Rhizobium Sophorae Sp. Nov. and Rhizobium Sophoriradicis Sp. Nov., Nitrogen-Fixing Rhizobial Symbionts of the Medicinal Legume Sophora Flavescens

International Journal of Systematic and Evolutionary Microbiology (2015), 65, 497–503 DOI 10.1099/ijs.0.068916-0

Rhizobium sophorae sp. nov. and Rhizobium sophoriradicis sp. nov., nitrogen-fixing rhizobial symbionts of the medicinal legume Sophora flavescens

Yin Shan Jiao,1,2,3 Hui Yan,1,2,3 Zhao Jun Ji,1,2,3 Yuan Hui Liu,1,2,3 Xin Hua Sui,1,2,3 En Tao Wang,1,2,3,4 Bao Lin Guo,5 Wen Xin Chen1,2,3 and Wen Feng Chen1,2,3

Correspondence 1State Key Laboratory of Agrobiotechnology, Beijing 100193, PR China Wen Feng Chen 2College of Biological Sciences and Rhizobia Research Center, China Agricultural University, [email protected] Beijing 100193, PR China 3MOA Key Laboratory of Soil Microbiology, Beijing 100193, PR China 4Departamento de Microbiologı´a, Escuela Nacional de Ciencias Biolo´gicas, Instituto Polite´cnico Nacional, Me´xico D. F. 11340, Me´xico 5Institute of Medicinal Plant Development, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100193, PR China

Five bacterial strains representing 45 isolates originated from root nodules of the medicinal legume Sophora flavescens were defined as two novel groups in the genus Rhizobium based on their phylogenetic relationships estimated from 16S rRNA genes and the housekeeping genes recA, glnII and atpD. These groups were distantly related to Rhizobium leguminosarum USDA 2370T (95.6 % similarity for group I) and Rhizobium phaseoli ATCC 14482T (93.4 % similarity for group II) in multilocus sequence analysis. In DNA–DNA hybridization experiments, the reference strains CCBAU 03386T (group I) and CCBAU 03470T (group II) showed levels of relatedness of 17.9–57.8 and 11.0–42.9 %, respectively, with the type strains of related species. Both strains CCBAU 03386T and CCBAU 03470T contained ubiquinone 10 (Q-10) as the major respiratory quinone and possessed 16 : 0, 18 : 0, 19 : 0 cyclo v8c, summed feature 8 and summed feature 2 as major fatty acids, but did not contain 20 : 3 v6,8,12c. Phenotypic features distinguishing both groups from all closely related species of the genus Rhizobium were found. Therefore, two novel species, Rhizobium sophorae sp. nov. for group I (type strain CCBAU 03386T5E5T5LMG 27901T5HAMBI 3615T) and Rhizobium sophoriradicis sp. nov. for group II (type strain CCBAU 03470T5C-5-1T5LMG 27898T5HAMBI 3510T), are proposed. Both groups were able to nodulate Phaseolus vulgaris and their hosts of origin (Sophora flavescens) effectively and their nodulation gene nodC was phylogenetically located in the symbiovar phaseoli.

Rhizobia isolated from the medicinal legume Sophora species in the same genus, Sophora alopecuroides, have been flavescens have not been surveyed previously, although the extensively studied. Rhizobia isolated from Sophora alope- genetic diversity of rhizobia associated with another wild curoides were classified into five genera and nine genos- pecies, with Mesorhizobium alhagi and Mesorhizobium Abbreviations: DDH, DNA–DNA hybridization; ML, maximum-likelihood; gobiense as the main groups, and Agrobacterium tume- MLSA, multilocus sequence analysis; NJ, neighbour-joining. faciens, Mesorhizobium amorphae, Phyllobacterium trifolii, The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA gene Rhizobium giardinii, Rhizobium indigoferae, Ensifer fredii sequences of Rhizobium sophorae CCBAU 03386T and Rhizobium (former Sinorhizobium fredii) and Ensifer meliloti (former sophoriradicis CCBAU 03470T are KJ831229 and KJ831225, Sinorhizobium meliloti) as the minor groups (Zhao et al., respectively. 2010). At the time of writing, 76 species have been One supplementary table and six supplementary figures are available classified and described in the genus Rhizobium (Euze´by, with the online Supplementary Material. 1997; Parte, 2014), including recently described species

068916 G 2015 IUMS Printed in Great Britain 497 Y. S. Jiao and others isolated from the medicinal legume Calliandra grandiflora: Sequence comparison of housekeeping genes is regarded Rhizobium calliandrae, Rhizobium mayense and Rhizobium as a powerful taxonomic tool for differentiating closely jaguaris (Rinco´n-Rosales et al., 2013). related prokaryotes and for the description of novel species (Martens et al., 2008; Zhang et al., 2012). In the In a previous study, 271 root nodule bacteria were isolated present study, three housekeeping genes, namely recA, glnII from Sophora flavescens grown in fields of different regions and atpD, were amplified using primer pairs recA41F/ of China (our unpublished data). Of these isolates, 98 were recA640R, glnII12F/glnII689F and atpD255F/atpD782R, clustered within the genus Rhizobium based on phylogeny respectively (Turner & Young, 2000; Vinuesa et al., of the housekeeping gene recA (encoding recombinase 2005a, b). The PCR products were sequenced directly A), including 45 within two groups showing recA gene using the forward primers of the primer pairs and the sequence similarities lower than 97.4 % to the type strains acquired sequences were compared with those of the of recognized species of the genus Rhizobium (our unpub- reference or type strains. The ML phylogenetic tree (Fig. 2) lished data). In the present study, five isolates representing and the NJ phylogenetic tree (Fig. S3) of combined these two groups were systematically studied with poly- sequences of these three genes were reconstructed using phasic methods to determine their taxonomic positions. T MEGA 5.05 software (Tamura et al., 2011). Strains of group I The five representative isolates (CCBAU 03386 , CCBAU and group II represented two distinct lineages in both the 03383 and CCBAU 03468 representing 20 isolates in group T phylogenetic trees. Multilocus sequence analysis (MLSA) I; CCBAU 03470 and CCBAU 03433 representing 25 similarities ranged from 99.0 to 100 % and from 99.2 to isolates in group II) were isolated from root nodules of 100 % among strains within group I and group II, Sophora flavescens using the standard isolation method and respectively (Table 1), which supported the grouping maintained on yeast mannitol agar (YMA) (Somasegaran results based on 16S rRNA gene sequence analysis. Strain & Hoben, 1994) at 4 uC or room temperature for short- CCBAU 03386T in group I had highest MLSA similarity of term storage (less than 1 month) and in YM broth 95.6 % to R. leguminosarum USDA 2370T. Strain CCBAU (Somasegaran & Hoben, 1994) supplemented with 20 % 03470T in group II had highest MLSA similarity of 93.4 % (w/v) glycerol at 280 uC for long-term storage. to Rhizobium phaseoli ATCC 14482T (Table 1). Genomic DNA extracted from each of the strains using the DNA–DNA hybridization (DDH) allows genome-wide method of Terefework et al. (2001) was used as a template comparisons between organisms and is a standard technique for amplification of different DNA fragments. BOX-PCR for the description of novel species (Graham et al., 1991; fingerprinting was used to evaluate the genetic relation- Hanage et al., 2006; Wilson, 1987). Strain CCBAU 03386T, ships among the five isolates using the BOXAIR primer and representing group I, showed a mean DDH value of 92.6 % the procedure described by Versalovic et al. (1994). Results with strain CCBAU 03383 in the same group and less of BOX-PCR showed that these five isolates had different than 57.8 % with the type strains of related species (Table 1), fingerprints, although the isolates CCBAU 03383 and indicating that it represents a distinct genomic species. CCBAU 03468 displayed very similar fingerprints (Fig. S1, Strain CCBAU 03470T, representing group II, showed a available in the online Supplementary Material), dem- DDH value of 90.1 % with strain CCBAU 03433 in the same onstrating that the five isolates were independent strains. group, and less than 42.9 % with the type strains of related The partial 16S rRNA gene was amplified with primers P1 species (Table 1), suggesting this group as another novel and P6 (Chen et al., 1995) and was sequenced bidirection- species. The DNA G+C content was determined according ally using the same primer pair as described by Hurek et al. to the method of Marmur & Doty (1962) and strains (1997). Sequences acquired in this study and reference CCBAU 03386T and CCBAU 03470T contained 59.2 and sequences matched using the SeqWath program in the RDP 59.4 mol% (Tm), respectively, which were within the range database (Cole et al., 2009) were aligned using the CLUSTAL reported for species of the genus Rhizobium (Ramı´rez- W program in the MEGA 5.05 software (Tamura et al., 2011). Bahena et al., 2008; Tian et al., 2008). Evolutionary distances were calculated according to Kimura’s For cellular fatty acid analysis, strains CCBAU 03386T and two-parameter model (Kimura, 1980). Phylogenetic trees CCBAU 03470T were grown on YMA plates at 28 uC and were inferred using neighbour-joining (NJ) and maximum- harvested at the late-exponential phase of growth (after likelihood (ML) analyses. Two novel lineages were identified approximately 3 days of incubation). Cellular fatty acids as members of the genus Rhizobium based upon both the ML were obtained from cells by extraction and methylation (Fig. 1) and the NJ (Fig. S2) phylogenetic trees of 16S rRNA according to the MIDI protocol (Sasser, 1990) and then gene sequences. Strains in group I shared 100 % similarity were identified and quantified with the MIDI System with each other (Table 1) and formed a cluster together with software (Ver. 6.0, Aerobe RTSBA6 method and RTSBA6 T T Rhizobium laguerreae FB206 , R. indigoferae CCBAU 71042 database) (Tighe et al., 2000). Both strains CCBAU 03386T and Rhizobium leguminosarum USDA 2370T (Fig. 1). Strains and CCBAU 03470T possessed 16 : 0, 18 : 0, 19 : 0 cyclo v8c, in group II shared 99.2–100 % sequence similarity with each summed feature 8 (18 : 1v7c/18 : 1v6c) and summed other and formed a group together with Rhizobium etli feature 2 (12 : 0 aldehyde/unknown 10.928) as major fatty CFN42T (Fig. 1), with which they showed 98.8–99.0 % 16S acids (Table 2), consistent with the major compositions rRNA gene sequence similarity (Table 1). found in members of the genus Rhizobium (Tighe et al.,

498 International Journal of Systematic and Evolutionary Microbiology 65 Two novel Rhizobium species

R. vallis CCBAU 65647T (FJ839677) 55/89 R. multihospitium CCBAU 83401T (EF035074) CIAT 899T (HQ850704) 98/99 88/89 R. tropici Model selected: GTR+G+I R. hainanense I66T (U71078) BIC = 15944.050 R. miluonense CCBAU 41251T (EF061096) -lnL = 7643.203 R. calliandrae CCGE524T (JX855162) AICc = 15412.642 R. lusitanum P1-7T (AY738130) (+I) = 0.53 R. rhizogenes ATCC 11325T (AY945955) T (+G) = 0.96 R. jaguaris CCGE525 (JX855169) R = 1.98 R. leucaenae LMG 9517T (X67234) Base frequencies: R. mayense CCGE526T (JX855172) freqA = 0.194 R. etli CFN42T (U28916) freqC = 0.181 CCBAU 03433 (KJ831227) Group II CCBAU 03470T (KJ831225) freqG = 0.322 (Rhizobium sophoriradicis sp. nov.) 58/69 CCBAU 33202T (DQ835306) freqT = 0.303 R. fabae R. pisi DSM 30132T (AY509899) Rate matrix: R. phaseoli ATCC 14482T (EF141340) R(a) [A-T] = 0.018 R. leguminosarum USDA 2370T (NR044774) R(b) [A-C] = 0.040 R. laguerreae FB206T (JN558651) R(c) [A-G] = 0.103 R. indigoferae CCBAU 71042T (AY034027) R(d) [C-G] = 0.064 99 CCBAU 03383 (KJ831228) R(e) [C-T] = 0.187 CCBAU 03468 (KJ831224) Group I (Rhizobium sophorae sp. nov.) R(f) [G-T] = 0.028 CCBAU 03386T (KJ831229) 52/89 R. tubonense CCBAU 85046T (EU256434) R. grahamii CCGE 502T (JF424608) T 58/88 R. tibeticum CCBAU 85039 (EU256404) 0.05 71/54 98/99 ‘R. cauense’ CCBAU 101002 (JQ308326) R. mesoamericanum CCGE 501T (JF424606) R. alamii GBV016T (AM931436) 97/57 R. mesosinicum CCBAU 25010T (DQ100063) R. sullae IS123T (Y10170) 62/73 R. indigoferae CCBAU 71042T (NR025157) R. loessense CCBAU 07190 BT (AF364069) R. mongolense USDA 1844T (U89817) R. gallicum R-602 spT (U86343) R. rhizoryzae J3-AN59T (EF649779) B. japonicum USDA 6T (AB231927)

Fig. 1. ML tree reconstructed based on 16S rRNA gene sequences showing the phylogenetic relationships of Rhizobium sophorae sp. nov. (Group I), Rhizobium sophoriradicis sp. nov. (Group II) (in bold) and related species of the genus Rhizobium under the best-fit model shown beside the tree. Bootstrap values equal to or higher than 50 % (ML/NJ) are given at branch points. Bradyrhizobium japonicum USDA 6T was used as an outgroup. Bar, 5 % sequence divergence.

2000). However, 20 : 3 v6,9,12c was not detected in all the Gao et al. (1994). Urease, H2S production and the Voges– strains in this study, although it was reported previously in Proskauer reaction were determined as described by some species of the genus Rhizobium at a concentration Smibert and Krieg (1994). Cell size and appearance were of 0.77–2.66 % (Tighe et al., 2000). Ubiquinone 10 (Q- visualized using a Hitachi S-3400 scanning electron 10) was detected as the major respiratory quinone, using microscope. Specimens used for scanning electron micro- reversed-phase HPLC (Komagata & Suzuki, 1987), in scopy were prepared using the following method. Cells in a T T both strains CCBAU 03386 (85.4 %) and CCBAU 03470 colony cultured on a YMA plate were picked out and (89.3 %). were fixed in 2.5 % glutaraldehyde for more than 2 h. The Phenotypic features of all the novel strains were deter- fixed specimen was washed with 0.1 M phosphate buffer mined and compared with those of the type strains of the three times, fixed in 1 % osmium tetroxide for 2 h and phylogenetically most closely related species of the genus dehydrated in a graded ethanol series (30, 50, 70, 80, 90 Rhizobium. The following parameters were included for and 100 %; 5–15 min each). Then the specimen was dried phenotypic characterization: utilization of carbon sources, in a critical point dryer (Leica EM CPD). After being resistance to antibiotics, ranges of NaCl concentration, coated with a thin layer of gold using a sputter coater (Eiko temperature and pH for growth, reduction of H2O2, Nile IB-3), the specimen was used for scanning electron micro- blue and nitrate, according to the methods described by scopy. Cell size and shape of strains CCBAU 03386T and http://ijs.sgmjournals.org 499 Y. S. Jiao and others

Table 1. Sequence similarity and DNA–DNA relatedness (DDH) of representative strains CCBAU 03386T and CCBAU 03470T compared with other related strains in the genus Rhizobium

R. indigoferae CCBAU 71042T and Rhizobium pisi DSM 3012T were not used in DDH experiments because they were considered to be synonymous species to R. leguminosarum and Rhizobium fabae, respectively (Zhang et al., 2012). ND, Not determined.

Strain Similarities or relatedness (%) with Similarities or relatedness (%) with CCBAU 03470T CCBAU 03386T

16S rRNA MLSA DDH 16S rRNA MLSA DDH

Group I (Rhizobium sophorae sp. nov.) CCBAU 03386T 100 100 100 99.0 92.8 40.8 CCBAU 03383 100 99.0 92.6 99.0 93.1 ND CCBAU 03468 100 100 ND 98.0 92.8 ND Group II (Rhizobium sophoriradicis sp. nov.) CCBAU 03470T 99.0 92.8 40.8 100 100 100 CCBAU 03433 98.9 92.9 ND 99.9 99.2 90.1 R. etli CFN42T 98.8 90.8 52.6 99.8 91.8 11.0 R. fabae CCBAU 33202T 99.3 93.4 17.9 99.6 92.4 17.7 R. vallis CCBAU 65647T 97.4 93.5 27.5 98.3 92.4 38.1 R. phaseoli ATCC 14482T 99.3 90.7 57.8 99.6 93.4 21.5 R. leguminosarum USDA 2370T 99.8 95.6 19.2 98.8 91.9 42.9 R. laguerreae FB206T 100 95.2 25.8 99.0 92.7 42.3 T R. indigoferae CCBAU 71042 100 95.2 ND 99.0 91.3 ND T R. pisi DSM 30132 99.3 93.6 ND 99.6 92.6 ND

CCBAU 03470T are shown in Fig. S4 and described in the strains might be microsymbionts for both Sophora flavescens description of the novel species. and Phaseolus vulgaris. The phenotypic characteristics that differentiated the Based on the phenotypic and genetic characteristics, we two novel species from the type strains of phylogenetically classify the studied strains as representatives of two novel related species are shown in Table S1 and in the species species of the genus Rhizobium, for which the names descriptions. Cluster analysis of the phenotypic characters Rhizobium sophorae sp. nov. and Rhizobium sophoriradicis based on Euclidean distance and unweighted pair-group sp. nov. are proposed. method with arithmetic means (UPGMA) (Fig. S5) using R software (R Core Team, 2012) confirmed that the two Description of Rhizobium sophorae sp. nov. novel representative strains differed from the type strains of related species. Rhizobium sophorae (so.pho9rae. N.L. fem. n. sophorae of Sophora, the Latin name of the Chinese medicinal herb Symbiosis and the formation of effective nitrogen-fixing Sophora flavescens, from the root nodules of which the type nodules on legumes are important features of rhizobia. A strain was isolated). cross-nodulation test (Graham et al., 1991) was applied to evaluate the nodulation and nitrogen fixation abilities and Gram-reaction-negative, aerobic rods (1.1–1.5 mm60.5– host range of the novel groups in this study. Seeds of 0.6 mm). Colonies are small (2 mm in diameter) and pearl- Phaseolus vulgaris, Medicago sativa, Glycine max, Trifolium white after 3 days of incubation on YMA at 28 uC, which is repens, Astragalus sinicus and Sophora flavescens were surface- the optimal growth temperature. The optimum pH for sterilized, germinated and inoculated with strains CCBAU growth is pH 7–7.5. Temperature range for growth is 10– 03386T and CCBAU 03470T, respectively, in a Leonard jar 37 uC. Cannot grow on YMA containing more than 1 % assembly with reference to the procedures and methods (w/v) NaCl. Catalase, oxidase and urease are positive. described by Somasegaran & Hoben (1994). Effective Hydrolysis of starch and 3-ketolactose, growth in nutrient nodules were observed only on roots of Phaseolus vulgaris broth and H2S production are negative. Uses L-arabinose, and Sophora flavescens. No nodule was formed on roots of cellobiose, i-erythritol, sucrose, pyruvic acid methyl ester, the other test plants. Phylogenetic analysis of the nodulation D-galactonic acid lactone, bromosuccinic acid, urocanic gene nodC using the primer pair nodCF540/nodCR1160 and acid, glycerol, xylitol and D-galactose as carbon sources and the amplication procedure of Sarita et al. (2005) showed that grows weakly on uridine, L-alaninamide, succinamic acid, the novel strains formed a cluster together with the Phaseolus D-gluconic acid, maltose, a-D-glucose and D-arabitol as sole vulgaris nodulating rhizobia designated symbiovar (sv.) carbon source. The major cellular fatty acids are 16 : 0, phaseoli (Fig. S6). The nodulation tests and the phylogenetic 19 : 0 cyclo v8c, summed feature 8 (18 : 1v7c/18 : 1v6c) relationships based on nodC demonstrated that these novel and summed feature 2 (12 : 0 aldehyde/unknown 10.928).

500 International Journal of Systematic and Evolutionary Microbiology 65 Two novel Rhizobium species

93/100 FB206T (JN55868, JN558671, JN558661) 57/84 R. laguerreae CCBAU 03468 (KJ831247, KJ831236, KJ831230) T Group I

99/100 CCBAU 03386 (KJ831252, KJ831241, KJ831235) Model selected: T92+G+I 61/60 (Rhizobium sophorae sp. nov.) BIC = 8097.827 CCBAU 03383 (KJ831251, KJ831240, KJ831234) -lnL = 3635.294 53/94 R. indigoferae CCBAU 71042T (EF027965, JN580717, GU552925) AICc = 7424.847 99/100 T (+I) = 0.68 R. leguminosarum USDA 2370 (AJ294376, AF169586, AJ294405) (+G) = 0.29 R. vallis CCBAU 65647T (GU211770, GU211771, GU211768) R = 1.17 T Base frequencies: R. fabae CCBAU 33202 (EF579941, EF579935, EF579929) freqA = 0.227 99/100 R. pisi DSM 30132T (EF113134, JN580715, EF113149) freqC = 0.273 ATCC 14482T (EF113136, JN580716, EF113151) freqG = 0.273 86/100 R. phaseoli freqT = 0.227 64/76 CCBAU 03470T (KJ831248, KJ831237, KJ831231) Group II Rate matrix: 99/100 CCBAU 03433 (KJ831250, KJ831239, KJ831233) R(a) [A-T] = 0.052 (Rhizobium sophoriradicis sp. nov.) R(b) [A-C] = 0.0563 R. etli CFN42T (CP000133, EU488776, CP000133) R(c) [A-G] = 0.148 CCBAU 05176T (GU128902, GU128895, GU128888) R(d) [C-G] = 0.063 R. vignae R(e) [C-T] = 0.123 ‘R. cauense’ CCBAU 101002 (JQ308335, JQ308332, JQ308329) R(f) [G-T] = 0.052 R. sullae IS123T (FJ816279, FJ816280, DQ345069) T 95/100 99/100 R. mongolense USDA 1844 (AY907358, AY929453, AY907372) R. gallicum R-602 spT (AY907357, EU488785, HM142762) 71/88 R. yanglingense SH 22623T (AY907359, AY929462, AY907373) R. lusitanum P1-7T (DQ431674, EF639841, DQ431671) 0.05 R. rhizoryzae J3-AN59T (KF384476, AY929470, EU732563) 55/55 R. hainanense CCBAU 57015T (HM047132, GU726294, GU726293) R. multihospitium CCBAU 83401T (EF490029, EF490040, EF490019) 77/100 R. miluonense CCBAU 41251T (HM047131, HM047120, HM047116) R. tropici CIAT 899T (EU488815, EU488791, CP004015) R. giardinii R-4385T (AM182123, EU488778, AM418780) B. japonicum USDA 6T (AM168341, AF169582, AM418753)

Fig. 2. ML tree reconstructed based on housekeeping gene recA, glnII and atpD sequences showing the phylogenetic relationships of Rhizobium sophorae sp. nov. (Group I), Rhizobium sophoriradicis sp. nov. (Group II) (in bold) and related species of the genus Rhizobium under the best-fit model shown beside the tree. Bootstrap values equal to or higher than 50 % (ML/NJ) are given at branch points. B. japonicum USDA 6T wasusedasanoutgroup. Bar, 5 % sequence divergence.

T T T The type strain, CCBAU 03386 (5E5 5LMG 27901 5 ester and xylitol, but not glycogen, N-acetyl-D-galactosa- T HAMBI 3615 ), was isolated from effective nodules of mine, D-arabitol, cellobiose, i-erythritol, L-fucose, gentio- Sophora flavescens in Wuxiang County (GPS coordinates: biose, a-D-glucose, myo-inositol, a-lactose, maltose, D- 113u 119 560 E, 35u 459 390 N), Changzhi City, Shanxi mannitol, D-mannose, raffinose, methyl b-D-glucoside, L- + Province, China. The G C content of the type strain is rhamnose, D-sorbitol, sucrose, trehalose, turanose, succinic 59.2 mol% (Tm). Effectively nodulates Sophora flavescens acid monomethyl ester, citric acid, D-galactonic acid and Phaseolus vulgaris. lactone, D-galacturonic acid, D-gluconic acid, a-hydroxy- butyric acid, a-ketoglutaric acid, DL-lactic acid, D-saccharic Description of Rhizobium sophoriradicis sp. nov. acid, succinic acid, bromosuccinic acid, succinamic acid, L- alaninamide, D-alanine, L-alanine, L-ornithine, L-serine, L- Rhizobium sophoriradicis sp. nov. [so.pho.ri.ra9di.cis. N.L. threonine, urocanic acid, uridine, D-galactose or glycerol as n. Sophora a plant species; L. n. radix,-icis a root; N.L. gen. sole carbon source. The major cellular fatty acids are 16 : 0, n. sophoriradicis of (from) the root of Sophora]. 19 : 0 cyclo v8c, summed feature 8 (18 : 1v7c/18 : 1v6c) Gram-reaction-negative, aerobic rods (~1.45 mm60.55– and summed feature 2 (12 : 0 aldehyde/unknown 10.928). 0.57 mm). Colonies are small (3 mm in diameter) and T T T u The type strain, CCBAU 03470 (5C-5-1 5LMG 27898 5 pearl-white on YMA after 3 days of incubation at 28 C, T which is the optimal growth temperature. Can grow at HAMBI 3510 ), was isolated from effective nodules of pH 4.0–10.0 and at ¡37 uC. Does not grow on YMA Sophora flavescens in Zhenxin New Area (GPS coordinates: containing more than 1 % (w/v) NaCl. Catalase, oxidase 113u 039 020 E, 36u 029 240 N) of Changzhi City, Shanxi and urease are positive. Hydrolysis of starch, production of Province, China. The G+C content of the type strain is 3-ketolactose, growth in nutrient broth and production of 59.4 mol% (Tm). Effectively nodulates Sophora flavescens H2S are negative. Utilizes L-arabinose, pyruvic acid methyl and Phaseolus vulgaris. http://ijs.sgmjournals.org 501 Y. S. Jiao and others

Table 2. Fatty acid profiles of the novel isolates and recognized species of the genus Rhizobium

Strains: 1, R. sophorae sp. nov. CCBAU 03386T;2,R. sophoriradicis sp. nov. CCBAU 03470T;3,R. fabae CCBAU 33202T;4,R. vallis CCBAU 65647T; 5, R. phaseoli ATCC 14482T;6,R. etli CFN42T;7,R. leguminosarum USDA 2370T;8,R. laguerreae FB206T. Values are percentages of the total fatty acids; values below 0.1 % are not shown or are shown as TR (trace); 2, not detected.

Fatty acid 1 2 3 45678 15 : 0 2-OH 2.3 1.6 1.3 1.2 1.0 0.4 0.7 1.0 16 : 0 4.7 3.9 3.1 7.7 2.9 3.8 3.6 5.3 16 : 0 3-OH 4.1 1.8 0.9 1.1 1.2 1.6 1.6 1.5 17 : 0 TR 0.3 2222 1.3 2 18 : 0 9.1 7.8 9.0 10.7 9.8 7.0 3.2 8.3 18 : 1v7c 11-methyl 3.6 4.4 0.5 5.8 1.1 2 4.4 1.6 18 : 0 3-OH 0.7 1.1 2.1 1.8 3.0 1.6 2.6 2.0 19 : 0 cyclo v8c 5.5 15.8 7.0 20.5 13.8 10.2 15.6 9.5 Summed feature 2* 11.3 4.4 5.1 4.5 4.8 4.9 6.5 6.0 Summed feature 3* 1.2 0.6 0.6 0.7 2 0.6 0.8 0.9 Summed feature 8* 56.1 55.7 69.7 42.7 61.2 68.7 38.5 63.0

*Summed feature 2 comprised 12 : 0 aldehyde/unknown 10.928; summed feature 3 comprised 16 : 1v6c/16 : 1v7c; summed feature 8 comprised 8:1v7c/18 : 1v6c.

Acknowledgements Hurek, T., Wagner, B. & Reinhold-Hurek, B. (1997). Identification of N2-fixing plant- and fungus-associated Azoarcus species by PCR- We are grateful to the culture collection centre of LMG for providing based genomic fingerprints. Appl Environ Microbiol 63, 4331–4339. reference strains. This work was supported by Chinese Universities Kimura, M. (1980). Scientific Fund (2014QT003) and National Natural Science Foundation A simple method for estimating evolutionary rates of China (31270052) to W. F. C. E. T. W. was financially supported by of base substitutions through comparative studies of nucleotide Instituto Politecnico Nacional, Mexico (projects SIP 20130828 and sequences. J Mol Evol 16, 111–120. 20140124). B. L. G. was supported by Northern China Standardized Plant Komagata, K. & Suzuki, K. (1987). Lipid and cell-wall analysis in Base Optimization and Upgrading and Series Products Comprehensive bacterial systematics. Methods Microbiol 19, 161–207. Development of Sophora flavescens (No. 2011BA107B03). We thank Marmur, J. & Doty, P. (1962). Determination of the base composition Bernhard Schink and Tian Shen Tao for their suggestions on the of deoxyribonucleic acid from its thermal denaturation temperature. epithets of the two novel species, and Mr Zha Gen Guan and Miss J Mol Biol 5, 109–118. Shang Ying Wu for their help with collection of nodules. Martens, M., Dawyndt, P., Coopman, R., Gillis, M., De Vos, P. & Willems, A. (2008). Advantages of multilocus sequence analysis for References taxonomic studies: a case study using 10 housekeeping genes in the genus Ensifer (including former Sinorhizobium). Int J Syst Evol Chen, W. X., Wang, E. T., Wang, S. Y., Li, Y. B., Chen, X. Q. & Li, Y. Microbiol 58, 200–214. (1995). Characteristics of Rhizobium tianshanense sp. nov., a Parte, A. C. (2014). LPSN–list of prokaryotic names with standing in moderately and slowly growing root nodule bacterium isolated from nomenclature. Nucleic Acids Res 42 (Database issue), D613–D616. an arid saline environment in Xinjiang, People’s Republic of China. R Core Team (2012). R: A Language and Environment for Statistical Int J Syst Bacteriol 45, 153–159. Computing. Vienna: R Foundation for Statistical Computing. Cole, J. R., Wang, Q., Cardenas, E., Fish, J., Chai, B., Farris, R. J., Ramı´rez-Bahena, M. H., Garcıá-Fraile, P., Peix, A., Valverde, A., Kulam-Syed-Mohideen, A. S., McGarrell, D. M., Marsh, T. & other Rivas, R., Igual, J. M., Mateos, P. F., Martıńez-Molina, E. & Vela´ zquez, authors (2009). The Ribosomal Database Project: improved align- E. (2008). Revision of the taxonomic status of the species Rhizobium ments and new tools for rRNA analysis. Nucleic Acids Res 37 leguminosarum (Frank 1879) Frank 1889AL, Rhizobium phaseoli (Database issue), D141–D145. Dangeard 1926AL and Rhizobium trifolii Dangeard 1926AL. R. trifolii Euze´ by, J. P. (1997). List of bacterial names with standing in is a later synonym of R. leguminosarum. Reclassification of the strain nomenclature: a folder available on the internet. Int J Syst Bacteriol 47, R. leguminosarum DSM 30132 (5NCIMB 11478) as Rhizobium pisi 590–592. sp. nov. Int J Syst Evol Microbiol 58, 2484–2490. Gao, J. L., Sun, J. G., Li, Y., Wang, E. T. & Chen, W. X. (1994). Numerical Rinco´ n-Rosales, R., Villalobos-Escobedo, J. M., Rogel, M. A., taxonomy and DNA relatedness of tropical rhizobia isolated from Martinez, J., Ormen˜ o-Orrillo, E. & Martıńez-Romero, E. (2013). Hainan Province, China. Int J Syst Bacteriol 44, 151–158. Rhizobium calliandrae sp. nov., Rhizobium mayense sp. nov. and Graham, P., Sadowsky, M., Keyser, H., Barnet, Y., Bradley, R., Cooper, Rhizobium jaguaris sp. nov., rhizobial species nodulating the medicinal J., De Ley, D., Jarvis, B., Roslycky, E. & other authors (1991). Proposed legume Calliandra grandiflora. Int J Syst Evol Microbiol 63, 3423– minimal standards for the description of new genera and species of 3429. root-and stem-nodulating bacteria. Int J Syst Bacteriol 41, 582–587. Sarita, S., Sharma, P. K., Priefer, U. B. & Prell, J. (2005). Direct Hanage, W. P., Fraser, C. & Spratt, B. G. (2006). Sequences, sequence amplification of rhizobial nodC sequences from soil total DNA and clusters and bacterial species. Philos Trans R Soc Lond B Biol Sci 361, comparison to nodC diversity of root nodule isolates. FEMS Microbiol 1917–1927. Ecol 54, 1–11.

502 International Journal of Systematic and Evolutionary Microbiology 65 Two novel Rhizobium species

Sasser, M. (1990). Identification of bacteria by gas chromatography of Versalovic, J., Schneider, M., de Bruijn, F. & Lupski, J. (1994). cellular fatty acids, MIDI Technical Note 101. Newark, DE: MIDI Inc. Genomic fingerprinting of bacteria using repetitive sequence based Smibert, R. M. & Krieg, N. R. (1994). Phenotypic characterization. In PCR (rep-PCR). Methods Mol Cell Biol 5, 25–40. Methods for General and Molecular Bacteriology, vol. 5, pp. 611–654. Vinuesa, P., Silva, C., Werner, D. & Martı´nez-Romero, E. (2005a). Edited by P. Gerhardt, R. G. Murray, W. A. Wood & N. R. Krieg. Population genetics and phylogenetic inference in bacterial mole- Washington, DC: American Society for Microbiology. cular systematics: the roles of migration and recombination in Somasegaran, P. & Hoben, H. J. (1994). Handbook for Rhizobia, Bradyrhizobium species cohesion and delineation. Mol Phylogenet Evol Methods in Legume-Rhizobium Technology. New York: Springer. 34, 29–54. Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. & Kumar, Vinuesa, P., Silva, C., Lorite, M. J., Izaguirre-Mayoral, M. L., Bedmar, S. (2011). MEGA5: molecular evolutionary genetics analysis using E. J. & Martı´nez-Romero, E. (2005b). Molecular systematics of maximum likelihood, evolutionary distance, and maximum par- rhizobia based on maximum likelihood and Bayesian phylogenies simony methods. Mol Biol Evol 28, 2731–2739. inferred from rrs, atpD, recA and nifH sequences, and their use in the Terefework, Z., Kaijalainen, S. & Lindstro¨ m, K. (2001). AFLP fingerprint- classification of Sesbania microsymbionts from Venezuelan wetlands. ing as a tool to study the genetic diversity of Rhizobium galegae isolated Syst Appl Microbiol 28, 702–716. from Galega orientalis and Galega officinalis. JBiotechnol91, 169–180. Wilson, K. (1987). Preparation of genomic DNA from bacteria. In Tian, C. F., Wang, E. T., Wu, L. J., Han, T. X., Chen, W. F., Gu, C. T., Gu, Current Protocols in Molecular Biology, pp. 2.4.1–2.4.5. Edited by J. G. & Chen, W. X. (2008). Rhizobium fabae sp. nov., a bacterium that F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, nodulates Vicia faba. Int J Syst Evol Microbiol 58, 2871–2875. J. A. Smith & K. Struhl. New York: Green Publishing & Wiley- Interscience. Tighe, S. W., de Lajudie, P., Dipietro, K., Lindstro¨ m, K., Nick, G. & Jarvis, B. D. (2000). Analysis of cellular fatty acids and phenotypic Zhang, Y. M., Tian, C. F., Sui, X. H., Chen, W. F. & Chen, W. X. (2012). relationships of Agrobacterium, Bradyrhizobium, Mesorhizobium, Robust markers reflecting phylogeny and taxonomy of rhizobia. PLoS Rhizobium and Sinorhizobium species using the Sherlock Microbial ONE 7, e44936. Identification System. Int J Syst Evol Microbiol 50, 787–801. Zhao, L. F., Deng, Z. S., Yang, W. Q., Cao, Y., Wang, E. T. & Wei, G. H. Turner, S. L. & Young, J. P. W. (2000). The glutamine synthetases of (2010). Diverse rhizobia associated with Sophora alopecuroides grown rhizobia: phylogenetics and evolutionary implications. Mol Biol Evol in different regions of Loess Plateau in China. Syst Appl Microbiol 33, 17, 309–319. 468–477.

http://ijs.sgmjournals.org 503