Bacillus aryabhattai AB211 was isolated using a functional screening based method described previously (Ghosh et al., 2007), from tea rhizosphere of Rohini, Darjeeling district, West Bengal, India. Strain AB211 was routinely grown in M9 minimal medium supplemented with glucose. The culture was incubated at 37°C on a rotary shaker (150 rpm) for desired period with or without amendments of antibiotic (tetracycline 35 μg ml^-1).

To test the ability of B. aryabhattai AB211 to use inorganic and organic insoluble phosphate as a phosphorous source, Pikovskaya agar plates were used (Pikovskaya, 1948). The reaction was considered positive when a clear halo surrounding the bacterial colonies was observed after 3–7 days of incubation at 37°C. Furthermore, the ability of the strain AB211 to solubilize inorganic phosphate was quantitatively assessed. 50 μl of overnight grown culture was inoculated in 5 ml Pikovskaya’s broth and incubated for 7 days. Uninoculated medium was used as a control. Following incubation, the cell suspension was centrifuged, and the available phosphate content of the supernatant was estimated by malachite green method (Ghosh et al., 2011). The experiment was performed in triplicate.

Chrome azurol sulfonate (CAS) agar solid medium was used to screen siderophore production (Alexander and Zuberer, 1991). The reaction was considered positive when an orange halo surrounding the bacterial colony appeared due to the removal of iron from CAS by the siderophore. Quantification of siderophore production was estimated by following the formula: % siderophore unit = [(Ar-As)/Ar] × 100, where, Ar = absorbance of reference (minimal medium + CAS assay solution), As = absorbance of the sample.

To detect the production of ammonia, the strain was grown in 4% peptone broth and incubated for 7 days at 37°C. Following incubation, 0.5 mL of Nessler’s reagent was added to the bacterial suspension. The development of deep yellow color with brown precipitation indicated ammonia production.

Production of indole-3-acetic acid (IAA) by the strain AB211, was estimated by growing the isolate in M9 medium, supplemented with L-tryptophan as a precursor of IAA at different concentrations (0, 50, 100, 200, 300, 400, and 500 mg l^-1), and incubated for 48 h. The supernatant of the culture fluid was mixed with Salkowski’s coloring reagent (50 ml of 35% HClO₄, 1 ml of 0.5 M FeCl₃) in the ratio of 2:1 and incubated in the dark for 30 min. The absorbance was measured at 540 nm.

The synthesis of exopolysaccharides (EPS) was estimated following the method described previously (Chiba et al., 2015). Quantification of proteins in the isolated EPS fractions was carried out using standard Bradford protein assay and total saccharide was estimated following phenol-sulfuric acid method.

Bacterial identification was conducted based on morphology and biochemical characterization. The morphological, cultural, and physiological characteristics of the isolated strain was compared with data from Bergey’s Manual of Determinative Bacteriology (Holtz, 1993). To identify the strain AB211 by carbon source utilization pattern, the GEN III MicroPlate^TM test (Biolog Inc., Hayward, CA, USA) was carried out, which provides a standardized micro-method using 94 biochemical tests to profile, and characterize a broad range of bacteria (Bochner, 1989). This technique analyzes a microorganism depending on phenotypic tests which include 71 carbon source utilization assays and 23 chemical sensitivity assays.

Total DNA was isolated from B. aryabhattai AB211 as described for B. subtilis according to the method of Bron and Venema (1972). Genome sequencing of B. aryabhattai AB211 was performed at Bionivid Technology Pvt. Ltd (Bengaluru, Karnataka, India) using an Illumina platform using HiSeq Illumina paired-end technology with 151 bp of reads. All reads were quality assessed using NGSQC toolkit. Primary genome assembly using Velvet (v 1.2.10) (Zerbino and Birney, 2008), and further scaffolding of primary assembly using SSPASE (v 3.0) (Boetzer et al., 2011) were performed. De novo genome validation and quality control were performed using Bowtie 2 (v 2.2.2) (Langmead et al., 2009).

Putative coding sequences (CDS) were identified by the RAST server (Aziz et al., 2008; Overbeek et al., 2014). All CDS identified were manually reviewed, and false CDS were flagged as “artifact.” The remaining CDS were then submitted to automatic functional annotation via BLAST searches against the UniProt databank to determine significant homology. Putative tRNAs were identified using ARAGORN (v 1.2.36) (Laslett and Canback, 2004), and rRNAs were identified using RNAmmer 1.2 server (Lagesen et al., 2007). The presence of plasmid derived sequence was verified using Webcutter (v 2.0), and Plasmid Finder (v1.3) (Carattoli et al., 2014). Detection of bacteriophage sequences was performed using PHAST (Zhou et al., 2011). The taxonomic identification was performed using MEGA6 (Tamura et al., 2013). Finally, genome finishing was carried out using CONTIGuator (V 2.7) (Galardini et al., 2011) applying closest homolog of the assembled genome as a reference.

For comparative genomic analysis of AB211 strain, genome sequences of seven other B. aryabhattai strains (Table 1) were downloaded from NCBI ¹. For identification of reference genome, a whole genome nucleotide based alignment (BLAST) with strain AB211 in the NCBI ‘nr’ database was performed with >95% identity and coverage values. Identified highly similar genomes (Table 1) were then subjected to ANI analysis, and the most identical complete genome sequence, viz. B. megaterium Q3, also having close evolutionary relationship with strain AB211 as evident in phylogenetic analysis, was selected as reference genome for genome alignment of AB211. To arrange the genomic assemblies of the draft genome sequences, contigs were ordered and oriented by promer (Kurtz et al., 2004) based on their genomic alignments with the complete genome sequence of B. megaterium Q3. Ordered contigs were then pasted together to form a pseudo-chromosome, where contig boundaries were separated by a spacer sequence, as suggested previously (Tettelin et al., 2005).

To identify the core and strain-specific genes of AB211, a pan-genome analysis of gene distribution of AB211 was carried out against other B. aryabhattai strains. In this study, five strains of B. megaterium (Table 1), having high genomic and evolutionary similarities with B. aryabhattai AB211, were also considered. The pan-genome analysis of these 13 genomes was carried out in BPGA pipeline (Chaudhari et al., 2016), with default criteria. The specific set of genes (present and/or absent) of AB211 was then subjected to Clusters of Orthologous Group (COG) category-wise classification scheme by performing PSI-BLAST against NCBI COG database in WebMGA analysis server with default parameters (Wu et al., 2011). The statistical significance of the abundance of individual COG categories was determined by a 2 × 2 contingency table in STATISTICA version 6 (StatSoft Inc., 2001).

To understand the genome-wise distribution of orthologous clusters and to find out whether recombination has played any role in the evolution of B. aryabhattai strains, a genome-wide synteny analysis was performed in GSV (Revanna et al., 2011) with various criteria. Multiple genome alignment utility in progressive Mauve (Darling et al., 2010) was also used for the same. The probable recombination regions were identified from these syntenic blocks. Individual gene sequences were extracted from these homologous clusters and further aligned with ClustalW in MEGA6 (Tamura et al., 2013) to search for any recombination signature present in them by RDP analysis (Martin et al., 2010).

Based on the obtained 16S rRNA gene sequence of B. aryabhattai AB211, and multiple alignments, a phylogenetic tree was constructed by the Neighbor-Joining method, and confidence level was estimated for 1,000 replicates using MEGA6 (Molecular Evolutionary Genetics Analysis) (Tamura et al., 2013). Unaligned regions and gaps were excluded from the analyses. Sequences of representative Bacillus strains and out-group were obtained from NCBI GenBank². Core and pan-genome phylogenetic studies were performed in BPGA pipeline with default criteria (Chaudhari et al., 2016).

The assay was performed in 96-well tissue culture plates (Nunclon Delta Surface, Thermo Scientific) to study the efficiency of the strain AB211 in biofilm formation under varied incubation time in M9 medium (Koerdt et al., 2011). After inoculation with the culture and incubation for 48, 72, and 96 h, respectively, the microtiter plates were cooled down to room temperature and the OD₆₀₀ of the planktonic cells from each well was measured using a plate reader (FLUOstar, OPTIMA) at a wavelength of 600 nm. Culture supernatant was then removed from each well and the cells attached to the well were quantitatively estimated using crystal violet (CV). Ten microliter of a 0.5% solution of CV was added to each well and incubated at room temperature for 10 min. Subsequently, the liquid supernatant was removed from each well and the biofilm cells attached to the well were washed with water. Hundred percentage ethanol was added to release the CV from the biofilm. The absorbance of CV from each well was measured at a wavelength of 570 nm. The percentage of cells within the biofilm was calculated by determining the correlation between the growth of the cells (OD_{600 nm}) and the absorbance of CV (OD_{570 nm}). Each set was performed in triplicate.

A 16 h grown culture of AB211 was inoculated (1% v/v) in M9 minimal medium and incubated for 24 h at 37°C. After incubation, cells were harvested by centrifugation and washed thrice with sodium phosphate buffer (pH 7.4). Cells were then fixed with 0.5% glutaraldehyde solution and washed thrice with sodium phosphate buffer. Dehydration of cells was carried out in a gradient of ethanol and finally incubated in 100% ethanol for 1 h. Two–three microliter of fixed cells were placed on a small glass slide and dried (Focardi et al., 2010; Banerjee et al., 2011). SEM observation was performed using a scanning electron microscope (Zeiss-EVO18).

For biofilm study, 16 h grown culture of the strain AB211 was inoculated on cover slips (1% v/v) kept in M9 minimal medium and incubated for 48, 72, and 96 h, respectively, at 37°C without disturbance. The biofilm was fixed following the same protocol as mentioned earlier. After complete dehydration, the cover slips were dried and viewed under SEM.

To assess the ability of B. aryabhattai AB211 to colonize on maize roots, 50 ml of M9 minimal medium (supplemented with 0.5% w/v glucose), inoculated with the strain AB211 was incubated overnight at 37°C under shaking conditions. The cells were harvested by centrifugation at 6000 rpm for 5 min and were washed twice in M8 buffer (22 mM Na₂HPO₄, 22 mM KH₂PO₄, 100 mM NaCl, pH 7.0). Finally, the harvested cells were suspended in 250 ml of M8 buffer. Roots of the maize seedlings were washed thoroughly with sterile water to get rid of attached soil particles, and then soaked in bacterial suspension for 1 h under aseptic conditions, and transferred back to sterile bottles. Control was maintained by following the same protocol with M8 buffer, excluding the cells. Both the plants were transferred to plant growth chamber at 28°C with a 16 h light regimen. The roots were then cut and fixed in 2.5% glutaraldehyde in 0.075 M phosphate buffer overnight, and processed further for dehydration and visualization as described above.

Based on the performance in the in vitro experiments, B. aryabhattai AB211 was further evaluated for its plant growth promoting potential on maize seedlings in pot trials. Surface sterilized maize seeds (variety: Early Golden Bantam) were sown in sterile pots (one seed/pot) filled with sterile soilrite (Keltech Energies Limited, Bangalore, India). After germination, rhizospheres of 7-day-old seedlings were inoculated with the isolate (inoculation with about 10⁸ cfu/ml of AB211 culture; and in soil, resulting cfu of AB211 was approximately 10⁷ cfu/gm). Control was maintained by applying sterile medium without culture on the plants. Sixteen replicates were maintained for both control and treatment sets. The plants were regularly irrigated. Observations were taken on the 15th day of treatment. Growth parameters such as total chlorophyll content, height, wet weight, dry weight of the shoot and root were measured and statistically analyzed.

The whole-genome shotgun project of B. aryabhattai AB211 was deposited at DDBJ/EMBL/GenBank under the accession number MCAN00000000. The accession number for submitted 16S rRNA gene sequence is KP896525.1.

The main features of the B. aryabhattai AB211 genome have been summarized in Table 1. The circular chromosome (5,403,026 bp) was found to be somewhat smaller to that of the closely related B. megaterium (Arya et al., 2014; Wang et al., 2016). The whole genome sequence of strain AB211 was obtained by an Illumina platform using HiSeq Illumina paired-end technology with 151 bp reads. Primary genome assembly using Velvet (v 1.2.10) (Zerbino and Birney, 2008), and further scaffolding of primary assembly using SSPASE (v 3.0) (Boetzer et al., 2011) revealed that the draft genome consists of 23 scaffolds with an average genome length of ∼5.4 Mbp, G+C content of 37.82%, and N₅₀ size of 4199117 (∼4.19 Mb) bp. Genome annotation by RAST server revealed that the draft genome has 5226 protein CDSs, 16 rRNA genes, 120 tRNAs, 8 ncRNAs, 58 non-protein coding genes, and 11 prophage regions. No plasmid was identified when analyzed using Webcutter (v 2.0) and Plasmid Finder (v1.3) (Carattoli et al., 2014). The taxonomic identification using MEGA6 (Tamura et al., 2013) revealed that the strain AB211 exhibits closest phylogenetic relationship to the B. megaterium strain Q3.

In RAST annotation, genes encoding for transport system, plant–bacterial interaction, secretion system, antibiotic resistance, surface appendages/exopolysaccharides synthesis, heavy metal resistance/mobilization, and stress response were observed. In general, B. aryabhattai AB211 genome revealed high metabolic diversity. In reference to plant-microbe interaction, the AB211 genome analysis revealed 300 proteins in carbohydrate metabolism, 54 proteins in flagella assembly, function and signaling; 19 proteins in nitrogen metabolism, 19 proteins in phosphorous metabolism, 29 proteins in siderophore biosynthesis and iron acquisition, 74 proteins in stress response, 43 proteins in antibiotic and heavy metal resistance, 17 proteins in plant hormone/volatile compound synthesis and 16 proteins in aromatic compound degradation pathway (Supplementary Table S1).

To understand the evolutionary relationship of B. aryabhattai AB211 with other Bacilli, a 16S rRNA Neighbor-Joining phylogeny of AB211 with 24 other known Bacillus species was performed. Phylogenetic analysis revealed the close evolutionary relationship of AB211 with B. megaterium (Figure 1). To further illustrate the evolutionary relationship, both core and pan-genomic phylogeny of B. aryabhattai strains were constructed with those strains of B. megaterium that have high sequence similarities with B. aryabhattai AB211 (>95% identity and coverage) (Table 1). Core and pan-matrix phylogeny revealed that, in genome scale, there was very little difference in B. aryabhattai and B. megaterium strains. In fact, in both core and pan-genome phylogeny, few strains of B. megaterium were found to be more related to AB211 than other B. aryabhattai strains (Supplementary Figure S1). Among these strains of B. megaterium, complete genome sequence of strain Q3, having maximum genomic similarity with AB211, and also having close relationship in both core and pan-matrix phylogeny, was being selected as a reference genome for comparative analysis. Average Nucleotide Identity (Goris et al., 2007) (ANI) value also indicated a high degree of reciprocal sequence similarity (Average ANI = 96.35%) between these two genomes (Supplementary Figure S2). This close homology of B. megaterium with B. aryabhattai strains have been well reported in the previous analyses (Ray et al., 2012), emphasizing a common evolutionary path of these two species of Bacilli.

For genomic comparisons, seven other B. aryabhattai genome sequences were downloaded from NCBI (Table 1). Genomic alignments were performed for all of these draft genomes against a complete genome sequence of B. megaterium strain Q3. Contigs were reordered, and pseudochromosomes were constructed for each of the draft genomes. To check the consistency of the draft genome sequences of B. aryabhattai strains, a whole genome-wide synteny analysis was performed both with the complete genome of B. megaterium Q3 and with each other (Supplementary Figure S3), applying different orthology criteria. Synteny analysis of strain AB211 with B. megaterium Q3 revealed an overall consistent genomic arrangement in each of these closely related species (Supplementary Figure S3A). Locally Collinear Blocks (LCB) in progressive Mauve (Darling et al., 2010) also showed no major recombination chunks among the eight strains of B. aryabhattai (Supplementary Figure S3B). Furthermore, a genome synteny analysis among these strains was performed with different criteria of identifying probable recombination blocks (such as e-value, base pair length etc.) in GSV (Revanna et al., 2011). Probable recombination regions were marked, and individual gene sequences from these regions were extracted and aligned with ClustalW in MEGA6 (Tamura et al., 2013) for recombination study with RDP (Martin et al., 2005). Three different recombination detection methods were tested in these sequences, viz. GENECONV (Sawyer, 2000), Bootscan (Martin et al., 2005) and MaxChi (Smith, 1992), but none of the above mentioned program showed any evidence of recombination in these gene alignments.

By a comparative BLASTN analysis with BLAST Ring Image Generator (BRIG), the newly sequenced genome of B. aryabhattai strain AB211 was compared with partially assembled genomes of seven other B. aryabhattai strains (Figure 2). Complete genome sequence of B. megaterium strain Q3, being phylogenetically close and having maximum genome identity (Supplementary Table S2) with strain AB211 (beside type strain), was taken as a reference genome. The overall genome of AB211 has a high degree of sequence similarities with other B. aryabhattai strains. The GC content and overall GC skew, when compared with B. megaterium Q3, showed certain patches of atypical GC usage, possibly implying regions of horizontal gene transfer (HGT).

To further understand the gene usage of AB211, a pan-genomic analysis was carried out among eight B. aryabhattai strains and five evolutionary as well as genetically similar B. megaterium strains (Table 1 and Supplementary Figure S4). In these 13 highly similar genomes, the core-genome constitutes of 3558 genes, whereas the pan-genome has 7392 genes (Supplementary Tables S3, S4). The power law regression in the BPGA pipeline estimated that the pan-genome of B. aryabhattai and B. megaterium was still open (Supplementary Table S4). The core genes, that were shared among all these strains, constituted of 3558 genes and the dispensable genes, which were shared by some but not all, were mainly responsible for differential characteristics of these strains. 158 gene families were found to be exclusively present in AB211 but absent in all other strains of B. aryabhattai (Supplementary Table S5). These 158 unique genes are probably foreign in nature and could be inherited by HGT events. GC% analyses of those genes showed that these genes have significantly lower G+C content than the genomic GC of strain AB211 (Supplementary Figure S5), supporting the probable HGT events from lower G+C content organism. Although most of these unique gene products were uncharacterized, the annotated ones were mostly related to transport of small molecules and ions, transcriptional regulators, and membrane proteins. Interestingly enough, when these genes were arranged in AB211 chromosome after contig re-ordering, they appeared to be contiguous in AB211 genome. Among these, a continuous stretch of ∼200 kbps region (Figure 2), encoding 85 proteins, were exclusively present in B. aryabhattai AB211. To functionally characterize these proteins, COGs analysis of the encoded proteins was performed, and their significant abundance levels were assessed with a 2 × 2 chi-square contingency table in STATISTICA 6.0 (StatSoft Inc., 2001). The significantly abundant COG categories (p < 0.05) have been shown in Figure 3. The most abundant COG categories among these exclusively present proteins of AB211, were COG category M (Cell wall/membrane/envelope biogenesis, P = 0.09), and COG category G (Carbohydrate transport and metabolism, p = 0.07). The other most abundant COG categories, although not statistically significant, were COG category K (Transcription), and COG category E (Amino acid transport and metabolism). Individual COG categories, and their relative abundances have been shown in Supplementary Table S6. Abundances of these specific functional categories indicate the unique functional attributes of AB211 within its distinct ecological niche, which require further thorough examination by biochemical studies.

Bacillus aryabhattai AB211 is a Gram-positive rod shaped spore forming firmicute. Classical biochemical test evidenced that the strain AB211 was positive for MR- VP test and indole production, and negative for catalase, amylase and gelatinase test (Supplementary Figure S6A). Results of other biochemical tests have been summarized in Supplementary Figures S6A,B. Besides, strain AB211 showed positive motility through diffused cloudy growth away from the line of inoculation in semisolid M9 agar medium (Supplementary Figure S7D). It also showed resistance to the antibiotic tetracycline at 35 μg/ml concentration. The isolate was resistant against all the tested heavy metals at 5 mM concentration. It was found that the strain AB211 can grow in M9 medium supplemented with either of 0.5% (w/v) sucrose, D-glucose, D-fructose, D-maltose, D-galactose, L-arabinose, and D-mannitol as a sole source of carbon, which was further confirmed using GEN III MicroPlate^TM test. Identification of the strain AB211 on the basis of carbon source utilization pattern, using the GEN III MicroPlate^TM test revealed that the strain AB211 could utilize several other carbohydrates and amino acids, grow at high salt concentration (at 8%), and is resistant to several antibiotics like- rifamycin, troleandomycin, lincomycin, nalidixic acid, and aztreonam (Supplementary Figure S6B).

Phosphate solubilizing bacteria increases phosphorus uptake of crop plants by releasing insoluble and fixed forms of phosphorus from soil (Rodriguez and Fraga, 1999). After 7 days of incubation, the strain AB211 was found to form a clear zone around the point of inoculation on Pikovskaya’s agar plate, indicating phosphate solubilization. In the quantitative estimation, strain AB211 was found to solubilize 199.09 ± 0.18 μg/ml of inorganic phosphate. The pH of the broth was found to decline to 4.8 from 7.0 (control), due to the bacterial activity. Furthermore, quantitative estimation of exopolysaccharides revealed that the isolated EPS fraction from 72 h grown AB211 contained about 272.5 μg/ml of protein and 185 μg/ml of carbohydrate.

The ability of B. aryabhattai AB211 to efficiently colonize on the surfaces of plant roots is a prerequisite for phytostimulation. Based on the genome analysis, B. aryabhattai AB211 seems well adopted to thrive in the plant rhizosphere as it encodes essential features required for its survival. In general, colonization of the root surfaces followed by phytostimulation involves different events: (a) movement (swimming) of bacteria toward plant root, (b) survival within the rhizospheric environment, e.g., survival in the presence of plant responses (oxidative stress and root exudates) and inter-species competition between microbial communities (antibiotic sensitivity), (c) adhesion and colonization of the root surfaces (biofilm), and finally (d) synergistic interactions with host plant (metabolic versatility) viz a viz plant growth promotion. An overview of each of these events as evident from genome analysis and supporting experimental results is presented in the subsequent section.