2019 Article 2013
2019 Article 2013
2019 Article 2013
https://doi.org/10.1007/s13205-019-2013-0
ORIGINAL ARTICLE
Received: 21 May 2019 / Accepted: 3 December 2019 / Published online: 10 December 2019
© King Abdulaziz City for Science and Technology 2019
Abstract
Endophytes confer unique ecological advantages to their host plants. In this study, we have characterized the diversity of
endophytic consortia associated with the GPU-28 (GPU) and Udurumallige (UM) finger millet varieties, which are resistant
and susceptible to the blast disease, respectively. Whole genome metagenome sequencing of GPU and UM helped to identify
1029 species (includes obligate endophytes) of microbiota. Among them, 385 and 357 species were unique to GPU and UM,
respectively. Remaining 287 species were common to both the varieties. Actinobacteria and other plant-growth promoting
bacteria were abundant in GPU as compared to UM. Functional annotation of genes predicted from genomes of endophytes
associated with GPU variety showed that many genes had functional role in stress response, secondary metabolism, aromatic
compounds, glutathione, and cysteine synthesis pathways as compared to UM. Based on in vitro and in planta studies, Bacil-
lus cereus and Paenibacillus spp. were found to be effective in suppressing the growth of blast disease pathogen Magnaporthe
grisea (strain MG03). In the future, these strains could serve as potential biocontrol agents to reduce the incidence of blast
disease in finger millet crop.
Introduction
13
Vol.:(0123456789)
15 Page 2 of 17 3 Biotech (2020) 10:15
Blast disease is a major constraint in finger millet culti- genotypes viz. GPU-28 (GPU) and Udurumallige (UM). The
vation as this crop is highly susceptible to the disease. The GPU is highly resistant to the blast disease with < 2% inci-
blast disease in finger millet is caused by Magnaporthe gri- dence of both neck and finger blast (Nagaraja et al. 2008),
sea (Hebert) Barr. (Anamorph: Pyricularia grisea Sacc.), whereas UM is highly susceptible (Shirke et al. 2016) to the
which is highly devastating and adversely affects the fin- disease. The whole genome metagenome (WMG) sequenc-
ger millet productivity. The pathogen infects all aerial parts ing approach was used in this study to understand the diver-
of finger millet plant causing leaf, neck, and finger blast sity and function of microbial niches, which could not be
resulting in yield losses, which may often exceed 50% (Sastri addressed using conventional methods (Handelsman 2005;
1950; Esele 2002). Although fungicides like carbendazim, Schloss and Handelsman 2005; Ravin et al. 2015). This
mancozeb, and tricyclazole are effective in controlling blast metagenomic study also helped to identify the putative endo-
disease (Netam et al. 2014), synthetic fungicides are hazard- phytes contributing specific to blast disease resistant variety
ous to human health and also leaves a negative impact on the GPU. Endophytes isolated from these varieties were subse-
environment (Budnik and Baur 2009). Repeated application quently evaluated in vitro and in planta for their antagonistic
of chemicals may also induce resistance in the pathogen. effect on M. grisea strain MG03, which was whole genome
Moreover, chemical control in blast disease management in characterized (Shirke et al. 2016).
finger millet is not viable for resource-poor farmers and in
organic production systems. The pathogen is highly vari-
able and breeding for resistant varieties could only control
Materials and methods
disease in the short run. In this perspective, it is desirable to
identify more eco-friendly, economical, and farmer-friendly
Leaf sample collection
approaches to control the blast disease in finger millet, which
could be equally applied to other crops to control various
Three healthy plants each of GPU and UM cultivars were
diseases.
collected randomly 21 days post sowing from the experi-
The use of biological agents in plant disease manage-
mental plots in the Department of Plant Pathology, GKVK,
ment in agricultural crop production is already in practice.
UAS, Bengaluru, India (13° 05″ N Latitude, 77° 34″ E
However, in recent decades, the novel research on biological
Longitude). The seedlings were raised under natural field
control of plant diseases investigated the role of a particular
conditions and the use of any plant protection chemicals
class of microbes that colonize the internal tissues of the
was strictly avoided during the study. The collected samples
host plant, referred to as endophytes, and hypothesized that
were sealed in sterile bags maintained at 4 °C and processed
they confer host plant resistance against fungal pathogens
immediately upon return to the laboratory.
(Arnold et al. 2003; Gunatilaka 2006; Mousa and Raizada
2013). These endophytes do not cause any apparent disease
symptoms by themselves (Braun and Hirsch 1992). How- Surface sterilization and genomic DNA extraction
ever, considerable evidence has been accumulated sup-
porting the beneficial role of endophytes in host plants in The plant samples were surface sterilized and washed three
promoting plant growth (Verma et al. 2001; Sziderics et al. times in sterile double-distilled water. An aliquot (100 µL)
2007) improved nutrient acquisition from soil (Upson et al. from final wash was plated onto the potato dextrose agar
2009) abiotic stress tolerance (Rodriguez et al. 2008), and (PDA) and nutrient agar (NA) media to confirm the effi-
enhanced production of plant defensive compounds by up- ciency of surface sterilization (epiphytic check). The inocu-
regulating the gene expression of biochemical pathways lated plates were maintained at 28 ± 2 °C for 3 days and
involved in plant defense mechanisms (Gond et al. 2015). only the tissue samples with no growth on epiphytic check
Various studies have reported several mechanisms by which plates were used for isolation of endophytes. The surface-
the host inhabiting endophytes confer host plant resistance sterilized leaf, stem and root fragments were processed
against multiple plant pathogens. They include competition individually as per Sessitch et al. (2012) for extraction of
for space and nutrition (Bolwerk et al. 2005), induction of endophyte cells from plant tissues. Extracted endophyte cells
host resistance genes (Waller et al. 2005), promotion of plant were collected separately and utilized for DNA extraction
growth and physiology (Chaturvedi and Singh 2016), hyper- using the D NASure® Plant Mini Kit (Genetix Biotech Asia
parasitism, predation (Grosch et al. 2006; Dutta et al. 2014), Pvt. Ltd.) as per the manufacturer’s instructions. The quality
and stimulation/production of natural biostatic and biocide and quantity of DNA were assessed using both the Qubit™
compounds (Shankar et al. 1994). dsDNA BR Assay Kit (Thermo Fisher Scientific, USA) and
In the present study, we investigated the diversity and Nanodrop® (DS-11 FX + , D eNovix® Spectrophotometer/
abundance of different endophytic communities coloniz- Fluorometer, DeNovix Inc., USA). The DNA was pooled
ing the host tissues of two different indigenous finger millet and whole genome metagenome sequencing was performed.
13
3 Biotech (2020) 10:15 Page 3 of 17 15
Metagenome library preparation and sequencing sequences). Rarefaction analysis was performed using the
Lowest Common Ancestor (LCA) algorithm by binning
NEBNext® Ultra™ DNA Library Prep Kit for Illumina® reads to taxa based on their BLAST-hits and rooted taxon
(NewEngland BioLabs® Inc., USA) was used to prepare the tree was constructed based on unique operational taxonomic
whole genome libraries of GPU and UM from the DNA units (OTUs) used in the analysis. Taxonomic profiling of
samples. The genomic DNA was fragmented using Cova- endophytes was performed using the NCBI taxonomy data
ris. The end-repaired samples were ligated with adapters sets and phylogenetic trees at species level were generated
and size selected. The library was further enriched using based on the neighbor-joining method using MEGAN (MEta
nine cycles of PCR. After purification, using A gencourt® Genome ANalyzer) software. Figtree (version 1.4.3) soft-
® ware was used to construct the phylogenetic tree.
AMPure XP beads (Beckman Coulter), the library was
quality checked using the Agilent 4200 TapeStation System
(Agilent Technologies Inc., USA) and the average library Comparative metagenome analysis
size was confirmed to be about > 600 bp. Final libraries were
sequenced in AgriGenome Labs (Cochin, Kerala, India) For functional profiling of genes associated with endophytes,
using HiSeq 2500 (Illumina Inc., USA). SEED subsystems and KEGG databases were applied
(Kanehisa et al. 2012). The clustered OTU BIOM table was
Sequence data quality check, pre‑processing normalized using the normalize_by_copy_number.py script
and assembly and processed with the predict_metagenomes.py script. The
predicted metagenome was exported to BIOM format for
The raw fastq paired-end reads were quality checked using further statistical analysis. Statistical Analysis of Metagen-
the FastQC program (https: //www.bioinf ormat ics.babrah am. omic Profiles (STAMP) software version 2.1.3 was used for
ac.uk/projects/fastqc/) and the low-quality sequence bases studying the taxon difference. Potential gene comparison
(PHRED score Q ≤ 20) were trimmed. Adapter dimers, GC- between samples was performed using two-sided Fisher’s
content, overlapping sequences, PCR duplicates, ambiguous exact test and Storey’s false discovery rate (FDR) method
bases, homopolymers, and chimeras were checked to avoid (Parks and Beiko 2010). p values were adjusted using Bon-
errors in the assembly. The Illumina adapters at 5′ and 3′ ferroni correction. Comparative metagenomics such as tax-
ends of sequence reads were trimmed using the Cutadapt onomic-to-phenotypic mapping and multivariate analysis
tool version 1.8.1 (Martin 2011). The pre-processed reads including principal component analysis (PCA), clustering,
were assembled using the Ray Meta metagenome assembler and classification were performed using the METAGENas-
with default parameters (Boisvert et al. 2012). The contigs sist web server by uploading BIOM files. R-package was
with a length < 150 bp were discarded to reduce false posi- used for generating the p value significant plots (Arndt et al.
tives during downstream analyses like gene prediction and 2012).
annotation.
Isolation and morphological characterization
Prediction of open reading frames and functional of endophytes
annotation
After epiphytic check, the surface-sterilized plant samples
Assembled contigs were submitted to the MetaGeneAnno- of GPU and UM were used to isolate endophytes using the
tator (Noguchi et al. 2008) to predict open reading frames serial dilution (Hung and Annapurna 2004) and tissue bits
(ORFs). The read depth of all predicted ORFs was quanti- (Gayathri and Muralikrishnan 2013) methods. In serial dilu-
fied to reveal the taxonomic abundance. Functional annota- tion method, 100 µL aliquot from each dilution ( 10−3 and
tion was performed by aligning ORF sequences against the 10−4 dilutions for fungi; 10−5 to 10−7 dilutions for bacte-
NCBI-nr protein database using the DIAMOND tool ver- ria) of tissue extracts crushed in 0.9% NaCl solution were
sion 0.7.9.58 (Buchfink et al. 2015). The DIAMOND output uniformly spread on potato dextrose agar (PDA) and nutri-
files were analyzed using MEGAN5 software (Huson et al. ent agar (NA) media plates and incubated at 28 ± 2 °C to
2011). The gene ontology (cellular, biological and molecular obtain colony growth. In case of the tissue bits method, the
functions) was assigned for annotated ORFs based on SEED surface-sterilized leaf, stem, and root fragments were first
(Overbeek et al. 2014) classification methods available in imprinted on the control plates (epiphytic check) (Schulz
MEGAN5. The key genes involved particularly in hydrocar- et al. 1998) containing PDA and NA media for 5 s. Later,
bon degradation were searched from the annotation results. tissue fragments were removed and placed on fresh culture
All the functionally annotated ORFs were assigned the plates in identical positions to the control plates. Four bits
taxonomic hierarchy from phylum to species level (includ- were placed on medium in two replicates and incubated at
ing Bacteria, Archaea, Eukaryota, Viruses, and unclassified 28 ± 2 °C until endophyte growth appeared on the periphery
13
15 Page 4 of 17 3 Biotech (2020) 10:15
of tissue fragments. The culture plates showing no growth rDNA sequences were searched with those on the NCBI
on control plates were considered for further use. database using the BLAST tool (https://blast.ncbi.nlm.nih.
Following overnight incubation, morphologically distinct gov/Blast.cgi) for species identification.
bacterial colonies were transferred to fresh NA medium
plates using the quadrant streak plate method. The pure cul-
tures were stored in NA slants and glycerol stocks for further In vitro and in planta evaluation of endophytes
use. Similarly, pure cultures of distinct fungal endophytes against rice blast pathogen Magnaporthe grisea
were obtained using PDA medium plates and subsequently
preserved in PDA slants and glycerol stocks. The in vitro interaction of different endophyte strains against
MG03 (Magnaporthe strain, Shirke et al. 2016) was studied
16S rDNA and 18S rDNA analysis of bacterial using dual-culture assay. The MG03 strain was inoculated
and fungal endophytes on one side of Luria–Bertani (LB) agar medium and a loop-
ful of bacterial endophyte culture was streaked/inoculated
The genomic DNA of bacterial endophytes was extracted on the other end of culture plate (Singh and Varma 2015).
using AMpurE Bacterial Genomic DNA Mini Spin kit The fungal endophytes were inoculated as per Benfradj et al.
(Amnion Biosciences, India). The isolates were first inocu- (2016). A culture plate with MG03 alone was maintained as
lated on to a nutrient broth (NB) medium and incubated a control. For the positive control, culture plate with actively
overnight in an orbitary shaker at 150 rpm at 28 ± 2 °C. growing MG03 was inoculated with Bacillus velezensis
The bacterial cells were harvested from broth by centrifu- strains A6 and P42, a known strains antagonistic to M. gri-
gation for 2 min at 14,000×g and the pelleted cells were sea (Lim et al. 2017). Three replicates were maintained for
processed for DNA isolation as per manufacturer’s instruc- each endophyte as well as for positive and negative controls,
tions. RNA contamination was removed by treating with and the culture plates were incubated at 28 ± 2 °C for 5 days.
5 µL of RNAse A (20 mg/mL) ( HIMEDIA®, India) solution. The percentage inhibition of pathogen was calculated using
The DNA quality was checked on both 0.8% agarose gel the formula given by Vincent (1947):
and using Nanodrop®. The isolated bacterial genomic DNA
(C − T) × 100
was used as template to amplify the 16S rDNA region using I = ,
C
universal primers for bacterial endophytes’ identification
and characterization. Each 25-µL reaction mixture contained where, I = percentage inhibition; C = radial growth in control
2.5 µL (50 ng/µL) of template gDNA, 1.25 µL (10 pmol/µL) plate (mm); T = radial growth in treatment (mm) plate.
each of forward (5′ GAGTTTGATCCTGGCTCAG 3′) and The seeds of GPU and UM were bio-primed with endo-
reverse (5′ AGAAAGGAGGTGATCCAGCC 3′) primers, phytes to study their potential role in suppressing the blast
and 17.5 µL EmeraldAmp® GT PCR Master Mix (TaKaRa disease in finger millet in planta. Endophyte-free seedlings
Bio Inc., Japan). Final volume was made up to 25 µL using of GPU and UM were obtained as per the previously estab-
nuclease-free water. The thermal cycling program on the lished procedures (Pandey et al. 2016, 2018). The seeds
Eppendorf Mastercycler® pro (Eppendorf AG, Hamburg, were thoroughly washed in sterile d dH2O and incubated for
Germany) included an initial denaturation at 94 °C for 3 min 24 h in 0.2% benomyl and 70 ppm each of streptomycin sul-
followed by 35 cycles of 94 °C for 45 s, 58 °C for 1 min, phate and tetracycline solutions to eliminate the seed-borne
and 72 °C for 90 s and a final extension at 72 °C for 10 min. endophytes. The treated seeds were washed five times with
Genomic DNA from fungal isolates was extracted as per sterile ddH2O and an aliquot of seeds were homogenized
Moller Bahnweg et al. (1992). The quality of extracted DNA using pestle and mortar. The homogenate was plated on PDA
was assessed on a 0.8% agarose gel and quantified using a and NA agar media plates and incubated at 28 ± 2 °C for
Nanodrop®. The 18S rDNA region in the fungal genome 7 days. When no microbial growth was observed on the incu-
was amplified using ITS2 (5′ GCTGCGTTCTTCATCGAT bated plates, the remaining treated seeds were considered
GC 3′) and ITS4 (5′ TCCT CCG CTT ATT
GAT ATG C 3′) uni- endophyte-free and used for bio-priming. For bio-priming,
versal primers (White et al. 1990) for fungal endophytes’ all the bacterial and fungal endophytes were individually
characterization. The composition of the PCR mixture was inoculated into sterile NB and potato dextrose broth (PDB)
same as mentioned earlier and the amplification program media and incubated for 24 h and 7 days, respectively. For
included an initial denaturation at 94 °C for 4 min followed each endophyte to be treated, 50 surface-sterilized treated
by 35 cycles of 94 °C for 30 s, 55 °C for 30 s, and 72 °C for seeds each of GPU and UM were soaked in these pre-inoc-
1 min, and a final extension at 72 °C for 5 min. The PCR ulated NB and PDB media for 24 h. An equal number of
amplicons of 16S rDNA (for bacterial endophytes) and 18S chemical-treated seeds of GPU and UM soaked in sterile
rDNA (for fungal endophytes) were sequenced by Sanger broth media was kept as a control (C0). The soaked seeds
dideoxy sequencing technology. The 16S rDNA and 18S were air-dried under aseptic conditions and later sown in
13
3 Biotech (2020) 10:15 Page 5 of 17 15
pots (15 cm diameter) filled with sterile growth medium and UM, respectively) using Ray Meta de novo assembler (ver-
maintained in a greenhouse. Each treatment was replicated sion 2.3.1) and only contigs of size > 150 bp were retained
twice and the temperature was maintained at 28 ± 2 °C dur- for downstream analysis (Table 1). Totally, 5,96,312 and
ing day and 25 ± 2 °C at night. The plants were transferred to 3,74,417 contigs having length > 150 bp among which only
the inoculation chamber 1 day prior to pathogen inoculation 5,62,769 (94.37%) and 3,50,015 (93.48%) contigs exclu-
for acclimatization (28 ± 5 °C day and 25 ± 5 °C night tem- sively belonging to bacteria and fungi were retained for
perature, and relative humidity > 90%). The seeds obtained downstream analyses in GPU and UM, respectively. The
from C0 were sown as a control. contigs not classified within bacterial and fungal taxa were
excluded from downstream analysis. The rarefaction curves
Pathogenicity test and disease scoring (Fig. S1) indicated that GPU had high species richness as
compared to UM.
Sporulation was induced in MG03 culture plates and the Taxonomic profiling of the metagenomes of GPU and UM
spore suspension was sprayed onto 20-day-old GPU and revealed that bacterial classes, Alphaproteobacter, showed
UM seedlings maintained in growth chamber (Mahesh eight genera of potential biocontrol agents that were seen
et al. 2016). The treated seedlings were further covered in both GPU and UM; however, GPU showed more species
with polythene bags for 1 week to maintain high humidity richness (Fig. 1). Bacteria were the most abundant domain
(~ 94%) conducive for disease development. The polythene among the endophytic consortia colonizing both the geno-
bags were removed on sixth day after inoculation (DAI) and types followed by the domains of eukaryotes and archaea.
seedlings were scored for disease severity as per the IRRI Among the bacterial domain, the phylum Proteobacteria
SES scale (Table S1) at 30 DAI. The percent disease sever- was the most abundant in both GPU and UM followed by
ity (PDI) was calculated as per Chiang et al. (Chiang et al. Actinobacteria and Bacteriodetes. The remaining phyla viz.,
2017): Cyanobacteria (blue green algae), Ascomycota and Basidi-
omycota (fungi) were less prominent in the metagenomes
𝛴 Sum of all numerical scores × 100
PDI = of GPU and UM. The domain Archaea in both the samples
Total number of leaflets observed × maximum score
was represented by two phyla viz, Euryarchaeota and Tau-
After disease development, the infected leaves were col- marcheota (Fig. 2).
lected, surface sterilized, and small infected leaf bits were Out of the total 1029 species of microbiota collectively
placed on PDA medium to re-isolate the test pathogen. The identified in GPU and UM, 287 species were identified as
culture plates were incubated at 28 ± 2 °C to promote the common, while 385 and 357 species were specific to GPU
growth of test pathogen. Later, the pathogen was purified and UM, respectively. The plant growth-promoting bacteria
and its identity with original culture of MG03 was confirmed like Sorangium cellulosum, Lechevalieria aerocolonigenes,
based on morphological characters. Thus, Koch postulates Stenotrophomonas maltophilia, and Streptomyces spp. were
were established. widely distributed in both GPU and UM, but their relative
abundance was much higher in GPU as compared to UM
Statistical analysis (Fig. 3). The genera Streptomyces in both UM and GPU was
dominated by the species, Streptomyces galbus, which is a
The data were analyzed using the appropriate statistical strong producer of antifungal compounds (Paul and Baner-
tools and percent values were arcsine transformed (Gomez jee 1983) and this organism was more prevalent in GPU
et al. 1984) before subjecting the data to analysis of vari- than UM. The other species in this group with documented
ance (ANOVA). The critical difference (CD) values were antifungal activity and identified exclusively in GPU were S.
calculated at 5% probability level. griseus (Anitha and Rebeeth 2009; Rabeeth et al. 2011), S.
halstedii (Frandberg et al. 2000), S. anulatus (Mander et al.
2016), and S. cluvuligerus (Doran et al. 1990). Bacilli were
Results
13
15 Page 6 of 17 3 Biotech (2020) 10:15
Fig. 1 Phylogenetic diversity of metagenomes identified in the two nome datasets of GPU and UM. The genus highlighted in colors is
finger millet genotypes. The circular phylogenetic trees (left—GPU reported as biocontrol agents against plant pathogens. The classes are
and right—UM) representing various taxa identified from metage- represented in the outer ring
another important group which are important in the context UM = 22), phosphorous metabolism (GPU = 22; UM = 12),
of this study. The species, Bacillus cereus, Paenibacilllus metabolism of aromatic compounds (GPU = 63, UM = 48),
mucilaginosus, and Lactobacillus sp. were identified exclu- and secondary metabolism (GPU = 11; UM = 12) were
sively in GPU and these species were reported to produce quantitatively higher in GPU as compared to UM (Fig. 4).
siderophores which promote plant growth and also possess The metabolic functions of genes present in the endo-
biocontrol activity against phytopathogenic fungi (Tokpah phytic microbiota were identified through KEGG analysis
et al. 2017; Amruta et al. 2018). and revealed seven major metabolic and enzyme pathways
at the metagenome level (Fig. S3). The contigs pertaining
Functional annotation of assembled metagenome to metabolism (carbohydrate metabolism, energy metabo-
contigs lism, lipid metabolism, nucleotide metabolism, amino acid
metabolism, and biosynthesis of secondary metabolites),
The SEED classification of annotated contigs affiliated to genetic information processing, and environmental infor-
endophytes helped to identify 23 major Level-1 sub-sys- mation processing were the major categories detected in
tems in the metagenome samples of GPU and UM (Fig. 4). both the cultivars. However, contigs belonging to many
Carbohydrates, protein metabolism, DNA metabolism, metabolism pathways hold a significant importance in
and respiration were among the major Level-1 sub-sys- key processes involved in disease resistance mechanism.
tems identified in both the samples. The other important Among the different metabolic pathways, reads belonging
Level-1 sub-systems relevant to the present study were to cysteine synthase (GPU = 11; UM = 7) and glutathione
stress response, clustering-based sub-systems, sulsur synthase (GPU = 4 and UM = 1) bearing ECs ‘2.5.1.47′
metabolism, phosphorous metabolism, metabolism of and ‘6.3.2.3,’ respectively, were more in GPU as com-
aromatic compounds, and secondary metabolism. Com- pared to UM. Glutathione synthase (Dubreuil-Maurizi and
parative analysis of these functional sub-systems (Fig. S2) Poinssot 2012) and cysteine synthase (Álvarez et al. 2012)
in GPU and UM showed that the reads encoding stress enzymes are reported to play an important role in plant
response (GPU = 76; UM = 63), clustering-based sub-sys- defense mechanism.
tems (GPU = 81; UM = 70), sulfur metabolism (GPU = 29;
13
3 Biotech (2020) 10:15 Page 7 of 17 15
Morphological characterization of culturable reaction). In GPU, four and in UM, three endophytes were
endophytes found to be aerobic (positive reaction).
Based on the morphological characteristics like myce-
Using serial dilution and tissue bits isolation methods, a lium type, spore characteristics and colony morphology, five
total of 75 bacterial and 5 fungal endophytes were isolated fungal endophytes were found belonging to genera viz. Lasi-
from the leaf, stem, and root tissues of GPU and UM. Out odiplodia, Talaromyces, Phoma, and Pencillium (Table S3).
of 75 bacterial isolates, 52 and 23 strains were obtained
from GPU and UM, respectively (Table 2). Maximum iso- Molecular characterization of culturable
lates were obtained from the leaf (22) tissues of GPU and endophytes
the lowest from stem (6) tissues of UM (Fig. 5).
A rich microbial diversity was observed among the bac- The amplified products of 16S rDNA (bacteria) and 18S
terial isolates with respect to their colony morphology, rDNA (fungi) were sequenced and based on maximum query
color, shape, and size (Table S2). The gram staining pro- coverage, the species were identified and accession numbers
cedure differentiated the 75 bacteria into 45 Gram negative were obtained from NCBI (Table S4). Based on molecular
(G−ve) and 30 Gram positive ( G+ve) isolates. Among them, phylogeny, the 75 bacterial endophytes were assigned to 12
19 and 11 G +ve; 33 and 12 G −ve isolates were obtained genera among which, six, four, and two genera were associ-
from GPU and UM, respectively. The oxidase test indi- ated to Gammaproteobacteria, Bacilli, and Actinobacteria,
cated that the majority of isolates were anaerobic (negative respectively (Table 2).
13
15 Page 8 of 17 3 Biotech (2020) 10:15
Fig. 3 Comparative distribution of top 33 species identified in the the two samples. The two icons on each radius of the circular chart
metagenome of GPU and UM. The relative abundance is mapped correspond to the respective genotypes indicated in the key
based on the number of reads pertaining to each species identified in
In GPU, the class Gammaproteobacteria (63.46%) was In vitro study of interaction between endophytes
abundantly distributed followed by Bacilli (32.69%). The gen- and M. grisea
era Cronobacter, Enterobacter, Escherichia, Pantoea, Aceni-
tobacter, and Pseudomonas were identified in Gammaproteo- The dual plate assay (Fig. 6) revealed that, among 56 endo-
bacteria, whereas Bacillus, Paenibacillus, and Lactobacillus phytes (52 bacteria and 4 fungi) isolated from GPU, 12
were found among the Bacilli group. In case of UM, Bacilli isolates inhibited the growth of M. grisea (strain MG03) by
(52.17%) was the predominant taxa followed by Gammapro- more than 50% (Table 3). The positive control, B. velezen-
teobacteria (47.83%). Three genera viz. Enterobacter, Pantoea, sis strains P42 and A6 inhibited the MG03 by 78.83% and
and Pseudomonas were identified in Gammaproteobacteria, 80%, respectively. The strains, B. cereus GPUL8 (71.93%),
while Bacillus and Lysinibacillus were detected in Bacilli Bacillus licheniformis GPUL23 and B. cereus GPUR10
group. Among the different culturable endophytes isolated in (66.67% each) inflicted maximum percentage inhibition
the present study, we found that the Enterobacter spp. and (PI) on MG03. The least inhibitory effect on the patho-
Bacillus spp. were the dominant bacterial taxa colonizing GPU gen was observed with the strain Enterobacter aerogenes
and UM (Table 2). GPUL5 (20.83%). Similarly, among the 24 endophytes
Based on the maximum query coverage value and percent- (23 bacteria and one fungi) isolated from UM, seven iso-
age identity, the fungal isolates were identified as Penicillium lates inhibited the growth of MG03 by more than 50%.
polonicum, Lasiodiplodia theobromae, Talaromyces trachy- The strain, Bacillus mojavensis UMR9 imposed maxi-
spermus, and Phoma sp. Among the five fungal genera identi- mum inhibitory effect (65%) on MG03 followed by the
fied, only Phoma was found common to both GPU and UM strains Bacillus sp. UML10, Bacillus sp. UMS5, and
(Table S4). Pseudomonas aeruginosa UML11 (61.67% each). The
strains Enterobacter mori UML4 and Enterobacter sp.
13
3 Biotech (2020) 10:15 Page 9 of 17 15
Fig. 4 SEED classification-based functional annotation of contigs groups, and pigments are represented by asterisk. The phages like
of GPU and UM endophytes. Relative abundance of contigs of GPU prophages, transposable elements, and plasmids are represented by
and UM endophytes affiliated to 23 functional categories identified double asterisk
through SEED classification. The cofactors like vitamins, prosthetic
The values in the parentheses indicate the percentage abundance of the genus in the genotype
UMR2 showed minimum growth inhibition (21.67% each) (44.44%). In general, the genera Bacillus and Paenibacillus
of MG03. Among the five fungal endophytes evaluated, were highly antagonistic and inhibited the growth of MG03
L. theobromae had the highest inhibition effect on MG03 under in vitro condition (Table 3).
13
15 Page 10 of 17 3 Biotech (2020) 10:15
Fig. 5 Diversity and distribution of culturable endophytes isolated from tissues. c Summary of richness of bacterial genera and number
from tissues of GPU and UM genotypes. a Relative richness of dif- of bacterial species isolated from leaf, stem, and root tissues of GPU
ferent bacterial genera isolated from the leaf, stem, and root tissues. and UM
b Number of culturable fungal and bacterial endophytes isolated
13
3 Biotech (2020) 10:15 Page 11 of 17 15
Table 3 In vitro and in planta evaluation of bacterial and fungal endophytes isolated from GPU and UM against M. grisea pathogen
Sl no. Strains Bacterium PI (%) PDI (%)
UM GPU
Bacteria
1 UML2 Lysinibacillus sp. 48.33 (44.04) 41.67 (40.20) 36.11 (36.92)
2 UML4 E. mori 21.67 (27.72) 69.44 (56.65) 33.33 (35.19)
3 UML5 Enterobacter sp. 33.33 (35.26) 36.11 (36.92) 25.00 (29.97)
4 UML6 E. cloacae subsp. dissolvens 26.67 (31.09) 22.22 (27.95) 47.22 (43.41)
5 UML10 Bacillus sp. 61.67 (51.75) 52.78 (46.59) 27.78 (31.81)
6 UML11 P. aeruginosa 61.67 (51.75) 61.11 (51.42) 27.78 (31.69)
7 UML12 Bacillus sp. 48.33 (44.04) 30.56 (33.54) 25.00 (29.97)
8 UMS1 B. tequilensis 56.67 (48.83) 63.89 (53.08) 22.22 (28.13)
9 UMS2 B. mojavensis 48.33 (44.04) 50.00 (45.00) 36.11 (36.92)
10 UMS3 P. agglomerans 33.33 (35.26) 33.33 (35.26) 25.00 (29.97)
11 UMS5 Bacillus sp. 61.67 (51.75) 58.33 (49.80) 33.33 (35.19)
12 UMS6 Pseudomonas sp. 28.33 (32.15) 75.00 (60.03) 75.00 (60.03)
13 UMS7 Bacillus sp. 41.67 (40.13) 52.78 (46.59) 63.89 (53.08)
14 UMR2 Enterobacter sp. 21.67 (27.59) 41.67 (40.20) 8.33 (16.55)
15 UMR3 E. cloacae 36.67 (37.25) 16.67 (24.09) 44.44 (41.79)
16 UMR4 Bacillus sp. 60.00 (50.77) 61.11 (51.42) 8.33 (16.55)
17 UMR5 E. cloacae 31.67 (34.18) 33.33 (35.19) 13.89 (20.88)
18 UMR6 P. agglomerans 31.67 (34.24) 44.44 (41.79) 55.56 (48.21)
19 UMR9 B. mojavensis 65.00 (53.78) 36.11 (36.92) 47.22 (43.41)
20 UMR10 Bacillus sp. 55.00 (47.87) 52.78 (46.59) 55.56 (48.21)
21 UMR12 Bacillus sp. 42.11 (40.43) 44.40 (41.81) 69.44 (56.46)
22 UMR13 Lysinibacillus sp. 40.35 (39.42) 66.67 (54.81) 58.33 (49.80)
23 UMR15 Enterobacter sp. 43.86(41.39) 63.89 (53.08) 61.11 (51.42)
24 GPUL1 E. cloacae 28.50 (32.26) 38.89 (38.58) 44.44 (41.81)
25 GPUL2 E. dissolvens 28.50 (32.26) 16.67 (23.80) 13.89 (21.78)
26 GPUL3 A. baumannii 24.40 (29.60) 47.22 (43.41) 22.22 (28.13)
27 GPUL4 Enterobacter sp. 31.25 (33.88) 30.56 (33.54) 16.67 (23.80)
28 GPUL5 E. cloacae 27.85 (31.82) 36.11 (36.92) 36.11 (36.92)
29 GPUL6 E. cloacae 37.50 (37.76) 33.33 (35.19) 44.44 (41.81)
30 GPUL7 Enterobacter sp. 31.25 (33.99) 33.33 (35.26) 25.00(29.68)
31 GPUL8 B. cereus 71.93 (58.01) 41.67 (40.20) 47.22 (43.41)
32 GPUL9 E. cloacae subsp. dissolvens 21.88 (27.83) 58.33 (49.80) 25.00 (29.97)
33 GPUL11 B. cereus 28.33 (32.15) 41.67 (40.20) 33.33 (35.19)
34 GPUL12 C. sakazakii 51.11 (45.64) 52.78 (46.66) 27.78 (31.81)
35 GPUL13 A. baumannii 25.56 (30.32) 30.56 (33.54) 16.67(24.09)
36 GPUL14 A. baumannii 28.89 (32.50) 44.44 (41.81) 19.44 (26.11)
37 GPUL15 Bacillus sp. 55.56 (48.19) 33.33 (35.19) 27.78 (31.81)
38 GPUL16 A. baumannii 25.00 (29.97) 52.78 (46.59) 16.67 (23.80)
39 GPUL17 B. cereus 51.67 (45.96) 30.56 (33.54) 13.89 (21.78)
40 GPUL18 A. baumannii 27.85 (31.82) 44.44 (41.79 36.11 (36.92))
41 GPUL19 E. cloacae 51.67 (45.96) 41.67 (40.13) 38.89 (38.54)
42 GPUL20 M. lylae 35.00 (36.27) 55.56 (48.21) 44.44 (41.81)
43 GPUL21 B. cereus 38.33 (38.25) 44.44 (41.79) 19.44 (26.11)
44 GPUL23 B. licheniformis 66.67 (54.74) 44.44 (41.79) 33.33 (35.19)
45 GPUL24 P. stutzeri 22.81 (28.46) 63.89 (53.08) 50.00 (45.00)
46 GPUS1 E. cloacae 20.00 (26.25) 38.89 (38.54) 19.44 (26.11)
47 GPUS2 E. cloacae subsp. cloacae 45.56 (42.45) 38.89 (38.54) 27.78 (31.81)
13
15 Page 12 of 17 3 Biotech (2020) 10:15
Table 3 (continued)
Sl no. Strains Bacterium PI (%) PDI (%)
UM GPU
13
3 Biotech (2020) 10:15 Page 13 of 17 15
Pathogenicity test and blast disease scoring are contrasting in their reaction towards blast disease, caused
for blast in finger millet by Ascomycetes fungi Magnaporthe grisea. The WMG
sequencing followed by metagenome assembly of GPU and
Artificial inoculation of MG03 produced characteristic and UM resulted into 7,76,412 and 4,95,014 contigs in GPU
typical spindle-shaped lesions on the leaves of GPU and UM and UM, respectively. The metagenome of both GPU and
seedlings at 6–7 DAI. The lesions initiated as tiny spots with UM cultivars was dominated by the sequences belonging
yellowish margin and grayish-green center became whitish to the domain bacteria in which, the phyla Proteobacteria
gray and disintegrated later. The percentage disease incidence and Actinobacteria were abundant. The previous studies
(PDI) was recorded on the infected GPU and UM seedlings at on the characterization of endophyte communities in rice
10 DAI and the disease severity was compared in both the fin- (Sessitsch et al. 2012; Sengupta et al. 2017), which reported
ger millet genotypes. Among the 52 bacterial isolates of GPU, high abundance of Proteobacteria, Firmicutes, and Actino-
seed treatment with the strains viz. Enterobacter dissolvens bacteria. Proteobacteria were abundant in UM, whereas the
GPUL2 (Disease severity 13.89% and 16.67%), Enterobacter Actinobacteria were highly prominent in GPU. Actinomy-
sp. GPUL4 (16.67% and 30.56%), B. cereus GPUL17 (13.89% cetes are soil microbes that colonize the host plant tissues as
and 30.56%), Acinetobacter baumannii GPUL13 (16.67% and endophytes (Madhurama et al. 2014; Jogaiah et al. 2016) and
30.56%), Bacillus sp. GPUR12 (16.67% and 38.89%), and produce antibiotics, promote plant growth, and also protect
Paenibacillus sp. GPUS13 (25% and 44.44%) showed less against phytopathogenic fungi (Smith 1957; Fiedler et al.
blast disease severity in GPU and UM cultivars (Table 3). The 2008; Ma et al. 2009; Vurukonda et al. 2018). Actinomy-
species belonging to E. cloacae, A. baumannii, Lactobacillus cetes are the largest microbial source for different bioactive
helveticus, E. dissolvens, and Enterobacter sp. were effective secondary metabolites, and among the genera of Actinomy-
in suppressing the pathogen in planta in both GPU and UM; cetes group, the Streptomyces spp. alone contributes to more
however, they exhibited a poor inhibition of the pathogen in than 70% of the total antibiotic production (Ningthoujam
in vitro assay. Among the different strains isolated from GPU, et al. 2009). This suggests that the relative abundance of
the species belonging to B. cereus, A. baumannii, L. helveti- Streptomyces spp. in GPU as compared to UM may underpin
cus, and Paenibacillus sp. were found unique to GPU. the host-plant resistance in GPU against M. grisea patho-
Out of 23 bacterial isolates of UM, seed treatment with gen. Similar antifungal activity of Streptomyces spp. against
the strain Bacillus sp. UML12 resulted in lowest disease M. grisea was also reported in earlier studies (Hwang et al.
severity (25% and 30.56% in GPU and UM, respectively) 1996; Lee et al. 2002; Yoshii et al. 2012; Van Minh et al.
in planta assay. However, seed treatment with the strains 2015).
belonging to E. cloacae UMR5 (13.89% and 33.33%), Enter- The species, S. cellulosum, identified in the metagen-
obacter sp. UML5 (25% and 36.11%), Bacillus sp. UML12 ome samples of both GPU and UM, but more abundant in
(25% and 30.56%) and Pantoea agglomerans UMS3 (25% GPU, was known to produce many antifungal compounds
and 33.33%) resulted in low PDI in in planta in both GPU like ambruticins, jerangolids (Gerth et al. 1995a; Knuat and
and UM, respectively (Table 3). Among the five fungal Reichenbach 2000), soraphens (Gerth et al. 1994), ratjadon
endophytes identified, L. theobromae YGL3 and T. trach- (Gerth et al. 1995b), and epothilones (Gerth et al. 1996),
yspermus PGR6 showed low PDI in both GPU and UM. which inhibit several phytopathogenic fungi. Another impor-
However, their antagonistic activity against the pathogen in tant endophyte S. maltophilia, reported to hold a strong
in vitro was found to be less effective. In a control (C0), antagonistic activity against M. grisea (Jha and Subramanian
the PDI was recorded as 52.56% and 55.56% in GPU and 2013; Etesami and Alikhani 2016), was found in GPU. The
UM, respectively. An overview of the results from in vitro production of alkaline serine proteolytic enzyme and host
and in planta studies have revealed that the GPU strains, B. systemic acquired resistance induced by S. maltophilia might
cereus GPUL8, Bacillus sp. GPUL15, B. cereus GPUL17, play a key role in its biocontrol activity against different phy-
Paenibacillus sp. GPUS13, Bacillus sp. GPUR12, B. cereus topathogens (Selim et al. 2016). The L. aerocolonigenes was
GPUR10, Bacillus sp. GPUR11, Bacillus sp. GPUR12, and identified in GPU, and was reported to produce thiobutacin
B. cereus GPUR16 were effectively suppressed the pathogen (Lee et al. 2004; Yeop et al. 2008) which is antagonistic
both in in vitro and in planta in GPU and UM cultivars. against oomycetes and some ascomycetes fungi. The high
abundance of these species in GPU might have a potential
role in the resistant reaction of this cultivar against the blast
Discussion pathogen. The functional annotation of genes of endophytes
identified in GPU and UM through SEED classification
In the present study, metagenomics approach enabled to gain revealed that the GPU associated endophytes harbored with
a deeper understanding of the diversity of endophytic con- genes for stress response, sulphur metabolism, phosphorus
sortia in the GPU and UM finger millet genotypes, which metabolism, secondary metabolites, and metabolism of
13
15 Page 14 of 17 3 Biotech (2020) 10:15
aromatic compounds. Earlier studies reported that phos- and described its antagonistic potential against several phy-
phorous metabolism (Dordas 2008), secondary metabolites topathogenic fungi such as Cryphonectria parasitica, Glom-
and metabolism of aromatic compounds (Aranega-Bou et al. erella glycines, Phytophthora capsici, Fusarium gramine-
2014; Llorens et al. 2017), and sulphur metabolism (Romero arum, Botrytis cinerea, and Rhizoctonia solani. Interestingly,
et al. 2014) play an important role in inducing host-plant metagenome data and isolation studies showed that the A.
resistance against several phytopathogens. Our in silico baumannii was present only in GPU cultivar. However, the
analysis suggested that the microbial consortia of GPU may in vitro and in planta studies showed variation in microbial
help to overcome the pathogen infection and impart disease composition with respect to genera and species in GPU and
resistance reaction. UM cultivars. The endophyte strains which effectively sup-
Similarly, the KEGG pathway analysis revealed that the pressed the pathogen in vitro (Bacillus sp. UML10, Bacillus
GPU accommodated more genes encoding glutathione syn- sp. UMS5, Bacillus sp. UMR4, Bacillus sp. GPUS16, and
thase pathway as compared to UM. The glutathione is an Cronobacter sakazakii GPUL12) showed poor suppression
important signaling compound involved in plant disease of disease in planta and vice versa (E. cloacae GPUL5, A.
resistance (Ghanta and Chattopadhyay 2011; Gullner and baumannii GPUL13, A. baumannii GPUL14, Bacillus sp.
Zechmann 2017) and is known to involve in antioxidative GPUS3, E. ludwigii GPUS9, and E. aerogenes GPUS11,
defense mechanism in plants. We also found that the genes etc.). In in vitro, the endophytes were directly exposed to
involved in cysteine synthase pathway were abundant in pathogen on synthetic media resulting in direct interac-
GPU. For most sulphur-containing metabolites, a direct anti- tion between the endophyte and the pathogen. But under in
fungal mode of action has been reported where it activates planta, each endophyte was evaluated under natural condi-
systemic acquired resistance thereby initiating hypersensi- tions in its original host where it becomes exposed to mul-
tive reaction (HR). Cysteine is the main precursor for all titude of factors and this compound effect could possibly
sulfur -containing compounds and is directly linked to stress be responsible for the variance in results observed under
response through its function related to systemic acquired the two experimental conditions. We also observed that the
resistance (Luckner 2013). Cysteine displays a regulatory same endophyte strain produced contrasting results in GPU
function in pathogen defense and it was reported that a spe- and UM against the blast pathogen in planta. Particularly,
cific cytosolic cysteine content is essential for the initiation in case of the strains belonging to Bacillus spp. (B. cereus,
of the plant immune response to pathogens and a link to the B. mojavensis, B. tequilensis, B. licheniformis, and Bacil-
hypersensitive response (Álvarez et al. 2012). lus sp.), we observed that the majority of the Bacillus sp.
To support the results of our metagenomics study, cultur- strains had high suppression of the blast disease in planta
able endophytes were isolated from GPU and UM cultivars. in GPU, while the same strains showed less blast incidence
52 bacterial and four fungal endophytes from GPU; 23 bac- in UM. There could be a specific host-endophyte relation-
terial endophytes and one fungal endophyte from UM were ship between the Bacillus species and GPU, which could
isolated. Molecular characterization of 75 bacterial (16S be derived through their evolutionary relationship and con-
rDNA sequencing) and 5 fungal (18S rDNA sequencing) tributing GPU to take advantage of the colonization of the
endophytes revealed that Enterobacter spp. and Bacillus spp. Bacillus species and use their antagonistic potential for the
were the dominant bacterial phylotypes colonizing GPU and suppression of the blast pathogen during infection.
UM. Based on metagenome sequencing and wet lab experi-
ments, we observed diversity with respect to the distribution
of endophyte communities in these two host habitats, i.e. Conclusion
GPU and UM. Enterobacter spp. and Bacillus spp. were
identified as more abundant in GPU as compared to UM. The information and knowledge of microbial community
This indicated that the nature of cultivar may contribute to dynamics found in the two genotypes of finger millet, GPU-
the structure of the endophytic communities associated with 28 and Udurumallige, was limited. This metagenomic study
the finger millet. had highlighted biocontrol agents, to provide a new genesis
Among all the culturable endophytes evaluated in both and inception of microbial diversity to protect the finger mil-
in vitro and in planta, the strains belonging to the GPU let against the blast disease caused by M. grisea. The WMG
native endophytes like B. cereus and Paenibacillus sp. were sequencing of GPU and UM revealed high structural diver-
found highly effective in suppressing the growth of blast sity and distribution of endophytic microbial communities
pathogen. The fungistatic effect of B. cereus and Paenibacil- in these two genotypes. This information was supported by
lus sp. on M. grisea identified in the present study was also the functional annotation of gene clusters of endophytic con-
reported in previous studies (Ladeuze et al. 2011; Xu et al. sortia which unveiled genes encoding stress response, sec-
2014; Huang et al. 2018). Liu et al. (2007) reported A. bau- ondary metabolites and metabolism of aromatic compounds,
mannii as a potential endophyte in Cinnamomum camphora phosphorous metabolism, and sulphur metabolism to be in
13
3 Biotech (2020) 10:15 Page 15 of 17 15
greater fraction in GPU as compared to UM. This com- Bhatt D, Negi M, Sharma P et al (2011) Responses to drought induced
parative study provides key insights into the structural and oxidative stress in five finger millet varieties differing in their
geographical distribution. Physiol Mol Biol Plants 17:347–353.
functional differences of endophytic communities associ- https://doi.org/10.1007/s12298-011-0084-4
ated with blast resistant and susceptible cultivars might have Boisvert S, Raymond F, Godzaridis É et al (2012) Ray Meta: scal-
putative role in contrast disease reactions to blast pathogen. able de novo metagenome assembly and profiling. Genome Biol.
The unique endophytes of GPU like B. cereus and Paeni- https://doi.org/10.1186/gb-2012-13-12-r122
Bolwerk A, Lagopodi AL, Lugtenberg BJJ, Bloemberg GV (2005)
bacillus sp. could be used as potential bio-control agents in Visualization of interactions between a pathogenic and a ben-
reducing the incidence of blast disease in finger millet crop. eficial Fusarium strain during biocontrol of tomato foot and
The results reported in this study could enhance the current root rot. Mol Plant-Microbe Interact 18:710–721. https://doi.
understanding of the application of endophytes in control- org/10.1094/MPMI-18-0710
Braun G, Hirsch U (1992) Communities of parasitic microfungi.
ling diseases in agriculturally important crops. In: Winterhoff W (ed) Fungi in vegetation science. Springer,
Dordrecht, pp 225–250
Author contributions HBM and MKP conceived and conceptualized Buchfink B, Xie C, Huson DH (2015) Fast and sensitive protein
the study. HBM and RUD collected the plant samples and isolated alignment using DIAMOND. Nat Methods 12:59–60. https://
genomic DNA. RUD and KT performed the in vitro experiments. RCP doi.org/10.1038/nmeth.3176
analyzed the metagenome data. HBM guided metagenome data analy- Budnik LT, Baur X (2009) The assessment of environmental and
sis. MEP assisted in maintaining cultures of endophytes and sequenc- occupational exposure to hazardous substances by biomonitor-
ing. HBM, MKP and BK wrote the manuscript. HBM, KSN and GVB ing. Dtsch Arztebl Int 106:91–97. https://doi.org/10.3238/arzte
edited the manuscript. KSN, KB and BP contributed in preparing bl.2009.0091
the figures. All authors read and approved the final manuscript for Chaturvedi H, Singh V (2016) Potential of bacterial endophytes as
publication. plant growth promoting factors. J Plant Pathol Microbiol. https
://doi.org/10.4172/2157-7471.1000376
Availability of sequenced data The raw fastq files for the metagenome Chiang KS, Liu HI, Bock CH (2017) A discussion on disease sever-
data were uploaded in NCBI SRA database with accession number ity index values. Part I: warning on inherent errors and sugges-
SRP105033. tions to maximise accuracy. Ann Appl Biol 171:139–154. https
://doi.org/10.1111/aab.12362
Doran JL, Leskiw BK, Aippersbach S, Jensen SE (1990) Isolation
Compliance with ethical standards and characterization of a beta-lactamase-inhibitory protein from
Streptomyces clavuligerus and cloning and analysis of the cor-
Conflict of interest Author Rajadurai, R. C. was employed by com- responding gene. J Bacteriol 172:4909–4918
pany AgriGenome Laboratory. All other authors declare no competing Dordas C (2008) Role of nutrients in controlling plant diseases in
interests. sustainable agriculture. A review. Agron Sustain Dev 28:33–46.
https://doi.org/10.1051/agro:2007051
Dubreuil-Maurizi C, Poinssot B (2012) Role of glutathione in plant
signaling under biotic stress. Plant Signal, Behav
References Dutta D, Puzari KC, Gogoi R, Dutta P (2014) Endophytes: exploi-
tation as a tool in plant protection. Braz Arch Biol Technol
57:621–629
Álvarez C, Bermúdez MÁ, Romero LC et al (2012) Cysteine homeo- Esele JP (2002) Diseases of finger millet: a global overview. In: Mur-
stasis plays an essential role in plant immunity. New Phytol phy LJ, Montorsi F (eds) Sorghum and finger millet diseases. Iowa
193:165–177. https://doi.org/10.1111/j.1469-8137.2011.03889.x State Press, Iowa, pp 19–26
Amruta N, Prasanna Kumar MK, Puneeth ME et al (2018) Exploring Etesami H, Alikhani HA (2016) Suppression of the fungal patho-
the potentiality of novel rhizospheric bacterial strains against the gen Magnaporthe grisea by Stenotrophomonas maltophilia, a
rice blast fungus Magnaporthe oryzae. Plant Pathol J 34:126–138. seed-borne rice (Oryza sativa L.) endophytic bacterium. Arch
https://doi.org/10.5423/PPJ.OA.11.2017.0242 Agron Soil Sci 62:1271–1284. https://doi.org/10.1080/03650
Anitha A, Rebeeth M (2009) In vitro antifungal activity of Streptomy- 340.2016.1139087
ces griseus against phytopathogenic fungi of tomato field. Acad Fiedler H, Bruntner C, Riedlinger J et al (2008) ORIGINAL ARTICLE
J Plant Sci 2:119–123 Proximicin A, B and C, Novel aminofuran antibiotic and antican-
Aranega-Bou P, de la OLeyva M, Finiti I et al (2014) Priming of plant cer compounds isolated from marine strains of the Actinomycete
resistance by natural compounds. Hexanoic acid as a model. Front Verrucosispora. J Antibiot 61:158–163
Plant Sci 5:1–12. https://doi.org/10.3389/fpls.2014.00488 Frandberg E, Petersson C, Lundgren LN, Schnurer J (2000) Strep-
Arndt D, Xia J, Liu Y et al (2012) METAGENassist: a comprehen- tomyces halstedii K122 produces the antifungal compounds
sive web server for comparative metagenomics. Nucl Acids Res bafilomycin B1 and C1. Can J Microbiol 46:753–758. https://doi.
40:88–95. https://doi.org/10.1093/nar/gks497 org/10.1139/w00-050
Arnold AE, Mejia LC, Kyllo D et al (2003) Fungal endophytes Gayathri P, Muralikrishnan V (2013) Isolation and characterization of
limit pathogen damage in a tropical tree. Proc Natl Acad Sci endophytic actinomycetes from mangrove plant for antimicrobial
100:15649–15654. https://doi.org/10.1073/pnas.2533483100 activity. Int J Curr Microbiol Appl Sci 2:78–89
Benfradj N, Tounsi S, Boughalleb-MHamdi N (2016) In-vitro evalua- Gerth K, Bedorf N, Irschik H et al (1994) The soraphens: a family of
tion of Antagonists and Fungicides in Controlling Citrus Gummo- novel antifungal compounds from Sorangium cellulosum (Myxo-
sis Caused by Phytophthora, Phytopythium and Pythium species bacteria). I. Soraphen A1 alpha: fermentation, isolation, biological
in Tunisia. Br Microbiol Res J 16:1–14. https://doi.org/10.9734/ properties. J Antibiot (Tokyo) 47:23–31. https://doi.org/10.7164/
BMRJ/2016/27247 antibiotics.47.23
13
15 Page 16 of 17 3 Biotech (2020) 10:15
Gerth K, Schummer D, Höfle G, Irschik H, Reichenbach H (1995a) Lee CH, Kim BJ, Choi GJ et al (2002) Streptomyces with antifungal
Ratjadon: a new antifungal compound from Sorangium cellu- activity against rice blast causing fungus, Magnaporthe grisea. J
losum (Myxobacteria) production, physico-chemical and bio- Microbiol Biotechnol 12:1026–1028
logical properties. J Antibiot (Tokyo) 48:973–976. https://doi. Lee JY, Moon SS, Yun BS et al (2004) Thiobutacin, a novel antifungal
org/10.7164/antibiotics.48.973 and antioomycete antibiotic from Lechevalieria aerocolonigenes.
Gerth K, Washausen P, Hfletf G et al (1995b) The jerangolids : a family J Nat Prod 67:2076–2078. https://doi.org/10.1021/np049786v
of newantifungal compounds from Sorangium cellulosum (Myxo- Lim SM, Yoon M, Choi GJ et al (2017) Diffusible and volatile anti-
bacteria) production, physico-chemical and biological properties fungal compounds produced by an antagonistic Bacillus velezen-
of jerangolid A1. Optim Antibiot Product 49:71–75 sis G341 against various phytopathogenic fungi. Plant Pathol J
Gerth K, Bedorf N, Hflen G, Reichenbach H (1996) Epothilons A 33:488–498. https://doi.org/10.5423/PPJ.OA.04.2017.0073
and B: antifungal and cytotoxic compounds from Sorangium Liu CH, Chen X, Liu TT et al (2007) Study of the antifungal activity of
cellulosum (Myxobacteria) production. Physico-chem Biol Prop Acinetobacter baumannii LCH001 in vitro and identification of its
1:560–563 antifungal components. Appl Microbiol Biotechnol 76:459–466.
Ghanta S, Chattopadhyay S (2011) Glutathione as a signaling molecule https://doi.org/10.1007/s00253-007-1010-0
another challenge to pathogens. Plant Signal Behav 6:783–788. Llorens E, García-Agustín P, Lapeña L et al (2017) Advances in
https://doi.org/10.4161/psb.6.6.15147 induced resistance by natural compounds: towards new options
Gomez KA, Gomez KA, Gomez AA (1984) Statistical procedures for for woody crop protection. Sci Agric 74:90–100. https://doi.
agricultural research. Wiley, New York org/10.1590/1678-992x-2016-0012
Gond SK, Bergen MS, Torres MS, White JF (2015) Endophytic Bacil- Luckner M (2013) Secondary metabolism in microorganisms, plants
lus spp. produce antifungal lipopeptides and induce host defence and animals. Springer Science & Business Media, Berlin
gene expression in maize. Microbiol Res 172:79–87. https://doi. Ma GBZ, Xia Z, Wu S (2009) Inhibiting effect of seven marine actin-
org/10.1016/j.micres.2014.11.004 omycete strains against vegetable pathogenic microorganisms.
Grosch R, Scherwinski K, Lottmann J et al (2006) Fungal antago- Crops 5:3–9
nists of the plant pathogen Rhizoctonia solani: selection, control Madhurama G, Sonam D, Urmil PG, Ravindra NK (2014) Diversity
efficacy and influence on the indigenous microbial community. and biopotential of endophytic actinomycetes from three medici-
Mycol Res 110:1464–1474. https: //doi.org/10.1016/j.mycre nal plants in India. Afric J Microbiol Res 8:184–191. https://doi.
s.2006.09.014 org/10.5897/AJMR2012.2452
Gullner G, Zechmann B (2017) Glutathione in plant growth, devel- Mahesh H, Meghana S, Shailaja H et al (2016) Acquisition of the grass-
opment, and stress tolerance. Springer, Cham. https : //doi. hopper retro transposon by rice Magnaporthe isolates indicates a
org/10.1007/978-3-319-66682-2 dynamic gene flow between rice and non-rice Magnaporthe popu-
Gunatilaka AAL (2006) Natural products from plant-associated micro- lation. J Pathol Microbiol 1(2):1011
organisms: distribution, structural diversity, bioactivity, and impli- Mander P, Cho SS, Choi YH et al (2016) Purification and characteriza-
cations of their occurrence. J Nat Prod 69:509–526. https://doi. tion of chitinase showing antifungal and biodegradation properties
org/10.1021/np058128n obtained from Streptomyces anulatus CS242. Arch Pharm Res
Handelsman J (2005) Metagenomics: application of genomics to uncul- 39:878–886. https://doi.org/10.1007/s12272-016-0747-3
tured microorganisms. Microbiol Mol Biol Rev 69:195. https:// Martin M (2011) Cutadapt removes adapter sequences from high-
doi.org/10.1128/MMBR.69.1.195.2005 throughput sequencing reads. EMBnet J 17:10
Huang W, Liu X, Zhou X, et al (2018) Bacillus cereus HS24 sup- Moller Bahnweg G, Sandermann H, Geiger HHEM (1992) A simple
presses conidia germination of Magnaporthe oryzae by inhibiting and efficient protocol for isolation of high molecular weight DNA
the Ca2+ signaling pathway. http://dx.doi.org/10.1101/310516 from filamentous fungi, fruit bodies, and infected plant tissues.
Hung PQ, Annapurna K (2004) Isolation and characterization of endo- Nucl Acids Res 20:6115–6116
phytic bacteria in soybean (Glycine Sp.). Omonrice 101:92–101 Mousa WK, Raizada MN (2013) The diversity of anti-microbial sec-
Huson DH, Mitra S, Ruscheweyh H et al (2011) Integrative analysis of ondary metabolites produced by fungal endophytes: an inter-
environmental sequences using MEGAN4. Genome Res 21:1552– disciplinary perspective. Front Microbiol 4:1–18. https://doi.
1560. https://doi.org/10.1101/gr.120618.111.Freely org/10.3389/fmicb.2013.00065
Hwang BK, Lee JY, Kim BS, Moon SS (1996) Isolation, structure elu- Nagaraja A, Gowda J, Krishnappa M, Gowda KTK (2008) GPU 28: a
cidation, and antifungal activity of a manumycin-type antibiotic finger millet variety with durable blast resistance. J Mycopathol
from Streptomyces flaveus. J Agric Food Chem 44:3653–3657. Res 46:109–111
https://doi.org/10.1021/jf960084o Netam RS, Tiwari RK, Bahadur ANDS (2014) In vitro and in vivo
Jha Y, Subramanian RB (2013) Root associated bacteria from the local efficacy of fungicides against Pyricularia grisea causing finger
variety of rice GJ-17 antagonized the growth of Magnaporthe millet blast disease. Int J Plant Prot 7:137–142
grisea. Adv Biores 4(1):70–77 Ningthoujam DS, Kshetri P, Sanasam S, Nimaichand S (2009) Screen-
Jogaiah S, Kurjogi M, Govind SR et al (2016) Isolation and evaluation ing, identification of best producers and optimization of extracel-
of proteolytic actinomycete isolates as novel inducers of pearl lular proteases from moderately halophilic alkalithermotolerant
millet downy mildew disease protection. Sci Rep 6:1–7. https:// indigenous actinomycetes. World Appl Sci J 7:907–916
doi.org/10.1038/srep30789 Noguchi H, Taniguchi T, Itoh T (2008) Meta gene annotator: detecting
Kanehisa M, Goto S, Sato Y et al (2012) KEGG for integration and species-specific patterns of ribosomal binding site for precise gene
interpretation of large-scale molecular data sets. Nucl Acids Res prediction in anonymous prokaryotic and phage genomes. DNA
40:109–114. https://doi.org/10.1093/nar/gkr988 Res 15:387–396. https://doi.org/10.1093/dnares/dsn027
Knuat P, Reichenbach H (2000) On the mechanism of action of the Overbeek R, Olson R, Pusch GD et al (2014) The SEED and the Rapid
myxobacterial fungicide ambruticin. J Antibiot (Tokyo) 53:1182– Annotation of microbial genomes using Subsystems Technology
1190. https://doi.org/10.7164/antibiotics.53.1182 (RAST). Nucl Acids 42:206–214. https://doi.org/10.1093/nar/
Ladeuze S, Lentz N, Delbrassinne L et al (2011) Antifungal activ- gkt1226
ity displayed by cereulide, the emetic toxin produced by Bacil- Pandey SS, Singh S, Pandey H et al (2018) Endophytes of Withania
lus cereus. Appl Environ Microbiol 77:2555–2558. https://doi. somnifera modulate in planta content and the site of withanolide
org/10.1128/AEM.02519-10
13
3 Biotech (2020) 10:15 Page 17 of 17 15
biosynthesis. Sci Rep 20:1–19. https://doi.org/10.1038/s4159 Smith GE (1957) Inhibition of Fusarium oxysporum f. lycopersici by
8-018-23716-5 a species of Micromonospora isolated from tomato. Phytopathol-
Parks DH, Beiko RG (2010) Identifying biologically relevant dif- ogy 47:429–432
ferences between metagenomic communities. Bioinformatics Sziderics AH, Rasche F, Trognitz F et al (2007) Bacterial endophytes
26:715–721. https://doi.org/10.1093/bioinformatics/btq041 contribute to abiotic stress adaptation in pepper plants (Capsi-
Paul AK, Banerjee AK (1983) A new antifungal antibiotic produced by cum annuum L.). Can J Microbiol 53:1195–1202. https://doi.
Streptomyces galbus. Folia Microbiol (Praha) 28:386–396. https org/10.1139/W07-082
://doi.org/10.1007/BF02879488 Tokpah DP, Li H, Newmah JT et al (2017) Biological control of poten-
Rabeeth M, Anitha A, Srikanth G (2011) Purification of an antifun- tial antagonistic bacteria isolates to restrict Magnaporthe grisea
gal endochitinase from a potential biocontrol agent Streptomy- infection on rice. Afr J Microbiol Res 11:1108–1119. https://doi.
ces griseus. Pak J Biol Sci 14:788–797. https://doi.org/10.3923/ org/10.5897/AJMR2017.8562
pjbs.2011.788.797 Upson R, Read DJ, Newsham KK (2009) Nitrogen form influences
Ravin NV, Mardanov AV, Skryabin KG (2015) Metagenomics as a the response of Deschampsia antarctica to dark septate root
tool for the investigation of uncultured microorganisms. Russ J endophytes. Mycorrhiza 20:1–11. https://doi.org/10.1007/s0057
Genet 51:431–439. https://doi.org/10.1134/S1022795415050063 2-009-0260-3
Rodriguez RJ, Henson J, Van Volkenburgh E et al (2008) Stress toler- Van Minh N, Woo EE, Kim JY et al (2015) Antifungal substances
ance in plants via habitat-adapted symbiosis. ISME J 2:404–416. from Streptomyces sp. A3265 antagonistic to plant pathogenic
https://doi.org/10.1038/ismej.2007.106 fungi. Mycobiology 43:333–338. https : //doi.org/10.5941/
Romero LC, Aroca MÁ, Laureano-Marín AM et al (2014) Cysteine and MYCO.2015.43.3.333
cysteine-related signaling pathways in Arabidopsis thaliana. Mol Verma SC, Ladha JK, Tripathi AK (2001) Evaluation of plant growth
Plant 7:264–276. https://doi.org/10.1093/mp/sst168 promoting and colonization ability of endophytic diazotrophs
Sastri BN (1950) The Wealth of India. A dictionary of indian raw from deep water rice. J Biotechnol 91:127–141. https://doi.
materials and industrial products. Raw materials. Wealth India A org/10.1016/S0168-1656(01)00333-9
Dict Indian Raw Mater Ind Prod III:160–166 Vetriventhan M, Upadhyaya HD, Dwivedi SL et al (2015) Finger and
Schloss PD, Handelsman J (2005) Metagenomics for studying uncul- foxtail millets. Elsevier Inc., Amsterdam
turable microorganisms: cutting the Gordian knot. Genome Biol Vincent JM (1947) Distortion of fungal hyphae in the presence of cer-
6:6–9. https://doi.org/10.1186/gb-2005-6-8-229 tain inhibitors. Nature 159:850
Schulz B, Guske S, Dammann U, Boyle C (1998) Endophyte-host inter- Vurukonda SSKP, Giovanardi D, Stefani E (2018) Plant growth pro-
actions. II. Defining symbiosis of the endophyte-host interaction. moting and biocontrol activity of Streptomyces spp. As endo-
Symbiosis Phila Pa (USA) 25:213–227 phytes. Int J Mol Sci. https://doi.org/10.3390/ijms19040952
Selim HMM, Gomaa NM, Essa AMM (2016) Antagonistic effect of Waller F, Achatz B, Baltruschat H et al (2005) The endophytic fun-
endophytic bacteria against some phytopathogens. Egypt J Bot gus Piriformospora indica reprograms barley to salt-stress toler-
56:613–626 ance, disease resistance, and higher yield. Proc Natl Acad Sci
Sengupta S, Ganguli S, Singh PK (2017) Metagenome analysis of the 102:13386–13391. https://doi.org/10.1073/pnas.0504423102
root endophytic microbial community of Indian rice (O. sativa White TJ, Bruns T, Lee S, Taylor JL (1990) Amplification and direct
L.). Genom Data 12:41–43. https: //doi.org/10.1016/j.gdata sequencing of fungal ribosomal RNA genes for phylogenetics.
.2017.02.010 Academic Press, San Diego
Sessitsch A, Hardoim P, Döring J et al (2012) Functional characteris- Xu SJ, Hong SJ, Choi W, Kim BS (2014) Antifungal activity of Paeni-
tics of an endophyte community colonizing rice roots as revealed bacillus kribbensis strain T-9 isolated from soils against several
by metagenomic analysis. Mol Plant-Microbe Interact 25:28–36. plant pathogenic fungi. Plant Pathol J 30:102–108. https://doi.
https://doi.org/10.1094/MPMI-08-11-0204 org/10.5423/PPJ.OA.05.2013.0052
Shankar M, Kurtböke DI, Gillespie-Sasse LMJ et al (1994) Possible Yeop LJ, Sherman DH, Hwang BK (2008) In vitro antimicrobial and
roles of competition for thiamine, production of inhibitory com- in vivo antioomycete activities of the novel antibiotic thiobutacin
pounds, and hyphal interactions in suppression of the take-all fun- Pest Management Science: formerly. Pestic Sci 64(2):172–177
gus by a sterile red fungus. Can J Microbiol 40:478–483. https:// Yoshii A, Moriyama H, Fukuhara T (2012) The novel kasugamy-
doi.org/10.1139/m94-077 cin 2′-N-acetyltransferase gene aac(2′)-IIa, carried by the IncP
Shirke MD, Mahesh HB, Gowda M (2016) Genome-wide comparison island, confers kasugamycin resistance to rice-pathogenic bacte-
of Magnaporthe species reveals a host-specific pattern of secretory ria. Appl Env Microbiol 78:5555–5564. https://doi.org/10.1128/
proteins and transposable elements. PLoS One 11:1–19. https:// AEM.01155-12
doi.org/10.1371/journal.pone.0162458
Singh N, Varma A (2015) Antagonistic Activity of Siderophore Pro-
ducing Rhizobacteria Isolated from the Semi-Arid Regions of
Southern India. Int J Curr Microbiol App Sci 4(9):501–510
13