Artículos de investigación

 

Abundance and diversity of legume nodulating Rhizobia in soils of Embu District, Kenya

 

Abundancia y diversidad de Rhizobia productora de nódulos en leguminosas en los suelos del Distrito de Embu, Kenia

 

G. M. Mwenda^a*, N.K. Karanja^c, H. Boga^a, J.H.P. Kahindi^b, A. Muigai^a and D. Odee^d

 

^a Jomo Kenyatta University of Agriculture and Technology, P.O. Box 62000-00200, Nairobi *Corresponding author Email: mutegi@gmail.com

^b United States International University, P.O Box 14634-00800, Nairobi.

^c University of Nairobi, P.O Box 30197-00100, Nairobi.

^d Kenya Forestry Research Institute, P.O. Box 20412-00200, Nairobi, Kenya.

 

Submitted February 16, 2010
Accepted July 7, 2010
Revised received July 12, 2010

 

Abstract

A major strategy towards addressing soil fertility depletion is the conservation and sustainable use of rhizobia that are able to fix nitrogen in the soil in association with legumes. The study assessed abundance and diversity of legume nodulating rhizobia (LNB) in soils collected from six different land use systems in Embu District, Kenya. The populations were estimated by the most-probable-number (MPN) plant infection technique using Macroptilium atropurpureum (DC.) Urban (Siratro) as the trap host species. Symbiotic effectiveness was measured for the isolates in association with Siratro. Isolated rhizobia were characterized morphologically and genetically by PCR-RFLP and partial sequencing of 16S rRNA genes.

The LNB populations in soils collected from the different land uses in Embu ranged from 0 to 2.3 x 10² cells g^-1 soil. There was apparent land use effect on abundance of LNB with fallow system giving high abundance. A total of 250 pure isolates were obtained from the root nodules of Siratro trap plants. The isolates were characterized on yeast extract mannitol mineral salts agar (YEMA) media containing bromothymol blue and grouped into fast growers (acid-producing) and slow growers (alkali-producing) (70% and 30 % of isolates respectively). PCR-RFLP analysis categorised the rhizobia into five species in the genera Rhizobium, Bradyrhizobium, Mesorhizobium and Agrobacterium. Land use system under tea had four of the five species found in the area whereas natural forests had two species. Land use significantly impacted on the diversity of rhizobia (PO.05) with soils under tea having the highest diversity with a mean Shannon diversity index of 1.304 compared to the lowest (0.297) recorded in natural forest. Isolated rhizobia strains formed effective nodules on Siratro. However, the level of nitrogen fixation varied among isolates while symbiotic efficiency ranged from 27-112%.

The findings indicate that abundance and diversity of rhizobia does not necessarily decrease with agricultural intensification as hypothesized but recommends further studies to obtain a clearer understanding of the relationship between soil rhizobia diversity and land use and management.

Key words: Rhizobia; most-probable-number (MPN); trap host; Macroptilium atropurpureum (Siratro); symbiotic efficiency.

 

INTRODUCTION

To achieve improved and sustainable agriculture, emphasis must be on the use and effective management of internal resources such as those presented by rhizobia-legume associations (Sparks, 2002). It is estimated that legume N₂ fixation accounts for about 40% of global N₂ fixation (Brockwell and Bottomley, 1995), but despite N₂ fixation by legume-rhizobia symbiosis being one of the most important processes in nature, rhizobia diversity is yet to be fully documented. Diverse indigenous rhizobia are present in various ecosystems of the world (Xu et al, 1995; Chen et al, 1997; Peng et al, 2002) and present opportunities for exploitation through the development of inoculants. The central highlands of Kenya, like many other parts of the world face the challenge of land fragmentation and consequent intensification of farming activities due to high population growth. Many agricultural practices, such as crop rotation, continuous cropping, and tillage, induce changes in microbial communities in soil (Lupwayi et al, 1998; Alvey et al, 2003) but specific microbial groups may respond differently. Tools to accurately and rapidly characterize rhizobia have only been developed over the last two decades and the impact of land management practices on the diversity of rhizobia is only beginning to be studied, especially in the tropics (Depret et al, 2004; Ngokota et al, 2008). There are several reports on studies on natural nodulation of agricultural pasture and grain legumes in cropping systems of Kenya (McDonald, 1935, Bumpus, 1957; Morrison, 1966; Souza, 1969). Most of these legumes have been reported to nodulate with varying levels of nodulation intensity- from poor or erratic to very profuse nodulation. However, these earlier studies did not quantify the abundance and characteristics of indigenous LNB populations. Odee et al. (1995) surveyed natural nodulation and determined the abundance of indigenous populations in a wide spectrum of agro-ecological zones mostly from indigenous woodlands. The isolates from these systems showed a wide range of phenotypic and genetic diversity, which also indicated that most of the described genera were present (Odee et al, 1997, 2002). This study assessed abundance and diversity of rhizobia in different land-use systems in soils from Embu district in Kenya.

 

MATERIAL AND METHODS

Study area and selection of sampling points

The area of study was Embu district. The District lies approximately between latitudes 0° 8 and 0° 35 South and longitudes 37° 19 and 37° 40 East and is divided into five divisions. The Central point of the study area (Mt. Kenya Forest near Irangi market and bordering agricultural lands) was traversed by longitude 37° 28 E and latitude 0° 20. Altitudes for the area range between 1500m and 4500 m. The soils within the area are well drained, extremely deep, dusky red to dark brown, friable soils with humic top soils. The soils are mainly classified as humic nitisols (FAO, 1989). Six land use types were identified as prevalent in the area. These were; maize-bean intercrop, tea; napier grass (Pennisetum purpureum); coffee (Coffea arábica); fallow or pasture; and natural undisturbed forest. Stratification of the area was done based on the land uses and sampling points, 200m apart, allocated on a systematic grid.

Soil samples were obtained from sampling points represented by a radius of 3 m and 6 m collected at a depth of 0-20 cm. Each soil was collected aseptically to avoid cross-contamination between soils from different sampling points. A composite soil comprising 24 cores per sampling point and weighing approximately 500 g was transported to the laboratory within the shortest time possible. Nodules of legumes growing in the sampling points were also collected. Soil sampling was undertaken to represent land use systems .A total of 60 sampling points in 3 windows were sampled representing various land-use systems.

 

Enumeration of LNB

Indigenous LNB populations were determined using the plant infection technique (Somasegaran and Hoben, 1994). Each composite soil sample was mixed thoroughly and quartered. Soil inocula were prepared by suspending 10 g of soil sample in 90 ml of sterile water in a 160-ml dilution bottle and shaken for 20 min in a wrist-action shaker at room temperature (~25°C). One ml of each suspension was aseptically pipetted into 9 ml sterile water diluents in McCartney bottle and shaken for 2 min. The resulting suspension was serially-diluted tenfold from 10^-1 to 10^-6 with four replications at each dilution level. Aliquots of one ml were used to inoculate 3-5 day old siratro seedlings previously pre-treated, germinated and aseptically transferred to sterile plastic pouches containing N-free nutrient solution (Broughton and Dilworth, 1971) with two plants per pouch. The pouches were supported in improvised racks. The plants were grown for 28 days in a glasshouse at temperature 30/18 °C (day/night) and natural light of ca. 12 h photoperiod. The number of nodulated plants at each dilution was recorded and used as an ordered code (from low to high dilution) and used to estimate the most-probable-number (MPN). The computer program, Most Probable Number (MPNES) by Bennet et al. (1990) was used to calculate the populations.

 

Isolation of LNB from nodules and characterization

All nodules were freshly isolated. Nodules were surface sterilized in 1 % NaOCl for 6 min, rinsed in several changes of sterile water, and then crushed with a flame-sterilized blunt-tipped pair of forceps. A loopful of the crushed nodule was then streaked across the surface of Petri dish containing yeast extract mannitol mineral salts agar (YEMA) as described by Vincent (1970). Some nodules had dual or multiple nodule occupancy; not all nodules produced isolates. Typical well-isolated colonies were re-isolated and characterized on YEMA containing 25 mg kg^-1 (w/v) bromothymol blue (BTB) as a pH reaction indicator. In addition, the growth of the isolates was characterized by the rate of colony emergence on YEMA/BTB media incubated at 28°C. Fast- and slow- growing LNB were described as emerging after 3-5 and 7- 10 days following inoculation, respectively. All isolates were stored in 16% glycerol yeast mannitol broth (YMB) at -70°C.

 

Symbiotic efficiency

Symbiotic efficiency was determined for isolates as described by Somasegaran and Hoben (1994). Isolates were used to inoculate Siratro in modified Leonard jars using vermiculate and nitrogen-free nutrient solutions. Non-inoculated nitrogen-free and nitrogen-supplemented plants were used as controls. Plants were grown in green houses for 8 weeks and symbiotic effectiveness determined according to Gibson (1987): Shoot dry weight (SDW) inoculated plants/SDW non-inoculated nitrogen supplemented control plants (140 ppm. nitrogen as KNO₃).

 

RFLP-PCR analysis of LNB isolates.

DNA was extracted using a standard phenol-chloroform procedure (Hermann and Frischauf, 1987). Genomic DNA obtained was then used as template for amplification of the 16S rRNA gene. Nearly full-length 16S rRNA genes were amplified using 8F (5'-AGAGTTTGATCATGGCTCAG-3') and 1492R (5'-GGTTACCTT GTTACGACTT-3') primers. Amplification was carried out in a 40 μl mixture with 1 μl template DNA, 1 ul dNTP (2mM), 0.8 μl of each primer (10mM), 4.8 μl MgC12 (25mM), 4 μl 10x PCR buffer (Biolabs), 0.4 μl Taq DNA polymerase (Biolabs) and 28 ul sterile PCR water. DNA was amplified in a 9800 Fast Thermal Cycler from Applied Biosystems programmed as follows: denaturation of DNA at 94°C for 5 min; 35 cycles of denaturation (45 s at 94°C), annealing (50 s at 55°C) and extension (90 s at 72°C) with a final extension time of 8 min at 72°C. Amplification products were visualized by horizontal gel electrophoresis on a 1% (w/v) agarose gel stained with Ethidium Bromide run in TBE (Tris-borate-EDTA) buffer at 80V for 60 minutes (Wang et al, 1999).PCR products were digested with Hae III restriction enzyme (Promega Corporation, Madison, USA) according to manufacturer's instructions. The DNA fragments were separated and visualized by gel electrophoresis on a 2% (w/v) agarose gel stained with ethidium bromide run in TBE (Tris-borate-EDTA) buffer at 80V for 60 minutes. The different banding patterns were noted, and the frequency of similar patterns was scored (Wang et al, 1999).Representative PCR products were purified using QIAquick PCR purification kit (Qiagen, Tiangen, China) according to the manufacturer's instruction and sequenced directly as reported previously (Hurek et al., 1997) using 8F and 1492R primers. Sequences were edited using Chromas software. The 16S rRNA gene sequences were compared to sequences in the public database using Basic Local Alignment Search Tool (BLAST) on the National Center for Biotechnology Information (NCBI) website in order to determine similarity to sequences in the Genebank database (Shayne et al, 2003). The 16S rRNA gene sequences with high similarities to those determined in the study were retrieved and added to the alignment (Clustal W) based on BLAST results. The evolutionary history was inferred using the Neighbor-Joining method (Saitou & Nei, 1987). The evolutionary distances were computed using the Maximum Composite Likelihood method (Tamura et al., 2004) and Phylogenic analyses were conducted in MEGA4 (Tamura et al., 2007). Bootstrap for 500 replicates was performed to attach confidence estimates for the tree topologies (Felsenstein, 1985).

 

RESULTS AND DISCUSSION

Indigenous LNB populations

The populations of indigenous LNB nodulating siratro varied with the land use systems (Table 1). Among the land use systems, coffee, tea, maize intercrop and fallow recorded a mean range of 1.1 -2.3 X 10² cells g^-1 soil. Land use systems under napier and natural forests had 6.1 x 10 and 0 cells g^-1 soil, respectively. However, some isolates were recovered from nodules collected from leguminous plants growing at sites in the natural forest.

The population sizes determined in these sampling points were within the ranges reported for LNB associated with native woody legumes (mainly Acacia spp.) occurring in diverse ecological regions of Kenya (Odee et al., 1995). It has also been demonstrated that indigenous common bean nodulating LNB occur in acid soils (pH ≤ 4.5) of Kenya and have a broad host range that include siratro (Anyango et al., 1995) which was used as the trap crop in this study. The soils in this study were classified acidic with pH ranging from 3.2 - 4.9. A total of 250 pure isolates were recovered from nodules of the trap crop and from nodules of field plants which reflected abundance and diversity of LNB in Embu. The pure isolates were characterized by growth rate on YEMA supplemented with BTB resulting into two major growth rate types namely; fast growers (acid-producing) and, slow growers (alkali-producing) which constituted 70% and 30 % of the isolates respectively. However, previous studies using cultural characteristics have shown that characterization of tropical LNB populations do not always conform to these groups due their diverse nature (Zhang et al, 1991; Odee et al, 1997; Bala et al, 2004).

 

Symbiotic efficiencies

Symbiotic efficiencies (SE) of 100 of the isolates differed significantly (p<0.005). SE ranged from a low 27% to a high of 112% (Figure 1). Sixty seven percent of the isolates had an SE of above 50%. Laranjo et al. (2001) tested thirty two rhizobia isolates for their SE with a winter variety chickpea and found only 9% to have an SE of above 50%. In yet another study with Portuguese isolates, none of thirty nine isolates tested had an SE of above 50% (Laranjo et al, 2002). Isolates tested in this study had good SE in comparison to those in studies elsewhere.

 

Ribotyping of isolates

Restriction of amplified 16S rRNA regions with Hae III generated a total seven ribotypes which were named T1 to T7. Tl and T6 were the most abundant of the ribotypes with 46.5% of isolates giving the either of the two ribotypes. T7 was the least common (Table 2).

Soils under tea had the highest total ribotype richness. Tea had five of the seven ribotypes. Land under napier grass and maize-based intercrop had four ribotypes each. Natural forests had the least number of ribotypes with only two ribotypes. Ribotype T1 was found in five of the six land uses lacking only in the natural undisturbed forest. Diversity of rhizobia as measured by the Shannon diversity index was highest in soils under tea and lowest in natural forest. Diversity as measured by this index was significantly different (p<0.001) among the land use types (Table 2).

Differences in evenness were significant (p<0.001) among the six land uses tested. Evenness in the occurrence of ribotypes was highest in napier grass and lowest in maize based intercrop (Fig 2). As expected, napier grass and natural forests were the most even. In comparison to maize, coffee and tea, napier grass and natural forests represent stable ecosystems that are less subject to anthropogenic disturbances.

Detection of rhizobia ribotypes increased with increase in number of soil samples taken (Fig. 3). However, as the curve indicates, all possible ribotypes were recovered in 20 samples, meaning that processing of additional samples would have yielded no further ribotypes.

 

Phylogenetic characterisation of isolates

Phylogenetic analysis showed that three of the ribotypes clustered within the Rhizobium branch, one within the Mesorhizobium lineage, one within the Bradyrhizobium group, while two were related to the Agrobacterium lineage (Fig. 4). Closest relatives of sequenced isolates are shown in Table 4. Within the Rhizobium group, ribotype T1 and T4 clustered with R. tropici but on different sub-branches while T6 was clustered withi?. leguminosarum. Ribotype T2 and T5 clustered on a unique branch. The representative isolate for T5 shared a 100% sequence similarity to both an Agrobacterium strain and a Rhizobium sp. strain while T2 had only a 92% sequence similarity to any published sequence. Type T2 and T5 shared equal similarity with Agrobacterium and Rhizobium sp. strains, but were clearly on a distinct branch from other Rhizobium species and therefore this lineage was regarded as Agrobacterium. Within the Mesorhizobium lineage, ribotype T7 formed a lineage with M. loti. Lastly, T3 clustered on the Bradyrhizobium branch with close affiliation to B. japonicum.

Differences in diversity seen among the land use types can be the result of a number of factors. The first is soil pH, as already suggested for Rhizobium populations by Harrison et al. (1989). Rhizobium populations were described as low in acid soils and high from soils with higher pH. In a separate study, Muya et al. (2009) found mean soil pH of the six land use types studied here do differ significantly (P<0.05). The land use types had mean soil pH ranging from a high of 4.28±0.04 in fallow fields to a low of 3.52±0.09 in tea plantations. Napier grass, fallow and maize-based fanning systems had the highest mean soil pH and also ranked among the top three land use types with highest diversity (Table 3). Excluding data on tea, diversity strongly correlated with pH (r=0.80), as recorded by Muya et al. (2009). Tea plantations had the lowest soil pH and yet were found to have the highest rhizobial biodiversity here. This may due to some crop related factor. Venkateswarlu et al. (1997) reported that crop related factors have more critical influence on the abundance of native rhizobial population than soil or climatic factors. Further supporting the crop-related factor theory are studies by Ngokota et al. (2008) and Depret et al. (2004). Ngokota et al. (2008) found rhizobia diversity to be highest in Cocoa monoculture from among several land use systems that included mixed fanning systems whereas Depret et al. (2004) reported highest level of diversity in soils under wheat monoculture. Soil nitrogen content has also been shown to influence diversity of rhizobia in soils. High levels of nitrogen in the soil are thought to decrease the diversity of rhizobia in the soil (Hirsch, 1996; Palmer & Young, 2000). In the area of study, the seven land use types had significantly different mean amounts of soil nitrogen (P<0..05) (Muya et al, 2009) and diversity was negatively co-related with soil n (r= -0.72).

Another possible explanation to diversity patterns observed was that, in the more cultivated areas, rhizobia may have been introduced together with legumes seeds or as inoculants. This has already been reported (Perez-Ramirez et al, 1998). The natural forest did not have any unique strains. All rhizobia groups identified were present in at least two land types. Origin of rhizobia found was not investigated and can only be speculated. But with the presence of legumes such as Phaseolus vulgaris in some of the land uses, the possibility of recent introduction with planting seeds or as inoculum cannot be ruled out. The common bean is a promiscuous host plant that can be nodulated by a wide range of rhizobia including most found in the study area (Sawada et al, 2003)

Soil amendments, which vary with land use type, also influence rhizobia diversity. Natural forests represent a land use system with relatively stable plant population. Arable soils of land use systems such as maize-based mixed systems and tea are subject to higher levels of soil amendments, fertilizers, herbicides, and pesticides than the Natural forest and had greater diversity of rhizobia. It is known that rhizobial numbers are affected by soil amendments, such as manure, lime, fertilizer application, phosphate (Lowendorf, 1980; Caballero-Mellado & Martinez-Romero, 1999; Anthony et ah, 2001).

 

CONCLUSION

This work has demonstrated the occurrence of varying population levels of LNB in soils from Embu district in Kenya, which were largely affected by the land use systems. Two major LNB groups were characterised: fast and slow growers were obtained. Diversity of the rhizobia isolates was observed.

 

ACKNOWLEDGEMENTS

Financial support from the Conservation and Sustainable Management of Below Ground Biological Diversity (CSM-BGBD) Project, through funds from the Global Environment Facility (GEF) is gratefully acknowledged.

 

REFERENCES

Alvey, S., Yang, C.H., Buerkert, A., Crowley, D.E. 2003. Cereal/legume rotation effects on rhizosphere bacterial community structure in West African soils. Biology and Fertility of Soils. 37:73-82.         [ Links ]

Anthony, G., Seaman, M., Maeral, A., Waite, I., Davies, J. 2001. Plant fertilizers as drivers of change in microbial community structure and function in soils. Plant and Soil. 232: 135-145.         [ Links ]

Anyango, B., Wilson, K.J., Beynon, J.L., Giller, K.E. 1995. Diversity of rhizobia nodulating Phaseouls vulgaris L. in two Kenyan soils with contrasting pHs. Applied and Environmental Microbioliogy. 61: 4016-4021.         [ Links ]

Bala, A., Murphy, P.J., Giller, K.E. 2004. Classification of tropical tree rhizobia based on phenotypic characters forms nested clusters of phylogenetic groups. West African Journal of Applied Ecology. 6: 9-19.         [ Links ]

Bennett, J.E., Woomer, P.L., Yost, R.S. 1990. Users manual for MPNES most-probable-number enumeration system ver. 1.0. NifTAL project and University of Hawaii.         [ Links ]

Brockwell, J., Bottomley, P. 1995. Recent advances in innoculant technology and prospects for the future. Soil Biology and Biochemistry. 27: 683-697.         [ Links ]

Broughton, W.J., Dilworth, M.J. 1971. Control of leghaemoglobin syntheisis in snake beans. Biochemical Journal. 125: 1075-1080.         [ Links ]

Bumpus, E.D. 1957. Legume nodulation in Kenya. East African Agricultural and Forestry Journal. 23:91-99.         [ Links ]

Caballero-Mellado, J., Martinez-Romero, E. 1999. Soil fertilization limits the genetic diversity of Rhizobium in bean nodules. Symbiosis. 26: 11-121.         [ Links ]

Chen, W.X., Tan, Z.Y., Gao, J.L., Li, Y., Wang, E.T. 1997. Rhizobiumn hainanense sp.nov., isolated from tropical legumes. International Journal of Systematic Bacteriology. 47: 870-873.         [ Links ]

Depret, G., Houot, S., Allard, M., Marie-Christine, M., Laguerre, G. 2004. Long-term effects of crop management on Rhizobium leguminosarum biovar viciae populations. FEMS Microbiology Ecology. 51: 87-97.         [ Links ]

Felsenstein, J. 1985. Confidence limits on phytogenies: An approach using the bootstrap. Evolution. 39: 783-791.         [ Links ]

Food and Agricultural Organization ( FAO). 1989. Forestry and Food Security, FAO Forestry Paper, FAO 90:1-2.         [ Links ]

Gibson, A. H. 1987. Evaluation of nitrogen fixation by legumes in the greenhouse and growth chamber. In A. H. Gibson, Symbiotic Nitrogen Fixation Technology (pp. 321-363). New York: Marcel Dekker.         [ Links ]

Harrison, S., Jones, D., Young, J. P. 1989. Rhizobium population genetics: genetic variation within and between populations from diverse locations. Journal of General Microbiology. 135: 1061-1069.         [ Links ]

Herrmann, B. , Frischauf, A.M. 1987. Molecular cloning techniques. Methods in Enzymolology. 152: 180-183.         [ Links ]

Hirsch, P. R. 1996. Population dynamics of indigenous and genetically modified rhizobia in the field. New Phytologist. 133: 159-171.         [ Links ]

Hurek, T., Wagner, B., Reihold-Hurek, B. 1997. Identification of N2-fixing plant-and fungus-associated Azoarcus species by PCR-based genomic fingerprints. Applied and Environmental Microbiology. 63: 4331-4339.         [ Links ]

Laranjo, M, Branco, C, Alho, L., Carvalho, M. D. 2002. Comparison of chickpea rhizobia isolates from diverse Portuguese natural populations based on symbiotic effectiveness and DNA fingerprint. Journal of Applied Microbiology. 92: 1043-1050.         [ Links ]

Laranjo, M., Rodrigues, R., Alho, L., Oliviera, S. 2001. Rhizobia of Chickpea from Southern Portugal: symbiotic efficiency and genetic diversity. Journal of Applied Microbiology. 90: 662-667.         [ Links ]

Lowendorf, H. S. 1980. Factors affecting survival of Rhizobium in soil. Advanced Microbial Ecology. 4: 87-124.         [ Links ]

Lupwayi, N. Z., Rice, W. A., Clayton, G. W. 1998. Soil microbial diversity and community structure under wheat as influenced by tillage and crop rotation. Soil Biology and Biochemistry. 30:1733-1741.         [ Links ]

McDonald, J. 1935. The inoculation of leguminous crops. East African Agricultural and Forestry Journal. 1:8-13.         [ Links ]

Morrison, J. 1966. Productivity of grass and grass/legume swards in Kenya highlands. East African Agricultural and Forestry Journal. 32: 19-24.         [ Links ]

Muya, E.M., Karanja, N., Okoth, P.F.Z., Roimen, H., Munga'tu, J., Mutsotso,B., Thuranira, G. 2009. Comparative description of land use and characteristics of belowground biodiversity benchmark sites in kenya. Tropical and Subtropical Agroecosystems. 11:263-275.         [ Links ]

Ngokota, L., Krasova-wade, T., Etoa, F. X., Sylla, D., Nwaga, D. 2008. Genetic diversity of rhizobia nodulating Arachis hypogaea L. in diverse land use systems of humid forest zone in Cameroon. Applied Soil Ecology. 40: 411-416.         [ Links ]

Odee, D.W., Haukka, K., Mclnroy, S.G., Sprent, J.I., Sutherland, J.M., Young,J.P.W. 2002. Genetic and symbiotic characterization of rhizobia isolated from tree and herbaceous legumes grown in soils from ecologically diverse sites in Kenya. Soil Biology and Biochemistry. 34: 801-811.         [ Links ]

Odee, D.W., Sutherland, J.M., Kimiti, J.M., Sprent, J.I. 1995. Natural populations and nodulation status of woody legumes growing in diverse Kenyan conditions. Plant Soil. 173:211-224.         [ Links ]

Odee, D.W., Sutherland, J.M., Makatiani, E.T., Mclnroy, S.G., Sprent, J.I. 1997. Phenotypic characteristics and composition of rhizobia associated with woody legumes growing in diverse Kenyan conditions. Plant Soil. 188: 65-75.         [ Links ]

Palmer, K. M., Young, J. P. W. 2000. Higher diversity of Rhizobium leguminosarum Biovarviciae in arable soils than in grass soils. Applied and Environmental Microbiology. 66: 2445-2450.         [ Links ]

Peng, G.X., Tan, Z.Y., Wang, E.T., Reinhold-Hurek, B., Chen, W.F., Chen, W.X. 2002. Identification of isolates from soybean nodules in Xinjiang region as Sinorhizobium xinjiangense and genetic differentiation of S. xinjiangense from Sinorhizobium fredii. International Journal of Systematic and Evolutionary Microbiology. 52: 457-462.         [ Links ]

Perez-Ramirez, N., Rogel-Herndez, M., Wang, E., Martinez-Romero, E. 1998. Seeds of Phaseolus vulgaris bean carry Rhizobium etli. Microbial Ecology. 26: 289-296.         [ Links ]

Saitou, N., & Nei, M. 1987. The neighbour-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution. 4: 406-425.         [ Links ]

Sawada, H., kuykendall, D., Young, J. 2003. Changing concepts in the systematics of bacterial nitrogen-fixing legume symbionts. Journal of General and Applied Microbiology. 49:155-179.         [ Links ]

Shayne, J. J., Hugenholtz, P., Sangwan, P., Osborne, C, Jansen, H. P. 2003. Laboratory cultivation of widespread and previously uncultured bacteria. Applied and Environmental Microbiology. 69: 7211-7214.         [ Links ]

Somasegaran, P., Hoben, H. J. 1994. Handbook for Rhizobia: Methods in Legams-Rhizobium technology. Berlin: Springer-Verlag.         [ Links ]

Souza, D.I.A. de. 1969. Legume nodulation and nitrogen fixation studies in Kenya. East African Agricultural and Forestry Journal. 34: 299-305.         [ Links ]

Sparks, D. 2002. Advances in Agronomy (6 ed., Vol. 76). Chicago: University of Chicago Press.         [ Links ]

Tamura, K., Dudley, I, Nei, M, Kumar, S. 2007. MEGA4: Molecular Evolutionary genetics Analysis (MEGA) software version 4.0. Molecular Biology and Evolution. 24: 1596-1599.         [ Links ]

Tamura, K., Nei, M., Kumar, S. 2004. Prospects for inferring very large phytogenies by using the neighbour-joining method. National Academy of sciences Conference (pp. 11030-11035). New York: National Academy of Sciences.         [ Links ]

Venkateswarlu, B., Hari, K., Katyal, J.C. 1997. Influence of soil and crop factors on the native rhizobial populations in soils under dryland fanning. Applied Soil Ecology. 7:1-10.         [ Links ]

Vincent, J.M. 1970. A manual for the practical study of root-nodule bacteria. Blackwell, Oxford.         [ Links ]

Wang, E. T., Van Berkum, P., Beyene, D., Sui, X., Chen, W., Martinez-Romero, E. 1999. Diversity of rhizobia associated with Amorpha isolated from Chinese soils and description of Mesorhizobium amorphae sp. International Journal of Systematic Bacteriology. 49: 51-65.         [ Links ]

Xu, L.M., Ge, C, Cui, Z., Li, I, Fan, H. 1995. Bradyrhizobium liaoningense sp. nov., isolated from the root nodules of soybeans. International Journal of Systematic Bacteriology. 45: 706-711.         [ Links ]

Zhang, X., Harper, R., Karsisto, M., Lindstrom, K., 1991. Diversity of Rhizobium bacteria isolated from the root nodules of leguminous trees. International Journal of Systematic Bacteriology. 41: 104-113.         [ Links ]