Introduction

Soybean (Glycine max) (L.) Merr. is the dominant and the most important commercial legume crop species, yielding food oil and animal meal as well as nutritious vegetable protein [1–3]. The plant was first introduced into USA agriculture during the mid-18th century and was mainly used as a forage crop until the 1920s [4]. The development of new cultivars, along with technological advances in soybean processing and increased demand for soybean products, has led to major increases in production during the 20th century [4].

As with most papilionoid legumes, soybean engages in a symbiotic relationship with dinitrogen-fixing soil bacteria known as rhizobia and is able to obtain on average 50–60% of its required nitrogen through symbiotic nitrogen fixation [5]. A greater understanding of the symbiosis between soybean and its cognate rhizobia is of direct relevance for maintaining environmentally sustainable high crop yields, which significantly contributes to the Sustainable Development Goals adopted in September 2015 as part of the UN’s development agenda ‘Transforming our world: the 2030 Agenda for Sustainable Development’ [6].

The soybean-nodulating bacteria, known as Rhizobium japonicum according to a 1929 classification scheme [7], were reclassified as Bradyrhizobium japonicum in 1982 because of several fundamental morphological and physiological differences with the genus Rhizobium [8]. The bacteria isolated from nodules of soybean had previously been shown to be phenotypically diverse, even though they were grouped together in the species Bradyrhizobium japonicum. One of the major methods that demonstrated this diversity was serology, which was used to classify individual isolates into 17 distinct serogroups [9]. This was accomplished by generating antisera to specific strains in the USDA collection in Beltsville and then using the sera to generate a serological scheme. One of the strains used to generate antisera was USDA 76T and all isolates that cross-reacted with the antiserum generated with this serotype strain were combined together in the 76 serogroup. The strain USDA 76T deposited in the Beltsville collection was a re-isolate from a greenhouse-grown plant inoculated with USDA 74 in Maryland. In turn, USDA 74 was a re-isolate of USDA 8 from a plant passage field test in California in 1956. The original parent culture of USDA 76T is USDA 8, which was isolated from soybean grown at the Arlington Farm, Virginia in 1915.

Differences among the soybean root nodule bacteria classified as B. japonicum were also demonstrated using molecular methods. Hollis et al. [10] reported the presence of three DNA homology groupings by analysis of 28 strains within the soybean rhizobia. Using this approach, nine of the 17 serogroups were assigned to three DNA homology groupings: group I, the closely related group Ia and the more divergent group II. Supporting evidence for these three groupings was obtained by Kuykendall et al. [11]. By sequence analysis of the 16S rRNA genes, each of the 17 serotype strains representing the serogroups were also placed into three closely related groups [12] that matched their separation by DNA homology. Since soybean strains could be distinguished phenotypically and by several approaches in molecular biology, Kuykendall et al. [13] proposed that DNA homology group II strains be separated from B. japonicum as the species Bradyrhizobium elkanii , with USDA 76T as the type strain.

Because of these distinguishing characteristics and its significance as a microsymbiont of the economically important legume soybean, B. elkanii USDA 76T was selected as part of the DOE Joint Genome Institute 2010 Genomic Encyclopedia for Bacteria and Archaea-Root Nodule Bacteria sequencing project [14, 15]. Here we present a summary classification and a set of general features for B. elkanii strain USDA 76T , together with a description of its genome sequence and annotation.

Organism information

Classification and features

Bradyrhizobium elkanii USDA 76T is a motile, non-sporulating, non-encapsulated, Gram-negative strain in the order Rhizobiales of the class Alphaproteobacteria . The rod shaped form has dimensions of approximately 0.5 μm in width and 1.0–2.0 μm in length (Fig. 1 Left and Center). It is relatively slow growing, forming colonies after 6–7 days when grown on ½ Lupin Agar [16], Modified Arabinose Gluconate [17] and modified Yeast Mannitol Agar [18] at 28 °C. Colonies on ½ LA are opaque, slightly domed and moderately mucoid with smooth margins (Fig. 1 Right).

Fig. 1
figure 1

Images of Bradyrhizobium elkanii USDA 76T using scanning (Left) and transmission (Center) electron microscopy as well as light microscopy to visualize colony morphology on solid media (Right)

Full size image

Sequence divergence among the 16S rRNA genes of the 33 type strains within the genus Bradyrhizobium was limited and ranged from no differences in many cases to a similarity of 98% between B. elkanii USDA 76T and B. neotropicale (Fig. 2) after accounting for 40 bp in gaps along the alignment length. Such high similarity values would question the reliability of defining species limits within the genus based on divergence of the 16S rRNA genes [19]. Bootstrap values for each of the nodes of the branches were low and none of the confidence values reached or exceeded 95%. Therefore, the placement of each of the taxa relative to the others in the tree is inconclusive.

Fig. 2
figure 2

Comparison of the 16S rRNA gene of Bradyrhizobium elkanii USDA 76T (shown in bold blue print) with those of other proposed Bradyrhizobium species and the serotype strains of the remaining 16 serogroups of the soybean bradyrhizobia. DNA homology affiliation of the different soybean serogroup strains are indicated within the rectangles. DNA homology values for the serogroup strains USDA 4, USDA 94, USDA 124, USDA 126, USDA 127, USDA 129 and USDA 135 were not reported. The sequences were initially aligned by using the software MEGA, version 5 [67]. Subsequently the alignment was manually inspected for errors and necessary corrections were made by using GeneDoc version 2.6.001 [68]. The outgroups Mesorhizobium loti LMG6125T and M. ciceri UPM-Ca7T were chosen because of the reported recombination events between the 16S rRNA genes of B. elkanii and Mesorhizobium [22]. Of the 1313 active sites of the alignment there were 40 gaps among the Bradyrhizobium taxa. The number of different base pairs among all the 35 aligned sequences (including the two Mesorhizobium species) was determined by using MEGA, version 5 [67] to generate a tree using the UPGMA algorithm. Bootstrap analysis [69] with 2000 permutations of the data set was used to determine support for each of the branches. Type strains are indicated by name in the figure. Strains in the figure with a genome sequencing project registered in GOLD [70] are as follows: B. daqingense (2596849087), USDA 110 (640700549), USDA 76 (2513649183), USDA 6 (2513666035), B. pachyrhizi (2655289729), and B. yuanmingense (2617374406)

Full size image

Genetic recombination resulting in a reticulate evolutionary history of the 16S rRNA gene is perhaps a likely explanation for the low bootstrap values. Therefore, an analysis for recombination was done with the aligned 33 Bradyrhizobium 16S rRNA genes using the pairwise homoplasy index test [20]. By using this test, statistically significant evidence for recombination among the 33 16S rRNA genes was detected (P = 0.003). The detection of genetic recombination within the rrn loci of rhizobia is not unprecedented since reticulate evolutionary histories of the 16S rRNA genes and the Internally Transcribed Spacer between the 16S and 23S rRNA genes has been described before [21, 22]. The 16S rRNA sequence of B. pachyrhizi was identical with those of the B. elkanii serogroup strains USDA 31, USDA 94 and USDA 130, which differed from B. elkanii USDA 76T by one bp (99.999% similar). The most divergent 16S rRNA gene within B. elkanii was that of the serogroup strain USDA 46 (99.996% similar), while the most divergence among the soybean serogroup strains was that between USDA 46 and USDA 110, which were 98.4% similar. Since the divergence of the 16S rRNA genes of the genus Bradyrhizobium is narrow, with evidence for the presence of a history of genetic recombination, it may be necessary to more precisely establish their phylogeny by comparing their entire genomes rather than individual genes. Such an approach may provide more fundamental insight into the evolutionary history of this class of symbiotic bacteria as well as impacting potential changes in their current proposed taxonomy. Minimum Information about the Genome Sequence of USDA 76T is provided in Table 1 and Additional file 1: Table S1.

Table 1 Classification and general features of Bradyrhizobium elkanii USDA 76T in accordance with the MIGS recommendations [71] published by the Genome Standards Consortium [72]
Full size table

Symbiotaxonomy

An investigation of the symbiotic properties of soybean began with the work of Brooks [23] in the late 19th century, when he observed that soybean grown in the fields of his experiment station in Massachusetts only nodulated when supplied with dust he had brought with him from Japan. This led to the theory that soybean-nodulating bacteria in the soils of the USA were imported from the Far East. Cotrell et al. [24] and Hopkins [25] reported the supporting evidence that soybean in Kansas nodulated with soil taken from the Massachusetts Experiment station, or in Illinois from soil collected from fields with a history of soybean cultivation. However, several decades later it became evident that rhizobia that nodulated native American legumes within the genera Apios , Amphicarpa , Crotalaria , Desmodium , Lespedeza , Baptisia , Cassia , Genista and Wisteria also nodulated soybean [26–28]. With the exception of USDA 6 and USDA 38, which are from Japan, all the remaining soybean serotype strains were recovered from nodules of soybeans grown in the USA, including USDA 8 (the original parent of USDA 76T ). Consequently, it is unclear whether these rhizobia obtained from nodules of USA-grown soybean originate from the Far East or are in fact native to the soils of America. Therefore, the possibility exists that USDA 76T may be able to nodulate and form a symbiosis with a wide variety of legumes, but this has not been thoroughly investigated. Unfortunately, the communication that included the proposal of USDA 76T as the type strain for B. elkanii did not include results of plant tests to describe its symbiotic range, but instead relied on distinction by phenotype and genotype [11]. An indication of the possible American origin of USDA 76T is its reported effectiveness in symbiosis with the native Apios americana Medik. and use as an inoculum for this potential leguminous crop [29]. Further evidence for this theory is the ability of USDA 76T to nodulate and fix nitrogen with the native American Amphicarpaea bracteata (L.) Fernald [30]. USDA 76T effectively nodulates the promiscuous Vigna unguiculata (L.) Walp. (cowpea), but is unable to nodulate the tropical American legume Phaseolus lunatus L. (Lima bean), which forms nodules with various other strains of bradyrhizobia [31]. To our knowledge, the only other reported information is that USDA 74 (parent of USDA 76T ) forms an effective symbiosis with Macroptilium atropurpureum (DC.) Urb. (Siratro) and Vigna unguiculata (L.) Walp [32].

In soybean, the Rj(s) or rj(s) genetic loci have been identified as controlling the ability of compatible rhizobia to nodulate with a particular cultivar (reviewed by Hayashi et al. [33]). USDA 76T is reported to form nodules (albeit in reduced numbers) on the cultivar Clark (rj1) and to nodulate and fix N2 with the isogenic lines BARC-2 and BARC-3, harboring the Rj4 and rj4 alleles, respectively, when tested in Leonard jars with sterile vermiculite or sand [30]. The symbiotic characteristics of B. elkanii USDA 76T on a range of selected hosts are summarized in Additional file 2: Table S2.

Genome sequencing information

Genome project history

This organism was selected for sequencing at the U.S. Department of Energy funded Joint Genome Institute as part of the Genomic Encyclopedia of Bacteria and Archaea-Root Nodule Bacteria project project [14, 15]. The root nodule bacteria in this project were selected on the basis of environmental and agricultural relevance to issues in global carbon cycling, alternative energy production, and biogeochemical importance. In particular, strain USDA 76T was chosen since it is a microsymbiont of the economically important legume soybean, but can also form symbioses with several legumes native to the USA. The USDA 76T genome project is deposited in the Genomes Online Database [34] and a high-quality permanent draft genome sequence is deposited in IMG [35]. Sequencing, finishing and annotation were performed by the JGI [36] and a summary of the project information is shown in Table 2.

Table 2 Genome sequencing project information of Bradyrhizobium elkanii strain USDA 76T
Full size table

Growth conditions and genomic DNA preparation

After recovery from permanent storage, the B. elkanii USDA 76T was streaked onto MAG solid medium and grown at 28 °C for 6 days to obtain well grown, well separated colonies, then a single colony was selected and used to inoculate 5 ml MAG broth. The culture was grown on a gyratory shaker (200 rpm) at 28 °C for 6 days. Subsequently 1 ml was used to inoculate 50 ml MAG broth and grown on a gyratory shaker (200 rpm) at 28 °C until an OD600nm of 0.6 was reached. DNA was isolated from the cells according to van Berkum [17]. Final concentration of the DNA was set to 0.5 mg ml−1. Culture identity was confirmed by partial sequence analysis of several housekeeping genes and the 16S rRNA gene using the prepared DNA as template for PCR.

Genome sequencing and assembly

The draft genome of B. elkanii USDA 76T was generated at the DOE Joint genome Institute (JGI) using the Illumina technology [37]. An Illumina short-insert paired-end library was constructed with an average insert size of 200 bp that when sequenced generated 312,796,730 reads. An Illumina long-insert paired-end library with an average insert size of 6505.78 +/− 3679.88 bp also was constructed that when sequenced generated 19,315,434 reads. The total amount of sequence data obtained with the Illumina was 34,177 Mbp. Library construction and sequence analysis were done at the JGI according to the protocols outlined on their website [38]. The first of two initial drafts, assembled with Allpaths version r38445 [39], contained 81 contigs in 17 scaffolds and subsequently a consensus was computationally shredded into 10 Kbp overlapping fake reads (shreds). The second draft assembled with Velvet, version 1.1.05 [40], resulted in consensus sequences that were computationally shredded into 1.5 Kbp overlapping fake reads (shreds). The data were assembled again with Velvet using the shreds from the first Velvet assembly to guide the next assembly. The consensus from this second Velvet assembly was shredded into 1.5 Kbp overlapping fake reads. The fake reads from the Allpaths and both Velvet assemblies together with a subset of the Illumina CLIP paired-end reads were assembled using parallel Phrap, version 4.24 (High Performance Software, LLC). Potential errors in the assemblies were corrected by manual editing with Consed [41–43]. Gap closure was accomplished using repeat resolution software (Wei Gu, unpublished) and sequence analysis of bridging PCR fragments with PacBio technology (Cliff Han, unpublished). Gaps were closed and the quality of the final sequence was improved with 35 PCR PacBio consensus sequences. The total size of the genome is 9.5 Mbp and the final assembly is based on 34,177 Mbp of Illumina draft data, which provides an average 3560x coverage of the genome.

Genome annotation

Genes were identified using Prodigal [44] that was followed by a round of manual curation using GenePRIMP [45] as part of the DOE-JGI genome annotation pipeline [46, 47]. The predicted CDSs were translated and used to search the National Center for Biotechnology Information (NCBI) non-redundant, UniProt, TIGRFam, Pfam, KEGG, COG, and InterPro databases. The tRNAScanSE tool [48] was used to find tRNA genes. Ribosomal RNA genes were found by searches against models of the ribosomal RNA genes built from SILVA [49]. Other non–coding RNAs such as the RNA components of the protein secretion complex and the RNase P were identified by searching the genome for the corresponding Rfam profiles using INFERNAL [50]. Additional gene prediction analysis and manual functional annotation were done within the Integrated Microbial Genomes-Expert Review system [51] developed by the Joint Genome Institute, Walnut Creek, CA, USA.

Genome properties

The genome of B. elkanii USDA 76T is 9,484,767 nucleotides long with a GC content of 63.70% (Table 3) and has been assembled into two scaffolds. Of the 9151 genes identified, 9060 are protein encoding and 91 are RNA only encoding genes. Of the 9151 total genes identified in USDA 76T , the majority (73.28%) were assigned a putative function and the remaining genes were annotated as hypothetical. The distribution of genes into COGs functional categories is presented in Table 4.

Table 3 Genome statistics for Bradyrhizobium elkanii USDA 76T
Full size table
Table 4 Number of protein coding genes of Bradyrhizobium elkanii USDA 76T associated with the general COG functional categories
Full size table

Insights from the genome sequence

Scaffold 1.1 of B. elkanii USDA 76T contains a low GC content for the region ~3,000,000–3,800,000 and the presence of symbiotic nod, nif and fix genes in this region indicates a symbiotic island integration (Fig. 3). Using the Phylogenetic Profiler tool within IMG, 356 genes were found to be unique to USDA 76T in a comparison with four other strains (587 [52], CCBAU43297, CCBAU05737 [53] and USDA 94) ascribed to the B. elkanii IMG clique. Of those that were unique, the majority (223 genes, representing 62.6%) were annotated as encoding hypothetical proteins. Out of the remainder, a significant number were phage related. Using the PHASTER algorithm [54], 22 of these genes were found to be co-located genes of an intact resident prophage (Fig. 4). Using this algorithm another incomplete phage gene set on the same scaffold was also identified.

Fig. 3
figure 3

Graphical map of the largest scaffold (9,116,505 bp) of USDA 76T (a) showing the location of common nodulation genes within the symbiotic island of this strain (b). From bottom to the top of the scaffold map: Genes on forward strand (color by COG categories as denoted by the IMG platform), genes on reverse strand (color by COG categories), RNA genes (tRNAs green, sRNAs red, other RNAs black), GC content, GC skew

Full size image
Fig. 4
figure 4

Resident prophage present in Bradyrhizobium elkanii USDA 76T imaged using PHASTER [54]. Prophage maps are not drawn to scale. Reference locus tag for Prophage Region 1 is BraelDRAFT_5594 terminase; ter); reference locus tag for Prophage Region 2 is BraelDRAFT_6751 (terminase; ter). Coat protein (coa), fiber protein (fib), phage-like protein (plp), portal protein (por), tail shaft protein (sha), and terminase (ter). All other genes encode hypothetical proteins

Full size image

Extended insights

Using the Phylogenetic Profiler tool, 7556 genes were found to be conserved in five B. elkanii strains (587, CCBAU43297, CCBAU05737, USDA 76T , USDA 94), including genes encoding a general secretion pathway and type II, III, IV and VI secretion system proteins. The Type III secretion system (T3SS) [55] can either promote or impair the establishment of symbiosis, depending on the legume host [56], and has been characterized as a host determinant for rj1, Rfg1, Rj2 and Rj4 soybean cultivars [33, 57, 58]. The dominant soybean genes Rj2 and Rj4 restrict nodulation with specific strains of Bradyrhizobium [33]. Most investigations of soybean host genes controlling the symbiosis have focused on the Rj4 soybean line that was originally identified by its inability to nodulate with USDA 61 ( B. elkanii , serogroup 31) [59]. The predicted Rj4 thaumatin-like protein is thought to be involved in conferring resistance to Bradyrhizobium strains producing specific T3SS effector proteins [60]. However, USDA 76T was reported to nodulate and form an effective nitrogen-fixing symbiosis with the isogenic lines BARC-2 (Rj4) and BARC-3 (rj4) [30, 61], suggesting that this strain does not produce the interacting T3SS effector protein(s). Conversely, the recessive soybean gene rj1rj1 [62], encoding a putative truncated Nod factor receptor protein [63], restricts nodulation by many Bradyrhizobium and Ensifer strains, although specific strains of B. elkanii , including USDA 76T , can form a limited number of nodules when tested with plants in Leonard jars containing sterilized vermiculite or sand [30, 59, 61].

USDA 76T genes encoding components required for a functional T3SS were identified within the integrated symbiotic island (Figs. 5 and 6). Although the nopA and nopC genes were not annotated in the USDA 76T genome, by using TBLASTN these genes were identified in the intergenic region between BraelDRAFT_3047 (sctD) and BraelDRAFT_3048 (hypothetical) that share 100% sequence similarity with nopA and nopC of the characterized Bradyrhizobium elkanii strain USDA 61 [57]. Although T3SS components can also be found in Bradyrhizobium strain USDA 110, this strain lacks the nopX gene encoding the translocon required to introduce effector molecules into host cells [56, 64]. This is in contrast to the presence of nopX in USDA 76T , which could extend its host range to otherwise incompatible hosts.

Fig. 5
figure 5

Comparison of the gene neighbourhood regions containing loci that encode type III secretion system components in the genomes of Ensifer fredii NGR234 and the Bradyrhizobium strains USDA 76T and USDA 110. The colour scheme is as follows: green, structural component; orange, pilus component; purple, regulatory component; red, translocon component; uncoloured, other genes; and yellow, effector component

Full size image
Fig. 6
figure 6

Schematic representation of the components constituting the T3SS present in Bradyrhizobium elkanii USDA 76T. The IMG product name is provided with the Yersinia Ysc-Yop T3SS ortholog shown in brackets. The relative secretion components were identified based on information provided by Galán et al. [55]

Full size image

Conclusions

B. elkanii USDA 76T originated from strain USDA 8, which was obtained in 1915 from an effective nodule of soybean grown on the USDA Arlington farm in Virginia. Its ability to nodulate the native North American legumes Apios americana Medik. and Amphicarpaea bracteata (L.) Fernald indicates a possible North American origin for this strain. USDA 76T was selected for genome sequencing [14] because of its significance as a microsymbiont of soybean. The genome size of USDA 76T was established as 9.5 Mbp, which falls within the range of 7.7 to 10.5 Mbp observed for other bradyrhizobial genomes. The genome of this N2-fixing microsymbiont contains nod, nif and fix genes located on an integrated symbiotic island, and genes encoding both an intact and an incomplete phage. According to ANI values, strain USDA 76T formed an ANI clique with four other B. elkanii soybean strains: USDA 94, 587, CCBAU 43297 and CCBAU 05737. Of particular interest was the discovery that these strains contain a T3SS that contains the NopCA pilus genes and the NopX translocon protein, which are essential for introducing effector molecules into host cells [55]. The T3SS has been shown to be an important host range determinant that enables the nodulation of some soybean cultivars and is detrimental to symbiosis with other cultivars [56]. Here we postulate that the presence of a functional T3SS is important in determining the host range of USDA 76T and enables it to form some nodules on the soybean cultivar Clark (rj1) when grown in Leonard jars with sterilized vermiculite or sand [65, 66]. Further analyses of Bradyrhizobium genomes, including that of USDA 76T , will increase our understanding of determinants that lead to the establishment and functioning of different Bradyrhizobium symbioses.