Bacteriophages

BACTERIOPHAGES
Bacteriophages
JOHN DOUGLAS
M.Sc. Ph.D. D.Le. M.LBiol
Department of Applied Biology
Brunel University
Springer-Science+Business Media, B.Y.

© 1975 John Douglas
Originally published by Chapman and Hall in 1975.
Typeset by Tek Art
ISBN 978-0-412-12640-6 ISBN 978-1-4899-3418-5 (eBook)

DOI 10.1007/978-1-4899-3418-5
This title is available in both hardbound and paperback

editions. The paperback edition is sold subject to the condition
that it shall not, by way of trade or otherwise, be lent, re-sold,
hired out, or otherwise circulated without the publisher's
prior consent in any form of binding or cover other than that
in which it is published and without a similar condition
including this condition being imposed on the subsequent
purchaser.
All rights reserved. No part of this book may be reprinted, or
reproduced or utilized in any form or by any electronic,
mechanical or other means, now known or hereafter invented,
including photocopying and recording, or in any information
storage and retrieval system, without permission in writir.zg
from the Publisher.
Distributed in the U.S.A.

by Halsted Press, a Division
of John Wiley & Sons, Inc., New York
Library of Congress Catalog Card Number 75-1288
Preface
The contribution of bacteriophages to the development of

modern biology cannot be overestimated yet, sixty years
after their discovery, they are as remote and mysterious to
many scientists as they are to most laymen. This book
endeavours to remedy the situation: an attempt has been
made to provide, in readily comprehensible form, a nucleus
of information essential to anyone embarking on the study
of bacteriophages or using them in their work for the first
time. It shows the range of bacteriophage structure and
behaviour; it illustrates the role of bacteriophage in molecular
biology; it surveys the current state of the art; it presents the
medical and industrial aspects. Some simple experimental
procedures are given in sufficient detail for the beginner to
attempt them successfully. Other, more sophisticated pro-
cedures are presented so as to impart a feeling of intimate
reality without dazzling the reader with technical complexity.
I hope that young readers will forgive me for assuming
that they have some knowledge of bacteria, nucleic acids,
antibodies and isotopes. Likewise I would ask mature workers
to excuse the omission of cherished specialities. To have
included all these, valuable though they are, might have put
this book beyond the re ach of the phage-novices for whom
it is intended. Specific references, save a few of particular
interest, have been omitted. Other books on bacteriophage
v
vi BACTERIOPHAGES
provide them in abundance. They are all much larger and

more expensive than this book, which in no way contrives to
compete with them. On the contrary, it should stimulate the
student to explore them and guide hirn in his selection. That
is not to say that only simple, non-controversial aspects are
considered here; that would deny the reader his share of the
current excitement of bacteriophage research. Brief summaries
of some recent, far-reaching achievements and speculations
were considered a necessary part of this irreducible minimum.
I wish to thank all those persons and organizations who
have allowed me to use their illustrations; Veronica Phillips
for preparing specimens for photography; Fergus Kirkham
for some of the line drawings and Dr. Susan E. Smith for
assistance with Chapter 7 and reading the final draft.
Brunel University 1974 John Douglas

Contents
Preface page v
1 Introduction 1
2 Lysis 7
3 Structure and function of the viron 21
4 The latent per iod and the ec1ipse 49
5 Lysogeny 65
6 Survival of phages 77
7 The importance of phage in fundamental biology 88
Phage genetics and the structure of the gene. Transduction.
The wider aspects: A miscellany
8 Applied aspects 119

9 Conc1usion 127
Appendices 129
A. Units of measurement in electron microscopy 129
B. Media for bacteriophage studies 129
C. The principal phage collections in Britain 130
D. Selected bibliography 131
Index 134
VB
1 Introduction
Bacteriophages, usually abbreviated to 'phages', are viruses

that infect bacteria. The first account of a bacteriophage
was published by F.W. Twort in 1915. He had noticed that
some colonies of a bacterium, Micrococcus, had undergone
what he termed a 'glassy transformation' in that, instead of
their normal, opaque creamy-white appearance, they had
become clear, like glass. When he transferred a minute speck
of material from such a glassy colony to anormal colony that
to~, in a day or so, became glassy. The glassy appearance
could be transmitted from colony to colony indefinitely in
this way and, because Twort could find no bacterial cells in
the glassy colonies and deduced that they must have lysed
(dissolved), he called the phenomenon 'transmissible lysis'.
Material from a glassy colony was found to bring about the
glassy transformation at high dilution, even after it had been
passed through a filter whose pores were fine enough to
restrain the smallest bacteria. Heating the filtrate destroyed
its ability to bring about the transformation. Only colonies of
living bacteria could be lysed by it and during lysis, it was
shown, the lytic agent increased many hundredfold in
quantity. On the basis of this evidence Twort cautiously
suggested that the agent might be, among other possibilities,
a virus similar to those that infect plants and animals, which
had been discovered a few years previously. Little notice was
2 BACTERIOPHAGES
taken of Twort's work at the time.

The word 'bacteriophage', which means 'bacteria-eater',
was coined by F. d 'Herelle who seems to have rediscovered
the phenomenon independen tly in 1917. Unlike Twort,
d'Herelle had no doubt that it was due to a virus. Further-
more, he soon envisaged the possibility of using phage for
the treatment of bacterial infections in Man and animals.
This, unfortunately, is not yet possible. Even though
bacteriophages are able to annihilate huge populations of
pathogenic bacteria in minutes in vitro, all attempts to get
them to do so in vivo have been uniformly unsuccessful. The
reasons for this are not dear, even today. I t may be that
bacterial populations in vivo rapidly develop resistance to
phage; alternatively the phage, being itself an exogenous
substance, may be eliminated by the body's defences before
it can significantly affect the pathogen. D'Herelle nevertheless
succeeded in arousing world-wide interest in bacteriophage
research which revealed a veritable cornucopia of biological
discoveries and continues unabated at the present time.
During the 1920's and 1930's many of the fundamental
properties of bacteriophages were investigated. Contrary to
the belief held by d'Herelle, there was found to be not just
one bacteriophage capable of attacking all bacteria but many
different phages, each having a limited range of host bacteria
that it could infect and lyse. Normally the bacterial hosts of
a particular strain of phage are dosely related variants of a
single species. Any one strain of bacteria is usually susceptible
to infection by a group of phages but these may differ widely
among themselves in other properties. The physical and
chemical factors affecting the growth and death (more
correctly 'multiplication' and 'inactivation') of bacteriophages
were studied using the plaque assay technique which will be
described in full in the following chapter.
Although bacteriophages are too small for their shapes to
be resolved in even theoretically perfeet light microscopes, it
was nevertheless possible to observe phages as dancing points
of light in a high-power, dark-field microscope and thus to
count the number of particles liberated by the bursting of a
single infected bacterial cell. By testing the ability of phages
INTRODUCTION 3
to pass through ultrafine filters of graded pore size, estimates

of the dimensions of phages were made which subsequently
turned out to be remarkably accurate. Controversy over
whether phages were organisms or chemieals in solution,
however, continued for many years.
The 1940's saw two developments that gave renewed
impetus and direction to bacteriophage biology. Firstly, the
development of the electron microscope reached the stage
where bacteriophage particles could be seen in clear outline
for the first time, confirming their particulate nature beyond
dispute and revealing so me of their diversity of size and shape.
Secondly, whereas previous workers had tended to study in
isolation, using a variety of heterogeneous phages, a group in
the U.S.A. decided to concentrate on the phages of Escbericbia
co/i, in particular those known as the 'T-phages'. This was
both a good and a not-so-good thing; good because by
standardising their organisms and, to a large extent, their
methods communication and cross-fertilization of ideas were
facilitated; not so good because the rapid advances in
bacteriophage morphology, biochemistry and genetics that
gathered pace and continued to the present time relate almost
entirely to the phages of E. co/i. Practically all bacteria that
have been investigated have proved to be susceptible to one
or more phages and some of these, like the lipid-containing
Pseudomonas phage PM2 [8], the first shown to have any
major structural component other than nucleic acid and
protein, have revealed new and hitherto unsuspected
properties.
The 1950's could be called the 'Golden Age' of phage
biology. Phage studies played a major role in the discovery
that deoxyribonudeic acid (DNA) was the primary repository
of genetic information, and in the elucidation of the genetic
code. Although of minor (but significant) economic import-
ance in themselves, phages provided a model of behaviour
for research es whereby viruses with greater influence on
human affairs could be better understood. Through the
1960's to the present, phages have proved to be ideal material
with which to investigate the molecular mechanisms of
genetics. If one wishes to breed a better strain of wheat or a
4 BACTERIOPHAGES
faster race-horse, then one must experiment with wheat and

horses but these organisms are too complicated for the study
of fundamental processes. Over the last seyenty years the
search for a model system easily susceptible to investigation
at the molecular level has led through fruit flies (Drosophila),
fungi (Neurospora) and bacteria (E. coli) to bacteriophages,
which include the smaUest biological entities endowed with
the power of self-replication. Phages, indeed, do little other
than replicate and therein lies their attraction. Unencumbered
by the need to maintain complex life-support systems since
these are borrowed from the host ceU, the phage genome can
comprise up to half the total weight of the particle. Thus
there is little to obstruct experimental procedures designed to
affect the genome directly and at the same time the very
limited phenotype aUows isolation of specific aspects for
study. Thus genome and phenotype can be related in a way
that is not possible with cells. The bacterial hosts are very
conveniently maintained in vitro under controlled conditions
requiring relatively little in terms of time, space and labour,
in contrast to the demands of plants and animals. Thus within
the space of a day one phage particle may be multiplied into
biIlions; these can be treated with a mutagenic chemical and
the rarest of mutants detected by examining the entire
population - aU on a few square feet of bench space. The
ability to handle huge populations economicaUy is a conse-
quence of bacteriophages' small size yet, if required for
biochemical analysis or manipulation, modern fermenter
technology permits gram quantities of bacteriophages or
their isolated nucleic acids, pro teins and enzymes to be
prepared.
For most purposes, until comparatively re cent years, the
phage phenotype consisted of the macroscopic appear.ance of
the phage-infected bacterial culture. However, other biological
entities may produce similar appearances and it is important
for the phage student to be aware of these. Both Bdellovibrio
bacteriovorus and colicin K can produce local clearings in
layers of otherwise confluent bacterial growth and both can
cause the turbidity of broth cultures of bacteria to be reduced.
But there the resemblance ends. Bdellovibrio is a ceIl, able to
INTRODUCTION 5
move, respire, increase in size and reproduce by fission.

Colicin K is simply a pro tein moleeule unable to reproduce
itself at all. Bacteriophages, as Figure 1.1 shows, are
intermediate in size between these two. They are also inter-
mediate in complexity. Are they to be regarded then as the
present-day representatives of an intermediate stage in
evolution? It is true to say that, at the lower end of the
:.:::0
lJ T3Q T4~
T5\ T6~
.~
<:
.~
CII
13
~ T7Q
~\
0
S13
0
.X174
Figure 1.1 Relative Sizes of a Bacterium (E. Coli) and an

Assortment of Biological Entities (Bdellovibrio, Bacteriophages and
a Colicin) wh ich attack it.
Colicin K is beyond the resolution of current electron microscopes.
The size depicted is based on indirect estimates.
6 BACTERIOPHAGES
biological size range, smallness necessitates simplicity of

structure and tends to be thought of as primitive. But was
the small size of phages arrived at by the elaboration of an
even smaller molecule, did they arise by the simplification
of a cell down to the barest requirements for a parasitic
life, or did they travel some evolutionary route that is
exdusively their own? Some dues may be gleaned from the
following pages but until the extent of the bacteriophage
world is more fuHy explored it would be best to reserve
judgement on this question.
2 Lysis
Lysis, that is the dissolution of the bacterial cell, is the

essential phenomenon whereby the presence of phage is
detected. Some filamentous phages simply leak out of the
cell over aperiod without killing it; certain defective phages
multiply in the infected cell but cannot lyse it; some phages
termed 'temperate' may integrate themselves into the
bacterial cell and lyse it only when induced to do so by
special treatments. In such cases phage is detected only with
difficulty and if all bacteriophages behaved in any one or
other of these ways they might not yet have been discovered.
Fortunately the virulent, lytic phages are very common. They
lyse the cell in which they have multiplied by causing the
production within it of an enzyme, lysozyme, that attacks
the murein of the cell wall, weakening it so that it bursts and
liberates the phage within. In so me cases the dissolution of
the wall is very nearly complete, in others a substantial
amount of debris remains. Lysis can be detected in several
ways.
2.1 Lysis on solid media

A plate of nutrient agar can be inoculated heavily and
uniformly with host bacteria by flooding its surface with a
broth culture, removing the excess liquid and allowing the
plate to dry. When incubated, a 'lawn' of bacterial growth
7
8 BACTERIOPHAGES
will develop. If drops of concentrated phage suspension are

placed on the plate soon after the inoculum has dried, the
lawn that develops will have bare patch~s corresponding to
where the drops lay. If instead a few phage particles are
mixed with the inoculum before it is applied, each particle
will produce a hole in the lawn in the following way: it first
of all infects a single cell, multiplies within it and causes it to
lyse, liberating many progeny phages; these diffuse away
from the original site, infect other cells and lyse them and
the process is repeated so that the area of lysis, known as a
'plaque', * grows to be visible eventually to the naked eye.
These simple techniques have their uses but there are
problems in ensuring the even distribution of inocula and the
time they take to dry. More reproducible results can be
obtained if the host bacteria and the phage are embedded in
a thin layer of agar. Phages can diffuse through agar gels to
form plaques. If a low concentration, for example 0'4 per
cent, of agar-agar is used the gel offers less resistance to
diffusion and plaques are larger. Such agar, known as 'semi-
solid', 'soft' or 'sloppy' agar, is normally employed as a thin
layer (c. 1- 2 mm) poured over a thicker layer of normal
solid agar wh ich provides, by diffusion, the main source of
nutrients for the bacteria in the sloppy layer.
By counting the plaques that arise from a known volume
of phage suspension one can estimate the number of phage
particles in it. This is known as the 'plaque assay technique'
and the full procedure is shown diagrammatically in Figure
2.1. It is simple, rapid and accurate. It is probable that
without the plaque assay phage research would hardly have
advanced beyond Twort and d'Herelle's original observations.
A similar technique for animal viruses was not developed
until1952 and one for plant viruses is stilliacking, which
goes a long way towards explaining why these branches of
virology lag so far behind phage.
Plaque formation by phages is not 100 per cent efficient;
that is to say, not every particle added to a lawn of bacteria
* In some virologists' vocabulary this word can also be used as a transitive verb
meaning "to cause to make plaques". Others use the verb "to plate" synony-
mously. Unfortunately microbiologists in general use "to plate" meaning simply
"to inoculate (bacteria or fungi) into a dish 01' agar".
LYSIS 9
forms a plaque. So me may infect dead bacteria for example,

in which case they are unable to develop further; others
encounter random hazards during manipulations that damage
them. Und er favourable circumstances some phages may
approach 100 per cent plaquing efficiency but others have an
intrinsically low efficiency of 10 per cent or less because of
innate defects.
O·lml O ·lml
Re ult
GGGG
()(~ .
( )
", ./
t)
Figure 2.1 Phage Assay by the 'Sloppy Layer' Plaque Technique
The phage suspension was serially diluted by transferring 0 .1 ml
portions through 9 .9 ml portions of diluent, making 10- 2 , 10-4 ,
10- 6 and 10- 8 dilutions of the original. Four sterile, empty test-tubes
10 BACTERIOPHAGES
were placed in the waterbath at 46°C ± 1°C and 0.1 ml of the

corresponding phage dilution added to each. To each was then added
0.2 ml of an overnight broth culture of the host bacterium and 3 ml of
'sloppy agar' (broth + 0.4 per cent agar-agar), which had been melted
and cooled to 46_o C in the waterbath. The contents of each tube
were then weIl mixed by swirling, poured over the surface of a plate
of normal, solid agar and allowed to set level. The plaques were
visible after six hours. The titre of the original suspension, calculated
from the 10-6 dilution, is 8 x 10 X 106 = 8 X 10' plaque-forming
units (p.f.u.) mr 1 .
Plaques have other uses besides assay. The plaque character-

istics are often useful in distinguishing different phages. The
size may vary within limits for any particular phage but
whereas some phages typically produce plaques of 5 mm
diameter or more, others rarely exceed 1 mm. The shape of
most plaques is circular but mutants producing sectored or
'star' plaques have been described. The plaque margin may
be sharp or diffuse and there may be a zone of turbidity
surrounding it (Figure 2.2). If a phage is plaqued using a
mixture of two different strains of bacteria the plaques will
be clear only if both strains are lysed. If only one of them
is lysed the plaques will be turbid due to the growth of the
other. This 'mixed indicator' technique is useful for isolating
host-range mutants. Sometimes a plaque is seen to have a
colony of bacteria exactly in the centre. This may be a
'lysogenic' colony that secretes phage to which it is immune
but the rest of the lawn is not.
The phages in a plaque usually constitute a clone, a pop-
ulation of genetically uniform individuals that have arisen
from a single individual by vegetative reproduction. Thus a
pure culture may be obtained from a mixture of phages by
plaquing it, taking a few particles from a well-isolated plaque
by stabbing it with a wire, and using them to initiate a new
culture.
An assay plate showing almost confluent plaques may be
used to prepare a concentrated phage suspension. The sloppy
layer is scraped off, macerated with water or other diluent
and the agar and bacterial debris removed by low speed
centrifugation. Suspensions with a titre of over 1012 plaque-
LYSIS 11
Figure 2.2 Diffusible Lysin

The plaque on the left (Staphylococcus aureus phage 42D) has a
weIl defined edge. That on the right (Streptococcus lactis 0712) is
surrounded by a halo of partially lysed ceIls, caused by the soluble,
lytic enzyme that diffuses from the centre faster than the phage partides
themselves Both plaques were grown in sloppy agar by the double-
layer technique and photographed at a magnification of 40x after
24 hours. The background lawn can be seen in each case to be
composed of crowded colonies of bacteria; the individual cells
cannot be seen at this magnification.
forming units (p.f.u.) per ml ean often be prepared by this

method. It is not always neeessary to remove the sloppy
layer. Adequate titres may sometimes be obtained simply by
washing its surfaee with a few ml of diluent.
Phage plaques are often diseernible, albeit with some
diffieulty, after as little as five hours ineubation. Surprisingly,
they may not inerease in size after that time even though
they become mueh clearer. A few eombinations of phage,
host and medium result in plaques that appear to grow
indefinitely but these are rare. No satisfaetory general
explanation as to why plaques stop growing has been advaneed.
Depletion of nutrients for baeterial energy requirements and
biosynthesis and the aeeumulation of toxie metabolites may
explain some, but not all eases. Mieroseopy of the plaque
margin is interesting. The light mieroseope often reveals
12 BACTERIOPHAGES
........
.... 10 8
\
\
"ci.
...:
//
50i
:1
...// \
./
o 2 3 4
Hours post-infection
Figure 2.3 Lysis in a Broth Culture

This shows what may be expected when sufficient phage is added to
a growing culture to infect about 1 per cent of the cells. The growth
of an uninfected (control) culture (dotted line) levels out as the
medium becomes exhausted. The infected culture (pecked line)
appears to begin similarly since only about 1 per cent of the cells
are affected; this level of infection will hardly register at all on a
nephelometer (Fig 2.5 B). After about an hour, however, these cells
burst and liberate enough phage to infect the other 99 per cent.
Growth then stops and eventually the whole culture dears. The
two cydes of infection are revealed more dearly if the number of
plaque-forming units (p.f.u., = phage partides plus infected cells) is
simultaneously followed (solid line). The second 'step' is less sharp than
the first since the progeny of the first cyde of infection are not all
released at the same time and thus the second cyde is not synchronous
in all cells. No third cyde is possible since there are no host cells left.
Although this simple experiment is instructive for the beginner, it
yields little precise information of value to the phage biologist.
Firstly the presence of uninfected cells at the end of the first cyde
means that there is a loss of p.f.u. due to two or more phages being
adsorbed onto the same cell; thus one does not know how many
were released. Secondly the continued growth of the uninfected
cells, superimposed on the growth of the phage, confuses the
biochemical picture. The more sophisticated 'one-step growth'
experiments, in which a single cyde of replication is isolated for
study, are shown in the following figure (2.4).
LYSIS 13
-A
++++ B
Burst size
;-._____L_a_t_en_t_p_er_i_Od_ _ _+-_-'-................. .L
Eclipse
I.j. ...---phase
5 10 15 20 25
Minutes post-infrction
i
i
i
Figure 2.4 One-step Growth Experiments
A. Normal Lysis (heavy line). A broth culture of the host
bacterium, growing in the logarithmic phase, is inoculated with
phage at an input ratio of one phage per 10 cells to ensure that
each cell that is infected is infected with only a single phage since
multiple infection may give a different result. After 5 minutes the
mixture is diluted in broth containing specific antiphage serum. The
serum inactivates any phage still free; dilution itself reduces the
probability of further adsorption so that one is dealing thereafter
with a culture in which infection is nearly synchronous. Five minutes
later the culture is diluted again to inactivate the antiserum. Sam pies
are removed for plaque assay every few minutes.
Initially each phage gives one plaque. When the phage adsorbs
onto and infects a cell, that too will give one plaque. The number
of plaque-forming units per cell remains unity during what is called
the 'latent period' up to the onset of lysis. When a cell Iyses it
releases a large number of phages, each of which can produce a
plaque, so that the number of p.f.u. per cell shows a massive
stepwise increase. The relative increase, representing the average
number of phages released from one cell, is known as the 'burst
size'. Because of the great dilution of the original mixture, the
probability of loss of p.f.u. by multiple adsorption is very small.
B. Premature, artifically-induced lysis (barred line). This experiment
is done to find out what is going on inside the infected cell during
the latent period. The technique is similar to that in A. but when
sampies are taken for assay they are first treated so as to break open
14 BACTERIOPHAGES
the cells and liberate any phages they contain. Of the many methods
of breaking open cells without inactivating the phages they contain,
shaking with chloroform has been most widely used. It is very
effective with E. coli and similar enteric bacteria but with others,
staphylococci for instance, it is totally ineffective and other methods,
such as mechanical rupture or explosive decompression, must be
used. Within a few minutes of infection the phage virtually disappears
from the culture; one cannot extract p.f.u. from the cells by any
method. This state of affairs persists until shortly before normal
lysis would occur. The p.f.u. increase rapidly in number until they
reach the level that normal lysis would produce. The period during
which no phage is extractable is known as the 'ec1ipse phase'.
grossly swoHen and distorted ceHs. Electron microscopy may

reveal very few phages at the margin, suggesting that the
latter stages of plaque growth are due, not to phage but to
the diffusion of excess lytic enzyme from the central area.
A few unlysed ceHs are usuaHy to be seen scattered through-
out the plaque.
2.2 Lysis in liquid media

This can be the most dramatic manifestation of phage activity.
If a growing broth culture is inoculated with a few particles of
a phage to which it is susceptible ("homologous" phage) the
culture appears to continue growing normally for so me hours
(or minutes, depending on the 'input ratio' of phages to
bacteria) then, in the space of as little as perhaps ten minutes,
the wh oIe culture be comes glass-clear as the eells Iyse. What
one is witnessing is the progressive multiplieation of the
phage, impereeptible at first sinee only a few eeHs are affected
but, as they lyse and liberate progeny phage, the proportion of
infected eeHs rises exponentiaHy until the last eyde of lysis
involves all that remain (Figure 2.3) so that clearing is sudden
and complete. Figure 2.5 shows instruments used to quantify
lysis in liquid media more precisely than purely visual
observation aHows. The absorptiometer measures turbidity
as the impedance offered to light passing through the suspen-
sion of organisms; the nephelometer measures turbidity as the
LYSIS 15
A B
b
a a
d e
Figure 2.5 Instruments for Measuring the Turbidity of Bacterial

Suspensions
A Absorptiometer. A beam of light, collimated by lenses a, a,
passes through the suspension in cuvette band impinges on photo-
cell c, thereby generating an electromotive force registered on
galvanometer d. In modern instruments the galvanometer scale is
usually. calibrated in units of extinction relative to a control
solution. The relationship between cell concentration and extinction
obeys the Beer-Lambert law fairly closely over a wide concentration
range.
B Nephelometer. The path of a single ray is shown. It passes
through the aperture of annular photocell a into culture rube band
strikes a bacterium at c which deflects it through 90°. It is then
reflected by hemispherical mirror d onto the surface of the photocell
and generates an electromotive force which is registered on
galvanometer e. The galvanometer scale is calibrated in arbitrary units
relative to a stable standard (usually a ground glass rod) and the
reading is known as the 'reflectance'. The relationship between cell
concentration and reflectance is complex.
The optical properties of bacterial suspensions vary with the
species and physiological condition. Both nephelometers and
absorptiometers require a new calibration curve for each set of
circumstances. Rectilinear relationships are not always obtained,
particularly at high bacterial concentrations where forward scattering
and reabsorption become significant. Curvilinear relationships may be
perfectly adequate for practical purposes. In suspensions that are
undergoing lysis and contain swollen and ruptured cells, the transla-
tion of optical data into cell numbers is of dubious value; these
instruments are nevertheless very useful for determining the times of
onset and completion of lysis and relative rates of change.
16 BACTERIOPHAGES
amount of light deflected sideways by the cells. The choice of

instrument depends on many factors. One advantage of
nephelometers is that they ar~ designed, as a rule, to accept
the culture in its normal tube instead of requiring it to be
transferred to a special cuvette as in most absorptiometers.
2.3 Lysis of individual cells

Lysis of individual ceils by phage may be observed with the
light mkroscope provided steps are taken to immobilise them.
This is conveniently done in the following way. A culture is
infected with phage and the turbidity followed with one of
the instruments mentioned above. When the turbidity ceases
to rise, indicating that massive lysis is imminent, a small
loopful (c. 0.001 ml) is removed and placed on a thin layer
of agar on a slide. The liquid soaks into the agar in seconds,
leaving the infected bacteria perched immobile on the surface.
Figure 2.6 Assay by 'Dilution to Extinction'

In the exampIe shown, 1.1 mI of a suspension of Streptococcus
lactis phage 0712 was added to the first 9.9 mI portion of litmus
milk in tube No. 1. 1.1 mI of this mixture was then transferred to
tube No. 2 and so on, making serial decimal dilutions down to
10- 12 . One drop of a IJO dilution of an overnight broth culture of
Streptococcus lactis strain 712 was then added to each tube and they were
incubated at 30°C overnight. Next day tubes 1 - 7 still showed the
LYSIS 17
characteristic purpie colour of uninoculated litmus milk but tubes

8 - 12 showed the pink acidic surface and bleached lower part
typical of the growth of Streptococcus lactis. Tube No. 7 contained
1 ml * of the previous, 10-6 , dilution. As it is the last tube in the
series to show inhibition of bacterial growth, that 1 ml must have
contained at least one phage particle. The titre of the original phage
suspension was therefore in the order of 10 6 particles per ml. This
technique is simple and instructive for beginners but it is
arithmetically 'messy', statistically naive and lacking in precision. With
even 50 replicates at each dilution its accuracy would still be only about
that of a single plaque assay. It is useful for making rough comparisons
of the amounts of phage in different preparations under circumstances
where plaque assay is not practicable but its main use is not as an assay
but for propagating phage from minimal inocula, by taking the last
tube be fore extinction to make a stock. The National Collection of
Dairy Organisms recommends it for this purpose. In the example
shown, the litmus milk in tube No. 7 would have to be decaseinated by
acidifying it with dilute lactic acid, centrifuging to remove the casein
thereby precipitated and carefully neutralizing the supernate for stock.
The use of litmus milk is usual for dairy phages. Dilution to extinction
can, of course, be used for any phage having an overt effect on the
host culture.
* It received 1.1. ml but 0.11 ml was removed in making the next dilution so
the volume remaining was actually 0.99 ml. So long as one is using the same
portion of liquid for making and testing a dilution, it is not possible to choose
transfer and diluent volumes that will make both the dilution-ratio and the
residual volumc convenient round figures. By calIing 0.99 ml 1 ml, an error of
I per cent is introduced. Considering the low intrinsie accuracy of this assay
and the uscs to which it is put, this error is not significant.
They can be covered with a coverslip and examined con-

tinuously until they lyse, using either a dark-fieId or phase-
contrast microscope. A high magnification, 1000 X at least,
is required if details of the rupture of the eells are to be
pereeived but if all one wishes to see is the sudden disappear-
anee of eaeh eell 400 x is quite adequate. Cells in the terminal
stages of infeetion are little affeeted by small ehanges in
temperature and the time at whieh the broth eulture in a
waterbath at, say, 30°C lyses is found to eorrespond weH
with the disappearance of cells in the sam pIe under the
mleroseope.
18 BACTERIOPHAGES
2.4 Stock lysates

A lysate may serve as a stock of phage for use in subsequent
experiments. The highest possible titre is desirable. This is
usually obtained by the sloppy agar method described above
but varies somewhat from phage to phage. Lysis of a
stationary broth culture can give yields of 109 p.f.u. mr 1 or
more and this may be increased by a factor of 10-1000 if
the culture is aerated in a bubbier tube. A typical bubbier
tube is shown in Figure 2.7. It is normally used for volumes
up to 25 ml but there is no limit to potential escalation.
The lysate, however prepared, is centrifuged at low speed
or passed through a bacteriological filter or both, to remove
unlysed bacteria and debris. Filters of the 'Millipore'
membrane type are preferable to thick pads since they retain
less phage. Sometimes chloroform is added as a preservative
but storage at low temperature (+4°C) is usually adequate.
Freezing is not normally necessary and may even result in
loss of titre. There is much to be said for dispensing the stock
phage into small ampoules containing about 1 ml or such
other quantity as will suffice for one experiment. On this 'use
and discard' system the possibility of the growth of contam-
inants in an opened stock bottle between experiments is
eliminated.
2.5 Bdellovibrio (Figure 1.1)

Bdellovibrio is not a phage but it i.s included here because it
is bacteriolytic and can produce plaques in a lawn of bacteria
like a phage. It is a very small bacterium, just recognisable as
such in the light microscope, which parasitises other bacteria,
usually gram-negative rods, and causes them to lyse. It can be
plaque assayed like a phage but since plaques of Bdellovibrio
take 3-5 days to appear there should never be any confusion
of the two. The name is derived from Latin words meaning
'leech' and 'shaker', which describe its modes of life and
locomotion very weIl.
It swims very rapidly with its single polar flagellum; the
initial contact with the host is violent and the host cell is
said to recoil. The parasite attaches itself by its front end and
LYSIS 19
Im----Rubber bung
Aluminium cop
1---fJ~-Posteur pipette
Figure 2.7 BubbIer Tube

BubbIer tubes are widely used for propagation of phages of aerobic
bacteria. A high rate of aeration ensures maximum available energy for
growth and reduces the accumulation of toxic metabolites. Several
tubes of the pattern shown may be worked simultaneously off a small
aquarium aerator. A cotton-wool filter may be inserted in the airline or
in the pasteur pipette to protect the culture from contamination.
Exhaust air escapes under the rim of the cap; it will contain phage and
bacteria in an aerosol; if thls is undesirable (when a patp.ogen is used,
for instance) the pipette may be introduced through a iirm cotton-wool
plug but convenience is lost.
squeezes into the host eell through a hole sm aller than its
own diameter whieh it makes, presumably, by enzymie means.
Onee inside, Bdellovibrio appears to draw nourishment from
the eytoplasm without aetually penetrating it sinee, in
eleetron mierographs of seetions of infeeted baeteria, it ean
be seen between the eell wall and the membrane. Cell division
oeeurs within the host, whieh eventually lyses. Bdellovibrio
20 BACTERIOPHAGES
can be maintained in broth cultures of host bacteria and

separated from them by means of a 'Millipore' filter of 0.44
micron size, through which the parasite can pass but normal
bacteria can not. Bdellovibrio bacteriovorus and host strains
of E. coli are maintained in the National Collection of
Industrial Bacteria (see Appendix C).
3 Structure and function
of the virion
The infectious phage particle, like the particles of other

viruses, is called a 'virion'. Figure 1.1 shows in outline the
virions of a range of phages of E. eoli ('coliphages'),
illustrating the diversity in size and shape of the virions of
phages of a single bacterial species. On analysis, practically
all phage virions are found to consist of protein and nucleic
acid only. The nucleic acid is cornrnonly said to be
exclusively either DNA or RNA but this rnay not be strictly
true. One coliphage, T5, contains rnainly DNA but probably
some RNA-like material as well [13] .
The elucidation of phage structure and its correlation
with function is a slow process. Coliphage T2 has probably
been investigated in more detail than any other but is not to
be regarded as typical in any way. It will nevertheless be
advantageous to study it first and use its properties as
reference points enabling the virions of other phages to be
seen in perspective.
3.1 Coliphage T2
This virion consists of a polyhedral head approximately
100 nm x 80 nrn and a tail about 100 nm long joined to it
by a short 'collar piece'. At the distal end of the tail is a
21
22 BACTERIOPHAGES
roughly hexagonal 'end plate' bearing at each corner a short

'tai! pin' and a long, jointed fibre. The main part of the tail
is eomposed of two eoncentrie tubes known as the 'eore' and
'sheath' respectively. The head comprises an outer protein
'membrane' enclosing a eentral mass of tightly packed DNA.
The manner of packing of the DNA is still in dispute. X-ray
diffraetion data suggests that it is arranged in parallel bundles;
some recent electron micrographs [12] of partially disrupted
phage heads indieate that it is wound in a ball, as if on a
spindIe. The two viewpoints are not mutuaUy exclusive. What
is beyond dispute is that packing must be orderly, not
haphazard. All the evidenee, microscopical, biochemical and
genetic, shows quite dearly that T2 phage DNA has the form
of a single strand, an unbranching linear moleeule more than
500 times as long as the head that eontains it, and that during
infeetion the DNA passes along the narrow central channel of
the tail eore. A very special packing arrangement is surely
neeessary for this to oeeur smoothly, without knotting or
tangling.
There is evidence that so me phage DNA's, under certain
conditions, are eircular. The word 'circular' is not used in the
striet geometrie sense but implies that the ends of the linear
mole eule are joined so that it forms a closed loop, which is
endless, like a cirde. Phage lambda (A) is one such phage. In
the virion its DNA is linear but on entering the host eeH the
ends join to form a cirde (Figure 5.2). The DNA of phage
0X174, however, is circular both in the virion and in the
host cello It has been suggested that the DNA of T4 (a dose
relative of T2) must circularise in the host cell in order to
replicate. Evidence for this view at the moment is tenuous.
How does the T2 virion find and attack a eell? In aqueous
medium the virion is constantly bombarded by moving water
moleeules that keep it in astate of continuous rapid motion.
The tai! fibres can be envisaged as flailing in all directions
under the same influence. When it chances to eome within
range of a suitable host, one of the phages tail fihres will stick
to it. As the phage and baeterium are then kept in dose
proximity by this attachment it is a matter of moments only
before all the fibres have adhered. The phage is then said to
STRUCTURE AND FUNCTION or THE VIRION 23
have 'adsorbed' to the bacterial ceIl, reversibly at first but

irreversibly soon afterwards. The end plate of the virion
attaches to the bacterial cell wall and the tail sheath contracts
so that the tail core penetrates the wall which may be
weakened by enzymes contained in the core canal. The
nucleic acid of the phage is then injected into the lumen of
the cell through the hollow tail.
Figure 3.1 Coliphage TZ

Negatively stained. (Courtesy of Dr. M. Perutz, Medica1 Research
Council Laboratory of Molecular Biology)
24 BACTERIOPHAGES
I--sonm-L
Head
(A compact mass of
1
100nm
DNA enclosed by
a protein membrane)
:=
Collar
Sheath
{ Core
-,
~ Corecanal------~----~.I~ 100nm
Fibre
End-plate
_I
Tail pin
Figure 3.2 Anatomy of the Coliphage T2 Virion

The prineipal parts are shown in their normal relation to one another.
The head is depieted so as to emphasise its polyhedral nature; the
geometry is based on its appearanee in eleetron mierographs but
theoretieal eonsiderations suggest that the true shape is more eomplex
(inset). The tail is shown in seetion revealing the helieally eonstrueted
sheath around a hollow eore. Dimensions are approximate. Older
estimates based on filtration and sedimentation are mainly of historieal
interest now but dimensions in eleetron mierographs also vary. This
eould be due to strain differenees but is more likely due to the teeh-
niques of preparation. During drying the head flattens and spreads to a
varying degree. The proteinaeeous eomponents eolleetively eonstitute
the 'eapsid' of the virion.
STRUCTURE AND FUNCTION OF THE VIRION 25
A B
Medium
--Bacterio! ce!! wal!
Lumen of the cel!
c D
Figure 3.3 The Mechanies of Infection by Coliphage T2

A. Free phage. B. Adsorption by taiI fibres. C. TaiI pins engage cell
wall. D. Sheath contracts, core penetrates celI wall and nucleic acid
moIecuIe is injected.
26 BACTERIOPHAGES
A B C
•
~~1
~
• •
• •
• •
•
Figure 3.4 Adsorption of Coliphage T5 to Purified Receptor In
Vitro
A. Coliphage T5. B. Receptor moleeules isolated from the ce!l wall
of E. coli B. C. Mixture of A and B after incubation. (based on e!ectron
micrographs by W. Weide! et al)
100
0-
.
..
50
.c
'"o
.....
a.
."
.I>
~
."
o
C
~
10
o 10 20 30
Minutes after mixing
Figure 3.5 The Kinetics of Adsorption

A suspension of Streptococcus lactis 712 containing about 108 cells
per ml was made in broth and about 10 7 p.f.u. of phage 0712 per ml
was added. 1 ml portions were removed at once and at intervals
thereafter, diluted 1: 100 in chilIed broth (to prevent further adsorption
and extend the latent period of the phage), the bacteria and adsorbed
phage removed by centrifugation and the unadsorbed phage in the
supernatant determined by plaque assay. (B. Cartwright, unpublished
data).
Adsorption is very host specific and depends on the
presence in the cell wall of specific receptor sites. In the
case of another phage, T5, the receptor has been isolated. It
is a lipoprotein molecule of high molecular weight. When
phage T5 and its receptor are incubated in vitra, electron
microscopy shows that the receptor molecules become
attached to the tip of the tail. In phage T4 the tail fibres
are wound tightly around the tail and cannot function in
tryptophan-free media. Tryptophan allows the fibres to
unwind.
Many phages have a very definite requirement for divalent
cations particularly calcium, which promotes adsorption and
without which infection may fail to occur at all. This explains
why they are inhibited by, for example, citrates and oxalates
which form insoluble compounds with calcium in the
medium and render it unavailable. Ethylene-diamine-tetra-
acetic acid (synonyms: EDTA, 'Versene') chelates calcium
and is highly inhibitory to calcium-requiring phages.
Magnesium may be used as areplacement for calcium by
so me but not all phages with this requirement.
The contraction of the T2 tail sheath, driving the core
through the cell wall, is a most rem ar kable process that has
been intensively studied. It is very doubtful whether
mechanical force alone is sufficient to effect penetration. If
a large number, say 100 virions, are allowed to adsorb simul-
taneously onto a single ceIl it lyses immediately. This is
known as 'lysis from without' (phages normally escape from
the host ceIl by 'lysis from within' but the term is hardly
ever used). It suggests the presence of an enzyme similar to
lysozyme in the tip of the phage tail to ass ist penetration.
The 'trigger' mechanism that controls sheath contraction
can be set off artificially by treating the phage with either
hydrogen peroxide or zinc cyanide. The natural stimulus is
unknown. The sheath of T2 is composed of 144 comma-
shaped protein subunits arranged in a helix having twelve
turns, each of twelve subunits. On contraction a rearrange-
me nt occurs giving six turns of 24 subunits, thus reducing it
to half its originallength. This is an energy requiring process.
The tail contains 144 molecules of ATP which are converted
28 BACTERIOPHAGES
to ADP during contraction. I t also contains calcium and
folie acid, whose function is uncertain. The passage of the
nucleic acid contained in the phage head through the hoHow
tail co re into the bacterial ceH deserves detailed consideration.
Table 3.1 The essential strategy of Hershey and Chase's (1952)

experiments.
Expt.l Expt.2
Bacteriophage propagated
in medium containing
radiolabelled . . . Phosphorus (p 32 ) Sulphur (S35)
Radiolable appears in
the phage's . . . Nucleic acid Protein
When such a phage is used

to infect bacteria, then
sheared off and the cells
separated from the medium
by centrifugation, most of
the label is found in the . .. Bacterial cells Supernatant
Hence, at the moment of

shearing, the labelled
material was. . . Inside the cells Adhering to the
exterior of the
cells.
It was established by Hershey and Chase in 1952, in a

series of experiments that has been widely and, according
to one author [16], frequently wrongly cited, that shortly
after infection it is possible to remove most of the phage
protein but hardly any of the phage nucleic acid from the
infected bacterium by vigorous mechanical agitation in a
'Waring Blendor', an instrument resembling a domestic
liquidiser. The subsequent multiplication of the phage is
unaffected. Clearly, the protein can be removed since it
comprises the head sheath and tail which remains attached
to the outside of the cel!. The nucleic acid cannot be
Figure 3.6 Adsorbed Phage

Part of a section of Bacillus subtilis showing three phage SP50tJ virions
adsorbed to the wall. The head of the central phage appears dark since
it still contains nucleic acid; that on the right has lost its nucleic acid
and appears 'ghostly'. (Courtesy of Dr. R.G. Milne and Plant Pathology
Laboratory , Rothamsted)
removcd by agitation ('sheared off') because it has entered

the cdl. This interpretation is supported by electron micro-
scopy. Phage heads are usually dark in electron micrographs
because their tightly packed nucleic acid is not easily
penetrated by electrons. Phages whose heads have lost their
nucleic acid appear light in electron micrographs and are
known as 'ghost phages'. Adsorbed phages freguently appear
30 BACTERIOPHAGES
ghostly. If a phage suspension is suddenly diluted in hypo-

tonie medium the resultant osmotic shock liberates the nucleic
acid into the medium. However it is very doubtful whether
any such effect could propel the nucleic acid through the
tail and into the host cell; in osmotic shock the nucleic acid
is presumed to be lost through cracks in the head. N either
is contraction of the phage head sheath a possible propulsive
mechanism since the diameters of free, intact phages and
adsorbed ghosts are not significantly different. The most
satisfying explanation of the movement of the nucleic acid
is provided by the following thermodynamic argument.
If a smalliength of nucleic acid protrudes from the phage
tail into a large space such as the lumen of a bacterial cell, it
is free to move under the influence of bombardment by
rapidly moving water molecules. The only restraint on its
movement is the fact that it is still attached to the rest of it
contained within the phage. Thus the end can acquire
momentum and, as it reaches the limit of its travel, that
momentum will be expressed as a tug on the rest, pulling
more of it through the tail. Were the phage head and the
bacterium the same size, similar movements inside the
phage head would puH it back and an equilibrium might be
established. The head being so much smaller, however, the
nucleic acid is restrained by its own dense packing which
ensures that it cannot acquire much momentum and thus
quickly enters the larger space. This explanation is satisfying
because it is camplete in itself and no special propulsive
organelle or chemie al reaction need be postulated; transfer is
the inevitable consequence of the observed structures. There
is, nevertheless, the problem of explaining how the first bit
gets through to pull the rest. Leaving aside the matter of the
nature of the force that propels the nucleic acid moleeule,
a model is needed to show how the strand of nucleic acid
may be packed tightly in the head yet in such a way that the
end can find the smaH opening into the hollow tail and the
rest of this extraordinarily long moleeule can follow it in a
minute or so, as it is known to do. On the first point, it may
be that the end of the nucleic acid mole eule is already in
the lumen of the tail by virtue of the tail having been
Figure 3.7 Phage End-plate

A. Tangential (grazing) section of Bacillus subtilis infected with
nurnerous SP50ß phage particles. The section shows heads, tails and,
in the centre, the star-shaped 'foot' or end-plate. Scale-bar = 100 nrn.
(Courtesy of Dr. R.G. Milne and Plant Pathology Laboratory,
Rothamsted)
B. Drawing to show the plane of the section A - A.
assembled round it when the virion was formed. Regarding

the second point, one is tempted to think of those commerc-
ially available balls of string, where one can puH string end-
lessly from the centre without rotating the ball, and speculate
on similarities.
In even the best electron micrographs the head sheath or
"membrane" of T2 appears to be smooth and structureless.
32 BACTERIOPHAGES
In the light of modern knowledge of nucleic acid function

this is mysterious. The ratio between the weight of a finite
quantity of nucleic acid and that of the molecule of protein
for which it can code is, very roughly, 9 : 1. If the protein of
a virion were all one molecule with a unique amino acid
sequence, then 90 per cent of the weight of the virion would
need to be nucleic acid. Yet even in those virions that contain
the greatest proportion of nucleic acid it never accounts for
as much as 50 per cent. The paradox is resolved if the protein
of the virion is present, not as a single molecule but as many
identical sm aller moleeules (subunits). This was realised so on
after the theory (now an accepted fact) of the genetic function
of nucleic acid was proposed and the finding that, in the very
smallest phages, a protein subunit structure could indeed be
revealed by electron microscopy provided valuable support
for the theory. Using the technique of negative staining, not
only could the knobs or 'capsomeres' that compose the shell
or 'capsid' of phages such as 0X174 and f2 be clearly seen
but also the 'monomers' from which the capsomeres themselves
are composed. A subunit structure, you will remember, is
known to be present in the tail of T2 - it can be seen with
the electron microscope - so the homogeneous head presents
a problem for which two possible answers have been debated.
One is that it is composed of subunits that are beyond the
limits of resolution of the electron microscope. It is difficult
to see how the precise geometry of the head could be main-
tained by the large number of subunits required unless the
bonds between them were unusually rigid. The alternative
is to assume that large subunits are present, maintaining the
shape of the head but invisible because they are 'plastered
over' by a layer of different, much smaller protein molecules.
Analysis of the head sheath proteins is not easy and is COI11-
plicated by the existence of a small amount of intern al
protein, possibly involved in the organization of the packing
of the nucleic acid. In the heads of certain highly defective
mutant phages ('polyheads') a subunit structure can indeed
be seen.
3.2 Other tailed phages

The contractile tails of coliphages T2, T4 and T6 (the 'T-even'
series) are virtually identical and there are many other phages
of other bacteria whose tails are similar although they may
vary in length and details of the collar, end-plate and fibres
(if any). The contractile tail is a very efficient mechanism and
those phages that have it tend to be adsorbed rapidly.
There are, however, several other types of tail. Coliphages
Tl, T5 and lambda, for example, have thin, flexible tails with
no sheath which are non-contractile. Coliphages T3 and T7
have just short, tapering stumps. Tails like these are quite
common and while there is no doubt that they serve as organs
of adsorption and injection it must be admitted that we have
litde knowledge of how they function.
3.3 Tail-Iess ('Minute') phages

All known tail-less phages are parasites of E. coli and other
Gram-negative rod-shaped bacteria. They are spherical, with
diameters in the range 22 nm - 25 nm. They are the smallest
forms of life and show the ultimate adaptation to smallness
in that their genetic material (DNA or RNA) is single-stranded.
The nucleic acid must code for the protein that surrounds
and protects it but, with diminishing size, the ratio of surface
to volume increases and the coding problem becomes more
acute. The number of structural proteins in these minute
virions may be just two. One of these may function also as a
lysozyme but, in addition to these, the nucleic acid must
code for the enzyme that catalyses its replication in the host
cell. The DNA in higher organisms, as weIl as that in the larger
phages, consists of the familiar double helix. One of the two
strands in such molecules is clearly redundant so far as
information storage is concerned since, given either strand,
the other could be formed upon it. By eliminating the
complementary strand the minute phages encode their
genomes in the most efficient way possible.
In phages 0Xl74 and SB the genetic material is single-
stranded DNA. This excited great interest when first
discovered but has since been found in several other viruses
34 BACTERIOPHAGES
as well. Single-stranded DNA differs from ordinary double-

stranded DNA in several important properties. Firstly the
radiation sensitivity is much greater since, when a quantum
of radiation causes the single strand to break, there is no
complementary strand to hold the broken ends in juxta-
position and so increase the chance of reunion. Secondly the
double helix has rigidity and tends to be brittle and break
up when exposed to the high shearing forces encountered
when a solution is pipetted through a narrow orifice; single-
stranded DNA is flexible and resists pipetting damage.
Perhaps the best evidence for single-strandedness was the
finding that in such DNA the ratios of the bases adenine:
thymine and guanine: cytosine are not unity, as they should
be if paired in complementary strands. Such evidence is hard
to obtain since it requires that a substantial quantity of pure
nucleic acid be prepared for analysis. Whereas specific
inhibitors of RNA and DNA synthesis may be used to
differentiate RNA and DNA viruses, there are no such re-
agents to distinguish single-stranded from double-stranded
DNA viruses as yet. There are, however, fluorescent staining
tests [2] that enable nucleic acid type to be determined with
a mlcrogram or so.
Phages f2 and Qß have RNA as the genetic material. This
is single-stranded as in all forms of life save certain plant and
animal viruses. Only in viruses, some plant, some animal,
some bacterial, is RNA the primary repository of genetic
information rather than a temporary carrier of it. Many people
have wondered whether the RNA viral genome might not
serve as a retro-active messenger for the synthesis of a DNA
genome to co-ordinate replication in the host cell but the
evidence does not support this idea.
Although tail-less, the minute phages may have an attach-
ment organ in the form of a single molecule of a different
protein (the "A-protein") present in the capsid. It has been
found possible to extract capsid protein, A-protein and
nucleic acid from aminute phage, purify them and
then remix them so that they reassemble and give rise to
infectious phage again. If protein and nucleic acid will self-
Table 3.2 Properties of some prominent coliphages
Group Phage Nucleic Other peculiarities

acid*
Phages with T2 2-DNA

contractile T4 2-DNA Contain HMC
tails T6 2-D NA
Phages with Tl 2-DNA

non-contractile T3 2-D NA
tails TS 2-DNA - - Multistage injection
T7 2-DNA
Lambda 2-DNA - - Temperate
Tail-less, 0X174 I-DNA

'minute' SB I-DNA
phages f2 l-RNA
Qj3 l-RNA
Filamentous
phages
f1
fd
I-DNA
I-DNA
} Male-specific
*prefix refers to strandedness
assemble in vitro there is no reason to suppose that any

cellular conveyor belts or assembly jigs are required in vivo.
3.4 Filamentous phages (Figure 1.1, f1 and fd)

Another group of phages that contain single-stranded DNA
are the filamentous androphages. 'Androphage' is derived
from the Greek words meaning "Man-eater" and refers to
the fact that they infect only male strains of bacteria. Male
strains possess a long thin tube known as the "F-pilus"
through which DNA is believed to be transferred to the
female cell in conjugation. Filamentous androphages adsorb
to the tip of this and presumably infect through it. The
minute RNA phages also are male-specific but they adsorb to
36 BACTERIOPHAGES
the sides of the F-pilus (Figure 3.8). The filamentous andro-

phages are exceedingly long and slender and unique in that
they do not lyse the host cell but leak out of it gradually.
Figure 3.8 Adsorption of Androphages

A male (F+) cell of E. coli is depicted with a single F-pilus. Three
filamentous (DNA) androphages are shown adsorbed to the tip of the
pilus, sixteen spherical (RNA) androphages are attached along its
length. The scale is approximate.
3.5 The ultracentrifuge

Because of their small size, phages are kept in suspension by
diffusion and will not sediment of their own accord no
matter how long apreparation is allowed to stand. Neither
can they be sedimented in ordinary bench centrifuges with
top speeds in the order of 5000 r.p.m.; the centrifugal force
is still not enough to overcome diffusion. Such centrifuges,
however, are useful for removing bacterial cells and debris
from a phage suspension. They are referred to as 'low speed
centrifuges' to distinguish them from 'ultracentrifuges' in
which phages can be sedimented into compact pellets using
centrifugal forces of a higher order of magnitude.
The centrifugal force relative to gravity (g) that is produced
in a centrifuge is a function of its speed in revolutions per
minute (n) and the radius (r), that is the distance from the
centre of rotation to the point at which the force is deter-
mined. They are related by the formula
g = 1'12 n 2 r x 10- 5
Since g increases with the square of n but only linearly with
r it is clearly more efficient to construct faster rather than
larger ultracentrifuges to produce higher forces. Modern ultra-
centrifuges have rotors with diameters in the range 17 to 27
cm and achieve forces of half-a-million g at speeds up to
75,000 r.p.m. The rotors of such centrifuges run in vacuum
chambers with refrigeration to protect the specimen from
the heat generated by friction.
Figure 3.9 An Ultracentrifuge

The thick steellid ha,s been moved aside to reveal the rotor in its
armoured bowl. On the worktop stands a spare angle rotor; its cover
has been removed and one tube partially withdrawn to show the
construction.
38 BACTERIOPHAGES
The ultracentrifuge is of inestimable value in phage work,

enabling phages to be washed and concentrated for analysis.
Simply sedimenting phages into pellets achieves only partial
purification since other matter may be co-sedimented. Better
purification may be achieved by using the ultracentrifuge in
one of the following more sophisticated modes, which can
also yield valuable information concerning the phage's
physical characteristics.
3.5.1 Rate zonal centrifugation

In this, a sm all volume of phage suspension is layered on top
of a sucrose solution whose concentration is graded from
12.5 per cent at the top to 52.5 per cent at the bottom.
During centrifugation phages travel in sharp bands at differ-
ent rates according to their shape and density. Each phage
can thus be separated and characterised by its sedimentation
rate. Most T-phages can be banded by centrifuging for 40
minutes at 100,000 gin a 30 ml tube.
3.5.2 Isopycnic banding (equilibrium density gradient

centrifugation)
If phage is suspended in caesium chloride solution of density
1'5 (0'830 g CsCI plus 1 ml of water) and centrifuged at
100,000 g for several hours the heavy caesium ions form a
density gradient from about 1'4 at the top to 1·6 at the
bottom and the phage forms a band corresponding to its
buoyant density which is usually in the region of 1· 5. The
simplest method of unloading the gradient is to prick the
bottom of the tube and allow the contents to drip out,
collecting successive fractions. If the phage concentration is
insufficient to give a visible band one may have to detect it
with a U.V. monitor working at 260 nm. At even lower
phage concentrations one may have to resort to infectivity
titrations or the use of radio-labelled phage and scintillation
counting. When the fr action containing most phage has been
located its density may be determined from its refractive
index using an Abbe retractometer.
1-62
1·60
1-58
~1'56
.~ 1·54
'"
01'52
1-50
1-48
1·46
...... '" ,.... ('t') U"l ...... C'f\"'- (Y)
~~~~~~~~~
~.;......;.....~~~~-~
Refractive index
(50dium light)
Figure 3.10 Properties of Aqueous Caesium Chloride Solutions

Grams CsCI per 100 grams Density relative to
of solution water at 20°C
32 1.3135
46 1.5158
56 1.6999
(Data from R.C. Weast's Handbook ofPhysics arid Chemistry, 54th

Edition published by The Chemical Rubber Co., 1973)
The decision whether to use rate zonal or equilibrium

density gradient centrifugation for a particular purpose is
sometimes difficult. The latter, performed as above, may
require a very lengthy centrifugation for the gradient to
form. This time may be reduced by using a preformed
gradient. Angle rotors are intrinsicaHy faster than their swing-
out counterparts and give exceHent results with caesium
chloride [7] . For large scale work the high cost of caesium
chloride may be prohibitive. The phage may be inactivated
by either solute; T-phages survive weH in both but an unknown
phage needs to be tested first. A major consideration with
caesium chloride is its corrosiveness. It must be rigorously
excluded from any aluminium structures. In arecent paper
[1] the relative merits of rate zonal centrifugation, isopycnic
40 BACTERIOPHAGES
banding and other procedures for the purification of phages
are critically compared. There is no generally superior
method.
w.'......
A Start-rotor stationary B During centrifugation
CentrifUgal~.:.
".
~;
force
"
~
.
C Finish-rotor stationary
117 t \\\
'(J Gravity '\:)
D Tubes removed
Gradient unloaded
L-J L-J L..J L-J.
Figure 3.11 Isopycnic Banding of Phage in an Angle Rotor: I

A. Before centrifugation the phage is uniformly distributed in a
homogeneous solution of esel. During centrifugation the centri-
fugal force, acting in the horizontal plane, produces a lateral density
gradient of esel and the phage is concentrated in a band corresponding
to its own density. C. When the rotor comes to rest the gradient
re-orientates from lateral to vertical with linie remixing. D. Each tube
is removed, a minute hole is made in its lower end and the gradient is
allowed to drip out slowly so that successive fractions can be collected.
Very concentrated phage is visible as a blue-brown, opalescent band.
The esel can be removed from the phage fraction by dialysis against
buffer.
CI>
1·0 [ 1-42 >-l
:;c
E 0 ·9 1'44 _
0 0 '8 1'467 R
>-l
-c 0 ·7 1'48 5 c:::
N ~ :;c
tTl
i~ 1 ' ~~
~ - >
~ 0 ·5 1'52 ~ Z
t:l
,g... 0-4 1·54 ~
~ 'Tl
Sl 0·3 1'56 0 c:::
D
0« 0 ·2 I-58 ?5
>-l
0 ·1 1·60 Ö
Z
2 4 6 8 10 12 14 16 18 2022 24 o'Tl
Froctlon no >-l
:::c
tTl
:::;
A B :;c
Ö
z
Figurc 3.12 Isopycnic Banding of Phage in an Angle Rotor: II
A. Band of phage in CsCl gradient. B. Thc gradient was collectcd as 25 equal fractions. The upper curve shows
the density of each fraction determined from its refractive index. The lower curve shows the V.V. absorption of each
fraction. Fraction 13 contains most phage (peak absorption> and has a dcnsity of 1.51. ....
Figure 3.13 Electron Microscopes
A. A 'conventional' e1ectron microscope with the filament in the top
of the column, the specimen chamber behind hatch a and the viewing
screen several inches behind window b, where it can be observed
c10sely through the binoculars. The separate power pack and vacuum
system are not shown. B. Close-up of A's specimen chamber (opened).
A single specimen is held in the end of tube c, which is lowered
automatically into position when the hatch is c1osed. The vacuum
must be broken and re-established each time the specimen is changed.
C. In this novel design the power pack and vacuum system are con-
tained in the console; by having the column below bench level
mechanical stability is improved and a very compact instrument is
possible. The filament is in the lower end of the column. The specimen
holder is inserted at d. The phosphor is applied directly to the back of
viewing window e, giving a very bright image. It can be viewed in
moderate room lighting, unlike A which must be used in almost total
darkness. D. The specimen holder for C takes four grids at f. They can
be viewed in any order without disturbing the vacuum. (Courtesy of
A.E.1. Ltd)
44 BACTERIOPHAGES
3.6 The electron microscope

Phage structure can be observed only with an electron micro-
scope. The principles of its construction closely parallel those
of the light microscope but technical complexities make it a
very costly instrument. Four decades of refinement have
produced instruments that are not at all difficult to maintain
and operate. Books on electron microscopy are numerous
and manufacturers are usually willing to give technical advice
so an äccount of the instrument itself here would hardly be
germane. Where the phage-novice is more likely to need help
is in the preparation of his specimen.
The electron beam has poor penetrating power, consequent-
ly the phage must be mounted on thin films of high electron-
transparency, themselves supported by sm all copper mesh
grids. The material known as 'Formvar', cast on glass, is
excellent for beginners on account of its toughness. Collodion
and evaporated carbon films are less grainy but more fragile.
Composite films, blending the advantages of two or all of
these materials can be made. The phage suspension is applied
either with a spray ('atomiser') or by putting a droplet on the
filmed grid and removing the excess with a point of filter
paper. The phage itself is essentially transparent to an electron
beam and various techniques have been used to make them
visible. The electron microscope operates in high vacuum.
Hence whichever technique is used the specimen must be free
from water and other volatiles and great care is necessary to
mini mise distortion as they are removed.
3.6.1 Sbadow casting

This involves the deposition, in a vacuum chamber, of a thin
film of heavy atoms (gold/palladium alloy was popular) from
a point source at a low angle. Only one side of the specimen
will be coated for atoms in vacuo travel in straight lines. The
accumulated metal impedes the electron beam, giving a strik-
ing 'light and shadow' effect. It was useful in early phage
studies, revealing the overall shapes at magnifications of c.
20 OOOx. Since the metal actually occludes the finer details
the technique is not much used nowadays except in the
'- t ,,/ Metal evaporqted
l
from electrically
• '," ~~a:ed fllamen
C::::::===~I~
t To vacuum
pump
'"
D
~~~ -';\ 0 ! 6 -Support film
Ir:
,: i G'd
rI b ar
li! Debosited layer of metal
'
'Shadow'
I
Partlcle
Figure 3.14 Shadow Casting

A. The specimen is tilted at an appropriate angle to the source of
metal atoms. B. To show formation of the 'shadow'.
46 BACTERIOPHAGES
A Phage in protein
film on surface of
liquid
... ..' '.:: :.: .: ..

B Phage head ., .: .~ .:.: . '-.: '.' ;: .
ruptures and : ........ : ' . .'.: '.'
spreading protein ' ,' .' :'. ' .... '. .
extends nucleic .... : .':.;';.. .:.;. . : -
acid moleeule . : . : :: :.:: .... . .
....
~ ':' .... :::'.;-.::. ::::.' ~..
:
Figure 3.15 Spreading of Phage Nucleic Acid in a Protein Film

A and B show successive stages in the process leading to the fully
displayed nuc1eic acid molecule in C.
'freeze-fracture' process and Kleinschmidt's method for the
examination of nucleic acid molecules (Figure 3.15). In the
latter, phage particles are osmotically shocked in a spreading
monolayer of protein (usually cytochrome C) at a water-air
interface. As the protein spreads the DNA is gently dispersed
over a confined region in the vicinity of the empty phage head.
The spread-out molecules are picked up on a filmed grid.
Platinum shadowing is applied at a very low ('grazing') angle
while the specimen is rotated to achieve deposition on all
sides of the molecule but not over it.
3.6.2 Negative staining

This is useful in revealing the fine details of phage structure,
including the shapes of some individual monomers. The
Electron beom
A . POSITIVE STAINING
Rl'froctl'd
stoined .Ieclrons
Image on -+----1- Beom passes through

I!ll!etron-lueent oreos
fluorne.nt
Grld bor sereen ond
on photo-
graphie pr int
- - ---
D:
B. NEGATIVE STAINING
Negative Rl'fracled
stain surrounds ppear,?nee
parti cl~ _ ~t_gr'd eleclrons
Beam pann through
electr on -Iueent areas
Image on
lIuoreseent
se reen and
on photo -
graphie print
Figure 3.16 Positive and Negative Staining for Electron Microscopy

Positive staining is hardly used in bacteriophage studies except in the
examination of sections of infected bacteria (see para. 4.7).
48 BACTERIOPHAGES
particles are mixed with a solution of an electron-opaque salt

such as potassium phosphotungstate or uranyl acetate before
application to the filmed grid. The salt does not penetrate
the specimen but produces electron-dense interstices between
partides and between the monomers that compose them so
that a reversed image is obtained when compared to positively
stained specimens.
It should be noted that shadow casting increases the appar-
ent size of particles due to the thickness of the metal coating
whereas negative staining re duces it as the stain lies above and
below the real edge of the structure.
4 The latent period
and the ecfipse
The one-step growth experiments shown in Figure 2.4 show

that phage multiplies inside the host ceH and is released at the
end of the latent period when it bursts, yet, if the ceH is
broken open, no phage can be extracted from it. Wh at
becomes of the phage during this 'eclipse phase'? The answer
to this question has been obtained from three main lines of
investigation. Again, the T-even phages predominate.
4.1 Electron microscopy

In ultra-thin sections of infected bacteria certain events can
be discerned. Within five minutes of infection the bacterial
nucleus, at first compact and discrete, breaks up into granules
that migrate to the periphery of the ceH and disappear.
Between five and ten minutes threadlike elements appear and
coalesce into spherical packets. At around the 15th minute
particles resembling mature phage with head membran es and
tails can be seen. If infected ceHs are broken open and the
contents examined ten minutes or more after infection it is-
sometimes possible to see structures that have been caHed
'doughnuts' on account of their appearance and are probably
head membranes containing no nucleic acid. Tail sheaths,
fibres and end plates mayaiso be seen.
49
50 BACTERIOPHAGES
Figure 4.1 Ultrathin Seetions of T4-Infected E. eoli

A. Darkly staining 'condensates' can be seen. B. Many 'tau-particles'
(arrowed) can be seen lying next to the cell membrane and a few free
in the cytoplasm. Some tubular structures (polyheads perhaps) can
THE LATENT PERIOD AND THE ECLIPSE 51
also be seen. Both the condensates and the tau-partic1es are described
as 'head-related' structures but whether either or both are true
precursors is not yet proved. (Courtesy of Prof. E. Kellenberger and
Heidi Wunderli)
52 BACTERIOPHAGES
NH 2
I
N~~'C/X
11 5 11
~Cz..3:C,
o N H
I
H
Figure 4.2 HMC

The DNA of coliphages T2, T4 and T6 contains, instead of cytosine,
its analogue S-hydroxymethyl cytosine (HMC). In cytosine X = H.
In HMC X = CH 2 OH.
4.2 Biochemical analysis

Bacterial DNA contains the four nucleotide bases, adenine,
thymine, guanine and cytosine found in higher organisms.
Coliphage T2, however, contains hydroxymethyl cytosine
(HMC) in place of cytosine and this makes its DNA readily
distinguishable from that of the host cell. Thus the synthesis
of the phage DNA and bacterial DNA can be studied indepen-
dently in the cell. Such studies show that almost immediately
on infection the synthesis of bacterial DNA ceases. The
synthesis of phage DNA is detectable between the 5th and
10th minutes and continues at an almost constant rate until
shortly before lysis.
4.3 Antigenie analysis

Bacteriophages are powerful antigens. The proteins that com-
pose the various parts of the virion are all antigenically dis-
tinct and antisera may be raised against them separately if
sufficient of the component can be purified. An antiserum
against intact phage will contain antibodies against all the
exposed protein components. The rem oval of any of these
antibodies (but particularly those elicited by the tail
proteins) will reduce its neutralising titre. Thus an antiserum
against whole phage will have a certain titre but if it is allowed
to react with isolated phage components that titte will be

reduced. The extent of the reduction is known as the 'serum
blocking power' of the preparation used. The detection of
serum blocking power in a bacterial exttact is therefore
evidence that phage proteins are present in it. By means of
this technique it was shown, for example, that head membrane
antigens were present after about the tenth minute post-
infection, which corre1ates remarkably weH with the
appearance of 'doughnuts' at about the same time.
4.4 The overall picture

The findings summarised in Figure 4.3 show that the phage
nucleic acid injected into the cells catalyses its own repro-
duction and brings about the synthesis of phage proteins
which are eventually assembled into mature virions. But how
is this achieved? Be10w is an outline answer to the whole
question foHowed by detailed consideration of two important
specific aspects.
Minutes post-inlection
o 5 10 15 20 25
... 1 : ; li 1111
Disappearanee 01 baeterial nueleus ~
Funetional baeterial DNA present
Synthesisol baeterial DNA eeases ~
Phage DNA deteetable
Phage DNA condensates appear
Net RNA synthesis eeases ..
Phage protein'
'Early sp8eilie detectable
RNA deteetable ::i::~:=::5:=
Pro teins 01 the virion detectable
Virion companents seen in E.M.
Infectious virions present
Lysis
Figure 4.3 Observations on T2-Infected Cells of E. coli During the

Latent Period
The diagram summarises the findings of several authors. Tapering of
the datum bars indicates the variation in timing of each phenomenon.
54 BACTERIOPHAGES
The bacterial cell's own enzymes transcribe a complemen-

tary RNA copy from one strand of parts of the parental
phage genome. This is called the 'early messenger'. It
associates with the cell's ribosomes and directs the synthesis
of the 'early proteins'. These are enzymes that (a) suppress
completely, within minutes, the cell's own RNA and protein
synthesis ('shut-off'), (b) degrade the bacterial DNA so that
breakdown products are available for the synthesis of phage
nucleic acid, (c) makes copies of the phage genome.
In the case of T-even phages, before the genome can be
copied enzymes in the early protein must produce HMC,
which is not present in bacterial cells. Two of the complex
battery of enzymes responsible, thymidylate synthetase and
deoxycytidylate hydroxymethylase, were the first enzymes
to b.e demonstrated in the early protein and started a hunt
for more that continues to the present.
Halfway through the latent period many DNA molecules,
copies of the parental genome, have been formed constituting
the 'replicating pool'. Units are steadily withdrawn and con-
densed for incorporation into virions but there is now some
doubt whether the condensates seen in early electron micro-
graphs are true precursors whereas Kellenberger's tau particles
probably are. DNA molecules in the replicating pool are not
entirely stable; breakage occurs frequently and although the
fragments reunite to form complete genomes they may do
so with the corresponding fragment of another molecule.
Where a cell is co-infected with two or more phages genetic
exchange is thus possible; indeed, specific enzymes known as
'nickase' , which weakens the DNA molecule by removing a
few nucleotides from one strand, and 'ligase', involved in
rejoining, have been postulated. The mechanism of breakage
and reunion must be a sophisticated process to ensure that
the rejoined fragments make up no more or less than a
complete genome.
The process of DNA replication heraids the transcription
of other parts of the phage genome which specify structural
pro teins of the virion. These and the DNA are then assembled
to form the mature, infectious virion. Finally, through the
action of phage-induced lysozyme, the murein of the bacterial
THE LATENT PERIOD AND THE ECLlPSE 55
cell wall is digested, the weakened wall bursts and the mature
particles are spilled out into the medium. Contrary to what
was at one time thought, no extracellular maturation appears
to be involved. Besides mature phage, lysis invariably releases
phage components that, for some reason, did not become
assembled ('excess antigen') and substantial amounts of free
lysozyme.
The eclipse phase is of considerable taxonomie significance
as it is probably the only unassailable criterion on which viruses
can be defined as a group. Of all organisms, only viruses go
through a stage in which they exist as nucleic acid only. How-
ever, irrefutable evidence for the existence of an eclipse pp ase
in the life cycle of viruses other than phages is hard to obtain
because of the difficulty of synchronising infection. Attempts
at one-step growth experiments using plant and animal viruses
have yielded results suggestive of an eclipse phase but open
to dispute. With improved technique this difficulty may
disappear.
4.5 Replication of the phage genome

A copy of 'replica' of the nucleic acid moleeule that was
introduced by the infecting phage in to the hast cell must be
manufactured for each of the progeny. Nowadays research
is concerned with the detailed biochemistry and enzymology
of r~plication but, lang before anyone knew even that the
total of inheritable information ('genome') of a phage or
any other organism was physically encoded in nucleic acid
base sequences, S.E. Luria had analysed the overall strategy
of copying. He realised that there are three fundamentally
different approaches to the problem of making numerous
copies of a unique original. They are shown diagrammatically
in Figure 4.4. In I, where all copies are made from the
original, an error would affect only that copy since no copies
are made of the copy. But this method might be expected to
engender severe wear and t~ar on the original. In 11, where
each successive copy is made from the preceding copy, the
original is subject to minimum use but the copies will
deteriorate as every blemish is cumulatively reproduced.
S6 BACTERIOPHAGES
11
1lI
Figure 4.4 Three Strategies for Copying a Genome

Each diagram traces .the lineage, through six generations, of each replica
of the parental genome. At each individual replication, shown as a
bifurcation, the original is placed to the leit and the replica to the
right. Open circles represent normal replicas, solid circles are erroneous
replicas. Note the erroneous fraction:
I. All copies are made from the original. An error in the third
generation affects 1/6 individuals in the sixth generation, decreasing.
II. Each copy is made from the preceding copy. An error in the
third generation affects 2/3 individuals in the sixth generation, increas-
ing.
III. Both originals and copies are copied. An err6r in any generation
affects a constant proportion of individuals in subsequent generations.
Occurring in the third generation it affects 1/4.
Both land 11 are inefficient processes because at any one
time there is only one genome from which copies are being
made. In 111, where both originals and copies are copied,
there is no such bottleneck. The mean quality of the copies
might be expected to be worse than in I but better than in
11. These ideas are not necessarily Luria's, nor does the
author claim them as his own. They would probably occur
to anyone faced with the task of making a large number of
copies of a precious document.
But Luria took positive steps to determine which of these
possible strategies represented the actual modus operandi of
a phage. First of aIl he showed on paper that the distribution
of mutant phages per infected ceIl would depend on the
nature of the copying mechanism. These distributions are
shown in Figure 4.5 with Roman numerals corresponding to
the mechanisms in Figure 4.4. Clearly they are very different.
He then set about determining the distribution experimentaIly.
Not aIl errors in the phage genome copy are detectable as
mutant phages but, since they occur at random, one particular
mutation should reflect the behaviour of all. Luria chose to
study the T2r mutant of coliphage T2, recognisable by its
exceptionaIly large, sharp-edged plaques. He examined the
bursts from a large number of individual bacteria and scored
the r mutants. His findings agreed very weIl with the distri-
bution predicted by mechanism 111. Everything that has
subsequently been learned about the T2 genome has tended
to confirm that it replicates in this exponential way. The
two principal papers [10], [11] in which this work is
described make a fascinating study in the conjunction of
theoretical and experimental analyses.
In phage 0X174, which may be typical of phages containing
single-stranded DNA, the replication mechanism seems to be
more like mechanism I. The original phage genome that enters
the ceIl is single stranded and circular and is known as the
'plus' strand. Its first function is to direct the synthesis of a
complementary "minus" strand and together the plus and
minus strands constitute a circular, double-stranded molecule
known as the 'replicative form' or 'RF'. This original RF
attaches itself to a unique site on the bacterial cytoplasmic
58 BACTERIOPHAGES
128
64
32
16
III
8
The number 01
mutants cannot
4 exceed the burst
1
1 2 4 8 16 32 64 128
Figure 4.5 Theoretical Distribution of Mutant Phages According to the

Mechanisms in Fig 4.4
Horizontal axis: number of mutant phages in the burst from a cell.
Vertical axis: number of cells yielding that number or more from their
burst. The Roman numeral on each curve corresponds to the relevant
mechanism in Fig 4.4.
membrane so that when it replicates the progeny RF1s must

remain free in the cytoplasm. There they act as master copies
in that the plus strand is split away to be incorporated into a
virion while simultaneously a new plus strand is formed on the
minus strand. It is not yet known whether the messenger RNA
that directs the synthesis of the phage pro teins is transcribed
from the parental or progeny RF's.
4.6 Morphogenesis of the virion

The linear order of the amino acids in a polypeptide chain, its
'primary structure', is determined by the sequence of
nucleotide bases in the corresponding gene. There is a wide
gap between a linear molecule and a complex three-dimension-
al structure like a phage but there are excellent reasons for
believing that the precise size and shape of the virion, possibly
of any biological structure, are a direct consequence of the

primary structures of its protein molecules.
4.6.1 Formation ofmonomers

A polypeptide chain can be envisaged as a long flexible
molecule constantly changing shape as it is jostled by other
molecules in the liquid medium. If it ever does exist in that
form it must be for an infinitely short time since there are
attractive forces between different parts of the molecule
that pull it into a more compact, coil-spring shape known as
an a-helix, the 'secondary structure'. This by no means
satisfies all the residual forces on the molecule. Others, hydro-
gen bonding and the formation of disulphide bridges between
adjacent cysteine residues, fold the a-helix into an even more
compact 'tertiary structure' in which two or more polypeptide
chains may participate. This is the basic monomer of which
pro tein structures are built. Gentle dissociative chemical
treatments can be used to separate the monomers composing
some phage organelles so that their shapes can be seen in the
electron microscope. Monomers of the T-even tail sheath, for
instance, are comma-shaped. The still unsatisfied residual
forces on the surface of the monomer form a pattern which,
when aligned with a complementary pattern on another
molecule, will cause them to bind together in a highly
specific orientation.
4.6.2 Assembly
At this point the reader must permit a lapse into analogy; it
may help to darify one of the central problems in biology.
Polypeptid"e chains can be regarded as different raw days and
monomers as bricks made to a specific shape according to
the day. Is it necessary to postulate conveyor belts to take
each brick to the appropriate place on site, bricklayers to
join them together and foremen to tell then where to start
and stop in order to produce a house of the required
dimensions? Microscopists and biochemists have rummaged
in the cell for many years to find the analogous components
60 BACTERIOPHAGES
but without success. Fortunately there is a very plausible

explanation: these things do not exist because they are not
necessary; protein monomers, unlike bricks, assemble them-
selves! Moleeules in solution are in astate of constant rapid
motion so that in an infinitesimal space of time any two
molecules in a ceU will come together in every possible
orientation. If their surface forces are such that when they
come together in one particular way they bind firmly in that
position a predetermined structure will grow. The basic
concept of such self-assembly has been amply confirmed by
the chemical disseetion and reconstitution of simple viruses
including phage.
A spherical or quasispherical capsid presents no problems.
Given monomers of the correct shape and bond-angle a finite
regular sheU will result. On the other hand a linear structure
like a phage tail, growing by the addition of successive
monomers to the end, is of potentiaUy infinite length unless
there is some mechanism for halting growth when the appro-
priate length has been reached. Figure 4.6 suggests one such
mechanism suitable in principle for a structure comprising
two collinear assemblies of monomers of different length,
like a tail core and sheath. It works on the 'Vernier' principle.
As the sheath and core grow by the addition of successive
monomers, starting from a level base at the coUar end, the
unequallengths ensure that the other end is never level until
there are twelve sheath monomers to seven core monomers.
The array of residual forces on the end is then complementary
to that on an end-plate jostling nearby, which 'locks on' and
thus finalises the elongation process. The degree of mismatch
necessary to prevent premature adhesion need be only very
slight since the residual forces on a protein moleeule operate
only over very short, precise distances. The 'Vernier' theory
is not new and has recently been joined by others [6] . In
one of these the length of the phage tail is determined by
assembling it around a specific length of nucleic acid hanging
out of the head. This 'tape-measure' theory has other
attractions; it would explain how the nucleic acid found its
way into the tail at injection. The 'cumulative strain' theory
also has its adherents; in this each successive monomer must
Figure 4.6 The 'Vernier' Model
be deformed a little more than its predecessor to fit the growing

tail, which terminates when the required deformation exceeds
the capacity of the monomer.
4.7 Ultramicrotomy
To observe intracellularphage development with the electron
microscope infected bacteria must be sectioned at about 100
nm thickness using an ultramicrotome. They are fixed at the
required stage in the latent period by the addition of osmic
acid (OS04), which also stains them, and spun down into a
pellet. The pellet is stirred with a few drops of molten 2 per
62 BACTERIOPHAGES
cent agar, allowed to set and cut into 1 mm cubes. A further

stain, uranyl acetate, may be applied at this stage. The cubes
are then dehydrated in aseries of alcohol or acetone solutions
of increasing concentration, ready for embedding.
The cubes of agar must be impregnated with and embedded
in a synthetic resin with the appropriate cutting qualities.
Methacrylate, 'Araldite' and 'Epon' have been used with
success; 'Vestopal' is perhaps the most widely used for
bacteriophage studies. The specimen is placed in a solution
of liquid resin which diffuses into it. It is then transferred to
a mixture of resin and a catalyst which when heated sets hard.
This is usually done in a polythene capsule about the size and
shape of a pistol bullet, with the specimen at the point. For
sectioning, the point is trimmed to an approximately
pyramidal shape with a flat top about one mm2 •
The knife used for cutting ultra-thin sections is normally
the edge of a piece of broken glass. Diamond knives are
available and are reputed to be economical in the long run
if used carefully. They are not, however, an automatie
choice, even for the wealthy. The coefficient of friction
between the resin and the knife is important. 'Vestopal' is
said to be less suitable for use with diamond than glass knives
for this reason. Glass knives are made from strips of high
quality glass ab out 25 mm wide by 6 mm thick. They are
first broken across the length to make squares. The squares
are then broken diagonally, thus producing four acute angles.
Although the break is made by a special machine it invariably
curves at the end so that only one of these angles is optimum
for section cutting. A trough of liquid is fixed behind the
cutting edge so that sections can float away from the edge as
they are produced. They can be further stained by floating
them on solutions of lead and uranium salts. Finally they are
picked up on aperforated copper grid that supports them in
the electron microscope.
Figure 4.7 shows the knife in its holder and the specimen
in its chuck. The rest of the ultramicrotome, a mirac1e of
ingenious design and precision engineering, is simply to
ensure that they move correct1y relative to each other during
the cutting operation. In that, nothing short of perfection
THE LATENT PERlOD AND THE ECLIPSE 63
Figure 4.7 Ultramicrotomy

The supporting grid held in the forceps will be used to collect the
sections by dipping it into the liquid and raising it under them as they
float dear of the knife edge. (Courtesy of LKB Instruments Ltd.)
64 BACTERIOPHAGES
is good enough. With a binocular attachment magnifying

about twenty times the sections can be seen coming off the
knife edge. Those showing silver or gold interference colours
are of a satisfactory thickness for most studies.
Ultramicrotomy is a very individual art. The foregoing
outlines a fairly general technique; most workers have their
own modifications in detail. Positive staining for electron
microscopy is particularly idiosyncratic. No colours can be
seen with an electron microscope, only light and dark
according to the impedence that the specimen offers to the
electron beam. Stains therefore consist of electron-dense
metals like osmium, uranium and lead which, under appro-
priate conditions of concentration and pH, selectively react
with nucleic acids or lipids and make them appear darker in
the final print.
5 Lysogeny
The relationship between bacteriophages and their hosts that

has been considered up to now hardly justifies the title of
'relationship' at all. It is completely one-sided in that, within
minutes of their first encounter, the phage has multiplied
about one hundredfold while the host is utterly destroyed.
Any parasite that continues to behave in this fashion will
very quickly render itself extinct, along with its host. It is
not surprising, therefore, that a different sort of relationship
has been found to exist, in which the destruction of the host
bacterium is a rare event. It is called 'lysogeny', and the
essential feature of it is that, after infection by phage, the
bacterial cell continues to live and multiply normally in a
manner for the most part indistinguishable from that of the
uninfected cell yet harbours within it the capacity to produce
a normal burst of phage under certain conditions. Such
bacteria are called 'lysogenic' and the phages that produce
the condition 'temperate', as opposed to the virulent phages
considered hitherto.
Lysogenic cultures can be recognised by the fact that they
always contain some free phage particles. The number may
be very low; one per million cells or less. They can be detected
by their plaque-forming ability on a different strain of
bacteria. A temperate phage will not form plaques on its
lysogenic host since lysogeny confers immunity to super-
infection by the same phage. The sensitive strain, known as
65
66 BACTERIOPHAGES
the 'indicator' strain, may be found by trial and error, testing
a large number of strains for sensitivity to the temperate
phage. This is a tedious business. Quite possibly a high pro-
portion of all bacterial cultures are lysogenic but the fact
has gone unrecognised for lack of suitable indicator strains.
One of the best places to look for indicator strains is in the
lysogenic culture itself. In any lysogenic culture a few
cells spontaneously 'eure' themselves of phage and thus
become sensitive to it. Ultraviolet irradiation increases the
frequency of curing.
At one time it seemed that lysogeny might be the result
of persistent re-infection by a slowly multiplying phage whose
growth rate could never catch up with that of the host so as
to infect and lyse all the cells. This situation may indeed
exist in certain 'phage-carrying' streptococci known to the
dairy industry but it was dramatically proved to be otherwise
in better known lysogenic cultures by the following experi-
ments. Culture of lysogenic bacteria in media containing
specific antiphage serum showed conclusively that extra-
cellular phage (which would have been inactivated by the
serum) was not essential to the maintenance of lysogeny.
Furthermore, by culturing individual cells in separate micro-
drops of medium for many generations and testing them for
lysogeny, it was shown that all the cells in a lysogenic culture
possessed the lysogenic property. The two experiments can
be combined, making an even more convincing demonstration.
It should have been apparent from the total immunity of the
lysogenic culture to super-infection that lysogeny affected
more than just those few cells that actually lysed. However,
it must be remembered that at the time when the concept of
lysogeny was put forward the attention of the scientific
community was focussed on the fascinating achievements of
the American workers using virulent phages. It was tempting
to dismiss lysogeny as 'experimental error' or 'misinterpreta-
tion '. or simply as a rare example of freak behaviour. It now
seems more likely that the normal phage-host relationship is
one of lysogeny and that it is the virulent phages which are
freaks.
The free phage found in a lysogenic culture arises by the
LYSOGENY 67
spontaneous lysis of an occasional cell, liberating anormal

burst of progeny. Certain special treatments will induce all
the cells to lyse, with the release of phage. The first of these
to be discovered was ultraviolet irradiation. The dosage must
be precisely regulated and the cells must be maintained in the
appropriate physiological condition. Done correctly,
irradiation results in massive, simultaneous lysis. Mutants of
lysogenic cultures are known in which a small increase in
incubation temperature will cause lysis to occur. Of the many
chemieals found to bring about lysis of a lysogenic culture,
the antibiotic mitomycin C currently enjoys the greatest
popularity for experimental purposes. These treatments,
physical and chemieal, are known as 'inducing agents'. Their
mechanisms are incompletely understood at the present time.
A further form of induction, 'zygotic induction', will be
described when the general properties of lysogenic cultures
have been set forth in more detail. One temperate phage,
coliphage P2, is said to be non-inducible; virions are released
spontaneously by a small number of cells.
Every cell in the lysogenic culture can lyse and liberate
phage; yet if the cells are broken open no phage can be
extracted from them nor can phage be seen in sections by
electron microscopy. Neither can phage proteins be detected
serologically. Lysogenic cultures thus behave as if they con-
tained a gene for phage production; a gene that is normally
quiescent, functioning only occasionally as if by accident or
in response to a specific stimulus. Long before the chemical
nature of genes was known, the name 'prophage' was given
to whatever it was that lysogenic cells contained. The pro-
phage is transmitted to the daughter cells at division and a
mathematical study of spontaneous curing suggested that
prophage was present at the rate of one per cell. Phages are
not very heat resistant. At 100°C their survival is measured
in fractions of a second. Nevertheless the spores of a lyso-
genie Bacillus species can be boiled and still germinate to give
the same lysogenic culture. Clearly, whatever mechanism
protects the bacterial genes is shared by the prophage. It is
now believed that the prophage does not simply behave like
a gene; it becomes so completely integrated into the cell that
68 BACTERIOPHAGES
it temporarily loses its separate identity and to all intents and

purposes it is a gene. An understanding of this situation has
been achieved mainly through the study of the inheritance
of lysogeny and at this juncture readers are asked to note
carefully the following caution. By far the greater part of
all our knowledge of lysogeny refers to one phage, coliphage
lambda (A), and its relatives, the lambdoids. The published
literature is vast, induding A.D. Hershey's 792-page book
devoted exdusively to lambda biology. Notwithstanding the
current importance of lambda, one should not assume that it
is typical of temperate phages in general. Too few temperate
phages are known in sufficient detail for any such generalisa-
tion to be made but there certainly are some whose behaviour
is very different. However, within the confines of this smaU
volume lambda must provide the framework of the ac count,
with other temperate phages being mentioned to illustrate
specific points where necessary.
5.1 Inheritance of lysogeny

In bacterial matings the chromosome of the male (donor) ceU
passes into the female (recipient) ceU via a conjugation tube
(the F-pilus). Transfer proceeds linearly at a uniform rate and,
by artificiaUy interrupting the mating after a specific interval
and detecting the last gene to be transferred, the linear order
of genes on the chromosome can be established. Where the
donor carries lambda the prophage is inherited like any other
gene and has been assigned with precision to a position on
the chromosome between the genes for galactose fermentation
and tryptophan synthesis (gal and trp loci).
Where the recipient is lysogenic but the donor is not, the
prophage state is maintained in the zygote. If, however, a
non-Iysogenic recipient receives a prophage from a lysogenic
donor one of two things may happen. The zygote may simply
inherit the prophage and thus become lysogenic; alternatively
it may succumb to avirulent infection and lyse. This latter is
the phenomenon of 'zygotic induction' and to understand it
one must first appreciate how the prophage state is maintained.
The prophage is the genome of the virion, comprising all
LYSOGENY 69
those individual genes necessary for synthesis of the virion

and lysis of the host cell. Why, then, do the cells not lyse?
Briefly, the prophage bears also the gene for a pro tein called
the 'repressor'. This gene is continuously active and the
repressor protein associates with the prophage and prevents
the other genes from being expressed. The cytoplasm of a
lysogenic bacterium contains repressor which is not affected
by conjugation so lysogeny is maintained in the lysogenic
recipient. If however the prophage enters a cell in which
there is no pre-existing repressor to bind to it, all its genes
will be able to express themselves normally and whether
lysis or lysogenization ensues depends on the relative rates of
production of the repressor and the other phage pro teins.
The balance is delicate and the outcome depends on environ-
mental factors.
The mechanism whereby the prophage was located on the
bacterial chromosome has occupied the attention of many
phage biologists. Figure 5.1 shows the principal theories that
1. Attachment
Prophage
Q
Part 01
bacter io l
chromosome
, with genes
" , ' Q and b
2. Insertion
/' , ,,,
'/
3. Insertion with deletion
bl
,
'~ ... '
Figure 5.1 Theories of Prophage Location

For explanation see text.
70 BACTERIOPHAGES
j Cohesive ends
unite
o j N;,ko." " ...d
o
Figure 5.2 Formation of 'Hershey Circles'
DNA extracted from lambda virions (A) is double-stranded except at
the ends where one strand projects twelve bases beyond the other.
The two protruding sequences are on opposite strands and are comple-
mentary. If such DNA is heated to 70°C and cooled slowly the ends
join to form a helix of base pairs and the whole molecule is thus 'cir-
cularised'. At this stage (B) the ends are held by base pairing only
since no phosphodiester linkage has yet been formed to elose the
'nicks' in the sugar-phosphate backbone of each strand. In vivo the
nicks may be elosed by a specific enzyme ('ligase'), to give complete,
circular DNA (C).
have been considered. In the first (1) the prophage was

thought of as a lateral attachment to the bacterial chromo-
some. The idea was never popular because it was hard to
envisage how the attachment occurred and how such a
structure could replicate. The main alternatives involved the
insertion of the prophage into the bacterial chromosome,
LYSOGENY 71
becoming indistinguishable from it except by the nucleotide

sequence. In one of these (2) the prophage was inserted at a
simple break. One might think it would be easy to test such
a hypothesis by seeing whether hacterial genes either side of
the insertion were genetically further apart after lysogeniza-
tion but it must be remembered that the prophage is only
about one hundredth the size of the bacterial chromosome
and genetic mapping techniques were not sensitive enough
to resolve such a small separation. In the other case (3) the
prophage was inserted as areplacement for a corresponding
deletion of bacterial genes. If this were so, the loss ,of
bacterial genes consequent on lysogenization should have
C irculor ized
prophoqe
" ",'
, "
/,'
Figure 5.3 Insertion of Prophage: Campbell's Model

72 BACTERIOPHAGES
been detectable, but it was not. The matter was hotly

debated against a growing background of biochemical and
genetic evidence. It is now widely accepted that lambda DNA
can form a circle (Figure 5.2) and be inserted into the
bacterial chromosome in the manner proposed by A. Campbell
[5] in 1962 (Figure 5.3), whereby there is simultaneous
breakage of the prophage and bacterial chromosomes with
reunion of the broken ends as shown. I t is further proposed
that on induc.tion the prophage is detached by 'looping out'
and 'excision' as shown in Figure 5.4. Norma11y the point of
excision corresponds to the point of insertion but sometimes
it is latera11y displaced so that the phage DNA incorporates
some of the bacterial genes which it can then transduce (see
Chapter 7) and leaves behind some of its own so that it is
defective and can no longer give rise to an infectious particle.
The phage genome can thus exist in three distinct states:
a mass of compact DNA in the virion; replicating in the
cytoplasm or integrated linearly into the chromosome of the
host. Not a11 temperate phages do however; PI prophage for
instance, cannot be assigned to any particular genetic locus
and is assumed to ex ist freely in the cytoplasm or possibly
attached to the cytoplasmic membrane. It is of interest to
note that no RNA phages have yet been shown to be tem-
perate; are they intrinsica11y unable to lysogenize? Integra-
tion of an RNA prophage into the bacterial chromosome
seems unlikely but the formation of a DNA copy of the RNA
genome cannot be discounted. Host preservation by lysogeni-
zation confers a great advantage on the phage so evolutionary
pressures might be expected to have selected strongly in
favour of temperance in a11 phages. Isolation of an RNA
phage and testing it for ability to lysogenize are both very
laborious procedures. The real reason why no temperate
RNA phages are known could be that the search has not
yet been sufficiently thorough.
Reference was made earlier to the apparent normality of
bacteria after lysogenization but there are exceptions. Lyso-
genization can convert non-toxigenic strains of Coryne-
bacterium diphtheriae into toxigenic strains and alter the
antigenicity of Salmonella strains.
LYSOGENY 73
Prophoge ,nserted into

boctllnol chromosomll
between genes a ond b
/."'.,.,,'
Dillect i ~e
~~~;:I©
pholJe genome
'Exc,s,on' restrieted to
tronsduction
genome 01 boeteriol
a b __-T_-,genll b
.. ...... ,~
"
--
Figure 5.4 Excision of Prophage: Normal and Defective

5.2 Techniques of induction

Procedures for V.V. induction vary in detail from one worker
to the next, each finding an optimum under the p.articular
circumstances. The following may serve as a reference point.
To induce lambda, 50 ml of a dense culture.is centrifuged
and the cells resuspended in 10 ml of saline. This is placed in
an uncovered glass petri dish and exposed for 20 seconds
with gende rocking to a 15 watt V.V. lamp emitting mainly
74 BACTERIOPHAGES
at 254 nm from a distance of about 40 cm. The culture is
then diluted back to its original volume with broth and re-
incubated. The intensity of irradiation is affected by the age
of the lamp filament and the nature of the diluent. Small
traces of broth in the saline can absorb much of the U. V.
radiation. Lysis should occur after two to three hours . .In
so me phage-host systems visible lysis may not occur and to
demonstrate induction at all it may be necessary to sediment
the bacteria and plaque the supernatant on the indicator
strain.
Mitomycin C will induce many temperate phages when
added to an exponentially growing culture at a concentration
of lllg mr 1 •
5.3 Bacteriocins
A colicin is a protein produced by one strain of E. coli and
lethai to another strain. It may be demonstrated by the
inability of the sensitive strain to grow in broth which has
been sterilised by filtration after the colicinogenic strain has
grown in it for a short while; the inhibition is maintained even
after substantial dilution. Another way is to grow colonies of
the colicinogenic strain on agar, kill them with chloroform
vapour and overlay with agar inoculated with the sensitive
strain. On continued incubation the overlay shows growth
except over and around the colicinogenic colonies. Colicins
can be assayed by finding the highest dilution that will
produce a clear area when applied to a lawn of a sensitive
strain. This phenomenon is not confined to E. coli. Similar
substances are produced by other bacteria and named, like-
wise, according to the species. Thus Bacillus megaterium
produces megacins, Pseudomonas pyocyanea produces
pyocins and so on. Collectively these substances are known
as 'bacteriocins'.
Bacteriocins differ from ordinary antibiotics in several ways,
the most striking of which is the narrow spectrum of activity.
Whereas an antibiotic is usually active against diverse organisms
taxonomically distant from that which produces it, bacterio-
cins are usually active against other strains of the same species.
LYSOGENY 75
Electron microscopy of bacteriocins [3] reveals that there

are two distinct types. There are the large bacteriocins that
look like phages or parts of phages and which probably are
phages, defective to a greater or lesser degree; then there are
those which are so small that they cannot be properly
resolved and whose sizes must be estimated indirectly from
studies on their rates of diffusion through various media. It
is to these small bacteriocins such as colicins K and V, with
molecular weights in the order of 5 x 104 daltons that the
rest of this account mainly relates.
One molecule of bacteriocin appears to be sufficient to
bring about the death of a bacterial cel!. It adsorbs to a
specific receptor site on the bacterial cell wall (which may
be the same as that for a phage) and remains there; whatever
the influence by which it kills the cell, it is exerted from
there. Treatment with trypsin can remove it and rescue the
cell if applied in time. The mode of action of bacteriocins
varies. Some interfere with energy-yielding metabolism,
others with pro tein or nucleic acid synthesis; the precise
mechanism is mysterious.
Whether or not a bacterium can produce bacteriocin
depends on the possession of a sm all genetic element known
as 'bacteriocinogen' ('C-factor' in the case of colicins), which
is composed of DNA and can replicate autonomously. Some
C-factors have been shown to pass from donor to recipient
during conjugation but, since they cannot be assigned to any
definite locus, they are presumed not to be integrated with
the chromosome. Others actually mediate chromosomal
transfer in conjugation and thus probably are integrated with
it like the bacterial sex-factor.
Mere possession of a bacteriocinogen does not mean that
a cell will necessarily produce bacteriocin. The medium and
growth conditions have been shown to be contributory
factors. The fact that ultraviolet irradiation and mitomycin
C can induce colicin production and its release by the cell
enhances the resemblance of bacteriocinogens to defective
prophages; inevitably students ask whether that is what they
are. Inducibility, the specificity of bacteriocin action, the
shared receptor and the killer action all point that way but
76 BACTERIOPHAGES
we may never know for sure. Bacteriocinogens may be

defective prophages representing the degenerate remains of
ancestral phages. On the other hand, if viruses are genes that
acquired means of escaping fram the contral of the nucleus
and, eventually, from the cell, bacteriocinogens may represent
an eady stage in their evolution.
6 Survival of phages
The phage virion is essentially a survival mechanism designed

to protect the phage genome from the rigors of the environ-
ment to which it is exposed when it destroys its host and to
get it into a new host when such becomesavailable. It is
clearly efficient, otherwise phages would long ago have
become extinct, yet it is common knowledge that phage
preparations in the laboratory gradually lose their infectivity
and eventually become useless. An understanding of the
factors that affect survival of phages is thus important for
anyone who intends to work with them. Furthermore,
valuable knowledge of the fundamental nature of a phage
may be gained by observing how it survives deliberate
insults. The term 'survival' can be construed in many ways,
making it important to define certain terms at the outset.
Survival may imply survival as an object recognisable in the
electron microscope as a phage, i.e. not utterly destroyed,
but most workers take it to mean the retention of infectivity
as shown by the formation of plaques or the clearing of
broth cultures. A phage that has lost the ability to infect
cannot be referred to as 'dead' since it is debatable wh ether
it was ever alive. There seems to be general agreement to use
the term 'inactivated' where loss of infectivity is due to a
specific treatment and simply 'loss of titre' for the natural
decline of infectivity during storage.
77
78 BACTERIOPHAGES
Survival of phages is usually exponential; that is to say

that if the numbers of survivors are plotted on an exponential
scale such as loglo against the duration of treatment they
fall on a straight line (Figure 6.1). This is known as 'first
order kinetics' and resembles the decay of a radioactive
element. As with a radioactive element, survival can be
expressed in terms of the half-life. For comparative purposes
the inactivation rate in logarithms per minute is equally
useful.
6.1 Survival in lysates

Even at room temperature many phages survive weH in
lysates and a working stock may often be left on the bench
for a week or so with negligible loss of titre. In the refrigera-
tor at about +4° C survival is prolonged and the loss of titre
may easily be less than one log per annum. Freezing and
thawing a lysate usually reduces the titre to so me extent but
once in the frozen state, storage at -20°C to -70°C maintains
the titre unchanged for years. Certain conditions must be
observed, however, to ensure good survival. Acids produced
by the host before it was lysed have a strong disinfecting
action; where the host is a vigorous acid producer, a lactic
streptococcus for instance, either the medium must be weH
buffered or the lysate neutralised for storage. Another cause
of loss is the presence of unlysed bacterial debris. Phages may
adsorb onto these and thus be unable to adsorb onto a new
host cello Low speed centrifugation will remove most of such
debris.
6.2 Disinfection
With a few conspicuous exceptions, chemicals active against
bacteria are active also against phages. Thus most general
disinfectants will deal adequately with phage contamination
when required. Disinfectants are very varied in their chemical
make-up and modes of action. Broadly speaking, they
inactivate phages by reacting with either their pro teins or
their nucleie acids but some react with both. Phenols and
surface-active agents destroy phages by stripping off the
SURVIV AL OF PHAGES 79
'0
...
E
.J::
·~O
C7>
o
. !
...J
~2'7~
min
•
Minutes irradiation
Figure 6.1 Inactivation of Phage by U1traviolet Irradiation

A suspension of Staphylococcus aureus phage 42D was made in Ringer
solution and a layer 2 mm deep was irradiated in a lidless petri dish
under a V.V. source emitting 15 ergs mm 2 sec- 1 at 254 nm. The source
was pre-run to stabilize the emission and dosage was controlled by
shielding the dish with a metal cover. Sampies (0.1 ml) were taken at
intervals for plaque assay. All operations were carried out in a
darkened room to prevent photoreactivation.
The logarithm of the percentage of the original population surviving
was plotted against time and a straight line fitted by eye. This is called
the 'survival curve'. From it the time corresponding to a tenfold (one
log) decrease was found to be 2.7 minutes and hence the rate of in-
adivation was 1/2.7 = 0.37 logs per minute. (M. Burnett, unpublished
data).
protein. The various oxidising agents, hypochlorite, iodine

and hydrogen peroxide for example, denature the protein by
oxidation. Silver and mercury compounds are very active
against phages; presumably they bind to their pro teins and
impair function as they do in bacteria.
Hydroxylamine and nitrous acid are chemicals that react
with nucleic acids in ways that are now weH known. At low
concentrations the modification to the molecule may be so
80 BACTERIOPHAGES
slight that the phage is mutated but survives. At higher

concentrations the extent of the change may be incompatible
with function. Formaldehyde reacts with both nucleic acids
and proteins. Single-stranded nucleic acids are more sensitive
to it than double-stranded, a property that has been used to
distinguish them. Metabolic poisons, cyanide for instance, are
ineffective against virions as they have no metabolism to be
poisoned. So also are ether and chloroform since these sub-
stances act by disrupting lipid structures and, with the
exception of PM2, phages have none.
6.3 Experimental insults

Using dilute solutions of phenols, detergents or alkalis, the
proteins that make up the virion can be progressively dissolved
away and at any stage what remains can be examined in the
electron microscope. This 'chemical dissection' has contributed
substantially to our knowledge of bacteriophage structure.
The capsid can be ruptured osmotically to reveal the nucleic
acid; artificial stimuli such as hydrogen peroxide will make
the tail sheath contract enabling the mechanism to be studied.
These procedures inactivate the phage completely.
Bacteriophages are about as sensitive to heat as the majority
of non-sporing bacteria. At 100°C they are inactivated almost
instantly. Between 65°C and 85°C inactivation rates can be
determined conveniently under laboratory conditions and
used to characterise phages. Below 65°C some phages are
inactivated fairly rapidly but others hardly at alL The medium
in which phages are heated has a great influence on the rate of
inactivation. Inactivation is most rapid in pure water; the
addition of salts, especially calcium or magnesium, or
pro teins reduces the rate of inactivation considerably.
The radiobiology of phages is worthy of a book to itself.
Briefly, there are two categories of radiation with which
phages may be treated experimentally: ionizing radiations
comprising principally X-rays and gamma rays, and non-ionizing
radiations comprising visible and ultraviolet light. Ionizing
radiations are highly damaging to bacteriophages, as to all
forms of life, and inactivate them by causing ionizations in
SURVIVAL OF PHAGES 81
or very dose to the nudeic acid moleeule, resulting in strand

breakage. If one of the strands of a double-stranded nudeic
acid mole eule is broken, the complementary strand holds the
broken ends in juxtaposition, enabling the break to be
repaired by specific cellular enzymes. Phages containing
double-stranded nudeic acid must therefore sustain a double
break to become inactivated and are thus much less sensitive
to radiation than those with single-stranded nudeic acid in
which every break is lethal. At one time, before the electron
microscope resolved the matter, attempts were made to
calculate the size of phages from their X-ray sensitivities. The
principle is sound: analogously one might estimate the size
of a distant bell in the dark by scattering pebbles at it and
listening for the hits, but the capacity of double-stranded
PNA for repair was then unknown and consequently,
except in the case of phages with single-stranded nudeic
acid (and that was not known at the time either) all phages
appeared to be much sm aller than they really are. Nor is it
possible to estimate the size of the radiation target by com-
paring its sensitivity with that of an object of similar compo-
sition and known size such as a cell nucleus. Secondary effects
whereby ionizations produce toxic substances at a distance
from the sensitive site to reach it by diffusion can make the
latter appear erroneously large.
The molecular mechanism whereby ionizations cause
nucleic acid moleeules to break is poorly understood. This
complex problem has great relevance to the treatment of
malignant disease with radiations and phages provide a useful
experimental model.
Vltraviolet irradiation has a very subtle effect on DNA. It
does not cause the strands to break. I t causes covalent bonds
to form, cross-linking adjacent pyrimidine bases (particularly
thymine) and preventing replication. Cells contain enzymes
capable of cleaving these 'dimers' and restoring function,
especially if exposed to strong visible light, hence the name
'photoreactivation'. V.V. irradiated phage can also be
reactivated by host cell enzymes that ex eise the dimer and
replace it with normal bases, not always the original bases
however, so that a mutation may occur. Alternatively some
82 BACTERIOPHAGES
A
o
11
H C M
'N/ ........ C. . . . .
I Thymine 11
c!'C'N/C'H
Symbols
I Deoxyribose M
® Phosphate New covalent bond
Figure 6.2 The Effect on Ultraviolet Irradiation on DNA (Dimerization)

A. Two adjacent thymine bases on the sugar-phosphate backbone of
normal DNA.
B. Ultraviolet irradiation has caused the formation of new covalent
bonds, creating a cyclo-butane ring and distorting the structure in that
area.
The exact nature of the distortion is conjectural at present. In any
case, although the diagram indicates correctly the type of bond
between atoms and groups, clear representation of ac cu rate bond
lengths and angles in the compact double helix is not possible on a
flat page.
hosts allow the dimerized DNA to be replicated, leaving a gap

in the strand corresponding to the dimer to be filled in at the
next cycle of replication. This also can lead to mutation.
Irradiation of phages with V.V. is a convenient method of
producing mutants. The phage may be irradiated as virions
in suspension or as replicating genomes inside infected
bacteria. In either case a substantial proportion of those
surviving a heavy dose of V.V. will be found to have mutated
in so me respect. If a plaque assay of survivors is performed,
the curious phenomenon of 'multiplicity reactivation' may
be observed. In this, the number of plaques ai each successive
dilution diminishes more than would be expected. The reason
is that a phage may be so damaged that it cannot, even though
it injects its genome into the cell, initiate avirulent infection.
If the phage concentration is high, so that cells become infec-
ted with two or more damaged phages each, each phage may
be able to compensate for the other's defect so that between
them they cause the production of all the proteins necessary
for infection and lysis.
A special form of internal irradiation known as the
'suicide technique' involves propagating the phage in a
medium containing either labelled phosphorus (P 32 ) or tritiated
thymidine (or other nucleoside in which His replaced by H3 ).
The isotope becomes incorporated into the phage nUcleic acid
and, as it decays, the radiation it produces may cause breakage
of the molecule. p 32 is particularly deadly; when it is incorpor-
ated into the 'backbone' of a nucleic acid molecule each disinte-
gration, in which phosphorus is transmuted to sulpnur ensures
breakage of the strand at that point. As with irradiation from
external sources, phages with single-stranded nucleic acid are
much more sensitive than those with double-stranded DNA.
6.4 Antisera
Phages will not infect animals but nevertheless when injected
into them the proteins of the virion elicit the production of
antibodies that will react with and inactivate ('neutralise') the
phage. The antibodies appear in the animal's serum, which is
then called an antiserum. Because of the high specificity of
84 BACTERIOPHAGES
the reaction of antisera with the proteins that evoke them,

antisera have played, and will continue to play, a valuable role
in bacteriophage studies. Antisera can be used to identify
phages; to inactivate a phage selectively in the presence of
bacteria or other phages; to detect the presence of free phage
proteins. It is therefore important that the budding phage
biologist should be aware at least of the way in which anti sera
are made and tested; details of antibody formation and mode
of action can be found in a modern textbook of immunology
if needed.
6.4.1 Production of antisera

Rabbits are the animals of choice. Beginners may find it
easiest to inject low titre lysates subcutaneously between the
shoulder blades; 10 mls can be administered in this way to an
average-sized rabbit without distressing it. If high titre lysates
(10 10 p.f.u. mr 1 ) are available a faster response is obtained by
injecting a small volume (around 1 ml) into an ear vein.
Four doses, given at weekly intervals, are usually enough to
produce a powerful antiserum. About a week after the last
dose the rabbit is bled by nicking the marginal vein of the ear
and allowing about 10 ml of blood to drip into a test tube.
On standing at room temperature, the blood separates into a
dot surrounded by pale yellow serum, which can be removed
gently with a fine pipette. Anyone proposing to make anti-
sera should first ascertain whether they are bound by any
legal prescriptions. In Britain the Cruelty to Animals Act of
1876 applies. The Horne Office is usually prepared to discuss
with bona fide individuals the type of licence that their work
requires. Supervision by an experienced worker is essential
until skill has been attained. All materials for injection
should, of course, be free from viable bacteria.
6.4.2 Testing the potency of antisera

It is important to know the potency of any antiserum
prepared. The simple technique of demonstrating antiphage
activity shown in Figure 6.3 can be turned into a rough assay
Figure 6.3 Neutralisation of Phage by Antiserum

A phage suspension was plaqued and, before incubation, a drop of
specific antiserum mixed with sloppy agar was placed on the surface
at A. A drop of control serum (from the same anima! before immuni-
sation) was placed likewise at B. Under the control serum plaques have
developed normally but the antiserum has neutralised the phage and
plaque formation is completely prevented. (Courtesy of A. Qanber-
Agha)
by applying serial dilutions of the serum to the plate to

determine the 'end-point'; that which just inhibits plaque
formation. The accurate determination of the inactivation
constant (k) of an antiserum is a valuable exercise for
beginners. Dilutions of the antiserum are made in broth and
a known concentration of phage is added. The mixture is
86 BACTERIOPHAGES
incubated at 37°C and after a standard interval (30 minutes

is often selected) the residual concentration is determined by
plaque assay. The constant is then calculated thus
k = 2·3 x Serum Dilution Factor x LOglO(initial p.f.u. mr 1 )
Incubation (minutes) residual p.f.u. mr 1
There are other formulae for calculating k which employ

naturallogarithms. The factor of 2·3 is incorporated into
formulae employing logarithms to base 10 (common
logarithms) so that values of k obtained by either formula
are directly comparable.
An antiserum raised against a lysate will contain, in
addition to antibodies against phage proteins, antibodies
against bacterial pro teins as well. These latter can be
removed if necessary by the addition of disintegrated
bacterial cells which precipitate the unwanted antibodies as
a complex that can be removed by centrifugation. The anti-
bodies that remain will consist only of those against phage-
induced pro teins but not all of these, it will be remembered,
are present in the virion; some are enzymic pro teins involved
in its production and release. Subsequent absorption of the
antiserum with virions to leave only these latter is possible
but technically difficult, requiring rigorous quantitative
control.
Under certain conditions the neutralization of a phage by
antiserum may be reversible. Papain will remove antibodies
from phage and restore infectivity; so can ultrasonic treat-
ment. Although some antigen-antibody complexes can be
dissociated by dilution, the infectivity of phages can rarely
be restored in this way.
6.5 Survival of phages in nature

Phages can be isolated from most natural materials in which
their host bacteria are found but of the hazards they en-
counter there relatively little is known. The host-to-host
distance is likely to be greater under natural conditions than
in laboratory cultures and the demands made upon the
virion by the environment correspondingly greater. A hazard

likely to be encountered in Nature but rarely in the laboratory
is desiccation, to which the resistance of phages is suprisingly
variable, the minute RNA phages being able to survive desic-
cation for several days at room temperature and others being
inactivated in an hour or so.
7 The importance of
phage in fundamental
biology
üf necessity this chapter is a long one because of the extent

to which materials, techniques and concepts stemming from
bacteriophage studies have permeated the science of biology
in the past two decades. The chapter falls naturally into three
subsections. First and foremost is the concept of the physical
nature of the gene that has been provided by studies using
phage. It is not true that Hershey and Chase's experiments
first showed that genes were composed of DNA - Avery's
work on the bacterial transforming principle should have
convinced most people years before but it was the realisation
that with phage one could introduce selected, uncluttered
genes into an environment (the cell) designed for their
exploitation that formed the linchpin of the genetic revo-
lution. We must therefore take a long look at phage as a
genetic system in its own right. But to what extent is genetic
knowledge obtained from phage studies applicable to other
organisms? In the second subsection we will consider the
facile exchanges of genetic material between bacteriophages
and bacterial cells that lead us to suppose that they operate
on the same genetic basis. Regarding the cells of plants and
animals, in which the DNA is tenaciously bound to protein
and other material, we must admit that there may be pro-
found differences yet very recently CH. Doy and his co-
88
THEIMPORTANCEOFPHAGE 89
•• •\
• t
Figure 7.1 Plaque Type Mutants

Plaques formed by a mixture of T2 phages carrying mutations at the b
and r loci, plated on a mixture of E. coli Band E. coli B/2. Phages with
the b+ allele produce turbid plaques since they lyse only E., coli B
while those with b produce clear plaques by lysing both E. coli strains.
Rapid lysis (r) mutants can be distinguished by the larger plaques they
produce. All possible combinations can be distinguished: b+ r+, b+r,
br+ and br. From Molecular Biology 01 Bacterial Viruses by Gunther
S. Stent: W.H. Freeman and Company. Copyright © 1963.
90 BACTERIOPHAGES
workers have claimed that bacterial genes conveyed to plant

cells by phage would function in them and another team led
by C. Merrill did likewise with anima! cells. The general term
'transgenosis' has been coined to cover all such phenomena
but the evidence for it is not as strong as one might wish. The
third subsection comprises a selection of diverse phage-based
achievements which will serve to illustrate their extent and
importance.
7.1 Phage genetics and the structure of the gene

Early geneticists demonstrated the linear order of genes in
higher organisms by crossing mutants and noting the degree
of recombination found. At that time the gene was regarded
as more or less indivisible and the establishment of the mole-
cular basis of genetics, with the recognition that recombination
can be detected within the gene, had to await not only the
development of new techniques but also the use of a new
kind of organism j one which had a minimum of genetic
characters, was haploid so that no gene could escape attention
through recessiveness and one in which rare events were
readily detectable by the rapid screening of populations
numbering billions. Bacteriophages met this requirement.
The genetics of phage, like that of all organisms, is based
on the study of mutants. In outline, sensitive bacteria are
simultaneously infected with two (or more) phage particles
which differ in severa! characters and the recombinant progeny
rI
r III
Figure 7.2 Relative Positions of the Rapid Lysis Linkage Groups on

the Genetic Map of T4 Phage
THE IMPORTANCE OF PHAGE 91
counted. Since bacteriophages are haploid, lethaI mutations are

unsuitable for genetic studies because the phage carrying them
cannot be propagated. Instead 'permissible' mutations, that
is those that permit propagation, are utilised. Permissible
mutations fall into two main classes: plaque-type mutations
which result in the production of plaques with changed
morphology and conditionallethal mutations which have a
lethaI effect under some circumstances ('restrictive conditions')
but not under others ('permissive conditions').
7.1.1 Plaque type mutants

The most famous mutants involving plaque morphology are
the rapid lysis (r) mutants of the T-even phages. Wild type
(r+) strains plated on E. coli B produce small plaques with
fuzzy edges. The fuzzy edges are a consequence of 'lysis
inhibition' in which reinfection with further phage of the
same type delays lysis considerably so that only a fraction of
the bacteria burst before the phase of active multiplication is
over. If a large number of plaques is examined about one in
10 000 can be distinguished as not being restrained by lysis
inhibition; these plaques, which are larger than the rest and
have sharper edges, are rapid lysis mutants. There are three
distinct groups of r mutation, which are designated rl' r II
and r m and map in different regions of the chromosome
(Figure 7.2). They are distinguishable phenotypically by their
Table 7.1 Wild type (r+) and rapid lysis (r) mutant
group plaque morphology
E. eoli strain
Phage group
used for plating
r+ rI rn rm
B wild r r r
K wild r no plaques wild
wild normal r+ plaques

r rapid lysis plaques
E. eoli K is E. eoli K12, lysogenic
for lambda phage
92 BACTERIOPHAGES
ability to plaque, and on the nature of the plaques formed,

on different strains of E. coli (Table 7.1). Other distinct,
genetically determined differences from wild type plaque
morphology inc1ude those determined by turbid mutants
which give plaques circumscribed by a turbid ring and 'star'
mutants.
7.1.2 Conditionallethal mutants

Host range (h) mutant phages were used in early genetic
studies. Bacteria readily acquire resistance to phage but
phages often reciprocate by acquiring the ability, also through
mutation, to attack these resistant bacteria. A technique for
the actual isolation of host range mutants is shown in Figure
7.3. Conditionally lethai host range mutants are easily
identified by plating out phage on mixed indicator strains
(E. coli Band E. coli B/2 for example). Non-mutant phage
lyse only their normal indicator and turbid plaques result
because the other strain, resistant to the phage, remains but
the h mutant lyses both strains of bacteria and consequently
produces c1ear plaques (see Figure 7.1 again).
Temperature sensitive (ts) mutants are characterised by
mutations that prevent the expression of their genes at, for
example, 42°C but not 30°C. This is probably because the
mutant pro tein is unable to retain a functional three-
dimensional structure at the higher temperature. Such
mutants may be isolated by plaquing at the restrictive temper-
ature, marking the plaques that develop and re-incubating at
the permissive temperature when any unmarked plaques
developing should be ts mutants. Alternatively the cultures
may be incubated at the permissive temperature first. On
transfer to the restrictive temperature, wild- type plaques
continue to become larger while ts mutants remain distinctively
small.
Suppressor sensitive (S US) mutations of a very special
kind have been described in T-even phages: these are known
as 'amber' (am) mutations. Unlike h mutants, whose host
range depends on whether or not they will adsorb to the
bacterial strain, am mutants will adsorb to both the permissive
A. Confluent
lysis with a
few resistant
bacterial
.. "."
colonies
(termed
E. co!; B/2)
E. co/; B
B. Confluent growth with E. co/; B/2 colony

a few plaques (phage from picked and
these are termed T2h) cultured
Pick plaque
C. Bacterial colonies
resistant to T2 h
arise (termed E. co!i B/2/h
o
Lysate of T2h
phage prepared
on E. co!; B/2
Figure 7.3 Scheme for the Isolation of Phage Resistant Bacteria and of
Host Range Phage Mutants
Confluent lysis results when E. eoli B is plated (after allowing time for
adsorption) with a gross excess of wild-type (h +) T2 phage (A). A few
resistant bacterial colonies will be present, these are termed B/2
bacteria, that is resistant to T2h + phage. If an excess of T2h + phage is
plated with B/2 bacteria no confluent lysis is found but a few plaques
arise from infection by mutant phage particles designated T2h (B).
Then if E. eoli B/2 and phage T2h are plated together, B/21h bacterial
mutants can be obtained resistant to both h + and h phage types (C).
and so on.
94 BACTERIOPHAGES
and restrictive host. They are characterised by a mutation

generating the 'amber' nonsense codon (UAG) in a gene with a
vital function so that they can multiply only in bacterial
strains carrying a suppressor of this codon. It is said that
'amber' is the English equivalent of 'Bernstein', the name of
someone associated with the discovery of these mutants, and
has no more significance than that. The mutants themselves
are of the greatest significance because they are readily
isolated and have helped fill in gaps all round the genetic map
where no other mutants were available.
7.1.3 The technique of genetic analysis by crossing phage

mutants
Once aseries of mutants has been isolated one can begin to
construct a genetic map for the region to which the mutants
pertain. In principle, bacteria are mixedly infected with two
mutant phage types and the percentage of wild type progeny
that result, if any, is recorded. For such recombination
experiments at least three particles of each mutant phage are
added per bacterial cell so that it stands a good chance of
being infected by both. Most of the progeny of such a 'cross'
will resemble one or other of the parents but in a few the
genetic information may have been recombined to give a
genome without either mutation, i.e. the wild type. The
further apart the mutational sites lie, it is assumed, the
greater is the chance of recombination occurring between
them so, by converse reasoning, the less frequent. the wild
type recombinant the doser the mutational sites. Where no
wild type recombinants occur it is taken that the mutational
sites are identical, at least within the limits of resolution of
the technique.
It must be remembered that with phage one is dealing with
a dynamic system in which recombination occurs randomly
within a multiplying popul:ition of phage chromosomes, .each
member of which can partake in a succession of recombina-
tional events. Under such circumstances one cannot hope to
discover the outcome of any single mating event and the only
reasonable approach is to regard phage recombination as a
problem in population genetics which requires mathematical

treatment for its solution. Besides recombination, mixed
infection mayaiso result in complementation, where both
mutants reproduce in a cell where neither could on its own,
and phenotypic mixing, where progeny virions consist of the
DNA of one mutant and the pro teins of another. Only
recombination, however, produces geneticaHy stable wild type
progeny.
7.1.4 Fine structure analysis ofthe rrr region

The first system of genetic fine structure analysis was evolved
by S. Benzer in the mid 1950's. The development of this
system was a direct consequence of his observation that, in
general, rrrmutants of T4 phage do not plaque on E. coli K,
although they infect and kill this strain. To remind you, rrr
mutants are easily recognised since they produce large, distinc-
tive plaques on E. coli B but do not normaHy plaque on
E. coli K. However, an rrr mutant can grow on E. coli K if the
ceH is co-infected with a wild type phage - it is as if the wild
type makes good some deficiency of the mutant. Benzer's
success in exploiting the rrr system of phage T4 depended to
a very large extent on the development of several distinct
types of mapping procedure which reduced the work load
enormously. One was the use of deletion mutants to order
point mutations; another was the complementation or 'cis-
trans' test which led to the recognition of units of genetic
function.
7.1.5 The use 0 f deletion mutants to order point mutations

Deletion mutants show no tendency to revert spontaneously
to the wild type and include those mutants in which large
deletions or alterations of the chromosome have occurred.
Point mutations, on the other hand, revert spontaneously at
measurable rates and behave as if their alterations were
localised at single points and are correctable. Conceivably
they arise when the DNA molecule of the phage undergoes
alteration of a single base pair; often this substitution can be
reversed with relative ease. Point mutations, not deletion
96 BACTERIOPHAGES
---'}
A ---~ IC
2 ---~ ,D ---2
3 ---~ ,E F -- - 3 Deletion mutants
A
4 ------I 1-------4
---~ ~--5
I
5
A B C D
•E F
I
G
Point mutant
::.: ::::
BnD ~~= == ==nt mutant
A E F G Deletion mutant Nö.3
__ _ 'U
C
G ______ u Jo;ot m"toot

- ---------il-----------
A G Deletion mutant No.5
Figure 7.4 The use of Deletion Mutants in Genetic Mapping

A. Aseries of deletion mutants with the extent of their deletions
indicated, and the relative position of a mutational site in a point
mutant. B. Pairing of chromosomes carrying point mutation with
deletion mutant No. 3. Wild type recombinants can result. C. Pairing of
chromosomes carrying point mutation with deletion mutant No. 5. No
wild type recombinants can result since if recombination does occur
with the deletion mutation the point mutation will not be involved.
Hence wild type recombinants may be formed when the point mutant
is crossed with deletion mutants 1, 2 and 3 but not with 4 and 5.
mutations, are the ones that must be mapped if the fine

details of genetic structure are to be probed. However, to test
thousands of point mutations against one another for
recombination in all possible pairs would require millions of
crosses and is obviously impracticable. Benzer's very neat
solution to this problem was to use deletion mutants to
assign the point mutations to certain regions of the chromo-
some and thus considerably reduce the requirement for
carrying out pairwise crosses with point mutations. In essence,
recombination tests are made between two mutants; one
with a point mutation, the other with adeletion. A negative
result, that is no wild type recombinant progeny, indicates
that the deletion overlaps the point mutation; a positive
result indicates that it does not overlap (Figure 7.4).
For his analysis Benzer isolated large numbers of point
and deletion mutants in the rn region and allocated each point
mutation to one of seven segments of the region by means of
a set of deletion mutants as shown. Then he further localised
the point mutations into one or other of 47 segments using
another series of deletion mutants wh ich intruded into the
deletion segments of those previously used. Mutants falling
into each of these small segments were sometimes further
localised by more deletion mutants and were then mapped by
pairwise crossing. Those point mutants that showed recom-
bination were concluded to be at different sites; the order of
sites within a segment can be established by making quanti-
tative measurements of the recombination frequencies with
respect to one another and ordering the point mutations
accordingly at appropriate distances apart on the chromosome.
7.1.6 The limits 0 f genetic fine structure analysis

Down to what level Can the gene be divided? Can it, for
example, be mapped at the level of individual nucleotides or
is the region over which recombination takes place much
larger than that?
When Benzer crossed members of his rn point mutation
population and selectively scored for the frequency of wild
type (rt) recombinants he was able to detect very rare recom-
binants between adjacent sites; theoretically those occurring
at frequencies of 0.0001 per cent. He found, however, that
the lowest frequency at which rt recombinants were 'detected
was 0.01 per cent even though the method was a hundred
times more sensitive than this. Hence Benzer seemed to have
98 BACTERIOPHAGES
discovered the minimum distance between two genetic sites

still separable by recombination. His figures indicated a
physical distance corresponding to about three nucleotides
(.one co don) as the minimum recombinational unit but recent
work has shown that this is not quite truej recombination
within a codon can occur but is relatively difficult to detect.
7.1.7 A utonomous functional units defined by the comple-

mentation or 'cis-trans' test (Figure 7.5)
The hereditary structure needed by the phage in order to

multiply consists of many parts, distinguishable by mutation
and recombination. But, one may ask, is the rn region, because
it controls one characteristic, namely whether or not the phage
can plaque on E. coli K, to be thought of as one or hundreds
of genes? Although one function is involved it could be argued
that growth in E. coli K requires aseries of biochemical
reactions, each controlled by a different portion (gene?) of
the rn region. The absence of any one step would block the
final result, the ability to plaque.
Mixed infection of E. coli K by wild type phage and a
phage carrying two rn mutations still results in the formation
of plaques containing both wild type and doubly mutant
phages. The wild type phage is able to promote the growth
of the rn mutant by making good both functional defects. On
the other hand, when E. coli K is mixedly infected with various
combinations of rn mutants in what is called a 'trans' arrange-
ment: ( T n a- b +) or ( T n a+ b-), neither of which yield plaques
on this host when plated alone, either no plaques or normal,
wild type plaques develop. The plaques that do arise are a
result of complementation, not recombination, since the
phage particles in them usually consist of the two mutant
types, not the wild type.
Using this knowledge, Benzer devised what is known as the
'complementation, test' which measures the ability of mutants
to complement each other, that is each make good the other's
defects. With this test he classified r mutants into two clear-
cut functional groups, A and B, each of which can function
Wild type
T4 phage Mutant l
AO~
Active: Le. plaques produced Inactive: no plaques result
Mutant2 Mutant 3
Inactive Inactive
Active (trans arrangement)
Inactive (cis arrangement)
,,~ -B
~ Cistrons 01 the r II region
AO BO Functional proteins
Figure 7.S The Complementation or Cis-Trans Test

Wild type cistrons produce wild type functional proteins while certain
mutant cistrons cannot. Therefore T4 is active in E. coli K only if both
A and B ci~trons are functional in an infected cel!. In this Figure E.
co li K is infected by T4 phages; either the wild type or mutants carry-
ing gene mutations in the TU region which result in the production of
non-functional protein in either the A or the B cistron. A and B
proteins result only when both cistrons, either on the same or separate
phage genomes, are present in a functional state in any one cello (only
the TU region of the phage genome is indicated)
100 BACTERIOPHAGES
independently of the other. An r mutant in the A region can

complement an rn mutant in the B region but two in the A or
two in the B region ('cis' arrangements) cannot do so. Hence
both A and B represent a single functional unit. These units
were termed 'cistrons'; it can be said that the rn region is com-
posed of two cistrons, the A cistron and the B cistron. The
approximate genetic lengths of these cistrons can be inferred
from the recombination frequencies of mutant loci identified
at their extreme ends, the A cistron being about 6 map units
long and the B cistron about 4 map units long. A map unit is
the genetic distance between two loci that recombine with a
frequency of 1 per cent. This work gave a new perspective to
genetics and it seemed that 'cistron' might replace the old
term 'gene'. Habits die hard, however, and geneticists still
refer to the unit of function as a 'gene' although they may
think of it as a cistron.
7.1.8 Complete mapping

The rn region, now mapped in considerable detail, is but a
small part of the T4 chromosome. A genetic map exists for
the whole of the T4 chromosome, albeit in somewhat less
det~il. It was not all arrived at in the same manner as the rn
regIOn map.
The bulk of the mapping, in fact, was accomplished using
ts and am mutants. R.S. Edgar and R.H. Epstein used
Benzer's cis-trans test to assign more than 1000 of these
mutants to different genes and some 60 genes had been
identified and mapped by the mid 1960's. Edgar, working
with ts mutants, would mixedly infect E. coli with two of
them and maintain it at a restrictive temperature. If the two
mutants complemented each other to produce lysis they were
assigned to different genes; if they did not so complement
each other they were assigned to the same gene. Epstein
similarly tested his am mutants, infecting restrictive E. coli
with them in pairs and noting whether lysis occurred. Finally
the outcome of mixed infection of non-permissive E. coli
with a ts and an am mutant allowed determination of their
mutual complementation capacity and hence the relative
Figure 7.6 The Genetic Map of Bacteriophage T4

Each small division of the inner circle represents 10 3 nucleotide pairs
- the totallength of the T4 DNA molecule determined by electron
microscopy is about 1.66 x 10 5 nucleotide pairs so the agreement is
dose. Gene positions are indicated by lines or stippled bars and
functions of essential genes can be inferred from tue consequences of
the corresponding gene defects, shown in the boxes. Functions or
designations of non-essential genes (all others) are defined on the out-
side of the cirde for example, ac: acriflavine resistance; imm: superin-
feetion immunity. (Reproduced by kind permission ofW.B. Wood
from Handbook of Genetics, Vol. 1, edited by R.C. King, Plenum,
New York, 1974)
102 BACTERIOPHAGES
gene assignment of the two mutant types.

This mapping demonstrated that the entire chromosome
of T4 is a single unit and is genetically circular. By examining
the contents of restrictedly infected ceUs biochemicaUy or in
the electron microscope it was possible to determine the
nature of so me of the restrictive lesions. These are shown in
the boxes on the map (Figure 7.6); for a fuU explanation of
aU the symbols the reader must consult Professor Wood's
article. Genes relating to·the same orassociated functions
are seen to be clustered but, and the significance of this is
considered more fuUy in the section on 'Control Mechanisms',
clusters relating to similar functions may be found on widely
separated segments of the map.
Another phage weIl characterised genetically is lambda;
special attention has been centered on the nature of the link
between the prophage and the bacterial chromosome. Early
genetic studies were carried out in the 1950's by F. Jacob and
E.L. WoUman using characters such as host range, plaque size
and plaque morphology. Other mutant types studied more
recently include the c mutants which are fairly readily
isolated and recognised. They interfere with the ability of
the phage to lysogenise so that clear, instead of turbid plaques
result on indicator strains (not mixed). These mutants fall
into three functional groups, Cl' C2 and C3 ' each of which is
now known to be a cistron. More recently A. CampbeU has
isolated and mapped aseries of amber mutants of lambda
which have defined the genetic map (and the behaviour of
prophage) more fuUy. A partial genetic and functional map
is given in Figure 7.7
7.2 Transduction
The importance of bacteriophages in genetics is twofold:
besides providing a model genetic system for study in their
own right, they can mediate the transfer of bacterial genes
from one ceU to another by the process of 'transduction',
which literaUy means 'leading across'. Transduction was
first noted in 1951 by Lederberg, Lederberg, Zinder and
Lively who were trying to repeat with Salmonella typhimu-
rium experiments that had previously been used to demon-
R \
L
m m, ;J
Inducer of
late functions
Y \
Cohesive sites at
end of vege.tative
chromosome Lysozyme
Figure 7.7 A Partial Genetic and Functional Map of Lambda Phage

Genes Cl, C2 and C3 are concerned with the establishment of prophage
and Cl is also concerned with the maintenance of lysogeny; hand i
determine host range and immunity respectively; the capitalletters
represent sites of genes based on the analysis of amber mutants by
Campbel!.
strate conjugation in E. coli K12. When they mixed two

auxotrophic strains of S. typhimurium, unable individually
to grow on minimal medium, prototrophic recombinants
able to grow on minimal medium arose. However recom-
binants resulted not only when cultures of the two Salmonella
strains were mixed but also when a culture of one strain was
treated with a cell-free filtrate of the other. This was in
104 BACTERIOPHAGES
distinct contrast to conjugation where cell-to-cell contact is

necessary. Zinder and J. Lederberg then turned their
attention to the nature of the agent in the filtrate and soon
rec.ognised it as a phage, named in this instance 'P22'. Trans-
duction has since been reported in genera as diverse as
Bacillus, Escherichia, Proteus, Pseudomonas, Shigella and
Staphylococcus.
Minimal medium Minimal medium
O
- prototrophic
~.:i.':
colonies deve/op
-no growth .
:~~((
Salmonella typhimurium
Strain A
(110··
. ....
(21 0:.
. ..
•
O-~ Baeteria -free filtrate

of Salmonella typhimurium
strain A culture
Figure 7.8 Procedure Adopted by Lederberg, Lederberg, Zinder and

Lively to Demonstrate the Process of Transduction, as Distinct from
Conjugation, in Salmonella typhimurium
Cultures are plated either singly or in various combinations on minimal
medium. The prototrophic colonies arising from combination (1) could
be due to a process similar to conjugation; combination (2) shows that
this is not so.
7.2.1 The mechanism o[ transduetion

When a phage lysate is prepared from a donor bacterium only
a small proportion of the progeny particles (10- 5 to 10-7 ) can
transmit genetic characteristics to a subsequent host. These
transducing phages carry fragments of donar chromosome in
place of their own viral genes. Hence these phage particles are
defective; they cannot reproduce themselves.
Often transduction can occur for any genetic markers of the
donor bacterium. However, since the size of the DNA fragment
that can fit inside a phage head is limited, only closely linked
markers are co-transduced by the same particle. The amount
of DNA in the viral genome is known to be about one
hundredth of that in the E. eoli or Salmonella chromosome
and consequently it seems reasonable to suppose that the
phage cannot transduce donor fragments appreciably larger
than this. Studies on the thr-Ieu-azi region of the bacterial
genome of E. eoli K12 have revealed that the thr-Ieu portion
represents the extreme length of donor fragment which can
be transduced by PI phage; this portion constitutes about
one fiftieth of the E. eoli genome.
7.2.2 Types o[ transducing phage

There are two types of transduction; generalised and restricted.
In generalized transduction any small region of the donor
chromosome can be transduced as in the P221 S. typhimurium
system outlined earlier. Restri.cted transduction on the other
hand is limited to phage that can lysogenise bacteria; transfer
of chromosomal genes is restricted to a cluster of genes
adjacent to the prophage location on the bacterial chromo-
some. Restricted transduction, therefore, is possible only with
lysates produced by induction of a prophage and not (in
contrast to generalised transduction) with lysates produced
by lytic infection. A good example of a restricted transducing
phage is lambda. On induction lambda prophage will occasion-
ally undergo exchange with the nearby gal segment of the
bacterial DNA and particles defective as phage but able to
transduce the gal genes will result. These particles are called
Adg (for lambda, defective, gal) and constitute only aminute
106 BACTERIOPHAGES
Generolised tronsducing phoge Restricted tronsducing phoge
S. ~
Infection of
bocteriol
cells
I
~~g~me gol+ Cell is
Iysogenrsed
_ _ by A-prophoge
I which is inserted
_-----~ by recombinotion
C=~ followingpoiring
goi
G 9
\ Very occosionolly,
0 n induction,o
~
901+ ~efective p~age
rncorporotlOg
the goi region
may be formed
G
\ by iII~itimote
~
Breokdown:(f Vporrrng
( _- ___ -
_
- ----)
_ _
bocteriol
chromosom
r - -
'--- _ _
~
~
t +
©D pno?J;~~::io"C?oq6Q)
t
Progeny
releosed
All progeny con tronsduce
Figure 7.9 Production of Transducing Phage

In the production of generalised transducing phage a proportion of the
phage produced by a cell may contain a fragment of the bacterial
genome and hence be able to transduce that character. Restricted
transducing phage results when the cell is lysogenised by Aprophage
inserted into the chromosome; on induction an occasional prophage
exercises abnormally, incorporating the bacterial gal+ gene. It replicates,
and all the progeny in that cell can transduce gal but, since such
abnormal excision is rare in a population, an LFT lysate results.
fraction of particles in the lysate, which is called an LFT (low

frequency transduction) lysate
7.2.3 Partial diploids

If a culture of agar mutant of E. coli K12 (one which cannot
utilise galactose), which may be either sensitive to or lysogenic
for lambda, is infected at high multiplicity with an LFT
preparation of "Adg phage obtained by induction of a lysogenic
;gal+ strain, clones of gal+ bacteria can be recognised after
growth on agar containing galactose together with suitable
indicator dyes. The majority of these gal+ transductants differ
from the original gal+ donor bacteria in two respects. Firstly
they are unstable for the gal character and gar progeny are
segregated with a probability of 1O-3 /cell/generation. Trans-
ductants must therefore be diploid for the gal region since
the gar gene must have persisted in order for it to segregate
later. Hence the transduced marker does not replace its
recipient allele but is added to the recipient genome. Partial
diploids of this sort are called 'heterogenotes'; they are
usually doubly lysogenic, containing both "Adg and regular
lambda, and are unstable for the "Adg region. Secondly, when
such cells are induced by ultraviolet light, they yield a high
frequency transduction (HFT) lysate, in which about half
the particles are "Adg and half are normal phage. Regular
lambda has to be present in these ceIls to act as a helper,
providing the functions of genes missing in "Adg.
7.2.4 Complete and abortive transduction (Pigures 7.10 and

7.11)
Transduction is said to be comptete when the fragment of
donor bacterial chromosome (called an 'exogenote') introduced
by the phage is inserted, by recombination, into the chromo-
some of the recipient cell. The inserted markers are then
108 BACTERIOPHAGES
c
o
....,
U
:J
"0
U'>
C
o
....,L..
QI
U;
Q.
E
o
U
c
o
....,
QI U
> :J
.- "0
tU'>
o C
.DO
«.!:;
Figure 7.10 Mechanisms of Complete and Abortive Transduction

In complete transduction (A) the fla + fragment is in corpora ted and
replicates so that all progeny are equally motile. In abortive trans-
duction (B) the fragment is inherited unilinearly; in those progeny
that receive it flagella are fully developed, in those that do not the
character is gradually diluted out.
transmitted regularly 10 all progeny. Sometimes however, the

exogenote does not become inserted into the bacterial
chromosome, does not replicate and is transmitted unilinearly
so that at any one time only a single cell of the clone of cells
produced from the transduced cell possesses the fragment.
This phenomenon is known as 'abortive' transduction. The
exogenote still expresses the function of the gene or genes it
possesses and it follows that the gene products (enzymes or
enzyme-forming systems) are synthesised in every cell through
which it passes. These products may be transferred cytoplas-
mically; derivatives of the cell may go on dividing for a
limited number of times until the gene products have been
diluted out, exhibiting what is known as 'phenotypic lag'.
Abortive transductants for a nutritional marker form
minute colonies on minimal medium~ these contrast with the
larger colonies of complete transductants. On the other hand,
when genes concerned with motility are transduced complete
transductants yield swarms of growth in soft gelatin agar
while abortive transductants give trails of colonies leading out
from the point of inoculation.
Transduction has proved an extremely valuable tool in
mapping the bacterial chromosome: the degree of linkage of
genetic markers can be inferred from the relative frequency
with which they are co-transduced; the greater the frequency
of co-transduction the closer the linkage. This enables the
bacterial chromosome to be mapped in finer detail than is
possible with conjugation experiments. In the latter, the
position of any particular gene can be determined with an
accuracy of about 1 per cent of the totallength of the
bacterial chromosome. Transduction allows the position of
a gene to be determined within a 1 per cent fragment so that
the two methods are complementary. The loci determining
tryptophan synthesis in Salmonella were mapped using the
transducing phage P22 as genetic vector, while the pathway
110 BACTERIOPHAGES
Figure 7.11 Transduction in Salmonella typhimurium

A non-motile strain (j1a-) of S. typhimurium was treated with phage
lysate from S. typhimurium LT2, a flagellated (j1a +) strain, and
inoculated in a broad streak (hatched) on a plate of semi-solid gelatin
medium. After 24 hours incubation at 37°C trails of small colonies of
abortive transductants (a) could be seen emanating from the inoculum
together with fla + swarms (b) each derived from one completely
transduced cello (Drawn from a photograph by Dr. S.E. Smith)
of L-arabinose utilisation in E. coli was shown, using mutant

E. coli strains and a transducing phage, to be mediated by
three genes, each specifying one enzyme, and a regulator gene
whose function was thereby defined.
Whether or not transduction occurs to a significant extent
in Nature is a matter for conjecture. It is widely argued that
bacteria tolerate phages because transduction confers evo-
lutionary flexibility. Restricted transduction would seem to
give the bacterium a small return for harbouring the seeds of
its own destruction. The fact that phages and their hosts, both
lysogenic and sensitive, can be isolated from the same environ-
ments suggests that some delicate equilibrium exists in which
each is necessary for the other's survival.
7.3 The wider aspects: A miscellany
7.3.1 Control mechanisms

In the 1930's and 40's Energy and Phosphorylation were the
centre of biological attention. Throtigh the 50's and 60's,
DNA and the Genetic Code reigned supreme. Now emphasis
is shifting to Control, with the ultimate objective of under-
standing Cell Differentiation and Morphogenesis. Somewhere
over the horizon, the other side of the millenium probably,
looms comprehension of the Brain and Consciousness - too
far away to concern us here. The basic problem of control is
as follows. At any particular time an organism may require
only part of its genome to function and different parts may
be required to function in a programmed sequence as, for
example, when a cell goes through the complexities of
division. If the organism is multicellular, different parts of
the genome may be required to function depending on the
differentiation of the ceIls, yet all the cells contain the same
genome, derived by successive replication from that of their
original unicell (spore, zygote, egg... ). So how are the required
parts of it promoted and the remainder suppressed in the
appropriate areas of space and time? The control systems that
undoubtedly exist in higher organisms res ist experimental
analysis because of their size and complexity. Even bacteria
have proved to be less amenable than was originally thought.
What more natural then, than to seek control systems in those
smallest and simplest of self-replicating entities, the phages,
and try to und erstand these. When something has been
accomplished in this direction, pro- and eu-karyotic cells
might be tackled with more elan.
The very speed of phage multiplication and the fact that
virion proteins are produced in approximately the proportions
required are advanced as indications that control systems
operate generally. The classification of virus-induced proteins
into 'early' and 'late' supports this. The case of lysozyme,
however, is ambiguous. Clearly it must function last of aIl,
otherwise unassembled phage components would be spilled
out prematurely into the medium but that does not mean that
112 BACTERIOPHAGES
it is necessarily synthesised last. A cell may contain a high

concentration of lysozyme yet fail to lyse. A very plausible
explanation (the 'equiIibrium hypothesis') is that lysozyme
is synthesised weIl before the end of the developmental process
and it begins to dissolve away the cell wall at once but that
repair is continuously effected until the cell's capacity for
re-synthesis is reduced to a criticallevel as a result of the
infection and lysis then ensues.
Some putative control mechanisms are analysed in Table
7.2. Positive control implies that a gene cannot be expressed
until the product of another gene is present; negative control
Table 7.2
Control mechanisms in phage development: some presumptive examples
Level oi control Type oi control Example
Injection Multi-stage T5
Replication of Negative P2
the genome
Transcription Positive Gene 1 of phage T7
Negative Lambda repressor
Translation Negative RNA-phage replicase
Gene-product Dynamic equi- Lysozyme
function librium
means that Gae gene is not expressed because the product of

another gene is present. Gene Cl of phage lambda produces a
protein (the 'repressor') that binds specifically at two points
on the prophage and prevents transcription of all the other
genes. This is perhaps the best known example of negative
control. In RNA phages the accumulation of coat protein
inhibits the translation of the replicase gene. Gene 1 of phage
T7 produces a protein that is thought to bind to RNA poly-
merase and alter its specificity so that it no longer transcribes
early genes but transcribes late genes instead. This can be
regarded as an example of positive control. Given that positive
and negative control are both possi1]le, one can envisage a
general "cascade" mechanism of regulation in which a

gene product specifically shuts off genes no longer required
and promotes the next in sequence when its concentration
reaches a certain level. Evidence for other positive and
negative controls is gradually accumulating but R. Calendar's
recent review of progress in this area [4] reveals that no
comprehensive control scheme is yet in sight for any phage.
At one time it seemed possible that the order of gene
expression might be a reflection of their linear order in the
nucleic acid molecule, wh ich passed through some cellular
organelle that read it as a computer reads a tape, with
appropriate pauses. This attractive hypothesis is seemingly
contradicted by the genetic map of phage T4 (Figure 7.6)
on which genes for related functions are often very far apart
and which bears no relation to the developmental sequence
in any circular permutation. Thus the cascade mechanism
depending not on linear order but on specific product binding
to other genes reached by diffusion, is now a more appealing
model. There is so me evidence, ltowever, that linear order
may be important in an entirely different mechanism in
phage T5. This phage injects its genome, not as one piece but
as two (or more?) pieces separated by a delay. This could be
a form öf control which limits injection to those genes
required at the time, leaving those not yet needed in the
phage head where they cannot function.
In at least one case, replication of the genome seems to be
under some sort of negative contro!. Superinfection of a
lysogenic bacterium which carries P2 prophage with avirulent
P2 mutant yields only the virulent phage. This is not a con-
sequence of prophage attachment because if the ceIl is
superinfected simultaneously with the virulent mutant and
another phage of the prophage type, again only the virulent
mutant is produced, i.e. the superinfecting temperate phage
is blocked as completely as its prophage. Hence all the
positive elements must be present, otherwise the virulent
mutant would not grow, and control of the temperate phage
must be negative.
114 BACTERIOPHAGES
7.3.2 Tbe Qß equation

Cortventional molecular biology assumed that all genetic
information was primarily stored as DNA, transcribed into
RNA and thence translated into protein so the existence of
viruses with an RNA genome was rather embarrassing at first
but has since turned out to be a blessing in disguise. A phage
genome consisting of a single RNA molecule can often be
extracted in one piece, and used to prime the synthesis of all
the proteins that it is known to specify in cell-free extracts
of E. coii. It thus earns the tide of 'polycistronic messenger'
and, more than any other natural messenger, has been used to
study details of translation such as polarity (from which end
does translation commence?) and termination (how is the
complete specification cut into one-protein pieces?). These
matters have been extensively reviewed [9] .
Phage Qß is a favourite experimental material on ac count
of the stability of its replicase. There appear to be only three
Qß proteins: the co at protein and the A-protein in the virion
and the repliease in the infected cello The RNA molecule in
the virion is known to contain about 3300 nucleotides,
sufficient to code for 1100 amino acids. The coat protein
monomer consists of 129 amino acids and the A-protein
accounts for 320 or so more, leaving 651 for the replicase.
Pure (as opposed to 'purified') replicase is not yet available
but first estimates suggest that it has at least twice that
number. There may be other Qß proteins as yet undiscovered
and equally there may be RNA sequences that serve some
purpose other than to be translated into protein but the over-
all impression gained from these figures is that an RNA phage
is likely to become the first organism about which nothing
remains to be learned. It has been suggested that the replicase
molecule is composed of two identical subunits, which would
reduce the co ding requirement, or that it is in part derived
from host-coded protein. The surprising thing is not that the
equation is as yet out of'balance but that the present state
of knowledge allows a balance sheet to be drawn up at all.
Two /ac-bearing genomes
t ra nsducing!g~ bacteriophage Phage A
transducing!g~bacteriophage
phageableto~g!t ransducebugs
phageableto~glt ransduce bugs Phoge B
Hybrid DNA
structure
lac Isolated lac gene

IQ~
Figure 7.12 Scheme for the Isolation of Pure Lac Operon DNA
7.3.3 Isolation of a 'gene'

The first isolation of a length of DNA corresponding to a
specific biochemical function (the lac operon) was accom-
plished using two transducing phages [14]. Both, as a result of
genetic manipulation had acquired the lac operon from their
bacterial hosts but inserted into their chromosomes in opposite
directions. Their DNA's were extracted and heated to separate
the strands of the double helices. These were then mixed and
annealed together, forming a hybrid structure as shown schema-
tically in Figure 7.12. Since the only region of homology was
the lac region, this was the only double-stranded part of the
hybrid. The four single-stranded ends were then digested away
with an enzyme (fungal nuclease) specific for single-stranded
DNA, leaving the lac operon in isolation. The process could
116 BACTERIOPHAGES
be monitored by electron microscopy which showed the four-

ended hybrid clearly. The brevity of this account must not be
allowed to impart a false impression of simplicity. This was
a technological triumph of great complexity requiring skills
and resources of a high order.
7.3.4 Grams of lac repressor

E. coli does not contain enzymes for fermenting lactose unless
they are induced by the presence of that sugar because the
lac genes are repressed by a protein (the lac repressor). Unfor-
tunately this interesting material is present in normal cells in
only vanishingly smaH amounts so that isolation of enough
to study is very laborious. The isolation has been greatly
facilitated by infecting the ceHs with a lambda phage that
has lost its late genes (those responsible for lysing the host) in
exchange for the lac genes of a previous host's chromosome
and is, in addition, temperature-inducible. On induction by
temperature-shift, many copies of the phage genome (including
the lac gene) are made. The ceH does not lyse so these serve
as additional templates for the synthesis of lac repressor,
resulting in such an increase in the amount formed that gram
quantities are not nowadays remarkable.
7.3.5 Precellular evolution

Purified replicase from Qß.-infected E. coli can use Qß-RNA as
a template to synthesise copies in vitro when provided with a
mixture of nucleotides and other defined chemicals. There is
no doubt that the entire Qß genome is replicated because
when this new RNA is applied to E. coli spheroplasts (cells
from which the wall has been removed), they become infected
and complete, plaquable Qß particles are formed. This
replicase (or 'RNA- dependent RNA polymerase', to give it
its full title) is specific for Qß-RNA, as predicted by S.
Spiegelman on the grounds that if it were not it would repli-
cate all the multifarious RNAs found in the bacterial cell
which would be of little value to the virus. Thus we have
replication of macromolecules in vitro, a situation which,
according to Spiegelman [15] , 'rnimics an early, precellular
event, when the environmental selection operated directly on

the genetic material rather than on the gene product.' One of
the most mysterious phases in evolution thus became access-
ible to experimental investigation.
Spiegelman designed an experiment to see whether selective
pressures in the test tube would evoke mutant nucleic acid
molecules under conditions in which translation into protein
did not occur. Essentially, the experiment consisted of serially
diluting the reaction mixture in selective, unprimed (RNA-Iess)
medium with such a short time interval between dilutions
that the original RNA species was diluted to extinction,
leaving only the faster-replicating mutants that arose. In one
experiment the medium was made selective by the use of
sub optimal nucleotide concentrations; in another an anti-
metabolite dn.ig was added. In both experiments mutant
RNA's did indeed emerge, having replication rates substan-
tially higher than that of the original when tested separately.
Spiegelman suggests that the differences in replication rate
may reflect different affinities for the replicase due to the
different secondary structures assumed by the mutant RNA
molecules. He further points out that there is a possibility of
using similarly modified viral nucleic acids therapeutically,
to compete with and arrest the replication of an established
pathogen.
7.3.6 The bacteriophage model

The ease with which the bacterial hosts can be manipulated
has allowed phage knowledge to get so far ahead of the rest
of virology that it is natural for plant and animal virologists
faced with a particular problem to look to related work with
phages as a source of ideas. It is best to think of the bacteri-
ophage model not in the sense of an ideal virus but as a basis
for speculation. It is quite legitimate to speculate on the
extent to which phages and other viruses may share a common
property provided that such speculation, without evidence,
is not mi staken for discovery. One may note, for example,
that bacteria can acquire new properties on transduction or
lysogenisation with phage and that animal ceHs can become
118 BACTERIOPHAGES
cancerous on treatment with other viruses and ponder on

whether a similar mechanism underlies both phenomena. At
the moment, such evidence as exists suggests that the two
processes are dissimilar - one cannot, for example, induce
provirus in cancer cells as one can in lysogenic bacteria. The
general term 'provirus' was coined many years ago but the
only proviruses substantiated to date are prophages. Where
the bacteriophage model is of considerable value is in the
testing of new techniques and the teaching of existing
methods if an that is required is a particle of viral compo-
sition and dimensions which is readily recognised and
enumerated using the plaque assay. Centrifuges, filters and
electron microscopes can an be proved with a phage
preparation and for the biology teacher, phage may be the
only available material for practical virology - the green-
house, the Horne Office licence, the CO 2 incubator, the
candling box, an the cumbersome trappings of eukaryote
virology that make the simplest demonstration time con-
suming and expensive, can be dispensed with and the
student can learn the elements of virology with only test
tubes, petri dishes and so me simple media.
8 Applied aspects
8.1 Phage nuisances

Whereas phages are virus es of minor importance compared to
those that infect Man and his domestic plants and animals,
there are certain industries where bacteriophage infection is
a nuisance, if not economically disastrous. These are industries
that depend at some stage in their processes on changes
brought about by specific bacteria.
8.1.1 Cheesemaking
In cheesemaking a culture of lactic streptococci known as
'starter' is first of all added to the milk. These bacteria
convert the milk sugar (lactose) into lactic acid. The acidity
of the milk is monitored closely during the process because
the cheesemaker knows that unless the acidity has risen to
the criticallevel for each subsequent stage of the process the
cheese will be abnormal, possibly not of marketable quality
if, indeed, it forms at all. Should bacteriophage able to attack
the starter cu hure get into the vat the bacteria may be
destroyed in a few hours and the acid level remains stationary.
The cheesemaker refers to such an occurrence as a 'pack-up'
because that is what he has to do. There is no point in adding
more starter because, in destroying the first lot, the phage
will have muhiplied many thousandfold and the milk then
119
120 BACTERIOPHAGES
Figure 8.1 A Cheesmaking Vat

The curd is being cut. In cases of phage infection this stage may not be
reached. (Courtesy of Unigate Ltd.)
APPLIED ASPECTS 121
teems with phage ready to destroy subsequent additions even

more quickly.The milk is thus wasted.
Two main remedies for the phage problem have been
developed. One is to use, not a pure culture of bacteria but
a mixture of several acid-producing strains. Since it is unlikely
that any phage would have a sufficiently wide host range to
infect them all, the usual result of phage infection is that only
part of the bacterial population is eliminated and the process
can proceed to completion, albeit slowly. It is just possible
that the vat could become infected with several phages active
against all the bacterial strains but such occurrences, fortun-
ately, are rare.
The other solution is more sophisticated. Phage infection
may occur through airborne contamination or infected
apparatus associated with the cheese-making vat; it is much
more serious if it occurs in the smaller vessel in which the
bulk starter is grown. Most of the phages likely to be
troublesome in cheesemaking require calcium ions for
infection. This makes it possible to design a phage resistant
medium (PRM) so that calcium is absent or held by chelating
agents in an unavailable form.
Mixed starters and PRM may be employed together but
there is still a need for good plant hygiene; careful disposal
of whey, which maycontain much phage, is important. The
build-up of phage infection in a creamery can be a slow and
insidious affair so that it is not noticed until it is firmly
established.
8.1.2 Antibiotic fermentations

Certain antibiotics, of which streptomycin is a prime example,
are produced by fermentation using actinomycetes. These are
small filamentous organisms which have the same diameter as
true bacteria and the same prokaryotic internaiorganisation.
They are attacked by viruses which are often indistinguishable
regarding their morphology and general behaviour from
bacteriophages and are known as 'actinophages'. Actinomy-
cetes require highIy aerobic conditions and the actinophage
used to gain access to the fermentation via the air suppIy.
122 BACTERlOPHAGES
Figure 8.2 Starter Culture Inoculation into a Bulk Starter Vessel

A shrouded, double-ended hypodermic needle is inserted through
rubber seals on both containers for aseptic transfer (Lewis's method).
(Courtesy of Unigate Ltd.)
APPLIED ASPECTS 123
It would then multiply rapidly, lysing or inhibiting the

culture and causing serious production losses. Being very
small, the actinophage could pass through the slag wool or
carbon granule filters that were perfectly adequate for
removing bacteria. The problem has been largely eliminated
nowadays by improvement of the filtration media or use of
he at sterilisation instead.
8.1.3 Solvent production

During the first World War and for some years afterwards,
acetone and butanol, two important industrial solvents, were
produced by a process using a bacterium, Clostridium
acetobutylicum, to ferment sugary materials under anaerobic
conditions. Bacteriophage was a frequent cause of slow
fermentation and low yields. Phage was poorly understood
in those days and one plant became so thoroughly infected
with phage that production was no longer economic. It was
abandoned and another plant erected many hundreds of miles
away. The process is hardly used at all nowadays, most
acetone and butanol being made chemosynthetically, but it
is worthy of note as one of the first applied phage problems.
8.1.4 Baking and brewing

Several accounts have been published of a virus that infects
yeast and may be the cause of the 'degeneration' of cultures
used in these industries. It is called 'zymophage'. The evidence
for it is tenuous.
8.2 Detection and identification of pathogenic bacteria

Phages have their positive side: although no longer contem-
plated as therapeutic agents, phages have an established place
in medicine as diagnostic tools. Phage-typing is routinely used
for the identification of pathogenic bacteria of which staphy-
lococci and typhoid bacilli are prime examples. The techniques
are delicate, requiring a high level of skill and judgement but
in capable hands can distinguish between strains of bacteria
inseparable by other methods. Analogies have been drawn
between this and the fingerprinting of human beings.
124 BACTERIOPHAGES
Figure 8.3 Determination of Routine Test Dilution (RTD)

Drops of phage dilutions increasing anticlockwise from 12 o'clock
were applied to the bacteriallawn. RTD (arrowed) is the last to show
confluent lysis. (Courtesy of Dr. E.A. Ashesov and Central Public
Health Laboratory, Colindale)
8.2.1 Salmonella typhi

Virulent strains of this organism may be identified using one
remarkable phage, the 'Vi-phage', which attacks many strains
but shows enhanced virulence towards the strain on which it
was last propagated. This phenomenon is known as 'host-
induced phage modification'. It is due to a phenotypic, not
genotypic change in the phage brought about by the methy-
lation of DNA bases in patterns that protect the phage DNA
from host enzymes that would otherwise destroy it ('restrictive
APPLIED ASPECTS 125
Figure 8.4 Mechanised Phage Typing

A. Lidwell's Applicator. The wire spirals a pick up phage suspension
from the wells in plastic block band deposit it on the lawn of bacteria
to be typed in dish c. They are re-sterilized over the multiple gas-
burner d. B. Resulting patterns of virulence on three different bacterial
strains. (Courtesy of Dr. E.A. Ashesov and Central Public Health
Laboratory, Colindale)
126 BACTERIOPHAGES
enzymes'). The molecular basis is as yet far from dear but

this does not detract from the practical value of Vi-phage.
Stocks of Vi-phage, propagated on known strains, are main-
tained and, when applied to an unknown strain at the
appropriate dilution, its identification is possible by matching
patterns of virulence.
8.2.2 Staphylococcus aureus

No system comparable to Vi-ph-age is available for this
pathogen; instead a large number of different phages, differing
in their host ranges is used. They rarely attack only a single
strain of staphylococcus so it is necessary to match patterns
of virulence on a lawn of the unknown strain as shown in
Figure 8.4. Each phage is applied at its Routine Test Dilution
(RTD), which is that which just produces confluent lysis of
the propagating strain. Lidwell's Applicator is a device for
applying a set of typing phages quickly and accurately to
many different lawns of bacteria.
8.2.3 Seed-borne plant pathogens

Phage can be used to detect the presence of bacterial plant
pathogens in seeds. The seed is incubated in nutrient broth
with a small, accurately known amount of phage specific
for the pathogen being sought. If the phage is found sub-
sequently to have multiplied, it may be inferred that the
host pathogen was present.
8.3 Radiation dosimetry

An entirely different application for phage is as a measure of
radiation dosage. The radiation sensitivities of certain phages,
T2 for instance, are known very accurately. By induding a
known amount of phage in a material that is to be irradiated,
or irradiating a phage control under identical conditions, the
radiation dosage may be calculated from the residual infect-
ivity. This is very valuable, giving a direct measure of the
biological effectiveness of a radiation treatm~nt that may
defy calculation from physical and chemical data in complex
systems.
9 Conclusion
The reader should by now have a fair idea of the nature and
extent of bacteriophage biology; the way in which it is
conducted; some of its achievements and ambitions. A
selective bibliography is appended for those who wish to
study it further. For anyone who has not yet read biology or
biochemistry up to degree standard, Stent's 'Molecular
Genetics' is strongly recommended, and Mathews' 'Bacterio-
phage Biochemistry' for the more erudite. After that, the
phage literature being so vast, some measure of specialisation
is advisable, if not already dictated by individual circumstances.
The student may be fortunate enough to find a review of his
particular field of interest in one of the specialised review
journals such as 'Bacteriological Reviews', 'An nu al Review
of Microbiology' or 'Advances in Virus Research'. To keep
abreast of current events it is advisable to read at least one
virological joornal thoroughly, 'Journal of General Virology'
say or 'Virology' and to maintain surveillance of the rest by
means of 'Current Contents', a journal that simply lists the
titles of papers in other journals. Promising titles may be
checked efficiently some months later in 'Biological Abstracts'
or 'Virological Abstracts' to see whether the paper is worth
reading in full. Finally one can find out precisely what is
going on in a particular line by selecting a few recent papers
of central importance and looking up in the 'Citation Index'
who has subsequently cited them.
127
128 BACTERIOPHAGES
It is important for biology students and teachers, even

those with no research ambitions, to maintain physical
contact with the material of their philosophy. Priests of old
claimed to derive superhuman knowledge from the inspection
of entrails - Science now claims to have discovered the
secrets of Life through the study of the invisible parasites of
E. coli, the invisible inhabitant of Man's bowels! Unless the
general ron of students find that they can verify at least some
of the doctrines of the phage-priests, Biology may become
an edifice of sacred dogma based, like an ancient religion, on
something that can be neither seen nor questioned except by
the chosen few. The humblest plaque or dilution assay made
by one's own hand is a confirmation of the credibility of our
latter-day catechism; the present author hopes that he has
shown the way to begin.
Appendices
APPENDIXA
UNITS OF MEASUREMENT IN ELECTRON MICROSCOPY
Linear dimensions of bacteriophages and comparable objects are given

in nanometers (nm). This unit is superceding the Angstrom Unit (0·1 nm)
and the millimicron 0·0 nm) for such purposes. 1 nm = 10-9 netres.
APPENDIX B
MEDIA FOR BACTERIOPHAGE STUDIES
Many special media have been described which give superior results in
particular applications. Details of these must be sought in the relevant
literature; here only general purpose media will be considered. A medium
in which the host grows weIl is usually suitable for the propagation and
assay of its phages provided that there is sufficient available calcium and
magnesium and that the pB is not extreme. For coliphages the following
medium is recommended:
grams per litre
Nutrient broth powder (Difco) 8
Sodium chloride 5
Agar (Difco) for solid medium 15
(for sloppy medium 7)
Autoclaving: 121°C for 15 minutes
The above medium is unsuitable for organisms that must have carbo-
129
130 BACTERIOPHAGES
hydrate, for example some streptococci. For these the following may
be useful. It contains glucose and a buffer to prevent it becoming acid
which does not render calcium unavailable.
Tryptone (Oxoid) 10
Yeast extract (Difco) 3
Glucose (bacteriological) 2
Sodium glycerophosphate 5~ H2 0
(B.D.H. Laboratory Grade) 10
Tris-(hydroxymethyl)-aminomethane 1
Calcium chloride 0.25
Agar (Oxoid No. 3) for solid medium 10
(for sloppy medium 4)
The pH is brought to 7·S with a few drops of dilute lactic acid. In a

domestic pressure cooker 15 pounds per square inch (121 0 C) for 15
minutes is required for sterilisation. In larger autoclaves a shorter
time, which must be determined experimentally, will be necessary to
avoid charring the glucose; alternatively the glucose may be sterilised
separately and added to the medium aseptically when cold.
APPENDIXC
THE PRINCIPAL PHAGE COLLECTIONS IN BRITAIN
The undermentioned organizations maintain stocks of phages for

distribution to bona lide workers. A fee may be payable.
Phages 01 industrial, research and general interest

National Collection of Industrial Bacteria,
Torrey Research Station, P.O. Box 31,
13 5 Abbey Road,
Aberdeen AB9 SDG,
Scotland.
Phages of bacteria of medical importance

National Collection of Type Cultures,
Central Public Health Laboratory,
Colindale Avenue,
London NW9 5HT.
APPENDICES 131
Phages of bacteria from milk and dairy products

National Collection of Dairy Organisms,
National Institute for Research in Dairying,
Shinfield, Reading,
Berkshire.
Phages of bacteria pathogenic to plants

National Collection of Plant Pathogenic Bacteria,
Plant Pathology Laboratory ,
Ministry of Agriculture, Fisheries and Food,
Hatching Green,
Harpenden, Herts.
APPENDIX D
SELECTED BIBLIOGRAPHY
1. Books on Bacteriophage and Related Subjects (an indication of

their nature is given where this is not evident from the title)
Adams, M.H. (1959) Bacteriophages, Interscience Publishers, Inc.,
New York. (Deals with all aspects of the subject, somewhat out
of date but still a valuable work of reference)
Cairns, J., Stent, G.S. and Watson, J .D. (Eds.) (1966) Phage and the
Origins of Molecular Biology, Cold Spring Harbour Laboratory of
Quantitative Biology, New York. (Gives an insight int<> the personal-
ities and attitudes of some prominent phage biologists)
Cohen, 5.5. (1968) Virus-induced Enzymes, Columbia Oniversity Press,
New York and London. (The title is misleading since the book is
almost entirely devoted to phage-induced enzymes).
Hayes, W. (1968) The Genetics of Bacteria and their Viruses, 2nd
edition, J. Wiley and Sons, New York.
Hershey, A.D. (Ed.) (1971) The Bacteriophage Lambda, Cold Spring
Harbour Laboratory, New York.
Mathews, C.K. (1971) Bacteriophage Biochemistry, Van Nostrand
Reinhold Company, New York.
Reeves, P. (1972) The Bacteriocins, Chapman and Hall Ltd., London.
Stent, G.S. (1963) Molecular Biology of Bacterial Viruses, W.H. Freeman
and Co., San Francisco and London. (A very readable and well
illustrated account of the historical development of this subject
from the very beginning)
Stent, G.S. (1971) Molecular Genetics, an Introductory Narrative,
132 BACTERIOPHAGES
W.H. Freeman and Co., San Francisco and London. (Phage occupies
only a modest proportion of this book but is weIl seen against the
background of genetics as a whole)
Tikhonenko, A.S. (1970) Ultrastructure of Bacterial Viruses, Plenum
Press, New York, London.
2. Specific References, referred to by number thus [] in the text.

[1) Bachrach, U. and Friedman, A. (1971) Practical procedures for
the purification of bacterial viruses. Applied Microbiology
22(4),706-715.
[2] Bradley, D.E. (1966) The fluorescent staining of bacteriophage
nucleic acids. Journal of General Microbiology 44(3), 383 - 391.
[3] Bradley, D.E. (1967) Ultrastructure of bacteriophages and
bacteriocins. Bacteriological Reviews 31,230 - 314.
[4] Calendar, R. (1970) The regulation of phage development.
Annual Review of Microbiology 24, 241 - 296.
[5] Campbell, A.M. (1962) Episomes. Advances in Genetics 11,
101 - 145.
[6] Kellenberger, E. (1972) Mechanisms of length determination in
protein assemblies. Polymerization in biological systems
(G. Wolstenholme and M. O'Connor, Editors) Ciba FoundatioQ
Symposium 7 (new series), 295 - 299.
[7] Fisher, W.D., Cline, G.B. and Anderson, N.G. (1964) Density
gradient centrifugation in angle-head rotors. Analytical Bio-
chemistry 9, 477 - 482.
[8] Franklin, KM., Hinnen, R., Schafer, R. and Tsukagoshi, N.
(1974) Biochemical aspects of .the structure and synthesis of
bacteriophage PM2. Proceedings of the Society for General
Microbiology 1(2),36 - 37.
[9] Kozak, M. and Nithans, D. (1972) Translation ofthe genome
of a ribonucleic acid bacteriophage. Bacteriological Reviews
36(1),109 - 134.
[10] Luria, S.E. (1945) Genetics of bacterium - bacterial virus
relationships. Annals of the Missouri Botanic Garden,
32,235 - 242.
[11] Luria, S.E. (1951) The frequency distribution of spontaneous
bacteriophage mutants as evidence for the exponential rate of
phage reproduction. Cold Spring Harbour Symposium on
Quantitative Biology 16,463 - 470.
[12] Richards, K.E., Williams, R.C. and Calendar, R. (1973) Mode of
DNA packing within bacteriophage heads. Journal of Molecular
Biology 78, 255 - 259.
INDEX 133
[131 Rosenkranz, H.S. (1973) RNA in coliphage T5. Nature 242,

327 - 329.
[141 Shapiro, J., MacHattie, L., Eron, L., Ihler, G., Ippens, K. and
Beckwith, J. (1969) Isolation of pure lac operon DNA. Nature
224,768 - 774.
[151 Spiegeiman, S. (1970) Extracellular evolution of replicating
molecules. The Neurosciences Second Study Program (F.O.
Schmitt, Editor) The RockefeIler Press, New York.
[161 Wyatt, H.V. (1974) How history has blended. Nature 249,
803 - 805.
Index
Terms that occur wideiy in the text are defined or explained on the
page given in heavy type.
a- helix 59 androphages 35
A - protein 34 angle rotors 39, 40
absorptiometer 15 antiserum 13, 83
acetone-butanol fermentation
123 bacteriocinogen 75
actinophage 121 bacteriocins 74
adsorption 23,25,26,27 bacteriophage typing,
agar, of pathogens 12 3
semi-solid, sloppy, solid, bacteriophages
soft 8 assay,
134 INDEX
by plaquing 9, virulent 7, 65
by dilution to extinction 16 Bdellovibrio 4, 18
defective 7,73,105,106 Benzer, S. 95
definition 1 bubbIer tubes 19
burst size 13
evolution 6
filamentous 7, 35
helper 107 C-factor 75
homologous 14 caesium chloride (CsCl) 39
male-specific 35 calcium 27,28,80,121
minute 33,35,87 Calendar, R. 113
pure cultures 10 Campbell, A. 71,72, 102
tail-less 33 cancer and provirus 118
Slze 5, 48 capsid 24, 32, 60, 80
species, capsomeres 32
by sigla only (see also cascade mechanism 112
coliphages) cheesemaking 119
f1 35 chemical dissection 80
f2 34,35 cis-trans test 95, 99
fd 35 cistron 99, 100, 102
X(lambda) 22, 68, 70, colicin 5, 74
103, 105, 112, 116 coliphages 5,21,35,129
Xdg 105,107 collar piece 21
<jJX174 32,33,35,57 complementation 95, 99
PI 72 cumulative strain theory 60
P2 67, 112, 113 curing 66
P22 104, 105, 109 dark-fjeld microscopy 2
PM2 3,80 deoxycytidylate hydroxy-
Qß 34,35,114,116 methylase 54
SB 33,35 desiccation 87
SP50IJ 29, 31 dilution to extinction 16
T-even 33, 59, 91 dimers 81 et seq.
T-phages 3, 38 disinfectants 78
Tl 33,35 DNA,
TZ 21 et seq., 35,53, circular 22,
89,126 injection of 23,35,30
T3 33 packing in head 22, 30,
T4 22,50,51,90,95, 32
101,113 replicative form (RF)
T5 21,26,27,112,
57
113 staining of 34
T6 33
doughnuts 49
T7 33,112 Doy, C.H. 88
0712 11, 16, 26
420 79 early messenger 54
temperate 7, 65 early protein 53, 54
INDEX 135
eclipse phase 13, 14,49 et latent period 13, 14,49 et. seq.
seq., 55 lawn of bacterial growth 7
Edgar, R.S. 100 Lederberg et. al. 102, 104
EDTA 27 Lidwell's Applicator 125
efficiency of plating (EOP), ligase 54, 70
synonym of: plaquing looping out 73
efficiency 9 Luria, S.E. 55 et. seq.
electron microscope 3,42, lysates, stock 18
43,44 et seq. lysin, diffusible 11
end plate 22, 31 lysis,
Epstein, R.H. 100 in broth cultures 12,
equilibrium hypothesis on solid media 7,
112 microscopy of 16,
excess antigen 55 premature 13
eXClSlOn 73 lysis from without 27
exogenote 107 lysis inhibition 91
lysogenic conversion 72
genome 55
lysogeny 65 et. seq.
ghost phages 29
lysozyme 7, 54, 111
glass knives 62
glassy transformation 1
magnesIUm 27
head, geometry 24 maps, genetic
d'Herelle, F. 2 T4 101,
Hershey circles 70 A 103
Hershey and Chase 28, 88 Merrill, C. 90
heterogenote 107 minus strand 57
hydroxymethyl cytosine mitomycin C 67,74
(HMC) 52 mixed indicators 10, 92
indicator strain 66 monomers 32, 59
inducing agent 67 multistage injection 35, 112,
induction, 113
techniques of 73, mutants,
zygotic 68 amber (am) 92,
ionizing radiation 80 conditionallethal 92,
isopycnic ban ding 38,40,41 deletion 95,
host range (h) 89,
Jacob and Wollman 102 isolation of 93,
point 95,
Kleinschmidt's technique 47
rapid lysis (r) 89,91,
star 10, 92,
lac operon 115,
suppressor sensitive (sus) 92,
repressor 116
turbid 92
lambda DNA 22,
mutations,
genetic map 103,
permissible 91
lysogeny 65 et. seq., 103
repressor 112 nanometer 129
136 INDEX
nephelometer 15 serum blocking power 53

neutralisation 85 shadow-casting 44
nickase 54 shut-off 54
nonsense codon 94 staining,
fluorescent,
one step growth 13 of nucleic acids 34,
osmotic shock 30 positive 47
partial diploids 107 negative 47
phage- (see bacteriophage) suicide technique 83
phage resistant medium (PRM) support films 44
121 tail,
phenotypic, components 24,
mlxmg 95, contraction of sheath 27
lag 109 tape-measure theory 60
photoreactivation 81 target size 81
plaque, tau particles 51,52,54
assay 8, 9, termination 114
definition of 8, thymidylate synthetase 54
size 11, transduction 102 et. seq.,
as a verb 8(footnote) abortive 107
plaquing efficiency 9 complete 107
plaque-forming unit (p.f.u.) 10, frequency of 107
12 generalised 105,
plus strand 57 restricted 105
polarity 114 transgenosis 90
polycistronic messenger 114 transmissible lysis 1
polyheads 32, 50 tryptophan 27
precellular evolution 116 Twort, F.W. 1
prophage 67, 118
provirus 118 ultracentrifuge 36, 37, 38
ultramicrotomy 61
radio biology 80 ultravoilet (V.V.) irradiation
radiation dosimetry 126 67, 73, 79, 81
rate zonal centrifugation 38
receptor, for T5 26, 27, Vernier model 61
for bacteriocin 75 'Versene' 27
recombination 94 Vi-phage 124
replicase 114, 116, virion,
gene for 112 definition 21,
replicating pool 54 assembly 59
repressor (lambda) 69, 112 Wood, W.B. 101
lac 116
routine test dilution (RTD) Zinder and Lederberg 104
124, 126 zygotic induction 68
zymophage 123
self-assembly 34

Bacteriophages

Uploaded by

Copyright:

Available Formats

Bacteriophages

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bacteriophages

Uploaded by

Copyright:

Available Formats

BACTERIOPHAGES

Springer-Science+Business Media, B.Y.

ISBN 978-0-412-12640-6 ISBN 978-1-4899-3418-5 (eBook)

This title is available in both hardbound and paperback

Distributed in the U.S.A.

The contribution of bacteriophages to the development of

provide them in abundance. They are all much larger and

Brunel University 1974 John Douglas

8 Applied aspects 119

Bacteriophages, usually abbreviated to 'phages', are viruses

taken of Twort's work at the time.

to pass through ultrafine filters of graded pore size, estimates

faster race-horse, then one must experiment with wheat and

move, respire, increase in size and reproduce by fission.

Figure 1.1 Relative Sizes of a Bacterium (E. Coli) and an

biological size range, smallness necessitates simplicity of

Lysis, that is the dissolution of the bacterial cell, is the

2.1 Lysis on solid media

will develop. If drops of concentrated phage suspension are

forms a plaque. So me may infect dead bacteria for example,

were placed in the waterbath at 46°C ± 1°C and 0.1 ml of the

Plaques have other uses besides assay. The plaque character-

Figure 2.2 Diffusible Lysin

forming units (p.f.u.) per ml ean often be prepared by this

Figure 2.3 Lysis in a Broth Culture

grossly swoHen and distorted ceHs. Electron microscopy may

2.2 Lysis in liquid media

Figure 2.5 Instruments for Measuring the Turbidity of Bacterial

amount of light deflected sideways by the cells. The choice of

2.3 Lysis of individual cells

Figure 2.6 Assay by 'Dilution to Extinction'

characteristic purpie colour of uninoculated litmus milk but tubes

They can be covered with a coverslip and examined con-

2.4 Stock lysates

2.5 Bdellovibrio (Figure 1.1)

Figure 2.7 BubbIer Tube

can be maintained in broth cultures of host bacteria and

The infectious phage particle, like the particles of other

roughly hexagonal 'end plate' bearing at each corner a short

have 'adsorbed' to the bacterial ceIl, reversibly at first but

Figure 3.1 Coliphage TZ

Figure 3.2 Anatomy of the Coliphage T2 Virion

Figure 3.3 The Mechanies of Infection by Coliphage T2

Figure 3.5 The Kinetics of Adsorption

Table 3.1 The essential strategy of Hershey and Chase's (1952)

When such a phage is used

Hence, at the moment of

It was established by Hershey and Chase in 1952, in a

Figure 3.6 Adsorbed Phage

removcd by agitation ('sheared off') because it has entered

ghostly. If a phage suspension is suddenly diluted in hypo-

Figure 3.7 Phage End-plate

assembled round it when the virion was formed. Regarding

In the light of modern knowledge of nucleic acid function

3.2 Other tailed phages

3.3 Tail-Iess ('Minute') phages

as well. Single-stranded DNA differs from ordinary double-

Table 3.2 Properties of some prominent coliphages

Group Phage Nucleic Other peculiarities

Phages with T2 2-DNA