2009Genetic diversity and biogeography of rhizobia associated with

Systematic and Applied Microbiology 32(2009)351–361

Genetic diversity and biogeography of rhizobia associated with Caragana species in three ecological regions of China

Yang Li Lu a ,Wen Feng Chen a,?,En Tao Wang b ,Su Hua Guan a ,Xue Rui Yan c ,Wen Xin Chen a

The State Key Laboratory for Agrobiotechnology/Department of Microbiology and Immunology,College of Biological Sciences,China Agricultural University,Beijing 100193,China b

Departamento de Microbiolog?

′a,Escuela Nacional de Ciencias Biolo ′gicas,Instituto Polite ′cnico Nacional,Me ′xico D.F.11340,Mexico c

College of Plant Protection,Shenyang Agricultural University,Shenyang 110161,China Received 13August 2008

Abstract

Twenty-two genospecies belonging mainly to Mesorhizobium ,and occasionally to Rhizobium and Bradyrhizobium ,were de?ned among the 174rhizobia strains isolated from Caragana species.Highly similar nodC genes were found in the sole Bradyrhizobium strain and among all the detected Mesorhizobium strains.A clear correlation between rhizobial genospecies and the eco-regions where they were isolated was found using homogeneity analysis.All these results demonstrated that Caragana species had stringently selected the rhizobia symbiotic genotype,but not the genomic background;lateral transfer of symbiotic genes from Mesorhizobium to Bradyrhizobium and among the Mesorhizobium species has happened in the Caragana rhizobia;and biogeography of Caragana rhizobia exists.Furthermore,a combined cluster analysis,based upon the patterns obtained from ampli?ed 16S rRNA gene and 16S–23S intergenic spacer restriction analyses,BOX PCR and SDS-PAGE of proteins,was reported to be an ef?cient method to de?ne the genospecies.

Keywords:Caragana ;Mesorhizobium ;Diversity;Eco-region;Combined cluster analysis

Introduction

Recently,the existence of biogeography for soil bacteria has been proven by several studies.The distribution of free-living soil bacterial communities depends on their adaptation to the ecosystem,and is mainly related to the soil pH [1].This conclusion

also ?ts the plant-associated symbiotic bacteria,such as the rhizobia [2].Collectively,rhizobia include soil bacteria from about 10genera of Alphaproteobacteria and Betaproteobacteria ,such as Rhizobium and Burkholderia ,which cause the forma-tion of nodules on roots and/or stems of some legumes and they reduce nitrogen to ammonia inside the nodules.Therefore,they are symbiotic nitrogen ?xers,a most important type of microbial resource [3]with great potential for improving the growth of their host plants [4,5].

www.elsevier.de/syapm

?Corresponding author.Tel.:+861062734009;

fax:+861062734008.

E-mail address:chenwf@https://www.360docs.net/doc/e72771300.html, (W.F.Chen).

Differently from the free-living soil bacteria,rhizobia additionally face selection stress related to their host plant speci?city and adaptation[6],as well as to the soil conditions.In our recent studies,correlations between the rhizobia genomic groups and their geo-graphic origins have been detected among the micro-symbionts of the faba bean(Vicia faba)[7]and epidemic legumes on the Qinghai–Tibet plateau[8].These studies and other research[2,9]demonstrated that both the abiotic conditions(temperature,soil pH,rainfall,etc.) and biotic conditions(genotypes of host plants and their distribution)might affect the diversity of the rhizobia species.

The legume genus Caragana contains about100 species native to northeastern Europe and central Asia, and they are remarkable for their cold-and drought-resistant properties and nitrogen-?xing ability.Approxi-mately70species have been recorded in China and most of them grow on hills,plains and plateaus,as well as in gullies and thickets[10].The Caragana species serve as forage resources for wild animals and are preferable plants for reforestation against deserti?cation and erosion[11,12].In addition,their?owers are a good source of honey production and their seeds are used as an herbal medicine in China[13].

Although slow-growing cowpea–soybean–lupine types of rhizobia were isolated by Beijerinck from Caragana arborescens one hundred yeas ago[14],the rhizobia associated with Caragana species were not studied systematically for a long period.Chen et al.[15] described several rhizobial strains isolated from C.polourensis grown in Xinjiang,China as Mesorhizo-bium tianshanense.About20bacterial strains isolated from Caragana spp.growing in the Ningxia and Inner Mongolia regions of China were identi?ed as Rhizobium [16,17].Thirty-three rhizobial strains isolated from Caragana intermedia growing in the Maowusu Desert in Inner Mongolia were reported as Rhizobium that were closely related to Agrobacterium tumefaciens[18].Most recently,109rhizobia strains isolated from Caragana spp.growing in Liaoning,northeast China,were designated into the genus Mesorhizobium[19].

All previous studies have suggested that distinct rhizobia might nodulate with Caragana species grown in different regions,but the association between these Caragana rhizobia and their geographic origins has not been statistically analyzed.To investigate the diversity of Caragana rhizobia in other geographical regions, and to elucidate the relationship between Caragana rhizobia populations and their geographic origins,more rhizobia needed to be surveyed and statistical analysis had to be applied.In the present study,the genetic diversity,biogeography and symbiotic relationships with the host legumes of rhizobia isolated from nodules of Caragana spp.grown in three eco-regions in China were analyzed.Materials and methods

Isolation and nodulation assays of rhizobia

To isolate rhizobia,root nodules were collected in July and August2005from the endemic Caragana species C.intermedia,C.jubata,C.bicolor,C.erinacea and C.franchetiana growing at17sampling sites in three eco-regions in China.Eco-region A had prairie with sandy soils in Eastern Inner Mongolia(a temperate region),whereas eco-region B corresponded to hills with saline/alkaline soil in Northern Shanxi(Loess Plateau in a temperate region),and eco-region C referred to the hillside/forestland with fertile soil in North-western Yunnan(a tropical region with high mountains)(see Supplementary Fig.A for detail).C.intermedia was the most endemic legume shrub in eco-regions A and B.The other four species were native to eco-region C and not found in the other two eco-regions.Rhizobia were isolated from root nodules and were puri?ed with a standard procedure on YMA medium[20].Nodulation on the host of origin of each strain was checked with the method of Vincent[20],by growing the inoculated seedlings in vermiculite moistened with N-free nutrition solution[21]under natural sunlight and temperature in August.Nodules were observed after1month of inoculation.All bacterial strains were kept on YMA medium at41C for temporary storage and in30%(v/v) glycerol atà701C for long-term storage.

Ampli?ed16S rRNA gene and16S–23S intergenic spacer(IGS)restriction analysis

DNAs extracted according to Terefework et al.[22] were used as templates to amplify the almost-complete 16S rRNA gene with the procedure and primers P1(50-AGA GTT TGA TCC TGG CTC AGA ACG AAC GCT-30)and P6(50-TAC GGC TAC CTT GTT ACG ACT TCA CCC C-30)described previously[23],and to amplify the16S–23S IGS with primers FGPS1490 (50-TGC GGC TGG ATC ACC TCC TT-30)and FGPL1320(50-CCG GGT TTC CCC ATT CGG-30) using the procedure described by Laguerre et al.[24]. The endonucleases Alu I,Hinf I,Hae III and Msp I (Takara Shiga,Japan)were used separately to digest the ampli?ed rRNA gene and Msp I,Hha I and Hae III to IGS.The restriction fragments were analyzed by electrophoresis in3%(w/v)agarose gels[25]supplied with0.5m g mlà1of ethidium bromide.The rRNA gene restriction patterns were visualized and photographed under UV light and were normalized according to the patterns of DNA molecular weight markers,which were included on the left,right and in the center of each gel. Strains with identical restriction patterns were de?ned as the same rDNA type or IGS type.

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361 352

BOX-PCR?ngerprinting

In order to investigate the genomic diversity, BOX-PCR was applied to the strains by using total DNA of bacteria as templates with primer BOXair (50-CTA CGG CAA GGC GAC GCT GAC G-30)[26] and the PCR procedure as described by Nick et al.[27]. The PCR products were separated by electrophoresis in 1.5%(w/v)agarose gels[27].The ampli?ed fragments were separated,visualized and normalized as described for16S rRNA gene restriction analysis.

SDS-PAGE of whole-cell proteins

The preparation of whole-cell soluble proteins of the bacteria and the SDS-PAGE was performed as described by Tan et al.[16].The protein patterns visualized by silver staining[23]were scanned and normalized according to the same patterns of protein markers as in16S rRNA gene restriction analysis. Cluster analysis

The Sj coef?cient[25]for the patterns of restriction analyses of the ampli?ed16S rRNA gene and IGS,and of protein SDS-PAGE,as well as the Dice coef?cient for BOX-PCR?ngerprints were calculated between each strain pair and were used in cluster analysis to generate a UPGMA dendrogram[28]for each analysis with the GelCompar II software package.The Optimization 2.00%parameter and Band2.00%comparison toler-ance were used in the cluster analysis.

To perform the combined cluster analysis,the DNA ?ngerprints and protein patterns obtained in each of the four methods were combined.Firstly,the proportion (weight)of each analysis in the combined cluster analysis was justi?ed by using the combined patterns of reference strains to estimate the optimum relative proportion of each analysis that could differentiate the reference strains according to their genera and species. Then,all the strains were included in the combined cluster analysis with the Sj coef?cient and the UPGMA method in the GelCompar II software package with the optimum proportions.

Sequencing and phylogenetic analysis of the16S rRNA gene and nodC

The16S rRNA gene of the representative strains for each combined cluster was ampli?ed with the same primers and procedure as in the restriction analysis,and was sequenced directly by using a PCR-based method [16].Primers for nodCF and nodCI were used to amplify the nodulation gene nodC[29].The acquired sequences and the related sequences extracted from the GenBank database with the Blast program were aligned with the ClustalW program using MEGA4software.Phyloge-netic trees were constructed with the neighbor-joining method and the Jukes–Cantor model,and were boot-strapped with1000replications of each sequence[30]. Correspondence analysis

The correlations of the rhizobia genospecies (combined clusters)and the geographic origins were examined with the correspondence analysis(Ver.1.0) using the SPSS12.0package(by the Data Theory Scaling System Group,Faculty of Social and Behavioral Sciences,Leiden University,The Netherlands).In this analysis,the geographic origin and rhizobial genospecies (combined clusters)were treated as two variables. Seventeen levels,corresponding to the sampling sites, were included in the variable of geographic origin.The 22combined clusters de?ned in the Caragana rhizobia were treated as levels in the variable of rhizobial species. The results of the correspondence analysis were pre-sented in a two-dimensional?gure,in which different levels of the variables were grouped.

Results

Isolations and nodulation assay of rhizobia

In this study,a total of174rhizobial strains were isolated from Caragana nodules:91from eco-region A, 67from eco-region B and16from eco-region C(Table1 and Supplementary Table A).More than150root nodules were collected from eco-region C,but only 16rhizobial isolates were obtained and survived successfully.This may be due to the non-culturability of rhizobia,as shown by Murseu et al.[31].Strain CCBAU03282was the sole slow growing and alkali-producing bacterium on YMA media,and the other173 strains were fast or moderately growing,acid-producing bacteria that formed colonies larger than1mm in diameter after5days of incubation.In nodulation tests, all the isolated strains were induced on the effective nodules of host plants,which were shown by the red color of nodules and the dark-green leaves.

Ampli?ed16S rRNA gene and IGS restriction analyses

A total of24rDNA types(Table1)were found among the Caragana rhizobia strains,which were divided into three main groups at a60%similarity level (?gure not shown):the Mesorhizobium group covering rDNA types1–12with144strains,the Rhizobium group

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361353

covering rDNA types13–29including29strains,and Bradyrhizobium strain CCBAU03282(rDNA type31). In IGS analysis,single fragments with sizes ranging from800to1500bp were ampli?ed from the strains. Several strains produced an extra adjacent fragment (data not shown)and they were used in restriction directly.In the dendrogram constructed with IGS restriction patterns,the Caragana rhizobia were divided into33clusters or single strains and most of the reference strains formed clusters according to their species at a similarity level of80%(Table1).In several cases,strains in one rDNA type were found in different IGS clusters.

BOX-PCR?ngerprinting

In this study,the cluster analysis of BOX-PCR patterns was performed separately for the strains of Mesorhizobium,Rhizobium and Bradyrhizobium,accord-ing to the restriction results of the16S rRNA gene. Great resolution was obtained from the BOX-PCR ?ngerprints,since most strains showed distinctive patterns.Taking the grouping results of reference strains as a control,the Caragana rhizobia formed52clusters at a similarity of88%(Table1).In most cases,strains within a16S rDNA type,or the same IGS cluster,were separated into several BOX clusters,but no strains of different rDNA types or IGS clusters were found in the same BOX cluster.

SDS-PAGE of whole-cell proteins

In the dendrogram constructed in this analysis(?gure not shown),a total of44protein clusters were obtained at a92%similarity level(Table1)and all the reference strains were grouped according to their species.The Caragana rhizobia were found in27clusters.Most protein clusters corresponded to the rDNA types/IGS groups.In general,this method had a better resolving power than ARDRA but less than BOX-PCR,and the grouping results were similar to PCR-RFLP of IGS. Combined cluster analysis

The proportion of33.3%for16S rRNA gene restriction patterns,16.7%for IGS restriction patterns, 16.7%for BOX-PCR patterns and33.3%for protein patterns proved to be the most suitable parameter, according to the grouping results of reference strains(data not shown).Most strains had distinctive combined patterns and all the reference strains were differentiated according to their species in the dendrogram(Fig.1).Considering the grouping results of reference strains,22clusters containing2–43 strains each and29single strains corresponding to Mesorhizobium,Bradyrhizobium and Rhizobium genospecies were de?ned at a90%similarity level (Fig.1).

Sequencing and phylogenetic analysis of the16S rRNA gene and nodC

Forty-three strains representing the22combined clusters were chosen for sequencing of the16S rRNA gene.The acquired16S rRNA gene sequences in this study had approximately1350nt.In the phylogenetic tree(Fig.2),the Caragana rhizobia were grouped into Mesorhizobium,Rhizobium and Bradyrhizobium,similar to the dendrogram of combined cluster analysis,implying that the results of the two methods were consistent.There was only one strain CCBAU65328in cluster3and it clearly appeared in a different place between the two methods(Figs.1 and2).

Ampli?cation of the nodC fragment from the Rhizobium strains failed.Similar sequences for nodC were found among the Mesorhizobium and Bradyrhizo-bium strains and these sequences showed more than 94.0%similarities with those in M.tianshanense A-1BS T,Mesorhizobium temperatum SDW018T and Mesorhizobium septentrionale SDW014T(Fig.3). Correspondence analysis

In the generated two-dimensional?gure(Fig.4),three groups were de?ned based upon the distances between the combined clusters and sampling sites.The Pearson correlation coef?cient was0.454(40.01), meaning that the correspondence was signi?cant.Four of the?ve sites in region A(except A5)and eight combined clusters formed a group.The nine sites in regions B,A5and C3formed another group together with nine combined clusters.Sites C1and C2,together with combined clusters7,10and15,formed the third group.These results showed a clear correlation between rhizobial genospecies and their geographic origins.

According to this correspondence analysis,Mesorhi-zobium genospecies I,II,IV,VI and VIII contained the strains isolated from Region A,while the genospecies M.temperatum,M.tianshanense,M.septentrionale, Mesorhizobium sp.III,Rhizobium yanglingense and Rhizobium sp.V were mainly from Region B. M.plurifarium and Mesorhizobium genospecies V and VII,and Rhizobium sp.IV were from Yunnan(Region C).Mesorhizobium amorphae genospecies included strains isolated from both regions A and B,and M.temperatum genospecies was found in both regions B and C.

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361355

Discussion

Currently,16S rRNA gene sequence analyses are used as the basal method to group bacteria at the genus level [25,32],whereas 16S–23S IGS restriction analysis [24],SDS-PAGE of whole-cell proteins and DNA–DNA hybridizations [23]are methods to de?ne bacterial species;and BOX-PCR ?ngerprinting is applied to reveal genetic diversity of bacterial popula-tions [18,33–35].Normally,the pro?les produced from each method are used to construct dendrograms separately and the strains are classi?ed based upon the consensus of the grouping results.The separated analysis produces various dendrograms that enlarge the paper size and complicate the de?nition of bacterial classi?cation.In the present study,to overcome these disadvantages,combined cluster analysis was tried,where the weight or proportion of each pattern is a determinant for obtaining reasonable grouping.The validity of the proportions used in this study was con?rmed by both the grouping results of reference strains and the 16S rRNA gene sequence comparison.

Fig.1.Simpli?ed dendrogram constructed by combining cluster analyses showing the relationships of Caragana rhizobia strains.The number in parenthesis refers to the number of strains represented by the strain.The full tree is given in Supplementary Fig.B.

Y.L.Lu et al./Systematic and Applied Microbiology 32(2009)351–361

356

These proportions re?ect the importance of each method in the de?nition of bacterial genus and species,and give the consensus a quantitative basis. The de?nition of22combined clusters and11single strains among the tested rhizobial strains,using the combined cluster analysis,demonstrated that this method has a high resolution for distinguishing bacterial species.Only one exception showed that strain CCBAU65328was grouped into cluster3and was adjacent to the de?ned species M.temperatum SDW018T(Fig.1),while its phylogenetic position based on the16S rRNA gene had Mesorhizobium ciceri as its closest neighbor(Fig.2).The close phylogenetic position between CCBAU65328 and M.ciceri was con?rmed further by sequencing of the three genomic housekeeping genes,recA (EU672502),glnII(EU672487)and atpD(EU672472) (unpublished data).

Fig.2.Neighbor-joining tree constructed from sequences of the16S rRNA gene showing the phylogenetic relationships of the representative Caragana rhizobia and related bacteria.Bootstrap values above50are indicated at the main nodes.The scale bar represents a1%nucleotide divergence.

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361357

Clearly,the Caragana species formed nodules mainly with Mesorhizobium spp.(82.8%)and occasionally with Rhizobium and Bradyrhizobium in the three eco-regions of China.These results were similar to our previous study on Caragana rhizobia in another eco-region in Northeastern China [19],but different from those in which Rhizobium /Agrobacterium -related strains were predominant in C.intermedia-associating rhizobia in the Maowusu Desert (adjacent to and north of eco-region B in the present study)[18].It seems that Mesorhizobium spp.are the main microsymbionts for temperate woody legumes considering the results of this and other studies on the rhizobia associated with Albizia kalkora [36],Caragana spp.[19],Cercis racemosa and Wisteria sinensis [37],as well as Robinia pseudoaccacia [38]and Sophora spp.[39].

Including the ungrouped strains,59.5%(85strains)of the Mesorhizobium strains might represent novel poten-tial genospecies,since they formed combined clusters or single strains distinct from the reference strains of de?ned species.In addition,the remaining 40.5%(59strains)were also different from the corresponding reference strains,although they were grouped with reference strains of M.temperatum ,M.plurifarium ,M.tianshanense ,M.amorphae ,M.septentrionale and Mesorhizobium huakuii .Further studies,including

Fig.3.Neighbor-joining tree constructed from sequences of partial nodC gene sequences showing the phylogenetic relationships of the representative Caragana rhizobia and related bacteria.Bootstrap values above 50are indicated at the main nodes.The scale bar represents a 2%nucleotide

divergence.

Fig.4.Homogeneity analysis showing the correlation between the rhizobial species (combined clusters)and their geographic origins (sampling sites).Dimensions 1and 2represent the linear combination of the two variables and may have no real meaning.Different levels in the same variable located in the same group meant that they might have similar characteristics.The distance between different variables in the same group represented their relative correlation.

Y.L.Lu et al./Systematic and Applied Microbiology 32(2009)351–361

358

phenotypic characterization and DNA–DNA hybridiza-tion,may be needed to clarify the exact taxonomic position of the genospecies.

According to our results,the Caragana rhizobia communities were different in the three eco-regions. The clear correlation between the geographic regions and the rhizobial genospecies may be attributed to the different environmental factors and soil characteristics of the sampling sites in this study(Fig.1),as reported for soil bacteria[1].The close correlation of sites A5and C3with eco-region B demonstrated the possibility that the soil characters might be more important that the climate factors in the determination of rhizobial diversity or community composition,therefore,the soil characters should be considered in any further study of the biogeography of rhizobia.Furthermore,the fact that the rhizobia from Yunnan Province were obtained from Caragana species different from those in eco-regions A and B might be another factor to differentiate the rhizobia community there from those of eco-regions A and B.

Considering both the results in the present study and previous studies on faba bean rhizobia[7],soybean rhizobia[40,41]and Caragana rhizobia[18,19]in different regions of China,the existence of rhizobia biogeography is clear in different eco-regions.The correlation between rhizobial species and their geo-graphic origin should be an important consideration for selection of rhizobial inoculates.

Nodulation genes in rhizobia have been widely used to perform evolutionary analysis and to estimate their host ranges[29,42,43].One of the nodulation genes, nodC,which encodes the enzyme involved in the?rst step of the Nod factor assembly has often been chosen as a nodulation marker in different studies because it is essential for nodulation in most of the rhizobial species examined so far[43–46].The nodC gene has also been described as a determinant of host range[42,47–49], though it has been reported recently that some photo-synthetic bradyrhizobial strains can nodulate in the absence of conventional nod genes[50].

In our present study,the nodC gene was ampli?ed from Mesorhizobium and Bradyrhizobium strains.The reason for the failure to amplify nodC from Rhizobium strains was not clear.The close phylogenetic relation-ships of nodC sequences in the Caragana rhizobia demonstrated that the Caragana species strongly selected the speci?c and similar symbiotic background of rhizobia,as opposed to the situation of soybean that is a host nodulating with rhizobia harboring diverse nodC genes[41].Considering their similar geological distribution,the harboring of similar nodC in different Mesorhizobium genospecies might be a result of frequent lateral transfer of symbiotic genes among these closely related rhizobia,as reported previously[29,42,49,51]. The high nodC similarity between Bradyrhizobium https://www.360docs.net/doc/e72771300.html,BAU03282and Mesorhizobium https://www.360docs.net/doc/e72771300.html,BAU03299 strongly indicated that the symbiotic genes had been transferred from Mesorhizobium to Bradyrhizo-bium.To better understand the relationships between the genomic genes and the symbiotic genes in rhizobia,it would be important and interesting in future work to study the nodulation gene type of the Rhizobium/ Agrobacterium-related strains[18],as well as the Rhizobium strains isolated from Caragana in present and previous studies[16].

In the nodC sequence analysis,similarities as great as 94–100%were found among the Caragana mesorhizo-bia and M.tianshanense A-1BS T,M.temperatum SDW 018T and M.septentrionale SDW014T,isolated from Glycyrrhiza pallidi?ora[15,42]and Astragalus adsurgens, respectively[42,52].The cross-nodulation among Caragana,Glycyrrhiza and A.adsurgens rhizobia [15,18,42]was consistent with the great similarities of their nodC,demonstrating that the nodC sequence similarity is a good estimation of host range for rhizobia,as suggested by other authors[48,52].

In conclusion,the results of the present study demonstrated that Caragana species could nodulate with distinctive rhizobial populations,mainly Mesorhi-zobium spp.,in the three eco-regions sampled;lateral transfer of a symbiotic gene(nodC)between the Mesorhizobium genospecies and from Mesorhizobium to Bradyrhizobium might have happened.The Mesorhi-zobium genospecies isolated from Caragana could nodulate with Glycyrrhiza,Astragalus,Gueldenstaedtia and Oxytropis species.It was also found that combined cluster analysis of different electrophoretic patterns was a rapid and convenient method to classify a large number of rhizobial strains.

Acknowledgments

This work was?nanced by the National Natural Science Foundation of China(Project nos.30870004, 30400001and30670001),National Program for Basic S &T Platform Construction(no.2005DKA21201-10) and National Basic Research Program of China (2006CB100206,2006AA10A213).We thank Dr.

C.X.Man,T.X.Han and C.F.Tian for their assis-tance in using the SPSS software,and thank Dr.X.H.Sui for technical assistance with strain maintenance.E.T.Wang was supported by the CGPI grants of IPN,Mexico.

Appendix A.Supporting Information Supplementary data associated with this article can be found in the online version at doi:10.1016/ j.syapm.2008.10.004.

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361359

References

[1]N.Fierer,R.B.Jackson,The diversity and biogeography

of soil bacterial communities,Proc.Natl.Acad.Sci.103 (2006)626–631.

[2]D.Diouf,R.Samba-Mbaye,D.Lesueur,A.T.Ba,B.

Dreyfus,P.de Lajudie,M.Neyra,Genetic diver-sity of Acacia seyal Del.rhizobial populations indige-nous to Senegalese soils in relation to salinity and pH of the sampling sites,Microb.Ecol.54(2007) 553–566.

[3]O.M.Aguilar,M.V.Ver o′ica,P.M.Riccillo,The diversity

of rhizobia nodulating beans in Northwest Argentina as a source of more ef?cient inoculant strains,J.Biotech.91 (2001)181–188.

[4]T.Becki,M.G.Julie,H.G.Peter,Selection of rhizobia for

prairie legumes used in restoration and reconstruction programs in Minnesota,Can.J.Microbiol.50(2004) 977–983.

[5]P.Bogino, E.Banchio, C.Bon?glio,W.Giordano,

Competitiveness of a Bradyrhizobium sp.strain in soils containing indigenous rhizobia,Curr.Microbiol.56 (2008)66–72.

[6]P.Normand,https://www.360docs.net/doc/e72771300.html,pierre,L.S.Tisa,J.P.Gogarten,N.

Alloisio, E.Bagnarol, C.A.Bassi, A.M.Berry, D.M.

Bickhart,N.Choisne, A.Couloux, B.Cournoyer,S.

Cruveiller,V.Daubin,N.Demange,M.P.Francino,E.

Goltsman,Y.Huang,O.R.Kopp,https://www.360docs.net/doc/e72771300.html,barre, A.

Lapidus, https://www.360docs.net/doc/e72771300.html,vire,J.Marechal,M.Martinez,J.E.

Mastronunzio, B.C.Mullin,J.Niemann,P.Pujic,T.

Rawnsley,Z.Rouy, C.Schenowitz, A.Sellstedt, F.

Tavares,J.P.Tomkins,D.Vallenet,C.Valverde,L.G.

Wall,Y.Wang, C.Medigue, D.R.Benson,Genome characteristics of facultatively symbiotic Frankia sp.

strains re?ect host range and host plant biogeography, Genome Res.17(2007)7–15.

[7]C.F.Tian,E.T.Wang,T.X.Han,X.H.Sui,W.X.Chen,

Genetic diversity of rhizobia associated with Vicia faba in three ecological regions of China,Arch.Microbol.188 (2007)273–282.

[8]B.C.Hou,E.T.Wang,Y.Li,R.Z.Jia,W.F.Chen,C.X.

Man,X.H.Sui,W.X.Chen,Rhizobial resource asso-ciated with epidemic legumes in Tibet,Microb.Ecol.57 (2009)69–81.

[9]L.Moulin,A.Munive,B.Dreyfus,C.Boivin-Masson,

Nodulation of legumes by members of the beta-subclass of Proteobacteria,Nature411(2001)948–950.

[10]M.L.Zhang,A dispersal and vicariance analysis of the

genus Caragana,Fabr.J.Integrat.Plant Bio.47(2005) 897–904.

[11]Y.Z.Su,T.H.Zhang,Y.L.Li,F.Wang,Changes in soil

properties after establishment of Artemisia halodendron and Caragana microphylla on shifting sand dunes in semiarid Horqin sandy land,northern China,Environ.

Manage.36(2005)272–281.

[12]S.Y.Fan, B.Freedman,J.X.Gao,Potential environ-

mental bene?ts from increased use of bioenergy in China, Environ.Manage.40(2007)504–515.

[13]T.Xiang,T.Uno,F.Ogino,C.Ai,J.Duo,U.Sankawa,

Antioxidant constituents of Caragana tibetica,Chem.

Pharm.Bull.53(2005)1204–1206.[14]O.N.Allen, E.K.Allen,The Leguminosae:A Source

Book of Characteristics,Uses,and Nodulation,Univer-sity of Wisconsin Press,1981.

[15]W.X.Chen,E.T.Wang,S.Y.Wang,Y.B.Li,X.Q.Chen,

Y.Li,Characteristics of Rhizobium tianshanense sp.nov.,

a moderately and slowly growing root nodule bacterium

isolated from an arid saline environment in Xinjiang, People’s Republic of China,Int.J.Syst.Bacteriol.45 (1995)153–159.

[16]Z.Y.Tan, E.T.Wang,G.X.Peng,M.E.Zhu, E.

Martinez-Romero,W.X.Chen,Characterization of bac-teria isolated from wild legumes in the north-western regions of China,Int.J.Syst.Bacteriol.49(1999) 1457–1469.

[17]A.M.Yan,W.X.Chen,Phenotypic feature diversity of

rhizobia isolated from Medicaga sp.,Melilotus sp.and Caragana sp.,Biodiv.Sci.7(1999)1–8.

[18]L.F.Gao,Z.A.Hu,H.X.Wang,Genetic diversity of

rhizobia isolated from Caragana intermedia in Maowusu sandland,north of China,Lett.Appl.Microbiol.35 (2002)347–352.

[19]X.R.Yan,W.F.Chen,J.F.Fu,Y.L.Lu,C.Y.Xue,X.H.

Sui,Y.Li,E.T.Wang,W.X.Chen,Mesorhizobium spp.

are the main microsymbionts of Caragana spp.grown in Liaoning Province of China,FEMS Microbiol.Lett.271 (2007)265–273.

[20]J.M.Vincent,A Manual for the Practical Study of Root

Nodule Bacteria.International Biological Programme Handbook No.15,Blackwell Scienti?c Publications, Oxford,England,1970.

[21]P.van Berkum,Evidence for a third uptake hydrogenase

phenotype among the soybean bradyrhizobia,Appl.

Environ.Microbiol.56(1990)3835–3841.

[22]Z.Terefework,S.Kaijalainen,K.Lindstr o′m,AFLP

?ngerprinting as a tool to study the genetic diversity of Rhizobium galegae isolated from Galega orientalis and Galega of?cinalis,J.Biotechnol.91(2001)169–180. [23]Z.Y.Tan,X.D.Xu,E.T.Wang,J.L.Gao,E.Martinez-

Romero,W.X.Chen,Phylogenetic and genetic relation-ships of Mesorhizobium tianshanense and related rhizobia, Int.J.Syst.Bacteriol.47(1997)874–879.

[24]https://www.360docs.net/doc/e72771300.html,guerre,P.Mavingui,M.R.Allard,M.P.Charnay,

P.Louvrier,S.I.Mazurier,L.Rigottier-Gois,N.Amar-ger,Typing of rhizobia by PCR DNA?ngerprinting and PCR-restriction fragment length polymorphism analysis of chromosomal and symbiotic gene regions:application to Rhizobium leguminosarum and its different biovars, Appl.Environ.Microbiol.62(1996)2029–2036.

[25]E.T.Wang,P.van Berkum,D.Beyene,X.H.Sui,O.

Dorado,W.X.Chen, E.Martinez-Romero,Rhizobium huautlense sp.nov.,a symbiont of Sesbania herbacea that has a close phylogenetic relationship with Rhizobium galegae,Int.J.Syst.Bacteriol.48(1998)687–699. [26]J.Versalovic,M.Schneider,F.J.De Bruijn,J.R.Lupski,

Genomic?ngerprinting of bacteria using repetitive sequence-based polymerase chain reaction,Methods Mol.Cell.Biol.5(1994)25–40.

[27]G.Nick,M.Jussila,B.Hoste,R.M.Niemi,S.Kaijalai-

nen,P.De Lajudie,M.Gillis, F.J.De Bruijn,K.

Lindstr o¨m,Rhizobia isolated from root nodules of

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361 360

tropical leguminous trees characterized using DNA–DNA dot-blot hybridisation and rep-PCR genomic?ngerprint-ing,Syst.Appl.Microbiol.22(1999)287–299.

[28]L.Vauterin,P.Vauterin,Computer-aided objective

comparison of electrophoresis patterns for grouping and identi?cation of microorganisms,Eur.Microbiol.1(1992) 37–41.

[29]https://www.360docs.net/doc/e72771300.html,guerre,S.M.Nour,V.Macheret,J.Sanjuan,P.

Drouin,N.Amarger,Classi?cation of rhizobia based on nodC and nifH gene analysis reveals a close phylogenetic relationship among Phaseolus vulgaris symbionts,Micro-biology147(2001)893–981.

[30]K.Tamura,J.Dudley,M.Nei,S.Kumar,MEGA4:

molecular evolutionary genetics analysis(MEGA)soft-ware version4.0,Mol.Biol.Evol.24(2007)1596–1599.

[31]R.Muresu,E.Polone,L.Sulas,B.Baldan,A.Tondello,

G.Delogu,P.Cappuccinelli,S.Alberghini,Y.Benhizia,

H.Benhizia,A.Benguedouar,B.Mori,R.Calamassi,

F.B.Dazzo,A.Squartini,Coexistence of predominantly

non-culturable rhizobia with diverse,endophytic bacterial taxa within nodules of wild legumes,FEMS Microbiol.

Ecol.63(2008)383–400.

[32]https://www.360docs.net/doc/e72771300.html,guerre,M.R.Allard,F.Revoy,N.Amarger,Rapid

identi?cation of rhizobia by restriction fragment length polymorphism analysis of PCR-ampli?ed16S rRNA genes,Appl.Environ.Microbiol.60(1994)56–63. [33]J.C.Cho,J.M.Tiedje,Biogeography and degree of

endemicity of?uorescent Pseudomonas strains in soil, Appl.Environ.Microbiol.66(2000)5448–5456.

[34]M.Healy,J.Huong,T.Bittner,M.Lising,S.Frye,S.

Raza,R.Schrock,J.Manry, A.Renwick,R.Nieto, Microbial DNA typing by automated repetitive-sequence-based PCR,J.Clin.Microbiol.43(2005)199–207. [35]J.L.W.Rademaker,B.Hoste,F.J.Louws,K.Kersters,J.

Swings,L.Vauterin,P.Vauterin, F.J.de Bruijn, Comparison of AFLP and rep-PCR genomic?ngerprint-ing with DNA–DNA homology studies:Xanthomonas as

a model system,Int.J.Syst.Evol.Microbiol.50(2000)

665–677.

[36]F.Q.Wang, E.T.Wang,Y.F.Zhang,W.X.Chen,

Characterization of rhizobia isolated from Albizia spp.

in comparison with microsymbionts of Acacia spp.and Leucaena leucocephala grown in China,Syst.Appl.

Microbiol.29(2006)502–517.

[37]J.Liu, E.T.Wang,W.X.Chen,Diverse rhizobia

associated with woody legumes Wisteria sinensis,QJ;Cer-cis racemosa and Amorpha fruticosa grown in the temperate zone of China,Syst.Appl.Microbiol.28 (2005)465–477.

[38]A.Ulrich,I.Zaspel,Phylogenetic diversity of rhizobial

strains nodulating Robinia pseudoacacia L,Microbiology 146(2000)2997–3005.

[39]B.S.Weir,S.J.Turner,W.B.Silvester,D.C.Park,J.M.

Young,Unexpectedly diverse Mesorhizobium strains and Rhizobium leguminosarum nodulate native legume genera of New Zealand,while introduced legume weeds are nodulated by Bradyrhizobium Species,Appl.Environ.

Microbiol.70(2004)5980–5987.

[40]T.X.Han,E.T.Wang,L.L.Han,W.F.Chen,X.H.Sui,

W.X.Chen,Molecular diversity and phylogeny of

rhizobia associated with wild legumes native to Xinjiang, China,Syst.Appl.Microbiol.(2008).

[41]C.X.Man,H.Wang,W.F.Chen,X.H.Sui,E.T.Wang,

W.X.Chen,Diverse rhizobia associated with soybean grown in the subtropical and tropical regions of China, Plant Soil310(2008)77–87.

[42]W.F.Chen,S.H.Guan,C.T.Zhao,X.R.Yan,C.X.Man,

E.T.Wang,W.X.Chen,Different Mesorhizobium species

associated with Caragana carry similar symbiotic genes and have common host ranges,FEMS Microbiol.Lett.

283(2008)203–209.

[43]T.Ueda,Y.Suga,N.Yahiro,T.Matsuguchi,Phylogeny

of Sym plasmids of rhizobia by PCR-based sequencing of

a nodC segment,J.Bacteriol.177(1995)468–472.

[44]M.Kalita,T.Stepkowski, B.Lotocka,W.Malek,

Phylogeny of nodulation genes and symbiotic properties of Genista tinctoria bradyrhizobia,Arch.Microbiol.186 (2006)87–97.

[45]G.Moschetti,A.Peluso,A.Protopapa,M.Anastasio,O.

Pepe,R.Defez,Use of nodulation pattern,stress tolerance,nodC gene ampli?cation,RAPD-PCR and RFLP-16S rDNA analysis to discriminate genotypes of Rhizobium leguminosarum biovar viciae,Syst.Appl.

Microbiol.28(2005)619–631.

[46]S.Sarita,P.K.Sharma,U.B.Priefer,J.Prell,Direct

ampli?cation of rhizobial nodC sequences from soil total DNA and comparison to nodC diversity of root nodule isolates,FEMS Microbiol.Ecol.54(2005)1–11.

[47]J.Donate-Correa,M.Leon-Barrios,M.Hernandez,R.

Perez-Galdona,M.del Arco-Aguilar,Different Mesorhi-zobium species sharing the same symbiotic genes nodulate the shrub legume Anagyris latifolia,Syst.Appl.Micro-biol.30(2007)615–623.

[48]X.Perret,C.Staehelin,W.J.Broughton,Molecular basis

of symbiotic promiscuity,Microbiol.Mol.Biol.Rev.64 (2000)180–201.

[49]A.Rincon,F.Arenal,I.Gonzalez,E.Manrique,M.M.

Lucas,J.J.Pueyo,Diversity of rhizobial bacteria isolated from nodules of the gypsophyte Ononis tridentata L.

growing in Spanish soils,Microb.Ecol.(2007)223–233.

[50]E.Giraud,L.Moulin,D.Vallenet,V.Barbe,E.Cytryn,

J.C.Avarre,M.Jaubert,D.Simon,F.Cartieaux,Y.Prin,

G.Bena,L.Hannibal,J.Fardoux,M.Kojadinovic,L.

Vuillet,https://www.360docs.net/doc/e72771300.html,jus,S.Cruveiller,Z.Rouy,S.Mangenot,B.

Segurens, C.Dossat,W.L.Franck,W.S.Chang, E.

Saunders, D.Bruce,P.Richardson,P.Normand, B.

Dreyfus,D.Pignol,G.Stacey,D.Emerich,A.Vermeglio,

C.Medigue,M.Sadowsky,Legumes symbioses:absence

of Nod genes in photosynthetic bradyrhizobia,Science 316(2007)1307–1312.

[51]R.Rivas,https://www.360docs.net/doc/e72771300.html,ranjo,P.F.Mateos,S.Oliveira, E.

Martinez-Molina,E.Velazquez,Strains of Mesorhizobium amorphae and Mesorhizobium tianshanense,carrying sym-biotic genes of common chickpea endosymbiotic species, constitute a novel biovar(ciceri)capable of nodulating Cicer arietinum,Lett.Appl.Microbiol.44(2007)412–418. [52]J.L.Gao,Z.Terefework,W.X.Chen,K.Lindstrom,

Genetic diversity of rhizobia isolated from Astragalus adsurgens growing in different geographical regions of China,J.Biotechnol.91(2001)155–168.

Y.L.Lu et al./Systematic and Applied Microbiology32(2009)351–361361