International Journal of Systematic and Evolutionary Microbiology (2013), 63, 2002–2007 DOI 10.1099/ijs.0.044362-0

Mesorhizobium qingshengii sp. nov., isolated from effective nodules of Astragalus sinicus

Wen Tao Zheng,1 Ying Li, Jr,1 Rui Wang,1 Xin Hua Sui,1 Xiao Xia Zhang,2 Jun Jie Zhang,1 En Tao Wang1,3 and Wen Xin Chen1

Correspondence 1State Key Laboratory for Agro-Biotechnology, College of Biological Sciences, China Agricultural Xin Hua Sui University, Beijing, 100193, PR China [email protected] 2Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, PR China 3Departamento de Microbiologı´a, Escuela Nacional de Ciencias Biolo´gicas, Instituto Polite´cnico Nacional, 11340 Me´xico DF, Mexico

In a study on the diversity of isolated from root nodules of Astragalus sinicus, five strains showed identical 16S rRNA gene sequences. They were related most closely to the type strains of loti, Mesorhizobium shangrilense, and Mesorhizobium australicum, with sequence similarities of 99.6–99.8 %. A polyphasic approach, including 16S– 23S intergenic spacer (IGS) RFLP, comparative sequence analysis of 16S rRNA, atpD, glnII and recA genes, DNA–DNA hybridization and phenotypic tests, clustered the five isolates into a coherent group distinct from all recognized Mesorhizobium species. Except for strain CCBAU 33446, from which no symbiotic gene was detected, the four remaining strains shared identical nifH and nodC gene sequences and nodulated with Astragalus sinicus. In addition, these five strains showed similar but different fingerprints in IGS-RFLP and BOX-repeat-based PCR, indicating that they were not clones of the same strain. They were also distinguished from recognized Mesorhizobium species by several phenotypic features and fatty acid profiles. Based upon all the results, we suggest that the five strains represent a novel species for which the name Mesorhizobium qingshengii sp. nov. is proposed. The type strain is CCBAU 33460T (5CGMCC 1.12097T5LMG 26793T5HAMBI 3277T). The DNA G+C content of the type strain is

59.52 mol% (Tm).

Astragalus sinicus Linn. (Chinese milk vetch), an herb- in turn improves the quality and flavour of rice (Lin & Gu, aceous legume originated in China, has been widely 1998). Although only Mesorhizobium huakuii has been cultured as a traditional green manure in wintry fallow described from the rhizobia of A. sinicus (Chen et al., 1991; paddy fields in southern China and Japan. It is also used as Murooka et al., 1993; Jarvis et al., 1997; Nuswantara et al., a soil fertilizer and in repairing the soil environment, which 1999), great diversity in the chromosomal genes has been revealed among the microsymbionts of this plant. In addition, the nodulation (nod) genes in the rhizobia of A. The GenBank/EMBL/DDBJ accession numbers reported herein are sinicus were found to be conserved (Murooka et al., 1993; as follows: JQ339788, JQ339776, JQ339785, JQ339775 and Guo et al., 1999; Zhang et al., 2000). JQ339793 for the 16S rRNA gene; JQ339818, JQ339806, JQ339815, JQ339805 and JQ339824 for the partial atpD gene; During a study of rhizobia nodulating A. sinicus in south-east JQ339851, JQ339839, JQ339848, JQ339838 and JQ339856 for the China (encompassing four provinces), 232 bacterial isolates partial glnII gene; and JQ339757, JQ339745, JQ339754, JQ339744 were isolated from root nodules collected in the fields by and JQ339762 for the partial recA gene for strains CCBAU 33460T, CCBAU 33430,CCBAU 33431,CCBAU 33446 and CCBAU 33455, standard methods (Vincent, 1970). All were classified within respectively. Plus JQ339911, JQ339899, JQ339908 and JQ339916 the Mesorhizobium based upon 16S rRNA gene for the partial nifH gene; and JQ339881, JQ339869, JQ339878 and sequence analysis (our unpublished data), with the methods JQ339886 for the partial nodC gene for strains CCBAU 33460T, described subsequently. Within these rhizobia, five isolates CCBAU 33430, CCBAU 33431 and CCBAU 33455, respectively. originating from Jiangxi province shared identical sequences Six supplementary figures and three supplementary tables are available of 16S rRNA genes and formed a unique lineage in a with the online version of this paper. subcluster together with the type strains of Mesorhizobium Abbreviations: Box-PCR, Box-repeat-based PCR; IGS, 16S–23S rRNA loti, Mesorhizobium shangrilense, Mesorhizobium ciceri and intergenic spacer; NJ, neighbour-joining; ML, maximum-likelihood. (Fig. 1). To clarify the taxonomic relationships of these five

2002 044362 G 2013 IUMS Printed in Great Britain Mesorhizobium qingshengii sp. nov.

CCBAU 33446 (JQ339775) CCBAU 33455 (JQ339793) CCBAU 33460T (JQ339788) M. qingshengii sp. nov. CCBAU 33431 (JQ339785) CCBAU 33430 (JQ339776) 74/76 Model selected: GTR+I+G M. ciceri LMG 14989T (U07934) -lnL=2790.6831 67/80 T K=10 M. loti NZP 2213 (NR_025837) AIC= 5601.3662 68/80 M. shangrilense CCBAU 65327T (EU074203) Base frequencies: 73/72 M. australicum WSM2073T (AY601516) freqA =0.2444 freqC =0.2331 M. alhagi CCNWXJ 12-2T (EU169578) freqG =0.3129 100/100 M. camelthorni CCNWXJ 40-4T (EU169581) freqT =0.2095 Substitution model: T M. albiziae CCBAU 61158 (DQ100066) Rate matrix M. chacoense LMG 19008T (AJ278249) R(a) [A-C] =1.4942 R(b) [A-G] =1.5203 81/86 M. robiniae CCNWYC 115T (EU849582) R(c) [A-T] =3.0340 91/88 M. muleiense CCBAU 83963T (HQ316710) R(d) [C-G] =0.4775 65/79 M. temperatum SDW018T (AF508208) R(e) [C-T] =2.8207 68/89 R(f) [G-T] =1.0000 T M. mediterraneum LMG 17148 (AM181745) Among-site rate variation M. caraganae CCBAU 11299T (EF149003) Proportion of invariable sites T (I) = 0.8315 M. gobiense CCBAU 83330 (EF035064) Variable sites (G) M. metallidurans STM 2683T (AM930381) Gamma distribution shape 69/95 M. tianshanense CCBAU 3306 T (AF041447) parameter = 0.7122 M. tarimense CCBAU 83306T (EF035058) M. septentrionale SDW014T (AF508207) 81/95 M. silamurunense CCBAU 01550T (EU399698) M. plurifarium LMG 11892T (Y14158) M. amorphae ACCC 19665T (AF041442) M. opportunistum WSM2075T (AY601515) M. thiogangeticum SJTT (AJ864462) M. huakuii CCBAU 2609T (FJ491264) S. fredii USDA 205T (AY260149)

0.1

Fig. 1. Maximum-likelihood tree reconstructed from 16S rRNA gene sequences showing the phylogenetic relationships of the representative strains of Mesorhizobium qingshengii sp. nov. under the best-fit model shown. Bootstrap values of 65 % or more (ML/NJ) are provided at the nodes. The sequence of Sinorhizobium fredii USDA 205T was used as an outgroup. Bar, 10 % sequence divergence. strains, a polyphasic approach was performed in this Mesorhizobium huakuii, and study. All the bacterial strains used in the present study Mesorhizobium tianshanense. were cultured in TY medium (tryptone, 5.0 g; yeast Amplification of 16S rRNA, atpD, glnII and recA genes with extract, 3.0 g; CaCl , 0.6 g; pH 7.0–7.2) at 28 uC (except 2 primer sets P1/P6 (Tan et al., 1997), atpD255F/atpD782R, where indicated). recA41F/recA640R and glnII12F/glnII689R (Vinuesa et al., For 16S–23S intergenic spacer (IGS) RFLP, the IGS 2005), respectively, was performed with the protocols fragment was amplified with primer pair FGPS6/23S-38 originally described. Amplification of partial nifH and and the PCR protocol of Rasolomampianina et al. (2005). nodC genes was performed with primers nifH1F/nifH1R The genomic DNA extracted from each strain following the (Laguerre, et al., 2001) and nodC540/nodC1160 (Sarita method of Terefework et al. (2001) was used as template. et al., 2005), respectively. All the amplified fragments were The amplified genes were digested with HaeIII, HhaI and directly sequenced as described by Hurek et al. (1997). The MspI as specified by the manufacturer and the digested sequences were aligned with those of defined Mesorhizobium fragments were separated and visualized as described by species using the CLUSTAL W program in the MEGA 5.0 Terefework et al. (2001). The five strains showed five software package (Kumar et al., 2008). Unrooted trees were different patterns and were clustered as a group at 72 % constructed with the neighbour-joining (NJ) method similarity in the cluster analysis using the DICE coefficient (Saitou & Nei, 1987) and Jukes–Cantor distance (Jukes & and UPGMA method (Fig. S1 available in IJSEM Online). Cantor, 1969), respectively, and were bootstrapped using They were further grouped at 60.6 % similarity with the 1000 replications (Felsenstein, 1985). Maximum-likelihood type strains of Mesorhizobium caraganae, M. ciceri, M. loti, (ML) trees were also constructed using the PhyML 3.0 http://ijs.sgmjournals.org 2003 W. T. Zheng and others program (Guindon & Gascuel, 2003). The robustness of ML Nodulation and nitrogen-fixation abilities and host ranges topologies was inferred by non-parametric bootstrap tests are important features for symbiotic rhizobial species. In based on 100 pseudo-replicates of the data (Felsenstein, this study, cross nodulation tests were performed in 1985). The nucleotide substitution model was selected Leonard jars filled with vermiculite moistened with N-free by Akaike’s information criterion, as implemented in solution (Vincent, 1970). Except strain CCBAU 33446, the Modeltest 3.7 (Posada & Crandall, 1998). remaining four strains were able to nodulate with A. sinicus and occasionally with Astragalus adsurgens (four of the 10 In the phylogenetic analysis based on 16S rRNA gene plant replications nodulated) under laboratory conditions. sequences (1270 nt), the phylogenetic topologies were None of the five strains was able to nodulate with Cicer similar in the ML tree (Fig. 1) and in the NJ tree (not arietinum, corniculatus, Arachis hypogaea, Glycine shown). The gene sequences of the five tested strains were max, Trifolium repens, Pisum sativum or Medicago sativa. identical, and shared similarities of 99.8–99.6 % with those of the type strains of M. loti, M. shangrilense, M. ciceri and As a standard method for species definition (Wayne et al., M. australicum. They shared 96.4–98.7 % sequence sim- 1987; Graham et al., 1991; Tindall et al., 2010), DNA–DNA ilarity with the type strains of the 20 other recognized hybridization was performed in this study, using strain Mesorhizobium species (Table S1). CCBAU 33460T as reference. Total DNA isolation and the renaturation-rate technology were used according to the Multilocus sequence analysis is a reliable method for methods of Marmur (1961) and De Ley et al. (1970), rhizobial species or genospecies description (Vinuesa et al. respectively. All hybridizations were performed in triplic- 2005; Sahgal & Johri, 2006; Rivas et al., 2009). The results ate. The level of DNA–DNA relatedness between strain of phylogenetic analyses with the combined sequences CCBAU 33460T and the other four strains in the novel of the atpD, glnII and recA genes were similar in the ML group varied between 82.03 and 93.49 % (Table S1). The and NJ trees. In the ML tree (Fig. S2), the strains of the T novel group comprised a distinct lineage in the genus data for strain CCBAU 33460 and the type strains of 12 Mesorhizobium, clearly separated from all defined species. related Mesorhizobium species were in the range 21.54– Similarities of the combined sequences were 97.2–100 % 41.41 % (Table S1). These data demonstrated that the novel among the five novel strains (Table S1). Meanwhile, they strains formed a distinctive genospecies according to the suggested species threshold (70 % DNA–DNA relatedness) showed highest similarities of 94.7 and 93.3 % with M. + ciceri LMG 14989T and Mesorhizobium opportunistum LMG (Wayne et al., 1987). The DNA G C content, measured 24607T, respectively. Similarities between the novel strains with the thermal denaturation method (De Ley et al., and the type strains of other Mesorhizobium species ranged 1970), was 59.5–62.2 mol% for the five novel strains (Table between 92.9 and 87.5 % (Table S1). These data confirmed S1), within the range of Mesorhizobium members (59– the relationships of the phylogeny of the 16S rRNA gene, 64 mol%) (Jarvis et al., 1997). implying that the five novel strains represented a potential To estimate the genetic diversity in the novel group, Box- novel species, according to the suggested threshold for PCR fingerprinting was performed for the five tested Bradyrhizobium species (Rivas et al., 2009). strains in comparison with reference strains of defined The symbiotic genes nifH and nodC are commonly Mesorhizobium species, as described by Versalovic et al. analysed for the description of novel rhizobial species. (1994). Four distinctive patterns were observed in the five The results in the present study showed that four of the five novel strains which were different from those of the tested strains had identical nodC or nifH genes (Figs S3 and reference strains (Fig. S5). Although identical Box-PCR S4). The nodC (346 nt) and nifH (485 nt) genes of the patterns were obtained from strains CCBAU 33431 and novel strains shared 70.4–83.2 and 89.9–96.8 % similarity, CCBAU 33455, they could be distinguished from each respectively (Table S1), with those of the type strains of other based on 16S–23S IGS-RFLP profiles (Fig. S1) and recognized Mesorhizobium species. The type strains of M. sequence discrepancy in the glnII and aptD genes (Fig. S2). huakuii, Mesorhizobium metallidurans and Mesorhizobium Therefore, the five strains were not clones. thiogangeticum were not included because they do not have The profile of cellular fatty acids of strain CCBAU 33460T these genes (Ghosh & Roy, 2006). In the phylogeny of nodC was assayed in comparison with M. loti NZP 2213T, M. ciceri genes, the four novel strains shared identical sequences LMG 14989T, M. metallidurans LMG 24485T, Mesorhizobium with 7653R (AJ249393) originating from A. sinicus (Zhang tarimense CCBAU 83306T and M. tianshanense CCBAU et al., 2000). It confirmed that the common nodulation 3306T. The strains were grown to the late-exponential phase genes of A. sinicus rhizobia were conserved despite the on YM medium (Vincent, 1970) at 28 uC. Fatty acid methyl chromosomal divergence. In this study, amplification of esters were prepared and separated according to Sasser the nifH and nodC genes failed for strain CCBAU 33446, (1990) and identified with the MIDI Sherlock Microbial suggesting the deletion or mutation of these genes or the Identification System (Sherlock licence CD v 6.0), using the existence of quite different nodulation or nitrogen-fixation TSBA6 database. Twenty fatty acids were detected, 10 of [ genes in the strain, as mentioned in other cases (De´narie´ which C16 : 0,C17 : 0 iso, C17 : 1v8c,C17 : 0,C18 : 1v9c,C18 : 0, et al., 1996). However, this difference does not modify its C18 : 1v7c 11-methyl, C19 : 0 cyclo v8c, summed feature 3 ] species definition. (C16 : 1v6c/C16 : 1v7c) and summed feature 8 (C18 : 1v6c)

2004 International Journal of Systematic and Evolutionary Microbiology 63 Mesorhizobium qingshengii sp. nov.

T were common to all the tested strains (Table S2). Summed C13 : 0 iso 3-OH, which was absent in M. ciceri LMG 14989 . feature 8 (37.06–63.93 %), C16 : 0 (11.29–14.25 %), C18 : 1v7c In contrast, C17 : 1v6c and C19 : 0 10-methyl were present in M. T T 11-methyl (10.18–15.82 %) and C19 : 0 cyclo v8c (3.61–20 %) ciceri LMG 14989 , but absent in strain CCBAU 33460 .In were predominant fatty acids in all the tested Mesorhizobium addition, strain CCBAU 33460T was clearly distinct from M. strains in this study. This result was similar to that of Tighe ciceri LMG 14989T, M. loti NZP 2213T and M. tarimense T et al. (2000). Eleven fatty acids were detected in strain CCBAU CCBAU 83306 based on levels of C19 : 0 cyclo v8c (3.61, 33460T, similar to that in M. ciceri LMG 14989T (12 14.57, 17.81 and 20.00 %, respectively). These results components). Sixteen to 18 different fatty acids were detected indicated that the fatty acid compositions can be used to in the other four reference strains. Strain CCBAU 33460T had distinguish strain CCBAU 33460T from closely related species.

Table 1. Differential phenotypic features of strains of Mesorhizobium qingshengii sp. nov. and the type strains of related Mesorhizobium species

Strains: 1, CCBAU 33460T; 2, CCBAU 33430; 3, CCBAU 33431; 4, CCBAU 33446; 5, CCBAU 33455; 6, M. loti NZP 2213T;7,M. ciceri LMG 14989T; T T T 8, M. metallidurans LMG 24485 ;9,M. tarimense CCBAU 83306 ; 10, M. tianshanense CCBAU 3306 . +, Positive; 2, negative; W, weak; ND, not determined. All data were obtained in this study unless otherwise indicated.

Characteristic 1 2 3 4 5 6 7 8 9 10

Carbon source utilization D-Arabinose, D-galactose +++++ + ++2 + D-Ribose +++++ 2 + 22+ D-Xylose +++++ 22+ 2 + Dextrin 22222 2 22++ Dulcitol +++++ 2 + 2 ++ L-Rhamnose 22222 + 2 + 22 L-Threonine + 22++ 2 + 22+ L-Glycine + 2 +++ 2 + 22+ L-Arginine +++++ 2222+ L-Proline +++++ 2 + 2 + + Lactose 22222 +++2 + Gluconate, malonate, inulin 22222 2 222+ meso-Erythritol, DL-malate +++++ 2 + 2 ++ Salicin, acetate, citrate 22222 2 222+ Sorbitol 22222 2 22+ 2 Formate, D-gluconate, hippurate 22222 2 222+ Succinate +++++ 2 + 22+ Sorbose 22222 2 2+ 2 + Nitrogen source utilization D-Threonine, DL-alanine, hypoxanthine 22222 2 222+ L-Arginine 22222 2 22++ L-Lysine +++++ 2 + 2 ++ D-Glutamic acid, L(+)-glutamic acid +++++ 2 + 22+ L-Isoleucine, L-valine +++++ 2 + 22+ Phenylalanine dehydrogenase 22222 ++222 Curd in litmus milk 22222 ++222 Reduction in litmus milk +++++ 22+ 22 Growth with 0.1 % erythrosine bluish +++++ + W 222 Resistance to (mgml21): Ampicillin (5) 22222 ++22+ Ampicillin (100) 22222 + 2222 Tetracycline hydrochloride (5) 22222 ++2 ++ Maximum growth temperature (uC) 35 35 35 35 35 ,39–40* 40* 37* 30* ND* Maximum NaCl for growth (%, w/v) 11111,2* 2* 1* 1* 1* pH range for growth 5.5–9.0 5.5–9.0 5.5–9.0 5.5–9.0 5.5–9.0 4–10* 5–10* 5–11* 5–9* ND* DNA G+C content (mol%) 59.52 62.22 60.56 61.39 59.73 59–64* 63–64* ND* 57.85* 59–63*

*Data for M. loti, M. ciceri and M. tianshanense from Jarvis et al. (1997), data for M. metallidurans from Vidal et al. (2009) and data for M. tarimense from Han et al. (2008). http://ijs.sgmjournals.org 2005 W. T. Zheng and others

The respiratory quinone, determined by HPLC as described by brown, and in nutrient broth. Unable to grow in medium Lee et al. (2001), was Q-10 in strain CCBAU 33460T and the supplied separately with 0.1 % bromothymol blue, acridine five reference strains mentionedabove.Theanalysisofpolar hydrochloride, methylene blue, methyl green, neutral red, lipids with two-dimensional TLC (Minnikin et al., 1984) failed gentian violet or sodium nitrite. Acid reaction in litmus for M. tarimense CCBAU 83306T because of the difficulty in milk. Positive for urease and catalase. In addition to the extracting enough sample. Diphosphatidylglycerol (DPG), carbon and nitrogen sources listed in Table 1, also utilizes phosphatidylglycerol (PG), phosphatidylethanolamine (PE) D-fructose, D-galactose, D-glucose, maltose, D-mannose, and phosphatidylcholine (PC) were the major phospholipids melibiose, inositol, mycose and sucrose as sole carbon of the analysed strains (Table S3, Fig. S6), similar to the data source, but not adipic acid, D-amygdalin, raffinose, DL- for recognized Mesorhizobium species (Choma & Komaniecka, asparagine, L-methionine, soluble starch, syringic acid or 2002) and for Bradyrhizobium species (Wang et al., 2013). The sodium tartrate as a sole carbon source. Unable to utilize L- major polar lipids of Mesorhizobium strains were very similar cystine or L-methionine as sole nitrogen sources. Can also in this study, only quantitative differences being observed. In be distinguished from recognized Mesorhizobium species by addition, type and content differences in minor unknown housekeeping genes and by DNA–DNA hybridization polar lipids were detected in all the tested strains. An (Table S1). More distinguishing features are presented in ornithine-containing lipid (OL) was not identified in this Tables S2 and S3. study although it was reported to be present in several The type strain, CCBAU 33460T (5CGMCC 1.12097T5LMG Mesorhizobium species (Choma & Komaniecka, 2002, 2003). T T This compound is normally synthesized by grown in 26793 5HAMBI 3277 ), was isolated from effective nodule medium deficient in phosphate. of A. sinicus growninYuanZhoudistrict,YiChunCity, Jiangxi Province of China. The DNA G+C content of the The phenotypic features of the five novel strains were type strain is 59.52 mol% (Tm). The type strain is resistant to determined according to the method described by Smibert (mgml21) bacitracin (300), chloramphenicol (5), erythro- T T & Krieg (1994). M. loti NZP 2213 , M. ciceri LMG 14989 , mycin (300), neomycin sulfate (5) and streptomycin (300). M. tianshanense CCBAU 3306T (Jarvis et al., 1997), M. metallidurans LMG 24485T (Vidal et al., 2009) and M. tarimense CCBAU 83306T (Han et al., 2008) were also Acknowledgements included. We tested features the utilization of sole carbon This work was supported by National Natural Science Foundation of and nitrogen sources, resistance to antibiotics, tolerance China (project no. 31170003 to X. H. S.). E. T. W. was supported of NaCl, and pH and temperature ranges for growth. financially by the Instituto Polite´cnico Nacional, Mexico (grants Biochemical tests, including activities of catalase, urease SIP20110424 and 20120760). We thank Dr Joseph Rochlin of Lehman and nitrate reductase and reaction in litmus milk, were also College, City University of New York, for revising the English text. performed. Distinctive features between the novel group and closely related reference strains are shown in Table 1 and additional features are reported in the species References description below. Chen, W. X., Li, G. S., Qi, Y. L., Wang, E. T., Yuan, H. L. & Li, J. L. The results of the present study demonstrate that the five (1991). Rhizobium huakuii sp. nov. isolated from the root nodules of strains represent a novel species of the genus Mesorhizobium, Astragalus sinicus. Int J Syst Bacteriol 41, 275–280. for which we propose the name Mesorhizobium qingshengii Choma, A. & Komaniecka, I. (2002). Analysis of phospholipids and sp. nov. ornithine-containing lipids from Mesorhizobium spp. Syst Appl Microbiol 25, 326–331. Description of Mesorhizobium qingshengii Choma, A. & Komaniecka, I. (2003). The polar lipid composition of Mesorhizobium ciceri. Biochim Biophys Acta 1631, 188–196. sp. nov. De Ley, J., Cattoir, H. & Reynaerts, A. (1970). The quantitative Mesorhizobium qingshengii (qing.sheng9i.i. N.L. gen. masc. measurement of DNA hybridization from renaturation rates. Eur J n. qingshengii in honour of Dr Qing-Sheng Fan, a respected Biochem 12, 133–142. soil microbiological scientist, and one of the first micro- De´ narie´ , J., Debelle´ , F. & Prome´ , J. C. (1996). Rhizobium lipo- biologists to study rhizobia isolated from Astragalus sinicus chitooligosaccharide nodulation factors: signaling molecules medi- in China). ating recognition and morphogenesis. Annu Rev Biochem 65, 503– 535. Cells are Gram-negative, aerobic, non-spore-forming rods Felsenstein, J. (1985). Confidence limits on phylogenies: an approach of 0.38–0.5260.75–2.12 mm. Colonies on YMA medium using the bootstrap. Evolution 39, 783–791. are circular, convex and translucent, and have a diameter Ghosh, W. & Roy, P. (2006). Mesorhizobium thiogangeticum sp. nov., a of 1–2 mm within 3–4 days at 28 uC. Generation times are novel sulfur-oxidizing chemolithoautotroph from rhizosphere soil of 5–7 h in YM broth. The pH range for growth is 5.5–9.0, an Indian tropical leguminous plant. Int J Syst Evol Microbiol 56, 91– with optimum growth at pH 7.0. Growth occurs between 97. 10 and 35 uC, with optimum growth at 28 uC. Growth Graham, P., Sadowsky, M., Keyser, H., Barnet, Y., Bradley, R., occurs in medium separately supplied with 1 % (w/v) Cooper, J., De Ley, D., Jarvis, B., Roslycky, E. & other authors (1991). NaCl, 0.1 % sodium deoxycholate, Congo red or Bismarck Proposed minimal standards for the description of new genera and

2006 International Journal of Systematic and Evolutionary Microbiology 63 Mesorhizobium qingshengii sp. nov. species of root-and stem-nodulating bacteria. Int J Syst Evol Microbiol Rivas, R., Martens, M., de Lajudie, P. & Willems, A. (2009). Multilocus 41, 582–587. sequence analysis of the genus Bradyrhizobium. Syst Appl Microbiol Guindon, S. & Gascuel, O. (2003). A simple, fast, and accurate 32, 101–110. algorithm to estimate large phylogenies by maximum likelihood. Syst Sahgal, M. & Johri, B. N. (2006). of rhizobia: current Biol 52, 696–704. status. Curr Sci 90, 486–487. Guo, X. W., Zhang, X. X., Zhang, Z. M. & Li, F. D. (1999). Saitou, N. & Nei, M. (1987). The neighbor-joining method: a new Characterization of Astragalus sinicus rhizobia by restriction fragment method for reconstructing phylogenetic trees. Mol Biol Evol 4, 406– length polymorphism analysis of chromosomal and nodulation genes 425. regions. Curr Microbiol 39, 358–0364. Sarita, S., Sharma, P. K., Priefer, U. B. & Prell, J. (2005). Direct Han, T. X., Han, L. L., Wu, L. J., Chen, W. F., Sui, X. H., Gu, J. G., Wang, amplification of rhizobial nodC sequences from soil total DNA and E. T. & Chen, W. X. (2008). Mesorhizobium gobiense sp. nov. and comparison to nodC diversity of isolates. FEMS Microbiol Mesorhizobium tarimense sp. nov., isolated from wild legumes Ecol 54, 1–11. growing in desert soils of Xinjiang, China. Int J Syst Evol Microbiol Sasser, M. (1990). Identification of bacteria by gas chromatography of 58, 2610–2618. cellular fatty acids, MIDI Technical Note 101. Newark, DE: MIDI Inc. Hurek, T., Wagner, B. & Reinhold-Hurek, B. (1997). Identification of Smibert, R. & Krieg, N. (1994). Phenotypic characterization. In N2-fixing plant- and fungus-associated Azoarcus species by PCR- Methods for General and Molecular Bacteriology, pp. 607–654. Edited based genomic fingerprints. Appl Environ Microbiol 63, 4331–4339. by P. Gerhardt, R. G. E. Murray, W. A. Wood & N. R. Krieg. Jarvis, B. D. W., Van Berkum, P., Chen, W. X., Nour, S. M., Fernandez, Washington, DC: American Society for Microbiology. M. P., Cleyet-Marel, J. C. & Gillis, M. (1997). Transfer of Rhizobium Tan, Z. Y., Xu, X. D., Wang, E. T., Gao, J. L., Martinez-Romero, E. & loti, Rhizobium huakuii, Rhizobium ciceri, Rhizobium mediterraneum, Chen, W. X. (1997). Phylogenetic and genetic relationships of and Rhizobium tianshanense to Mesorhizobium gen. nov. Int J Syst Mesorhizobium tianshanense and related rhizobia. Int J Syst Bacteriol Bacteriol 47, 895–898. 47, 874–879. Jukes, T. H. & Cantor, C. R. (1969). Evolution of protein molecules. In Terefework, Z., Kaijalainen, S. & Lindstro¨ m, K. (2001). AFLP Mammalian Protein Metabolism, vol. 3, pp. 21–132. Edited by H. fingerprinting as a tool to study the genetic diversity of Rhizobium N. Munro. New York: Academic Press. galegae isolated from Galega orientalis and Galega officinalis. J Biotechnol 91, 169–180. Kumar, S., Nei, M., Dudley, J. & Tamura, K. (2008). MEGA: a biologist-centric software for evolutionary analysis of DNA and Tighe, S. W., de Lajudie, P., Dipietro, K., Lindstro¨ m, K., Nick, G. & protein sequences. Brief Bioinform 9, 299–306. Jarvis, B. D. W. (2000). Analysis of cellular fatty acids and phenotypic relationships of Agrobacterium, Bradyrhizobium, Mesorhizobium, Laguerre, G., Nour, S. M., Macheret, V., Sanjuan, J., Drouin, P. & Rhizobium and Sinorhizobium species using the Sherlock Microbial Amarger, N. (2001). Classification of rhizobia based on nodC and Identification System. Int J Syst Evol Microbiol 50, 787–801. nifH gene analysis reveals a close phylogenetic relationship among Phaseolus vulgaris symbionts. Microbiology 147, 981–993. Tindall, B. J., Rossello´ -Mo´ ra, R., Busse, H. J., Ludwig, W. & Ka¨ mpfer, P. (2010). Notes on the characterization of prokaryote strains for Lee, J. S., Shin, Y. K., Yoon, J. H., Takeuchi, M., Pyun, Y. R. & Park, taxonomic purposes. Int J Syst Evol Microbiol 60, 249–266. Y. H. (2001). Sphingomonas aquatilis sp. nov., Sphingomonas koreensis sp. nov., and Sphingomonas taejonensis sp. nov., yellow-pigmented Versalovic, J., Schneider, M., De Bruijn, F. J. & Lupski, J. R. (1994). bacteria isolated from natural mineral water. Int J Syst Evol Microbiol Genomic fingerprinting of bacteria using repetitive sequence based 51, 1491–1498. PCR (rep-PCR). Meth Cell Mol Biol 5, 25–40. Lin, D. H. & Gu, R. S. (1998). Chinese Milk Vetch. Fujian. Fujian Vidal, C., Chantreuil, C., Berge, O., Maure´ , L., Escarre´ , J., Be´ na, G., Science and Technology Publishing House. I, 1–5. (in Chinese). Brunel, B. & Cleyet-Marel, J. C. (2009). Mesorhizobium metallidurans sp. nov., a metal-resistant symbiont of Anthyllis vulneraria growing on Marmur, J. (1961). A procedure for the isolation of deoxyribonucleic metallicolous soil in Languedoc, France. Int J Syst Evol Microbiol 59, acid from . J Mol Biol 3, 208–218. 850–855. Minnikin, D. E., Odonnell, A. G., Goodfellow, M., Alderson, G., Vincent, J. M. (1970). Manual for the Practical Study of the Root- Athalye, M., Schaal, A. & Parlett, J. H. (1984). An integrated Nodule Bacteria, IBP Handbook 15. Oxford: Blackwell. procedure for the extraction of bacterial isoprenoid quinones and Vinuesa, P., Silva, C., Werner, D. & Martı´nez-Romero, E. (2005). polar lipids. J Microbiol Methods 2, 233–241. Population genetics and phylogenetic inference in bacterial molecular Murooka, Y., Xu, Y., Sanada, K., Araki, M., Morinaga, T. & Yokota, A. systematics: the roles of migration and recombination in (1993). Formation of root nodules by Rhizobium huakuii biovar Bradyrhizobium species cohesion and delineation. Mol Phylogenet renge, bv. nov. on Astragalus sinicus cv. Japan. Ferment Bioeng 78, 38– Evol 34, 29–54. 44. Wang, J. Y., Wang, R., Zhang, Y. M., Liu, H. C., Chen, W. F., Wang, E. T., Nuswantara, S., Fujie, M., Yamada, T., Malek, W., Inaba, M., Kaneko, Sui, X. H. & Chen, W. X. (2013). Bradyrhizobium daqingense sp. nov. Y. & Murooka, Y. (1999). Phylogenetic position of Mesorhizobium isolated from nodules of soybean grown in Daqing City of China. Int J huakuii subsp. rengei, a symbiont of Astragalus sinicus cv. Japan. Syst Evol Microbiol 63, 616–624. J Biosci Bioeng 87, 49–55. Wayne, L., Brenner, D., Colwell, R., Grimont, P. A. D., Kandler, O., Posada, D. & Crandall, K. A. (1998). MODELTEST: testing the model Krichevsky, M. I., Moore, L. H., Moore, W. E. C., Murray, R. G. E. & of DNA substitution. Bioinformatics 14, 817–818. other authors (1987). Repor6t of the ad hoc committee on Rasolomampianina, R., Bailly, X., Fetiarison, R., Rabevohitra, R., reconciliation of approaches to bacterial systematics. Int J Syst Be´ na, G., Ramaroson, L., Raherimandimby, M., Moulin, L., De Bacteriol 37, 463–464. Lajudie, P. & other authors (2005). Nitrogen-fixing nodules from Zhang, X. X., Turner, S. L., Guo, X. W., Yang, H. J., Debelle´ , F., Yang, rose wood legume trees (Dalbergia spp.) endemic to Madagascar host G. P., De´ narie´ , J., Young, J. P. & Li, F. D. (2000). The common seven different genera belonging to a- and b-. Mol Ecol nodulation genes of Astragalus sinicus rhizobia are conserved despite 14, 4135–4146. chromosomal diversity. Appl Environ Microbiol 66, 2988–2995. http://ijs.sgmjournals.org 2007