Indian Journal of Experimental Biology Vol. 56, June 2018, pp. 373-384

Molecular characterization of microsymbionts associated with root nodules of Crotalaria burhia Buch.-Ham. ex Benth., a native keystone legume species from Thar Desert of India

Indu Singh Sankhla1,2, Raju Ram Meghwal1, Sunil Choudhary1,3, Sonam Rathi1, Nisha Tak1, Alkesh Tak1 & Hukam Singh Gehlot1*

1BNF and Microbial Genomics Laboratory, Department of Botany, Jai Narain Vyas University, Jodhpur-342 033, Rajasthan, India 2Department of Botany, University of Rajasthan, Jaipur-302 004, Rajasthan, India 3ICFRE-Arid Forest Research Institute (AFRI), Jodhpur-302 005, Rajasthan, India Received 19 April 2016; revised 14 July 2017

Establishment of legume-rhizobia symbiosis has ample agronomic and ecological significance. Characterization of native rhizobia could enhance our understanding of their natural distribution and co-evolution. The Great Indian Thar Desert is an ecologically significant unique habitat with its flora and fauna. Crotalaria spp. is an economically important legume widely distributed in the Thar Desert and can be considered its one of the bioresources, particularly for biological nitrogen fixation with their symbiotic rhizobia. Here, we examined the legume Crotalaria burhia Buch.-Ham. ex Benth. in search of potential novel rhizobial species. Out of 72 root nodule bacterial (RNB) strains isolated from C. burhia, 51 rhizobia-like strains were examined for genetic diversity based on ARDRA and RAPD patterns. BLASTn sequence similarity results based on 16S rRNA gene of selective thirteen strains representing four ARDRA types revealed that they were related to genera Ensifer, and Bradyrhizobium. In 16S rRNA gene phylogeny, five (CB5, CB17, CB36, CB44, CB56) strains were closer to Ensifer kostiensis, three (CB6, CB12, CB32) to E. terangae and CB11 showed similarity with E. kostiensis and E. saheli. Strain CB4 was similar to Bradyrhizobium yuanmingense and three (CB29, CB31, CB46) strains were closer to species of Rhizobium (R. etli, R. sullae and R. borbori respectively). Symbiotic (nodA and nifH) genes phylogeny of Ensifer sp. CB56 was incongruent and showed close similarity with E. fredii whereas sym gene phylogeny of Bradyrhizobium sp. CB4 was congruent with 16S rRNA gene phylogeny. In Rhizobium strains sym genes could not be amplified and they failed to nodulate host. Our study suggests that C. burhia is nodulated by diverse strains of Ensifer and Bradyrhizobium in alkaline soil of Thar Desert and these strains effectively cross-nodulated crop Vigna radiata.

Keywords: ARDRA, Bradyrhizobium, Burhia Rattlepod, Ensifer, nifH, Nitrogen fixation, nodA, Nodulation, Root nodule bacterial (RNB) strains

Legume-rhizobia symbiosis known for biological The genus Crotalaria L. is the third largest in nitrogen fixation have long term agronomical and Papilionoideae subfamily of Leguminosae having ecological significance and is one of the alternative to more than 700 species3. It is distributed in tropical and nitrogen fertilizers. Rhizobia are Gram negative, sub-tropical regions of the world with the majority of aerobic, soil with ability to mostly form N species (543) native to tropical Africa and fixing nodules on roots of legumes1. Most of the Madagascar. In India, it is the largest legume taxa legume-rhizobia symbiosis studies have been carried comprising 93 species of which 27 are endemic4. out on either agricultural crops (soybean, mungbean, Many species of Crotalaria have great economic and chickpea) or pasture legumes (clover and importance in terms of insecticides, fibres, silage and alfalfa). From last two decades such studies have been green manure having agronomic values4,5. Crotalaria extended to wild/native legumes from tropical and burhia Buch.-Ham. ex Benth., (Burhia Rattlepod) semi-tropical climatic zones that resulted in locally called ‘Shinio’, is a perennial bushy legume discovering several novel species and genera belonging widely distributed on sandy areas throughout the Thar to class and betaproteobacteria2. Desert of India and adjoining Pakistan. The Indian ______Thar Desert is characterized by erratic precipitation, *Correspondence: high irradiation, high temperature and saline tracts6. Phone: +91 94141 24939 (Mob.) E-mail: [email protected] Overall, the area is alkaline and the soil has been 374 INDIAN J EXP BIOL, JUNE 2018

classified as ‘desert soil’ poor in nitrogen and available Methylobacterium nodulans21. Furthermore, Liu et al.22 phosphorus. Therefore the plants, animals, and microbes isolated Rhizobium and Bradyrhizobium strains from growing in Thar Desert are continuously under biotic nodules of C. pallida in China, while Rocha23 isolated and abiotic stresses. Gehlot et al.7 suggested some Mesorhizobium sp. from root nodules of traditional and modern scientific approaches for C. spectabilis. On the basis of Amplified r-DNA characterization, conservation and sustainable utilization Restriction Analysis (ARDRA) Singha et al.24 identified of bioresources of the Thar Desert. Nodulation in three groups of rhizobial strains (Rhizobium, number of wild/native legumes of Thar Desert have Bradyrhizobium and Mesorhizobium) from C. pallida in been reported as well as root nodule microsymbionts of Assam, India. Sankhla et al.12 reported that several legumes have been characterized at molecular C. medicagenia is nodulated by Ensifer strains in sandy and genomic level recently8-19. areas of Indian Thar Desert. In this study, we The C. burhia is one of the keystone species of investigated genetic diversity of root nodule bacteria Thar Desert and has a great potential in terms of (RNB) associated with C. burhia and characterized its restoration of soil fertility through nitrogen fixation rhizobia at molecular level including symbiotic genes. and also in prevention of desertification through its soil binding and sand dune stabilizing properties, Materials and Methods hence it have great ecological importance. Nodule sampling and isolation of rhizobia Crotalaria has been reported to be nodulating by Root nodules of C. burhia were collected from 25 Bradyrhizobium strains20. However, Crotalaria sites in five districts of Western Rajasthan (Table 1). podocarpa is nodulated by a novel rhizobia The nodules were collected during the monsoon and

Table 1 — Nodulation status of Crotalaria burhia growing at different sampling sites throughout Thar Desert and origin of root nodule bacterial (RNB) strains District Soil sampling site Geographical coordinates pH OC Total N P Avg. no. of Purified root nodule (%) (%) (kg/ha) nodules per plant bacterial strains Jodhpur Amritlal stadium 26°20'25.40"N/73° 3'3.09"E 8.2 0.16 0.0078 9.2 08 CB1, CB2, CB45, CB46r, CB47, CB48 JNVU New 26°14'49.85"N/73° 8.2 0.18 0.0091 12.4 09 CB5e, CB7, CB27, CB28, Campus 1'18.65"E CB29r, CB30 Kailana 26°18'5.83"N/72°58'29.38"E 8.2 0.10 0.0052 8.2 09 CB8, CB55 Osian 26°44'26.57"N/72°53'49.26"E 8.9 0.14 0.0071 18.4 14 CB38, CB39, CB40, CB42, CB44e Pratapnagar 26°17'29.28"N/73° 0'6.36"E 8.8 0.23 0.0048 16.3 09 CB49, CB50, CB51 Shergarh 26°19'10.58"N/72°18'1.42"E 8.7 0.19 0.0112 9.2 09 CB67, CB68 Nagaur Alai 27°19'24.91"N/73°35'6.99"E 8.2 0.24 0.0092 6.4 08 CB10, CB61 Baghnada 27° 8'42.94"N/73°48'32.42"E 8.2 0.26 0.0111 11.2 11 CB6e, CB15, CB16 Deh 27°18'30.40"N/73°54'53.51"E 8.4 0.28 0.0069 8.2 10 CB3, CB4b Harima 27°15'26.84"N/73°51'39.74"E 8.5 0.29 0.0071 7.8 14 CB21, CB22, CB23 Inana 27° 8'2.63"N/73°49'39.20"E 8.1 0.22 0.0114 10.8 12 CB18, CB19, CB20 Tausar 27° 6'36.85"N/73°46'30.60"E 8.4 0.23 0.0121 9.8 10 CB24, CB25, CB26 Barmer Barmer 25°45'10.09"N/71°26'5.31"E 8.2 0.14 0.0079 10.2 08 CB33, CB43 Bhuka 25°39'54.66"N/72° 0'54.03"E 8.8 0.13 0.0095 9.2 09 CB31r, CB32e, CB34 Chohtan 25°29'44.82"N/71° 5'6.90"E 8.1 0.16 0.0048 11.2 05 CB35, CB36e, CB37 Nimdee 26° 4'2.97"N/71°19'48.46"E 8.6 0.18 0.0081 11.9 07 CB9, CB11e, CB12e Jaisalmer Chandhan 26°59'18.13"N/71°18'13.62"E 8.8 0.17 0.0117 6.4 07 CB52, CB41 Jaisalmer 26°53'49.55"N/70°56'31.18"E 8.3 0.19 0.0050 6.9 07 CB63, CB64 Kuldhera 26°52'28.36"N/70°46'34.28"E 8.6 0.19 0.0094 5.5 09 CB56e, CB57 Longewala 27°30'8.57"N/70° 7'28.76"E 8.5 0.29 0.0098 7.1 07 CB65, CB66 Pokaran 26°58'25.49"N/71°54'6.92"E 8.7 0.22 0.0069 5.1 05 CB71, CB72 Sum 26°50'31.14"N/70°32'53.19"E 8.1 0.11 0.0069 8.7 09 CB58, CB62, CB69, CB70 Bikaner Bikaner 28° 1'49.04"N/73°15'30.63"E 7.9 0.21 0.0047 10.2 09 CB53, CB54 Deshnok 27°47'20.09"N/73°20'22.46"E 8.7 0.21 0.0117 11.9 10 CB13, CB14, CB17e Nokha 27°34'11.22"N/73°27'36.13"E 9.0 0.29 0.0092 11.2 08 CB59, CB60 [CB, Crotalaria burhia; Strains identified as eEnsifer sp., bBradyrhizobium sp. and rRhizobium sp. on the basis of 16S rRNA gene sequences] Sankhla et al.: CHARACTERIZATION OF MICROSYMBIONTS ASSOCIATED WITH CROTALARIA ROOT NODULES 375

post-monsoon season (July-October) in the years of Taq DNA polymerase, 1.2 μM of each of the 6 2011, 2012 and 2013. The Flora of the Indian Desert primers, 2 mM MgCl2, 150 μM of each dNTP and 1X and regional Botanical Survey of India were referred PCR buffer. The PCR temperature profiles were as for identification of C. burhia plants in the field. follows: initial denaturation at 94ºC for 5 min Whole plant were excavated with intact root system followed by 35 cycles of 94ºC for 30 s, 53ºC for 30 s, and then thoroughly washed with tap water. The 72ºC for 60 s and a final extension at 72ºC for 5 min. nodulation status of excavated plants was recorded Amplified PCR products along with 500 bp marker and 4-5 root nodules with roots were kept in moist were run on 0.89% (w/v) agarose gel, prepared in 1X soil and brought to the laboratory for preservation of TAE buffer and pre-stained with ethidium bromide. nodules and isolation of rhizobia. The viable seeds of The bands were visualized under BIO-RAD Gel Doc C. burhia were also collected to perform host System (Bio Rad SR+, USA Inc.). The amplified PCR authentication experiments. Root nodules of C. burhia products were quantified using Denovix (USA) DS11 were surface sterilized and bacterial strains were spectrophotometer (nanodrop). isolated and purified according to standard procedure as described by Vincent25 and Somasegaran and Molecular fingerprinting Hoben26. All bacterial strains were cultured on Congo Genetic variability among bacterial strains was Red-Yeast Extract Mannitol Agar (CR-YEMA) plates studied using following DNA fingerprinting patterns: and incubated at 28ºC. The CR-YEM agar plates were checked regularly to record colony characteristics of ARDRA (Amplified r-DNA Restriction Analysis) pattern isolates and to confirm their purity. The restriction endonucleases MspI (Genei Bangalore) was used to digest the amplified PCR Soil sampling and analysis products (16S rRNA gene). The reactions were carried Rhizosphere soil of C. burhia was collected from out in a final volume of 20 μL containing 2 μL of 10X each sampling sites at the time of nodule sampling for buffer, 0.25 μL of MspI enzyme (10U/μL), 10 μL chemical analysis. Chemical properties of the aliquots of PCR products and 7.75 μL of nuclease free collected soil such as pH (aqueous extract), total water. The reaction mixture was incubated at 37ºC for nitrogen (N), organic carbon (OC) and available overnight. Digested PCR products along with 100 bp phosphorus (P) were determined by using various DNA ladder (Genei Bangalore) were run on 2.0% standard methods described in Gehlot et al.8. agarose gel at 80 V for 1-2 h for separation of restricted fragments and were visualized by staining with ethidium Molecular characterization bromide using BIO-RAD Gel Doc System.

Isolation of genomic DNA RAPD (Random Amplification of Polymorphic DNA) pattern Purified bacterial strains were grown in YEM broth RPOI primer (nif gene directed primer widely used and were used for isolation of genomic DNA using 27 for study of genetic diversity) was used to randomly method described by Cheng and Jiang . Bacterial amplify genomic DNA as described by Richardson et al.29. cells were washed with TE (Tris-EDTA) and STE The reactions were carried out in a final volume of (Sodium chloride-Tris-EDTA) buffer, and lysed directly by adding phenol. The supernatant was Table 2 — List of primers used in this study for treated with chloroform to remove traces of phenol. molecular characterization The supernatant (containing DNA) was used as Primer Oligonucleotide sequence (5'→ 3')a Reference template for amplification of 16S rRNA and RPOI AATTTTCAAGCGTCGTGCCA 30 symbiotic (nodA and nifH) genes through thermo 18F AGAGTTTGATCCTGGCTCAG 29 cycler (BioRad T100). 1492R CTACGGCTACCTTGTTACG 29 800F GTAGTCCACGCCGTAAACGA 31 Amplification of 16S rRNA gene 820R CATCGTTTACGGCGTGGACT 31 Table 2 enlists various primers used in this study nodA1 TGCRGTGGAARNTRNNCTGGGAAA 32 for molecular characterization. Nearly full length 16S nodA2 GGNCCGTCRTCRAAWGTCARGTA 32 rRNA gene of selected bacterial strains was amplified nifHF TACGGNAARGGSGGNATCGGCAA 33 using universal primers (18F and 1492R)28. Each nifHI AGCATGTCYTCSAGYTCNTCCA 33 reaction was carried out in a final volume of 20 μL [aA, C, G, T =standard nucleotides; N=A, C, G or T; R=A or G; containing: 1 μL of template DNA (90 ng/μL), 0.6 U Y= C or T; S=G or C and W=A, T] 376 INDIAN J EXP BIOL, JUNE 2018

20 μL containing: 1.5 μL of template DNA (90 ng/μL), N- controls, respectively. The entire experimental set 1U of Taq DNA polymerase, 3 μM of RPOI primer, up of inoculated and controls plants were designed in 3.75 mM MgCl2, 150 μM of each dNTP and 1X PCR triplicates. Seedlings were grown under natural buffer. The thermal cycling condition were as sunlight and temperature (28±2ºC day and night) in follows: initial denaturation at 94ºC for 5 min glass house. Similar procedure was applied for followed by 5 cycles at 94ºC for 30 s, 50ºC for 60 s, V. radiata except scarification of seeds. Plants were 72ºC for 90 s and then 30 cycles at 94ºC for 30 s, harvested and checked for nodulation after 6-8 weeks 55ºC for 25 s and 72ºC for 90 s and a final extension of inoculation as described by Somasegaran and at 72ºC for 5 min. The amplified DNA fragments Hoben26. Rhizobia were re-isolated and purified from were resolved on 2.0% agarose gel at 80 V for 1-2 h root nodules of inoculated plants and compared with along with DNA ladder of 100 bp and generated the parental strains on the basis of colony morphology fingerprints were visualized under BIO-RAD Gel Doc and genetic fingerprints (RAPD pattern) using RPOI System. primer29.

Sequencing of 16S rRNA gene Amplification and sequencing of symbiotic genes (nodA and nifH) On the basis of ARDRA and RAPD patterns, 13 In selected isolates nodA and nifH genes were isolates were selected for sequencing of their 16S amplified. Primers nodA1 and nodA231 were used for rRNA gene. Amplified sample in adequate quantity PCR amplification of an internal fragment of size (50 μL having 20-30 ng/μL DNA) were sent to 650 bp of nodA gene (codes for N-acyl transferase Xcleris Genomics Labs Ltd., Ahmedabad for nodulation protein). The reactions were carried out in sequencing. Universal external (18F and 1492R)28 and a final volume of 20 μL containing: 1 μL of template internal (800F and 820R)30 primers were used to DNA, 0.6U of Taq DNA polymerase, 1 μM of each of obtain nearly full length (1.5 kb) nucleotides sequence the primers, 3.12 mM MgCl2, 150 μM of each dNTP, of 16S rRNA gene. The sequencing was done on and 1X PCR buffer. PCR cycling conditions were as Applied Biosystems platform using Big Dye version follows: initial denaturation at 94ºC for 5 min, 5 3.1 terminator and 5X buffer. Gene Tool Lite 1.0 cycles at 94ºC for 30 s, 55ºC for 30 s and 72ºC for (2000) software (Doubletwist, Inc., Oakland, CA, 60 s followed by 30 cycles at 94ºC for 30 s, 62ºC for USA) was used for analysis of all raw sequences. 45 s, 72ºC for 90 s and a final extension at 72ºC for Sequence similarity searches for nucleotide 7 min. Amplification of nifH region (750 bp codes for sequences were performed at the National Centre for Fe protein of nitrogenase enzyme) was carried out using primers nifHF and nifHI32 and the reaction Biotechnology Information (NCBI) server using basic local alignment search tool (BLAST). After mixture was prepared same as for nodA. PCR cycling completion of analysis, sequences were submitted to conditions were as follows: initial denaturation at 94ºC for 5 min followed by 25 cycles at 94ºC for 30 s, NCBI database using Sequin. 57ºC for 30 s, 72ºC for 30 s and a final extension at Nodulation test 72ºC for 7 min. PCR products along with DNA ladder Selective strains were evaluated for their ability to of 100 bp (Genei Bangalore) were run on 1.0% form nodules on their original host (C. burhia) as well agarose gel at 80 V for 1 h to confirm the as crop legume Vigna radiata. Seeds of C. burhia amplification of targeted gene. The amplified were scarified and surface sterilized by treatment with products were sequenced with the corresponding 95% (v/v) ethanol for 2 min and 0.1% HgCl2 for 3 min primers using an Applied Biosystems sequencer. followed by several washes with sterile distilled water and then allowed to germinate in petri plates, Phylogenetic analysis containing sterilized moist filter paper. Three to four Nucleotide sequences of RNB isolated from C. burhia germinated seedlings were planted aseptically into in this study were submitted in the GenBank database. each plastic pot containing washed and sterilized The accession numbers have been specified in the sand. Each seedling was inoculated with 1 mL corresponding phylograms. The multiple sequence (109cells/mL) of bacterial strains at exponential alignment program CLUSTALW33 was used to align growth phase. Un-inoculated seedlings either supplied the nucleotide sequences obtained in present study with mineral nitrogen (as 0.1% KNO3 in nutrient (16S rRNA and symbiotic genes sequences), together solution) or grown without nitrogen served as N+ and with related sequences of type strains retrieved from Sankhla et al.: CHARACTERIZATION OF MICROSYMBIONTS ASSOCIATED WITH CROTALARIA ROOT NODULES 377

the NCBI database. The phylogenetic trees were indeterminate and branched on maturity (Fig. 1F). generated using MEGA 6 software34 with maximum The morphology and internal structure of C. burhia likelihood method based on a GTR+G+I model. root nodules have been described in our previous A bootstrap for 1000 replicates was performed to study8. Soil characteristics such as total N, P, organic obtain the confidence values for the tree topologies. carbon and pH of various sampling sites were not significantly different as descried earlier8. The pH of Results soils in all sampling sites in Thar Desert was alkaline

Nodulation status and soil analysis and ranged from pH 7.9 to 9 (Table 1); however, soil Soil samples and nodules of C. burhia were texture, annual rainfall (100-500 mm), mean daily collected from all the sampling sites in five districts of temperature in summer (25-35ºC) and winter (7-15ºC) Western Rajasthan (Table 1). The C. burhia plants were found varied. growing in natural habitat is shown in Fig. 1A-C; with DNA fingerprinting (ARDRA and RAPD) and sequencing of flower (Fig. 1D); collected seeds (Fig. 1E) and root 16S rRNA gene nodules (Fig. 1F). The nodulation in C. burhia was A single band of approx. 1.5 kb of 16S rRNA gene observed at all sampling sites, although the average was amplified in all strains. The 51 rhizobial strains number of nodules per plant varied from site to site. formed four groups on the basis of the ARDRA The highest average number of nodules per plant was (Table 3). The largest ARDRA group-I had 42 strains recorded fourteen in Osian (Jodhpur) and Harima and group II had five strains which were identified as (Nagaur) sampling sites (semi-arid) and lowest was species of Ensifer and Bradyrhizobium respectively, five in Chohtan (Barmer) and Pokaran (Jaisalmer), on the basis of 16S rRNA gene sequencing and both arid to hyper arid sites (Table 1). Initially, BLASTn results. The group III and IV each nodules were globular but became elongated, containing 2 strains were identified as species of Rhizobium. All the isolates showed considerable variation in RAPD pattern and were distributed in 14 groups. The largest ARDRA group-I that comprised of 42 strains Table 3 — Grouping of selective Crotalaria burhia-RNB strains based on genetic fingerprinting RPOI genetic Strains ARDRA genetic groups groups I CB7, CB8, CB11e, CB14, CB15, I (42 strains) CB17e, CB27, CB28, CB30, CB40, CB42, CB72 II CB1, CB2, CB3, CB5e, CB44e III CB6e, CB36e, CB59 IV CB10, CB60, CB61, CB62, CB63, CB66 V CB54, CB55, CB56e, CB68 VI CB57, CB58, CB64, CB67, CB69 VII CB32e,CB70, CB71 VIII CB52, CB53 IX CB12e X CB20 XI CB4b, CB9, CB43, CB49, CB65 II (5 strains) XII CB29r III (2 strains) r XIII CB46 XIV CB31r, CB51 IV (2 strains) Fig. 1 — (A) The Crotalaria burhia plants growing in natural e b habitat during rainy season; (B) in dry season; (C) closer view; (D) [Strains identified as Ensifer sp., Bradyrhizobium sp. and r a flower; (E) collected seeds; and (F) indeterminate root nodules. Rhizobium sp. on the basis of 16S rRNA gene sequences] 378 INDIAN J EXP BIOL, JUNE 2018

(including nine sequenced Ensifer strains) was further (CB5, CB6, CB11, CB12, CB17, CB32, CB36, CB44 resolved into 10 RAPD groups (Table 3). This and CB56), three as species of Rhizobium (CB29, suggests that RPOI based fingerprinting gives better CB31 and CB46) and one strain (CB4) as species of resolution of genetic diversity when compared with Bradyrhizobium (Table 4). ARDRA. Strains CB4, CB9, CB43, CB49 and CB65 Authentication and host range showed similar banding patterns in both ARDRA and Five Ensifer strains (CB5, CB17, CB36, CB44 and RAPD. Based on ARDRA and RAPD profiles, 13 CB56) and a single Bradyrhizobium strain (CB4) representative strains (CB4, CB5, CB6, CB11, CB12, were found nodulating their original host C. burhia CB17, CB29, CB31, CB32, CB36, CB44, CB46 and and crop V. radiata, while remaining four Ensifer CB56) were selected and sequenced for their 16S strains (CB6, CB11, CB12 and CB32) and three rRNA gene. On the basis of BLASTn sequence Rhizobium strains (CB29, CB31 and CB46) failed to similarity search results for 16S rRNA gene out of 13 nodulate their original host as well as V. radiata. strains, nine were identified as species of Ensifer Phenotypically, nodulated plants appeared dark-green

Table 4 — Percentage sequence similarities of Crotalaria burhia-RNB strains based on 16S rRNA and symbiotic (nodA and nifH) genes with closest type/reference strain. Strains NCBI GenBank Closest type/reference strain Sequence Biological and geographical origin of accession (GenBank accession number) similarity closest type/reference strain number (%) 16S rRNA gene CB4 KJ871655 Bradyrhizobium yuanmingense CCBAU 10071T 99.8 Lespedeza cuneata, China (AF193818) CB5 KF938904 Ensifer kostiensis HAMBI 1489T (Z78203) 99.9 Senegalia senegal, Sudan CB6 JN832576 Ensifer terangae LMG 7834T (X68388) 99.7 Senegalia laeta, Senegal CB11 KF938905 Ensifer kostiensis HAMBI 1489T (Z78203) 99.7 Senegalia senegal, Sudan Ensifer saheli ORS 609T (X68390) Sesbania cannabina, Senegal CB12 KJ871656 Ensifer terangae LMG 7834T (X68388) 99.7 Senegalia laeta, Senegal CB17 KF938906 Ensifer kostiensis HAMBI 1489T (Z78203) 99.9 Senegalia senegal, Sudan CB29 KF938907 Rhizobium aegyptiacum 1010T (NR_137399) 100 Trifolium alexandrinum, Egypt Lens Rhizobium bangladeshense BLR175T (NR_137241) culinaris, Bangladesh Rhizobium binae BLR195T (NR_137242) Lens culinaris, Bangladesh CFN 42T (NR_029184) Phaseolus vulgaris, Mexico Rhizobium lentis BLR27T (NR_137243) Lens culinaris, Bangladesh CB31 KF938908 Rhizobium sullae IS 123T (NR_029330) 97.3 Hedysarum coronarium, Spain CB32 KF938909 Ensifer terangae LMG 7834T (X68388) 99.7 Senegalia laeta, Senegal CB36 KF938910 Ensifer kostiensis HAMBI 1489T (Z78203) 99.9 Senegalia senegal, Sudan CB44 KF938911 Ensifer kostiensis HAMBI 1489T (Z78203) 99.9 Senegalia senegal, Sudan CB46 KM044262 Rhizobium borbori DN316T (EF125187) 97 Activated sludge, China CB56 KF938912 Ensifer kostiensis HAMBI 1489T (Z78203) 99.9 Senegalia senegal, Sudan nodA gene CB4 KF437384 Bradyrhizobium yuanmingense CCBAU 10071T 97.1 Lespedeza cuneata, China (AM117557) CB56 KJ018181 Ensifer fredii USDA 205T (EU292000) 91.2 Glycine max, China Ensifer xinjiangensis CCBAU 110T (EF457965) Glycine max, China nifH gene CB4 KJ018170 Bradyrhizobium yuanmingense CCBAU 10071T 98.4 Lespedeza cuneata, China (EU818927) CB56 KJ018169 Ensifer fredii USDA257 (CP003565) 95.1 Glycine max, China Ensifer xinjiangensis CCBAU 110T (DQ411933) Glycine max, China Ensifer sojae CCBAU 05684T (GU994077) Glycine max, China Sankhla et al.: CHARACTERIZATION OF MICROSYMBIONTS ASSOCIATED WITH CROTALARIA ROOT NODULES 379

in comparison to control and non-nodulated plants. (99.8%). In addition, the strain CB4 also showed After recording the nodulation status 2-3 re-isolates 99.9% sequence similarity with B. yuanmingense were successfully isolated and purified from TF17 isolated from root nodules of Tephrosia excavated root nodules of each host (C. burhia and falciformis from Thar Desert (Fig. 3). The three non- V. radiata) plant. The colony characteristics and nodulating Rhizobium strains (CB29, CB31 and RAPD patterns (using RPOI primer) of re-isolates were exactly similar to corresponding inoculated parental strains (Fig. 2). The results of host authentication, suggest that C. burhia is nodulated by both, species of Ensifer and Bradyrhizobium.

Sequencing of symbiotic genes Symbiotic (nodA and nifH) genes of selective nodulating isolates CB56 (Ensifer strain) and CB4 (Bradyrhizobium strain) were successfully amplified Fig. 2 — The RPOI-DNA fingerprinting gel images of nodulating and sequenced. The symbiotic genes could not be parental (P) Crotalaria burhia (CB)-Ensifer strains compared amplified in the case of non-nodulating Rhizobium with re-isolates from host Crotalaria burhia (R1) and Vigna radiata (R2 and R3). (CB29, CB31 and CB46) and Ensifer (CB6, CB11, CB12 and CB32) strains, although various primer sets were used (Table 2).

Phylogenetic studies

16S rRNA gene phylogeny In phylogenetic analyses, the Ensifer strains formed three 16S rRNA types clustering into separate clades and a single lineage (Fig. 3). The first clade (16S rRNA type I) comprised by five nodulating strains (CB5, CB17, CB36, CB44 and CB56) showing 100% sequence similarities with Ensifer sp. TF7, TW10, RA9 and PC2 isolated from native legumes of the Thar Desert as well as with Ensifer sp. K15 (LK936546) and Vr38 (LN851899) isolated from chickpea and mung bean nodules, respectively, from Pakistan. These five strains were very close to E. kostiensis HAMBI 1489T (99.9% sequence similarities) as well as to E. saheli LMG 7837T (=ORS 609T) with 99.7% sequence similarities. The strain CB11 alone formed a separate lineage (16S rRNA type II) and showed 99.7% sequence similarity with type strains E. kostiensis HAMBI 1489T and E. saheli LMG 7837T. The second clade (16S rRNA type III) comprised of 3 strains (CB6, CB12 and

CB32) showing 99.7% sequence similarities to Ensifer sp. E60 (isolated from acacias in Algeria), and Fig. 3 — Phylogenetic tree constructed using 16S rRNA gene T sequences of Ensifer and Bradyrhizobium strains isolated from the type strain E. terangae LMG 7834 (Fig. 3 and Table 4). Indian native legume Crotalaria burhia together with those of The phylogenetic analyses of Bradyrhizobium strain type strains and close relatives. [The tree was built using a CB4 revealed that it had 100% sequence similarity to Maximum Likelihood (ML) method and bootstrap values strains PRNB-26, M11 and GX5 isolated from calculated for 1000 replications are indicated at internodes. The Pongamia pinnata (India), V. radiata (Nepal) and scale bar indicates 2% substitutions per site. Accession numbers from GenBank are in parenthesis. (Abbreviations: B, Bradyrhizobium; V. radiata (China), respectively and was adjacent to CB, Crotalaria burhia; E, Ensifer; JNVU, Jai Narain Vyas T type strain B. yuanmingense CCBAU 10071 University and T indicates type strain)] 380 INDIAN J EXP BIOL, JUNE 2018

CB46) formed three distinct types (Fig. 4). The strain sequence similarity with E. xinjiangensis CCBAU CB29 showed 100% similarity with following ty 110T and E. fredii USDA 205T forming a distinct pe strains: R. aegyptiacum 1010T, R. bangladeshense lineage (Fig. 5 and Table 4). The nodA gene of BLR175T, R. binae BLR195T, R. etli CFN 42T and Bradyrhizobium strain CB4 clustered together with R. lentis BLR27T. Strain CB31 showed 97.3% NGB-SR15 isolated from soybean root nodules similarity with R. sullae IS 123T and strain CB46 was from Egypt and showed 97.1% sequence similarity close to R. borbori DN316T with 97% similarity. to B. yuanmingense CCBAU 10071T. The phylogeny of nifH gene was consistent with Symbiotic gene phylogeny that of nodA gene. Like nodA gene phylogeny, the The symbiotic genes of two strains CB56 and nifH gene of Ensifer sp. CB56 showed highest CB4 each representing group of other similar sequence similarity (95.1%) with E. fredii USDA nodulating strains of Ensifer and Bradyrhizobium, 257, E. sojae CCBAU 05684T and E. xinjiangensis respectively, were successfully amplified and CCBAU 110T (Fig. 6A and Table 4). Like 16S sequenced to study their phylogeny. Our results rRNA and nodA phylogeny, the nifH gene of showed that the phylogeny of nodA gene of CB4 Bradyrhizobium sp. CB4 was adjacent to B. was congruent with its species phylogeny (close to yuanmingense CCBAU 10071T and showed 98.4% B. yuanmingense based on 16S rRNA gene). While sequence similarity. The strain CB4 also showed such congruence was not seen in the case of strain 99.4% sequence similarity with strains such as CB56 (close to E. kostiensis based on 16S rRNA TF17, D1 and SR42 isolated form T. falciformis, gene). The nodA gene of CB56 showed 91.2% Glycine max and V. radiata, respectively (Fig. 6B).

Fig. 4 — Phylogenetic tree constructed using 16S rRNA gene Fig. 5 — Phylogenetic tree constructed using nodA gene sequences of non-nodulating Rhizobium strains isolated from sequences of Ensifer and Bradyrhizobium strains isolated the Indian native legume Crotalaria burhia together with from the Indian native legume Crotalaria burhia together those of type strains. [The tree was built using a Maximum with those of type strains. [The tree was built using a Likelihood (ML) method and bootstrap values calculated for Maximum Likelihood (ML) method and bootstrap values 1000 replications are indicated at internodes. The scale bar calculated for 1000 replications are indicated at internodes. indicates 1% substitutions per site. Accession numbers from The scale bar indicates 10% substitutions per site. Accession GenBank are in parenthesis. (Abbreviations: CB, Crotalaria numbers from GenBank are in parenthesis. (Abbreviations: burhia; JNVU, Jai Narain Vyas University; NR, NCBI B, Bradyrhizobium; CB, Crotalaria burhia; E, Ensifer; Reference sequence; R, Rhizobium and T indicates type JNVU, Jai Narain Vyas University and T indicates type strain)] strain)] Sankhla et al.: CHARACTERIZATION OF MICROSYMBIONTS ASSOCIATED WITH CROTALARIA ROOT NODULES 381

showed that C. burhia is widely distributed throughout Thar Desert in rain-fed open area and on marginal lands that indicates its well adaptation to harsh climatic conditions of Thar Desert. The plants of C. burhia were found nodulating at all sampling sites although the number of nodules per plant varied from site to site. The average number of nodules per plant was poor at the sampling sites of arid and hyper- arid regions (Barmer and Jaisalmer), where soil was more sandy, poor in texture as compared to other sampling sites (semi-arid areas). Our results are in accordance with the previous studies35 that edapho- climatic conditions such as rainfall, salinity, pH and temperature may affect the number of effective root nodule bacteria (RNB) in the soil that establish functional nitrogen-fixing symbiosis and therefore number of nodules per plant varied. Molecular techniques such as ARDRA and RAPD have been extensively used in rhizobial ecology for genetic grouping and identification of strains. In the present investigation, our results on the basis of DNA fingerprinting indicates that fast growing Ensifer strains are genetically more diverse than the slow growing Bradyrhizobium strains. The more Fig. 6 — Phylogenetic tree constructed using nifH gene sequences occupancy of Ensifer strains compared to other of Ensifer (A); and Bradyrhizobium (B) strains isolated from the rhizobial (Bradyrhizobium and Rhizobium) strains in root Indian native legume Crotalaria burhia together with those of nodules of C. burhia indicates that Ensifer species are type/reference strains. [The trees were built using a Maximum Likelihood (ML) method and bootstrap values calculated for 1000 well adapted to alkaline soils of Thar Desert and is replications are indicated at internodes. The scale bar indicates dominant microsymbiont of C. burhia and other native 2% substitutions per site. Accession numbers from GenBank are legumes in the Thar Desert8,9,12,13,15,16,17,19. There are in parenthesis. (Abbreviations: B, Bradyrhizobium; CB, several reports8,9,12,13,15-19, from different legumes Crotalaria burhia; E, Ensifer; JNVU, Jai Narain Vyas University suggesting that alkaline soil and arid conditions and T indicates type strain)] favours more incidences of Ensifer species in root nodules rather than slow-growing species of Discussion Bradyrhizobium. Genomic studies done by Tian Like other native legumes, such as Acacia and et al.36 also strengthen such assumptions and reported Prosopis, the C. burhia is also well adapted and that several gene clusters involved in osmoregulation nodulates in the nutritionally poor alkaline soil of and adaptation to alkaline pH are present in the Thar Desert of India. Analysis of soil from the various species of Ensifer as compared to Bradyrhizobium. sampling sites in present investigation supports our The nine Ensifer strains sequenced in the present 8 previous report and further confirms that soil in Thar study were classified into three types on the basis of Desert is alkaline throughout and poor in total N and phylogenetic analysis of 16S rRNA gene, which were available P. Although the soil is not significantly mainly close to one of the three type strains differing in nutritional characters at various sites but E. kostiensis HAMBI 1489T (isolated from root the sampling sites in the different districts belongs to nodules of Senegalia senegal, Sudan)37, E. saheli semi-arid and hyper arid regions and varied in ORS609T (isolated from Sesbania cannabina, average annual rainfall and mean temperature in Senegal)38 and E. terangae LMG 7834T (isolated from summer and winter. The texture is also different and Senegalia laeta, Senegal)38; and also have close purely sandy in Barmer, Bikaner and Jaisalmer similarity with Ensifer strains isolated from other districts as compared to other sites. The field survey native legumes (Tephrosia spp., Rhynchosia aurea 382 INDIAN J EXP BIOL, JUNE 2018

and Prosopis cineraria) of the Thar Desert8,13 as well originally nodulating, but may have lost the ability to as with Ensifer strains from Pakistan, which indicate nodulate or the related nod genes during the their Asiatic origin. On the other hand, strain CB4 subculturing. Surprisingly, some strains of Ensifer showed similarity with Bradyrhizobium sp. isolated (CB6, CB11, CB12 and CB32) belonging to 16S from Tephrosia species from the semi-arid regions of rRNA type II and III failed to nodulate their host western Rajasthan8. Strain CB4 also showed close C. burhia. This lack of nodulation efficiency can be similarity with B. yuanmingense CCBAU 10071T explained by assumptions, such as (i) these strains of (isolated from the root nodules of Lespedeza cuneata Ensifer may have opportunistically entered along with from China)39 but formed a separate lineage. In other nodulating rhizobia; and (ii) these strains might addition, B. yuanmingense type of strains has also have lost their symbiotic traits during symbiosis or been isolated in India from the root nodules of once purified onto artificial YEMA media41. soybean from different agricultural-ecological- 32,42 40 Similar to other studies , the phylogeny based on climatic regions of India . The first report of isolation symbiotic genes (nodA and nifH) of Ensifer sp. CB56 of B. yuanmingense is from wild legume L. cuneata 39 was incongruent with its 16S rRNA gene phylogeny. from China and the same is the case in our present Strain CB56 (close to E. kostiensis HAMBI 1489T in study, and the previous reports from our group where 16S rRNA gene phylogeny) was closer to E. fredii such strains have been isolated from wild native USDA 205T43 and E. xinjiangensis CCBAU 110T44 in legumes (C. burhia, Tephrosia spp. and Vachellia 8,16 its symbiotic gene phylogeny. Such incongruence leucophloea) of Thar Desert where no soybean suggests that the sym genes in it have been acquired cultivation history is found. This suggests that from other symbiotic Ensifer species/strains present in B. yuanmingense is important unexplored microbial the soil through horizontal gene transfer (HGT). In resource of India that has to be further studied for its contrary, Bradyrhizobium sp. CB4 was close to wide host range. The occurrence of non-nodulating B. yuanmingense CCBAU 10071T in both 16S rRNA Rhizobium strains in the present investigation is and symbiotic gene phylogenies. interesting as these non-nodulating strains showed close similarity with few nodulating Rhizobium Conclusion strains, such as Phaseolus vulgaris microsymbiont Crotalaria burhia is nodulated by both, the fast (R. etli) from Mexico; R. aegyptiacum (isolated from growing species of Ensifer that has close similarity root nodules of Trifolium alexandrinum, Egypt) and with the Old World Ensifer strains (E. kostiensis, three type strains (R. bangladeshense, R. binae and E. saheli and E. terangae) and the slow growing R. lentis) isolated from the root nodules of Lens species of Bradyrhizobium close to B. yuanmingense. culinaris, Bangladesh. A single strain CB46 was close The Ensifer strains in the present study are genetically to non-nodulating R. borbori isolated from activated diverse as compared to Bradyrhizobium strains and sludge, China. both type of strains cross-nodulated crop V. radiata. The result of nodulation test suggests that C. The genetic diversity and incongruence between burhia is nodulated by both Ensifer and species (16S rRNA) and symbiotic gene phylogeny is Bradyrhizobium strains but not by species of the result of HGT due to stressful conditions Rhizobium. Interestingly CB-Ensifer and CB- prevailing in Thar Desert of India. Few strains in the Bradyrhizobium strains are nodulating crop legume V. present study are phylogenetically divergent from the radiata. This is the first report on molecular existing type strains which indicate the presence of characterization of N fixing microsymbiont of C. potential novel species of rhizobia in the desert burhia, an important native legume and a good soil region. Occurrence of non-nodulating Ensifer and binder of arid regions of the Thar Desert. The Rhizobium strains suggests that root nodules harbours isolation of non-nodulating Rhizobium from root both symbiotic as well as opportunistic bacteria. nodules of native legumes in our case is similar to Furthermore polyphasic approach including multi other researchers who isolated non-nodulating locus sequence analysis (MLSA) of conserved Rhizobium strains from a wide range of plant taxa. protein-coding housekeeping genes, DNA-DNA The symbiotic genes could not be amplified in non- hybridization, BIOLOG, FAME analysis and whole nodulating Rhizobium strains. There could be a genome sequencing will be required for the possibility that these Rhizobium strains were description of a formal new species of rhizobia Sankhla et al.: CHARACTERIZATION OF MICROSYMBIONTS ASSOCIATED WITH CROTALARIA ROOT NODULES 383

nodulating C. burhia in the Thar Desert of India. Genome sequence of Ensifer sp. TW10; a Tephrosia Host-range of these effective N fixing strains needs to wallichii (Biyani) microsymbiont native to the Indian Thar Desert. Stand Genomic Sci, 9 (2013) 304. be studied for using them as inoculums in future. 11 Panwar D, Tak N & Gehlot HS, Nodulated Native Legumes Such basic studies on identification, in an Arid Environment of Indian Thar Desert. In: MH characterization, screening of broad host range and Fulekar & RK Kale (eds.) Recent Trends in Plant Sciences efficient native desert rhizobia will be useful in (IK International Publishing House Pvt. Ltd. New Delhi, preparing consortium of agriculturally important India), 2014, 284. 12 Sankhla IS, Meghwal RR, Tak N, Tak A & Gehlot HS, rhizobial strains and will also form a platform for Phenotypic and molecular characterization of more detailed genomic studies to understand the microsymbionts associated with Crotalaria medicagenia: a mechanism of adaptability to changing environment native legume of the Indian Thar Desert. Plant Archives, 15 and spread of aridity. (2015) 1003. 13 Tak N, Awasthi E, Bissa G, Meghwal RR, James EK, Sprent JS & Gehlot HS, Multi locus sequence analysis and Acknowledgement symbiotic characterization of novel Ensifer strains nodulating Author Indu S. Sankhla acknowledges the Tephrosia spp. in the Indian Thar Desert. Syst Appl University Grants Commission (UGC), New Delhi, Microbio, 39 (2016) 534. for junior and senior research fellowships. The work 14 Gehlot HS, Ardley J, Tak N, Tian R, Poonar N, Meghwal RR, Rathi S, Tiwari R, Adnawani W, Seshadri R & was also supported by grants from UGC-SAPII-CAS- Reddy TB, High-quality permanent draft genome sequence I, UGC-BSR Start-Up-Grant (F.30-16/2014) and the of Ensifer sp. PC2, isolated from a nitrogen-fixing root Department of Biotechnology, Govt. of India nodule of the legume tree (Khejri) native to the Thar Desert (BT/PR11461/AGR/21/270/2008). of India. Stand Genomic Sci, 11 (2016) 43. 15 Sankhla IS, Tak N, Meghwal RR, Choudhary S, Tak A, Rathi S, Sprent JI, James EK & Gehlot HS, Molecular References characterization of nitrogen fixing microsymbionts from root 1 Sprent J, Legume nodulation: A Global Perspective, (John nodules of Vachellia (Acacia) jacquemontii, a native legume Wiley & Sons, Inc), 2009, 65. from the Thar Desert of India. Plant Soil, 410 (2017) 21. 2 Peix A, Ramírez-Bahena MH, Velázquez E & Bedmar EJ, 16 Choudhary S, Meghwal RR, Sankhla IS, Tak N & Gehlot HS, Bacterial associations with legumes. Crit Rev Plant Sci, 34 Molecular characterization and phylogeny of novel diverse (2015) 17. nitrogen fixing microsymbionts associated with Vachellia 3 Lewis GP, Schrire B, Mackinder B & Lock M, Legumes of (Acacia) leucophloea in arid and semiarid regions of the world, (Kew Royal Botanic Gardens, Kew) 2005. Rajasthan. Indian Forester, 143 (2017) 266. 4 Ansari AA, Crotalaria L. in India, (Bishen Singh Mahendra 17 Rathi S, Gaur S, Tak N, Tak A & Gehlot HS, Genetically Pal Singh, Dehra Dun), 2008, 52. diverse root nodule bacteria associated with Alysicarpus 5 Ranjith M, Bajya DR & Manoharan T, Field study on vaginalis from alkaline soil of Rajasthan, India. Plant repellent efficacy of Crotalaria burhia Buch.-Ham. ex Archives, 17 (2017) 495. Benth. and Anacardium occidentale L. against Odontotermes 18 Le Quéré A, Tak N, Gehlot HS, Lavire C, Meyer T, Chapulliot D, obesus (Rambur). Indian J Nat Prod Resour, 6 (2015) 288. Rathi S, Sakrouhi I, Rocha G, Rohmer M & Severac D, 6 Bhandari MM, Flora of the Indian Desert, (MPS Report, Genomic characterization of Ensifer aridi, a proposed new Jodhpur), 1990, 103. species of nitrogen-fixing rhizobium recovered from Asian, 7 Gehlot HS, Tak N, Dagla HR & Davis TD, Indigenous and African and American deserts. BMC Genomics, 18 (2017) 85. modern scientific strategies for characterization, conservation 19 Choudhary S, Tak N & Gehlot HS, Phylogeny and genetic and sustainable utilization of bio-resources of the Indian Thar diversity assessment of Ensifer strains nodulating Senegalia Desert. J Arid Land Stud, 24 (2014) 5. (Acacia) senegal (L.) Britton. in arid regions of Western 8 Gehlot HS, Panwar D, Tak N, Tak A, Sankhla IS, Poonar N, Rajasthan, India. Microbiology, 87 (2018) 127. Parihar R, Shekhawat NS, Kumar M, Tiwari R, Ardley J, 20 Samba RT, de Lajudie P, Gillis M, Neyra M, Spencer-Baretto James EK & Sprent JI, Nodulation of legumes from the Thar MM & Dreyfus B, Diversity of rhizobia nodulating Crotalaria desert of India and molecular characterization of their spp. from Senegal. Symbiosis, 27 (1999) 259. rhizobia. Plant Soil, 357 (2012) 227. 21 Jourand P, Giraud E, Bena G, Sy A, Willems A, Gillis M, 9 Gehlot HS, Tak N, Kaushik M, Mitra S, Chen WM, Poweleit N, Dreyfus B & de Lajudie P, Methylobacterium nodulans sp. Panwar D, Poonar N, Parihar R, Tak A, Sankhla IS, Ojha A, nov., for a group of aerobic, facultatively methylotrophic, Rao SR, Simon MF, dos Reis Junior FB, Perigolo N, legume root-nodule forming and nitrogen-fixing bacteria. Int Tripathi AK, Sprent JI, Young JPW, James EK & J Syst Evol Microbiol, 54 (2004) 2269. Gyaneshwar P, An invasive Mimosa in India does not adopt 22 Liu XY, Wang ET, Li Y & Chen WX, Diverse bacteria the symbionts of its native relatives. Ann Bot, 112 (2013) 179. isolated from root nodules of Trifolium, Crotalaria and 10 Tak N, Gehlot HS, Kaushik M, Choudhary S, Tiwari R, Mimosa grown in the subtropical regions of China. Arch Tian R, Hill Y, Bräu L, Goodwin L, Han J, Liolios K, Microbiol, 188 (2007) 1. Huntemann M, Palaniappan K, Pati A, Mavromatis K, 23 Rocha AL, Isolation and characterization of bacterial symbionts Ivanova N, Markowitz V, Woyke T, Kyrpides N & Reeve W, from Crotalaria spectabilis grown on trichloroethene 384 INDIAN J EXP BIOL, JUNE 2018

contaminated soil, Dissertation, Missouri University of Science 36 Tian CF, Zhou YJ, Zhang YM, Li QQ, Zhang YZ, Li DF, and Technology, Missouri, United States, 2011. Wang S, Wang J, Gilbert LB, Li YR & Chen WX, 24 Singha B, Mazumder PB & Pandey P, Characterization of Comparative genomics of rhizobia nodulating soybean plant growth promoting rhizobia from root nodule of suggests extensive recruitment of lineage-specific genes in Crotalaria pallida grown in Assam. Indian J Biotechnol, 15 adaptations. Proc Natl Acad Sci, 109 (2012) 8629. (2016) 210. 37 Nick G, de Lajudie P, Eardly B, Soumalainen S, Paulin L, 25 Vincent JM, A manual for the practical study of root-nodule Zhang X, Gillis M & Lindstrom K, Sinorhizobium arboris bacteria, (Blackwell Scientific Publications, Oxford), 1970. sp. nov., and Sinorhizobium kostiense sp. nov., isolated from 26 Somasegaran P & Hoben HJ, The handbook for Rhizobia: leguminous trees in Sudan and Kenya. Int J Syst Bacteriol, methods in legume Rhizobia technology, (Springer Verlag, 49 (1999) 1359. New York), 1994. 38 de Lajudie P, Willems A, Pot B, Dewettinck D, Maestrojuan G, 27 Cheng HR & Jiang N, Extremely rapid extraction of DNA Neyra M, Collins MD, Dreyfus B, Kersters K & Gillis M, from bacteria and yeasts. Biotechnol lett, 28 (2006) 55. Polyphasic taxonomy of Rhizobia: Emendation of the genus 28 Weisburg WG, Barns SM, Pelletier DA & Lane DJ, 16S Sinorhizobium and description of Sinorhizobium meliloti ribosomal DNA amplification for phylogenetic study. J comb. nov., Sinorhizobium saheli sp. nov., and Bacteriol, 173 (1991) 697. Sinorhizobium teranga sp. nov. Int J Syst Bacteriol, 44 29 Richardson AE, Viccars LA, Watson JM & Gibson AH, (1994) 715. Differentiation of Rhizobium strains using the polymerase 39 Yao ZY, Kan FL, Wang ET, Wei GH & Chen WX, chain reaction with random and directed primers. Soil Biol Characterization of rhizobia that nodulate legume species of the Biochem, 27 (1995) 515. genus Lespedeza and description of Bradyrhizobium 30 Yanagi M & Yamasato K, Phylogenetic analysis of the yuanmingense sp. nov. Int J Syst Evol Microbiol, 52 family and related bacteria by sequencing of (2002) 2219. 16S rRNA gene using PCR and DNA sequencer. FEMS 40 Appunu C, N’Zoue A & Laguerre G, Genetic diversity of Microbiol Lett, 107 (1993) 115. native bradyrhizobia isolated from soybeans (Glycine max L.) in 31 Haukka K, Lindstrom K & Young JPW, Three phylogenetic different agricultural-ecological-climatic regions of India. groups and nodA and nifH genes in Sinorhizobium and Appl Environ Microbiol, 74 (2008) 5991. Mesorhizobium isolates from leguminous trees growing in 41 Brom S, Garcia-de los Santos A, Stepkowsky T, Flores M, Africa and Latin America. Appl Environ Microbiol, 64 Davila G, Romero D & Palacios R, Different plasmids of (1998) 419. Rhizobium leguminosarum bv. phaseoli are required for 32 Laguerre G, Nour SM, Macheret V, Sanjuan J, Drouin P & optimal symbiotic performance. J Bacteriol, 174 (1992) Amarger N, Classification of rhizobia based on nodC and 5183. nifH gene analysis reveals a close phylogenetic relationship 42 Vinuesa P, Silva C, Werner D & Martinez-Romero E, Population among Phaseolus vulgaris symbionts. Microbiology, 147 genetics and phylogenetic inference in bacterial molecular (2001) 981. systematics: the roles of migration and recombination 33 Thompson JD, Higgins DG & Gibson TJ, CLUSTAL W: in Bradyrhizobium species cohesion and delineation. Mol improving the sensitivity of progressive multiple sequence Phylogenet Evol, 34 (2005) 29. alignment through sequence weighting, position-specific gap 43 Chen WX, Yan GH & Li JL, Numerical taxonomic study of penalties and weight matrix choice. Nucleic Acids Res, 22 fast-growing soybean rhizobia and a proposal that Rhizobium (1994) 4673. fredii be assigned to Sinorhizobium gen. nov. Int J Syst 34 Tamura K, Stecher G, Peterson D, Filipski A & Kumar S, Bacteriol, 38 (1988) 392. MEGA6: Molecular Evolutionary Genetics Analysis Version 44 Peng GX, Tan ZY, Wang ET, Reinhold-Hurek B, Chen WF 6.0. Mol Biol Evol, 30 (2013) 2725. & Chen WX, Identification of isolates from soybean nodules 35 Zahran HH, Rhizobium–legume symbiosis and nitrogen in Xinjiang Region as Sinorhizobium xinjiangense and fixation under severe conditions and in an arid climate. genetic differentiation of S. xinjiangense from Sinorhizobium Microbiol Mol Biol Rev, 63 (1999) 968. fredii. Int J Syst Evol Microbiol, 52 (2002) 457.