subsequent generations. Among sexually propagated plants
the handling procedures are simpler for those that are hermaphroditic or monoecious
than for dioecious plants. On the other hand, in asexually or apomictically propagated
plants the somatic sieve is the only important one since the gametic sieve does not

144
operate. Again, among apomicts there is a difference between those that are solely
propagated through vegetative parts and those that are seed propagated without there
having been through any meiosis in the cells that give rise to the functional
megaspores. The sexually and the apomictically propagated plants will have to be
considered separately, since the problems of recovery are quite unique to each of these
groups.

In rice and wheat, the number of initials in a meristem may vary from one to two or
more. When only a single cell is involved in the mutation process, then the entire
inflorescence will carry the mutation. Alternatively, if several cells are involved, the
inflorescence could appear as chimeric. M1 segregation ratios will help estimate the
number of inflorescence initials. The expected segregation ratio of normal to mutant
phenotypes will be 25 per cent (3: 1) if a single cell initial alone is involved; when
more than one initial exists, since all of them may not mutate simultaneously, the
segregation ratio will be less than 25 per cent in the M2 generation with a deficiency
of recessives (Cheema and Atta, 2003).

Figure 5.9. Schematic diagram illustrating the effect of diplontic (somatic) and haplontic (meiotic)
selection of mutated cells, after irradiation treatment of the multi-cellar embryo, on the mutants
M2 seeds.

145
When a plant produces several tillers, as in barley or wheat, the first-formed tillers
may carry more different mutations than later-formed ones. This is because of the
greater probability that the first-formed tillers arise from several initials already
present in the dormant embryo subjected to treatment. Thus, a reduction in tilling
achieved through close planting can increase the frequency of mutants recovered in
M2. It has been found that close spacing of barley plants raised from irradiated seeds
reduced the average number of initial cells involved in spike organization, thereby
increasing the area of mutated sectors (Singh, 2016). However, diplontic selection
does not seem to play the same role under all experimental conditions and in all plants.
Frydenberg and Jacobsen, (1966) established that the spikes of secondary shoots in
barley carried even more mutations than the main spike. The situation is even more
complex in dicotyledonous plants. Scheibe and Micke, (1967) also demonstrated that
the mutation frequency in sweet clover after seed irradiation was higher in basic
secondary shoots but diluted in the main shoot by extensive branching before the
formation of the flowers.

Corresponding observations were made in mutation experiments with perennial

grasses like Alopecurus, where the yield of mutations was highest in the first seed
harvest from M1 plants but dropped sharply to the second harvest in the next year. For
these reasons, it is always preferable to keep the seeds of each inflorescence separately
and carry them forward to the M2 generation as ear or inflorescence progenies rather
than whole plant progenies. This will help avoid a numerical dilution of the mutants
occurring in the population and thereby facilitate their detection.

5.2.2. Genetic structure

The genetic structure of the plant greatly influences the rigour of both the somatic and
haplontic sieves. Where duplicate loci are present for the basic metabolic factors, as
in many polyploids, both mutations and chromosome aberrations survive to a greater
extent than in plants with a strictly disomic genetic constitution. Thus, polyploidy may
facilitate the later recovery of induced mutations. On the other hand, the phenotypic
expression of an induced mutation may be masked because of the buffering effect of
the duplicate factors. Therefore, in many polyploids chlorophyll mutations are rare in
the M2 generation. However, while chlorophyll mutations may be rare or absent,
viable morphological mutations may occur at a high frequency in polyploids like
bread wheat (Hancock, 2012).

The nature of polyploidy is also important in relation to both the discovery and the
expression of an induced mutation. An understanding of the genetic constitution
would help to formulate suitable handling procedures. For example, in an
autotetraploid with a genotype AAAA at a locus the M1 plant may become AAAa if the
mutation A to a occurs. In the M2 only plants with the genotypes AAAA, AAAa and

146
AAaa will occur (unless there is random chromatid segregation) and hence, the
population will be phenotypically homozygous. A recessive phenotype may only arise
in M3, therefore in such cases the screening should be carried on to M3 and later
generations.

5.2.3. Function of locus

The function of the locus concerned and its neighbouring genes will also determine
the viability and frequency of the transmission of mutations. It is obvious that a
mutation in a locus with an important metabolic function will be less likely to survive
than one not concerned with growth and survival of the plant.

5.2.4. Mutability

It is important to note that different genes are not equally mutable. For example, when
studying the rate of spontaneous mutations in maize found it was that the gene r (for
colour) is relatively mutable, whereas the genes for waxy (wx) and shrunken (sh)
endosperm are relatively stable (Bennetzen and Hake, 2009). Cox, (1972) suggested
that mutation rates at different loci in the same cell may differ considerably and that
these rates are genetically controlled. To ascertain this assumption, the author argued
that genes in higher organisms may be organized not according to their function, but
into different segments of the genome that have different intrinsic mutation rates.

5.3.IDENTIFICATION, EVALUATION AND DOCUMENTATION OF

MUTANTS

5.3.1. Identification of mutants

For practical plant-breeding purposes, the question of whether a useful variant

selected after mutagenic treatment represents an induced mutant, a spontaneous
mutant, a line already intrinsic in the untreated material or a product of segregation
after selfing or outcrossing, is of minor concern. However, for mutation research and
for the evaluation of the efficiency of mutation breeding as compared with other
breeding methods this question is of great importance.

Although it appears impossible to state with absolute certainty that a variant has been
induced by the mutagen used in the particular experiment, typical macro-mutants of
self-pollinating crops such as: dwarf and semi-dwarf, spelt, early maturing, disease
resistant types and other major changes may be considered as being most probably
due to an induced mutation, mainly if the same variants are not found in comparable
control populations. Various precautions and procedures may help to verify the
identity of recognized mutant.

147
 x Mutagenic treatment of highly inbred material 'pure line' and, if possible, of
a single plant progeny selfed repeatedly under controlled conditions (by field
isolation or bagging).
x Mutagenic treatment of a 'pure line' marked by alleles not occurring in the
other breeding materials.
x Strict isolation of the M1 generation.
x Pedigree breeding from M2 to M3 or M4.
x Re-selection within the selected variant.
x Comparison of the variant with similar strains and varieties.
x Testing the race spectrum of disease resistance.
x Analysis of reciprocal crosses between variant and mother strain, if possible
combined with fingerprinting and/or DNA and RNA sequencing.
x Segregation analysis up to the M3, M4.
x Test crosses of the variant with other genotypes showing the variant
character.
x Mapping the mutant locus.
x Cytological investigations to reveal induced chromosomal rearrangements.
x DNA testing/sequencing of candidate genes when the causal gene is
deliberately targeted or can be predicted from the phenotype.

5.3.2. Propagation and evaluation of useful mutants

When a mutant appears promising it will take one or both of the two courses illustrated
in Figure 10; (1) direct increase by repeated cycles of selfing to produce sufficient
seeds for evaluation in replicated trials, or (2) use in crossing schemes, either back to
the parental line or to other elite lines to transfer the mutant trait into contemporary
breeding material. In many species, the rate of genetic progress means that the parent
cultivar will be inferior to more contemporary elite cultivars, even if the parent
cultivar was the best available at the time of planning the mutation experiment. The
isolated mutant is also likely to carry other undetected mutations that may result in an
inferior phenotype in other locations so some form of transfer by hybridisation and
selection is generally desirable. The seeds of the mutant needs to be multiplied prior
to proceeding to multi-location trials. The mutant, the mother genotype and other
prevailing varieties with which it is intended to be compared, should be propagated
under the same conditions to produce comparable seed properties for field testing.

148
The methods of testing seed propagated mutants in comparative trials are essentially
the same as for any other newly developed genotype (Hertel and Lobell, 2014;
Johnson et al., 2017). Mutants exhibiting variations in vernalization requirement,
photoperiod response, growth habit, stature, resistance to biotic and abiotic stresses
and yield components should be tested in a wide range of environments, i.e. various
locations, soils, water and nutrient conditions, seed rates, planting distances, sowing
dates, etc. For the first trials the number of replications per mutant is often reduced in
favour of a higher number of treatments and locations. Inter-varietal competition is
expected when neighbouring plots differ widely in stature and/or growth pattern and
the mutant growing type is rare among the varieties tested. It is not predictable
whether, for example, the taller or the shorter varieties have an advantage in a
particular trial. Competition effects can be minimised by an appropriate shape and
size of the plots and by excluding border rows from plot evaluation.

Important details for documentation include: origin (pedigree) and descendance of the
treated material (number of generations after a cross), material mutagenized, pre-
and/or post-treatments applied, mutagen type and dose used, mutant generation of the
first selection and the morphological and physiological differences between the
mutant and its mother genotype It should be noted that in most national variety
registration and release systems there are established guidelines for evaluation of new
genotypes for the purpose of release, including mutant varieties.

5.3.3. Recording of experiments and documentation of mutants

5.3.3.1.Recording of experiments

All relevant facts and information of the mutation breeding experiments should be
stated when reports are written, and the results should always be presented in a clear,
understandable way and should include all important details. It is especially important
that careful consideration be given to the protocols formulated for conducting the
experiments that could then be reported in publications, according to a standardized
format. There are many excellent examples, with variable layout, that illustrate
experiment documentation and publication. As an illustration, a general outline is
given below.

I. Experiment title
II. Experiment identification: location, main investigators, experiment
number, date, etc.
III. Specific objectives
IV. Material and methods

149
 A. Constants in experiment

1. Material

a. Biological parent material (name, identification

number and pedigree if appropriate); indicate source,
composition (bulk or lines), etc.
b. Ideally, a sub-sample of the parental material should
be logged in a gene-bank.
c. Mutagen(s), source, energy, dose rate, tests to ensure
purity, etc.

2. Methods

a. Pre-treatment(s), including preparation for

treatments.
b. Treatment(s)
x Mutagen dose(s): (i) Radiations, rate and
time(s), distance from source; (ii) Chemicals,
concentration(s), time(s), composition and
amount of treatment solution, etc.
x Conditions of treatment(s).
c. Post-treatments; handling of treated material,
storage, planting, etc.

B. Experimental variables: Record treatment and replicate numbers

in sequence depending on the specific experimental variables,
presenting the detail of experimental design relative to specific
objectives; arrange variables (whether materials and/or methods,
treatments and replicates) as per the above listing.

V. Results
VI. Summary and conclusions
VII. References

5.4.FACTORS INFLUENCING THE SUCCESS OF MUTATION BREEDING

The success of the mutation breeding programme is measured mainly by the

production of superior varieties, but also by the spectrum and quality of mutants

150
induced, identified and recovered from a segregating mutant population. Even with a
full consideration of the requirements for the mutation experiments there are other
factors which could limit the success in recovering the targeted mutant trait. These
mainly include, the situations described below.

5.4.1. Differences due to the genotype

Much evidence exists that genetic differences, even when they are as small as single
gene differences, can induce significant changes in radio-sensitivity, which in turn
influences not only the total rate but also the spectrum of recoverable mutations and
the degree of background (Zaman et al., 2007). Although nobody is able to predict
the influence of a particular genotype on the mutation spectrum, the choice of the
parent material is a key factor of any programme in mutation breeding (Bradshaw,
2016).

More definite information is available with regard to the influence of the ploidy level
on the mutation spectrum. In diploid species the great majority of mutations occur in
single recessive genes. However, deviation from the normal 3:1 ratio due to deficiency
of recessives has been very frequently observed. Dominant vital mutations hardly
occur, in fact that they are mostly lethal or semi-lethal in the homozygous condition,
in contrast to diploid organisms, as the dose required to produce them is unlikely to
result in viable plants. Many genes are re-duplicated in polyploids, which increases
their ability to bear a high mutational load, including gross chromosome aberrations,
with no apparent negative effects. This results in the more frequent discovery of
dominant and semi-dominant mutations amongst such species.

Phenotypic buffering is another property of polyploids that restricts mutability of

many characters, especially those essential for the whole life of the plant: e.g. the
process of chlorophyll formation. Thus, chlorophyll mutations decrease with the
increasing level of ploidy Stadler, 1929 cited by Jankowicz-Cieslak, Mba and Till,
(2017); however, the total rate of mutation increases. For example, in Triticum spp.
the total mutation rate was about three times higher in hexaploid wheat than in the
tetra and diploid genotypes (Rajarajan et al., 2014). Differences in mutagenic
response exist also between species of the same level of ploidy and between varieties
within the same species. Different monosomics of Triticum aestivum showed
differences in mutation frequency owing to factors controlling the chlorophyll
development (Lundqvist, 2014; Umavathi and Mullainathan, 2016; 3URWLü et al.,
2013). In a review on mutation breeding in plants Gottschalk and Wolff, (2012)
reported results confirming that the closer the varieties are in their genotypes, the
greater is the similarity in their spectra and frequency of mutation.

In studies of induced mutations in quantitative characters differences in the ploidy

level were not as important as the genotype at the same ploidy level (Bharathi

151
Veeramani et al., 2005). Thus, it seems that genetic variability in the background of a
genotype is an important factor. According to several authors those characters that
showed greater variability in the background could be improved more easily and give
better expectation of mutant improvement.

In biennial or perennial plant species it seems preferable to use early flowering

genotypes for mutagenic treatment and to harvest seeds from the M1 plants in the first
season. Any valuable mutant characters can be transferred later quite easily to late
maturing genotypes, if they are of particular breeding value (Wani et al., 2014).

Heterozygosity as a genotypic property can also influence the type and frequency of
mutation. Many polyploids are less sensitive to chromosome aberrations if they are in
heterozygous condition (Bradshaw, 2016). As Gregory (1960) stated “The chief
limiting factor in mutation production and mutant recovery is the genetic constitution
of the experimental organism and not the type of mutagen used. Thus, for the plant
breeder, knowledge of what might be called mutant expectations in his material may
be more important than a resolution of the mechanism of mutational change at the
sub-microscopic level”.

5.4.2. Type of mutagen and dose

The difference in mutation spectrum among different sources of irradiation is obvious

in the spectrum of induced flower colour changes following mutagen treatment (Jain
and others, 2010). For instance, densely ionizing radiations such as different sources
of ion beam produce relatively more chlorophyll mutations of the albina, striata, and
xantha type (Figure 5.11), whereas the frequency of the viridis type is highest
following gamma-ray treatment. Thus, the chance of selecting desired mutants might
be considerably increased by broadening the choice of mutagens. However, as
discussed earlier, besides the mutagen, other factors also affect the mutations
spectrum and the quality of induced mutants.

Another problem in the mutant quality is the number of mutation events that occur in
the same meristematic cell at the time of treatment that are transmitted to later
generations. The number of desirable event is far less than the undesirable ones and
consequently the number of mutant plants that carry only desirable changes will
further decrease if more than one mutation per cell is induced. Several measures can
be taken to avoid this undesirable result. Firstly, one should not apply too high a dose
of any mutagen. Secondly, one should seriously consider that super-mutagens, which
give mutation rates of at least 50 percent on the basis of plant or spike progenies, may
not be at all advantageous for mutation-breeding purposes. Thirdly, if high mutation
rates have been induced, they should be allowed to segregate, and selection for useful
types should be conducted in M3 or later generations (Hansel, Simon and Ehrendorfer,
1972). However, one should realize that the latter technique will not eliminate those

152
undesired mutations that are closely linked with a desired character in the same
chromosome.

In recurrent irradiation experiments the number of recoverable mutants can be

increased, but since the number of multiple mutations also increases, the mutant
quality from the breeding point of view may decrease (Micke, 1969). It would seem
wiser to allow segregation of mutation material, or to put the same effort as for
repeated mutagenic treatments into crosses of deviating plants selected in the
Ml generation for purification of the mutant character, or transfer of the mutated locus
into another genetic background. Both actions might change the quality of a mutant
in the desired direction.

5.4.3. Pleiotropy and linkage

Generally, it seems to be nearly impossible to find a mutation in an organism that

results in only one single divergent phenotype compared with its initial wild genotype.
For instance, mutations resulting in pale green plants also result in reduction in general
plant growth and delayed maturity and, in most cases a group of distinct variants can
be observed and this group as a whole is transferred from one mutant generation to
the next showing mostly a 3:1 segregation ratio. Theoretically, there are three possible
interpretations for this behaviour:

a. a single mutant gene is responsible for the whole complex of deviating

characters;
b. a tiny portion of a chromosome has been lost containing several genes;
c. and several closely linked or neighbouring genes have mutated.

Monohybrid segregation will occur in all these cases but only the first one is a true
example of a pleiotropic gene action. The other two events simulate a pleiotropic
effect of one gene, although several genes are lost or altered. It is practically
impossible, in most cases, to state which of these possibilities is realized; therefore,
the term 'pleiotropic gene action' is commonly used in the literature for the whole
group of these phenomena.

153
Figure 5.10. Schematic diagram showing options for advancing identified mutant in M2 or M3 through
either direct selfing and evaluation (right course) or backcrossing to parent or elite lines and cultivars
(left course) to multi-location trials, registration and release to farmers.

A B

C D

Figure 5.11. Examples of different types of irradiation-induced mutations on spikes (A-Hood spikes)
and leaves (B-Albino, C- Striata, and D- Xantha) in barley (Hordeum vulgare). Courtesy of L. Gomez-
Pando.

154
Quite often, true pleiotropism may be realized but there is no doubt that not only
minute deficiencies but also a more or less simultaneous mutation of closely linked
genes frequently arises by application of mutagenic agents (Gottschalk and Wolff,
2012). Although there is plenty of evidence for a differential chromosome-breaking
ability of different mutagens, there are so far, no systematic studies regarding the
extent to which pleiotropic effects are associated with mutations induced by a
particular mutagen. Advances in whole genome sequence of the mutants and their
parents will enable precise identification of the genes affected by mutation and
associate them with the expressed mutant phenotype (Caldwell et al., 2004; Jannink,
Lorenz and Iwata, 2010, see also Chapter 8.3).

This 'pleiotropic gene action' is a serious handicap in the practical performance of

mutation breeding. A large number of progressive mutants of different cultivated
plants cannot be used for practical breeding purposes because some negative features
are combined with the useful character. If the whole spectrum of phenotypic
alterations is really due to the action of one single gene, there may be no hope of using
the positive feature for practical breeding, because it cannot be separated from the
negative ones.

However, there is some evidence from experiments that specific detail of such a true
pleiotropic spectrum can be altered by transferring the mutated gene into a different
genetic background including the parent genotype (Gottschalk and Wolff, 2012).
Therefore, if a really important and valuable new character is part of a pleiotropic
spectrum, the mutant should be crossed with a large number of different varieties or
genotypes of the respective species in order to reduce the intensity or the strength of
the negative characters of the spectrum in a particular genotypic composition.

If 'pleiotropic mutation' is due to a deficiency, there is no possibility of a partial repair.

But if such a complex is caused by two or more independently functioning
neighbouring genes it is, in principle, possible to separate the positive and negative
characters by means of crossing and target selection. Some examples are known in
which such a separation could be achieved (Gottschalk and Wolff, 2012). However,
the frequency of such a recombination event will be extremely low because of the
very close linkage of the genes in question. It may therefore be only worthwhile
searching for such rare recombination event for particular important breeding goals
that cannot be achieved by other means.

Finally, it should be mentioned that many characters of a plant organism are controlled
by polygenic systems and that different genes of such a system often develop
distinguishing pleiotropic spectra, as could be shown, for instance, for the numerous
erectoides loci in barley .XF]\ĔVNDet al., 2013; Lundqvist, 2014).

155
156
 6. MUTATION BREEDING FOR VEGETATIVELY PROPAGATED
CROPS

6.1.APPLICATION OF MUTATION TECHNIQUES

A wide range of crops are not propagated through seeds, these are mostly plants grown
for their edible roots, tubers, leaves, and fruits, the propagation is obtained by
multiplying stem, root, stolon, tuber, and fruit cuttings, and even leaf fragments.
Multiplication is thus asexual, and such crops are known as vegetatively propagated
crops (VPCs). These plants include many economically important crops such as
banana, cassava, potato, sugarcane, many ornamentals, fruit trees and other
commodity crops such as tea and rubber. As an example, most edible bananas are
triploid and sterile, thus seedless, hence conventional breeding methods are difficult
if not impossible to implement. Mutation induction offers a useful tool for the
improvement of many VPCs, including banana using in vitro together with
greenhouse propagation techniques, i.e. cuttings and/or grafting.

Conventional plant breeding depends on the availability of utilizable genetic variation

for crossing and selection of desired genotypes. Genetic variability utilised by
conventional breeders is either found within the elite breeding pool (preferred
breeding material), exotic germplasm (second choice) or wild species (third and most
difficult breeding option) and transferred into advanced lines through hybridization
and the natural meiotic processes of recombination. The desired recombinant
genotype is later screened from segregating populations, and then through trials and
selection procedures, a selected potential mutant-line may be released as a new
variety. However, these procedures are restricted to sexually propagated crops (see
Chapter 4). Most VPCs are not suitable for such methods and even when feasible, the
application of cross breeding to VPCs is often too time consuming and remains
difficult since most VPCs exhibit a high level of heterozygosity, aneuploidy and/or
polyploidy and complicated biology (dormancy, long seasonal cycles, long juvenile
stage, fertilization incompatibility, etc.). In addition, it is often impracticable to grow
out large populations that exhibit sufficient genetic variation for screening due to
space, time and cost considerations.

Thus, for VPCs mutation breeding offers an alternative approach in widening the
genetic variability and producing novel traits by increasing the frequency of mutations
(over the spontaneous rates) and inducing desired genetic mutations (such as
unmasking recessive traits) that lead to superior performance and the development of
a new variety. As stated by Donini and Sonnino, (1998), in most of the vegetatively

157
propagated species, hybridization is not an easy option, and induction of mutation is
the only method for improvement.

VPCs are more complicated to handle as they present a wide range of targets (plant
parts), each of which requires specialised methods for the induction of mutation.
However, in recent years there has been new developments in enabling
biotechnologies, particularly in tissue culture, that provide efficient methods for
mutation breeding in VPCs. These include in vitro micro-propagation to raise
populations for mutation induction, mutant screening and mutant line development,
and have been reviewed recently by (Bado et al., 2016).

6.2.SELECTION OF PARENTAL LINES AND MUTAGENIC TREATMENTS

6.2.1. Selection of parental lines

Breeders of all crops need to carefully choose the variety/genotype to be mutagenized;

this involves consideration of its genetic characteristics and uniformity, and the
possible availability of the desired traits in the genetic pool of available germplasm.
In practice, the chosen variety/genotype should be well adapted, have a superior
agronomic performance and require the least number of genetic modifications,
e.g. resistance to a specific disease, resistance to lodging, short and stiff straw, etc.
The parental material is important because the genetic background of the plant
generally defines the range of mutant traits that can be obtained (Suprasanna and
Nakagawa, 2012a). It is especially important in VPCs and the methods used often
involve tissue culture and therefore the selected target genotype must be amenable to
in vitro techniques. Most VPCs are highly heterozygous (Aa in case of diploids, Aaa
for triploid and Aaaa for tetraploids) and a major aim of mutation breeding is to
knock-out the dominant allele to un-mask recessive traits. It should be noted that,
albeit rare dominant mutations can also be produced. There has been much discussion
on the status of ploidy of parent plant material to be used for mutagenesis (van Harten,
1998). In some cases, cultivars with higher ploidy gave higher number of mutants as
in the case of octoploid Dahlia, hexaploid Chrysanthemum and tetraploid Begonia
(Broertjes and van Harten, 1987) . However, it was also observed that in some
polyploids e.g. tetraploid Freesia, no mutants could be recovered. Optimal treatment
conditions will have to be considered carefully on a case-to-case basis and from other
related relevant reports before designing a mutation induction experiment for a VPC.

Plant material for mutagenic treatment should be disease-free (including virus-free),

the plant parts should also be uniform in terms of developmental stage and origin, and
be representative of the chosen clone, variety or genotype. Preferably the materials
should come from or be derived from the same ortet, otherwise materials may need
further cloning in the field, greenhouse or in vitro.

158
There are several plant targets for mutagenic treatment, these include:

x shoot tip or meristem tip from in vivo vegetative buds, stem cuttings, leaves,
petioles etc.;
x adventitious buds from explants of root cuttings, stem cuttings, leaves,
petioles, pedicel or somatic tissues; and
x cuttings from roots, stems, leaves, petioles, pedicels, etc.

In general, freshly induced/formed buds are highly appreciated for mutation induction
of vegetatively propagated crops. Both apical and axillary buds in fixed positions on
the plant and newly formed adventitious buds may be submitted to mutation induction
under specific in vitro or in vivo culture conditions. In all cases, the plant materials to
be treated should be selected according to the aim of the breeding programme, which
is usually to develop a genotype differing by only one single trait from the parental
germplasm (van Harten, 1998).

As stated in the previous chapters mutations induced on meristems, often leads to

chimeric structures, thus it is mandatory when considering vegetatively propagated
crops to take into account the handling of chimeras. This is also closely related to the
preferred mode of propagation, e.g. budding or grafting and the degree of
heterozygosity level of the selected crop. Only uniform and representative material of
the clone or variety should be selected for mutation induction purpose as the aim is to
produce a stable (non-chimeric) mutant.

6.2.2. Population size

In any breeding process the success in obtaining a desired mutant depends on the
ability to develop a sizable population that can be mutagenized in order to allow the
screening for desired traits. Here, the randomness and low occurrence of mutations in
a plant genome needs some consideration. Furthermore, the frequency of mutations
in M1V2 and M1V3 generations is often related to the position of the axillary buds
taken from the M1V1 shoot. It could also vary depending on whether the newly formed
bud was pre-existing on the plant material, prior to irradiation or it arose after
irradiation. Thereby, it could be expected that, in each species or variety, a given
portion of M1V1 shoots will have a higher or a lower chance of recovery of somatic
mutations. Once the most suitable axillary buds have been located preemptive action
may be taken to force their growth either by cutting back the other buds on the shoot
cutting (or by using these axillary buds directly for in propagation through in vitro
culture and grafting – this action is often used for mutagenesis of woody species,
e.g. fruit trees – for further details see pictures 3 and 4.

159
The success in obtaining desired mutants depends on the mutated population size, by
increasing the number of treated plant materials one may expect a higher probability
of success. A population size of 800 treated buds, which could give rise after
vegetative propagation to a population of about 4000 shoots in the M1V2 generation,
is considered adequate. For each dose 50 scions (with two or three buds) are needed,
the same number should be used as a control population, e.g. for apple and cherry fruit
trees at least 2000 M1V2 generation are needed (Donini and Sonnino, 1998; Micke
and Donini, 1993). In another example on chrysanthemum, considering that a stem
cutting of 10cm has about 10 axillary buds, it was therefore defined that 80 – 100
cuttings would be needed for inducing positive mutations
(http://www.fnca.mext.go.jp).

A common induced mutation frequency is in the order of 0.5 percent, i.e. 5 desired
mutants in a population of 1000 plants (Predieri and Di Virgilio, 2007). Theoretically,
one should aim to induce mutations in a minimum of 500 plant propagules, but in
practice at least 800 plant targets appear to be the minimum starting population to
provide a realistic probability for selection, as stated above. This will be followed by
deriving useful information on the number of propagules that can be generated
through vegetative means or regeneration/proliferation rate (in case of in vitro
cultures), number of vegetative multiplication cycles or sub-cultures after irradiation,
probability of well-developed rooting and high frequency plant survival. The number
of vegetative multiplication cycles or sub-cultures after irradiation may vary from a
minimum of 3 to 5 or more, depending on how many chimeric structures remain after
each round. The rooting rate may be only 80 percent of the regular frequency obtained
for control (non-treated) plants. Plant survival can be calculated on the basis of
experience with the specific plant material. Predieri and Di Virgilio, (2007) proposed
the following formula to calculate the number of shoots (X) to be submitted to
mutation induction treatment:
୔
X = [(ୟ×ୠ)×ୡ)×ୢ]

Where: P is the estimated number of plants to be planted in the field; a) the expected
proliferation rate; b) the number of subcultures, c) the expecting rooting percentage,
and d) the expected survival rate.

In another study Danso et al., (1990) applied gamma irradiation at 25 and 30 Gy to

1425 cassava (Manihot esculentus) cuttings, selection at the M1V4 stage produced a
mutant variety named “Tekbankye” which had good poundability, high dry matter
content (40 percent) and low incidence of Cassava Mosaic Virus (CMDV). Vegetative
buds of cocoa (Theobroma cacao) were also irradiated with 15, 20 and 25 Gy and
subsequently grafted onto rootstocks to generate M1V1 shoots. At M1V3 the shoots
were screened for Cocoa Swollen Shoot Virus (CSSV)-resistance, stable mutant lines

160
were selected in the M1V5 generation with high yield and bean quality. These were
then tested in multi-locational on-farm trials in Ghana for over 10 years with no
symptoms of the CSSV virus disease (Danso et al., 1990).

6.2.3. Mutagenic treatments

The choice of mutagenic treatment is related to its effectiveness and efficiency in

terms of mutation frequency, the availability of mutagenic treatments/facilities and he
population size and physical size of the material to be treated. More than 90 percent
of released VPC mutant varieties have been developed using physical mutagens
(http://mvd.iaea.org). Acute, semi-chronic or chronic, and recurrent exposure may be
applied to induce somatic mutations, however chronic irradiation has proven to be
more useful for mutation breeding in vegetatively propagated crops because the
mutants obtained undergo less radiation damaged and may be used directly as a new
cultivar. All types of ionizing radiations have been used, including sparsely ionizing
radiation such as X- and gamma-rays, which can easily penetrate plant tissue, and in
contrast UV light, which has low penetration power is normally used only for small
and sensitive plant parts (e.g. in single cells and thin tissue layers irradiation). Densely
ionizing radiation such as thermal or fast neutrons, usually cause drastic changes
(large deletions and gross chromosomal aberrations), which are frequently
deleterious.

Chemical mutagens, on the other hand, are known to favour micro-mutations i.e. point
mutations, which may be preferred as they are most likely to generate changes in the
DNA structure and function and thus, lead to inheritable mutations. However,
chemical mutagenesis, unlike physical mutagenesis, usually produces wide spread
mutation events, and therefore background mutational load can be an issue. Chemical
mutagens commonly used in VPCs include ethylmethanesulphonate (EMS), N-
(nitrosomethyl) urea (NMU) and N-Nitroso-N-ethylurea (ENU). These alkylating
agents are efficient but are light-sensitive and additional precautionary measures need
to be taken because of their higher volatility. Bulky plant materials, such as bulbs,
stolon fragments, scions etc., are difficult to mutagenize in a reproducible way using
chemical mutagens (Broertjes and van Harten, 1987) Chemical mutagens usually
have low penetration capability into target plant parts and this could explain their low
efficiency in the in vivo mutation systems of VPCs. As stated in Chapter 2, strict health
and safety precautions need to be observed when dealing with chemical mutagens.

6.2.4. Selection of optimal mutagenic dose

One of most critical prerequisites for successful mutation breeding is the

determination of the optimal mutagen dose. The dose required for a particular
experiment depends on the desired effects but may be restricted by undesirable effects
of the mutagenic treatment, which could lead to sterility and/or lethality. There is a

161
strong correlation between the genotype and the sensitivity of the plant material to the
mutagenic treatments in plants. The dose increase causes drastic mutations, such as
chromosomal aberrations, and can cause cell damage in apical meristem thereby lower
doses are usually preferred. So, it is recommended to conduct preliminary assays to
determine the appropriate doses for each plant material. Radiation and chemical
sensitivity tests should always be carried out to determine the mutagen dose that
results in a 50 percent reduction in plant height, root initiation, etc., this value is
known as RD50 or EMD (Efficient Mutation Dose) and is widely used to predict the
most effective and most efficient mutagen dose (Figures 6.1a and 6.1b). In practice, a
breeder applying irradiation treatment on vegetatively propagated crops may decide
to settle for a growth reduction of 30 – 50 percent (RD30 – 50) for M1V1 plants or a
survival rate of 40 – 60 percent (LD40 – 60) depending on the sensitivity of the plant
material.

It is generally advised to use the doses which correspond to 60 percent (higher dose)
and 40 percent (lower dose) of the determined LD50 and for each dose and to treat
30 – 50 shoot meristems bud-scions, stem cuttings, or rooted scions, etc., at a time
(Donini and Sonnino, 1998). An equal number of control materials for the comparison
should be planted as the same time. Measurement on shoot length reduction, of in vivo
treated material should be recorded 30 – 60 days after the treatments (Bado et al.,
2015), depending on the species, for example 60 days are needed for the evaluation
of the shoot length reduction for sweet cherry (Kunter et al., 2012). An extensive table
showing radio-sensitivity doses and estimated of LD50 or RD50, of various plant
species submitted to fast neutrons and to both acute and chronic gamma irradiation
can be found in (Shu et al., 2012). Table 6.1 below presents some chemical mutagens
and the recommended concentrations and table 6.2 summarises the recommended
doses for gamma-rays treatment of vegetative propagules used for VPS mutation
induction (Suprasanna et al., 2012; Bado et al., 2015).

162
Figure 6.1a. Cassava stem cuttings of the variety Sepang 1 (S-1) readied for irradiation using gamma-
rays at different doses; 0, 10, 20, 30, 40, 50, 60, and 70 Gy. Courtesy of F. Ahmad, M. Akil, N. Talib
and R. Ibrahim.

Figure 6.1b. Effects of gamma irradiation on budding, shoot length, and leaf size on cassava var.
Sepang 1 (S-1) stem cuttings with different doses of gamma rays 20 days after planting. Courtesy of F.
Ahmad, M. Akil, N. Talib and R. Ibrahim.

163
TABLE 6.1. EXAMPLES OF CHEMICAL MUTAGENS AND CONCENTRATIONS USED FOR
MUTATION INDUCTION IN VPCS (adapted from Donini and Sonnino, 1998)

Recommended Treatment
Species Plant material Mutagen Concentration
Potato (Solanum tuberosum) Buds on Tubers EMS 100 – 500 ppm
Grape (Vitis vinifera) Dormant buds EMS 0.15 – 0.20%
Sweet potato (Ipomea batatas) Shoot tips EMS 0.5%
Apple (Malus domestica) Growing shoots EMS 1%
Carnation (Dianthis caryophyllus) Rooted cuttings EMS 2.5%
Rose (Rosa spp.) Budwood EMS 2.5%

TABLE 6.2. ORNAMENTAL PLANTS IN WHICH VEGETATIVE PROPAGULES ARE USED

FOR MUTATION INDUCTION (Donini and Sonnino, 1998).

Plant Vegetative propagules Mutagen - Gamma-rays

Amaryllis Bulb 250 rad* – 5 Krad
Bougainvillea Stem cuttings 250 – 1250 rad
Canna Rhizome 2 and 4 Krad
Gerbera Rooted plantlet 1 and 2 Krad
Gladiolus Bulb 250 rad – 5 Krad
Hibiscus Stem cuttings 1 – 2 Krad
Narcissus tazetta Bulb 250, 500, 750 rad
Perennial portulaca Stem cuttings 250 – 1250 rad
Polianthus tuberosa Bulb 250 rad – 8 Krad
Rosa spp. Stem with budding eyes 2 – 6 Krad
Tagetes erecta Rooted cuttings 500 rad – 2 Krad
Lantana depressa Stem cutting 1 – 4 Krad
Chrysanthemum sp. Rooted cuttings 15, 20, 25 Gy

*1 Gy=100 rad

6.3.CHIMERAS

In previous chapters the origin and structure of chimeras have been largely described
and discussed. Thus, this chapter will focus on the handling of chimeric structures in
vegetatively propagated crops (VPCs).

164
Both solid mutants and periclinal chimeras can be used as new varieties in VPCs
(Suprasanna and Nakagawa, 2012a). Experiments using appropriate materials must
be carried out to clarify (disassociated) the chimeric situation. For VPCs, various
methods have been developed which involve tissue isolation and dissection during
post-mutagenesis aiming at reducing the genotypic complexity of the resulting plants.

In banana, to dissociate chimeras, Roux et al., (2001) evaluated three different in vitro
propagation systems (shoot-tip culture technique, multi-apexing culture technique and
corm slice culture technique). The average percentage of cyto-chimeras was reduced
from 100 percent to 36 percent after three sub-cultures using shoot tip culture, from
100 percent to 24 percent when propagating by the corm slice culture technique, and
from 100 percent to 8 percent after the same number of sub-cultures using the multi-
apexing technique. Although none of the systems studied eliminated chimerism
completely, the study showed the possibilities to reduce chimeras depending on the
type of shoot produced (axillary or adventitious) and the multiplication rate (number
of new shoots produced per subculture). Nevertheless, in all cases after three
subcultures the proportion of the number of chimeras tends to stabilize
(see Chapter 2).

6.4.HANDLING MUTANT POPULATIONS AND RELEASE OF NEW

VARIETIES

Since mutagen application causes some physiological and genetic injury to the plants,
M1V1 plants should be grown in non-stressed conditions otherwise the M1V2
population obtained may be insufficient for selection and mutant line development.
Care should be taken to ensure optimal watering, temperature; lighting and fertiliser
(particularly nitrogen) should also be optimised as described for growing the first M1
generation of seed propagated crop mutation induction.

In general the development of VPC mutant populations is done entirely by vegetative

propagation and flowering, meiosis and seed production are usually avoided, so that
the mutation will be inherited only by the lineage of the cells which have been mutated
with no other change for the rest of the plant itself (Suprasanna and Nakagawa, 2012).
The entire process is summarized in Figure 2 and described below

First year (M1V1): M1V1 generation is often not suitable for mutant selection since it
is not possible to detect homohistant mutants due to chimerism, which then need to
be dissolved (see Chapter 8-A). As the detection and isolation of induced somatic
mutations may present considerable difficulties, appropriate selection methods must
be used prior to clonal propagation.

165
Second year (M1V2): Chimeric structures may continue in the M1V2 generation so
this generation must be closely monitored in identifying any ‘deviant’ from the
standard material. Further visual selection and measurements are needed to identify
mutants such as plant growth habit, internode length, branching type, fruit
characteristics and number of buds. Selected individual M1V2 mutated shoots are
propagated for further observation to confirm their characteristics and to assess
uniformity and stability in M1V3 generation (Drake et al., 1998).

Third year (M1V3): Preliminary evaluation can start here, since this generation
should have solid mutants whose uniformity must be assessed. Non-uniform mutant
clones must undergo a further round of propagation to reach uniformity. In this
generation the evaluation of main desired traits, which may include yield, quality,
biochemical/mineral content, seed/fruit size, flower traits, weight could be undertaken
or delayed until more advanced generations.

Fourth to Ninth year (M1V4 to M1V9): Uniform clones may be propagated and
planted in experimental trials to test their performance for desired traits such as
biotic/abiotic stress. As early as in M1V4 generation, replicated trials of selected
mutants may be conducted using parental or local varieties as checks. The M1V5 and
M1V6 generations can be used in multi-locational trials and tested for performance in
a range of environments and agronomic traits.

Final assessment can be made in M1V9 to M1V10 generations depending on plant

species, the desired mutant clone or clones will be released as a new improved mutant
variety. With ornamental plants, selection of new mutant varieties can be quicker,
depending on plant type, mutagen and explants treated as the desired changes
(different flower colours and shapes and growth habit) may be reached at any time
during the process. Many mutant varieties of ornamentals such as achimenes,
chrysanthemum, carnation and rose have been reported
(http://mvgs.iaea.org/Search.aspx). Most of these mutant varieties were derived from
irradiating rooted stem cuttings, detached leaves and dormant plants (Ahloowalia,
1998).

In fact, in the case of flowers, chimeras might also be used for commercial purposes
due to their peculiar phenotypes. Many mutant varieties had been developed in
vegetatively propagated fruit trees (see example on sweet cherry below). One of the
specific and classic success is seedless fruit in citrus, grapefruit, lemon, mandarin and
oranges, which were obtained by exposing bud wood to gamma rays, thermal
neutrons, or X-rays (Bado et al., 2016). Figure 6.4 shows the use of budding and
grafting in developing mutant lines.

Mutation breeding programmes aimed at improving abiotic and biotic stress tolerance
require appropriate screening methods. It is also important to note that such methods

166
have to be rapid and economical, considering a large number of individual plants to
be screened in the post irradiation-handling from M1V2 or later stages. Any valuable
pre-screening techniques which can reduce the number of potential mutants to
reasonable levels prior to replicated field trialling may be considered.

6.5.SCREENING TECHNIQUES FOR ABIOTIC STRESS

Major stresses include salinity, drought, alkalinity, heavy metals and high/low
temperature. Screening methods need to be devised for rapid, sensitive, efficient and
preferably non-destructive testing. Screening for abiotic stress tolerance can be
evaluated by assessing variations in chlorophyll fluorescence, net photosynthetic rate,
transpiration rate, stomatal conductance, water-use efficiency, free proline content,
etc. Besides, physiological parameters, biochemical and molecular markers may, also,
be developed (see Chapter 8-C). Two approaches are presented below.

Indirect methods of screening for abiotic stress tolerance can also be conducted
in vitro, in the greenhouse or in field conditions. Various physiological and
biochemical indicators are suggested, for example photosynthesis rate, stomatal
conductance, chlorophyll fluorescence, lipid peroxidation, electrolyte leakage and
relative water content. Chlorophyll fluorescence and thermal imaging are well-
established, powerful, non-destructive, and rapid techniques for detecting and
diagnosing plant stresses in the field by providing information on both stomatal and
photosynthesis-related parameters, which are the key factors that determine plant yield
(Li, Zhang and Huang, 2014).

In comparison to screening for stress tolerance under in vivo conditions (field or

greenhouse), screening under controlled conditions can be advantageous by using
stress inducing agents. Pre-field screening to reduce the number of potential
candidates to reasonable numbers prior to replicated field trialling is also valuable (for
practical examples, which may be applied to VPCs see, (Bado et al., 2016).

167
 Phases Material handling Progress
Induction of Mutagenic application: Plant organ treated:
mutations and X- or ܵ-rays, chemical mutagens.
handling of chimeras Shoot meristems bulbs, Adventitious buds, somatic
tubers, cuttings, etc. embryo, single cell.

M1V1 generation
x Vegetative propagation
cutting back of the M1V1
shoot or bud grafting. Chimeric plant/shoot Uniform mutated plant
(mericlinal, periclinal) (homohistant)
x Growth in vitro of the shoot
and micropropagation of
axillary buds to eliminate
chimerism.

M1V2 generation
Selection of induced somatic Identify uniformly mutated Verify genetic uniformity of
mutations, cutting back of non- scions, branch, and tree clones.
mutated shoots and M1V2 shoots plant.
from chimeric plants, further
micro-propagation of axillary
buds and establishment of rooted
plantlets.

Identification of
induced mutations M1V3 generation
Selection and propagation, Achieving the uniformity Establish and test clones.
preliminary evaluation of the within a mutated clone.
mutant plants.

Agronomical M1V4 up to M1V9 generations

evaluation of the
mutants Vegetative propagation, Evaluation of genetic stability, yield and traits and testing in
maintenance of selected clones multi-location trials for identification of candidate mutants
and evaluation based on for release.
agronomic performance for
desired traits.
Release of the new
variety/-ies M1 V9 generation

Official testing and releasing of the mutant variety/-ies

Figure 6.2. Mutation breeding scheme for the improvement of VPCs (as modified from Donini and
Sonnino, 1998).

168
 Irradiation of M0 seeds with optimal dose

Irradiated seeds advanced as M1 Seeds from the M1 plants are The procedure is repeated for M3,
population sown again as M2 population M4 populations to stabilise any
putative mutation

169
a) seed propagated crops (ex: 60Co gamma ray irradation)

Irradiation of cuttings Development of irradiated Development of the new The procedure is repeated
axillary bud into M1V1 shoot axillary bud into M1V2 shoot several times to M1V3,
M1V4, etc…
b) Vegetatively propagated crops
Figure 6.3(a,b). General mutation induction principles for seed versus VPCs. (as modified from Broertjes and van Harten, 1987).
 1 - Each end of year, irradiated annual shoots are prepared for grafting onto parental root stocks for the new
generation the following year (1 to 8)

Irradiation of buds M1V1 M1V2 M1Vn

on cuttings located Budding
from the mid- Pruning
section of the plant back
stem

1-8 1-8

New shoots (M1V2) are 1 a. M1V1 scions are grafted or The process may be repeated
pruned back. clefted again for M1V2 several times

170
2. Pruning back of M1V1 1b. Each scion may be monitored and new shoots from M1V2 to
scions M1Vn assessed for mutations

Grafting of
irradiated scions
onto root-stock Each scion is pruned back to force budding and new shoots monitored from M1V2 to M1Vn.

Figure 6.4. Isolation of somatic mutations in vegetatively propagated fruit trees (as modified from Broertjes and van Harten, 1987).
6.6.SCREENING TECHNIQUES FOR BIOTIC STRESS

Vegetatively propagated crops such as banana pose unique problems compared to

cereals because they have a reduced genetic diversity as they have been grown in
monoculture for years and cannot be selfed easily nor cross pollinated to enhance
variation. Furthermore, because of its triploid parthenocarpic nature, bananas do not
produce seeds and conventional breeding is thus a long process that often results in
varieties not accepted by consumers.

Fusarium oxysporum f. sp. cubense is a fungus, which threatens the Cavendish banana
production worldwide. A mutation breeding experiment for resistance to Fusarium
wilt in banana clones was carried out by Mak et al., (2004) using the double tray
technique to screen banana plantlets in greenhouse conditions (Figure 5). The
technique consists of a perforated upper tray which contains sterilized sand media to
grow banana plantlets and a lower tray to accumulate nutrient solution and pathogen
derived wash. Meristem derived from two-months old plantlets (10-15cm long) of
mutant banana clones ‘Intan’ (Pisang Berangan, AAA), 'Gold Finger' (AAAB),
'Novaria' (Cavendish, AAA), and 'Mutiara' (an improved Pisang Rastali, AAB) were
tested against Fusarium oxysproum f. sp. cubense (FOC) race 4 under greenhouse
conditions. Susceptible plants showed both foliage and rhizome a symptom within
10 – 30 days, this technique is therefore accepted as a rapid method for early screening
against Fusarium wilt disease.

First layer: plantlets are placed in

sand filled small pots with a hole in
the bottom. The pots are then
placed on top of a medium
containing the fungus filtrate,
which is then absorbed by the
roots.

Second layer: medium

containing the fungus filtrate,
which is in contact with the
roots.

Figure 6.5. Double layer method modified from Mak et al., (2004).

171
6.7.PRACTICAL EXAMPLES OF THE APPLICATIONS OF MUTATION
BREEDING IN VPCS

There are many varied methods for mutation breeding in VPCs. Several of these
involve in vitro techniques and protocols are presented for banana, sugarcane and
melon in Chapter 8-A. Here we show an example in a fruit crop; sweet cherry (Figures
6.6a,b) and a novel method recently developed using micro-tubers in potato (Figure
6.7a,b).

6.7.1. Practical example of mutation breeding in Cherry

Figure 6.6a. New mutant sweet cherry (Prumus avium) varieties: ALDAMLA and BURAK, Turkey -
7XUNLVK $WRPLF (QHUJ\ $XWKRULW\ DQG $WDWUN +RUWLFXOWXUDO 5HVHDUFK &HQWHU % .XQWHU 0 %Dú
Y. .DQWR÷OXDQG0%XUDN2013.

172
 25 G
50 Gy
45 Gy
40 Gy

30 Gyy
35 Gy
2. EMD

Gyy
1. Irradiation
Determination

Mutant Variety
BURAK
6. M1V5 –
M1V8 3. M1V1

Mutant Variety
ALDAMLA
5. Juvenile 4. M2V2
Phase

Figure 6.6b. summary of the procedures for sweet cherry mutation induction.

173
 6.7.2. Practical example of mutation breeding in potato

3
2
8 4

5
7
6

Figure 6.7a. Mutation Breeding for Potato (Solanum tuberasum L.) variety NAHITA improvement
using gamma ray irradiation. 1. Radiosensitivity test conducted directly in the field (0, 25, 35, 45 Gy)
and the Effective dose determined; 2. The parental potato variety: Marfona is then irradiated with the
selected effective dose: 35 Gy; 3. The M1 V1 Generation was planted together with the parental variety
in the field and visual observations for abnormalities conducted: changes in flower types and colours,
in tubers sizes, shapes and numbers recorded; 4. All tubers from an individual M1V1 plant were planted
in row as M1V2 for further observation of uniformity and yield increase potential; 5. The propagation
is pursued following the same process from M1V3 to M1V8 while the evaluation of the quality of the
mutants is thoroughly conducted and, for example, all early yellow sprouting lines discarded; 6. Some
outperforming mutant clones were then micropropagated in vitro for population increase and
transferred to the greenhouse; 7. The selected mutants with excellent yield and tuber quality are
confirmed and the application for release placed in M1V9; 8. Released of the new potato mutant variety:
Nahita. (Sekerci et al 2016-Turkey).

174
Figure 6.7b. Summary of the procedures for potato mutation breeding. ùHNHUFLHWDO

175
176
 7. MAJOR TRAITS TO BE IMPROVED BY MUTATION BREEDING

The strategies utilized to select and improve desirable and specific characteristics in
a plant breeding vary with plant species, the environment where the plant is to be
cultivated, the farmers’ cultivation methods, and the utility and demand of the end-
product. The objectives of a mutation breeding programme are basically the same as
those of any other breeding methods, the advantages are that mutation breeding aims
to improve and already elite genotype usually for a single trait (improved yield,
quality, stress resistance, agronomy, etc.). Mutation induction and detection of desired
mutants represents an accelerated means of reaching the desired superior variety
(Bado et al., 2015) . In theory all genetic traits can be targeted by mutation breeding
and in this chapter, we will attempt to display the wide range of mutant varieties that
have been produced in a wide range of crops for a wide range of traits, world-wide.

7.1.YIELD IMPROVEMENT

7.1.1. Yield and yield components

Stable and high yield potential over a range of environmental conditions is probably
the most important objective of most plant breeding programmes. Yield is a complex
trait strongly influenced by other breeding objectives, such as, plant architecture,
maturity, nitrogen utilization efficiency, resistance to biotic and abiotic stresses, etc.
It is difficult to use mutation breeding to improve the yield potential of crops that are
well established and which have been subject to intense and refined breeding over
long periods of time.

Despite the difficulty of detecting yield mutants there is no doubt about their existence
and a number of mutant varieties with increased yield have been released (Table 7.1).
Positive yield mutations are formed at a low frequency, perhaps 1/1000 to 1/500 plants
in an M2 population (Saeed and Hassan, 2009). Thus, when planning mutation
induction for yield improvement large populations are needed in order to increase the
probability of finding yield mutants.

Since yield is so highly influenced by environmental fluctuations, one cannot expect

to recognize mutants in yield performance from observations on a single-plant basis.
Selection methods, therefore, deserve more attention than do the selection methods
for isolating mutants for qualitative characters. The selection method for a character
like yield is further complicated by the fact that spontaneous or induced mutants can
react differently from the mother genotype to environmental changes. However,
changes in genotype × environment interactions may be utilized in practical
cultivation of the crops. If, for instance, the environmental changes are represented by

177
placement or level of nitrogen dressing, the problem is how to pick out those mutants
that compete best at a given placement or at a particular level of fertilizer. Another
important point is that of competition between plants; most cultivars are grown as a
homogeneous genotype and do not compete with other genotypes; therefore,
screening must be done with this in mind. For example, if an assortment of wild type
tall plants grown together with induced semi-dwarf plants, the latter will be shaded
out, but this interaction will never occur in normal agricultural conditions where often
a semi-dwarf cultivar has a yield advantage.

There are two different selection methods for mutants with improved yield. In the first
method, mutants for qualitative characters are selected and isolated. Their yielding
ability is then tested in subsequent generations. The idea behind this method is that
some genes for qualitative characters may exhibit a positive pleiotropic effect on
yield, or that changes in such genes may be linked with other mutations affecting
yield. The method has led to positive results in many cases. As an example, the
erectoides mutants in barley should be mentioned. The erectoides character is
associated with high yield in some cases, especially under heavy nitrogen fertilization,
however, the method has clear limitations. This erectoides character is governed by
many loci in which mutations can occur, and not all of them are associated with high
productivity. From general reasoning it should be expected that yield is governed by
many loci, each of them having a relatively small effect. Not all of them, however,
can be expected to be associated with visible qualitative changes in the phenotype.

The second selection method is direct, and thereby considers the limitations
mentioned above. It is analogous with selection methods applied in populations
derived from cross breeding; the selection procedure starts with progeny testing of
individual plants. Since groups of plants are needed to detect yield mutants, the
progeny testing can only start in the M3 families at the earliest.

In practical breeding work on self-pollinating crops, it may sometimes, be advisable

to delay the progeny testing until later generations, M5 or M6. By that stage a fairly
high degree of homozygosity (uniformity) is attained, and selected families need not
be re-selected before they are handed over for testing on a large scale in field trials.

Since the influence of the environment on yield is so high, efforts should be made to
minimize this effect as much as possible, particularly in experiments with single-plant
progenies. By large spacing of the population plants to be tested sufficient seeds can
be obtained for replications in the experiment with single-plant progenies.
Experiments of this type will, in general, have a great number of families. To control
the experimental error special designs are required, such as the widely used split-plot
design introduced by Gaul (1964). In this case each plot includes a control row and
test-rows from single plants of untreated and treated material of the same genotype.

178
All measurements are taken on a split-plot basis. Data from such an experiment can
be used to construct distribution curves for continuously variable characters (Kusaksiz
and Dere, 2010).

As far as yield is concerned, the breeding aim is almost always to increase the mean
(directional selection). The left-hand half of the distribution curve is therefore of little
or no interest. If yield ability is considered alone, only those mutant families that yield
more than the mother genotype are of interest. The practical value of a variety does,
however, depend on the performance in several characters, all of which should be
considered in a selection experiment. This and other problems related to selection
intensity are exactly the same in mutation trials as they are in cross breeding.

When selection is continued in the following generations, more seeds are available
from each family. The plot size can therefore be increased and replication including
multi-location trials can begin, and the screening between families can thereby be
more precise. Even if an experiment with M4 families can be made quite precisely,
one should not place too much emphasis on the results from single experiments. The
reason for this is, again, genotype × environment interaction. Genotype × year
interactions are also common among families derived from mutagen-treated material.
This type of interaction cannot be utilized in practice and must be treated as
experimental error. For later selection, e.g. in M4 and M5, it is better to base the
selection on the mean performance over two or more years.

Statements presented so far, apply mainly to the self-pollinating annual crop species.
At present there are far less experimental results and practical experiences with
induction of mutational changes in yield ability and other quantitative characters for
cross-pollinating, perennial and vegetatively propagated crops. For vegetatively
propagated species, e.g. potato, variability experiments of the type described should
also be a suitable basis for the initiation of selection among clones (for further
information see Chapter 6).

179
 7.1.2. Examples of mutants with improved yield
TABLE 7.1. EXAMPLES OF MUTANTS WITH IMPROVED YIELD

Crop Trait Method Landmark Reference

mutants/cultivars
(country)
Rice High yielding gamma Zhefu – 8 cultivars: MBNL* Nos. 25 and
Oryza sativa rays 1985 – 2005 (China) 26, 1985

Bread wheat High yielding Jauhar – 78. 1979 Ahloowalia et al.,

Triticum aestivum (Pakistan) 2004

Barley Semi-dwarf Gamma Golden Promise Sigurbjörnsson and

Hordeum vulgare (GPert), (UK) Micke, 1974
malting quality
Barley Semi-dwarf X-ray Diamant Ahloowalia,
Hordeum vulgare (sd1) (Czech Republic) Maluszynski and
Nichterlein, 2004
Banana High yielding Gamma Al Beely (Sudan) PBGNL** Nos. 16
Musa sp. and 17, 2006

Groundnut High yielding Gamma TAG24 (India) Kale, Badigannavar

Arachis hypogea ray and Murty, 1999

Blackgram High yielding Gamma TAU-1 (India) (Ahloowalia,

Vigna mundo L. ray Maluszynski and
Nichterlein, 2004)
Cotton High yielding Gamma NIAB 78 Ahloowalia,
Gossypium sp. ray (Pakistan) Maluszynski and
Nichterlein, 2004
*MBNL: Mutation Breeding Newsletter; **PBGNL: Plant Breeding and Genetics Newsletter

7.2.TOLERANCE TO ABIOTIC STRESS

Abiotic stresses encompass several unfavourable environmental conditions such as

soil salinity, drought, extreme pH, flooding and harsh weather. The approaches
utilized in plant mutation breeding for such traits are often quite simple. Recently, two
candidate salt tolerant rice lines induced using carbon and neon ion beams have been
reported (Abe et al., 2007).

Though many of the physiological mechanisms of abiotic tolerance are unknown, the
threat of global warming and climate change forces new approaches in adapting crops
to fluctuating environments. Huge numbers of potentially useful genotypes have been

180
generated from mutation breeding research and breeding programmes and are
available in germplasm collections, e.g. tolerance to cold, heat, day length and
drought, available in China, Japan, USA, etc., (http://mvd.iaea.org).

The methods of mutation breeding have hitherto been used only to a very small degree
for developing varieties with tolerance to low temperature, heat, drought and salinity
of the soil. Therefore, a brief survey of the results of conventional breeding methods
and some information on the genetic bases of the tolerance are given.

7.2.1. Drought

Climate change is responsible not only for global temperature increase but also for
region-specific increases or decreases in rainfall. Water shortage has a negative
impact on agricultural production and this is particularly acute in developing
countries. Crop plants cannot grow without water, it is essential for all stages of crop
development from germination, to vegetative growth and reproductive periods (fruit
and seed development). Changes in rainfall (duration and timing) can impact all stages
in crop production and crops of all climatic regions, arid, temperate, tropical etc. Table
7.2 provides some examples of released crop mutant varieties with drought tolerance.

7.2.2. Salinity

There are two ways to develop new mutant varieties with salinity tolerance, one is to
mutate a variety that is high yielding but susceptible to salinity, the other is to mutate
a variety that is low yielding but tolerant to salinity, such as a traditional local variety
grown in the affected area. In such a case, because it is much easier to improve
agronomic performance than to enhance salinity tolerance, the latter is the better
option. However, it is important in such situation to carefully choose the most suitable
mutagen and technology to achieve this specific breeding goal. The final step is to
deploy an efficient method for mutant screening and validation. As in all mutation
breeding programmes, following selection, it is important to verify and confirm if the
selected mutated trait is heritable.

7.2.3. Temperature

Much work has been done during the past decades on tolerance to low temperature,
particularly in cereals, potato, fruit and forest plants. The genetic basis of frost
resistance is very complicated and evidently not uniform considering the different
types of cultivated plants. In barley winter hardiness is controlled not only by
dominant but also by recessive genes and there is a correlation between the degree of
dominance and the average winter hardiness. According to some authors, the action
of a polygenic system must be assumed, probably supplemented by series of multiple
alleles. Also, in bread wheat whole groups of different genes are effective and show

181
transgressive inheritance. Winter hardiness seems to be dominant in wheat and there
is obviously a variance in the degree of dominance in relation to the parents used for
hybridizations. The action of seven dominant genes influencing the character
'tolerance to low temperature' in rice (Oryza sativa) has been identified (Cruz et al.,
2013).

Further indications of a broad genetically conditioned variance in respect of this

character were observed in cabbage (Brassica olerocea), radish (Raphanus sativus),
buckwheat (Fagopyrum spp.), pea (Pisum sativum), alfalfa (Medicago sativa) and in
different species of lupin (Lupinus spp.). Moreover, it should be mentioned that
certain wild potatoes and some primitive cultivated potatoes are characterized by a
considerably high degree of winter hardiness. A polygenic system could also be found
in these species and the feature 'winter hardiness' is inherited partly in a dominant,
partly in an intermediate way. Furthermore, it could be shown that genes causing a
certain tolerance to low temperature are even present in the genomes of cultivated
plants that grow in relatively warm zones of the earth. This may be valid with regard
to cotton (Gossypium sp.), tobacco (Nicotiana sp.), maize (Zea mays) and to some
wild-growing tomato species such as tomato (Lycopersicon hirsutum and
L. peruvianum).

The first results, using methods of mutation breeding for abiotic stress were for
resistance to cold in barley. Winter varieties of barley were produced by the
application of X-rays to summer varieties during the early epoch of mutation research
(van Harten, 1998): winter hardiness was found to be a recessive character in this
plant material. There are also winter-hardy mutants of oat (Avena sativa) displaying
more ascorbic acid compared with their initial parental lines, which were susceptible
to low temperatures.

A gamma-ray induced mutant strain of soybean (Glycine max) was obtained that can
germinate at a temperature of 4°C, while the germinating temperature of normal
soybean varieties is at about 8°C (Khan and Tyagi, 2013). There is no doubt that this
mutant will represent a valuable basic material for developing strains and varieties
that may be cultivated in cooler regions.

A comparable situation exists in the field of tolerance to heat. Heat tolerant genotypes
have been selected in several crops. In fact, this type of tolerance is being looked at
with enhanced scrutiny in the wake of climate change and its probable effects on
agriculture.

182
TABLE 7.2. EXAMPLES OF MUTANTS WITH IMPROVED ABIOTIC STRESS TOLERANCE

Crop Trait Method Landmark Reference

mutant/cultivar
(country)
Rice Salt tolerance Gamma ray NIAB-IRRI-9 MBNL No. 45, 2001
Oryza sativa (India)

Rice Salt tolerance Ion beam Japan Abe et al., 2007

Oryza sativa irradiation

Rice Salt tolerance Gamma VND95-20 Do et al., 2009

Oryza sativa (Viet Nam)
Bread wheat Drought Gamma Njoro BW1 IAEA Bulletin, 50 – 1
Triticum aestivum tolerance (Kenya)
Maize Drought Gamma ray Kneja 698W PMR*, 2012
Zea mays tolerance (Bulgaria)

Barley High altitude Gamma UNA La Molina MBNL No. 43, 1997
Hordeum vulgare (harsh weather), (Peru) Gomez-Pando et al.,
early maturity 2009
Amaranth High altitude Gamma Centenario Gómez-Pando et al.,
Amaranthus (harsh weather) (Peru) 2009
cuadatus L.
Rice Tolerance to cold Gamma ray Kahmir Basmati Ahloowalia,
Oryza sativa (Pakistan) Maluszynski and
Nichterlein, 2004
Rice Tolerance to heat Gamma ray Nagina 22 (India) Poli et al., 2013
Oryza sativa
Soybean Tolerance to Gamma ray Heinong 26 Khan and Tyagi, 2013
Glycine max cold, drought and (China)
water logging
*PMR: Plant Mutation Report

7.3.TOLERANCE/RESISTANCE TO BIOTIC STRESS

7.3.1. Resistance to disease

Biotic stresses are primarily diseases caused by fungi, bacteria, and viruses, and the
damages induced by insects, animals, nematodes, weeds, and any other biological
causes. Mutation breeding has been successful in improving disease resistance; the
success in tackling pest resistance has yet to be taken to such level. Diseases involve
a complex inter-play between a host plant and a pathogen. Therefore, the
resistance/susceptibility response can involve several components; this implies that
there are many targets for mutation improvement. For example, induced mutations

183
may change the interaction and inhibit certain steps in the mechanism of infection.
Numerous mutants have been developed through mutation induction, showing
enhanced resistance to various diseases: virus, bacterial, and to some extent fungi
(http://mvd.iaea.org/ ; Lebeda and Svabova, 2010).

Unlike diseases, there is little interaction between host plants and their insect pests, as
one pest may attack other plant species or even different plant genera. This predator-
to-host interaction may explain why there are fewer examples of mutant varieties
carrying induced pest resistance. Tolerance towards insects is a quantitative reaction
and may include characteristics such as plant vigour, the ability to produce many
shoots and many roots, as well as strength of stem tissue and avoidance (little or no
vegetative growth when insect pests are prevalent). Recently, there has been some
progress in identifying resistance gene(s) to brown plant hopper in rice (Fujita, Kohli
and Horgan, 2013), as such, it may be possible to induce resistant rice mutants
efficiently by targeting the identified gene(s) related to the resistance mechanism. In
fact, this aspect was the foundation for transgenic breeding for pest resistance success,
e.g. GMO crops carrying the Bt toxin gene.

The constant challenge in breeding for disease and pest resistance is to deal with and
overcome the development of new aggressive strains of the pathogens. The advances
in molecular technology and recent findings in cloning of disease resistance (R) genes
should allow the improvement of crop disease resistance by applying a holistic
approach including traditional breeding, genomics, transgenic deployment, and/or
mutagenesis tools. The discovery of genetic resistance in plants is attributed to Orton
who at the end of the last century selected cotton for resistance to wilt caused by
Fusarium oxysporum f. sp. vasinfectum as cited by Epstein et al., (2017). For recent
reviews on the genetics of resistance (host) and pathogenicity (disease organism) see
Boyd et al., 2013; and Servin et al., 2015.

But as the disease is the product of an interaction between the host organism and the
pathogenic organism, genetic variability can be expressed in the former as well as in
the latter. In the pathogen the variability is expressed by means of genes for virulence
or avirulence whose differentiation on a set of differential varieties allows the
identification of the phenotype known as 'physiologic' or 'pathogenic races'. This kind
of interaction is identified in the host as specific resistance, which in many instances,
has proven to be of a temporary value in breeding because the pathogenic organism
creates new virulent races even while the plant breeder is developing novel resistant
varieties, hence creating a dynamic equilibrium. Thus, when dealing with plant
disease resistance one has to also consider both partners, and have a holistic approach
including plant pathology and other related fields. Barah and Bones, (2015) in an
extensive review of studies on plant-insect interactions using different biological tools
stated that most probably, in the near future, the combination of high-throughput

184
profiling techniques, bioinformatics tools, and published data from ecological studies
will provide ways by which to achieve a comprehensive systems-level understanding
of various aspects related to plant defence responses, which in turn can be used to
develop better crop varieties using all breeding tools currently available including
mutation induction.

7.3.1.1.Induction of mutations and mutant selection

The first report on the induction of mutations for disease resistance is attributed to
Freis and Lein, 1942 as cited by Gupta (1998). These authors isolated a mutant in
Haisa barley simultaneously resistant to three races of powdery mildew in Germany,
as a result of a treatment with X-rays, the survey mentioned about 12 000 progenies.

As in other characters, the success in induced mutagenesis programmes for resistance

to diseases depends upon careful selection of the parent variety, an effective mutagen
treatment and a specific and discriminating screening method. The latter should be
the most reliable and efficient manner of separating susceptible from resistant plants.
For this purpose, it is essential to know the genetic sources of variation of the
pathogen, thus permitting the choice of the most adequate race or races of the
pathogen for inoculation.

From the host-parasite interactions it is evident that a mutation induced in a

susceptible host can be expressed or may remain cryptic, according to the type of
inoculation employed in the screening process. As a result, the mutation frequency
observed will not be identical with that actually induced. By working with a range of
disease races the proportion of cryptic mutations increases and a wider spectrum of
resistance will be detectable. However, later generations will give a higher probability
for selecting a desired resistance mutation since it will appear in more plants,
particularly as the mutations are likely to be recessive.

Both physical and chemical agents have been employed successfully for inducing
disease resistant mutants, although irradiation has been used more frequently
(Table 7.3).

In 2010, the FAO/IAEA Joint Division published a compilation of studies realized

under CRPs and/or TCPs relating to screening mutant and non-mutant crops for
tolerance/resistance to disease, with an emphasis on fungus derived disease (Lebeda
and Svabova, 2010).

7.3.2. Resistance to pests

Since the dawn of agriculture breeders and farmers have succeeded in selecting and
growing crops with various types of resistance in order to ensure thriving crop growth

185
and productivity. The contribution of agronomists and soils scientists has provided
enhanced protection with the use of herbicides and/or insecticides. However, the long
use of these products has weighed on the environment and subsequently on the human
health, thus the development of plant resistant to insects and to herbicides has become
of significance in protecting the environment and human health. Oerke, (2006) in an
extensive review stated that among crops, the total global potential loss due to pests
varied from about 50 percent in wheat to more than 80 percent in cotton production,
26 – 29 percent for soybean, wheat and cotton, and 31, 37 and 40 percent for maize,
rice and potatoes, respectively. The author also noted that overall; weeds produced
the highest potential loss (34 percent), with animal pests and pathogens being less
important (losses of 18 and 16 percent). Thus, enhanced efforts are being assigned in
mutation induction to address that situation. Several mutants resistant to insect have
been registered in the IAEA - MVD database: rice, maize, black gram, white lupine,
fodders beet, rapeseed, hybrid maize and bitter gourd in China, the Russian
Federation, Vietnam and India respectively (http://mvd.iaea.org/).

186
 7.3.3. Examples of mutants with improved biotic tolerance/resistance
TABLE 7.3. EXAMPLES OF MUTANTS WITH IMPROVED BIOTIC STRESS TOLERANCE

Crop Trait Method Landmark Reference

mutant/cultivar
(country)
Rice Resistance to Gamma Camago 8 MBNL No. 43,
Oryza sativa blast and virus ray (Costa Rica) 1997
diseases
Bread wheat Resistance to Gamma EldoNgano-I PBGNL Nos. 32
Triticum aestivum black stem (Kenya) and 33, 2014
rust (Ug99)
Chick pea Blight Gamma Hassan-2K Hassan et al. 2001
Cicer arietinum resistance Pakistan
Lentil Blight Gamma NIAB Sadiq et al., 2008
Lens culinaris resistance MASOOR
Medik. 2006
Pakistan
Barley Powdery X-ray Comtesse MBNL No. 33,
Hordeum vulgare mildew (Germany) 1989 and No. 36,
1990
Barley Mildew EMS Betina Sigurbjornsson and
Hordeum vulgare resistance (France) Micke, 1974

Japanese pear Black spot Gamma Gold Nijisseiki Saito, 2016

Pyrus pyrifolia resistance (Japan)
Nakai
Pepermint Wilt disease Neutron Murray Todd, Green and
Menta × piperita L. resistance irradiation Mitcham; Horner, 1977
(USA)

7.4.QUALITY

7.4.1. Quality, nutrition and functionality

With respect to food, quality usually refers to the composition of organic compounds
produced and stored by plants, such as starch, protein, fatty acid, vitamins, and other
nutrients (Table 7.4). Enhancing the nutritious value of harvestable products is

187
therefore an important objective for plant breeding. The simplest route for modifying
quality traits using mutation is by inducing knock-outs in genes involved in the
metabolic pathways; thus, increasing the synthesis of upstream substances and
decreasing the production and concentration of downstream substances or producing
novel gene products, e.g. altered fatty acid composition of many oil crops via mutating
genes. Apart from oil palm, all oil crops have been improved by mutation (Vollman
and Rajcan, 2009). Canola and others oil producing brassicas have been mutagenized
using either chemical or physical mutagens for the improvement of several important
traits including high quality edible oil, obtained through the reduction of the levels of
toxins (glucosinolates) and of the erucic acid (Cheng, 2014; Singh and Verma, 2015).
Recently, mutation breeding has been used for enhancing bio-availability of important
nutrients in certain crops. For example, crops with low phytic acid content are
preferred because the bioavailability of mineral elements and phosphorus can be
significantly increased. In this regard, two barley mutant varieties have recently been
released for commercial production for additional information see review by
(Raboy, 2009).

Mutation breeding has also been used for breeding crops with special functionality.
In Japan, for example, rice mutant varieties with low glutelin content, such as LGC-1
and its derivatives, have been developed for people who must restrict protein intake,
as it is the case of patients with kidney disease. In Indonesia a mutation induction
programme on sorghum improvements showed that mutated genotypes presented high
nutrition values in terms of protein and starch contents so that it could be used an
alternative food source (Soeranto et al., 2001).

In addition to nutritional traits, quality also relates to medical (e.g. drugs) and
industrial traits (e.g. starches and oils). The first important problem the breeder is
faced with, in connection to quality is to define precisely the desired quality
parameters and to establish their order of priority in his/her particular situation. This
may be challenging as in most breeding programmes yield, disease resistance, and
climatic adaptations are higher on the agenda. This is mostly a matter of importance,
but in other cases it may simply come from the recognition of the fact that it is more
convenient for the breeder to select first for the field characteristics. This will have
also a positive economic effect, as it reduces the number of samples that have to
undergo time-consuming and costly laboratory tests. Restricting the population before
screening for quality traits may, of course, also reduce the probability of identifying
genotypes with the quality factors desired.

The next critical problem is to develop simple, speedy and cheap screening techniques
for the characters to be improved. Such techniques should be simple enough and
require only little material so that primary screening in the early generations can be
done in the breeder’s facilities (greenhouses and fields). In some instances, screening

188
may best be executed by a series of steps, beginning with a coarse, rapid and
sometimes only suggestive analysis, which must be followed by a more vigorous
chemical analysis once more material is available, and ultimately by testing in
production environments (glasshouses or fields) on a larger scale and with methods
close to the end use of the product (e.g. milling, baking tests, brewing, feeding trials,
etc.). The latter tests will almost always be performed at specialized laboratories.

As a first step to improve the quality factors of cultivars, the breeder normally screens
available varieties, breeding lines and germplasm collections. After detecting the
desired traits, the breeder has to choose the most economical way to incorporate such
traits into top varieties and to combine these traits with other desired traits.

As far as screening for nutritional quality of cereals is concerned, the results of

analysing large germplasm collections have not been, to date, very satisfactory. In
particular, the search for genotypes with significantly increased levels of particular
amino acids, such as lysine, has been somewhat disappointing as there is a large yield
penalty. This is understandable since from the standpoint of natural evolution a high
percentage of lysine seems to carry no beneficial fitness value. However, a few
genotypes with higher lysine content in the grains have been found in collections of
maize, barley and sorghum. Induced mutations for quality characters may have an
advantage if they are produced in a good genetic background, suitable for modern
agricultural production and easy to utilize in cross breeding. Seed quality mutants may
be pre-screened rapidly using soft X-ray imaging and near infrared spectroscopy as
demonstrated in barley and rice mutants by Jankowicz-Cieslak et al., (2013).

Quality traits may be subject to modification by environment, for example,

quantitative characters such as protein content are highly susceptible to the influence
of soil, moisture, fertilizer, temperature, and light conditions. This interaction makes
selection difficult where replicated trials cannot yet be carried out. Therefore, caution
is advised in drawing conclusions about the success of selection and results should
only be accepted after robust testing.

7.4.2. Starch

The polysaccharides, of which starch is the principal one, constitute the major
component of cereal grains and are probably the most important compound for energy
storage. Starches as a class of compounds have considerable diversity in both their
starch grain structure and their chemical composition. The classification of
carbohydrates in crops includes sugars (monosaccharides and disaccharides),
oligosaccharides, starch (amylose and amylopectin) and non-starch polysaccharides
usually found in the cell walls, such as pentosans. Most of this diversity is reflected
in nutritional or technological characteristics, such as digestibility, suitability for
bread making and malting, and cooking characteristics. This is important as a quality

189
parameter with regard to both human and animal nutrition. For example, cell wall
polysaccharides of cereal grains contribute to dietary fibre intake and are thus an
important health factor (Lafiandra, Riccardi and Shewry, 2014). In recent years there
has been a renewed interest in the improvement of crops, mainly cereals with
enhanced polysaccharides nutritional qualities (Lafiandra, Riccardi and Shewry,
2014).

In maize great variation is found in the content of starch, varying from 9 – 74 percent.
Rice varieties with high content of resistant starch are being developed in China for
dietary therapy of patients with type 2 diabetes. These varieties have about 10 times
higher resistant starch than normal rice varieties and preliminary tests have shown
they are effective for controlling the glycemic index when added into diets, though
more studies are needed (Shu and others, 2009). It is expected that more foods with
such novel functions will be developed through mutation breeding.

7.4.3. Protein

In many parts of the world, especially in the lower income groups, there is a dietary
shortage of protein and/or certain essential amino acids. Plant products, particularly
cereal grains are the basic food of most of the world's population living in the regions
with large vulnerable populations; they are the main source of protein as well. Cereal
grains are relatively low in protein content and their amino acid composition in
general does not suit human requirements. Therefore, where cereal grains constitute
the major part of a human diet, the supply of protein and essential amino acids will be
inadequate. Increasing the quantity and nutritional quality of grain protein, e.g. to
include essential amino acids such as lysine, is a potential means of combating
malnutrition.

To estimate the protein quantity, a number of methods are available based upon
nitrogen determination including Kjeldahl and Biuret techniques, etc., (De Mey et al.,
2008). Some of them have been more or less automated. In general, however, due to
the cost of the equipment needed the use of these methods has been a limiting factor
for an ambitious programme for protein improvement in crops. Therefore, more rapid
methods have been sought and in recent years many researchers have used the protein
dye binding property as a means of measuring the protein amount. The dye most often
used is acrilan orange, which is specific for the basic amino acids (lysine. histidine,
arginine). The use of this method implies the assumption that the proportion of basic
amino acids to the total protein content is constant. Such an assumption can only be
correct if one deals with material that does not genetically deviate in its proportion of
basic amino acids.

Micro-Kjeldahl, Lowry and Bradford procedures were compared for determining the
protein content of common bean (Phaseolus vulgaris) seeds during their development

190
in order to establish their respective efficiency. Several other nuclear technologies
have been developed for determining the total content as well as the amino acid
distribution of proteins in seeds and general quality, for example in dry beans
(Vakali et al., 2017).

7.4.4. Fats, oils and fatty acids

Vegetable oils are one of the most valuable agricultural products in terms of energy
source, provision of essential fatty acids, carriers of fat soluble vitamins and as
resource for many industrial products. Their value and utility primarily depends upon
their fatty acid composition (Kramer, 2012). Therefore, breeding plant for quantity
and quality of vegetable oils is likewise important. As for other quality traits, rapid,
precise and cheap analytical methods are a requirement. Traditional oil extraction
methods, e.g. Soxhlet method, need to be scaled up and are generally too time-
consuming for screening a larger number of samples. The resonances associated with
the protons of the oil component may be selected from the nuclear magnetic resonance
(NMR) spectrum so that the instrument's 'signal' is a function of seed oil content.
Digital readout of the NMR signal permits rapid screening of small seed samples
(even single seeds) with a relative error in the oil determination of less than 1 percent,
in seeds like soybean, with a very high negative correlation between oil and protein
content, NMR also offers an indirect means of selecting for protein content
(Weir et al., 2005). Liquid chromatography is also used for both quantitative and
qualitative determination of fatty acids. The method is so sensitive that the fatty acids
of as little as half a rapeseed may be determined and thus, has been developed into a
rapid and safe serial test and more rapid screening method (Bromke et al., 2015).

So far, mutation induction for altering fatty acid composition has been the most
frequently carried out and the mutants obtained had increased or decreased values of
linolenic acid without any change in linoleic acid (IAEA-TECDOC-781, 1994).

7.4.5. Toxins and anti-nutritional factors

Many plant species produce toxic, noxious or bad tasting substances and store them
in their tissues in such high concentrations that animals are repelled from eating
leaves, seeds or other organs. However, careful investigations reveal that some
variability always exists within and between species with regard to the content and
genotypes can be found that are practically free from noxious substances. A rather
reasonable theory assumes that the wild-type plants originally contained only
negligible amounts of bitter or toxic substances but that spontaneous mutations
induced a metabolic block which led to enrichment and storage of such compounds.

Breeding has been used in an attempt to revert to low 'wild-type' toxin types, but it
has so far been successful only in a few cases. A more serious problem than the

191
induction process itself, is the lack of efficient methods for the screening of large plant
populations for the desired character. Furthermore, the breeder may be faced with the
problem that plants that show the desired chemical change suffer from other metabolic
alterations and are low yielding.

The cruciferous and the leguminous families often contain noxious substances in the
form of glucosinolates, alkaloids and glucosides. Lupinus species produce several
different alkaloids and the utilization of these protein-rich plants was nearly
impossible until successful screening experiments for spontaneous mutants free of
alkaloids were performed (von Sengbusch, 1938 as cited by Boersma (2007).
MeliIotus albus, the white sweet clover contains a glucoside of the o-oxycinnamic
acid class, which converts into coumarin and the toxic dicoumarol. Large-scale
screening experiments for non-bitter plants failed for a long time, but this was finally
achieved in selecting non-bitter mutants after treatment with mutagenic chemicals or
ionizing radiation. A prerequisite for this successful mutation experiment was the
development of a rather simple method for mass screening, which made the
investigation of 1000 plants per day feasible on the basis of a half leaflet per plant.
While all non-bitter mutants exhibited a lower vitality, it was possible to increase their
vitality even above the original strain by heterosis breeding.

Rapeseed (Brassica napus), mustard (B. campestris), Indian mustard (B. juncea) and
rocket (Eruca sativa) are widely cultivated oil crops, but their seed meal and green
matter are also valuable protein sources for which they have, so far, only been partially
exploited. One of the main reasons for this is their content of glucosinolates, which
yield degradation products with toxic effects on animals and through the milk, also
on children. The first attempts to lower the glucosinolates content by conventional
breeding have produced interesting results (Röbbelen, 1973, cited by Jambhulkar,
2015) and the process has been pursued with mutation supported by more precise and
sophisticated screening methods (Bjerg, Sørensen and Wathelet, 1987).

Pea (Lathyrus sativus) is used in some parts of India for human consumption but
contains a neuro-toxic component that seriously injures children, causing the disease
known as: Lathyrism. The toxic factors have been studied and success in isolating
mutants practically devoid of the neurotoxic principle ß-N-oxalyamino alanine have
been reported (van Harten, 1998). The use of cassava (Manihot esculentus) as an
important item in human and livestock nutrition in the tropics is hampered by chronic
toxicity caused by the build-up of cyanogenic glucosides after harvest. Screening
techniques to check large numbers of samples are available (Tivana et al., 2014).
However, vegetative propagation practised throughout centuries has notably limited
the genetic variability of cassava and mutation induction is forcibly being applied as
it appears as a worthwhile approach when combing in vitro techniques and more
precise screening methods.

192
 7.4.6. Examples of mutants with improved quality
TABLE 7.4. EXAMPLES OF MUTANTS WITH IMPROVED QUALITY

Crop Trait Method Landmark Reference

mutant/cultivar
(country)
Rice Grain quality Gamma ray VND95-20 Do et al., 2009
Oryza sativa (Viet Nam)
Rice Grain quality Gamma ray Shwewartun MBNL Nos. 11 and
Oryza sativa (Myanmar) 12, 1978
Ahloowalia,
Maluszynski and
Nichterlein, 2004
Rice Glutinous Gamma ray RD6 Ahloowalia,
Oryza sativa endosperm (Thailand) Maluszynski and
Nichterlein, 2004
Bread wheat Grain colour Jauhar 78 MBNL No. 2, 1973
Triticum aestivum (Pakistan)
Cassava Cooking Gamma ray Tekbankye MBNL No. 44, 1999
Manihot esculentus quality (Ghana)

Sorghum Grain colour Gamma ray Djeman MBNL No. 44, 1999
Sorghum sp. (Mali)
Sunflower High oleic acid Gamma ray NuSun Ahloowalia,
Helianthus annuus (USA) Maluszynski and
Nichterlein, 2004
Tobacco Pale green X-ray Chlorina F1 Sigurbjornsson and
Nicotiana tabacum (Indonesia) Micke, 1974

Grapefruit Red fruit flesh Retrotransposon Rio Star MBNL No. 37, 1991
Citrus paradisi and juice (USA)
Macf.
Chrysanthemum Reduced Ion beam Aladdin 2 Shirao et al., 2013
Chrysanthemum sp. axillary buds (Japan)
Osteospermum Flower colour Ion beam Vient flamingo ; Sekiguchi, Hase and
Osteospermum Vient labios Tanaka, 2009
ecklonis (Japan)
Various flowers, i.e. New colours Gamma ray Golden Creman; MBNL Nos. 33 and
Canna lilies, Moss- and forms Cream Prapanpong; 34, 1989,
rose Orange Siranut; Wongpiyasatid,
Pink peeranuch; Hormchan et al., 2000
Yellow arunee
(Thailand)

193
7.5.AGRONOMIC TRAITS

7.5.1. Flowering and ripening time

Early and late maturing mutants are frequently induced through mutagenesis and are
easily identified. Early maturity in cereal and legume crops is one of the most useful
characteristics for cultivation in cool temperate regions, offering the opportunity to
flower in frost-free conditions, harvest prior to frost, and, in drought-prone regions
the ability to produce a viable crop prior to drought conditions. The flowering time
and maturity of cereals and other crops are controlled by the plants’ ability to sense
season temperature and day-length signals controlled by vernalisation and
photoperiodic sensitivity genes. Sweden was home to pioneering work on mutation
breeding in barley where the barley variety Mari, a mutant variety exhibiting early
maturity and semi-dwarfism was developed. Mari is a direct mutant resulting from
irradiation of Bonus Barley in 1960 (Lundqvist, 2014) from which many valuable
indirect-mutant varieties were bred. Early flowering and early maturity mutants have
been induced in a number of other crops; notable examples include banana, cotton,
pearl millet, rice, and soybean.

Quite often several additional characters are changed in early maturing mutants. The
yielding capacity of early maturing mutants is generally reduced. However, early
maturing mutants, especially slightly early maturing ones, with yielding capacities
equivalent to or higher than their original varieties have been induced in several crops.
It has been reported that correlations between total yield (grain + straw) and ripening
time are positive, but correlation between grain yield and total yield is negative in
early ripening barley mutants. Thus, in the evaluation of the practical value of early
maturing mutations, data on yield per day during their growth period should be
collected.

Plant height can also change in early maturing mutants and significant positive
correlations between the two characters have been reported. Reductions of internode
number, short basal internode or increase of upper internode length were observed in
early maturing barley mutants. Furthermore, other characters, such as panicle length,
1000 grain weight, panicle number, straw-stiffness and protein content, were changed
in early maturing mutants according to Gottschalk and Wolff, 1983 as cited by
(Datta, 2014).

Early heading mutants can be identified by simple observation or measurement, and

their screening and isolation can be conducted rather effectively because of relatively
high heritability (in the broad sense) of heading date. It is acknowledged that, the
degree of dominance of induced mutations influences efficiency of screening and
isolation of the mutations. Early heading macro-mutations are, in most cases,
recessive to the original genes. However, dominant or partially dominant early

194
heading mutations have also been isolated in barley, oat and other crops
(Dumlupinar et al., 2015). Differential responses of early heading mutants of spring
barley to light (quality and quantity) and temperature have also been reported. Studies
with Arabidopsis gave evidence for the occurrence of mutations that had lost the light
requirement for germination ability (Franklin and Quail, 2010).

7.5.2. Adaptability

Wide geographical adaptability has become an important character for crop varieties
in recent years, mostly in relation to the foreseen climate change. An FAO study
(Burke, Lobell and Guarino, 2009) on the shift in African climates by 2050 and the
implications for crop improvement, extensively examined the variation both in the
possible shift of isohyets and the difficulty to identify “analogue crops” to be used for
better adaptability in different regions.

Photoperiod insensitivity is a prerequisite for wide adaptability when cultivars are to

be grown in different localities at different latitudes, and mutants insensitive to
photoperiodic differences have been reported. Wide adaptability is achieved,
however, not only by photoperiodic response, but requires the variety to respond or
remain neutral to a number of different other environmental conditions. It is therefore
related to many physiological characters including those involve in the reproductive
biology. Durum wheat mutant varieties showed wider adaptability than their mother
variety, especially in regions where both soil fertility and water resources were not
limiting factors (Donini and Sonnino, 1998) . Mutants in practical breeding projects
should be tested at different locations under different environmental conditions. One
of the most promising potential uses of induced mutations is to break through barriers
at the outer limits of the adaptability of a species. A number of practical examples of
this already exist, e.g. Mari a barley mutant variety (Sigurbjörnsson, 1975; Xia et al.,
2017). Thus, barley can now be grown under short season in Scandinavia and in
equatorial regions with short day lengths. It also spans from sub-sea level land
locations to the high Andes.

7.5.3. Plant type, growth habit and architecture

Mutation breeding is frequently used to alter plant architecture. Plant architecture

includes phenotypic traits such as plant height (i.e. culm length of cereals), plant-type,
branching habit (e.g. number of tillers), size, number, shape and orientation of leaves,
stolon characters, size and number of flowers. Prime examples are mutants conferring
reduced plant height, which generally have positive effects on increasing yield by
reducing lodging and increasing tillering. One of the most successful utilization of
plant height mutations is the deployment of semi-dwarfism in annual cereals crops,
e.g. rht (reduced height) genes in wheat, sdw in barley and sd1 in rice. The sd1 mutant
gene of rice produces short and stiff stems, which also confers increased grain yields

195
through a reduction in lodging and greater mobilization of resources to grain. Semi-
dwarf rice and wheat varieties were leading examples of breeding progress as defined
by the “Green revolution”, for which Norman Borlaug won the 1970 Nobel Peace
Prize. The first semi-dwarf mutant variety of rice var. Reimei was induced through
gamma irradiation Futsuhara, 1968 as cited by Kikuchi and Ikehashi, (1984), the
height was reduced by at least 15cm when compared to the original mother variety:
Fujiminori. The variety Reimei carries the same sd1 (semi-dwarf) allele as do the
spontaneous mutant var. Dee-Geo-Woo-Gen and the induced mutant variety
Calrose 76. All these varieties are widely utilized across Asia and America (Lestari,
2016). Since the identification of the sd1 gene, in Japan, 80 of the 229 registered rice
varieties represent descendants of Reimei. In the USA, greater use has been made of
the semi-dwarf trait from cv. Dee-Geo-Woo-Gen derivatives; however, additional
induced or identified spontaneous mutations have also been successful (Rutger and
Mackill, 2001). Similar achievements have also been made in wheat, oat, barley, but
also in some fruit trees such as apple, peach, cherries.

Barley mutants have been used to study the basic architectural building blocks of
plants, in about 100 phytomer mutants (Forster et al., 2007). They can be used to study
and predict the type of organ to be formed by studying the ontogeny of meristems at
various positions (apical or branch) and at various stages in the plant life cycle. The
most basic phytomer unit consists of a stem segment with a leaf and/or root attachment
which can be replicated in an apical or side direction. Studies of phytomer mutants
can also be used in taxonomical studies in defining what type of structures is
developed from a meristem in a given species. For example, a classic taxonomic
difference between wheat (Triticum) and barley (Hordeum) genera is that spikelet
production in wheat is determinate whereas it is indeterminate in barley, and
conversely floret production in wheat is indeterminate, but determinate in barley. This
leads to wheat having multiple florets per spikelet whereas barley is restricted to one.
Wheat therefore produces far more seeds per spike (ear) than barley. However, one
phytomer mutant in barley, known as the “wheat mutant” exhibits indeterminate floret
production and therefore of interest in increasing yield (Forster et al., 2007); it is also
of taxonomic interest as it indicates that this particular taxonomic descriptor is
governed by a single gene.

7.5.4. Resistance to lodging and stem breakage

Lodging susceptibility and stem weakness are serious problems in many crops
including: wheat, oats barley, rice, corn, sugar cane, sorghum, flax, cotton, soya, broad
beans, etc. Lodging has been analysed by several investigators especially in cereals.
As a general rule, lodging is caused by either (1) uprooting of the plants; (2) breakage
of the stems near the ground level; and/or (3) leaning and bending of the stems: the
last cause being the most common one and often caused by harsh weather such as high

196
wind and heavy rain. Lodging is a complex event bound to the concomitance of
several external and internal factors. Apart from external factors, such as
environmental conditions, agronomic procedures, diseases, the most important plant
characters that can influence lodging are: strength and elasticity of the stem, structure
and development of the root system, and stem length. Although lodging resistance
and stem-length reduction are not always associated, the shortening of the stem is a
characteristic that is worth selecting for. Additionally, short cultivars are usually
better adapted to mechanical (e.g. combine) harvesting.

The general selection procedure for lodging-resistant mutants can be outlined as

described in this paragraph. Owing to the close association between stem length and
lodging resistance, short-stem plants can be isolated visually in the M2 or M3
generation. Once the mutations for stem shortening are confirmed and, if they have a
potential interest, in the following generation they can be multiplied in a single plot
or in small plots in order to derive useful information on agronomic value. Field trials
can then take place, generally starting from the M5 generation, possibly in different
locations and, for cereals, under high nitrogen dressing levels, which can be expected
to exercise a strong selection pressure. In this way both the lodging resistance and the
yielding ability and other important characters (earliness, quality, etc.) associated with
outstanding varieties can be tested in the M5 and subsequent generations.

Numerous lodging-resistant mutant lines have been developed from plant material
treated with chemical or physical mutagens (http://mvd.iaea.org).

From a general review of the results so far available the following remarks can be
made.

1. Mutations for lodging resistance are relatively easily induced in cereal crops,
and often induced in many other crops.

2. The increase in standing ability is mostly due to reduced height of the stem;
other changes can occur, parallel with stem shortening, namely changes in
the number of internodes, modifications in the relative length of internodes;
modifications of the root system can also take place. Further work will be
needed to evaluate the relative importance of these factors in determining the
behaviour of the selected lines.

3. The morphological and anatomical changes mentioned above can occur

without evident changes in important agronomic characteristics such as
inflorescence characters, yield potential, seed quality, date of maturity,
disease resistance etc.

197
 4. Parallel to the increase in standing ability the capacity of the selected mutant
line to endure higher amounts of nitrogen fertilizers with positive influence
on yielding ability is generally ascertained.

5. Numerous progressive stiff-straw mutants have so far been isolated in several

cereal species and have performed better than their mother variety in terms
of yield. This is the case, for instance, of mutants obtained in oat, barley, rice,
and in bread and durum wheats (Kato, 2008). New varieties have been bred
from some of these foundation mutants.

6. On the other hand, the utilization of lodging-resistant mutants in cross

breeding is also interesting because of the possibility of removing
undesirable pleiotropic effects (e.g. excessive spike density) possibly carried
by the original mutant.

7. The difficulty inherent in the genetic pattern of lodging and stem breakage,
which is also very complex in a diploid species such as barley, has prevented
any real advance in the knowledge of the genetic nature of the mutation for
lodging resistance.

8. The appearance as early as M2 of such complex phenotypes, which breed true

in the following generations, may indicate that simple heritable changes may
be responsible for the control of important traits.

7.5.5. Shattering and shedding resistance

A great amount of potential harvested products (fruits or seeds) are lost every year
due to wind and rain prior to harvest. Specific anatomical and histological features of
crops are responsible for these losses, which occur in one of the following ways:

x by shattering of the fruits (in legumes, crucifers, but also in sesame,

buckwheat and other species),
x by shedding of seeds, fruits or whole inflorescences (in cereals, legumes,
grasses and others).

Mutation breeding has not been exploited much in developing shattering and shedding
resistant varieties, but several resistant types have been selected from natural
populations during the past decades. Shattering-resistant spontaneous mutants of lupin
(Lupinus lure us) have been known since 1935; they were used as valuable partners
in lupin cross breeding. Analogous findings were obtained in Lupinus lutescens,
L. polyphyllus, L. mutabilis, and L. angustifolius, and also in beans (Vicia faba). In
lupin, the resistance to shattering is associated with the formation of a very thin layer

198
of fibres in the wall of the pod. Recessiveness is assumed, but the character in question
is very complex and it is not clear whether the resistance is really due to one single
gene (Maluszynski and Kasha, 2002).

Shattering resistance is an important breeding aim in crucifers’ oil plants, large

differences could be found with regard to this character when comparing different
species and varieties of this group of cultivated plants. Shattering resistant rapeseed
varieties possess a large number of specific thin-walled and lignified cells in their fruit
walls. Corresponding results were obtained in mustard (Brassica juncea); after X-rays
treatments a shattering-resistant mutant was produced with thick fruit walls.
Shattering resistant plants were also selected in Sesamum indicum (Ji et al., 2006;
Boureima et al., 2012).

Two genes have been found responsible for seed shedding in sorghum
(Sorghum virgarum) and resistance is triggered by recessivity of both these genes. It
would be of considerable importance if wild grasses and wild legumes such as lupin
(Lupinus perennis) and many bean species (Vicia spp) could acquire such characters,
which are considered standard for cultivated plants. The first results in this direction
were obtained in forage grasses (Phalaris) and blackgrass (Alopecurus) after X-ray
treatments and in soybean after neutron irradiation (Khan and Tyagi, 2013).

The semi-dwarf trait is generally related to modern production systems. This has two
major advantages: 1) it changes the source-sink relationship so that more energy is
directed towards seed production and less to vegetative growth, and 2) semi-dwarf
crops are amenable to mechanical (combine) harvesting. Some other characteristics
of plant architecture, such as increased branch number in cereal and legume crops,
more compact growth habit such as high plant density, shorter stolon length of potato
etc., can also affect the efficiency of growth and harvest, particularly in modern,
mechanized agriculture.

In the flower industry traits such as: reduced number of flowers, change in shape and
flower colour etc., are usually welcome because they readily add value to the plant.
One unique mutation in chrysanthemum with a lower number of axillary flower buds
was induced by ion beams in Japan (Shirao et al., 2013). This mutant is very useful
for the flower industry because in non-mutant chrysanthemum large flowers can only
be produced by removing (hand picking) axillary flower buds from the plant. This
mutation negates the need for axillary bud removal. Notable successful
chrysanthemum varieties carrying this mutation include var. Imajin (Imagine) and var.
Alajin (Aladdin).

Another important plant character induced by mutagenesis is the non-shattering rice.

This mutation, induced by gamma-ray irradiation utilizing a shattering indica variety,
turned out to be to be useful for release of a forage rice variety: “Minami-yutaka” in

199
Japan (Hiroshi, 2008). The mutation, which may also be induced by gamma-ray
irradiations in other gramineous species, is useful for improving other forage crops
which currently have high levels of seed shattering in their inflorescences. To make
agriculture greener, traits such as super-nodulation, enhanced water use efficiency,
and healthier nutrients uptake from soils will become more and more important, in
relation to the movement towards healthier and organic-based living style. The
success in breeding and release of a super-nodulating soybean variety is a significant
advance in this field (Takahashi et al., 2005).

For species relatively new to agriculture, e.g. blueberries, jatropha and those that, up
to now, have received little attention of plant breeders, e.g. medicinal plants and
culinary herbs and spices, there is an urgent need to identify, develop and establish
domestication traits. For species that have been in domestication for thousands of
years many agronomic traits have been provided through spontaneous mutants and
selection and have been incorporated into the crop as they arise. The small grain
cereals, such as rice, wheat, barley, etc., provide good examples. The grassy wild
progenitors of these crops possess natural seed dispersal mechanisms whereby the
seed head shatters into pieces at maturity and individual units carrying seeds fall to
the ground and are dispersed by hooking onto passing animals. Such dispersal
mechanisms are inappropriate to agriculture and cereal crops were not established
until mutants for non-shattering were found (Ji et al., 2006). Interestingly the barbed
awn trait useful for natural dispersal has persisted, however, barbs are now thought to
be associated with the dust produced during mechanical harvesting that causes
‘Farmers lung’ disease, smooth awn mutants are therefore of interest. These
respiratory diseases associated with occupational inhalation of dust during harvest and
processing have been, for example, described in hop (Humulus lupulus) cultivation,
and may then be addressed by developing non-shattering and/or awn-free mutants
varieties (Reeb-Whitaker and Bonauto, 2014).

7.5.6. Other agronomic traits

Mutants are usually named after the most distinguishably changed character, and
several hundreds of differently named mutants with common changes in growth habit
and plant type appear in articles on induced mutations (see Chapter 3). The plant type
must be considered an integration of variations in individual characters and must be
described by discriminate selected criteria taking into account existing variations of
other individual characters and their contribution to the characteristic feature.
Different plant types result from different patterns of growth and differentiation of
plants. Dwarf mutants, for example, are very frequently observed in various plant
species. They are characterized by shortened height and at the same time in many
instances reduced number plant organs, indicating a growth-rate reduction in many or
all plant parts throughout their entire life. However, there are examples in which plant

200
parts or organs are disproportionately reduced in size, e.g. semi-dwarfs in barley: the
GPert mutation in barley bestows gibberellin insensitivity that affects most tissues
and cells to produce semi-dwarfism whereas the sdw mutation in barley mainly
reduces the length of the first internode. Mutants in which the vertical axis is
disproportionately shortened and called brachytic types have been reported. In such a
'stout' mutant of sorghum the plant height is reduced to three–fourths while the stem
diameter is twice as large as the original form. Similar changes of relative size and
length/width ratio in various organs have been observed frequently in other crops.
Giant mutants have also been reported in pea, tomato and barley. In these cases,
several or all plant parts were enlarged.

A distinct change of growth habit observed in mutants is ageotropism or laziness.

Ageotropic or 'lazy' mutants have been found in maize, rice, barley, lupin, and peas in
the late 1930s and described in a review by (Howard III et al., 2014). These mutants
have mostly been used for genomic studies including transposon-tagging in maize. In
pear (Pyrus communis L.), Nashima et al., (2013) worked on the identification of
genes expressed differentially between the mutant: giant La France (GLaF) and the
parental variety: ‘La France’ based on the increased size of fruit. One conclusion
drawn from this observation was that, considering the bud mutation is a quite localized
event leaving the rest of the genome untouched, this result might open the way to
more investigation to identify key genes governing other important traits.

Dense organ formation in mutants is observed in various parts of the plant. Abundant
tillering or branching, often accompanied with dwarfism, dense node or multi-nodal,
fascinated, double leaf, dense compact or erect ear, increased row-number of ear (in
barley) and multi-florous or multi-ovary types are mentioned as examples. Such plant-
type mutants are classified or named by the most characteristic changes, but they may
also show changes in other characters. Table 7.5. below present some valuable mutant
varieties developed for their agronomic traits.

201
 7.5.7. Examples of mutants with improved agronomic traits
TABLE 7.5. EXAMPLES OF MUTANTS WITH IMPROVED AGRONOMIC TRAITS

Crop Trait Method Landmark Reference

mutant/variety
(country)
Rice Semi-dwarf Gamma ray Reimei (Japan) Das, Patra and Baek,
Oryza sativa 2017
Rice Short stature Gamma ray Calrose 76 Rutger, Peterson and
Oryza sativa (USA) Hu, 1977
Rice Short stature, early Gamma ray TNDB 100 MBNL No. 45, 2001
Oryza sativa maturity (Viet Nam)
Basmati rice Short stature Gamma ray CRM 2007-1 (India) PMR No. 1 (1 and 2),
Oryza sativa 2006
Durum wheat Short, resistant to X-rays Creso (Italy) MBNL No. 6, 1973
Triticum aestivum lodging
Durum wheat Resistant to lodging Gamma ray Gergana MBNL No. 37, 1991
Triticum aestivum (Bulgaria)
Barley Early flowering Gamma Mari (Sweden) Sigurbjönsson, 1975
Hordeum vulgare (Eam8)
Rice Semi-dwarf (Sd) Gamma Calrose-76 Lestari, 2016
Oryza sativa (USA)
Barley Semi-dwarf (Sdw) X-ray Diamant Ahloowalia,
Hordeum vulgare (Czech Republic) Maluszynski and
Nichterlein, 2004
Basmati rice Non-lodging Gamma CRM 2007-1 PMR Nos. 1 and 2, 2006
Oryza sativa (India)
Groundnut Early maturity, semi- Gamma ray TAG24 (India) Patil et al., 1995
Arachis hypogea dwarf
Chrysanthemum Reduced axillary Ion beam Allajin (Aladdin) Shirao et al., 2013
Chrysanthemum sp. buds (Japan)
Rice Non-shattering Gamma ray Mini-yutaku Hiroshi, 2008
Oryza sativa Japan
Rye Short life cycle Gamma ray Soron (Peru) Gómez-Pando et al.,
Secale cereale 2009
Banana Short stature Gamma Novaria (Malaysia) MBNL No. 44, 1999,
(Musa acuminata) Mak et al., 1996
Rice Herbicide resistance Gamma ray Rice (USA) Maluszynski,
Oryza sativa Ahloowalia and
Sigurbjörnsson, 1995
Corn Herbicide resistance Gamma ray Corn (USA) Mabbett, 1992
Zea mays
Rice Herbicide resistance Gamma ray IRAT 239 MBNL No. 33, 1981
Oryza sativa (Guyana)
Cherry Compact growth Gamma ray Aldamla and Burak Kunter et al., 2012
Prunus avium

202
7.6.MUTANTS TO FACILITATE PLANT BREEDING

Plant mutation breeding is already a form of accelerated breeding as it can induce

required changes directly into elite germplasm with no or little penalty. However,
there are mutant traits that could be exploited further in plant breeding.

x Increased recombination inter- and intra-specific, (see Chapter 3).

x Limited recombination – introduction of “alien” recombinants that do not
recombine (see Chapter 3).
x Haploid inducer genes in various species, new work in Arabidopsis
(see Chapter 8).
x Outcrossing, cytoplasmic male sterility, promotion of open pollination.
x Long lived pollen.
x Off-season flowering (e. g. cassava), not restricted to season.
x Elimination/overcoming self-incompatibility.
x Mutants for cultivability/tissue culture response, e.g. Golden Promise barley
a good genotype for transformation in barley.
x Markers for traits (SNPs).

203
204
 8. SPECIFIC TECHNIQUES FOR INCREASING EFFICIENCY OF
MUTATION BREEDING

This chapter will consider the various technologies that may be used to enhance the
out-come of mutation induction either by increasing the population size at any given
level from M0 to Mn, such as: A) in vitro techniques, or B) by more rapidly stabilising
the mutations in a fully homozygous state, such as the haploid/doubled-haploid
techniques and finally C) the application of molecular markers techniques for the
identification and screening for mutations in the DNA sequence, which may be related
to crop improvement.

A) IN VITRO METHODS IN PLANT MUTATION BREEDING CHANGE

8.1.BRIEF REVIEW OF PLANT TISSUE CULTURE

Plant tissue culture is generally defined as the process whereby small pieces of plant
materials (explants) are isolated and grown aseptically in vitro. Plant cells have both
plasticity and totipotency (Haberlandt, 1902), and are therefore suited to in vitro
culture. The culture of plant parts can be done at various levels: cells, tissues, organs
and whole plants. Each plant part (buds, leaves, flowers, fruits, roots etc.) is composed
of cells from the same origin and thus any part of the plant may be sourced for culture
purposes

In the early 1900s, the advent of in vitro techniques opened new avenues for
examining and manipulating cells, tissues and organs grown in culture. It was
observed that, when exposed to changes in culture conditions (culture media,
temperature, light, etc.) different responses were achieved; of particular interest was
regeneration, i.e. the production of new plants. One major break-through was the
production embryos in culture from de-differentiated callus. These “somatic
embryos” (as opposed to zygotic embryos) could be produced in large number and
grown through maturity.

In vitro methods have been an important part of plant biology and crop breeding since
the early 1960s. Plant tissue culture comprises a set of methods and strategies that
involve: the transfer of in situ plant parts to sterile in vitro culture, manipulations
under a laminar air flow (sterile air) and the culture on well-defined media (solid or
liquid) which comprises carbohydrates, major and trace mineral elements, vitamins
and growth regulators, (Murashige and Skoog, 1962; Grafi, 2004; Thorpe, 2006).
Specialised methods have been developed the culture of protoplasts, anthers,

205
microspores, ovules and embryos. In addition to tissue culture being used for basic
studies in cell biology and plant development it has been major tool in accelerating
plant breeding. The ability to manipulate cells, tissues and organs in culture has
allowed: the synchronous production of large plant populations, i.e. clonal production,
the production of genetically pure lines (haploids and doubled haploids), the
production of pathogen-free propagules, the genetic transformation and other new
methods in mutation induction.

8.2.PLANT REGENERATION SYSTEMS

8.2.1. Micro-propagation

In vitro propagation mainly consists of the proliferation of apical or axillary buds into
newly developed shoots, which in turn provide buds for subsequent rounds of culture
and multiplication. The technique is often called micro-propagation and is a cloning
strategy (Figure 8.1). Micro-propagation is used routinely in commercial production
of true-to-type plants, which retain the exact same genetic background as the donor
plant. Initially, the process was used for studies of plant development (anatomy,
histology and cytology, for example), but it rapidly became an industrial tool for mass
production of high price valued plants, mostly ornamentals, flowers and fruits trees
as well as vegetatively propagated crops, e.g. potato. For ornamental plants the
association of micro-propagation with mutation induction also proved to be of high
economic interest (Ahloowalia, 1998; Jain and Spencer, 2006).

Figure 8.1 Different types of explants for plant micro-propagation.

206
 8.2.2. Meristem culture

In the 1950s it was observed, that when a very small section of the apical dome in the
meristematic zone – a few cells – is cultured, a good rate of clonal multiplication
ensues and in addition, the propagules produced were virus-free; (Barba et al., 1994)
This finding found major application in commercial production of “virus-free plants”
in many ornamental, food crops and even trees of commercial value. The apical
meristematic dome, which is composed of non-differentiated cells, is not linked to the
vascular systems and thus contains a few or no virus. Additionally these cells are
genetically stable and responsive to culture, and thus, this approach is among the
in vitro techniques that are also very useful for a rapid mass production of mutated
plants (Ahloowalia and Maluszynski, 2001).

8.2.3. In vitro morphogenesis

Although micro-propagation is known to have the potential to produce hundreds, and

even thousands, of plants, it is still limited by the number of pre-existing buds from
the initial plant material. Morphogenesis is an efficient way of increasing the chances
to get new buds and thus new plantlets. There are two types of morphogenetic
processes. In “direct morphogenesis” the newly formed buds emerge directly from the
tissue in culture: i.e. leaf, stem and root tissues – cambium, epidermis – or any other
plant tissue (Figure 8.2a). Conversely, in “indirect morphogenesis” the buds usually
arise from a transitional structure of de-differentiated cells into a callus (Figure 8.2b).
The process might also result in a more sophisticated developmental process – somatic
embryogenesis – leading to a newly formed embryos (Thorpe, 2012).

Progress continues to be made in our understanding of the genetic basis underlying

tissue culture response (Phillips, 2004). Advances in in vitro systems are readily taken
up and exploited in plant improvement schemes including mutation induction,
mutation selection and mutant line development.

207
 Leaf fragments are plated Cytokinin (hormone) is added and after a few Developed shoot plated on a
onto a culture medium weeks: buds/shoots are formed rooting medium

Figure 8.2a. Direct organogenesis and/or embryogenesis: Direct shoots/buds formation in cassava
(Duclercq et al., (2011).

Figure 8.2b. Indirect organogenesis and/or embryogenesis: Callus formation on cassava leaf
fragments; and immature embryo culture followed by bud and/or secondary embryo development. As
modified from (Duclercq et al., 2011).

8.2.4. Haploid and doubled haploid production

In some mutation induction experiments, the traits to be exploited are either difficult
to identify or not noticeable in heterozygous genotypes, this is because the vast
majority of mutations are recessive. The advent of haploidy and doubled haploidy has
opened up a new area for mutation breeding as it provides a means for producing
mutants in haploids and then fixing these in a homozygous state of doubled-haploid

208
mutant plants. The potential outcomes from haploid culture in mutation breeding are
wide ranging from stabilization of homozygous recessive traits, immediate fixation of
mutated genes, increase selection efficiency, the ability to screen for desired mutants
in vitro and huge savings in experimental times. For further information see
Chapter 8-B.

8.3.PLANT TISSUE CULTURE FOR MUTATION BREEDING

When applied in combination with mutation induction, plant tissue culture increases
the overall efficiency of the mutagenic treatments. It enables the creation of new
genetic variation, it facilitates the handling of large plant populations, it allows early
in vitro selection, and it provides a means for easily cloning selected variants. Plant
tissue culture also provides the ability to handle these large mutagenized populations
in a laboratory scale, thus allowing the development and implementation of efficient
and reliable methodologies to screen for biotic and abiotic stress responses (Pathirana,
2011).

8.3.1. Micro-propagation of mutated plants

Here selected putative or confirmed mutant plants are entered into a micro-
propagation or regeneration programme. In the case of seed propagated crops, the
process of in vitro culture can be considered, but only if it provides some additional
advantages over normal field handling methods such as: developing pure genetic lines
(doubled haploidy) or facilitating and speeding up the selection through the
establishment of easy and reliable screening processes, or clonal production. In the
case of vegetatively propagated crops, in vitro techniques offer an exceptional
advantage over greenhouse and/or field experiments as large mutant populations can
be handled in a relatively small area, i.e. in the laboratory.

8.3.2. In vitro mutation induction

Since the 1970s, in vitro mutagenesis has gained increased popularity as it has
overcome major limitations of conventional mutagenesis, especially in the production
and handling of large mutant populations and the establishment of reliable and easy
to apply screening methods (Maluszynki et al. 1995; Suprasanna et al., 2012). In
addition, direct adventitious buds formation and somatic embryogenesis allow for
rapid dissolution of chimeras and facilitate the development of homohistant mutants
(Geier, 2012; Jankowicz-Cieslak and Till, 2017).

In vitro mutagenesis is a powerful tool for rapid multiplication of any new and
favourable trait obtained; however, there are also some constraints that should be
taken into consideration, in order to maximize the benefits.

209
Advantages of in vitro mutagenesis

x High mutation frequency.

x Uniform and repeatable mutation treatments.
x Use of single cell systems, and application of selective agents to provide
homogeneous and synchronised cultures.
x Less space to handle large population within short time frames and the
production of disease-free plants.
x Multiplication via tissue culture is fast and independent of seasons.
x In vitro selection and rapid cloning of selected mutants.
x Tissue culture takes place in special conditioned areas making the whole
process easier to control and monitor.
x Tissue culture provides high phytosanitary conditions, which is an excellent
system for obtaining healthy starting material and is maintained throughout
the plant regeneration process and clean plants are produced to satisfy
quarantine regulations.
x Plant tissue culture allows a wide choice of plant materials for mutagenic
treatment.

Some limitations and problems relating to in vitro mutation induction

x Specialized laboratory, equipment and trained staff is needed.

x Establishing cell cultures showing good regeneration may be difficult from
a technical point of view.
x Some methods are genotype dependent and limit widespread use.
x Cultured cells and whole plants often express different sets of genes
according to cultural medium and environments.
x Selection at the cellular level is not possible for many traits of agronomic
importance.
x Not all observed variation is genetic in nature, which complicates the
selection process.
x Effective selection of desired mutants is often hampered by inadequate
knowledge about biochemical pathways and developmental processes.
x Ultimately plants produced and selected need to be tested in the field.

210
 8.3.3. Types of explants for in vitro mutagenesis

In vitro mutagenesis can be applied immediately before, during and/or after the
in vitro culture. The in vitro mutagenesis processes involve important prerequisite
steps including the selection of the proper target material, the choice of explants and
the deployment of appropriate culture medium. In addition, there is also a need to
consider the genetic constitution and ploidy of the selected material which could
seriously affect the success of the programme – some genotypes are recalcitrant to
in vitro methods.

The choice of the mutagenic treatment and the determination of the optimal dose are
similar to methods for seed propagated plants (see Chapter 5), with the important
difference regarding the initial dose, which is usually significantly reduced due to the
high-water content of in vitro plant materials.

8.3.4. Mutagens used in vitro

As stated in Chapters 1 and 2, physical as well as chemical mutagens have been

applied successfully on in vitro derived plant materials. However, among the physical
agents, gamma- and X-rays and also UV radiation are the most popular choices. EMS
and sodium azide have been largely used as chemical mutagens in tissue culture
(Suprasanna and Nakagawa, 2012; Oladosu et al., 2016). Predieri and Di Virgilio
(2007) for example, stated that X- and gamma- rays are the most convenient and
easiest to use radiation types for inducing mutation in tissue culture, mostly in regards
to safety, environmental issues and post treatment handling as compared to chemical
mutagens. More than 90 percent of released in vitro mutant varieties are derived from
physical irradiation (http://mvgs.iaea.org/Search.aspx).

8.3.5. In vitro radio-sensitivity testing

One of the first steps in mutagenic treatment is the estimation of the most appropriate
mutagenic treatment. This involves the determination of radio-sensitivity and the dose
that produces a 50 percent reduction of vegetative growth (RD50). The radio-
sensitivity is usually estimated through the physiological response of the irradiated
material (Figures 8.3a,b,c)). This estimation may be carried out in the same way as
for in vivo mutagenesis. Usually at least 20-30 cultures are tested for each dose over
a range that covers 50 percent lethality or 50 percent growth reduction, LD50 and
RD50, respectively. However, it is sometimes necessary for in vitro mutagenesis of
certain crops to lower this value to take into account the fragility of some tissue culture
derived plant material, and in this case a LD30 may be selected (Patade and
Suprasanna, 2008). The breeder has the final say in the dose choice as he/she knows
his/her material best and how to handle the subsequent mutant populations. Radio-
sensitivity may vary with the species, the cultivars and genotype, with the

211
physiological conditions of the plants, organs, as well as with the types of mutagenic
treatment to be used: physical or chemical, and finally with the conditions surrounding
the experiment (Table 8.1).
TABLE 8.1. SOME EXAMPLES OF DOSE RANGE FOR RADIOSENSITIVITY TESTING OF
IMPORTANT CROPS FOR IN VITRO MUTAGENESIS (Shu et al., 2012)

Crop species Treated material Mutagen and dose (LD50 Gy)

Chrysanthemum Rooted cuttings Ȗ-rays, 25
Banana Shoot tips Carbon ion beam, 0.5 to 128
Banana Shoot tips Ȗ-rays, 60
Banana var. Lakatan Latundan Shoot tips Ȗ-rays, 25 – 40
Banana sp. Embryogenic cell Ȗ rays, 10 – 40
suspension
Pineapple var. Queen Crowns Ȗ rays, 0 – 45
Begonia In vitro cultured leaflets Ȗ-rays, 100
Weigela In vitro shoots Ȗ-rays, 40
Potato Callus cultures Ȗ-rays, 30 – 50
Potato Micro tubers Ȗ-rays, 10 – 30
Sugarcane Buds / callus cultures Ȗ-rays, 20 – 25
Cassava Somatic embryos Ȗ-rays
Cassava In vitro stem cuttings with Ȗ-rays, 25 – 3
two nodes
Peppermint Stolons and rhizomes Ȗ-rays, 30 – 40
Sweet potato Embryogenic suspensions Ȗ-rays, 80
Pear In vitro shoots Ȗ-rays, 3.5
Yam Stem cutting Ȗ-rays, 20 – 50
Yam Microtuber Ȗ-rays, 40
Dendrobium orchids Protocorm like bodies Ȗ-rays, 35

212
 (a)

(b)

0 Gy 5 Gy 10 Gy 15 Gy 20 Gy 25Gy 30
Gy
(c)

Figure 8.3. Radio-sensitivity testing, an example from chrysanthemum in vitro mutation induction: (a)
20 days old in vitro plants irradiated by 60Co gamma irradiator;(b) Irradiated in vitro plantlets
transferred onto fresh MS medium to mitigate against toxic effects of irradiation; (c) Radio-sensitivity
tests: doses 0 (control), 5, 10, 20, 25 and 30 Gy on shoot growth and development applied on 30-days
old plantlets. Courtesy of G. HaspRODW%.XQWHUDQG<.DQWR÷OX.

213
 8.3.6. Chimeras

Following an irradiation or a treatment with chemical mutagens of a plant tissue,

usually only the cells hit by the mutagen agent will exhibit a mutation, i.e. changes in
the DNA structure, consequently only cells derived from these mutated cells will
inherit the DNA modifications. Chimeras are therefore produced, which are populated
by cells of different origins (wild type and a range of mutants).

In seed propagated crops, the transmission of mutations to the next generation requires
that mutant cells enter the germline and are transmitted to eggs and pollen grains. For
vegetatively propagated plants the mutation must be transmitted to the vegetative
propagule, such as buds.

Figure 8.4 illustrates the shoot apical meristems with the three distinct cell layers: the
epidermal (L1) and the sub-epidermal (L2) layers form the outer layers called the
tunica, while the inner L3 layer forms the corpus. The two inner layers, L2 and L3
contribute cells to form the body of the plant with the proportion of cells derived from
each layer varying in different organ types. The chimera situation generates:
i) genetically different cells (mutated or not) within a layer = mericlinal chimera;
ii) genetically different cells involving several layers = sectorial chimera, and
iii) genetically uniform mutated cells in a layer = periclinal chimera (Figure 8.4).

After irradiation the explant (M1V0) carries mutated and non-mutated cells, when the
explant is sub-cultured (M1V1) a chimeric structure appears, the dissection of the
variant and additional sub-cultures engenders: i) a non-mutated bud, or ii) a mutated
bud at M1V1 (Figure 8.4a). On a transverse section of the irradiated bud several sectors
would appear (Figure 8.4b). The challenge for the breeder working with in vitro
mutation induction is to isolate the desired mutants. The process of separating and
selecting only the mutated cell lines is referred to as: chimera disassociation or
dissolution (Figure 8.5). In seed propagated crops the dissociation will occur naturally
during the normal processes of sexual reproduction (see Chapter 5). In vegetatively
propagated crops or in the case of in vitro mutation induction the disassociation of
chimeras normally involves consecutive rounds of sub-culturing: M1V1, M1V2, M1V3,
etc., (Geier, 2012).

8.4.HANDLING MUTATED PLANT POPULATIONS IN VITRO

The number of sub-cultures needed for isolating mutated sectors depends on the
species, the plant regeneration methods used, and the types of mutagenized plant
material.

Following the treatment with mutagenic agents, the newly formed shoot tips, for
example, are transferred individually to fresh medium plates and incubated under

214
normal controlled culture environments. After about 4 weeks, the percentage of
surviving plantlets is recorded and the ED (effective dose) level estimated. Individual
shoots are dissected from M1V1 cultures and transferred onto shoot multiplication
medium to generate the M1V2 generation. About one month after the transfer, a
number of data are collected: mean shoot length, average number of shoots per explant
and percentage of plants showing leaf abnormalities. All morphological changes
(chlorophyll deficiency, morphological abnormalities, etc.) are carefully recorded this
process might be repeated until, maybe the 5th vegetative generations (M1V5) to
identify any induced morphological variation which appears stable and transmissible
to the next generation. The shoots dissected from M1V5 cultures may then be plated
on a rooting medium for regenerating the whole plantlets. The time needed for roots
initiation, development rate and the average number of roots/explant should be
recorded. Well rooted mutated plants are then transferred and hardened in the
greenhouse and maintained at normal temperatures for about a week to enhance the
acclimatization process and carefully monitored until possible flowering and seed
setting stages.

Initial morphometric analysis and evaluation for new phenotypes can be completed in
the greenhouse at the whole plant level. Figure 8.6 illustrates the strategy for in vitro
mutagenesis (from mutagenized organogenic calli or shoot meristems), handling of
the mutated population and mutation recovery in a vegetatively propagated plant. The
isolated putative mutants in the M1V2, and/or M1V3 generation can be evaluated for
stability and multiplied to test their agronomic performance.

In addition to rounds of micro-propagation, Jain et al., (2010) reported that

embryogenic cultures (cell suspension or callus cultures) could be suitable for
inducing mutations and provides mutated somatic seedlings in a short period, thus
avoiding chimeras ,which otherwise require to multiply plants up to M1V4 generation
for chimera dissociation.

Mass propagation of large numbers of shoot tips is used in order to induce direct shoot
organogenesis and prevent callus formation. This system can be used for mutant plant
multiplication in large numbers for further evaluation in which the target
crop/genotype is recalcitrant to somatic embryogenesis. Successive rounds of
isolation and division are performed to reduce genotypic heterogeneity, the number
of individuals normally doubles or get even a larger increase (multi-apexing) at each
generation. Tissue may be collected from M1V6 individuals, DNA extracted and
screened (genotypically) for any induced mutations. The inheritance of isolated
mutations is evaluated and confirmed in the M1V6 and subsequent generations.

215
Figure 8.4. The development of mutant sectors in meristems: (a) Structure of the shoot meristem;
(b) Effects of mutation on the mutated bud, and (c) Sectorial chimeras generated by mutagen treatment.

216
Figure 8.5. Disassociation of chimeras through successive sub-cultures.

217
Figure 8.6. Theoretical schematic for the implementation in vitro mutagenesis programme (as modified
from Duclercq et al., 2011).

218
8.5.IN VITRO MUTATION SCREENING METHODS

8.5.1. In vitro screening techniques for abiotic stress

In an extensive and well documented review by Rai et al., (2011) the description and
assessment of various methods applied for developing stress tolerant plants using
in vitro techniques is provided. The responses to both biotic and the abiotic stresses
have been investigated by applying the adequately selecting stress inducing agents,
e.g. NaCl for salt tolerance (Figure 8.8), and polyethylene glycol (PEG) or mannitol
for drought tolerance. Nikam et al., (2015) reported on the selection of salt tolerant
mutant clones of sugarcane (Saccharum officinarum L.) that were cultured on MS
medium supplemented with 0, 50, 100, 150, 200 and 250mmol/l of sodium salt (NaCl)
and displayed increased brix percentage and improved agro-morphological
characters. Luan et al. (2007) also obtained salt tolerant cultivars of sweet potato
(Ipomoea batatas L.) from EMS induced mutation by repeatedly (5 times every 20
days) screening for tolerant calli on MS medium supplemented with 200mM NaCl,
prior to the regeneration of somatic embryos.

Vanhove et al., (2012) devised a protocol using sorbitol as an agent to increase

osmolality for drought experiments on various banana genotypes. The plantlets were
cultured on liquid medium consisting of: (1) a standard control medium (containing
0.09M sucrose) and (2) a stress medium containing 0.09M sucrose and 0.21M
sorbitol. The media were regularly refreshed every 2 weeks and the growth evaluated
by calculating the difference between the fresh weight of the plantlets at the beginning
and at the end of the experiment, i.e. after 48 days. Masoabi et al., (2017) exposed
16mM EMS-treated sugarcane callus to different polyethylene glycol concentrations
and reported the procedure as a suitable in vitro osmotic selection regime. The in vitro
selected plants were further stressed by reducing the water supply in greenhouse pot
trials to confirm drought tolerance.

8.5.2. In vitro screening for biotic stress

The advent of plant tissue culture techniques appeared early on in the mid-seventies
as an excellent tool for developing and screening for biotic resistance on in vitro
derived explants from (Rai et al., 2011). The review described experiments on in vitro
screening methods applied to organogenic and/or embryogenic calli, shoots, somatic
embryos or cell suspensions by exposing them to toxins produced by the pathogen, to
pathogen culture filtrate or to pathogen itself. Saxena et al., (2008) reported in vitro
screening of callus cultures of rose-scented geranium, Pelargonium graveolens cv.
Hemanti against leaf blight disease caused by the fungal pathogen Alternaria
alternate. Callus and regenerated plants were confirmed for resistance by exposure to
culture filtrate of A. alternate. Semal (2013) described an easy and reliable method
for screening resistance to fungus (Figure 8.8). Lebeda and Svabova, (2010) also

219
reported on mass screening methods for the selection of disease resistant dessert and
plantain banana (Musa spp.); apple (Malus domestica); pineapple (Ananas comusus
[L.] Merr.); and also, for pea (Pisum sativum); melon (Cucumis spp.); lettuce (Lactuca
sativa L.); chickpea (Cicer arietinum L.); and various other tropical crops.

Jain et al., (2010) in turn, presented results on irradiation with gamma-rays of axillary
buds excised from in vitro-grown strawberry plants; where 5 percent of the plants
survived the selection pressure of Phytophthora cactorum crude extract and these
plants were also able to withstand drought for 5 – 6 days. Shoot tips and in vitro grown
proliferating buds of banana cv. Rasthali (Silk, AAB) were treated with EMS, NaN3
and DES and the mutated explants were screened in vitro for response to fusaric acid
and fungus filtrate. Selected lines were confirmed for resistance to the disease in pot
experiments and three resistant mutants were selected (Saraswathi et al., 2016).

It is however, highly recommended to consider in vitro screening for both biotic and
abiotic stress only as a pre-screen procedure, this is because quite often results may
not continue in the field where the pressure can differ and other adverse factors may
hamper the plant response.

Figure 8.7. Screening for tolerance to abiotic stress (salt), as modified from Djilianov et al., (2003).

220
In the dual flasks, the lower level is filled with the nutrient solution to which salt is
added in various concentrations. A piece of cloth connects to the upper level
containing only the nutrient solution. The plantlets cultured onto this medium will
then grow and response to the salt concentration may be visualized.

Phase 1
x Inoculation and mycelium development on the first
layer medium 1.
M
Mycelium
Medium 1
M

Phase 2
x Addition of second layer of medium spiked with
fungicide.
x Toxic material diffuses into the upper layer.
c M
Medium 2
x The depth of the second layer may be adjusted to ba
Mycelium
M
meet a given toxic effect (a, b, c). Medium 1
M

Phase 3
x Callus culture on the second layer of medium 2.
x The Benlate (22 ppm) or Mycostatin (80 ppm) Callus culture
added into medium 2; inhibit the development of Medium 2
the mycelium. M
Mycelium
Medium 1
M

Figure 8.8. Screening for resistance to biotic stress (fungus) – Double Layer Technique (modified from
Semal, 2013).

8.6.SOMACLONAL VARIATION

It has been abundantly documented that during the process of in vitro tissue culture
some noticeable variations may appear, especially when long culture periods are
involved; these in vitro induced variations are called somaclonal variations.
Somaclonal variation is generally defined as genetic, epi-genetic and/or phenotypic

221
variations observed among the plants or plant progenies derived from in vitro cells,
tissues and organs culture, probably as a result of gene spontaneous mutations or epi-
genetic phenomenon such as DNA methylation (Delgado-Paredes et al., 2017).
Somaclonal variants are not always considered solid mutants; however, from the
breeder’s point of view these variants may show improved traits and are still valuable
in crop improvement. Tripathy et al., (2016) described somaclonal variation among
regenerants in four genotypes of grass-pea (Lathyrus sativus L.) including a large
seeded somaclone: NGOG 5 having high seed yield and low neurotoxin content,
which was considered as a desirable candidate for future breeding programme of the
species.

Thakur and Ishii, (2014) identified two narrow-leaf phenotypes among a population
of hybrid poplar (Populus sieboldii × P grandidentata). These two variants appeared
to be identical but arose from different parental lines. Using DNA markers, the authors
demonstrated that these somaclonal variants, presented noticeable genetic and
heritable variation for both qualitative and quantitative traits. Several authors have
documented cytogenetic changes, single genes changes and also transposon mediated
mutations in variants from tissue culture. In some crops somaclonal variation caused
by methylation can be screened for, by simple DNA analysis. In fact, with the
advances in molecular technologies, the genetic characterization of the mutations is
becoming an attainable objective and could then lead to the validation of these variants
as mutants.

Somaclonal variants have been isolated for a variety of valuable traits like disease
resistance, insect resistance, stress (drought, salt, low temperature) resistance,
improved yield anefficient nutrient uptake etc., (Tripathy et al., 2016). Variations in
flower colour and morphology has been frequently observed in horticultural plants
such as cyclamen (Cyclamen persicum), carnation (Dianthus caryophyllus),
chrysanthemum (Chrysanthemum indicum), gerbera (Gerbera spp.), and torenia
(Torenia fournieri) have been commercially exploited through somaclonal variation
(Singh, Kumar and Verma, 2017).

8.7.PROCEDURE FOR MUTATION INDUCTION IN MUSA SPP.

Banana and plantain (Musa spp.) are two important crops in which all cultivars are
derived from spontaneous mutation without any breeding. Induced mutation
techniques are particularly important for bananas as there is limited sexual
reproduction and the technique can generate valuable genetic variation (Roux et al.,
2001).

The Musa spp. mutation induction system, based on in vitro techniques to recover
mutant plants and micro-propagate desirable mutants, was pioneered by Novák and

222
collaborators in the in the Joint FAO/IAEA Plant Breeding and Genetics Laboratory.
It is now applied worldwide in several Musa breeding programmes in dessert banana
(Musa acuminata and Musa balbisiana) and plantain banana (Musa
paradisiacal × Musa spp.), (Roux, 2004).

Step 1 – The process usually starts with in vitro the propagation of the plant material
using: shoot tips, corms, and also embryonic cells suspensions. Shoot tips are found
to be the most suitable and are the most commonly used hence, the protocol described
below is primarily for mutation induction of shoot tips. The shoot apices can be
excised either from suckers or male inflorescences. If embryogenic cell suspension is
to be used, immature male flowers and/or shoot tip derived samples are generally used
as initial explants. In these two methods, a few hundred initial explants are cultured
over a period of 3 – 4 months followed by several sub-cultures to obtain good quality
embryogenic callus and shoot developed. The choice between these methods will
depend on the need and capacity of the laboratory.

Step 2 – The determination of an optimal mutagenic dose for physical and/or chemical
mutagenesis. Success of any in vitro mutagenesis programme will depend on mutation
induction at a workable frequency followed by selection for desirable mutants. An
initial assessment of sensitivity of the plant part to specific mutagen treatments is
normally contacted from which the optimal dose is determined in developing a mutant
population.

Physical mutagenesis

x Shoot tips are treated with 10 doses: 10 to 100 Gy using a 60Co gamma
irradiation source at a dose rate of 44 Gy/min. For each Musa accession, at
least 200 explants are typically treated for radio-sensitivity testing and 20 non-
irradiated explants are used as controls. Immediately after irradiation, the
explants are plated onto fresh semi-solid MS medium with 20 μM BAP.
Radiation sensitivity and post-irradiation recovery are assessed by measuring
survival and multiplication rates, shoot height and fresh weight 40 days after
irradiation (Roux, 2004). The optimal dosage for mutation induction depends
on parameter studied, although survival rate and fresh weight are preferred.
The lethal dose at 50 percent (LD50) is calculated but relatively low doses are
advised because they produce less chromosomal damage and less negative
side effects than stronger treatments.
x After radio-sensitivity testing, the irradiation can be performed on about 2000
shoot tips, e.g. in batches of a few hundred depending on available resources,
labour, space in field or in the greenhouse, etc. It should be noted that optimal
dose depends on the variety/genotype but also on tissue culture conditions and
handling that are laboratory specific.

223
Chemical mutagenesis

x Shoot tips propagated in vitro can also be submitted to chemical mutagen

treatments. The optimal dosage for mutation induction is calculated as for
physical mutagens. Sodium azide (NaN3), diethyl sulphate (DES) and EMS
can be applied as chemical mutagens, Bhagwat and Duncan, (1998)
compared the effect of these three chemical mutagens, at various
concentrations on shoot tips of in vitro grown cultures of banana considering
two criteria as indicated in below formula: the number of apices that survive
the treatment and the number of regenerated shoots, coupled with the factor
of effectiveness (FE).

Total number of variations

‫( ܧܨ‬%) = × 100
Total number of apices treated

Step 3 – The disassociation of chimeras and in vitro sub-cultures: M1V1 to M1V4

Several rounds of vegetative propagation should be carried out to disassociate

chimeras, but the minimum number of cycles required depends on various conditions.
The M1V4 stage or if needed, M1V6 stage, are commonly used to complete the process
and these plantlets are transferred to a rooting medium.

Step 4 – Screening and selections may be conducted for biotic and/or abiotic stress in
the laboratory or in the greenhouse.

Step 5 – Acclimatization and planting in field for screening/selection and assessment

of agronomic characteristic.

Step 6 – Selected clones are propagated and evaluation is extended to field

performance, this will include multiplication of the mutated clones (under field or
laboratory conditions), confirmation of identified clones and evaluation for yield and
yield components.

Step 7 – Multi-locational trials: selected clones should be tested in different locations

before official testing and release (Table 8.2).

224
TABLE 8.2. THE CURRENT IN VITRO MUSA SPP. MUTATION-INDUCTION PROCESS USING
SHOOT-TIP CULTURE AND FIELD SELECTION

Steps
Dissection of sucker from the field
and shoot tip culture

Radio-sensitivity test
(minimum 200 shoot-tips/dose)

Mutagenic treatment of
minimum 2000 shoot tips with a LD50 dose

Micro-propagation M1V1

Micro-propagation M1V2

Micro-propagation M1V3

Rooting M1V4

Acclimatization and transfer to soil

Field selection

Genetic confirmation and agronomic evaluation

Micro-propagation of desired plants

Multi-location trials

225
8.8.PRACTICAL EXAMPLES OF IN VITRO MUTAGENESIS

Case Study 1: In vitro mutation induction for the improvement of Melon

(Cucumis melo cv. Yuva) initiated from the irradiation of plantlets and leaf or
cotyledon fragments, at the Aegean Research Institute, and the Turkish Atomic
Energy Authority, Sarayköy Nuclear Research and Training Centre, Turkey –
Y. .DQWR÷OXDQGFROODERUDWRUV (Figure 8.9).

Figure 8.9. Scheme for mutation induction in melon.

A. In vitro culture and irradiation

1. Irradiation of 7 days-old in vitro plantlets using a vertical 60Co gamma

ray irradiator.
2. Radio-sensitivity tests performed on 30 days old (M1V1) plantlets and
optimal dose determined as: 22 Gy.

226
 B. Screening for resistance to wilt disease

3. Filtrates of Fusarium oxysporum melonis race1, 2 were prepared for

screening for the tolerance to the Fusarium wilt disease. After the
treatment with the filtrate, 6-month-old rooted shoots and 3 weeks old
embryogenesis derived surviving shoots and/or plantlets plants were
sub-cultured several times and transferred to hardening conditions in
the greenhouse.

C. Acclimatization and transfer to the field

4. 36 to 42-month-old M1V6 plantlets transferred to the greenhouse are

one more time subjected to inoculation with Fusarium oxysporum
fungus filtrate in pots.
5. A few plants, which maintained tolerance/resistance and set fruits;
were then tested for agronomic and commercial value in various fields.

Case Study 2: Sugarcane improvement through in vitro mutagenesis – Bhabha

Atomic Research Centre, India (Suprasanna et al., 2008; Suprasanna, 2010)
(Figure 8.10).

Embryogenic callus cultures of a widely grown sugarcane (Saccharum

officinarum L.) var. Co 86032 were established from spindle leaf discs on callus
induction medium. Gamma radiation was applied to cultures at 10, 20, 30, 40 or 50 Gy
(~ 100 embryos per dose) and the LD50 dose was determined as 20 Gy. Plantlets were
regenerated from the callus and rooted on specific rooting medium, and then
transferred for hardening in the greenhouse. About 5,000 putative mutant plants were
planted in the field and at maturity data were recorded on agronomic traits including
number of millable canes, stool weight, number of internodes, cane weight, cane
diameter and Brix (a measure of total soluble solids). A total of 900 variants that
performed better in comparison to check varieties were selected and re-evaluated in
the field during the 2007 – 2008 to 2011 – 2012 growing seasons. Eleven selected
clones were isolated that were found to be superior for
morphological, quality and yield contributing characters (Table 8.3).

227
 1
7 2

6 3

5 4

Figure 8.10. Radiation mutagenesis in sugarcane (Saccharum officinarum L.) var. Co 86032. 1. Leaf
discs of the var. CoC671, 86,032 were plated onto specific embryogenic medium; 2. Callus was
irradiated at 10, 20, 30, 40, or 50 Gy and the LD50 dose determined as 20 Gy; 3. Callus cultures were
sub-cultured on new medium containing added NaCl (salt) at various concentrations for three
subsequent subcultures; 4. Tolerance to salt was measured by various parameters including cell
membrane damages, electrolyte leakage and free proline production; 5. After 30 days somatic embryos
which survived the salinity stress were transferred to embryo induction medium; 6. Fully developed
plantlets were scored and then transferred to acclimatization conditions then to hardening; 7. Selected
vigorous plants were later transferred to the field for trials and assessment of agronomic and
commercial value.

228
TABLE 8.3. MUTAGENESIS PROGRAMME IN SUGARCANE

Steps involved Generation Time duration

Establishment of in vitro embryogenic callus V0 2-3 months

cultures from young leaf roll explants collected
from field grown mother plants
Establishment of high frequency in vitro plant V0 2-4 months
regeneration and mass multiplication
Establishment of radio-sensitivity tests with 100 V0 2-4 months
cultures (per dose) and optimization of LD50 dose
Mutagenic treatment of 500 – 1000 cultures with M1V0 6-8 months
LD50 dose and selection
First sub-culture of selected callus lines M1V1 to M1 V2 2 months
Third sub-culture of selected callus lines M1V3 to M1 V4 2 months
Plant regeneration from selected irradiated cultures M1V4 2-4 months
Hardening of plantlets and evaluation in the nursery M1V4 4-6 months
on the basis of quality and phenotype
Evaluation under row trials on the basis of quality M1V4 12 months
and phenotype
Evaluation of selected clones in replicated trial M1V4 12 months
based on agronomic and biochemical characters
Vegetative propagation and maintenance of M1V4 12 months
selected clones –based on agronomic characters
and selection for stable phenotypic character
Vegetative propagation and maintenance of M1V4 12 months
selected clones based on agronomic characters and
selection for stable phenotypic characters
Multi-location trials, selection of clones and M1V4 12 months
evaluation in replicated trials based on agronomic
characters and stable new phenotypic character

229
B) HAPLOIDS AND DOUBLED HAPLOIDS IN MUTATION BREEDING

8.9.INTRODUCTION

A general principle in biology is that most organisms contain genetic contributions

from a male (sperm) and a female (egg) parent. The male and female gametes (sperm-
and egg cells, respectively) are the products of meiosis and contain half, i.e. n or one
set of chromosomes rather than the complete 2 sets or 2n chromosomes of the parent
somatic cells. The sporophyte of higher plants has paired chromosomes in their
genomes, abnormal plants with only a single (gametic) set of chromosomes are called
haploids and these can occur naturally at very low frequencies or be experimentally
induced in large numbers (see Section 8.2.2.).

Dunwell (2010) has provided as excellent recent review on haploids (Hs) and doubled
haploids (DHs). The first haploid described was a dwarf mutant form of cotton with
half the normal chromosome number (n), discovered in the early 1920s (Dunwell,
2010). In another pioneering study spontaneous haploids of Datura stramonium were
described by Blakeslee et al. (1922), that had 12 (n) chromosomes instead of the
normal 24 (2n) complement. It took four more decades until the first experiments were
under-taken in haploid induction in plants (Guha and Maheshwari, 1964) using
anthers of Datura innoxia. Haploidy/doubled haploidy in plant genetics and breeding
was first discussed seriously in a specially organized symposium in Guelph, Canada
in 1974 after the production of the first cultivar derived from a rapeseed (Brassica
napus), cv. Maris Haplona in the UK in the early 1970s (review by Forster et al.,
2007).

The main characteristic and advantage of haploids (Hs) is their potential for
conversion into doubled haploids (DHs) as these are genetically pure, homozygous
and fertile; haploids are of little use (other than their conversion to DHs or as
vegetatively propagated ornamentals) as they are often weak and infertile. Doubled
haploidy is the fastest route to homozygosity, a condition prized by plant breeders as
DHs can be the end products – cultivars – in the breeding of many crops such as rice,
wheat, barley, and rapeseed or provide parental lines in the production of F1 hybrid
genotypes (cultivars) such as maize and various vegetables, e.g. eggplant, pepper,
melon and tomato. Doubled haploids can also be maintained indefinitely by selfing or
clonal propagation.

In mutation breeding, it is known that most mutations in plants, spontaneous or

induced are genetically recessive and, accordingly, are only expressed in the
homozygous condition (see Chapters 1 and 2). This is one reason why mutant traits
are rarely observed in M1 populations as the individuals are usually heterozygous. The
development of homozygosity is therefore important in observing and assessing

230
mutant phenotypes. Thus, the targeting of haploids for mutation induction and their
conversion into doubled haploids is of immense interest in plant mutation breeding.

8.10. METHODS FOR HAPLOID/DOUBLED HAPLOID PRODUCTION

The main methods in H/DH production in crop plants are described below.

8.10.1. Haploids via androgenesis

As stated earlier, the first experimentally produced haploid plants were obtained
through anther culture; these cultures were, in vitro induced to produce embryos from
male gametic cells (androgenesis) from Datura innoxia (Guha and Maheshwari,
1964). Androgenesis is to date, the easiest and most common procedure for the
generation of haploid in large numbers and in a wide range of plants. Careful
histological studies showed that the haploid embryos produced arise from microspores
at the mid to late uni-nucleate stage in the development of pollen grains (Szarejko,
2012). Specialized cultural conditions of these microspores with specific hormones
and nutrients divert the natural development from pollen grains into haploid embryos
(Figures 8.11 and 8.12). Great care is required as anther walls include diploid
(parental) tissue which may be confused with DH production. To circumvent this,
scientists initiated in vitro culture of isolated microspores obtained through a gentle
homogenization of anthers using a mortar and pestle or a blender for cell separation
(Szarejko, 2012).

This technology (Figures 8.13 and 8.14) requires skilled operators to produce high
quality, of healthy isolated microspores. Szarejko and her team with extensive
experience in the use of haploids in mutation breeding emphasized the need to confirm
the haploidy stage of the plants using either direct cytological or flow cytometry
techniques for chromosome counting and/or ploidy-level determination (Szarejko
et al., 1995; Szarejko, 2003; Szarejko, 2012). Ploidy confirmation may also be
verified using simpler, indirect methods based on guard cell size and plastids
dimensions, which are correlated with ploidy-level (Yuan et al., 2009). DNA markers
can be used to confirm homozygosity in potential DHs, one useful method is enzyme
mismatch cleavage of amplified DNA described by Till et al., (2004) and applied later
for DH determination in tef (Eragrostis tef), (Till et al., 2017). These methods provide
evidence that informative markers (unlinked and polymorphic in parental lines) are
homozygous and that the line is a doubled haploid with the following probability.

231
 Number of informative Probability of being a DH
homozygous markers
1 50.000%
2 75.000%
3 87.500%
4 93.750%
5 96.875%
6 98.437%
7 99.218%
8 99.608%
9 99.803%
10 99.900%

Plants responsive to androgenesis include: rice (Oryza sativa), common wheat

(Triticum aestivum), durum wheat (Triticum durum), maize (Zea mays), barley
(Hordeum vulgare), triticale (x triticosecale), rye (Secale cereale), timothy
(Phleum spp.), ryegrass (Lolium multiflorum), rapeseed (Brassica napus), broccoli
(Brassica oleracea), tobacco (Nicotiana tobacum), potato (Solanum tuberosum),
linseed (Linus usitatissimum), asparagus (Asparagus officinalis), and apple
(Malus domestica), aspen (Populus spp.), oak (Quercus spp.) and citrus (Citrus spp.).

232
 Figure 8.11. In vivo gametophytic development of a microspore.

Figure 8.12. Alternative Pathways for haploid development from a microspore (modified from
Bhojwani and Razdan, 1983).

233
Figure 8.13. Isolation of microspores from barley spikes. (A) Barley donor plants growth in a chamber
DW ై& DQG K SKRWRSHULRG % 0RUSKRORJLFDO IHDWXUHV DVVRFLDWHG ZLWK PLFURVSRUH
developmental stage. (C) Microspore at mid-late developmental stage. (D) Pretreatment of spikes for
2 weeks at 4 °C. (E) Cut spikes on 1cm fragments before blending. (F) Placement of spikes to the
blender chamber. (G) Blending spikes for 15 secs at low speed in 0,4 M mannitol. (H) Filter through
100 μm nylon mesh. (I) Centrifugation for 10 min at 110 × g (J). Viable microspores located in the
interphase after centrifugation in density gradient. (K) Determination of the number of microspores
using counting chamber. (L) Microspores suspension in an appropriate amount of induction medium.
(M) Supplementation of microspores with induction medium. Courtesy of M. Gajecka.

234
Figure 8.14. Microspore in vitro culture. (A) Microspores after 7 days of culture in induction medium.
(B) Supplementation with medium after 7 days. (C) Continuation of culture at 65 rpm and 25°C in the
dark for 2 weeks. (D) Microspore culture in induction medium after 21 days. (E) Microspores-derived
embryos after 14 days of culture on differentiation medium. (F) Androgenic embryos cultured on
regeneration medium at 25°C in the dark for 5 days and continued in the light. (G) Growth of
androgenic plantlets in in vitro culture. (H) Growth of plantlets in the soil. Courtesy of M. Gajecka.

235
 8.10.2. Haploids via gynogenesis

Haploid plants may also be induced from maternal gametic cells, e.g. the haploid cells
produced after meiosis of the megaspore which comprise the ovule. However, these
cells, typically the egg, are produced in very small numbers compared to those
produced for pollen production. Thus, the frequency of haploids via gynogenesis is
often far lower that of androgenesis, but considering that some species are recalcitrant
to androgenesis, gynogenesis offers a good alternative (Dunwell, 2010; Chen et al.,
2011; Germana, 2012).

Plants responsive to gynogenesis include: onion (Allium cepa), sugar beet (Beta
vulgaris), pepper (Capsicum annuum), maize (Zea mays), sweet potato (Ipomea
batatas), tulip (Tulipa generiana), barley (Hordem vulgare) and cucumber (Cucumis
sativus).

8.10.3. Haploids via wide crosses followed by chromosome elimination

In many species, egg cells may be stimulated to develop into haploid embryos after
pollination with pollen of another-, but related species as it is the case, for example,
when barley (Hordeum vulgare) is pollinated by the related species, H. bulbosum. The
fertilization occurs, but after the chromosomes from H. bulbosum get eliminated
during the early cell divisions of the embryo development (Kasha and Kao, 1970).
Similarly, wheat crosses with maize (Zea mays) lead to fertilization of an egg cell by
maize sperm followed by elimination of maize chromosomes in subsequent mitotic
divisions at early stages of embryo development.

These haploid embryos need to be rescued and cultured in vitro as the endosperm does
not developed and the embryo frequently aborts in vivo (although some of them may
survive, see Spontaneous haploids, 8.2.2.4. below).

Plants responsive to chromosome elimination after pollination with alien pollen

include: wheat (Triticum spp.), barley (Hordeum vulgare), pearl millet (Pennisetum
spp), sunflower (Helianthus annuus), strawberry (Fragaria spp), and maize
(Zea mays). This method is usually deployed in situations where other H/DH methods
fail, however in wheat the pollination with maize is commonly used in breeding as
the main method of DH production. For reviews see, among others: Devaux and
Pickering( 2005); Forster et al. (2007); Dunwell (2010); (Dunwell et al., 2010) and
Szarejko (2012).

8.10.4. Spontaneous haploids

Reproduction in higher plants is normally characterized by double fertilization:

resulting in 1) a zygote with 2n chromosomes (union of sperm- and egg cells) and

236
2) the endosperm with 3n chromosomes (union of sperm and two central nuclei of the
ovule), both occur in the ovule, which then develops into a seed. However, in some
species seeds may contain more than one embryo, polyembryony, in quite a few
instances the additional twin embryo is a haploid. Polyembryony was first described
in 1719 by Leeuwenhoek in orange seeds containing twin embryos. The phenomenon
was later observed in numerous species, genera and families of both gymnosperms
and angiosperms (Koltunow, Hidaka and Robinson, 1996). This is a natural
phenomenon, which occurs in many species a rate of about 1 in 100,000 seeds, for
example in oil pam (Forster et al., 2007). This frequency was considered too low for
practical purposes, but in recent years high-throughput detection methods, for
example the use of flow cytometry have allowed the efficient screening and detection
of haploids in oil palm (Nasution et al., 2013), and such methods may be applied to
other species.

Species amenable to haploid production via spontaneous production in seeds include,

among others: oil palm (Elaeis guineensis), pepper (Capsicum annuum), coffee
(Coffea arabica), and cotton (Gossypium hirsutum).

8.10.5. Haploids via aberrant pollination

In the early 1900s, Hertwig (1911) observed that frog sperm exposed to radium
retained their ability to fertilize eggs but lost their genetic function. This was called
then the “Hertwig Effect”. This observation demonstrated that eggs may be stimulated
to develop spontaneously without being fertilized after what is called: aberrant
pollination. Later, the technology was adapted to tobacco (Nicotina tabaccum) by
Pandey and Phung, 1982 as cited by (Sato et al., 2000; Germana, 2012) who worked
on carnation (Dianthus caryophyllus) and fruit crops, respectively. The examples of
successful application of radiation-stimulated parthenogenesis include citrus
(Citrus clementina) and rose (Rosa sp.). Gynogenetic haploid plants in muskmelon
(Cucumis melo) were also induced by pollination with pollen irradiated by 60Co
gamma-rays and in vitro culture of fertilized ovules or immature embryos by (Sauton
and Dumas de Vaulx, 1987). This methodology was later improved by Yetisir and
Sari (2003), who found that ܵ-ray doses higher than 30 Krads (300 Gy) were necessary
to avoid normal fertilization and obtain only haploid embryos (all haploids showed
expected maternal phenotypes).

Further studies in maize (Zea mays) showed the development of DHs expressing
various types of mutation, namely chromosomes aberrations, and some of these
mutants were viable and fertile (Viccini and de Carvalho, 2002). The same effect was
obtained in a few other crops, including: apple (Malus domesticus), and barley
(Hordeum vulgaris).

237
 8.10.6. Haploid inducer genes

Ravi and Chan (2010) indicated that haploid plants in the model species, Arabidopsis
thaliana can be regenerated by crossing cenh3 null mutants expressing altered
CENH3 proteins with wild type. This was an important breakthrough for Arabidopsis
research as Arabidopsis is recalcitrant to other H/DH systems described above.
Chromosomes from the mutant (cenh3) line are eliminated as they are not able to
attach to the metaphase spindle apparatus.

This method is yet to be applied to crop plants, though homologous of the cenh3
mutation are being searched for in crop plants.

8.11. MAIN APPROACHES IN HAPLOID AND DOUBLED HAPLOID

MUTAGENESIS

In recent years, haploid (H) and doubled haploid (DH) production of plants gained
increased popularity and is now applied to a wide range of crops, mostly for breeding
cereals, grasses, oil and other industrial crops, tubers and vegetables, trees, and
ornamentals. It is important to call plants experimentally doubled from haploids:
doubled haploids (DHs) as they are completely homozygous and true breeding and
thus, distinct from normal diploids, which have the same ploidy but may contain
heterozygous loci. The doubling of the chromosome sets can occur spontaneously
during mitosis, or be induced, e.g. by a treatment with colchicine (Yuan et al., 2009).
Maluszynski, et al., (2003) provided many protocols for H/DH production in crop
plants and extensively described their use in plant breeding.

There are many examples of successes in crop breeding using H/DH methods and
these include an important number in mutation breeding (Maluszynski, et al., 2003;
Jain and Spencer, 2006; Szarejko and Forster, 2007; Dunwell, 2010; Szarejko, 2012;
and Mba et al., 2012).

Vos et al., (2009) described induced mutations from spontaneous doubled haploid
embryos in oranges (Citrus spp.). The mutants displayed excellent fruit
characteristics, good yield, and also resistance to some diseases. An alternative
approach for inducing mutation from haploid cells is to irradiate or treat with chemical
mutagen agents, anthers or microspores cultures in vitro (Szarejko, 2003 and 2012).

8.11.1. Doubled haploid production from mutant plants

Doubled haploidy has relevance in plant mutation breeding as a means of fixing

mutations in a homozygous state. The DH lines can be produced from plants at any
generation from M1 to Mn, but for accelerated breeding the sooner the better.
Theoretically, all the methodologies cited earlier (Section 8.2.2.) for haploid plant

238
production can be used for inducing haploids/doubled haploids; however, in practice
most of the published examples of H/DH mutant genotypes have been obtained
through androgenesis, notably in rice, wheat, barley, vegetables and medicinal plants
(Szarejko, 2012). M1 plants are therefore of interest, but there are two major
constraints: 1) preferably only plants carrying interesting mutants should be selected
and this would require genotypic screening (see Section 8.3.), and 2) M1 plants are
often weak and therefore not good donors for haploid/doubled haploid methods.
Despite this, DHs can be produced, see Section 8.2.5.1., for a protocol. Stable mutants
have been produced from fertilized egg cells of a japonica rice (Oryza sativa) variety
treated with ethyl methane sulphonate (EMS) and N-methyl-N-nitrosourea (MNU).
The doubled haploids (DHs) derived from anther culture of these M1 plants were
stable and could be used as new breeding materials (Lee, Cheong & Kim, 2003).

However, M2 plants (and subsequent generations) provide a more practicable option.

In addition to fixing the mutant gene the H/DH plants produced will be the result of
one round of meiosis from M1 plants, two rounds for M2, three form M3 etc., and
therefore DHs produced carrying the mutation of interest can also be screened for
variation in background mutation load, i.e. selection of mutant lines with the minimal
background disturbance (see Section 8.3.).

The production of haploids from the M1 generation still offers great promise, as any
mutation existing in a homozygous genotype is likely to be confirmed and visualized
phenotypically as soon as the doubling of the chromosomes number is achieved as
there would be no segregation. This also means that in only one generation after
irradiation stable mutants can be obtained, thus short cutting considerably the
breeding cycles needed for obtaining true breeding material. Methods for mutant DH
production from M1 populations vary depending on the species, but the basic
protocols are the same and here we take barley protocols as an example.

8.11.2. Mutagenesis of haploid cells

Irradiation of haploid cells (anthers and/or pollen grains) has been used for:
1) enhancing the understanding of pollen grain germination and its possible impact of
flowering and maturity in crops; and 2) for the induction of mutation of haploid plant
materials. In general, these studies showed that when working with haploid cells in
mutagenesis attention must be paid to:

x the methods for the H/DH production;

x the choice of genotype (should be responsive to H/DH method);
x the choice of the mutagenic treatment (physical, chemical, biological);
x the assessment of effective mutagenic doses;

239
 x the screening process to be used to detect mutants (genotypic and/or
phenotypic); and
x the selection of the optimal developmental stage to be used for mutation
induction.

8.12. HAPLOID CULTURE AND MUTATION BREEDING

The advantages of the application of haploid methods in plant breeding have been
reviewed extensively (Maluszynski et al., 2003); Szarejko and Forster, 2007;
Dunwell, 2010). For plant breeding purpose, the development of haploids is exploited
to produce:

x pre-breeding plant material;

x rare varieties or when variety improvement is not attainable from
conventional crosses, in developing stable pure (homozygous) lines, and
x accelerated breeding cycles.

Coupling H/DH technology with mutation induction provides a rapid means of

producing mutant lines in a homozygous condition, thus speeding up the development
of new varieties with mutant traits.

Care is required for physical irradiation and chemical mutagenic treatments as

reproductive cells and tissues are fragile. Although all the known physical and
chemical mutagens have been applied to haploid cells, the most used agent is a gentle
treatment by UV light (see Chapter 1). This mutagen agent has many advantages as it
is readily available in most plant tissue culture laboratories, it is relatively cheap, and
the radiation doses are generally in the lower energy range. When a treatment is too
severe (e.g. gamma- or X-rays) it may cause sterility and viability problems and may
result in a low density of mutations, which in turn will require more screening to
identify the specific mutation. The low penetration of UV light irradiation requires a
large amount of pollen grains and the strict use of a one cell layer to ensure efficient
induction of mutation.

The irradiated material then needs to go through regeneration steps to produce H/DH
plants that would then be subjected to mutant gene/trait screening and selection
processes (Maluszynski, et al., 2003); Forster et al., 2007). It is therefore advisable to
mutagenize several batches over a range of mutagen dose rates and concentrations,
then choose the batch that shows the best trade-off between, stability, fertility and
mutation damage/induction.

240
8.13. PROTOCOLS IN MUTAGENESIS USING HAPLOID/DOUBLED
HAPLOID SYSTEMS

Haploids have been targeted by plant mutation breeders since the first
haploid/doubled haploid technologies were developed as they provide many
advantages. Chief among these is the immediate production of homozygous mutant
alleles on doubling haploid cells/tissues treated with a mutagen. Mutation induction
using microspores was first deployed using responsive model species such as barley
and brassicas see Maluszynski, et al., (2003); Forster et al., (2007); and Szarejko,
(2012) for reviews. Further studies in maize showed the development of DHs
expressing various types of mutations, namely chromosomes aberrations, and some
of these mutants were viable and fertile.

8.13.1. Barley: mutant haploid/doubled haploid production from M1

populations

In this approach, the M1 mutant population is produced by standard methods

(see Chapters 1, 2 and 4), but DHs are produced from M1 plants. The example given
here involves mutation induction via seed mutagenesis (Figure 8.15).

x Mutagenic treatment
Seeds (M0) of an elite line are selected for mutagenesis and treated as
described in Chapters 1, 2 and 4 to produce M1 seeds. Care is needed to select
the optimum dose treatment (usually in the range LD30 – LD50) to obtain a
mutant population in which the frequency of the desired mutant(s) is
detectable.

x Propagation of M1 plants
The M1 plants suffer from physiological disorders as well as chimeras and
they need to be grown up in optimal conditions, normally a growth chamber
or a greenhouse with controlled temperature and lighting. The M1 also carries
mutations in a heterozygous condition. It is essential that healthy plants are
produced as these are the best donors for haploid production.

x Doubled haploid production

Doubled haploid methods are applied directly on plants from the
M1 population. The choice of method (androgenesis, gynogenesis, aberrant
pollination etc.) depends on the species and genotype, a wide range of species
specific protocols is provided in the manual: Doubled Haploid Production in
Crop Plants (Maluszynski et al., 2003). Ideally only plants carrying

241
 mutations of interest should be selected as donors for DH production, but in
the M1 this is restricted to genotypic selection (see Section 8.3.).

x Selection of DH mutants
The DHs produced may be selected during the culturing process both
genotypically and phenotypically. Screening may also be done genotypically
and phenotypically on hardened plants produced from in vitro culture,
normally in a greenhouse. However, it should be noted that phenotypic
testing in culture and immediately on plants derived from culture is un-
reliable as the plants can suffer from physiological disorders and somaclonal
variation, therefore phenotypic testing is best done on later generations.

8.13.2. Barley: in vitro haploid/doubled haploid mutant production

Microspore culture is the best system, though anther culture can also be used and can
be applied to any species where these methods work. Mutation induction is targeted
at the single cell microspore stage shortly after isolation (uni-nucleate stage, see
Figure 8.16). The factors to consider are listed below.

x Mutagenic treatment
The mutagenic treatment must be applied at the uni-nucleate stage of
microspore development. Since the microspore stage is normally stage
material is sampled mutagenic treatment must be applied immediately or
soon after sampling. If the microspore is allowed to develop and produce
more nuclei they approach becomes use-less as more than one nucleus may
carry mutations and chimeras will develop. It should be noted that the
mutation treatment (physical or chemical) will drastically reduce the viability
of microspores and subsequent embryogenesis. This is one of the reasons
why UV light is a favoured (gentle) mutagen for microspores. If gamma is
used then a gamma source with a low activity should be used. Radio-
sensitivity of microspores is normally determined by measuring lethality,
“killing curves”. The dose selected should allow for sufficient
embryogenesis in producing a DH M1 population that can be screened
in vitro and eventually in the field.

Physical and chemical mutagens can be used though chemical mutagens are
more difficult to handle in terms of treatment and washing to remove
chemicals from cultures, discard of waste toxic materials and human health
and safety (see Chapter 2). EMS, ENU, MNU and sodium azide have been
used in barley and Brassica microspore mutagenesis (summarized by
Szarejko, 2003).

242
 Although microspores are the preferred materials for haploid/doubled
haploid production other reproductive tissues and cells can be targeted such
as immature inflorescences and isolated ovules.

x Selection in culture
Some traits may be selected for in culture, e.g. drought and salinity
(see Section 8.1.), these are usually classed as pre-screens as the acid test for
phenotypic traits is screening in the field.

x Regenerate doubled haploids (DH M1 mutants)

In some species, notably barley, there is a high rate of spontaneous doubling
in vitro during embryogenesis. This can attain more than 60 percent and thus,
precludes the need for artificial doubling with chemical agents such as
colchicine and oryzalin, which may be applied in vitro, or later on seedling
plant stage.

x Primary evaluation of plants in field conditions

A great advantage of DH mutants is that DHs can be multiplied and tested
repeatedly, e.g. in replicated, multi-locational and multi-season trials.
However, sufficient materials must first be generated and this is usually done
by growing the first generation from tissue culture in optimal greenhouse
conditions to maximize seed production.

243
Figure 8.15. Development of mutant lines using seed mutagenesis and DH systems, modified from
(Szarejko, 2003).

244
Figure 8.16. In vitro mutagenesis using isolated microspore culture from Szarejko (2003).

8.13.3. Doubled haploids in pedigree mutation breeding

Each breeding method is based on a wide genetic background. In case of mutation

breeding, there are different methods to increase the genetic variation/segregation via
different mutagenic agents. Two breeding methods are proposed. The first is to use
DH techniques in M1/M2 generations. In some exemption, when heterozygosity still
exists in M3-4 generations, alternative method is to release DHs from later generations
(M3/M4), when the breeder has important positive data about the selected but
segregated mutant lines (Pauk et al. 2004). The success of the techniques depends on
the used protocol (see Maluszynski, et al., 2003) and the routine and skills in the
particular laboratory. It should be noted that, such methods are continually updated
and verified against new publications to improve the laboratory protocol step by step
standard procedures.

245
In the mutation pedigree breeding methods two breeding alternatives are integrated.
The first breeding alternative is the early-generation haploid induction, and the other
one is the late-mutant generation induction. These two breeding alternatives have
different advantages and disadvantages as stated below.

x Releasing mutant DH lines from early generation (M1 or M2) of mutated

population
Spontaneous and colchicine treated mutant DH lines after one-year
evaluation and seed propagation are passed to the replicated and multi-
location performance tests. By this alternative 3 – 5 selection steps can be
saved in the breeding process.

x Releasing mutant DH lines from late generation (M3 or M4) of mutated

population
The second alternative shows, the selection in early generations (M1, and M2)
and starts the H/DH lines induction from only later mutant (M3 or M4)
generations. In this alternative the DHs are induced from selected population
genetically more valuable population than in the previous alternative.

x Evaluation of using DH lines in mutation breeding

Benefit of the 1st breeding alternative is the perfect homogeneity,
significantly shorter breeding time. Disadvantages are high number of
unusable DH mutants relatively, because of the lack of selection for
agronomic characters in selection process.

Benefits of the 2nd breeding alterative are perfect homogeneity, more usable
DH lines (positive mutants) in breeding, compared with the previous
alternative. Disadvantage of the 2nd alternative is that, there is no significant
time saving (only some years) in the breeding process.

8.14. SCREENING FOR HAPLOID/DOUBLED HAPLOID MUTANTS

8.14.1. Phenotyping

As in any other crop breeding programme, the production of haploid then doubled-
haploid mutant plants needs a thorough screening for the selection of the desirable
mutant trait. Haploids of higher plants can be distinguished from their diploid
equivalent in many ways. Most obviously from the perspective of phenotype, as they
are usually smaller in appearance, partly because of their smaller cell size; bearing in
mind that in general terms, cell volumes in plants are positively correlated to ploidy
level. The most widely used of these phenotypic methods is the measurement of

246
stomatal guard cell length and chloroplast content in these cells, nevertheless none of
these phenotypic predictors of haploidy is absolutely reliable. Methods providing
direct measurements of genome size provide are a far more reliable diagnosis of
haploid status. These include direct measurement of the chromosome number, using
conventional chromosome counting techniques (cytology) and measurement of the
DNA content using micro densitometry, or flow cytometry (Dunwell, 2010; Szarejko,
2012).

It should be noted that the frequency of mutated DH lines, be it in natural conditions

or experimentally induced is generally very low or low: ~ 0.00001 in natural
occurrences and 0.8 to15.8 percent in haploid experimental production. Increasing the
number of haploids is the key priority when searching for a viable and ultimate
haploid. Several methodologies have been assayed for increasing this rates; Szarejko
et al., (1995) reported rates at up to 25 percent resulting from the use of sodium azide
and NMU (N-methyl-N-nitrosourea). Nasution et al., (2013) with a rigorous and high
throughput approach using flow cytometry screening obtained up to 1/1000 seedlings
among natural off-types oil palm individuals.

Extensive review books and articles are available among which provide instrumental
protocols and tips for developing and identifying haploid and doubled haploid plants
for breeding purposes (Maluszynski et al., 2003; Forster et al., 2007; and Szarejko,
2012).

8.14.2. Genotyping

For scientists, as well as for breeders it is imperative to validate, at the earliest stage
possible the genetics of the lines in the pipeline for selection and/or for genetic
research. Thus, the plant populations on which to perform such validation where
formally defined including: F2, recombinant inbred lines, near-isogenic lines, etc.
Haploid and doubled-haploid populations have been recognized as the best plant
material for such purpose as they have simpler genome (n) or stable homozygous
genome (2n). In mutation breeding this becomes even more important as any recessive
mutation in the haploid stage readily appears in the doubled-haploid stage. So,
numerous tentative for applying genotyping techniques have been made. However,
they were not always successful nor were they worthwhile to apply in terms of costs
and labor.

The first approaches for validating haploid plants were: cytology, flow cytometry, and
from the onset of the tools in molecular biology and genetic markers some
acceleration was observed. Genetic maps for various crops where then developed
using different types of markers, including to haploid and doubled-ãhaploid plants
(Kuchel et al., 2007). DHs are invaluable in genetic mapping and trait mapping as
they ae “immortal” and can be used repeatedly, especially for new marker and new

247
trait screening. Numerous DH mapping populations exist in barley, rapeseed, and
wheat but are rare in triticale oat, rye, and others (Tuvesson et al., 2007; Seymour,
Taylor and Tucker, 2012). Genome sequencing and high throughput genotyping
methodologies contributed to an additional push and allowed the development of
markers for specific genes and/or QTL.

Close et al., (2009) in an in-depth critical review of the development of genetic

linkage maps in barley, provided a high-density consensus map based only on
complete and error-free datasets supported by a readily available SNP genotyping
resource. Till et al. (2017) likewise developed a low cost protocol for validation of
doubled -haploid plants by enzymatic mismatch cleavage in barley.

Genomic selection (GS) is an emerging as a new tool in plant breeding. Thus, GS uses
comprehensive marker information to calculate breeding values for complex crops
traits (Cros et al., 2015; Heffner et al., 2010). Genotypic selection is particularly
relevant for DH mutation programmes as DNA can be extracted from cultured
materials (callus, embryos, and plantlets) and early selection made for advancing
material of interest. Also, newly regenerated plants grown on in greenhouse
conditions often suffer from physiological disorders and therefore cannot be screened
reliably for phenotypic traits, but DNA analysis and selection for mutation can be
performed.

248
C) DNA MARKER AND GENOTYPING APPLICATIONS FOR MUTATION
BREEDING

8.15. INTRODUCTION

The value of genetic markers as indirect selection indicators in plant breeding has
been known for over 90 years. However, it was not until the mid-1980s that abundant
molecular markers became available for reliable selection of agronomically important
traits in breeding programmes. Since then, indirect selection using DNA markers has
significantly increased the efficiency and speed of plant breeding. The turn of the 21st
century witnessed another major leap forward with the advent of automated
technologies, next-generation DNA sequencing and enabling statistical and
bioinformatics tools. With regards to plant mutation breeding, the imminent impact
of these new concepts and approaches will be in: 1) marker-assisted backcrossing for
introgression or pyramiding of mutant alleles; 2) increased speed and precision to
detect mutations in genes underlying important traits enabling genotypic selection;
and, 3) improved design of mutation breeding programmes. For example, reverse
genetics techniques are providing new insights into the landscape of induced
mutations in plants, enabling a more rational choice and dosage of the mutagenic
agent. Identification of causative mutations at agronomically important loci can now
be performed with unprecedented speed and precision. This knowledge can then be
translated into functional markers, which show complete linkage with trait locus
alleles. In one application, genotyping assays can be used for marker-assisted
selection of mutant traits, in a similar way as done in cross breeding. In addition, direct
molecular screening for induced mutations in known target genes can enable selection
of candidate mutants at the initial stages of a mutation breeding programme. This can
significantly enhance the efficiency of mutant selection and expand the scope of
mutation breeding to crops or trees that have a long juvenile stage and which have
hitherto lagged, behind the annual crops. This chapter introduces selected concepts
and applications in the use of molecular tools and techniques for experimental and
applied plant mutagenesis. Two protocols are provided as examples: 1) a procedure
for identifying small sequence variations in large mutant populations using high-
throughput sequencing of PCR amplicons that can be applied to diploid crops;
2) genotyping assays for diagnostics and marker-assisted selection of shell thickness
in fruits of oil palm (Elaeis guineensis Jacq.).

8.16. ADVANTAGES AND USE OF MOLECULAR TECHNIQUES FOR PLANT

MUTATION BREEDING

As set out at the beginning of this manual, mutations are heritable changes in the
DNA. DNA-based molecular markers are polymorphic DNA fragments/sequences

249
that have been used widely for genotyping and diversity analysis (Staub, Serquen and
Gupta, 1996).

DNA-based markers offer several advantages compared to morphological markers:

1) Increased reliability: phenotypic assays are affected by environmental

conditions, heritability of the trait and other factors; thus, genotyping scores
based on DNA markers tend to be more reliable compared to measurements
of phenotypes.

2) Increased efficiency: DNA markers can be scored at seedling stages thus

considerable time and space can be saved, especially for traits expressed at
later stages of development (e.g. flower, fruit or seed characteristics).

3) Reduced cost: for example, PCR-based assays can be cost-effective

compared to phenotyping, especially in a high-throughput setting.

4) Uniqueness: DNA markers are unique for each gene/allele and have the
power to enable identification of more than one mutation for the same trait.
Thus, markers or genotyping assays can be used to screen simultaneously for
multiple genes or alleles which are useful for example in gene pyramiding.

A review of the types and characteristics of DNA markers and their applications in
mutation is provided by Wu et al. (2012).

The rapid development of DNA-marker techniques and genomics over the past
2 – 3 decades is changing the way traditional and mutation-based plant breeding is
being practiced. For example, acquisition of nucleotide variations via sequencing
allows the development of simple PCR-based markers for genotyping assays such as
allele-specific amplification, high-resolution melt analysis (HRM), cleaved amplified
polymorphic sequences (CAPS), and other assays.

Molecular methods for crop improvement are becoming increasingly automated and
reliable, and costs have come down. Significantly for mutation breeding, high
throughput methods are available which enable thousands of samples to be analysed
in a matter of weeks. While many methods have been described, the trend is that direct
DNA sequencing is becoming the standard platform from which new approaches are
being developed. The use of sequencing therefore provides new and precise tools, and
concomitantly new strategies, for both mutation detection and marker-assisted
selection.

Figure 8.16 illustrates the different stages of a plant mutation breeding programme
from the initial step of mutation induction to mutant selection and the subsequent

250
utilization of mutant alleles for introgression or pyramiding into elite germplasm.
DNA marker techniques and high-throughput sequencing methods can be applied at
different stages to facilitate some of the investigations or accelerate specific steps in
the mutation breeding programme. Note that in case of functional markers, the
molecular marker shows complete linkage with the mutant allele (Andersen and
Lübberstedt, 2003).

With respect to plant mutation breeding, two important applications of DNA markers
and high-throughput DNA sequencing are: 1) detection of mutations in target genes
of known sequence that control important traits to enable genotypic screening at the
early stages of a mutation breeding programme; and, 2) the use of markers for
introgressing or pyramiding important mutant traits through marker-assisted
backcrossing. For marker development, appropriate experimental populations should
be developed from mutants and wild-type parents to enable identification of the
causative mutation(s) and to confirm linkage with the trait (phenotype).

Additional general applications of DNA markers for mutation breeding include, for
example, the identification of off-types or contaminants that may have been
introduced during the mutation breeding programme (see Chapter 5) and the use of
markers to reduce the non-desirable mutations and maintain the elite genetic
background.

These applications are described briefly in the following sections followed by two
examples of protocols. In addition, genotyping using DNA markers can be considered
the most reliable method for the identification of mutant lines and varieties. Therefore,
molecular tools are also useful for identification of the genotype/cultivar, and when
appropriate, for seed purity analysis prior to mutagenesis.

251
Figure 8.16. Molecular marker and high-throughput sequencing applications in plant mutation
research or breeding. Adapted from Wu, Shu and Li (2012).

252
8.17. MARKER-ASSISTED BACKCROSSING

Marker-Assisted Selection (MAS) is a process whereby a marker is used for indirect

selection of a genetic determinant(s) of a trait of interest. The method involves
selection of plants carrying genomic regions that are involved in the expression of the
targeted trait(s) through the application of molecular markers. The availability of
molecular markers along with dense genetic maps enabled MAS for major genes and
quantitative trait loci. Thus, with the application of MAS, individual plants can be
selected based on their genotype during the selection process. MAS is widely used to
increase the effectiveness of backcross breeding to accelerate the introgression or
pyramiding of traits of agronomic interest from a donor parent to an elite, recurrent
parent (Das, Patra and Baek, 2017). Marker applications are particularly significant
for the selection of mutant traits as these are predominantly recessive and can only be
recognised phenotypically in the homozygous state. The process of selecting for new
traits from a donor parent using markers that are tightly linked to the gene of interest
is referred to as ‘foreground selection’ while ‘background selection’ refers to using
markers to select against other DNA from the donor parent, i.e. to maintain the elite
genetic background of the recurrent parent.

An important mutant trait in oil palm is shell thickness in fruit. The wild-type fruit
has a thick protective shell around the kernel (Dura fruit type, genotype Sh/Sh,
Figure 8.17a); mutant fruit is shell-less (Pisifera, genotype sh/sh, Figure 8.17b).
Commercial oil palm is the heterozygote between these two and has thin-shelled fruit
(Tenera, Sh/sh, Figure 8.17c) and is produced from Dura × Pisifera crosses. Tenera
fruits yield 30% more oil than Dura fruits. The mutant Pisifera suffers from female
sterility and is used as a male (pollen) parent in commercial Tenera seed production.
The shell thickness gene (Sh) is the most economically important gene in oil palm
production and has been the subject of intense genetic studies (Singh et al., 2013a).
The Sh gene has been sequenced and DNA diagnostic markers have been developed
for wild-type and mutant alleles (Singh et al., 2015a).

253
Figure 8.17. Shell thickness phenotypes of oil palm fruits: a) Dura wild-type fruit with a thick shell
protecting the kernel; b) Pisifera mutant-type fruit with no shell around the kernel; and c) Tenera thin-
shelled commercial fruit. Courtesy of B.P. Forster.

Genetic markers are important in screening for fruit type in oil palm as phenotyping
can only be done on mature palms and this takes 4 – 5 years from sowing seed.
Genotyping saves time and space, and can be done on seedlings, and/or selections,
made in the nursery, i.e. prior to field planting. Screening for fruit type is important
for various reasons: 1) to check the purity of Teneras in commercial plantations; 2) to
check for purity of Teneras in commercial seed production (Kelanaputra et al., 2018
in press); 3) to select for fruit types in progenies from breeding programmes and select
specific types for field trialling (Setiawati et al., 2018 in press); and 4) to screen for
new induced mutations in the shell thickness gene of oil palm (Nur et al., 2018 in
press).

8.18. GENOTYPIC SELECTION

Phenotypic selection of mutants is the cornerstone of a mutation breeding programme,

typically requiring extensive resources and time to select and progress mutant
populations up to the stage of advanced mutant lines or varietal release. Direct
(phenotypic) selection remains the main process by which desirable mutants are
selected and advanced. Indeed, it is a sensible way to proceed as ultimately it is the
traits that are important in developing superior varieties.

However, molecular screening for induced mutations in known target genes allows
genotypic selection of candidate mutants in a mutation breeding programme. Indeed,
Next Generation Sequencing (NGS) platforms have facilitated high-throughput
discovery of induced mutations in plant genomes (Tsai et al., 2011; Du et al., 2014;
Yuan et al., 2014; Li et al., 2016; Gupta et al., 2017; Datta et al., 2018). Importantly,
genotypic selection requires knowledge of the DNA sequence of the allele(s) that are
causative for the desired trait. The efficiency of traditional gene mapping and cloning

254
for marker development can be greatly improved using whole genome sequencing to
evaluate segregating populations (Abe et al., 2012; Schneeberger et al., 2009).
Alternatively, reverse genetic methods can be employed to create and identify
mutations in specific genes to assay gene function. Mutation detection in targeted
genes of large populations using complex DNA pools has been adopted for this
purpose. This method essentially combines chemical mutagenesis with high
throughput sequencing of DNA fragments amplified through PCR, or enriched via
probe capture assays (Tsai et al., 2011; Henry et al., 2014; Krasileva et al., 2017). In
this way, large mutant populations comprising thousands of individual plants can be
effectively screened for natural or induced sequence variations in a short time (see
Section 8.20.1).

Chemical mutagens have been most commonly used in reverse genetic studies
because they induce primarily single point mutations, and current technologies can be
easily adapted for their discovery (Jankowicz-Cieslak and Till, 2015b). Treatment of
plants with physical mutagens can result in more diverse spectra of lesions including
SNPs, small and large insertions and deletions and genomic rearrangements
(Yuan et al., 2014; Henry et al., 2015; Li et al., 2016a, 2016b; Datta et al., 2018). As
the cost of sequencing drops, whole genome sequencing approaches can be applied to
recover all types of induced lesions. It is likely that varying dosages will affect the
types of mutations that accumulate and with stable inheritance. Therefore, dosage
optimizations may be carried out to enrich the allele types desired for a specific
reverse genetics project.

As stated earlier, the prerequisite for genotypic selection is a full understanding of the
molecular and genetic architecture of the trait of interest. Candidate genes are
currently available for many traits of interest such as, for example, traits involved in
crop domestication. Still, more efforts are needed to link allelic diversity to
phenotypes in germplasm used in crop improvement programmes.

Importantly, new genotyping methods can be applied to select mutants at the early
stages of the mutation breeding programme, thus saving years of this process. As
described in Chapter 5, the M1 population consists of chimeric tissues. In case of
autogamous seed crops, this chimerism is largely resolved at the M2 stage after selfing
of the M1 plants. Therefore, genotypic selection is preferable, and may be performed
at the M2 stage where all mutations identified should be meiotically heritable.

It is also possible to select at the M1 stage. This approach of cataloguing potentially

interesting mutant alleles in the M1 is particularly relevant for perennial crops or trees
that have a long juvenile stage (in some cases 5 – 10 years). In addition, for large
perennial crops such as trees, space can be saved in growing out only the selected
(10 – 30) individual M1 plants and not the complete M1 population (normally about a

255
thousand individuals) as selection can be made from leaf samples taken from nursery
plants before transplanting to the field (Nur et al., 2018 in press). However, the risk
is that some mutations selected may not be heritable and thus may not be passed on
to the next generation, but at least some will be. Also, possibly some interesting
mutations may be missed when performing the selection at the M1 step.

Three major factors have contributed to genotyping as the choice method for mutation
detection: 1) the availability of sequence data of plants and crops species; 2) reverse
genetic studies that relate genes to functions; and, 3) a drastic reduction in cost of
genotyping assays. Together these methods provide sequences of target genes and
information on sequence variations produced through mutations (phenotypes of the
various mutant alleles). Selected individuals with sequence variation in the target
genes will be sorted out for the evaluation of expression of the traits. It is, however,
expected that not all the sequence variants will be expressed in the targeted trait.

Before deciding to set up a genotypic screen, practical concerns and comparative cost-
benefit analysis with direct (phenotypic) selection need to be evaluated. If a cost-
effective and efficient phenotypic screen is available that can test thousands of plants
in a mutant population, it is advisable to carry out phenotypic selection. This is
especially the case if there is no knowledge of the genes involved in the trait of
interest. In addition, genotypically selected mutants need to be tested in the field
eventually to confirm the capture of the desired phenotype and to assess field
performance. As knowledge about the genes underlying traits of interest grows, more
gains in streamlining and efficiency will be achieved through application of genotypic
selection.

8.19. ADDITIONAL APPLICATIONS OF DNA MARKERS FOR MUTATION

BREEDING

8.19.1. Excluding off-types using DNA markers

During the process of mutant population development, there is ample opportunity to

introduce off-types or contaminants from external sources such as inadvertent
outcrossing (see Chapter 5). Such contaminants may also result from contamination
of the seed stock that was used for mutation induction. Often a comparative analysis
between a mutant and its parent needs to be conducted. Such comparative analysis
may serve different purposes. For example, to identify causative mutations underlying
important traits, it may be necessary to conduct comparative genomics between the
mutant and its parent variety (Abe et al., 2012; Schneeberger et al., 2009). The
scientific rigour of such comparative studies relies entirely on the true origin of the
mutant and the parent. In this regard, standard DNA markers such as SSRs can be
used to ensure the mutant being studied is truly a direct mutant of the parental variety.
Indeed, Fu, Li and Shu, (2008) demonstrated the occurrence of outcross-derived

256
contaminants which were frequently selected as ‘mutants’ in several rice populations.
They further showed that such false mutants can be easily identified using SSR
markers. For example, in case of SSRs if a mutant shows a high degree of
polymorphism (e.g. >5%) compared to its parent, it may not be a true mutant and is
more likely a contaminant (Fu et al., 2007).

8.19.2. Background selection for clean-up of mutational load

Another area where molecular markers can have a major impact in plant mutation
breeding is in background selection, i.e. the selection and preservation of elite genetic
backgrounds of the parents with a minimal mutational load. In addition to marker-
assisted selection for a desired mutant gene (foreground), the genetic background can
be monitored and selected efficiently, thus reducing the rounds of selfing,
backcrossing or top-crossing. Molecular markers can also be used in genome scans to
select those individuals that contain both the introgressed allele and the greatest
proportion of the recurrent parent genome, thus combining foreground and
background selection in a single scan.

8.20. EXAMPLE PROTOCOLS

8.20.1. Identification of SNPs and indels in large mutant populations

using high-throughput amplicon sequencing applicable to diploid plant
species

The below protocol was adapted at the Joint FAO/IAEA Plant Breeding and Genetics
Laboratory, Seibersdorf, Austria using the Illumina MiSeq sequencing platform. The
protocol has been used in the context of TILLING experiments (Gupta et al., 2017)
to analyse M2 populations from mutated tomato and barley, and also to evaluate
natural nucleotide variation in cassava (Duitama et al., 2017). Mutant populations
were developed using chemical mutagenesis to enrich for single nucleotide mutations.
Screening assays are typically carried out on several hundred mutant lines
simultaneously. Guidelines and protocols for chemical mutagenesis are outlined in
Chapter 2 of this manual.

Note: Before proceeding with Next Generation Sequencing (NGS) approaches for
mutation identification in specific target genes, please review the newest kits available
on the market and carefully read the technical notes from the respective kits used at
different stages of this protocol. Additional protocols for the recovery of chemically
induced mutations by NGS can be found in (Burkart-Waco et al., 2017).

This protocol consists of four major steps outlined in Figure 8.18. The first step, DNA
extraction and polymerase chain reaction (PCR) amplification is the longest and
critical for the subsequent steps and for effective mutation detection. Step 2: Library

257
Preparation and Sequencing, is being conducted using commercial kits. This step can
take between 1 to 2 days plus the run time of the DNA sequencer. In Step 3: Variant
Calling, various platforms can be used for mutation detection in pooled samples
(Gupta et al., 2017). In the final Step 4, candidate mutations are validated using
Sanger sequencing (Sanger, 1981) and phenotyping methods.

Figure 8.18. Flowchart outlining the different steps for high-throughput detection of SNPs and small
Indels in large mutant populations of diploid plants.

258
Step 1 – DNA extraction and PCR amplification

Step 1.1 – DNA extraction, quantification, normalization and 3-dimensional pooling

Proceed with collection of tissue, e.g. leaf tissue from individual M2 plants. Care
should be taken to avoid cross-contamination and mislabelling. Several DNA
extraction methods are available. Test these with your selected crop. Genomic DNA
yield and stability at different storage temperatures should be established. We
recommend testing with approximately 10 primer pairs in PCR reactions to ensure
that the chosen DNA extraction method does not produce any contaminants that may
inhibit PCR. Usually the M2 population consists of a minimum of few hundreds up to
thousands of plants. Therefore, a kit such as Qiagen 96 where 96 samples can be
processed in parallel is recommended.

The DNA quality and quantity is very important to obtain high quality sequence data.
Accurate determination of the concentration of template DNA is essential to ensure
that all samples in a pool of genomic DNA are represented equally so that mutations
from every pool can be identified in the assay. To verify DNA quality and quantity
we suggest using agarose gel electrophoresis as it allows evaluation of DNA
degradation as well as determination of concentration. Free software can be used to
determine DNA concentration from gels, and tools have been developed to facilitate
the normalization of all DNA samples to a single concentration prior to pooling
(Huynh et al., 2017) . If choosing another method of DNA concentration
determination, such as spectroscopic approaches, we suggest testing several methods
side-by-side to ensure that the chosen method is both accurate and precise for the
measurement of intact genomic DNA. It is advisable to normalize DNAs to a
concentration higher than that used in PCR. This allows flexibility in experimental
design. Array DNA of the M2 population into 8 × 8 grid format plates (Till et al.,
2006). For DNA pooling and multiplexing follow the protocol described by Tsai et al.
(2011). This pooling strategy utilizes 12 plates of 8 × 8 grid format. This equates to
768 individuals being pooled for one MiSeq run with 44 sequencing libraries. There
are a range of different pooling strategies that can be applied. For example, higher-
fold pooling has been described that can increase throughput and reduce costs
(Duitama et al., 2017; Gupta et al., 2017; Pan et al., 2015) . Symmetrically pooled
samples may also be advantageous. However, we suggest starting with the 12-plate
approach to establish optimal parameters for your plant material prior to attempting
higher-fold pooling experiments.

Step 1.2 – PCR amplification of genes of interest, quantification, normalisation,

pooling and fragmentation

Design primers to amplify specific fragments. Note that direct sequencing of smaller
PCR fragments (500 – 600bp) is possible when using the MiSeq (Gupta et al., 2017;

259
Pan et al., 2015). However, throughput is increased by designing primers to amplify
1500 bp or more of the targeted genes followed by fragmentation of the PCR products
(Slota, Maluszynski and Szarejko, 2017). Carry out PCR using pooled DNA (in this
case, 44 pools) for every target separately. Quantify PCR amplicons using for example
agarose gel electrophoresis. Throughput can be increased using multi-channel
pipettors and 96-well format gels. Normalise all PCR products to the same
concentration. It is essential that the concentration of all amplified genes in all pools
is the same to ensure accurate mutation calling. Carefully pool amplicons of all gene
targets resulting in 44 pools: all amplicons produced from the same genomic DNA
pool are pooled.

Fragment every individual pool through sonication using e.g. the Covaris
ultrasonicator M220 (Covaris, Inc. USA). The following settings are recommended:
run time 30 sec; peak power 50; duty factor 40, cycles/burst 200. The fragmented PCR
products can be visualised via standard gel electrophoresis, or by using automated
equipment such as the Advanced Analytical Fragment AnalyzerTM (ThermoFisher
Scientific). The pools containing the fragmented PCR amplicons are then used for the
library preparation.

Step 2 – Library preparation and sequencing

Step 2.1 – NGS library preparation, quantification and preparation for MiSeq run

Proceed with library preparation. In this protocol indexed DNA library for NGS is
prepared using the TruSeq Nano DNA HT Library Preparation Kit (Illumina, Inc.)
with 200 ng of input PCR product.

Check the quality and quantity of prepared library. Library sizing can be performed
using gel electrophoresis or automated equipment such as the Fragment Analyzer with
High Sensitivity NGS Fragment Analysis Kit (1 – 6000bp) as illustrated in Figure
8.19.

260
Figure 8.19. Qualitative and quantitative visualisation of NGS libraries with the use of Fragment
Analyser and the High Sensitivity NGS Fragment Analysis Kit.

The concentration of each prepared library should be determined and samples

normalized to a common concentration prior to pooling. While DNA concentration
can be determined using qPCR or with the Fragment Analyzer, we typically use
Qubit® dsDNA HS Assay Kits (ThermoFisher Scientific). The concentration of each
library is converted from ng/μl into nM, based on the size of the DNA amplicons.

Dilute each library to a concentration higher than that recommended by the kit
protocol. For example, with the current sequencing chemistry a final concentration of
4 nM is required. In this case libraries are diluted to 6nM. This allows for variations
in the pipetting process without producing a library below the set concentration. Next,
pool the indexed and normalised libraries (same volume and same concentration) in a
single Eppendorf tube. This will result in all 44 libraries being pooled together. Re-
quantify using the Qubit® (ThermoFisher Scientific) and adjust the pooled library to
the required concentration. Follow the library denaturation and dilution guidelines for
the kit and sequencing chemistry used.

Step 2.2 – Sequencing of amplicons on a MiSeq

Proceed with sequencing of denatured and diluted libraries. Follow guidelines for
preparing the sequencing sample sheet. If using the Illumina MiSeq, a 2 × 300 PE
chemistry can be employed. Note that shorter read lengths are needed for higher-
throughput Illumina platforms. When running the MiSeq, reads can be de-multiplexed
automatically with FASTQ files produced for downstream analysis. Note that when
outsourcing sequencing some facilities will not provide FASTQ files but rather
unaligned BAM files. FASTQ files can be generated (e.g., using bamtofastq), taking

261
care of appropriate parameters such as paired end reads. Note also that some tools
such as GATK (The Broad Institute) require read-group information.

Step 3 – Variant calling

TILLING by Sequencing data produced by the Illumina machines can be processed

further with the use of numerous bioinformatic platforms. Prior to the analysis,
evaluate the run statistics on the MiSeq. Important parameters to evaluate the quality
of your run are cluster density and Phred base quality scores. Optimal values for both
parameters may vary depending on machine and sequencing chemistry used.

Note that at the time of writing, the MiSeq software can produce variant call files
(VCF) using GATK. However, it does not take into account the complex pooling
scheme incorporated into the project run. You will have to make new files with
software where the pooling strategy can be addressed.

Transfer fastq.gz files from the MiSeq sequencer (for a full description of file types,
see manufacturer’s website and
https://www.ncbi.nlm.nih.gov/sra/docs/submitformats/). Prepare a reference file (.fa).
You have the option of using a whole genome or simply amplicons. The easiest
method is to make one file with a list of all amplicons as this works with the multiple
pipelines and is required for CAMBa (http://web.cs.ucdavis.edu/~filkov/CAMBa/).
Process fastq.gz and reference (.fa) files following the below presented workflow
(Figure 8.20). Note that most software is run on the command line and detailed
instructions are available for each tool. If you have not yet established a computational
platform for data analysis, we suggest the Linux operating system owing to its ease of
use and the fact that most tools are compiled for this operating system.

Useful links: Many tools are available from the GitHub development platform
(e.g., https://github.com/lh3/bwa, https://github.com/broadinstitute/picard).

CAMBa and related tools can be found at

http://comailab.genomecenter.ucdavis.edu/index.php/TILLING_by_Sequencing and
under “The TILLING Pipeline” click Pipeline download.

262
Figure 8.20. Bioinformatic analysis utilising CAMBa and GATK pipelines. Figure adapted from Gupta
et al., 2017.

Step 4 – Molecular and phenotypic validation

Step 4.1 – Validation of mutations using Sanger sequencing

Candidate sequence variations are amplified using standard PCR techniques and
sequenced using standard Sanger DNA sequencing to verify the mutations (Sanger,
1981) .

Step 4.2 – Phenotypic evaluation of mutant plants

Individual plants selected with sequence variations in the target gene will need to be
phenotyped for the targeted trait as compared with the wild parent. Previous studies
suggest that approximately 10% of recovered chemically induced SNP mutations
identified in a gene-specific PCR amplicon may affect gene-function. Thus, the
number of putative mutants that show altered phenotypes in the laboratory and screen-
house and finally confirmed in the field is expected to be small.

263
 8.20.2. Genotyping for shell thickness selection in oil palm

Single nucleotide polymorphisms in the shell thickness (Sh) gene are responsible for
the difference between Dura (thick shell) and Pisifera (shell-less) fruit types
(Figures 8.17; 8.21) and these can be detected using High Resolution DNA Melt
(HRM) curves. Cleaved amplified polymorphism (CAP) analysis can be used to verify
the genotypes. The protocol below is that used by the Plant Genomics Laboratory at
Verdant Bioscience, Indonesia.

Figure 8.21. Single nucleotide polymorphism between Dura and Pisifera fruit types in the Sh gene.

Step 1 – Look for SNP variation in the shell thickness gene sequence in the oil palm
genome

The shell thickness gene sequence of oil palm can be found in the oil palm genome
GenBank (https://www.ncbi.nlm.nih.gov/) by typing in the keywords “elaeis
guineensis shell thickness” in the search box. This will show the complete oil palm
shell-specific gene information that has been deposited in GenBank. By performing a
multiple sequence alignment for all coding sequences of shell thickness the position
of single nucleotide polymorphisms (SNPs) can be found. SNPs which are responsible
for shell thickness phenotypes are identified and DNA primers are designed targeting
the SNPs (Figure 8.21).

Step 2 – Design primers for shell thickness alleles

A wide selection of web-based primer design tools can be applied to generate allele-
specific PCR primers. For example, Primer 3 [bioinfo.ut.ee/primer3-0.4.0/] is one
such tool. Primer dimers and hairpin loops of designed primers can be checked using
web-based tools such as OligoAnalyzer web tool from IDT SciTools (Owczarzy et al.,
2008) and OligoCalc (http://biotools.nubic.northwestern.edu).

264
Step 3 – Oil palm leaf sampling

Suitable material for the extraction of genomic DNA is young leaf tissue with a
sample size approximately 3cm2. Samples are typically taken from seedlings in the
nursery and placed in zip-sealed plastic bags and deep frozen in a -80°C freezer if not
used immediately.

Step 4 – DNA Extraction

The leaf samples are punched using a leaf puncher tool and the leaf discs (2 – 3 to
5mm diameter discs) are placed in tubes. Samples are homogenized in a tissue
disruptor. DNA extraction is carried out with a DNeasy Plant Mini Kit (Qiagen)
according to the manufacturer’s instructions. The quantity and quality of DNA can be
checked by spectrophotometry (e.g. NanoDrop). Purified genomic DNA is used as a
DNA template for genotyping.

Step 5 – Shell thickness detection using the HRM technique, validated by CAP
analysis

High Resolution Melt PCR can be used for shell thickness genotyping (Dura, Pisifera
and Tenera). Multiplex PCR can be performed which uses 3 pairs of allele-specific
primers. From HRM melt curves, three genotypes of shell thickness can be
distinguished (Figure 8.22). HRM requires real time PCR machines, such as the
Rotor-Gene Q and can be used for large sample sizes in high-throughput screening.

Figure 8.22. HRM curves of Dura, Pisifera and Tenera are easily distinguished (here using Tenera as
a base line).

265
If a SNP occurs within the recognition site of a restriction enzyme, it is much easier
to use cleaved amplified polymorphic sequences (CAPS). Shell thickness gene
sequences have a specific restriction site for HindIII, this restriction site allows CAPS
to distinguish shell alleles. Figure 8.23 shows a clear-cut differentiation of Dura,
Pisifera and Tenera genotypes using CAPS markers. CAPS analysis involves a simple
PCR protocol using conventional PCR equipment and banding patterns are visualized
in simple gel-based systems. However, CAPS analysis is slow compared to HRM and
not suited to high-throughput.

Figure 8.23. Banding patterns of CAPs in verifying Dura, Pisifera and Tenera fruit types.

In addition to the in-house protocols, such as that described above, Sh determinations

can be done using outside services such as Orion Biosains (www.orionbiosains.com).
Orion Biosains also provides services to screen for another important mutation in oil
palm, virescent fruit (Vir), which changes the fruit colour during ripening from black-
red to green-yellow (Singh et al., 2015b).

266
 REFERENCES

Abdulrazaq, A. & Ammar, K. 2015. Effect of the Chemical Mutagens Sodium Azide on
Plant Regeneration of Two Tomato Cultivars under Salinity Stress Condition in vitro. Journal
of Life Sciences, 9: 27–31.

Abe, A., Kosugi, S., Yoshida, K., Natsume, S., Takagi, H., Kanzaki, H., Matsumura, H.,
Mitsuoka, C., Tamiru, M., Innan, H., Cano, L., Kamoun, S. & Terauchi, R. 2012.
Genome sequencing reveals agronomically important loci in rice using MutMap. Nature
Biotechnology, 30: 174–8. https://doi.org/10.1038/nbt.2095 nbt.2095 [pii].

Abe, T., Kazama, Y., Ichida, H., Hayashi, Y., Ryuto, H. & Fukunishi, N. 2007. Plant
breeding using the ion beam irradiation in RIKEN. Cyclotrons and Their Applications: 222–
224.

$GOLHQơ '  $GO\Wơ 5 2017. Dosimetry principles, dose measurements and radiation
protection. Institute of Nuclear Chemistry and Technology: 55.

Ahloowalia, B. 1998. In-vitro techniques and mutagenesis for the improvement of

vegetatively propagated plants. Somaclonal variation and induced mutations in crop
improvement, pp. 293–309. Springer.

Ahloowalia, B. & Maluszynski, M. 2001. Induced mutations–A new paradigm in plant

breeding. Euphytica, 118(2): 167–173.

Ahloowalia, B., Maluszynski, M. & Nichterlein, K. 2004. Global impact of mutation-

derived varieties. Euphytica, 135(2): 187–204.

Akerberg, E. & Hagberg, A. 1963. Plant Breeding. (Book Reviews: Recent Plant Breeding
Research: Svalof, 1946-1961). Science, 142: 1451.

Amaldi, U. 2000. The importance of particle accelerators. europhysics news, 31(6): 5–9.

Amin, R., Laskar, R.A. & Khan, S. 2015a. Assessment of genetic response and character
association for yield and yield components in Lentil (Lens culinaris L.) population developed
through chemical mutagenesis. Cogent Food & Agriculture, 1(1): 1000715.

Andersen, J.R. & Lübberstedt, T. 2003. Functional markers in plants. Trends in plant
science, 8(11): 554–560.

Ashikari, M. & Matsuoka, M. 2006. Identification, isolation and pyramiding of quantitative

trait loci for rice breeding. Trends in plant science, 11(7): 344–350.

267
Auerbach, C. 1946. XXV.—Chemically Induced Mosaicism in Drosophila melanogaster.
Proceedings of the Royal Society of Edinburgh, Section B: Biological Sciences, 62(2): 211–
222.

Auerbach, C. & Robson, J. 1946. Chemical production of mutations. Nature, 157(3984):

302.

Badigannavar, A., Kumar, A.A., Girish, G. & Ganapathi, T. 2017. Characterization of

Post-Rainy Season Grown Indigenous and Exotic Germplasm Lines of Sorghum for
Morphological and Yield Traits. Plant Breeding and Biotechnology, 5(2): 106–114.

Bado, S., Forster, B.P., Ghanim, A.M., Jankowicz-Cieslak, J., Berthold, G. & Luxiang,
L. 2016. Protocols for Pre-Field Screening of Mutants for Salt Tolerance in Rice, Wheat and
Barley. Springer.

Bado, S., Forster, B.P., Nielen, S., Ghanim, A., Lagoda, P.J., Till, B.J. & Laimer, M.
2015. Plant mutation breeding: current progress and future assessment. Plant breeding
reviews, 39: 23–88.

Badr, A., El-Shazly, H.H. & Halawa, M. 2014. Cytological Effects of Gamma Radiation
and Its Impact on Growth and Yield of M1 and M2 Plants of Cowpea Cultivars. Cytologia,
79(2): 195–206.

Bakri, A., Heather, N., Hendrichs, J. & Ferris, I. 2005. Fifty years of radiation biology in
entomology: lessons learned from IDIDAS. Annals of the Entomological Society of America,
98(1): 1–12.

Barah, P. & Bones, A.M. 2015. Multidimensional approaches for studying plant defence
against insects: from ecology to omics and synthetic biology. Journal of experimental botany,
66(2): 479–493.

Barba, M., Cupidi, A., Loreti, S., Faggioli, F. & Martino, L. 1994. In Vitro Micrografting:
A Tecnique to Eliminate Peach Latent Mosaic Viroid from Peach. XVI International
Symposium on Fruit Tree Virus diseases 386. pp. 531–535. Paper presented at, 1994.

Bennetzen, J.L. & Hake, S.C. 2009. Handbook of maize: genetics and genomics. Springer.

Bhagwat, B. & Duncan, E. 1998. Mutation breeding of banana cv. Highgate (Musa spp.,
AAA Group) for tolerance to Fusarium oxysporum f. sp. cubense using chemical mutagens.
Scientia horticulturae, 73(1): 11–22.

Bharathi Veeramani, B., Prakash, M., Jagadeesan, S., Kavimani, S., Saravanan, K. &
Ganesan, J. 2005. Variabiuty studies in M2 generation of groundnut (Arachis hypogaea L.)
var. VRI 2.

268
Bhat, I.A., Pandit, U.J., Sheikh, I.A. & Hassan, Z.U. 2017. Physical and Chemical
Mutagenesis in Linum usitatissimum L. to Induce Variability in Seed Germination, Survival
and Growth rate traits. Current Botany, 7: 28–32.

Bhojwani, S. & Razdan, M. 1983. Developments in Group Science 5. Elsevier, Amsterdam.

Bjerg, B., Sørensen, H. & Wathelet, J. 1987. Glucosinolates in Rapeseeds: Analytical

Aspects. World Crops: Production, Utilization, Description, pp. 125–150. Martinus Nijhoff,
Kluwer Academic, Dordrecht/Boston/Lancaster.

Blakeslee, A.F., Belling, J., Farnham, M.E. & Bergner, A.D. 1922. A haploid mutant in
the Jimson Weed," Datura Stramonium". Science, 55(1433): 646–647.

Boersma, J.G. 2007. Contributions to the Molecular Genetics of the Narrow-leaf Lupin
(Lupinus Augustifolius L.): Mapping, Marker Development and QTL Analysis. University of
Western Australia.

Boudou, C., Biston, M.-C., Corde, S., Adam, J.-F., Ferrero, C., Estève, F. & Elleaume,
H. 2004. Synchrotron stereotactic radiotherapy: dosimetry by Fricke gel and Monte Carlo
simulations. Physics in medicine and biology, 49(22): 5135.

Boureima, S., Oukarroum, A., Diouf, M., Cisse, N. & Van Damme, P. 2012. Screening
for drought tolerance in mutant germplasm of sesame (Sesamum indicum) probing by
chlorophyll a fluorescence. Environmental and Experimental Botany, 81: 37–43.

Boyd, L.A., Ridout, C., O’Sullivan, D.M., Leach, J.E. & Leung, H. 2013. Plant–pathogen
interactions: disease resistance in modern agriculture. Trends in genetics, 29(4): 233–240.

Bradshaw, J.E. 2016. Mutation Breeding. Plant Breeding: Past, Present and Future, pp.
529–560. Springer.

Bradshaw, V.A. & McEntee, K. 1989. DNA damage activates transcription and
transposition of yeast Ty retrotransposons. Molecular and General Genetics MGG, 218(3):
465–474.

Brock, R., Andrew, W., Kirchner, R. & Crawford, E. 1971. Early-flowering mutants of
Medicago ploymorpha var. Polymorpha. Crop and Pasture Science, 22(2): 215–222.

Broertjes, C. 2012. Application of mutation breeding methods in the improvement of

vegetatively propagated crops. Elsevier.

Broertjes, C. & van Harten, A. 1987. Application of mutation breeding methods to

vegetatively propagated plants. Improving vegetatively propagated crops, pp. 335–348.
Academic Press.

269
Bromke, M.A., Hochmuth, A., Tohge, T., Fernie, A.R., Giavalisco, P., Burgos, A.,
Willmitzer, L. & Brotman, Y. 2015. Liquid chromatography high-resolution mass
spectrometry for fatty acid profiling. The Plant Journal, 81(3): 529–536.

Brunner, H. 1995. Radiation induced mutations for plant selection. Applied radiation and
isotopes, 46(6–7): 589–594.

Bui, Q.T. & Grandbastien, M.-A. 2012. LTR retrotransposons as controlling elements of
genome response to stress? Plant transposable elements, pp. 273–296. Springer.

Burkart-Waco, D., Tsai, H., Ngo, K., Henry, I.M., Comai, L. & Tai, T.H. 2017. Next-
Generation Sequencing for Targeted Discovery of Rare Mutations in Rice. Biotechnologies
for Plant Mutation Breeding, pp. 323–340. Springer, Cham. (also available at
https://link.springer.com/chapter/10.1007/978-3-319-45021-6_20).

Burke, M.B., Lobell, D.B. & Guarino, L. 2009. Shifts in African crop climates by 2050,
and the implications for crop improvement and genetic resources conservation. Global
Environmental Change, 19(3): 317–325.

Butelli, E., Licciardello, C., Zhang, Y., Liu, J., Mackay, S., Bailey, P., Reforgiato-
Recupero, G. & Martin, C. 2012. Retrotransposons control fruit-specific, cold-dependent
accumulation of anthocyanins in blood oranges. The Plant Cell, 24(3): 1242–1255.

Byrne, J. 2013. Neutrons, nuclei and matter: an exploration of the physics of slow neutrons.
Courier Corporation.

Caldwell, D.G., McCallum, N., Shaw, P., Muehlbauer, G.J., Marshall, D.F. & Waugh,
R. 2004a. A structured mutant population for forward and reverse genetics in Barley
(Hordeum vulgare L.). The Plant Journal, 40(1): 143–150.

Castronuovo, D., Tataranni, G., Lovelli, S., Candido, V., Sofo, A. & Scopa, A. 2014. UV-
C irradiation effects on young tomato plants: Preliminary results. Pak. J. Bot, 46(3): 945–949.

Chaudhury, A.M. 1993. Nuclear genes controlling male fertility. The Plant Cell, 5(10):
1277.

Chawade, A., Sikora, P., Bräutigam, M., Larsson, M., Vivekanand, V., Nakash, M.A.,
Chen, T. & Olsson, O. 2010. Development and characterization of an oat TILLING-
SRSXODWLRQDQGLGHQWLILFDWLRQRIPXWDWLRQVLQOLJQLQDQGȕ-glucan biosynthesis genes. BMC
Plant Biology, 10(1): 86.

Cheema, A.A. & Atta, B.M. 2003. Radiosensitivity studies in basmati rice. Pak. J. Bot,
35(2): 197–207.

270
Chen, J.-F., Cui, L., Malik, A.A. & Mbira, K.G. 2011. In vitro haploid and dihaploid
production via unfertilized ovule culture. Plant Cell, Tissue and Organ Culture (PCTOC),
104(3): 311–319.

Chen, P., Liu, W., Yuan, J., Wang, X., Zhou, B., Wang, S., Zhang, S., Feng, Y., Yang,
B., Liu, G. & others. 2005. Development and characterization of wheat-Leymus racemosus
translocation lines with resistance to Fusarium Head Blight. Theoretical and applied genetics,
111(5): 941–948.

Chen, P., You, C., Hu, Y., Chen, S., Zhou, B., Cao, A. & Wang, X. 2013. Radiation-
induced translocations with reduced Haynaldia villosa chromatin at the Pm21 locus for
powdery mildew resistance in wheat. Molecular Breeding, 31(2): 477–484.
https://doi.org/10.1007/s11032-012-9804-x.

Cheng, J. 2014. A mutation breeding program to improve the quality of the oil crop Crambe
abyssinica. Wageningen University.

Chopra, V. 2005. Mutagenesis: Investigating the process and. Current Science, 89(2).

Chu, C.-G., Tan, C., Yu, G.-T., Zhong, S., Xu, S. & Yan, L. 2011. A novel retrotransposon
inserted in the dominant Vrn-B1 allele confers spring growth habit in tetraploid wheat
(Triticum turgidum L.). G3: Genes, Genomes, Genetics, 1(7): 637–645.

Chu, I., Kurtz, S. & others. 1990. Commercialization of plant micropropagation. Handbook
of plant cell culture. Volume 5. Ornamental species.: 126–164.

Close, T.J., Bhat, P.R., Lonardi, S., Wu, Y., Rostoks, N., Ramsay, L., Druka, A., Stein,
N., Svensson, J.T., Wanamaker, S. & others. 2009. Development and implementation of
high-throughput SNP genotyping in barley. BMC genomics, 10(1): 582.

Cooper, J.L., Till, B.J., Laport, R.G., Darlow, M.C., Kleffner, J.M., Jamai, A., El-
Mellouki, T., Liu, S., Ritchie, R., Nielsen, N. & others. 2008. TILLING to detect induced
mutations in soybean. BMC plant biology, 8(1): 9.

Cox, E.C. 1972. On the organization of higher chromosomes. Nature, 239(92): 133–134.

Cros, D., Denis, M., Sánchez, L., Cochard, B., Flori, A., Durand-Gasselin, T., Nouy, B.,
Omoré, A., Pomiès, V., Riou, V. & others. 2015. Genomic selection prediction accuracy in
a perennial crop: case study of oil palm (Elaeis guineensis Jacq.). Theoretical and Applied
Genetics, 128(3): 397–410.

Croteau, D. & Bohr, V.A. 2013. Overview of DNA Repair Pathways. DNA Repair and
Cancer: From Bench to Clinic, pp. 1–24. CRC Press.

271
Cruz, R.P. da, Sperotto, R.A., Cargnelutti, D., Adamski, J.M., FreitasTerra, T. & Fett,
J.P. 2013. Avoiding damage and achieving cold tolerance in rice plants. Food and Energy
Security, 2(2): 96–119.

Dahmani-Mardas, F., Troadec, C., Boualem, A., Leveˆque, S., Alsadon, A.A., Aldoss,
A.A., Dogimont, C. & Bendahmane, A. 2010. Engineering melon plants with improved fruit
shelf life using the TILLING approach. PloS one, 5(12): e15776.

Danso, K., Safo-Katanka, O., Adu-Ampomah, Y., Oduro, V., Amoatey, H., Asare, D.,
Ayeh, E.O., Amenorpe, G., Amiteye, S. & Lokko, Y. 1990a. Application of induced
mutation techniques in Ghana: Impact, challenges and the future. collections: 270.

Das, G., Patra, J.K. & Baek, K.-H. 2017. Corrigendum: Insight into MAS: A Molecular
Tool for Development of Stress Resistant and Quality of Rice through Gene Stacking.
Frontiers in plant science, 8: 1321.

Datta, S. 2014. Induced mutagenesis: basic knowledge for technological success. Exploring
genetic diversity of crops: 97.

Datta, S., Jankowicz-Cieslak, J., Nielen, S., Ingelbrecht, I. & Till, B.J. 2018. Induction
and recovery of copy number variation in banana through gamma irradiation and low
coverage whole genome sequencing. Plant Biotechnology Journal.
https://doi.org/10.1111/pbi.12901.

De Mey, M., Lequeux, G.J., Maertens, J., De Muynck, C.I., Soetaert, W.K. &
Vandamme, E.J. 2008. Comparison of protein quantification and extraction methods suitable
for E. coli cultures. Biologicals, 36(3): 198–202.

De Micco, V., Arena, C., Pignalosa, D. & Durante, M. 2011. Effects of sparsely and densely
ionizing radiation on plants. Radiation and Environmental Biophysics, 50(1): 1–19.

Delgado-Paredes, G.E., Rojas-Idrogo, C., Chanamé-Céspedes, J., Floh, E.I. & Handro,
W. 2017. Development and Agronomic Evaluation of In Vitro Somaclonal Variation in Sweet
Potato Regenerated Plants from Direct Organogenesis of Roots. Asian Journal of Plant
Science and Research, 7(1): 39–48.

Delhaize, E., Ma, J.F. & Ryan, P.R. 2012. Transcriptional regulation of aluminium tolerance
genes. Trends in plant science, 17(6): 341–348.

Devaux, P. & Pickering, R. 2005. Haploids in the improvement of Poaceae. Haploids in

crop improvement II, pp. 215–242. Springer.

Díaz, A., Zikhali, M., Turner, A.S., Isaac, P. & Laurie, D.A. 2012. Copy number variation
affecting the Photoperiod-B1 and Vernalization-A1 genes is associated with altered flowering
time in wheat (Triticum aestivum). PLoS One, 7(3): e33234.

272
Djilianov, D., Prinsen, E., Oden, S., van Onckelen, H. & Müller, J. 2003. Nodulation under
salt stress of alfalfa lines obtained after in vitro selection for osmotic tolerance. Plant Science,
165(4): 887–894.

Do, K. & others. 2009. Socio-economic impacts of mutant rice varieties in Southern
Vietnam. Induced plant mutations in the genomics era. FAO, Rome: 65–68.

Dong, C., Dalton-Morgan, J., Vincent, K. & Sharp, P. 2009. A modified TILLING method
for wheat breeding. The Plant Genome, 2(1): 39–47.

Dong, X., Li, W., Liu, R. & Gu, W. 2017. Brief Overview of Sweet Sorghum Irradiated by
Carbon Ion Beam. Journal of Agricultural Science, 9(8): 74.

Donini, P. & Sonnino, A. 1998. Induced mutation in plant breeding: current status and future
outlook. Somaclonal variation and induced mutations in crop improvement, pp. 255–291.
Springer.

'UDJDQLü, *XSWD% 1973. Dosimetry in Agriculture, Industry, Biology and Medicine.

IAEA-SM-160/93, p. 351. International Atomic Energy Agency Vienna.

Drake, J.W., Charlesworth, B., Charlesworth, D. & Crow, J.F. 1998. Rates of
spontaneous mutation. Genetics, 148(4): 1667–1686.

Du, Y., Li, W., Yu, L., Chen, G., Liu, Q., Luo, S., Shu, Q. & Zhou, L. 2014. Mutagenic
effects of carbon-ion irradiation on dry Arabidopsis thaliana seeds. Mutat Res, 759: 28–36.
https://doi.org/10.1016/j.mrgentox.2013.07.018.

Duclercq, J., Sangwan-Norreel, B., Catterou, M. & Sangwan, R.S. 2011. De novo shoot
organogenesis: from art to science. Trends in plant science, 16(11): 597–606.

Duitama, J., Kafuri, L., Tello, D., Leiva, A.M., Hofinger, B., Datta, S., Lentini, Z.,
Aranzales, E., Till, B. & Ceballos, H. 2017. Deep assessment of genomic diversity in
cassava for herbicide tolerance and starch biosynthesis. Computational and structural
biotechnology journal, 15: 185–194.

'XPOXSLQDU=.DUDNX]XOX+'HPLUWDú0%8÷XUHU, M., Gezginç, H., Dokuyucu,

T. & Akkaya, A. 2015. A Heterosis Study for Some Agronomic Traits in Oat. 7DUÕP%LOLPOHUL
Dergisi, 21(3): 414–419.

Dunwell, J.M. 2010. Haploids in flowering plants: origins and exploitation. Plant
Biotechnology Journal, 8(4): 377–424.

Dunwell, J.M., Wilkinson, M.J., Nelson, S., Wening, S., Sitorus, A.C., Mienanti, D.,
Alfiko, Y., Croxford, A.E., Ford, C.S., Forster, B.P. & others. 2010. Production of
haploids and doubled haploids in oil palm. BMC plant biology, 10(1): 218.

273
Eduardo deAlmeida, C., Ochoa, R., de Lima, M.C., David, M.G., Pires, E.J., Peixoto,
J.G., Salata, C. & Bernal, M.A. 2014. A Feasibility Study of Fricke Dosimetry as an
Absorbed Dose to Water Standard for 192Ir HDR Sources. PloS one, 9(12): e115155.

Ehrenberg, L., Lundqvist, U. & Ström, G. 1958. The mutagenic action of ethylene imine
in barley. Hereditas, 44(2–3): 330–336.

Ekberg, I. 1974. Cytogenetic studies of three paracentric inversions in barley. Hereditas,

76(1): 1–30.

Elkonin, L. & Tsvetova, M. 2008. Genetic and cytological analyses of the male sterility
mutation induced in a sorghum tissue culture with streptomycin. Russian Journal of Genetics,
44(5): 575–583.

Epstein, L., Kaur, S., Chang, P.L., Carrasquilla-Garcia, N., Lyu, G., Cook, D.R.,
Subbarao, K.V. & O’Donnell, K. 2017. Races of the Celery Pathogen Fusarium oxysporum
f. sp. apii Are Polyphyletic. Phytopathology, 107(4): 463–473.

Farkash, E.A., Kao, G.D., Horman, S.R. & Prak, E.T.L. 2006. Gamma radiation increases
endonuclease-dependent L1 retrotransposition in a cultured cell assay. Nucleic acids
research, 34(4): 1196–1204.

Forney, C.F. & Brandl, D.G. 1992. Control of humidity in small controlled-environment
chambers using glycerol-water solutions. HortTechnology, 2(1): 52–54.

Forster, B., Franckowiak, J., Lundqvist, U., Thomas, W., Leader, D., Shaw, P., Lyon, J.
& Waugh, R. 2012. Mutant phenotyping and pre-breeding in barley. Plant mutation breeding
and biotechnology. CABI, Wallingford: 327–346.

Forster, B.P., Franckowiak, J.D., Lundqvist, U., Lyon, J., Pitkethly, I. & Thomas, W.T.
2007a. The barley phytomer. Annals of botany, 100(4): 725–733.

Forster, B.P., Heberle-Bors, E., Kasha, K.J. & Touraev, A. 2007b. The resurgence of
haploids in higher plants. Trends in plant science, 12(8): 368–375.

Franklin, K.A. & Quail, P.H. 2010. Phytochrome functions in Arabidopsis development.
Journal of experimental botany, 61(1): 11–24.

Fricke, H. & Hart, E. 1966. Radiation dosimetry. Radiation Dosimetry. Academic Press,
New York: 167.

Friebe, B., Hatchett, J., Gill, B., Mukai, Y. & Sebesta, E. 1991. Transfer of Hessian fly
resistance from rye to wheat via radiation-induced terminal and intercalary chromosomal
translocations. TAG Theoretical and Applied Genetics, 83(1): 33–40.

274
Frydenberg, O. & Jacobsen, P. 1966. The mutation and the segregation frequencies in
different spike categories after chemical treatment of barley seeds. Hereditas, 55(2–3): 227–
248.

Fu, H.-W., Li, Y.-F. & Shu, Q.-Y. 2008. A revisit of mutation induction by gamma rays in
rice (Oryza sativa L.): implications of microsatellite markers for quality control. Molecular
Breeding, 22(2): 281–288.

Fu, H.W., Wang, C.X., Shu, X.L., Li, Y.F., Wu, D.X. & Shu, Q.Y. 2007. Microsatellite
analysis for revealing parentage of gamma ray-induced mutants in rice (Oryza sativa L.).
Israel journal of plant sciences, 55(2): 201–206.

Fujita, D., Kohli, A. & Horgan, F.G. 2013. Rice resistance to planthoppers and leafhoppers.
Critical reviews in plant sciences, 32(3): 162–191.

Gaul, H. 1959. A critical survey of genome analysis. Proc 1st Int Wheat Genet Symp.
University of Winnipeg, Winnipeg, Canada. pp. 194–206. Paper presented at, 1959.

Gautam, S., Saxena, S. & Kumar, S. 2016. Fruits and Vegetables as Dietary Sources of
Antimutagens. J. Food Chem. Nanotechnol., 2: 97–114.

Geier, T. 2012. Chimeras: properties and dissociation in vegetatively propagated plants.

Plant Mutation Breeding and Biotechnology. CABI, Wallingford, UK: 191–201.

Germana, M. 2012. Use of irradiated pollen to induce parthenogenesis and haploid

production in fruit crops. Plant mutation breeding and biotechnology. CABI, FAO,
Oxfordshire, UK: 411–421.

*LFKQHU73WiþHN26WDYUHYD'$:DJQHU(' 3OHZD0- 2000. A comparison

of DNA repair using the comet assay in tobacco seedlings after exposure to alkylating agents
or ionizing radiation. Mutation Research/Genetic Toxicology and Environmental
Mutagenesis, 470(1): 1–9.

Gill, S.S., Anjum, N.A., Gill, R., Jha, M. & Tuteja, N. 2015. DNA damage and repair in
plants under ultraviolet and ionizing radiations. The Scientific World Journal, 2015.

Gómez-Pando, L., Eguiluz, A., Jimenez, J., Falconí, J. & Aguilar, E.H. 2009. Barley
(Hordeun vulgare) and kiwicha (Amaranthus caudatus) improvement by mutation induction
in Peru. Induced Plant Mutations in the Genomics Era, Food and Agriculture Organization
of the United Nations, Rome: 371–374.

Gottschalk, W. & Wolff, G. 2012. Induced mutations in plant breeding. Springer Science &
Business Media.

275
Gottwald, S., Bauer, P., Komatsuda, T., Lundqvist, U. & Stein, N. 2009. TILLING in the
two-rowed barley cultivar’Barke’reveals preferred sites of functional diversity in the gene
HvHox1. BMC research notes, 2(1): 258.

Grafi, G. 2004. How cells dedifferentiate: a lesson from plants. Developmental biology,
268(1): 1–6.

Greene, E.A., Codomo, C.A., Taylor, N.E., Henikoff, J.G., Till, B.J., Reynolds, S.H.,
Enns, L.C., Burtner, C., Johnson, J.E., Odden, A.R. & others. 2003. Spectrum of
chemically induced mutations from a large-scale reverse-genetic screen in Arabidopsis.
Genetics, 164(2): 731–740.

Gregory, W. & others. 1960. The peanut NC 4x, a milestone in crop breeding. Crops and
soils, 12(8): 12–13.

Greiner, S., Sobanski, J. & Bock, R. 2015. Why are most organelle genomes transmitted
maternally? BioEssays, 37(1): 80–94.

Griffiths, A.J., Miller, J.H., Suzuki, D., Lewontin, R.C. & Gelbart, W.M. undated. An
Introduction to genetic analysis. 2000. WH Freeman.

Gruszka, D., Szarejko, I. & Maluszynski, M. 2012. Sodium azide as a mutagen. Plant
mutation breeding and biotechnology. CABI, Wallingford: 159–166.

Guha, S. & Maheshwari, S. 1964. In vitro production of embryos from anthers of Datura.
Nature, 204: 497.

Gupta, P. 1998. Mutation breeding in cereals and legumes. Somaclonal Variation and
Induced Mutations in Crop Improvement, pp. 311–332. Springer.

Gupta, P., Reddaiah, B., Salava, H., Upadhyaya, P., Tyagi, K., Sarma, S., Datta, S.,
Malhotra, B., Thomas, S., Sunkum, A., Devulapalli, S., Till, B.J., Sreelakshmi, Y. &
Sharma, R. 2017. Next-generation sequencing (NGS)-based identification of induced
mutations in a doubly mutagenized tomato (Solanum lycopersicum) population. The Plant
Journal: For Cell and Molecular Biology, 92(3): 495–508. https://doi.org/10.1111/tpj.13654

Halliwell, B. & Gutteridge, J.M. 2015. Free radicals in biology and medicine. Oxford
University Press, USA.

Hamill, S., Smith, M. & Dodd, W. 1992. In vitro induction of banana autotetraploids by
colchicine treatment of micropropagated diploids. Australian Journal of Botany, 40(6): 887–
896.

Hancock, J.F. 2012. Plant evolution and the origin of crop species. CABI.

Handro, W. 2014. Mutagenesis and in vitro selection. MPKV; Maharastra.

276
Haninger, T. & Henniger, J. 2016. Dosimetric properties of the new TLD albedo neutron
dosemeter AWST-TL-GD 04. Radiation Protection Dosimetry, 170(1–4): 150–153.

Hansel, H., Simon, W. & Ehrendorfer, K. 1972. Mutation breeding for yield and kernel
performance in spring barley. Induced Mutations and Plant Improvement: 221–235.

Hao, M., Luo, J., Zhang, L., Yuan, Z., Yang, Y., Wu, M., Chen, W., Zheng, Y., Zhang,
H. & Liu, D. 2013. Production of hexaploid triticale by a synthetic hexaploid wheat-rye
hybrid method. Euphytica, 193(3): 347–357.

Harrison, L. 2013. Radiation induced DNA damage, repair and therapeutics. DNA Repair
and Cancer: 92–136.

van Harten, A.M. 1998. Mutation breeding: theory and practical applications. Cambridge
University Press.

Hayashi, K. & Yoshida, H. 2009. Refunctionalization of the ancient rice blast disease
resistance gene Pit by the recruitment of a retrotransposon as a promoter. The Plant Journal,
57(3): 413–425.

Heffner, E.L., Lorenz, A.J., Jannink, J.-L. & Sorrells, M.E. 2010. Plant breeding with
genomic selection: gain per unit time and cost. Crop science, 50(5): 1681–1690.

Henry, I.M., Nagalakshmi, U., Lieberman, M.C., Ngo, K.J., Krasileva, K.V., Vasquez-
Gross, H., Akhunova, A., Akhunov, E., Dubcovsky, J., Tai, T.H. & Comai, L. 2014.
Efficient Genome-Wide Detection and Cataloging of EMS-Induced Mutations Using Exome
Capture and Next-Generation Sequencing. Plant Cell.
https://doi.org/10.1105/tpc.113.121590.

Henry, I.M., Zinkgraf, M.S., Groover, A.T. & Comai, L. 2015. A System for Dosage-
Based Functional Genomics in Poplar. The Plant Cell Online: tpc.15.00349.
https://doi.org/10.1105/tpc.15.00349.

Hertel, T.W. & Lobell, D.B. 2014. Agricultural adaptation to climate change in rich and poor
countries: Current modeling practice and potential for empirical contributions. Energy
Economics, 46: 562–575.

Heslop-Harrison, J.S. & Schwarzacher, T. 2007. Domestication, genomics and the future
for banana. Annals of botany, 100(5): 1073–1084.

Hiroshi, K. 2008. Development of rice varieties for whole crop silage (WCS) in Japan. Japan
Agricultural Research Quarterly: JARQ, 42(4): 231–236.

Hoffmann, A.A. & Rieseberg, L.H. 2008. Revisiting the impact of inversions in evolution:
from population genetic markers to drivers of adaptive shifts and speciation? Annual review
of ecology, evolution, and systematics, 39: 21–42.

277
Howard III, T.P., Hayward, A.P., Tordillos, A., Fragoso, C., Moreno, M.A., Tohme, J.,
Kausch, A.P., Mottinger, J.P. & Dellaporta, S.L. 2014. Identification of the maize
gravitropism gene lazy plant1 by a transposon-tagging genome resequencing strategy. PloS
one, 9(1): e87053.

Htun, K.W., Min, M. & Win, N.C. undated. Evaluation of Genetic Variability For
Agronomic Traits In M2 Generation of Sorghum Through Induced Mutation. International
Journal of Technical Research and Applications.3 (6): 145 – 149.

Hu, J. & Rutger, J. 1991. A streptomycin induced no-pollen male sterile mutant in rice
(Oryza sativa L.). J. Gen. Breed, 45: 349–352.

Huiru, Z., Yunhong, G., Huiyuan, Y., Zhen, J. & Guangyong, Q. 2009. Differential
Expression of Retrotransposon WIS 2–1A Response to Vacuum, Low-Energy N+
,PSODQWDWLRQ DQG &RȖ-ray Irradiation in Wheat. Plasma Science and Technology, 11(1):
104.

Huynh, O.A., Jankowicz-Cieslak, J., Saraye, B., Hofinger, B. & Till, B.J. 2017. Low-cost
methods for DNA extraction and quantification. Biotechnologies for Plant Mutation
Breeding, pp. 227–239. Springer.

ICRU, F.Q. 1998. Units for Ionizing Radiation. ICRU Report, 60.

ICRU, Q. 1993. Units in Radiation Protection Dosimetry. ICRU Report, 51.

Jain, S. 2008. In vitro mutagenesis in banana (Musa spp.) improvement. IV International

Symposium on Banana: International Conference on Banana and Plantain in Africa:
Harnessing International 879. pp. 605–614. Paper presented at, 2008.

Jain, S. & Spencer, M. 2006. Biotechnology and mutagenesis in improving ornamental

plants. Floriculture and ornamental biotechnology: advances and tropical issues: 1749–
2036.

Jain, S.M. & others. 2010. Mutagenesis in crop improvement under the climate change.
Romanian biotechnological letters, 15(2): 88–106.

Jambhulkar, S.J. 2015. 4 Induced Mutagenesis and Allele Mining. Brassica Oilseeds:
Breeding and Management: 53.

Jan, C. & Vick, B. 2006. Registration of seven cytoplasmic male-sterile and four fertility
restoration sunflower germplasms. Crop science, 46(4): 1829–a.

Jan, S., Parween, T. & Siddiqi, T. 2012. Effect of gamma radiation on morphological,
biochemical, and physiological aspects of plants and plant products. Environmental Reviews,
20(1): 17–39.

278
Jankowicz-Cieslak, J., Huynh, O.A., Brozynska, M., Nakitandwe, J. & Till, B.J. 2012.
Induction, rapid fixation and retention of mutations in vegetatively propagated banana. Plant
biotechnology journal, 10(9): 1056–1066.

Jankowicz-Cieslak, J., Kozak-Stankiewicz, K., Seballos, G., Razafinirina, L.,

Rabefiraisana, J., Rakotoarisoa, N., Forster, B., Vollmann, J. & Till, B. 2013. Application
of soft X-ray and near-infrared reflectance spectroscopy for rapid phenotyping of mutant rice
seed. Proceedings transformation technologies (plants and animals), Plant Gen. Breed.
Technol. Vienna: 37–40.

Jankowicz-Cieslak, J., Mba, C. & Till, B.J. 2017. Mutagenesis for Crop Breeding and
Functional Genomics. Biotechnologies for Plant Mutation Breeding, pp. 3–18. Springer.

Jankowicz-Cieslak, J. & Till, B.J. 2015. Forward and reverse genetics in crop breeding.
Advances in plant breeding strategies: breeding, biotechnology and molecular tools, pp. 215–
240. Springer.

Jankowicz-Cieslak, J. & Till, B.J. 2016. Chemical mutagenesis of seed and vegetatively
propagated plants using EMS. Current Protocols in Plant Biology: 617–635.

Jankowicz-Cieslak, J. & Till, B.J. 2017. Chemical Mutagenesis and Chimera Dissolution in
Vegetatively Propagated Banana. Biotechnologies for Plant Mutation Breeding, pp. 39–54.
Springer.

Jannink, J.-L., Lorenz, A.J. & Iwata, H. 2010. Genomic selection in plant breeding: from
theory to practice. Briefings in functional genomics, 9(2): 166–177.

Jardim, S.S., Schuch, A.P., Pereira, C.M. & Loreto, E.L.S. 2015. Effects of heat and UV
radiation on the mobilization of transposon mariner-Mos1. Cell Stress and Chaperones,
20(5): 843–851.

Jayaramachandran, M., Kumaravadivel, N., Eapen, S. & Kandasamy, G. 2010. Gene

Action for yield attributing characters in segregating generation (M2) of Sorghum (Sorghum
bicolor L.). Electronic Journal of Plant Breeding, 1(4): 802–805.

Ji, H.-S., Chu, S.-H., Jiang, W., Cho, Y.-I., Hahn, J.-H., Eun, M.-Y., McCouch, S.R. &
Koh, H.-J. 2006a. Characterization and mapping of a shattering mutant in rice that
corresponds to a block of domestication genes. Genetics, 173(2): 995–1005.

Ji, H.-S., Chu, S.-H., Jiang, W., Cho, Y.-I., Hahn, J.-H., Eun, M.-Y., McCouch, S.R. &
Koh, H.-J. 2006b. Characterization and mapping of a shattering mutant in rice that
corresponds to a block of domestication genes. Genetics, 173(2): 995–1005.

Johnson, S.D., Taylor, D.R., Tai, T.H., Jankowicz-Cieslak, J., Till, B.J. & others. 2017.
Field Evaluation of Mutagenized Rice Material. Biotechnologies for Plant Mutation
Breeding, pp. 145–156. Springer.

279
Kale, D., Badigannavar, A. & Murty, G. 1999. Groundnut variety, TAG 24, with potential
for wider adaptability.

Kamaruddin, N.Y., Abdullah, S. & Harun, A.R. undated. The Effect of Gamma rays on
the Radiosensitivity and Cytological Analysis of Zingiber officinale Roscoe Varieties
Bentong and Tanjung Sepat.

Kanazawa, A., Liu, B., Kong, F., Arase, S. & Abe, J. 2009. Adaptive evolution involving
gene duplication and insertion of a novel Ty1/copia-like retrotransposon in soybean. Journal
of molecular evolution, 69(2): 164–175.

Kannan, B., Davila-Olivas, N.H., Lomba, P. & Altpeter, F. 2015a. In vitro chemical
mutagenesis improves the turf quality of bahiagrass. Plant Cell, Tissue and Organ Culture
(PCTOC), 120(2): 551–561.

Karelin, Y.A., Gordeev, Y.N., Karasev, V.I., Radchenko, V.M., Schimbarev, Y.V. &
Kuznetsov, R.A. 1997. Californium-252 neutron sources. Applied radiation and isotopes,
48(10–12): 1563–1566.

Kasha, K. & Kao, K. 1970. High frequency haploid production in barley (Hordeum vulgare
L.). Nature, 225(5235): 874–876.

Kato, H. 2008. Development of rice varieties for whole crop silage (WCS) in Japan. Japan
Agricultural Research Quarterly: JARQ, 42(4): 231–236.

Keightley, P.D. & Halligan, D.L. 2009. Analysis and implications of mutational variation.
Genetica, 136(2): 359.

Khan, M.H. & Tyagi, S.D. 2013. A review on induced mutagenesis in soybean. Journal of
Cereals and Oilseeds, 4(2): 19–25.

Khan, S., Al-Qurainy, F. & Anwar, F. 2009. Sodium azide: a chemical mutagen for
enhancement of agronomic traits of crop plants. Environ. We Int. J. Sci. Tech, 4: 1–21.

Khan, S., Wani, M.R. & Parveen, K. 2004. Induced genetic variability for quantitative traits
in Vigna radiata (L.) Wilczek. Pakistan Journal of Botany, 36(4): 845–850.

Kharkwal, M. 2012. A brief history of plant mutagenesis. Plant Mutation Breeding and
Biotechnology. CABI, Wallingford: 21–30.

Kharkwal, M., Pandey, R. & Pawar, S. 2004. Mutation breeding for crop improvement.
Plant Breeding, pp. 601–645. Springer.

Kikuchi, F. & Ikehashi, H. 1984. Semidwarfing genes of high-yielding rice varieties in

Japan. Rice Genetics Newsletter, 1: 93–94.

280
Kirk, R.E. & Gorzen, D.F. 2008. X-ray tube with cylindrical anode. Google Patents.

Klassen, N.V., Shortt, K.R., Seuntjens, J. & Ross, C.K. 1999. Fricke dosimetry: the
difference between G (Fe 3+ ) for 60 Co ##IMG##
[http://ej.iop.org/icons/common/gamma.gif] gamma -rays and high-energy x-rays. Physics in
Medicine & Biology, 44(7): 1609.

Kleinhofs, A., Kleinschmidt, M., Sciaky, D. & Von Broembsen, S. 1975. Azide
mutagenesis. In vitro studies. Mutation Research/Fundamental and Molecular Mechanisms
of Mutagenesis, 29(3): 497–499.

Klekowski, E.J. 2011. Plant clonality, mutation, diplontic selection and mutational
meltdown. Biological Journal of the Linnean Society, 79(1): 61.
https://doi.org/10.1046/j.1095-8312.2003.00183.x.

Kobayashi, S., Goto-Yamamoto, N. & Hirochika, H. 2004. Retrotransposon-induced

mutations in grape skin color. Science, 304(5673): 982–982.

Koltunow, A.M., Hidaka, T. & Robinson, S.P. 1996. Polyembryony in Citrus

(Accumulation of Seed Storage Proteins in Seeds and in Embryos Cultured in Vitro). Plant
physiology, 110(2): 599–609.

Konzak, C. & Mikaelsen, K. 1977. Induced mutation techniques in breeding the seed
propagated species. Manual on Mutation Breeding, Tech. Rept. Ser., 119: 125–138.

Kovacs, E. & Keresztes, A. 2002. Effect of gamma and UV-B/C radiation on plant cells.
Micron, 33(2): 199–210.

Kovalchuk, O., Arkhipov, A., Barylyak, I., Karachov, I., Titov, V., Hohn, B. &
Kovalchuk, I. 2000. Plants experiencing chronic internal exposure to ionizing radiation
exhibit higher frequency of homologous recombination than acutely irradiated plants.
Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis, 449(1): 47–56.

Kozgar, M.I., Hussain, S., Wani, M.R. & Khan, S. 2014. The role of cytological aberrations
in crop improvement through induced mutagenesis. Improvement of Crops in the Era of
Climatic Changes, pp. 283–296. Springer.

Kramer, J.K. 2012. High and Low Erucic Acid in Rapeseed Oils. Academic press.

Krasileva, K.V., Vasquez-Gross, H.A., Howell, T., Bailey, P., Paraiso, F., Clissold, L.,
Simmonds, J., Ramirez-Gonzalez, R.H., Wang, X., Borrill, P., Fosker, C., Ayling, S.,
Phillips, A.L., Uauy, C. & Dubcovsky, J. 2017. Uncovering hidden variation in polyploid
wheat. Proceedings of the National Academy of Sciences, 114(6): E913–E921.
https://doi.org/10.1073/pnas.1619268114.

281
Kuchel, H., Fox, R., Reinheimer, J., Mosionek, L., Willey, N., Bariana, H. & Jefferies,
S. 2007. The successful application of a marker-assisted wheat breeding strategy. Molecular
Breeding, 20(4): 295–308.

.XF]\ĔVND $ 6XUPD 0 $GDPVNL 7 0LNR\lajczak, K., Krystkowiak, K. &
Ogrodowicz, P. 2013. Effects of the semi-dwarfing sdw1/denso gene in barley. Journal of
applied genetics, 54(4): 381–390.

Kumar, D.S., Chakrabarty, D., Verma, A.K. & Banerji, B.K. 2011. Gamma ray induced
chromosomal aberrations and enzyme related defense mechanism in Allium cepa L.
Caryologia, 64(4): 388–397.

Kunter, B., Bas, M., Kantoglu, Y. & Burak, M. 2012. Mutation breeding of sweet cherry
(Prunus avium L.) var. 0900 Ziraat. Plant Mutation breeding and biotechnology. CAB
International, Oxfordshire: 453–463.

Kurowska, M., Daszkowska-Golec, A., Gruszka, D., Marzec, M., Szurman, M.,
Szarejko, I. & Maluszynski, M. 2011. TILLING-a shortcut in functional genomics. Journal
of applied genetics, 52(4): 371.

Kusaksiz, T. & Dere, S. 2010. A study on the determination of genotypic variation for seed
yield and its utilization through selection in durum wheat (Triticum durum Desf.) mutant
populations. Turkish Journal of Field Crops, 15(2): 188–192.

Kwasniewska, J. & Kwasniewski, M. 2013. Comet-FISH for the evaluation of plant DNA
damage after mutagenic treatments. Journal of applied genetics, 54(4): 407–415.

Lababidi, S., Mejlhede, N., Rasmussen, S.K., Backes, G., Al-Said, W., Baum, M. &
Jahoor, A. 2009. Identification of barley mutants in the cultivar ‘Lux’at the Dhn loci through
TILLING. Plant breeding, 128(4): 332–336.

Lafiandra, D., Riccardi, G. & Shewry, P.R. 2014. Improving cereal grain carbohydrates for
diet and health. Journal of cereal science, 59(3): 312–326.

Lagoda, P. & others. 2009. Networking and fostering of cooperation in plant mutation
genetics and breeding: role of the joint FAO/IAEA Division. Induced plant mutations in the
genomics era. FAO, Rome, 487: 27–30.

Lagoda, P., Shu, Q., Forster, B., Nakagawa, H. & Nakagawa, H. 2012. Effects of radiation
on living cells and plants. Plant Mutation Breeding and Biotechnology. CABI Publishing,
Wallingford: 123–134.

L’Annunziata, M.F. 2016. Radioactivity: Introduction and History, From the Quantum to
Quarks. Elsevier.

282
Law, C., Snape, J. & Worland, A. 1987. Aneuploidy in wheat and its uses in genetic
analysis. Wheat Breeding, pp. 71–108. Springer.

Lebeda, A. & Svabova, L. 2010. In vitro screening methods for assessing plant disease
resistance. International Atomic Energy Agency (IAEA), IAEA.

Lee, L.S., Till, B.J., Hill, H., Huynh, O.A. & Jankowicz-Cieslak, J. 2014. Mutation and
mutation screening. Cereal genomics: methods and protocols: 77–95.

Lee, S., Cheong, J. & Kim, T. 2003. Production of doubled haploids through anther culture
of M1 rice plants derived from mutagenized fertilized egg cells. Plant cell reports, 22(3):
218–223.

Legrand, C., Hartmann, G. & Karger, C. 2012. Experimental determination of the effective
point of measurement for cylindrical ionization chambers in 60Co gamma radiation. Physics
in medicine and biology, 57(11): 3463.

Leitão, J. 2012. Chemical mutagenesis. Plant mutation breeding and biotechnology. Italy:
CAB International and FAO: 135–58.

Leonard, A., Jacquet, P. & Lauwerys, R. 1983. Mutagenicity and teratogenicity of mercury
compounds. Mutation Research/Reviews in Genetic Toxicology, 114(1): 1–18.

Li, G., Chern, M., Jain, R., Martin, J.A., Schackwitz, W.S., Jiang, L., Vega-Sánchez,
M.E., Lipzen, A.M., Barry, K.W., Schmutz, J. & Ronald, P.C. 2016a. Genome-Wide
Sequencing of 41 Rice (Oryza sativa L.) Mutated Lines Reveals Diverse Mutations Induced
by Fast-Neutron Irradiation. Molecular Plant, 9(7): 1078–1081.
https://doi.org/10.1016/j.molp.2016.03.009.

Li, L., Zhang, Q. & Huang, D. 2014. A review of imaging techniques for plant phenotyping.
Sensors, 14(11): 20078–20111.

Li, S., Zheng, Y., Cui, H., Fu, H., Shu, Q. & Huang, J. 2016b. Frequency and type of
LQKHULWDEOH PXWDWLRQV LQGXFHG E\ Ȗ UD\V LQ ULFH DVrevealed by whole genome sequencing.
Journal of Zhejiang University. Science. B, 17(12): 905–915.
https://doi.org/10.1631/jzus.B1600125.

Lisch, D. 2013. How important are transposons for plant evolution? Nature Reviews.
Genetics, 14(1): 49.

Liu, L., Guo, H., Zhao, L., Wang, J., Gu, J., Zhao, S. & others. 2009. Achievements and
perspectives of crop space breeding in China. Induced Plant Mutations in the Genomics Era:
213–15.

283
Liu, W., Chen, P. & Liu, D. 1998. Development of Triticum aestivum-Leymus racemosus
translocation lines by irradiating adult plants at meiosis. Acta Botanica Sinica, 41(5): 463–
467.

Liu, W.-X., Chen, P.-D. & Liu, D.-J. 2000. Studies of the development of Triticum
aestivum-Leymus racemosus translocation lines by pollen irradiation. Yi chuan xue bao=
Acta genetica Sinica, 27(1): 44–49.

Lokesha, R., Hegde, S.G., Shaanker, R.U. & Ganeshaiah, K.N. 1992. Dispersal mode as
a selective force in shaping the chemical composition of seeds. The American Naturalist,
140(3): 520–525. (also available at
http://search.ebscohost.com/login.aspx?direct=true&db=mdc&AN=19426054&site=eds-
live).

Lowry, D.B. & Willis, J.H. 2010. A widespread chromosomal inversion polymorphism
contributes to a major life-history transition, local adaptation, and reproductive isolation.
PLoS Biol, 8(9): e1000500.

Luan, Y.-S., Zhang, J., Gao, X.-R. & An, L.-J. 2007. Mutation induced by
ethylmethanesulphonate (EMS), in vitro screening for salt tolerance and plant regeneration of
sweet potato (Ipomoeabatatas L.). Plant cell, tissue and organ culture, 88(1): 77–81.

Lundqvist, U. 2014. Scandinavian mutation research in barley–a historical review.

Hereditas, 151(6): 123–131.

Lundqvist, U., Franckowiak, J. & Forster, B. 2012. Mutation categories. Plant mutation
breeding and biotechnology. CABI, FAO, Oxfordshire, UK: 47–55.

Luo, X., Tinker, N.A., Jiang, Y., Xuan, P., Zhang, H. & Zhou, Y. 2013. Suitable dose of
&R Ȗ-ray for mutation in Roegneria seeds. Journal of Radioanalytical and Nuclear
Chemistry, 295(2): 1129–1134.

Mabbett, T.  +HUELFLGH WROHUDQW FURSVʊ,&, VHHGV OHDGV WKH ZD\ International pest
control, 34(2): 49–50.

Mahamune, S. & Kothekar, V. 2012. Induced mutagenic frequency and spectrum of

chlorophyll mutants in French bean. International Multidisciplinary Research Journal, 2(3).

Mak, C., Ho, Y., Tan, Y. & Ibrahim, R. 1996. Novaria-A new banana mutant induced by
gamma irradiation. InfoMusa, 5(1): 35–36.

Mak, C., Mohamed, A., Liew, K., Ho, Y. & others. 2004. Early screening technique for
Fusarium wilt resistance in banana micropropagated plants. Banana Improvement, 18: 219–
227.

284
Maluszynska, J. 2003. Cytogenetic tests for ploidy level analyses—chromosome counting.
Doubled Haploid Production in Crop Plants, pp. 391–395. Springer.

Maluszynski, M., Ahloowalia, B.S. & Sigurbjörnsson, B. 1995. Application of in vivo and
in vitro mutation techniques for crop improvement. Euphytica, 85(1): 303–315.

Maluszynski, M. & Kasha, K. 2002. Mutations. Vitro and molecular techniques for
environmentally sustainable crop improvement. Kluwer Academic Publishers, Dordrecht,
246p.

Maluszynski, M., Kasha, K. & Szarejko, I. 2003. Published doubled haploid protocols in
plant species. Doubled Haploid Production in Crop Plants, pp. 309–335. Springer.

Maluszynski, M. & Szarejko, I. 2003. Induced mutations in the Green and Gene
Revolutions. International Congress “In the wake of the double helix: From the Green
Revolution to the Gene Revolution. pp. 27–31. Paper presented at, 2003.

Maluszynski, M., Szarejko, I., Bhatia, C., Nichterlein, K., Lagoda, P.J. & others. 2009.
Methodologies for generating variability. Part 4: Mutation techniques. Plant breeding and
farmer participation: 159–194.

Mandal, A., Chakrabarty, D. & Datta, S. 2000. In vitro isolation of solid novel flower
colour mutants from induced chimeric ray florets of chrysanthemum. Euphytica, 114(1): 9–
12.

Manova, V. & Gruszka, D. 2015. DNA damage and repair in plants–from models to crops.
Frontiers in plant science, 6: 885.

Marcu, D., Cristea, V. & Daraban, L. 2013. Dose-dependent effects of gamma radiation on
lettuce (Lactuca sativa var. capitata) seedlings. International Journal of Radiation Biology,
89(3): 219–223.

Masoabi, M., Lloyd, J., Kossmann, J. & van der Vyver, C. undated. Ethyl
Methanesulfonate Mutagenesis and In Vitro Polyethylene Glycol Selection for Drought
Tolerance in Sugarcane (Saccharum spp.). Sugar Tech: 1–10.

Mba, C. 2013. Induced mutations unleash the potentials of plant genetic resources for food
and agriculture. Agronomy, 3(1): 200–231.

Mba, C., Afza, R., Shu, Q., Shu, Q., Forster, B. & Nakagawa, H. 2012. Mutagenic
radiations: X-rays, ionizing particles and ultraviolet. Plant mutation breeding and
biotechnology. CABI, Oxfordshire: 83–90.

Mehta, K. & Parker, A. 2011. Characterization and dosimetry of a practical X-ray

alternative to self-shielded gamma irradiators. Radiation Physics and Chemistry, 80(1): 107–
113.

285
Micke, A. 1969. Improvement Of Low Yielding Sweet Clover Mutants by Heterosis
Breeding. International Atomic Energy Agency, Vienna.

Micke, A. & Donini, B. 1993. Induced mutations. Plant Breeding, pp. 52–62. Springer.

Micke, A. & Wohrmann, K. 1960. On the problem of the radiation sensitivity of dry seed.
Atompraxis (West Germany) Incorporated in Kerntechnik, 6.

Miller, M.W. & Miller, W.M. 1987. Radiation hormesis in plants. Health physics, 52(5):
607–616.

Minoia, S., Petrozza, A., D’Onofrio, O., Piron, F., Mosca, G., Sozio, G., Cellini, F.,
Bendahmane, A. & Carriero, F. 2010. A new mutant genetic resource for tomato crop
improvement by TILLING technology. BMC research notes, 3(1): 69.

Moiseenko, V., Khvostunov, I.K., Hattangadi-Gluth, J.A., Muren, L.P. & Lloyd, D.C.
2016. Biological dosimetry to assess risks of health effects in victims of radiation accidents:
Thirty years after Chernobyl. Radiother Oncol, 119: 1–4.

Moody, D. 2015. Breeding for imidazolinone tolerant barley varieties: industry issues and
concerns. Grain Research & Development Corporation.

Morgan Jr, D.T. 1950. A cytogenetic study of inversions in Zea mays. Genetics, 35(2): 153.

Mukai, Y., Friebe, B., Hatchett, J.H., Yamamoto, M. & Gill, B.S. 1993. Molecular
cytogenetic analysis of radiation-induced wheat-rye terminal and intercalary chromosomal
translocations and the detection of rye chromatin specifying resistance to Hessian fly.
Chromosoma, 102(2): 88–95. https://doi.org/10.1007/BF00356025

Murashige, T. & Skoog, F. 1962. A revised medium for rapid growth and bio assays with
tobacco tissue cultures. Physiologia plantarum, 15(3): 473–497.

Nairy, R.K., Bhat, N.N., Anjaria, K., Sreedevi, B., Sapra, B. & Narayana, Y. 2014. Study
of gamma radiation induced damages and variation of oxygen enhancement ratio with
radiation dose using Saccharomyces cerevisiae. Journal of Radioanalytical and Nuclear
Chemistry, 302(2): 1027–1033.

Naito, K., Zhang, F., Tsukiyama, T., Saito, H., Hancock, C.N., Richardson, A.O.,
Okumoto, Y., Tanisaka, T. & Wessler, S.R. 2009. Unexpected consequences of a sudden
and massive transposon amplification on rice gene expression. Nature, 461(7267): 1130.

Nashima, K., Takahashi, H., Nakazono, M., Shimizu, T., Nishitani, C., Yamamoto, T.,
Itai, A., Isuzugawa, K., Hanada, T., Takashina, T. & others. 2013. Transcriptome analysis
of giant pear fruit with fruit-specific DNA reduplication on a mutant branch. Journal of the
Japanese Society for Horticultural Science, 82(4): 301–311.

286
Nasution, O., Sitorus, A.C., Nelson, S.P., Forster, B.P., Caligari, P.D. & others. 2013. A
high-throughput flow cytometry method for ploidy determination in oil palm. Journal of Oil
Palm Research, 25(2): 265–271.

Neuffer, M. 1994. Mutagenesis. The maize handbook, pp. 212–219. Springer.

Nielen, S., Guzman, M. & Zapata-Aries, F. 2000. Studies on Tos17 retrotransposon in rice
plants derived from irradiated seeds and in gametoclonal variants. Abs. for ISPMB meeting.
Paper presented at, 2000.

Nikam, A.A., Devarumath, R.M., Ahuja, A., Babu, H., Shitole, M.G. & Suprasanna, P.
2015. Radiation-induced in vitro mutagenesis system for salt tolerance and other agronomic
characters in sugarcane (Saccharum officinarum L.). The Crop Journal, 3(1): 46–56.

Nilan, R., Sideris, E., Kleinhofs, A., Sander, C. & Konzak, C. 1973. Azide—a poten
mutagen. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis,
17(1): 142–144.

Oerke, E.-C. 2006. Crop losses to pests. The Journal of Agricultural Science, 144(1): 31–43.

Okamura, M., Tanaka, A., Momose, M., Umemoto, N., da Silva, J.A.T. & Toguri, T.
2006. Advances of mutagenesis in flowers and their industrialization. Energy, 20(260): 0–3.

Oladosu, Y., Rafii, M.Y., Abdullah, N., Hussin, G., Ramli, A., Rahim, H.A., Miah, G. &
Usman, M. 2016. Principle and application of plant mutagenesis in crop improvement: a
review. Biotechnology & Biotechnological Equipment, 30(1): 1–16.

Olsen, O., Wang, X. & von Wettstein, D. 1993. Sodium azide mutagenesis: preferential
generation of AT–> GC transitions in the barley Ant18 gene. Proceedings of the National
Academy of Sciences, 90(17): 8043–8047.

Ortiz, R. & Peloquin, S. 1992. Recurrent selection for 2n gamete production in 2x potatoes.
Journal of Genetics and Breeding, 46: 383–383.

Owais, W. & Kleinhofs, A. 1988. Metabolic activation of the mutagen azide in biological
systems. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis,
197(2): 313–323.

Owczarzy, R., Tataurov, A.V., Wu, Y., Manthey, J.A., McQuisten, K.A., Almabrazi,
H.G., Pedersen, K.F., Lin, Y., Garretson, J., McEntaggart, N.O. & others. 2008. IDT
SciTools: a suite for analysis and design of nucleic acid oligomers. Nucleic acids research,
36(suppl_2): W163–W169.

287
Palom, Y., Suresh Kumar, G., Tang, L.-Q., Paz, M.M., Musser, S.M., Rockwell, S. &
Tomasz, M. 2002. Relative toxicities of DNA cross-links and monoadducts: new insights
from studies of decarbamoyl mitomycin C and mitomycin C. Chemical research in
toxicology, 15(11): 1398–1406.

Pan, L., Shah, A.N., Phelps, I.G., Doherty, D., Johnson, E.A. & Moens, C.B. 2015. Rapid
identification and recovery of ENU-induced mutations with next-generation sequencing and
Paired-End Low-Error analysis. BMC genomics, 16(1): 83.

Paran, I. & Zamir, D. 2003. Quantitative traits in plants: beyond the QTL. Trends in
Genetics, 19(6): 303–306.

Patade, V. & Suprasanna, P. 2008. Radiation induced in vitro mutagenesis for sugarcane
improvement. Sugar Tech, 10(1): 14–19.

Paterson, A.H., Bowers, J.E., Bruggmann, R., Dubchak, I., Grimwood, J., Gundlach, H.,
Haberer, G., Hellsten, U., Mitros, T., Poliakov, A. & others. 2009. The Sorghum bicolor
genome and the diversification of grasses. Nature, 457(7229): 551–556.

Paterson, A.H., Lander, E.S., Hewitt, J.D., Peterson, S., Lincoln, S.E. & Tanksley, S.D.
1988. Resolution of quantitative traits into Mendelian factors by using a complete linkage
map of restriction fragment length polymorphisms. Nature, 335(6192): 721–726.

Pathirana, R. 2011. Plant mutation breeding in agriculture. Plant sciences reviews: 107–126.

Pelsy, F. 2010. Molecular and cellular mechanisms of diversity within grapevine varieties.
Heredity, 104(4): 331.

Peterson, P. 1991. The transposable element-En-four decades after Bikini. Genetica, 84(2):
63–72.

Phillips, G.C. 2004. In vitro morphogenesis in plants–recent advances. In vitro Cellular and
Developmental Biology-Plant, 40(4): 342–345.

Piffanelli, P., Droc, G., Mieulet, D., Lanau, N., Bès, M., Bourgeois, E., Rouvière, C.,
Gavory, F., Cruaud, C., Ghesquière, A. & others. 2007. Large-scale characterization of
Tos17 insertion sites in a rice T-DNA mutant library. Plant molecular biology, 65(5): 587–
601.

Podgorsak, E. & others. 2005. Radiation oncology physics. a handbook for teachers and
students/EB Podgorsak.–Vienna: International Atomic Energy Agency, 657.

Poli, Y., Basava, R.K., Panigrahy, M., Vinukonda, V.P., Dokula, N.R., Voleti, S.R.,
Desiraju, S. & Neelamraju, S. 2013. Characterization of a Nagina22 rice mutant for heat
tolerance and mapping of yield traits. Rice, 6(1): 36.

288
Prabhukumar, K., Thomas, V., Sabu, M., Prasanth, A. & Mohanan, K. 2015. Induced
mutation in ornamental gingers (Zingiberaceae) using chemical mutagens viz. Colchicine,
Acridine and Ethylmethanesulphonate. Journal of Horticulture, Forestry and Biotechnology,
19(2): 18–27.

Predieri, S. & Di Virgilio, N. 2007. In vitro mutagenesis and mutant multiplication.

Protocols for micropropagation of woody trees and fruits, pp. 323–333. Springer.

Pretorius, Z., Singh, R., Wagoire, W. & Payne, T. 2000. Detection of virulence to wheat
stem rust resistance gene Sr31 in Puccinia graminis. f. sp. tritici in Uganda. Plant Disease,
84(2): 203–203.

Prina, A., Pacheco, M. & Landau, A. 2012. Mutation induction in cytoplasmic genomes.
Plant Mutation Breeding and Biotechnology: 201–206.

3URWLü57RGRURYLü*3URWLü1.RVWLü0'HOLü')LOLSRYLü0)LOLSRYLü9
8JUHQRYLü9 RWKHUV. 2013. Variation of Grain Weight Per Spike of Wheat Depending
On Variety And Seed Size. Romanian Agricultural Research(30): 51–55.

Puchta, H., Dujon, B. & Hohn, B. 1996. Two different but related mechanisms are used in
plants for the repair of genomic double-strand breaks by homologous recombination.
Proceedings of the National Academy of Sciences, 93(10): 5055–5060.

Purnamaningsih, R. & Hutami, S. 2016. Increasing Al-Tolerance of Sugarcane Using Ethyl

Methane Sulphonate and In Vitro Selection in the Low pH Media. HAYATI Journal of
Biosciences, 23(1): 1–6.

Qüesta, J., Fina, J. & Casati, P. 2013a. DDM1 and ROS1 have a role in UV-B induced-and
oxidative DNA damage in A. thaliana. Frontiers in plant science, 4: 420.

Raboy, V. 2009. Approaches and challenges to engineering seed phytate and total
phosphorus. Plant Science, 177(4): 281–296.

Rai, M.K., Kalia, R.K., Singh, R., Gangola, M.P. & Dhawan, A. 2011. Developing stress
tolerant plants through in vitro selection—an overview of the recent progress. Environmental
and Experimental Botany, 71(1): 89–98.

Rajarajan, D., Saraswathi, R., Sassikumar, D. & Ganesh, S. 2014. Effectiveness and
efficiency of gamma ray and ems induced chlorophyll mutants in rice ADT (R) 47. GJBAHS,
3(3): 211–218.

Rakshit, S., Hariprasanna, K., Gomashe, S., Ganapathy, K., Das, I., Ramana, O.,
Dhandapani, A. & Patil, J. 2014. Changes in area, yield gains, and yield stability of sorghum
in major sorghum-producing countries, 1970 to 2009. Crop Science, 54(4): 1571–1584.

289
Ravi, M. & Chan, S.W. 2010. Haploid plants produced by centromere-mediated genome
elimination. Nature, 464(7288): 615–618.

Reeb-Whitaker, C.K. & Bonauto, D.K. 2014. Respiratory disease associated with
occupational inhalation to hop (Humulus lupulus) during harvest and processing. Annals of
Allergy, Asthma & Immunology, 113(5): 534–538.

Robinson-Beers, K., Pruitt, R.E. & Gasser, C.S. 1992. Ovule development in wild-type
Arabidopsis and two female-sterile mutants. The Plant Cell, 4(10): 1237–1249.

Roux, N. 2004. Mutation induction in Musa–review. Banana Improvement: Cellular,

Molecular Biology, and Induced Mutations. Enfield: Sci Pub, Inc. p: 21–29.

Roux, N., Dolezel, J., Swennen, R. & Zapata-Arias, F.J. 2001. Effectiveness of three
micropropagation techniques to dissociate cytochimeras in Musa spp. Plant Cell, Tissue and
Organ Culture, 66(3): 189–197.

Roux, N., Toloza, A., Radecki, Z., Zapata-Arias, F. & Dolezel, J. 2003. Rapid detection of
aneuploidy in Musa using flow cytometry. Plant Cell Reports, 21(5): 483–490.

Rutger, J. & Mackill, D. 2001. Application of Mendelian genetics in rice breeding. Rice
Genetics IV: 27.

Rutger, J., Peterson, M., Hu, C. & Lehman, W. 1976. Induction of useful short stature and
early maturing mutants in two japonica rice cultivars. Crop Science, 16(5): 631–635.

Rutger, J.N. 1992. Impact of mutation breeding in rice.

Rutger, J.N., Peterson, M.L. & Hu, C. 1977. Registration of Calrose 76 Rice1 (Reg. No.
45). Crop Science, 17(6): 978–978.

Rybianski, W. 2000. Influence of laser beams on the variability of traits in spring barley.
International Agrophysics, 14(2): 227–232.

Sacerdot, C., Mercier, G., Todeschini, A.-L., Dutreix, M., Springer, M. & Lesage, P.
2005. Impact of ionizing radiation on the life cycle of Saccharomyces cerevisiae Ty1
retrotransposon. Yeast, 22(6): 441–455.

Sadiq, M.F. & Owais, W.M. 2000. Mutagenicity of sodium azide and its metabolite
azidoalanine in Drosophila melanogaster. Mutation Research/Genetic Toxicology and
Environmental Mutagenesis, 469(2): 253–257.

Sadiq, M.S., Haidar, S., Haq, M.A. & Abbas, G. 2008. A high yielding and disease resistant
mutant of lentil developed through seed irradiation of an exotic germplasm. Canadian Journal
of Pure and Applied Sciences: 411.

290
Saeed, I. & Hassan, M.F. 2009. High yielding groundnut (Arachis hypogea L.) variety
“Golden”. Pak. J. Bot, 41(5): 2217–2222.

Saito, T. 2016. Advances in Japanese pear breeding in Japan. Breeding Science, 66(1): 46–
59. https://doi.org/10.1270/jsbbs.66.46.

Sanamyan, M.F., Petlyakova, J.E., Sharipova, E.A. & Abdurakhmonov, I.Y. 2011.
Cytogenetic characteristics of new monosomic stocks of cotton (Gossypium hirsutum L.).
Genetics research international, 2011.

Sanders, P.M., Bui, A.Q., Weterings, K., McIntire, K., Hsu, Y.-C., Lee, P.Y., Truong,
M.T., Beals, T. & Goldberg, R. 1999. Anther developmental defects in Arabidopsis thaliana
male-sterile mutants. Sexual plant reproduction, 11(6): 297–322.

Sanger, F. 1981. Determination of nucleotide sequences in DNA. Bioscience reports, 1(1):

3–18.

Saraswathi, M., Kannan, G., Uma, S., Thangavelu, R. & Backiyarani, S. 2016.
Improvement of banana cv. Rasthali (Silk, AAB) against Fusarium oxysporum f. sp. cubense
(VCG 0124/5) through induced mutagenesis: Determination of LD50 specific to mutagen,
explants, toxins and in vitro and in vivo screening for Fusarium wilt resistance.

Sato, S., Katoh, N., Yoshida, H., Iwai, S. & Hagimori, M. 2000. Production of doubled
haploid plants of carnation (Dianthus caryophyllus L.) by pseudofertilized ovule culture.
Scientia horticulturae, 83(3): 301–310.

Sauton, A. & Dumas de Vaulx, R. 1987. Induction of gynogenetic haploid plants in

muskmelon (Cucumis melo L.) by use of irradiated pollen. Agronomie (France).

Saxena, G., Verma, P.C., Rahman, L., Banerjee, S., Shukla, R. & Kumar, S. 2008.
Selection of leaf blight-resistant Pelargonium graveolens plants regenerated from callus
resistant to a culture filtrate of Alternaria alternata. Crop protection, 27(3): 558–565.

Scheibe, A. & Micke, A. 1967. Experimentally induced mutations in leguminous forage

plants and their agronomic value. Induced Mutations and their Utilization. Akademie-Verlag
Berlin: 231–236.

Schneeberger, K., Ossowski, S., Lanz, C., Juul, T., Petersen, A.H., Nielsen, K.L.,
Jorgensen, J.E., Weigel, D. & Andersen, S.U. 2009. SHOREmap: simultaneous mapping
and mutation identification by deep sequencing. Nat Methods, 6: 550–1.
https://doi.org/10.1038/nmeth0809-550

Schreiner, L. 2004. Review of Fricke gel dosimeters. Journal of Physics: Conference Series.
p. 9. Paper presented at, 2004.

291
Sears, E. & others. 1956. The transfer of leaf-rust resistance from Aegilops umbellulata to
wheat. The transfer of leaf-rust resistance from Aegilops umbellulata to wheat.

Sekiguchi, M., Hase, Y. & Tanaka, A. 2009. Osteospermum plant named ‘Vient Flamingo’.
Google Patents.

Semal, J. 2013. Somaclonal variations and crop improvement. Springer Science & Business
Media.

Serrat, X., Esteban, R., Guibourt, N., Moysset, L., Nogués, S. & Lalanne, E. 2014. EMS
mutagenesis in mature seed-derived rice calli as a new method for rapidly obtaining TILLING
mutant populations. Plant Methods, 10(1): 5.

Servin, A., Elmer, W., Mukherjee, A., De la Torre-Roche, R., Hamdi, H., White, J.C.,
Bindraban, P. & Dimkpa, C. 2015. A review of the use of engineered nanomaterials to
suppress plant disease and enhance crop yield. Journal of Nanoparticle Research, 17(2): 92.

Seymour, G.B., Taylor, J.E. & Tucker, G.A. 2012. Biochemistry of fruit ripening. Springer
Science & Business Media.

Shirao, T., Ueno, K., Abe, T. & Matsuyama, T. 2013. Development of DNA markers for
identifying chrysanthemum cultivars generated by ion-beam irradiation. Molecular breeding,
31(3): 729–735.

Shu, Q., Forster, B.P., & Nakagawa, H. 2012. Plant mutation breeding and biotechnology.
CABI.

Shu, Q. & others. 2009. Turning plant mutation breeding into a new era: molecular mutation
breeding. Induced plant mutations in the genomics era. FAO, Rome: 425–427.

Sigurbjörnsson, B. 1975. Methods of mutation induction, including efficiency, and

utilization of induced genetic variability. Barley Genetics III: 84–95.

Sigurbjornsson, B. & Micke, A. 1974. Philosophy and accomplishments of mutation

breeding. Polyploidy & Induced Mutations in Plant Breeding Proceedings.

Simonson, R., Baenziger, P. & others. 1991. Response of different wheat tissues to
increasing doses of ethyl methanesulfonate. Plant Cell, Tissue and Organ Culture, 26(3):
141–146.

Singh, K.P., Kumar, R. & Verma, P.K. 2017. Opportunities in floriculture for livelihood
security. Advances in Floriculture and Landscape Gardening: 66.

Singh, R., Abdullah, M.O., Ti, L.L.E., Nookiah, R., Manaf, M.A.A. &
Sambanthamurthi, R. 2015. SureSawitTM SHELL-A Diagnostic Assay to Predict Fruit
Forms of Oil Palm. Oil Palm Bulletin, 70: 13–16.

292
Singh, R., Low, E.-T.L., Ooi, L.C.-L., Ong-Abdullah, M., Ting, N.-C., Nagappan, J.,
Nookiah, R., Amiruddin, M.D., Rosli, R., Manaf, M.A.A. & others. 2013a. The oil palm
SHELL gene controls oil yield and encodes a homologue of SEEDSTICK. Nature, 500(7462):
340.

Singh, R., Ong-Abdullah, M., Low, E.-T.L., Manaf, M.A.A., Rosli, R., Nookiah, R., Ooi,
L.C.-L., Ooi, S., Chan, K.-L., Halim, M.A. & others. 2013b. Oil palm genome sequence
reveals divergence of interfertile species in Old and New worlds. Nature, 500(7462): 335–
339.

Singh, R.J. 2016. Plant cytogenetics. CRC press.

Singh, R.P., Hodson, D.P., Huerta-Espino, J., Jin, Y., Bhavani, S., Njau, P., Herrera-
Foessel, S., Singh, P.K., Singh, S. & Govindan, V. 2011. The emergence of Ug99 races of
the stem rust fungus is a threat to world wheat production. Annual review of phytopathology,
49: 465–481.

Singh, S. & Verma, A.K. 2015. A review on efforts of induced mutagenesis for qualitative
and quantitative improvement of oilseed brassicas. Journal of Pharmacognosy and
Phytochemistry, 4(2).

Sjödin, J. 1971. Induced paracentric and pericentric inversions in Vicia faba L. Hereditas,
67(1): 39–54.

Slota, M., Maluszynski, M. & Szarejko, I. 2017. Bioinformatics-Based Assessment of the

Relevance of Candidate Genes for Mutation Discovery. Biotechnologies for Plant Mutation
Breeding, pp. 263–280. Springer.

Soeranto, H., Manurung, S. & others. 2001. The use of physical/chemical mutagens for
crop improvements in Indonesia.

Spielmeyer, W., Ellis, M.H. & Chandler, P.M. 2002. Semidwarf (sd-1),“green revolution”
rice, contains a defective gibberellin 20-oxidase gene. Proceedings of the National Academy
of Sciences, 99(13): 9043–9048.

Staub, J.E., Serquen, F.C. & Gupta, M. 1996. Genetic markers, map construction, and their
application in plant breeding. HortScience, 31(5): 729–741.

Stephenson, P., Baker, D., Girin, T., Perez, A., Amoah, S., King, G.J. & Østergaard, L.
2010. A rich TILLING resource for studying gene function in Brassica rapa. BMC plant
biology, 10(1): 62.

Studer, A., Zhao, Q., Ross-Ibarra, J. & Doebley, J. 2011. Identification of a functional
transposon insertion in the maize domestication gene tb1. Nature genetics, 43(11): 1160–
1163.

293
Suprasanna, P. 2010. Biotechnological Interventions in sugarcane improvement: Strategies,
methods and progress. BARC Newsletter: 47–53.

Suprasanna, P. & Nakagawa, H. 2012. Mutation breeding of vegetatively propagated crops.

Plant Mutation Breeding and Biotechnology. Food and Agriculture Organization of the
United Nations: 347–358.

Suprasanna, P., Rupali, C., Desai, N. & Bapat, V. 2008. Partial desiccation augments plant
regeneration from irradiated embryogenic cultures of sugarcane. Plant Cell, Tissue and Organ
Culture, 92(1): 101–105.

Suresh, D., Poonguzhali, S., Sridharan, S. & Rajangam, J. 2017. Determination of Lethal
Dose for Gamma Rays Induced Mutagenesis in Butter Bean (Phaseolus lunatus L) Variety
KKL-1. Int. J. Curr. Microbiol. App. Sci, 6(3): 712–717.

Suzuki, T., Eiguchi, M., Kumamaru, T., Satoh, H., Matsusaka, H., Moriguchi, K.,
Nagato, Y. & Kurata, N. 2008a. MNU-induced mutant pools and high performance
TILLING enable finding of any gene mutation in rice. Molecular Genetics and Genomics,
279(3): 213–223.

Suzuki, Y., Sano, Y., Ise, K., Matsukura, U., Aoki, N. & Sato, H. 2008b. A rice mutant
with enhanced amylose content in endosperm without affecting amylopectin structure.
Breeding Science, 58(3): 209–215.

Szarejko, I. 2003. Anther culture for doubled haploid production in barley (Hordeum vulgare
L.). Doubled Haploid Production in Crop Plants, pp. 35–42. Springer.

Szarejko, I. 2012. Haploid mutagenesis. Plant mutation breeding and biotechnology. CABI,
Wallingford: 387–410.

Szarejko, I. & Forster, B. 2007. Doubled haploidy and induced mutation. Euphytica, 158(3):
359–370.

Szarejko, I., Guzy, J., Jimenez Davalos, J., Roland Chavez, A. & Maluszynski, M. 1995.
Production of mutants using barley DH systems. Induced mutations and molecular techniques
for crop improvement. IAEA, Vienna: 517–530.

Szarejko, I., Szurman-Zubrzycka, M., Nawrot, M., Marzec, M., Gruszka, D., Kurowska,
M., Chmielewska, B., Zbieszczyk, J., Jelonek, J. & Maluszynski, M. 2017. Creation of a
TILLING population in barley after chemical mutagenesis with sodium azide and MNU.
Biotechnologies for Plant Mutation Breeding, pp. 91–111. Springer.

Tai, T.H., Chun, A., Henry, I.M., Ngo, K.J. & Burkart-Waco, D. 2016. Effectiveness of
Sodium Azide Alone Compared to Sodium Azide in Combination with Methyl Nitrosurea for
Rice Mutagenesis. Plant Breeding and Biotechnology, 4(4): 453–461.

294
Takahashi, M., Shimada, S., Nakayama, N. & Arihara, J. 2005. Characteristics of
nodulation and nitrogen fixation in the improved supernodulating soybean (Glycine max L.
Merr.) cultivar ‘Sakukei 4’. Plant production science, 8(4): 405–411.

Talamè, V., Bovina, R., Sanguineti, M.C., Tuberosa, R., Lundqvist, U. & Salvi, S. 2008.
TILLMore, a resource for the discovery of chemically induced mutants in barley. Plant
biotechnology journal, 6(5): 477–485.

Tanner, G.J., Blundell, M.J., Colgrave, M.L. & Howitt, C.A. 2016. Creation of the first
ultra-low gluten barley (Hordeum vulgare L.) for coeliac and gluten-intolerant populations.
Plant biotechnology journal, 14(4): 1139–1150.

Tardieu, F., Cabrera-Bosquet, L., Pridmore, T. & Bennett, M. 2017. Plant phenomics,
from sensors to knowledge. Current Biology, 27(15): R770–R783.

Thakur, R. & Ishii, K. 2014. Detection and fingerprinting of narrow-leaf mutants in micro-
propagated hybrid poplar (Populus sieboldii x P grandidentata) using random amplified
polymorphic DNA. International Journal of Farm Sciences, 2(1): 79–84.

Thorpe, T.A. 2006. History of plant tissue culture. Plant cell culture protocols: 9–32.

Thorpe, T.A. 2012. In vitro embryogenesis in plants. Springer Science & Business Media.

Till, B.J., Cooper, J., Tai, T.H., Colowit, P., Greene, E.A., Henikoff, S. & Comai, L. 2007.
Discovery of chemically induced mutations in rice by TILLING. BMC plant biology, 7(1):
19.

7LOO %- +RILQJHU %- ùHQ $ +X\QK 2$ -DQNRZLF]-Cieslak, J., Gugsa, L. &
Kumlehn, J. 2017. A Protocol for Validation of Doubled Haploid Plants by Enzymatic
Mismatch Cleavage. Biotechnologies for Plant Mutation Breeding, pp. 253–262. Springer.

Till, B.J., Reynolds, S.H., Greene, E.A., Codomo, C.A., Enns, L.C., Johnson, J.E.,
Burtner, C., Odden, A.R., Young, K., Taylor, N.E. & others. 2003. Large-scale discovery
of induced point mutations with high-throughput TILLING. Genome research, 13(3): 524–
530.

Till, B.J., Reynolds, S.H., Weil, C., Springer, N., Burtner, C., Young, K., Bowers, E.,
Codomo, C.A., Enns, L.C., Odden, A.R. & others. 2004. Discovery of induced point
mutations in maize genes by TILLING. BMC plant biology, 4(1): 12.

Till, B.J., Zerr, T., Comai, L. & Henikoff, S. 2006. A protocol for TILLING and Ecotilling
in plants and animals. Nature protocols, 1(5): 2465.

Tivana, L.D., Francisco, J.D.C., Zelder, F., Bergenst\a ahl, B. & Dejmek, P. 2014.
Straightforward rapid spectrophotometric quantification of total cyanogenic glycosides in
fresh and processed cassava products. Food chemistry, 158: 20–27.

295
Todd, W., Green, R. & Horner, C. 1977. Registration of Murray Mitcham Peppermint1
(Reg. No. 2). Crop Science, 17(1): 188–188.

Toker, C., Yadav, S.S. & Solanki, I. 2007. Mutation breeding. Lentil, pp. 209–224. Springer.

Tripathy, S.K., Panda, A., Nayak, P.K., Dash, S., Lenka, D., Mishra, D., Kar, R.K.,
Senapati, N. & Dash, G. 2016. Somaclonal variation for genetic improvement in grasspea
(Lathyrus sativus L.). Legume Research, 39(3): 329–335.

Tsai, H., Howell, T., Nitcher, R., Missirian, V., Watson, B., Ngo, K.J., Lieberman, M.,
Fass, J., Uauy, C., Tran, R.K., Khan, A.A., Filkov, V., Tai, T.H., Dubcovsky, J. & Comai,
L. 2011a. Discovery of Rare Mutations in Populations: TILLING by Sequencing. Plant
Physiology, 156: 1257–1268. https://doi.org/DOI 10.1104/pp.110.169748

Tsubaki, M., Mogi, T., Anraku, Y. & Hori, H. 1993. Structure of the heme-copper binuclear
center of the cytochrome bo complex of Escherichia coli: EPR and Fourier transform infrared
spectroscopic studies. Biochemistry, 32(23): 6065–6072.

Tsuda, M., Kaga, A., Anai, T., Shimizu, T., Sayama, T., Takagi, K., Machita, K.,
Watanabe, S., Nishimura, M., Yamada, N. & others. 2015. Construction of a high-density
mutant library in soybean and development of a mutant retrieval method using amplicon
sequencing. BMC genomics, 16(1): 1014.

Turelli, M. 1984. Heritable genetic variation via mutation-selection balance: Lerch’s zeta
meets the abdominal bristle. Theoretical population biology, 25(2): 138–193.

Tuvesson, S., Dayteg, C., Hagberg, P., Manninen, O., Tanhuanpää, P., Tenhola-
Roininen, T., Kiviharju, E., Weyen, J., Förster, J., Schondelmaier, J. & others. 2007.
Molecular markers and doubled haploids in European plant breeding programmes. Euphytica,
158(3): 305–312.

Uauy, C., Paraiso, F., Colasuonno, P., Tran, R.K., Tsai, H., Berardi, S., Comai, L. &
Dubcovsky, J. 2009. A modified TILLING approach to detect induced mutations in tetraploid
and hexaploid wheat. BMC plant Biology, 9(1): 115.

Umavathi, S. & Mullainathan, L. 2016. Induced mutagenesis in Chickpea (Cicer arietinum

(L.) with special reference to the frequency and spectrum of chlorophyll mutations. Journal
of Applied and Advanced Research, 1(1): 49–53.

Vakali, C., Baxevanos, D., Vlachostergios, D., Tamoutsidis, E., Papathanasiou, F. &
Papadopoulos, I. 2017. Genetic Characterization of Agronomic, Physiochemical, and
Quality Parameters of Dry Bean Landraces under Low-Input Farming. Journal of Agricultural
Science and Technology.

296
Vanhove, A.-C., Vermaelen, W., Panis, B., Swennen, R. & Carpentier, S. 2012. Screening
the banana biodiversity for drought tolerance: can an in vitro growth model and proteomics
be used as a tool to discover tolerant varieties and understand homeostasis. Frontiers in plant
science, 3: 176.

Viccini, L.F. & de CARVALHO, C.R. 2002. Meiotic chromosomal variation resulting from
irradiation of pollen in maize. Journal of applied genetics, 43(4): 463–470.

Vollman, J. & Rajcan, I. 2009. Oil crops. Springer, London.

Wang, H., Liu, Z., Chen, P. & Wang, X. 2012. Irradiation-facilitated chromosomal
translocation: wheat as an example. Plant mutation breeding and biotechnology. CABI, FAO,
Oxfordshire, UK: 223–229.

Wang, Z., Zou, Y., Li, X., Zhang, Q., Chen, L., Wu, H., Su, D., Chen, Y., Guo, J., Luo,
D., Long, Y., Zhong, Y. & Liu, Y.-G. 2006. Cytoplasmic Male Sterility of Rice with Boro
II Cytoplasm Is Caused by a Cytotoxic Peptide and Is Restored by Two Related PPR Motif
Genes via Distinct Modes of mRNA Silencing. The Plant Cell, 18(3): 676–687.
https://doi.org/10.1105/tpc.105.038240.

Wani, M.R., Kozgar, M.I., Khan, S., Ahanger, M.A. & Ahmad, P. 2014. Induced
mutagenesis for the improvement of pulse crops with special reference to mung bean: A
review update. Improvement of Crops in the Era of Climatic Changes, pp. 247–288. Springer.

Waterworth, W.M., Drury, G.E., Bray, C.M. & West, C.E. 2011. Repairing breaks in the
plant genome: the importance of keeping it together. New Phytologist, 192(4): 805–822.

Weil, C.F. & Monde, R.-A. 2009. EMS mutagenesis and point mutation discovery.
Molecular Genetic Approaches to Maize Improvement, pp. 161–171. Springer.

Weir, A., Omielan, J., Lee, E. & Rajcan, I. 2005. Use of NMR for predicting protein
concentration in soybean seeds based on oil measurements. Journal of the American Oil
Chemists’ Society, 82(2): 87–91.

Wertz, E. 1940. Über die Abhängigkeit der Röntgenstrahlenwirkung vom Quellungszustand

der Gewebe, nach Untersuchungen an Gerstenkörnern. Urban & Schwarzenberg.

White, M. 1978. Chain processes in chromosomal speciation. Systematic Biology, 27(3):

285–298.

Wijnker, E. & de Jong, H. 2008. Managing meiotic recombination in plant breeding. Trends
in plant science, 13(12): 640–646.

Wongpiyasatid, A., Hormchan, P. & others. 2000. New mutants of perennial Portulaca
grandiflora through gamma radiation. Kasestart Journal (Natural Science), 34: 408–416.

297
Wu, D., Shu, Q. & Li, C. 2012. Applications of DNA marker techniques in plant mutation
research. CABI Publishing.

Xia, T., Zhang, L., Xu, J., Wang, L., Liu, B., Hao, M., Chang, X., Zhang, T., Li, S.,
Zhang, H. & others. 2017. The alternative splicing of EAM8 contributes to early flowering
and short-season adaptation in a landrace barley from the Qinghai-Tibetan Plateau.
Theoretical and Applied Genetics, 130(4): 757–766.

Xiao, H., Jiang, N., Schaffner, E., Stockinger, E.J. & van der Knaap, E. 2008. A
retrotransposon-mediated gene duplication underlies morphological variation of tomato fruit.
science, 319(5869): 1527–1530.

Ya, H., Zhang, H., Feng, W., Chen, Q., Wang, W. & Jiao, Z. 2011. Implantation with low-
energy N+ beam changes the transcriptional activation of the expression sequence tags (ESTs)
from the transposable elements in rice. African Journal of Agricultural Research, 6(18):
4318–4326.

Yan, L., Fu, D., Li, C., Blechl, A., Tranquilli, G., Bonafede, M., Sanchez, A., Valarik,
M., Yasuda, S. & Dubcovsky, J. 2006. The wheat and barley vernalization gene VRN3 is an
orthologue of FT. Proceedings of the National Academy of Sciences, 103(51): 19581–19586.

Yetisir, H. & Sari, N. 2003. A new method for haploid muskmelon (Cucumis melo L.)
dihaploidization. Scientia Horticulturae, 98(3): 277–283.

Yuan, L., Dou, Y., Kianian, S.F., Zhang, C. & Holding, D.R. 2014. Deletion mutagenesis
identifies a haploinsufficient role for gamma-zein in opaque2 endosperm modification. Plant
Physiol, 164: 119–30. https://doi.org/10.1104/pp.113.230961.

Yuan, S., Liu, Y.-M., Fang, Z.-Y., Yang, L.-M., Zhuang, M., Zhang, Y.-Y. & Sun, P.-T.
2009. Study on the relationship between the ploidy level of microspore-derived plants and the
number of chloroplast in stomatal guard cells in Brassica oleracea. Agricultural Sciences in
China, 8(8): 939–946.

Zaman, S., Fitzpatrick, M., Lindahl, L. & Zengel, J. 2007. Novel mutations in ribosomal
proteins L4 and L22 that confer erythromycin resistance in Escherichia coli. Molecular
microbiology, 66(4): 1039–1050.

Zengquan, W., Hongmei, X., Jianping, L., Shibin, Y., Yan, F. & Zhongkui, X. 2003.
Application of heavy ion beams in induced mutation breeding and molecular modification.
Nuclear Physics Review, 20(1): 38–41.

ĩPLHĔNR $ 6DPHODN $ .R]\lowski, P. & Figlerowicz, M. 2014. Copy number
polymorphism in plant genomes. Theoretical and applied genetics, 127(1): 1–18.

298
 ADDITIONAL READING

IAEA, 1973. Neutron irradiation of seeds. Technical reports N° 141. Available at:
www.iaea.org/inis/collection/NCLCollectionStore/_Public/25/052/25052009.pdf

IAEA, 1977. Manual on Mutation Breeding. Second Edition. Technical Reports

Series N°. 119.

IAEA, 2004. Directory of Gamma Processing Facilities in Member States. Available

at: https://www-pub.iaea.org/books/iaeabooks/6914/Directory-of-Gamma-
Processing-Facilities-in-Member-States

IAEA, 2010. International Conference on Operational Safety Experience and

performance of Nuclear Power Plants and Fuel Cycle Facilities: 21-25 June 2010,
Vienna, Austria. Available at:
https://wwwpub.iaea.org/MTCD/Publications/PDF/Pub1716web-18398071.pdf

IAEA, 2014. Radiation Protection and Safety of radiation sources: International

basic Safety standards. Basic Standard Series. CD-ROM. Available at:
https://www-pub.iaea.org/MTCD/publications/PDF/Pub1578_web-57265295.pdf

IAEA, 2016. Safety of Research Reactors. Series No. SSR-3. Available at:
https://www-pub.iaea.org/MTCD/Publications/PDF/P1751_web.pdf

299
 Manual on Mutation Breeding
THIRD EDITION

Edited by Madeleine Spencer-Lopes, Brian P. Forster and Ljupcho Jankuloski

Co-published by the International Atomic Energy Agency (IAEA) and

the Food and Agriculture Organization (FAO) of the United Nations.

The previous (2nd) edition of the Manual on Mutation Breeding was published in 1977. After nearly
40 years it is time to up-date this manual as there have been several major developments in the
intervening years. There has been an upsurge in the use of induced mutations – for both classical crop
improvement and for functional genomic studies – and at the same time, there is an ever increasing
and urgent need to generate novel variation in improving crop production: more nutritious, hardy,
input use-efficient and productive crop varieties, to underpin ‘green’ sustainable food production for
the 21st century, especially in safe-guarding food security which is challenged by climate change,
hunger and world population growth.

This 3rd edition of the Manual on Mutation Breeding that you now peruse, describes advances in
plant mutation breeding, including basic irradiation techniques as well as chemical mutagenesis, in
both seed-propagated and vegetatively propagated crops. The manual provides comprehensive
overviews and guidelines for new high-throughput screening methods – both phenotypic and
genotypic – that are currently available to enable the detection of rare and valuable mutant traits and
reviews techniques for increasing the efficiency of crop mutation breeding. Most of all the manual
provides practical, hands on, methods in plant mutation breeding techniques, with clear illustrated
step-by-step protocols.

The Manual on Mutation Breeding is founded on the experience and knowledge of the staff of the
Plant Breeding and Genetics Section and Laboratory of the Joint FAO/IAEA Division – past and
present – and has been written in collaboration with external scientists who are internationally
renowned experts in specific fields. The combined contributions, backed by the richness of the
Mutant Variety Database of the Joint FAO/IAEA Division, represents a valuable resource for all those
interested in the resurgence of plant mutation breeding.

Print j-5109 E5 Cover Page 2.indd 1 22.05.18 10:13