Research | Open | Published:
Genetic snapshots of the Rhizobiumspecies NGR234 genome
Genome Biologyvolume 1, Article number: research0014.1 (2000)
In nitrate-poor soils, many leguminous plants form nitrogen-fixing symbioses with members of the bacterial family Rhizobiaceae. We selected Rhizobium sp. NGR234 for its exceptionally broad host range, which includes more than 112 genera of legumes. Unlike the genome of Bradyrhizobium japonicum, which is composed of a single 8.7 Mb chromosome, that of NGR234 is partitioned into three replicons: a chromosome of about 3.5 Mb, a megaplasmid of more than 2 Mb (pNGR234b) and pNGR234a, a 536,165 bp plasmid that carries most of the genes required for symbioses with legumes. Symbiotic loci represent only a small portion of all the genes coded by rhizobial genomes, however. To rapidly characterize the two largest replicons of NGR234, the genome of strain ANU265 (a derivative strain cured of pNGR234a) was analyzed by shotgun sequencing.
Homology searches of public databases with 2,275 random sequences of strain ANU265 resulted in the identification of 1,130 putative protein-coding sequences, of which 922 (41%) could be classified into functional groups. In contrast to the 18% of insertion-like sequences (ISs) found on the symbiotic plasmid pNGR234a, only 2.2% of the shotgun sequences represent known ISs, suggesting that pNGR234a is enriched in such elements. Hybridization data also indicate that the density of known transposable elements is higher in pNGR234b (the megaplasmid) than on the chromosome. Rhizobium-specific intergenic mosaic elements (RIMEs) were found in 35 shotgun sequences, 6 of which carry RIME2 repeats previously thought to be present only in Rhizobium meliloti. As non-overlapping shotgun sequences together represent approximately 10% of ANU265 genome, the chromosome and megaplasmid may carry a total of over 200 RIMEs.
'Skimming' the genome of Rhizobium sp. NGR234 sheds new light on the fine structure and evolution of its replicons, as well as on the integration of symbiotic functions in the genome of a soil bacterium. Although most putative coding sequences could be distributed into functional classes similar to those in Bacillus subtilis, functions related to transposable elements were more abundant in NGR234. In contrast to ISs that accumulated in pNGR234a and pNGR234b, the hundreds of RIME elements seem mostly attributes of the chromosome.
Many different Gram-negative bacteria colonize the nutrient-rich rhizospheres of plant roots. Some bacteria are pathogenic, whereas others form beneficial associations. In nitrate-poor soils, strains of Azorhizobium, Bradyrhizobium, Mesorhizobium and Rhizobium (collectively known as rhizobia), form nitrogen-fixing symbioses with leguminous plants. In compatible interactions, invading rhizobia penetrate their hosts through infection threads, which develop centripetally. At the same time, new structures called nodules develop from meristems induced in the cortex of infected roots. When infection threads reach nodule cells, rhizobia are released as symbiosomes into the cytoplasm of infected cells where they eventually enlarge and differentiate into nitrogen-fixing bacteroids. Continuous exchange of chemical signals between the two symbionts coordinates expression of bacterial and plant genes required for a symbiotic development. Flavonoids released by legume roots are amongst the first signals exchanged in this molecular dialog. By interacting with rhizobial regulators of the NodD family, flavonoids trigger the expression of nodulation genes (nod, noe and nol). In turn, most nodulation genes participate in the synthesis and secretion of a family of lipochito-oligosaccharide molecules, the Nod factors that are required for bacterial entry into root hairs. Little is known about how rhizobia migrate inside the infection threads, although it seems likely that genetic determinants of both partners are again involved (see [1,2]). Once within the cortex, the rhizobia differentiate into bacteroids where low free-oxygen tensions help coordinate the expression of genes involved in nitrogen fixation (nif and fix) .
Taxonomic proposals based on DNA sequences of highly conserved genes indicate that rhizobia are a group of genetically diverse soil bacteria . Other data suggest that in populations of soil bacteria, natural genetic mechanisms exist which can transform isolates with widely different chromosomal backgrounds into nodulating bacteria (that is, rhizobia) (for review see ). Comparisons of genomes of soil bacteria will help define the pools of symbiotic genes. Unfortunately, genomic studies of this kind have been hindered by the relatively large size of rhizobial genomes (6.5 to 8.7 Mb for R. meliloti and B. japonicum, respectively). Instead, as many symbiotic loci are often clustered on large plasmids in Rhizobium strains, or in chromosomal 'symbiotic islands' as in B. japonicum  and M. loti , physical and genetic analyzes of symbiotic plasmids or 'islands' prevailed. Rhizobium sp. NGR234 was selected for its exceptionally broad host range, which includes more than 112 genera of legumes in addition to the non-legume Parasponia andersonii [7,8]. As in R. meliloti, the genome of NGR234 is partitioned into three replicons, a chromosome of about 3.5 Mb, a megaplasmid of more than 2 Mb (pNGR234b) and pNGR234a, a 536 kb symbiotic plasmid [9,10,11]. Although various experiments have shown that most symbiotic genes are amongst the 416 open reading frames (ORFs) identified in the complete sequence of pNGR234a [9,12,13], others are carried by the chromosome and/or the mega-plasmid [10,14].
Many ways of finding genes exist, but with the rapid advances in genomics, among the most effective are those that involve sequencing parts of or entire genomes. Although contiguous sequences of several symbiotic islands/plasmids will be released in the near future, R. meliloti strain 1021 as well as the phytopathogens Ralstonia solanacearum and Xanthomonas citri are the only plant-interacting microbes currently being sequenced [15,16,17]. The cost of sequencing a complete genome is still well beyond the capability of most laboratories, however. Nevertheless, extensive information on the structure and content of genomes can be gained by randomly sequencing libraries made from total DNA [18,19,20,21]. Here, we have used this approach to analyze the megaplasmid and chromosome of NGR234. A total of 2,275 individual shotgun sequences of ANU265 (a derivative strain of NGR234 cured of its symbiotic plasmid ) were searched for protein and/or DNA homologies, and putative coding sequences were grouped into 28 classes according to their putative function. In addition, clones carrying various Rhizobium-specific repeated elements such as RIME1 and RIME2 were also analyzed.
Results and discussion
Random sequencing of the ANU265 genome
Total genomic DNA of ANU265 was used to construct an M13 library with inserts ranging in size from 0.9 to 1.5 kb. Of the 2,856 random clones analyzed, 80% (2,275) produced high-quality DNA sequence with an average read length of 253 bp (Table 1). In this way, more than 575 kb of total nucleotide sequence was collected, which corresponds to approximately 10% of the ANU265 genome . At 61.2 mol%, the mean G+C content of these sequences is similar to that found for the entire genome , but is also significantly higher than the value of 58.5 mol% calculated for pNGR234a . This pool of 2,275 sequences was then screened for redundancy. A total of 381 overlapping sequences were identified, and grouped into 195 contigs (sets of overlapping sequences) of two to four elements each: 154 contigs represent pairs of clones, whereas the remaining 73 sequences belong to 23 groups of three elements and one of four clones. Because of the many highly conserved sequences repeated throughout the NGR234 genome [9,11,24], it was not possible to determine if overlapping clones represent contiguous sequences or DNA fragments from distinct repeats. Nevertheless, truly unique sequences represent 92% of the total number of clones. With an average insert size of 1.2 kb, clones tagged with non-overlapping sequences represent more than 40% (2.5 Mbp) of the ANU265 genome.
RIME- and IS-like sequences
Homology searches against nucleotide databases (BLASTN ) showed that 35 ANU265 sequences carried Rhizobium-specific intergenic mosaic elements (RIMEs). First identified in R. meliloti, R. leguminosarum bv. viciae and NGR234, RIME1 elements are 108 bp repeats characterized by two large palindromes, whereas RIME2 sequences are 109 bp repeats thought to be present only in R. meliloti . RIMEs have many features of the short interspersed repeated elements that are non-coding, intercistronic sequences of less than 200 bp found in many prokaryotic genomes . Of the 2,275 shotgun sequences of ANU265 collected, 29 contained RIME1 elements and 6 carried RIME2 repeats. Although Southern hybridizations indicated that approximately 20 copies of RIME1 were present in the genomes of. R. meliloti and NGR234 , our data indicate that there are many more. Among the 29 clones with RIME1 sequences, most (23) carry repeats that are very similar to the consensus ( and Figure 1). In another six (Figure 1, clones 27d06, 29g08, 0lf01, 11b07, 25e07 and 13c06), only one of the two large palindromic structures is conserved, however. This suggests that, in some cases, individual palindromes constitute independent repeats, not necessarily associated to form RIME1 elements. In the eight clones that code for putative proteins (Figure 1), RIME1 sequences are found immediately downstream of predicted ORFs (data not shown), indicating that these elements are probably confined to intergenic regions. Surprisingly, no RIME2 and a single RIME1 repeat were found on pNGR234a [9,11]. If these elements were regularly distributed throughout the NGR234 genome, more than a single RIME1 would have been expected on the 536 kb of pNGR234a. Thus, current data suggest that RIMEs preferentially accumulate on specific replicons, and that NGR234 carries possibly as many as 200 RIME-like elements.
In contrast to pNGR234a, which carries many IS sequences, only 2.2% (51) of the 2,275 ANU265 sequences were predicted to encode transposon-related functions. Although several clones that did not match database homologs may also carry sequences of yet uncharacterized IS elements, these results suggest that in proportion to their size, chromosome and megaplasmid carry fewer transposable elements than pNGR234a. Nevertheless most of the 51 clones (70%) matched ISs that were first identified in pNGR234a . For example, ten sequences highly homologous to NGRIS-4 were found. This 3,316 bp element is duplicated in pNGR234a , whereas chromosome and megaplasmid carry two and five copies of NGRIS-4 respectively [11,24].
Identification of putative genes
To assign putative functions to the cloned DNA fragments, sequences were compared to protein and nucleotide databases [25,28]. BLAST analyses showed that about 50% (1,130) of the 2,275 sequences matched protein-coding ORFs, three were homologous to rDNA and four to tRNA loci (see Table 1). Of the 1,130 putative protein-coding sequences, 208 (or 9% of the 2,275 sequences) were similar to hypothetical genes with no known function (pioneer sequences) of rhizobia and other organisms. Thus, together with the 1,109 clones which showed no significant similarity to entries in nucleotide and amino-acid databases (see Table 1), functions could not be assigned to 58% of the shotgun sequences. To provide an overview of the genetic organization of the ANU265 genome, predicted protein-coding sequences were grouped into various classes according to their putative function (Table 2).
A genetic snapshot of the ANU265 genome
In total, 922 of the 2,275 sequences were grouped into 28 functional categories (Table 2). Interestingly, comparison of this data with that derived from the complete sequence of the Bacillus subtilis genome  showed a similar distribution of genes in both organisms. Although B. subtilis is a Gram-positive bacterium, it is commonly found in soil, water sources and in associations with plants. Thus, with the exception of one homolog of a sporulation gene (which was not expected in rhizobia), the comparative analysis presented in Table 2 suggests that the number of shotgun sequences is probably sufficiently large to form a representative selection of ANU265 loci. All 1,130 sequences for which significant matches were found in database searches are classified by function in Table 3.
As in other bacterial genomes, such as that of Escherichia coli , the largest functional class represents transport and binding proteins (see Tables 2 and 3). A number of essential genes, including those required for replication, transcription and translation as well as those linked to primary metabolism, were also found. As expected of a soil-borne prokaryote, many loci (18%) involved in carbon and nitrogen metabolism were identified (encoding enzymes for the assimilation of nitrate/ammonia, the tri-carboxylic acid cycle, or transporters of dicarboxylic acids, and so on). In B. subtilis, 19% of the protein-coding genes are devoted to the metabolism of carbohydrates, amino acids and related molecules (Table 2). This is in contrast to microorganisms such as Haemophilus influenzae and M. genitalium that are not able to grow on many nitrogen and carbon sources (only 10% of their predicted genes code for such metabolic functions ). Interestingly, homologs of various chaperones such as GroES/GroEL, DnaJ, and other small heat-shock proteins (sHsps), were identified (Table 3, clones 308 to 318). The presence of multiple sHsps is not common in prokaryotes, but was shown to be widespread in rhizobia .
Obviously, the ability of rhizobia to respond to plant compounds that stimulate their growth contributes to successful colonization of the root  and absence of vitamins often limits the growth or rhizobia. Furthermore, the ability to either take up or synthesize vitamins is thought to be an essential characteristic of rhizobia . For these reasons, it is not surprizing that several ANU265 sequences matched genes for biotin and thiamine utilization, such as that coding for a homolog of bioS (clone 745), a biotin-regulated locus of R. meliloti . In R. meliloti, bioS is part of an operon which includes the surE and IppB/nlpD genes that are also found in ANU265 (clones 744 and 183). Homologs of thiamine biosynthetic genes thiCG of R. etli (clones 512 and 513) were also found. Miranda-Rios et al.  reported a direct correlation between the expression of thiC and the production of the symbiotic terminal oxidase cbb3, which is required for bacteroid respiration under conditions of low oxygen.
Putative symbiotic genes include loci involved in exopolysaccharide (EPS) biosynthesis and/or export, which are encoded by pNGR234b , as well as genes involved in the elaboration of acidic capsular polysaccharides (K-anti-gens), lipopolysaccharides and cyclic ß-glucans (Table 3, clones 245 to 270). A sequence homologous to fixN of R. meliloti was also identified (clones 208 and 209). The chromosomal fixNOPQ locus encodes an oxidase complex that is probably active during nitrogen fixation. Although sequences of the regulatory fixK genes  were identified (clone 683), no significant match to the oxygen-responsive system encoded by fixLJ was found. Members of other symbiotic two-component regulatory systems were detected in ANU265, however, including homologs of the sensor histi-dine kinase exoS (clone 200) and the response regulator chvI (clone 717). Both are necessary for regulating production of succinoglycans that are important in R. meliloti-Medicago sativa symbioses . Similarly, the nwsA locus (clone 202) encodes a putative sensor kinase that is involved in the expression of nodulation genes in Bradyrhizobium strains .
It has been postulated that genes responsible for the synthesis (mos) and catabolism (moc) of rhizopines confer a competitive advantage on their host rhizobia . Rhizopines are synthesized in nodules of M. sativa inoculated with R. meliloti strain L5-30, and can be used as growth substrates by certain rhizobia. Although mos and moc genes were thought to be limited to R. meliloti strains , homologs of mocABC, and mosA genes were also found in ANU265 (clones 543 to 549). Propagation of rhizobia in the soil, and hence their symbiotic efficiency, probably also depends on their tolerance to osmotic changes. It is thus notable that homologs of the R. meliloti betABC genes, which are involved in the osmoregulatory choline-glycine betaine pathway , were also found (clones 726 to 730).
Other putative symbiotic loci include homologs of the phbC and prsDE genes of R. meliloti, which encode a poly-3-hydroxybutyrate synthase  and a type I secretion system  (clones 741 to 743, and 298 to 301, respectively). Interestingly, PrsD and PrsE of R. meliloti are involved in the secretion of enzymes that modify succinoglycans , whereas a similar type I secretion system seems to be responsible for the export of the nodulation-signaling protein NodO in R. leguminosarum bv. viciae [44,45]. Although the role of these prsDE homologs in NGR234 is not clear, it is possible that more than one type of protein secretion system has a symbiotic role in this bacterium .
Random sequencing of ANU265 followed by homology searches of public databases resulted in the identification of 1,130 putative protein-coding sequences, of which 922 (41%) could be classified into functional groups. Comparison of these data with those derived from the complete sequence of the B. subtilis genome showed a similar distribution of putative coding sequences, except perhaps for functions related to transposable elements (Table 2). In fact, the genome of ANU265 carries more putative transposases and other IS-related functions (5.5% of all identified genes, and 2.2% of all shotgun sequences) than that of B. subtilis. Nevertheless, in proportion to their size, the chromosome and megaplasmid of NGR234 carry fewer IS sequences than pNGR234a. Furthermore, hybridization data indicate that the density of known transposable elements is higher in pNGR234b than on the chromosome (order of IS accretion is: pNGR234a > pNGR234b > chromosome) . This suggests that IS elements preferentially accumulate on plasmids, possibly because they are less likely to disrupt essential functions. In contrast, the many RIME elements present in NGR234 are clearly more abundant on the chromosome and megaplasmid than on pNGR234a. Together, the distinct G+C contents and structural features of the symbiotic plasmid, megaplasmid and chromosome suggest that different evolutionary constraints and histories contributed to shape these three replicons.
'Skimming' the genome of Rhizobium sp. NGR234 has given new insights into the evolution of its replicons and the integration of symbiotic functions in the genome of a soil bacterium. It also reinforced the assumption, which originated from host-range extension experiments [12,47], that pNGR234a carries most of the symbiotic genes. Although few nod, nif and fix homologs were found amongst the random clones, it is likely that additional chromosome- and megaplasmid-encoded functions contribute to successful symbioses between NGR234 and its many host plants. In this respect, transcriptional analyses using shotgun sequences as hybridization templates  will help identify such new symbiotic loci.
Materials and methods
Rhizobium strain ANU265 , a strain of Rhizobium sp. NGR234  cured of pNGR234a, was grown in Rhizobium minimal medium supplemented with succinate (RMM) . Escherichia coli was grown on SOC or in TY . Subclones in M13mp18 vectors  were grown in E. coli strain DH5aF'IQ .
Preparation of the random genomic library and M13 templates
Genomic DNA of Rhizobium strain ANU265 was prepared as in Perret and Broughton . ANU265 genomic DNA (15 µg) was sheared by sonication and incubated for 10 min at 30°C with 30 units of mung bean nuclease. The resulting digest was extracted with phenol/chloroform (1:1) and precipitated with ethanol. Fragments ranging in size from 900 to 1,500 bp were purified from agarose gels and ligated into SmaI-digested M13mp18 vector DNA. Ligation mixtures were electroporated into E. coli strain DH5aF'IQ [48,52], and transformants were plated on 5-bromo-4-chloro-indoyl-ß-D-galactoside (X-Gal) and isopropyl-ß-thiogalactopyranoside (IPTG)-containing petri dishes . Fresh 1 ml cultures of E. coli DH5aF'IQ were infected with phages from randomly selected white plaques, and grown for 6 h at 37°C in TY medium. Phages were precipitated from 600 µl of the culture supernatant by adding 150 µl 2.5 M NaCl/20% polyethylene glycol (PEG-8,000) (20 min at 25°C). Afterwards, they were centrifuged for 20 min at 3,000g at 25°C, and resuspended in 20 µl Triton-TE extraction buffer (0.5% Triton X-100; 10 mM Tris-HCl, 1 mM EDTA pH 8.0). Following 10 min incubation at 80°C and ethanol precipitation, single-stranded phage DNA was recovered in 50 µl H2O.
Dye-terminator cycle sequencing of individual M13 sub-clones, gel electrophoresis and sequence editing was performed as described by Freiberg et al. . Shotgun sequences were checked for redundancy using the XGAP program  and for significant homologies with BLASTX-BLASTN software  using nonredundant databases at NCBI .
Broughton WJ, Perret X: Genealogy of legume-Rhizobium symbioses. Curr Opin Plant Biol. 1999, 2: 305-311. 10.1016/S1369-5266(99)80054-5.
Perret X, Staehelin C, Broughton WJ: Molecular basis of symbiotic promiscuity. Microbiol Mol Biol Rev. 2000, 64: 180-201. 10.1128/MMBR.64.1.180-201.2000.
Fischer HM: Genetic regulation of nitrogen fixation in rhizobia. Microbiol Rev. 1994, 58: 352-386.
Martínez-Romero E, Caballero-Mellado J: Rhizobium phylogenies and bacterial genetic diversity. Critical Rev Plant Sci. 1996, 15: 113-140.
Kündig C, Hennecke H, Göttfert M: Correlated physical and genetic map of the Bradyrhizobium japonicum 110 genome. J Bacteriol. 1993, 175: 613-622.
Sullivan JT, Ronson CW: Evolution of rhizobia by acquisition of a 500-kb symbiosis island that integrates into a phe -tRNA gene. Proc Natl Acad Sci USA. 1998, 95: 5145-5149. 10.1073/pnas.95.9.5145.
Trinick MJ: Relationships amongst the fast-growing rhizobia of Lablab purpureus, Leucaena leucocephala, Mimosa spp., Acacia farnesiana, and Sesbania grandiflora and their affinities with other rhizobial groups. J Appl Bacteriol. 1980, 49: 39-53.
Pueppke SG, Broughton WJ: Rhizobium sp. strain NGR234 and R. fredii USDA257 share exceptionally broad, nested host ranges. Mol Plant-Microbe Interact. 1999, 12: 293-318.
Freiberg C, Fellay R, Bairoch A, Broughton WJ, Rosenthal A, Perret X: Molecular basis of symbiosis between Rhizobium and legumes. Nature. 1997, 387: 394-401. 10.1038/387394a0.
Flores M, Mavingui P, Girard L, Perret X, Broughton WJ, Martinez-Romero E, Davila G, Palacios R: Three replicons of Rhizobium sp. strain NGR234 harbor symbiotic gene sequences. J Bacteriol. 1998, 180: 6052-6053.
Perret X, Viprey V, Broughton WJ: Physical and genetic analysis of the broad host-range Rhizobium sp. NGR234. In Prokaryotic Nitrogen Fixation. Edited by Triplett EW. Wymondham: Horizon Scientific Press,. 2000, 679-692.
Broughton WJ, Heycke N, Meyer zAH, Pankhurst CE: Plasmid linked nif and nod genes in fast-growing rhizobia that nodulate Glycine max, Psophocarpus tetragonolobus, and Vigna unguiculata. Proc Natl Acad Sci USA. 1984, 81: 3093-3097.
Perret X, Freiberg C, Rosenthal A, Broughton WJ, Fellay R: High-resolution transcriptional analysis of the symbiotic plasmid of Rhizobium sp. NGR234. Mol Microbiol. 1999, 32: 415-425. 10.1046/j.1365-2958.1999.01361.x.
Perret X, Broughton WJ, Brenner S: Canonical ordered cosmid library of the symbiotic plasmid of Rhizobium species NGR234. Proc Natl Acad Sci USA. 1991, 88: 1923-1927.
Ralstonia solanacearum - Complete genome sequencing. [http://www.genoscope.cns.fr/externe/English/Projets/Projet_Y/Y.html]
Sinorhizobium meliloti strain 1021 sequencing projects. [http://sequence.toulouse.inra.fr]
Xanthomonas citri Genome Project. [http://genoma4.iq.usp.br/xanthomonas]
Brenner S, Elgar G, Sandford R, Macrae A, Venkatesh B, Aparicio S: Characterization of the pufferfish (Fugu) genome as a compact model vertebrate genome. Nature. 1993, 366: 265-268. 10.1038/366265a0.
Avalos J, Corrochano LM, Brenner S: Genomic organization of the fungus Phycomyces. Gene. 1996, 174: 43-50. 10.1016/0378-1119(96)00304-6.
Kim CW, Markiewicz P, Lee JJ, Schierle CF, Miller JH: Studies of the hyperthermophile Thermotoga maritima by random sequencing of cDNA and genomic libraries. Identification and sequencing of trpEG(D) operon. J Mol Biol. 1993, 231: 960-981. 10.1006/jmbi.1993.1345.
Peterson SN, Schramm N, Hu PC, Bott KF, Hutchison CA: A random sequencing approach for placing markers on the physical map of Mycoplasma genitalium. Nucleic Acids Res. 1991, 19: 6027-6031.
Morrison NA, Hau CY, Trinick MJ, Shine J, Rolfe BG: Heat curing of a Sym plasmid in a fast-growing Rhizobium sp. that is able to nodulate legumes and the nonlegume Parasponia sp. J Bacteriol. 1983, 153: 527-531.
Broughton WJ, Dilworth MJ, Passmore IK: Base ratio determination using unpurified DNA. Anal Biochem. 1972, 46: 164-172.
Perret X, Viprey V, Freiberg C, Broughton WJ: Structure and evolution of NGRRS-1, a complex, repeated element in the genome of Rhizobium sp. strain NGR234. J Bacteriol. 1997, 179: 7488-7496.
BlastX and BlastN analyses on non-redundant nucleic and aminoacids databases. [http://www.ncbi.nlm.nih.gov/BLAST/]
Østeras M, Stanley J, Finan TM: Identification of Rhizobium -specific intergenic mosaic elements within an essential two-component regulatory system of Rhizobium species. J Bacteriol. 1995, 177: 5485-5494.
Lupski JR, Weinstock GM: Short, interspersed repetitive DNA sequences in prokaryotic genomes. J Bacteriol. 1992, 174: 4525-4529.
Swissprot protein database. [http://www.expasy.ch]
Kunst F, Ogasawara N, Moszer I, Albertini AM, Alloni G, Azevedo V, Bertero MG, Bessieres P, Bolotin A, Borchert S, et al: The complete genome sequence of the Gram-positive bacterium Bacillus subtilis. Nature. 1997, 390: 249-256. 10.1038/36786.
Blattner FR, Plunkett G, Bloch CA, Perna NT, Burland V, Riley M, Collado-Vides J, Glasner JD, Rode CK, Mayhew GF, et al: The complete genome sequence of Escherichia coli K-12. Science. 1997, 277: 1453-1474. 10.1126/science.277.5331.1453.
Fraser CM, Gocayne JD, White O, Adams MD, Clayton RA, Fleischmann RD, Bult CJ, Kerlavage AR, Sutton G, Kelley JM, et al: The minimal gene complement of Mycoplasma genitalium. Science. 1995, 270: 397-403.
Münchbach M, Nocker A, Narberhaus F: Multiple small heat shock proteins in rhizobia. J Bacteriol. 1999, 181: 83-90.
Streit WR, Joseph CM, Phillips DA: Biotin and other water-soluble vitamins are key growth factors for alfalfa root colonization by Rhizobium meliloti 1021. Mol Plant-Microbe Interact. 1996, 9: 330-338.
Streit WR, Phillips DA: A biotin-regulated locus bioS, in a possible survival operon of Rhizobium meliloti. Mol Plant-Microbe Interact. 1997, 10: 933-937.
Miranda-Rios J, Morera C, Taboada H, Davalos A, Encarnacion S, Mora J, Soberon M: Expression of thiamin biosynthetic genes (thiCOGE) and production of symbiotic terminal oxidase cbb3 in Rhizobium etli. J Bacteriol. 1997, 179: 6887-6893.
Cheng HP, Walker GC: Succinoglycan production by Rhizobium meliloti is regulated through the ExoS-ChvI two-component regulatory system. J Bacteriol. 1998, 180: 20-26.
Grob P, Michel P, Hennecke H, Göttfert M: A novel response-regulator is able to suppress the nodulation defect of a Bradyrhizobium japonicum nodW mutant. Mol Gen Genet. 1993, 241: 531-541.
Murphy PJ, Saint CP: Rhizopines in the legume-Rhizobium symbiosis. In Molecular Signals in Plant-Microbe Communications. Edited by Verma DPS. Boca Raton, Florida: CRC Press,. 1992, 377-390.
Rossbach S, Rasul G, Schneider M, Eardly B, de Bruijn FJ: Structural and functional conservation of the rhizopine catabolism (moc) locus is limited to selected Rhizobium meliloti strains and unrelated to their geographical origin. Mol Plant-Microbe Interact. 1995, 8: 549-559.
Pocard JA, Vincent N, Boncompagnie E, Smith LT, Poggi MC, Le Rudulier D: Molecular characterization of the bet genes encoding glycine betaine synthesis in Sinorhizobium meliloti 102F34. Microbiology. 1997, 143: 1369-1379.
Willis LB, Walker GC: The phbC (poly-beta-hydroxybutyrate synthase) gene of Rhizobium (Sinorhizobium) meliloti and characterization of phbC mutants. Can J Microbiol. 1998, 44: 554-564. 10.1139/cjm-44-6-554.
York GM, Walker GC: The Rhizobium meliloti exoK gene and prsD/prsE/exsH genes are components of independent degradative pathways which contribute to production of low-molecular-weight succinoglycan. Mol Microbiol. 1997, 25: 117-134. 10.1046/j.1365-2958.1997.4481804.x.
Finnie C, Zorreguieta A, Hartley NM, Downie JA: Characterization of Rhizobium leguminosarum exopolysaccharide glycanases that are secreted via a type I exporter and have a novel heptapeptide repeat motif. J Bacteriol. 1998, 180: 1691-1699.
Sutton JM, Peart J, Dean G, Downie JA: Analysis of the C-terminal secretion signal of the Rhizobium leguminosarum nodulation protein NodO; a potential system for the secretion of heterologous proteins during nodule invasion. Mol Plant-Microbe Interact. 1996, 9: 671-680.
Finnie C, Hartley NM, Findlay KC, A. DJ: The Rhizobium leguminosarum prsDE genes are required for secretion of several proteins, some of which influence nodulation, symbiotic nitrogen fixation and exopolysaccharide modification. Mol Microbiol. 1997, 25: 135-146. 10.1046/j.1365-2958.1997.4471803.x.
Viprey V, Del Greco A, Golinowski W, Broughton WJ, Perret X: Symbiotic implications of type III protein secretion machinery in Rhizobium. Mol Microbiol. 1998, 28: 1381-1389. 10.1046/j.1365-2958.1998.00920.x.
Broughton WJ, Wong CH, Lewin A, Samrey U, Myint H, Meyer zAH, Dowling DN, Simon R: Identification of Rhizobium plasmid sequences involved in recognition of Psophocarpus, Vigna, and other legumes. J Cell Biol. 1986, 102: 1173-1182. 10.1083/jcb.102.4.1173.
Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning: A Laboratory Manual, second edn. Cold Spring Harbor, NY: Cold Spring Harbor University Press,. 1989
Yanisch-Perron C, Ira J, Messing J: Improved M13 phage cloning vectors and host strains: Nucleotide sequences of M13mp18 and pUC19 vectors. Gene. 1985, 33: 103-119. 10.1016/0378-1119(85)90120-9.
Hanahan D: Studies on transformation of Escherichia coli with plasmids. J Mol Biol. 1983, 166: 557-580.
Perret X, Broughton WJ: Rapid identification of Rhizobium strains by targeted PCR fingerprinting. Plant Soil. 1998, 204: 21-34. 10.1023/A:1004370725605.
Dower WJ, Miller JF, Ragsdale CW: High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 1988, 16: 6127-6145.
Freiberg C, Perret X, Broughton WJ, Rosenthal A: Sequencing the 500-kb GC-rich symbiotic replicon of Rhizobium sp. NGR234 using dye terminators and a thermostable "sequenase": a beginning. Genome Res. 1996, 6: 590-600.
Dear S, Staden R: A sequence assembly and editing program for efficient management of large projects. Nucleic Acids Res. 1991, 19: 3907-3911.
Altschul SF, Gish W, Miller W, Myers EW, Lipman D: Basic local alignment search tool. J Mol Biol. 1990, 215: 403-410. 10.1006/jmbi.1990.9999.
We thank S. Brenner, C. Freiberg, S. Taudien and D. Gerber for their help with many aspects of this work. Financial assistance was provided by the Fonds National Suisse de la Recherche Scientifique (Grant No. 31-45921.95) and the Université de Genève.