Genetic Diversity and Symbiotic Efficiency of Nodulating Rhizobia Isolated from Root Nodules of Faba Bean in One Field.

Thirty-one nodulating rhizobium strains were collected from root nodules of spring and winter type faba bean cultivars grown in micro ecoarea, i.e. the same field in Chengdu plain, China. The symbiotic efficiency and phylogeny of these strains were studied. Effectively nitrogen fixing strains were isolated from both winter type and spring type cultivars. Based on phylogenetic analysis of 16S rRNA gene and concatenated sequence of atpD, glnII and recA genes, the isolates were assigned as Rhizobium anhuiense and a potential new Rhizobium species. The isolates were diverse on symbiosis related gene level, carrying five, four and three variants of nifH, nodC and nodD, respectively. Strains carrying similar gene combinations were trapped by both winter and spring cultivars, disagreeing with the specificity of symbiotic genotypes to reported earlier faba bean ecotypes.

Introduction

Biological nitrogen fixation (BNF) is important for agriculture worldwide. Legume-rhizobia symbiosis accounts 60% of the total BNF [1, 2]. In the symbiosis, rhizobia form nodules on the roots or stems of the host plant. Rhizobia reduce atmospheric dinitrogen into ammonia inside the nodules.

Faba bean is grown worldwide as a source of protein and starch [3]. In 2013, the biggest faba bean producing countries were China, Australia, France and Egypt [4]. Besides yielding food and fodder, faba bean provides assimilated nitrogen to other crops grown in crop rotation [5]. Faba bean are divided into three main ecotypes: winter faba bean that is sown in autumn and grown in Mediterranean and the south of China, spring faba bean that is sown in spring and grown mostly in Europe and the north of China, and the Chinese Yunnan ecotype that is sown in both autumn and spring [6, 7]. In symbiosis, faba bean is considered as a selective legume. Previous reports indicated that the isolations from faba bean root nodules contain Rhizobium leguminosarum symbiovar viciae and trifolii, R. etli, R. fabae, R. laguerreae, R. mesosinicum, R. anhuiense and Agrobacterium tumefaciens [8-14]. The symbiotic genotype of the rhizobia determines the success in nodulating faba bean [15]. The faba bean ecotypes have been proposed to determine the distribution of rhizobial genomic and nodC-nodD types [9]. In addition, based on the phylogenetic analysis of nodD gene, faba bean symbionts were divided into groups roughly related to the division of winter and spring faba bean ecotypes [2].

In addition to the host cultivars, environmental conditions may affect the diversity of nodulating strains. Earlier we assessed the diversity of faba bean nodulating rhizobia on 21 sites with different environmental conditions across the Sichuan hilly areas, China [11]. For an insight into local diversity under uniform environmental conditions, and faba bean cultivars introduction experiment was coincidentally performed on one field in Sichuan agricultural university farm in Chengdu, so we collected nodules from fifteen faba bean cultivars grown on the same field in Chengdu, China, isolated nodulating rhizobia, and assessed their N2-fixation ability and phylogeny.

Materials and Methods

Isolation of the strains

Nodules were collected from fifteen faba bean cultivars grown on the research field of Sichuan Agricultural University in Chongzhou, Chengdu, Sichuan in 2013. Pink nodules were selected for isolation. The soil in the field was paddy soil with pH 6.3 and contained organic matter 37.6 g kg-1, total N 2.0 g kg-1, available N 136.0 mg kg-1, Olsen-P 20.4 mg kg-1, and exchangeable K 101.0 mg kg-1. After surface sterilization as described by Xu et al. [16], nodules were crushed and inoculated on YMA (yeast mannitol agar) medium supplemented with 25 mg L-1 congo red as described [17]. Isolates were purified by streaking on YMA medium and incubated at 28°C. Purified strains were maintained on YMA at 4°C for temporary storage and in 25% glycerol at -80°C for long-term storage.

Plant nodulation and symbiotic efficiency test

The plant nodulation and symbiotic efficiency test was carried out using hydroponically grown native winter type cultivar Hanyuan Dabaidou. The seeds were surface sterilized with 95% ethanol (5 min) and 0.2% HgCl2 (3 min) followed by washing seven times with sterilized distilled water (5 min per time). Seeds were soaked overnight in sterilized water to soften the thick and hard seed coat. Sprouting, transplanting and inoculating were done as described previously by Xu et al. [16]. After 50 days, the plants were harvested and shoot dry mass and nodule numbers were measured. Excel 2010 (Microsoft, Redmond, USA) and SPSS 17.0 (SPSS Inc., Chicago, USA) were used to calculate the one-way analysis of variance with a least significant difference (LSD) analysis (P ≤ 0.05) of the mean values.

Bacterial DNA extraction and PCR-RFLP of 16S rRNA gene and intergenic spacer region (IGS)

DNA was extracted as described by Little [18]. Primer pair P1 and P6 was used for amplification of 16S rRNA gene [11]. IGS was amplified using primer pair pHr (F) and p23SR01(R) [11]. Target sequences were amplified in Bio-RAD MyCyclerTM with 20 pmol of each primer pair and 50 ng total DNA as template using a Golden Easy PCR System (TIANGEN, Beijing, China). The PCR protocols were as described by Xu et al. [16], except that the annealing temperature was 60°C for IGS.

16S rRNA gene and IGS PCR-RFLP were done using restriction endonucleases TaqI, HaeIII, HinfI and MspI. 5 μl of PCR product was digested by the restriction endonucleases separately in a 10 μl reaction volume following the manufacturer’s instructions (Fermentas, EU). The fragments were separated in 2% agarose gel containing 0.5 μg ml-1 ethidium bromide at 80 V for 3 hours in 1×TAE (Tris-Acetate-EDTA) buffer. Gels were documented with a Gel Document System (Bio-rad, USA). 16S RFLP and IGS RFLP analyses were done using the UPGMA clustering algorithm of NTSYSpc program [19].

Sequence analyses

Eight isolates were selected for sequencing of housekeeping and symbiosis related genes. 16S rRNA gene, atpD, glnII, recA, nifH, nodC and nodD were amplified using the primer pairs P1/P6, atpDF3/atpDR [20], glnII5/glnII6 [21], recAF3/recAR3 [20], nifHctg/nifHR [22], nodCF/nodCR [23] and NBA12F/NBA12R [24], respectively, using the related protocols. The PCR products were directly sequenced at BGI Tech (Shenzhen, China). The sequences were compared with sequences in the NCBI database using BlastN to find reference sequences. The reference sequences and the sequences from the representative strains were aligned using ClustalW in MEGA 6.0 [25]. Phylogenetic trees were built using Neighbor-joining (NJ) method with 1000 resampling bootstrapping in MEGA 6.0. The percentage similarity of the genes was estimated using Distance in MEGA 6.0 [16]. In the multilocus sequence analysis of the concatenated housekeeping genes atpD, recA and glnII, we applied 97% similarity as the threshold for defining genospecies [26, 27].

Results

Isolation and symbiotic efficiency

Rhizobial bacteria were isolated from four spring type and eleven winter type faba bean cultivars grown in a single field in Chengdu plain (Table 1). The isolates formed nodules on the root of the native faba bean Hanyuan Dabaidou with the average nodule numbers ranging from 19 to 110 per plant. Compared to the uninoculated control, nine isolates increased significantly (P ≤ 0.05) the shoot dry mass of the plant, and were considered as potential inoculant strains (Table 1). The potential inoculant strains were isolated from three local Sichuan winter type faba bean cultivars and three spring type faba bean cultivars.

Table 1

Rhizobial strains isolated in this study and their symbiotic and genotypic characteristics.

Straina	Host cultivarb	Cultivar origin	Cultivar ecotype	Shoot dry mass	No. of nodules	16S-RFLP genotyped	IGS-RFLP genotypee
				(g plant-1)c	per plant
SCAUf68	DB	Sichuan	Winter	0.924±0.0724	53	A	XI
SCAUf65	HN	Henan	Spring	1.049±0.0621↑*	36.7	A	XI
SCAUf64	HN	Henan	Spring	0.696±0.0214	62	A	VII
SCAUf73	JX	Shanxi	Spring	1.176±0.0922↑*	52.7	A	X
SCAUf76	JX	Shanxi	Spring	1.019±0.0643↑*	64.3	A	XII
SCAUf75	JX	Shanxi	Spring	0.849±0.144	40.7	A	XII
SCAUf74	JX	Shanxi	Spring	0.844±0.0767	22	A	XI
SCAUf70	LP	Shandong	Spring	1.073±0.0860↑*	19.3	A	XI
SCAUf72	LP	Shandong	Spring	0.962±0.0656	54.7	A	XI
SCAUf71	LP	Shandong	Spring	0.685±0.0447	98.3	A	IX
SCAUf55	S1	Sichuan	Winter	0.829±0.264	87	A	II
SCAUf59	S2	Sichuan	Winter	1.006±0.0872	50.3	A	XI
SCAUf57	S2	Sichuan	Winter	0.907±0.0809	47.7	A	XI
SCAUf58	S2	Sichuan	Winter	0.899±0.138	44	A	XI
SCAUf56	S2	Sichuan	Winter	0.650±0.0893	38.3	A	XI
SCAUf61	S3	Sichuan	Winter	1.201±0.0914↑*	21.3	A	XII
SCAUf60	S3	Sichuan	Winter	1.035±0.168↑*	69.3	A	XII
SCAUf62	S3	Sichuan	Winter	0.862±0.141	59	A	VI
SCAUf63	S4	Sichuan	Winter	0.718±0.245	74.3	A	XII
SCAUf77	S5	Sichuan	Winter	1.051±0.255↑*	111.3	A	XII
SCAUf78	S5	Sichuan	Winter	0.899±0.0552	46	A	I
SCAUf79	S5	Sichuan	Winter	0.764±0.210	45	A	XII
SCAUf81	S6	Sichuan	Winter	1.054±0.190↑*	124.3	A	IV
SCAUf80	S6	Sichuan	Winter	1.013±0.0880↑*	114	A	IV
SCAUf82	S7	Sichuan	Winter	0.819±0.0187	19.7	A	VIII
SCAUf83	S8	Sichuan	Winter	0.787±0.0987	84.3	A	XI
SCAUf84	S9	Sichuan	Winter	0.845±0.158	14	A	V
SCAUf85	S9	Sichuan	Winter	0.636±0.0633	17.7	A	XIII
SCAUf67	YCDB	Zhejiang	Winter	0.925±0.0691	40.5	A	XI
SCAUf66	YCDB	Zhejiang	Winter	0.883±0.0235	34	A	XI
SCAUf69	ZZL	Tianjin	Spring	0.980±0.0664	110.3	A	III
NC	-	-	-	0.661±0.115	-	-	-

a The strains were deposited in Sichuan Agricultural University (SCAU) strain collection center. NC = uninoculated control in the symbiotic efficiency test.

b Faba bean cultivars: S1-S9 = Local Sichuan faba bean cultivars; JX = Jingxuan; LP = Lüpi; HN = Henan; ZZL = Zhenzhulü; YCDB = Yicundabai; DB = Dabai.

c↑* = the plant dry weight was significantly higher than that of NC according to the LSD test (P = 0.05). Data is shown as average ± standard error (n = 3).

d,e 16S-RFLP and IGS-RFLP genotype represents the combination of restriction patterns obtained by enzymes used (HaeIII, HinfI, MspI, and TaqI).

RFLP analyses

The 16S rRNA gene of all the 31 isolates was successfully amplified and approximately 1500 bp single band was obtained. In the 16S rRNA gene RFLP analysis all the strains represented the same 16S genotype (Table 1). In the IGS PCR, one band was amplified from all the isolates except SCAUf57, SCAUf59, SCAUf66 and SCAUf69 for which two bands were detected. According to the IGS-RFLP, the isolates represented thirteen genotypes that were divided into four IGS groups with 10, 5, 2, and 14 isolates at 93% similarity (Fig 1). The nine significantly growth promoting isolates were distributed into groups IGS1 (4 strains), IGS2 (2 trains) and IGS4 (3 strains).

Fig 1

IGS PCR-RFLP analysis by four restriction endonucleases (MspI, HinfI, HaeIII and TaqI) of strains isolated from faba bean nodules.

Strain number is followed by the name, the origin and the ecotype of the host of isolation. Abbreviations are as in the footnote of Table 1. The IGS PCR-RFLP groups are labeled with Roman numerals I to III. Isolates in bold were selected for sequencing of housekeeping and symbiosis genes.

Phylogenetic analysis of 16S rRNA gene

Based mainly on IGS RFLP while also considering the hosts of isolation, eight representative strains were selected for sequencing of the housekeeping and symbiosis genes. In accordance with the IGS RFLP (Fig 1), the strains were divided into three groups related to Rhizobium in the 16S rRNA gene phylogenetic tree (Fig 2). SCAUf82 clustered with R. leguminosarum USDA 2370T with 99.8% similarity (group R1). SCAUf61, SCAUf62 and SCAUf76 clustered with R. laguerreae FB206T, R. leguminosarum USDA 2370T, R. anhuiense CCBAU 23252T, R. sophorae CCBAU 03386T and R. gallicum R602spT with 100% similarity (group R2). SCAUf59, SCAUf67, SCAUf68 and SCAUf70 clustered separately (group R3).

Fig 2

The phylogenetic relationships between the representative faba bean strains and reference strains based on 16S rRNA gene sequences (1278 nt).

Genbank accession numbers are in parentheses. Bootstrap values above 50% are shown on the branches. Scale bar represents 0.2% nucleotide substitutions. R: Rhizobium.

Phylogenetic analyses of housekeeping genes

Three housekeeping genes (atpD, glnII and recA) were successfully amplified from all representative strains. In the multilocus sequence analysis (MLSA) tree that was based on concatenated gene sequences (Fig 3) and single gene trees (S1–S3 Figs) the representative strains were divided into two clusters related to Rhizobium. SCAUf59, SCAUf67, SCAUf68 and SCAUf70 were 96.6% similar with the type strain R. sophorae CCBAU 03386T and were assigned as Rhizobium sp. SCAUf61, SCAUf62, SCAUf76, and SCAUf82 clustered with R. anhuiense CCBAU 23253T with 98.8% to 99.3% similarity in the MLSA, and were thus assigned as R. anhuiense.

Fig 3

The phylogenetic relationships between the representative faba bean strains and reference strains based on multilocus sequence analysis (MLSA) of atpD (411 nt), glnII (514 nt), and recA (384 nt) genes.

Genbank accession numbers are in parentheses. Bootstrap values above 50% are shown on the branches. Scale bar represents 1% nucleotide substitutions. R: Rhizobium.

Phylogenetic analyses of symbiosis genes

Approximately 700 bp, 900 bp and 1500 bp fragments of nifH, nodC and nodD, respectively, were amplified from all the representative strains. In the nifH phylogenetic tree, the strains were distributed into five groups (Fig 4). The Rhizobium sp. SCAUf68, R. anhuiense SCAUf62 and R. anhuiense CCBAU 23252T nifH genes in the group H1 were 100% similar. Rhizobium sp. strain SCAUf59 clustered alone as group H2. In the group H3, R. anhuiense strains SCAUf61 and SCAUf76 carried a nifH gene 100% similar with R. fabae CCBAU 33202T. R. anhuiense strain SCAUf82 clustered with R. leguminosarum USDA 2370T and R. pisi DSM 30132T with 98.4% similarity in group H4. Rhizobium sp. strains SCAUf67 and SCAUf70 clustered separately in group H5.

Fig 4

Phylogenetic tree based on nifH (346 nt) gene of representative faba bean strains and reference strains.

Genbank accession numbers are in parentheses. Bootstrap values above 50% are shown on the branches. Scale bar represents 1% nucleotide substitutions. R: Rhizobium.

In the nodC phylogenetic tree, the strains were distributed into four groups (Fig 5). R. anhuiense strains SCAUf61, SCAUf76 and SCAUf82 clustered with R. fabae CCBAU 33202T in group C1. R. anhuiense SAUf62 and Rhizobium sp. SCAUf68 clustered with R. anhuiense CCBAU 23252T with 100% similarity in group C2. Rhizobium sp. strains SCAUf67 and SCAUf70 clustered separately as group C3. SCAUf59 clustered with Pisum sativum symbionts in group C4 (Fig 5).

Fig 5

Phylogenetic tree based on nodC (534 nt) gene of representative faba bean strains and reference strains.

Genbank accession numbers are in parentheses. Bootstrap values above 50% are shown on the branches. Scale bar represents 2% nucleotide substitutions. R: Rhizobium.

The nodD sequences were divided into three groups (Fig 6). R. anhuiense strains SCAUf61, SCAUf76 and SCAUf82 clustered with R. fabae CCBAU 33202T with 100% similarity in group D1. R. anhuiense SCAUf62 and Rhizobium sp. SCAUf68 carried nodD genes 100% similar to that of CCBAU 53093–2 in group D2. Rhizobium sp. strains SCAUf59, SCAUf67 and SCAU70 clustered with R. leguminosarum CCBAU 65264 and CCBAU 81100 with 100% similarity in group D3 (Fig 6).

Fig 6

Phylogenetic tree based on nodD (847 nt) gene of representative faba bean strains and reference strains.

Genbank accession numbers are in parentheses. Bootstrap values above 50% are shown on the branches. Scale bar represents 0.5% nucleotide substitutions. R: Rhizobium.

Discussion

Biological N2 fixation (BNF) is a key component in sustainable agriculture since BNF can replace nitrogen fertilizer in growing legume crops. Improving legume yields requires efficiently nitrogen-fixing symbiotic bacteria, rhizobia. Efficient rhizobium strains may be selected and applied as inoculants to increase yields [20, 28].

Faba bean cultivars are divided into three ecotypes: winter faba bean, spring faba bean and Yunnan ecotype [6, 7]. Based on the sequence analysis of symbiosis gene nodD, faba bean symbionts were divided into groups roughly related to the division of their hosts to winter and spring ecotypes [2]. The faba beans grown in Sichuan are the winter type. Earlier we isolated faba bean rhizobia from 21 sites covering most of the faba bean growing area in Sichuan [11]. The nodulating strains carried nodC similar to those of strains that nodulate winter type faba bean and field pea [11]. In the present study, we collected 31 rhizobium strains from root nodules of fifteen faba bean cultivars grown on the same field. The growth of a local faba bean cultivar was significantly promoted by nine isolates, implying that these strains were potential inoculants. Potential inoculant strains were isolated from both winter type and spring type cultivars. The applicability of these strains must be further tested in field experiments.

Faba beans are mainly nodulated by R. leguminosarum, R. etli, R. fabae and R. anhuiense [9, 10, 13, 29]. In our earlier study we found that in Sichuan, in addition to R. leguminosarum, faba bean was nodulated by five Rhizobium species [11]. R. leguminosarum strains are quite diverse, and the Sichuan R. leguminosarum strains were divided to two distantly related groups [11, 30]. In this study, based on both 16S rRNA gene and multilocus sequence analyses of three housekeeping genes, the representative isolates were assigned as R. anhuiense and a potential new Rhizobium species.

16S rRNA gene and housekeeping gene sequence analyses are usually applied to assess the phylogenetic position of rhizobia, whereas analyses of symbiosis related genes reveal the symbiotic genotype of rhizobia. Comparison of housekeeping and symbiosis genes gives crucial information on how rhizobia evolve [31]. The symbiosis related genes are commonly on mobile genetic elements that are transferrable between strains [32]. The nifH gene, encoding the nitrogenase Fe protein, is essential for symbiotic nitrogen fixation [33]. The nodC is essential for the synthesis of the aminosugar backbone of rhizobial signaling molecules [34].The nodD encodes a transcriptional regulator of nod genes [2]. The faba bean nodulating Sichuan R. leguminosarum strains carried three different nifH and nodC variants [11]. In addition to them, we found two and one new variants of nifH and nodC, respectively. Legumes may be classified as promiscuous or selective according to the number of rhizobia they are compatible with. For example, in Sichuan the promiscuous Leucaena leucocephala was nodulated by seven rhizobial species [35]. However, on nodulation gene (nodC) level there was less diversity [16]. Interestingly, the species and nodulation gene diversity of the faba bean symbionts was the opposite: two species that carry four variants of nodC, raising an intriguing question on what ultimately governs the host specificity. Both R. anhuiense and Rhizobium sp. included strains with respective three different nifH-nodC combinations (Table 2), indicating that the genes had been horizontally transferred (Table 2). The nodD variants D3 and D1were found together with two different nifH-nodC combinations suggesting a possible partial transfer of symbiotic island or plasmid (Table 2).

Table 2

The phylogenetic relationships of strains isolated from fifteen faba bean cultivars grown on the same field in Chengdu, China.

Straina	IGS group	The most similar species (sequence similarity with type strain, %)b
		atpD	glnII	recA	MLSA	nodC	nifH
R. anhuiense (17)
SCAUf61, 60, 63, 69, 75,	IGS1	Ra(98.5%)	Ra(99.6%)	Ra(98.2%)	Ra(98.8%)	Rf(100%)	Rf(100%)
SCAUf76, 77, 79, 84, 85		Ra(98.5%)	Ra(99.6%)	Ra(98.2%)	Ra(98.8%)	Rf(100%)	Rf(100%)
SCAUf62, 55, 78, 80, 81	IGS2	Ra(99.7%)	Ra(99.2%)	Ra(98.8%)	Ra(99.3%)	Ra(100%)	Ra(100%)
SCAUf82, 64	IGS3	Ra(99.7%)	Ra(98.3%)	Ra(100%)	Ra(99.3%)	Rf(100%)	Rle(98.4%)
Rhizobium sp.(14)
SCAUf59, 56, 57, 58	IGS4	Rla(95.5%)	Rle(98.1%)	Rle(97.6%)	Rs(96.6%)	Rle(96.0%)	Ra(99.4%)
SCAUf67, 65, 66		Rla(95.5%)	Rle(98.1%)	Rle(97.6%)	Rs(96.6%)	Rle(95.3%)	Rle(96.8%)
SCAUf68, 74, 83		Rla(95.5%)	Rle(98.1%)	Rle(97.6%)	Rs(96.6%)	Ra(100%)	Ra(100%)
SCAUf70, 71, 72, 73		Rla(95.5%)	Rle(98.1%)	Rle(97.6%)	Rs(96.6%)	Rle(95.3%)	Rle(96.8%)

a Strains in this study(number of isolates). The isolates in bold face were representative strains and were used in 16S rRNA, recA, glnII, atpD, nifH, and nodC genes sequencing.

b MLSA: Multilocus sequence analysis of concatenated atpD, glnII and recA genes. Ra = R. anhuiense; Rf = R. fabae; Rla = R. laguerreae; Rle = R. leguminosarum; Rs = R. sophorae.

The genotypes nod-I and nod-III are the dominant nodD genotypes of winter type and Yunnan-China type faba beans, respectively [2, 29]. In our study, the isolates carried three nodD variants. The variants D1 and D2 clustered with nod-I, and the variant D3 with nod-III. Strains carrying similar nodD gene variants and nifH-nodC combinations were trapped by both winter and spring cultivars, disagreeing with the division of nodD to cultivar specific variants. Obviously, analysis of a greater number of strains is needed to conclude whether the different cultivars favor certain nodD gene variants. Instead of cultivar specificity, the division of nodD types may be geographical. Since Sichuan is in between the spring type, winter type and Yunnan type faba bean cultivation areas, finding rhizobia carrying different nodD genotypes may be expected.

In all, the soil hosted R. anhuiense and a potential new Rhizobium species that were diverse on the symbiosis related gene level. The strains nodulated both winter and spring type cultivars. Even though the rhizobia may be co-introduced with the host [36], isolating similar strains from both winter and spring type cultivars indicates that the strains originated from soil, not seed. The local genetic diversity under uniform environmental conditions was equal to diversity of nodulating strains collected from many sampling sites with different conditions and cultivars, indicating that the symbiotic specificity of the faba bean rhizobia was related to the cultivars.

Phylogenetic tree based on atpD (396 nt) genes of the representative strains isolated from fababean and reference strains.

Genbank accession numbers are in parentheses. Bootstrap values ≥ 50% areshown on the branches. Scale bar represents 1% nucleotide substitutions. R: Rhizobium.

(JPG)

Click here for additional data file.

Phylogenetic tree based on glnII (483 nt) genes of the representative strains isolated from fababean and reference strains.

Genbank accession numbers are in parentheses. Scale bar represents 1%nucleotide substitutions. Bootstrap values ≥ 50% are shown on the branches, R: Rhizobium.

(JPG)

Click here for additional data file.

Phylogenetic tree based on recA (345 nt) genes of the representative strains isolated from fababean and reference strains.

Genbank accession numbers are in parentheses. Bootstrap values ≥ 50% areshown on the branches. Scale bar represents 1% nucleotide substitutions. R: Rhizobium.

(JPG)

Click here for additional data file.