- Email: [email protected]
The diversity of rhizobia nodulating beans in Northwest Argentina as a source of more efficient inoculant strains
Journal of Biotechnology 91 (2001) 181– 188 www.elsevier.com/locate/jbiotec
The diversity of rhizobia nodulating beans in Northwest Argentina as a source of more efficient inoculant strains O. Mario Aguilar *, M. Vero´nica Lo´pez, Pablo M. Riccillo Instituto de Bioquı´mica y Biologı´a Molecular, Facultad de Ciencias Exactas, Uni6ersidad Nacional de La Plata, Calles 47 y 115, 1900 La Plata, Argentina Received 24 November 2000; received in revised form 30 March 2001; accepted 4 April 2001
Abstract The common bean (Phaseolus 6ulgaris L.) is cultivated widely in Central and South America and particularly in the Northwest of Argentina. In order to describe the diversity of the common bean nodulating rhizobial population from the bean producing area in Northwest Argentina (NWA), a collection of about 400 isolates of common beans recovered from nodules and soil samples from NWA were characterized by using nifH-PCR, analysis of genes coding for 16S rRNA and nodC, and REP-fingerprinting, respectively. It was found that species Rhizobium etli is predominant in common bean nodules although a high degree of diversity was found within the species. Other bean nodulating genotypes recovered from soils by using Leucaena sp. as the trapping host was found to have the 16S rDNA alleles of species such as Sinorhizobium fredii, Sinorhizobium saheli, Sinorhizobium teranga, Mesorhizobium loti, and Rhizobium tropici. Some of the bean genotypes that were found to be more efficient in green house experiments were selected and assayed in two successive bean-cropping seasons in the field environment in NWA, and an increase in yields with inoculation was found. The performance of strains isolated from the region indicates potential for exploiting the diversity. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Bean rhizobia; Diversity; Inoculation; Yields
1. Introduction Common beans (Phaseolus 6ulgaris L.) are an important crop in South America and in Africa, supplying in some cases up to 20% of the protein intake per person. Beans are grown in a great variety of farming systems, ranging from a highly mechanized cropping in Northwest Argentina * Corresponding author. Tel.: + 54-221-4250497. E-mail address: [email protected] Aguilar).
(O.M.
(NWA) to the low-input subsistence cultivation in Africa and Central America. Generally, the nitrogen requirements of legumes can be met either by the inorganic N from the soil or from the chemical fertilizers applied as well as by biological nitrogen fixation (BNF) which results from the symbiotic association with the soil bacteria rhizobia. The rhizobial isolates that are able to nodulate bean roots form a very heterogeneous group. Until now five species have been described; however some other isolates, all of which are also able to nodulate common beans with different degrees
0168-1656/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 5 6 ( 0 1 ) 0 0 3 3 6 - 4
182
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
of effectiveness, still appear to represent other distinct phylogenetic lineages (Martı´nez-Romero et al., 1991; Segovia et al., 1993; Amarger et al., 1997; Herrera-Cervera et al., 1999). Species Rhizobium tropici, which is able to nodulate a broad range of tropical trees including the Leucaena spp., was first isolated in South America but, has also been associated with P. 6ulgaris in France by Amarger et al. (1994) and in acid soils in Kenya by Anyango et al. (1995). The predominance of the Rhizobium etli-like 16S rRNA allele in the native rhizobial populations associated to wild beans in NWA was reported by Aguilar et al. (1998b). It is generally accepted that P. 6ulgaris L. is native to the Americas. Domestication of wild beans took place independently in the Mesoamerican center of origin (Mexico, Central America and Colombia) and in the Andean center in South America (Gepts and Bliss, 1988; Gepts, 1990; Kami et al., 1995). The southernmost region of domestication in the Southern Andes in the NWA is also the main region of Argentine common bean production. With the discovery of America, the primitive variety of wild beans were collected by explorers and taken to Europe where after breeding bigger sized grain cultivars were selected. Later on at the beginning of century XX, European immigrants brought these agriculturally convenient bean varieties back to NWA where settlers found the region to be appropriate for bean cultivation. Thus, currently in this region there exist areas of virgin land that supports the growth of wild beans as well as areas of cultivated land that account for the annual common bean production of about 300 000 ton (Burkart, 1952). Lie et al. (1987) showed that the center of diversity for peas also contained a great diversity of pea rhizobia; therefore it seems likely that a rather small proportion of the genetic diversity in bean nodulating rhizobia has been examined. In order to increase grain yields the inoculation of beans with appropriate rhizobia strains represents an agriculturally sustainable approach to meet needs. However, common beans are often considered as rather poor nitrogen fixers, although there are reports indicating high levels of fixation as well as the isolation of more efficient
bean rhizobia (Buttery et al., 1992; Hungrı´a et al., 2000). As bean is a short-season crop any environmental constraint affecting symbiosis may result in a low seasonal level of fixed nitrogen (Buttery et al., 1992). Other factors such as the low competitiveness of the introduced strains against native rhizobia may result in the lack of response to inoculation (Buttery et al., 1992; Streeter, 1994). In this work, our objective was to examine the diversity of rhizobia nodulating common bean in NWA to select more efficient strains in the field environment in the NWA cropping region.
2. Material and methods
2.1. Rhizobial collection All rhizobia were isolated from common beans collected in NWA or were retrieved in the laboratory from field soils. The procedure applied for isolation was that described for the isolation of rhizobia from wild beans by Aguilar et al. (1998b). Soil samples from Cerrillos (24°55%S, 65°29%W), Meta´ n (24°55%S, 65°27%W), Santa Clara (23°55%S, 64°29%W), Campichuelo (23°15%S, 65°29%W) and Pichanal (23°32%S, 64°58%W) were transported from the field to the laboratory for the recovery of rhizobial isolates by using the trapping host Leucaena leucocephala.
2.2. nifH-PCR The isolates were assigned as either type I or type II bean rhizobia by applying the test described by Aguilar et al. (1998a) in which R. etli and Rhizobium leguminosarum bv. phaseoli type I strains but not type II strains, yield a nifH PCR amplification product of 570 bp.
2.3. Analysis of 16S rRNA genes The procedure described by Laguerre et al. (1994) to identify the restriction sites in a PCRamplified 16S rDNA region of about 1.5 kb that encompasses conserved and variable regions to permit identification of the individual species, was
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
used. The endonucleases assayed were HaeIII, MspI, HinfI, and CfoI.
183
incubated in a dry oven at 60 °C until constant weight. Five plants per treatment were pooled and subjected to dry weight determination.
2.4. REP-PCR 2.7. Field experiments The DNA sequences of the repetitive extragenic palindromic (REP) and enterobacterial repetitive intergeneric consensus (ERIC) primers have been reported by de Bruijn (1992). The amplification reactions using the REP and ERIC primers were performed as described by de Bruijn (1992). The electrophoretic patterns were ordered by similarity by the unweighted pair group (UPGMA) method of clustering using the GELCOMPAR 4.1 software package from Applied Maths, Kortrijk, Belgium.
2.5. nodC RFLP The nodC sequence from rhizobial isolates was PCR amplified by using nodC-specific primers (Laguerre et al., 2001). The sequence of the primers is as follows: NodCF (5%-AYGTHGTYGAYGACGGTTC-3%) and NodCI (5%-CGYGACAGCCANTCKCTATTG-3%). The restriction fragment length polymorphism analysis was performed by electrophoresis in 2% agarose gel after restriction with endonucleases HinfI, CfoI, MspI, NdeII, RsaI, HaeIII, respectively.
2.6. Selection of strains under greenhouse conditions Bean seeds (P. 6ulgaris L. cultivar Negro Camilo, Instituto Nacional de Tecnologı´a Agropecuaria — INTA Estacio´ n Experimental Salta, Argentina) were surface sterilized for 3 min in 96% ethanol, followed by 15 min in 20% commercial bleach, and then washed five times with sterile water. Seedlings were germinated on top of 1.5% agar in water, and then individually transferred into 500 ml plastic pots containing washed sterilized vermiculite. After 5 days, seedlings were inoculated with rhizobial suspensions and watered twice with N-free mineral nutrient solution and with sterile distilled water as described earlier by Aguilar et al. (1998b). After 6 weeks of growth, the aerial part of the plants were harvested and
Two field experiments were performed during the bean seasons in years 1999 and 2000 in Cerrillos, province of Salta. The main chemical characteristics of the soil were as follows: pH 6.4; nitrogen content, 0.1%; organic carbon, 1.03%; organic matter, 1.78%; phosphorus, 10 ppm. The numbers of rhizobia nodulating beans were estimated by the soil dilution-plant infection most probable number method (Vincent, 1970). The experimental design consisted of randomized block design with four replicates with plots 5× 2 m2, with 0.5 m between lines. Inoculation was carried out by mixing 50 g of a peat-based inoculant (109 rhizobia per gram), 50 ml of rhizobial broth (109 rhizobia per ml) and 720 g (about 3000 seeds) of common bean seeds cultivar Negro Camilo (INTA, Estacio´ n Experimental Salta, Salta, Argentina). Sucrose 0.3% (w/v) was added to promote adherence. Sowing was carried out with 16 seeds per meter. The experiments also included the non-inoculated controls with or without N-fertilizer (25 kg of N per hectare as urea at sowing time and 25 kg of N per hectare at preflowering time). About 100 days after sowing the pod fill was completed and yield was evaluated as grain weight (11.5% humidity). Data were subjected to statistical analysis of variance.
3. Results and discussion
3.1. Characterization of isolates from common beans A collection of 347 isolates were obtained from common bean nodules collected in the NWA from five sites located in the bean producing area in NWA. In an attempt to obtain isolates that represented a full spectrum of genetic diversity in the soil populations in NWA, we used common beans and leucaena which were grown in the laboratory after inoculation with soil suspensions
184
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
prepared with samples brought from the fields. From these experiments, 250 isolates were retrieved from beans and 107 isolates were recovered from leucaena which added to our collection. The bean nodulating population density in soil at several sites in NWA was found to be high between 104 and 105 rhizobia per gram of dry soil, whereas the leucaena nodulating rhizobia were found to be lower (102 per gram of dry soil). The two populations isolated from beans and leucaena, respectively, were examined by applying the same criteria of genomic analysis. In order to determine whether the isolates belong to the type I group of bean nodulating rhizobia, we used a nifH PCR probe as it was described earlier by Aguilar et al. (1998a). It was found that the 570 bp nifH fragment could be amplified for all the bean isolates investigated indicating that most likely the isolates are related to R. etli or R. leguminosarum bv. phaseoli. On the contrary, no amplification with nifH primers was observed in the case of isolates of leucaena (data not shown). In order to attempt species assignment of our isolates, further investigation by RFLP analysis of the 16S rDNA region was performed by using the procedure described by Laguerre et al. (1994), and in a few cases by DNA sequencing of a 260 bp fragment as described earlier (Aguilar et al., 1998b). It was found that all the isolates from common beans had RFLP patterns identical to that of the reference strain R. etli CFN42. A similar prevalence of the R. etli allele among rhizobia isolated from wild bean nodules was observed by Aguilar et al. (1998b). On the other hand the result of the 16S rDNARFLP analysis of the isolates (n = 107) recovered from soils under laboratory conditions, by using leucaena as the trapping host was found to be more diverse as alleles of 16S rRNA which appear identical to the strains Sinorhizobium fredii USDA191 (n= 58), Sinorhizobium saheli USDA4893 (n=11), Sinorhizobium teranga USDA4894 (n =15), R. leguminosarum USDA2671 (n=3), R. tropici CIAT899 (n= 16) and Mesorhizobium loti USDA3455 (n =4), were identified in this population. As it was assessed by comparing the aspect of bean plants inoculated
with these genotypes and non-inoculated control plants, we concluded that all of them were able to induce effective nodules in beans under greenhouse inoculation conditions. This indicated that in the rhizobial soil complexity in NWA, species other than R. etli may also establish symbiosis with beans, and that multiple trap hosts are required to estimate diversity in soils. However, our results also show that P. 6ulgaris is highly discriminating in the microsymbiont to which it associates in nature. In order to assess the population for the intraspecies diversity a genomic analysis comprising REP-PCR DNA fingerprinting, and RFLP of the nodC sequence was performed as follows. The total genomic DNA from each of the isolates was used as the template for PCR with either Rep or ERIC primers to produce DNA fingerprints. The patterns were found highly heterogeneous, i.e. no two isolates of the same 16S rRNA allele gave identical profiles of fragments produced by PCR. The maximum similarity between isolates was at a degree of relatedness of about 90%. To illustrate this diversity, a representative pattern of 18 bean nodulating isolates from Cerrillo, the place where the field experiments were performed, is shown in Fig. 1. The nodC sequence was characterized in a collection of 80 bean rhizobial isolates and reference strains having the 16S rRNA allele of R. etli and R. leguminosarum, respectively. RFLP analysis of the PCR products resulted in three groups which for convenience were identified here as A, B, and C. The different patterns obtained are shown in Fig. 2. Strains CFN42 and Viking 1 which represent two lineages of species R. etli originating from Mexico and Belize, respectively, were distinguished in patterns A and B, whereas the large majority of NWA isolates shared the pattern C. R. leguminosarum strains were found to have patterns A and B. By putting these results together it is possible to conclude that the bean nodulating rhizobia present in the NWA soils are genetically diverse; however, the population represented by species R. etli is prevalent in nodulation of common beans under the natural conditions in NWA.
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
185
3.2. Strain selection under greenhouse conditions To select more efficient bean strains from the NWA diversity, a procedure based on the recovery of rhizobia from bean plants that were inoculated in the laboratory by using serial dilutions of soil suspensions from NWA was applied. Four weeks after inoculation, plants were examined for nodulation. Rhizobia from 52 nodules that formed in the plants inoculated with the undiluted soil suspension as well as those from 50 nodules
Fig. 2. RFLP analysis of the nodC sequence of common bean nodulating rhizobia from NWA. Isolates harboring the R. etli 16S rRNA allele were examined for polymorphism in the PCR-amplified nodC sequence by using endonucleases. The figure shows the result of digestion with endonuclease HinfI. Characteristic patterns were obtained for R. etli strains CFN42 (A) and Viking 1 (B) and for most of the isolates from NWA (C).
Fig. 1. Diversity in the bean nodulating rhizobia from NWA. ERIC-PCR patterns for 18 isolates of common beans from Cerrillos at NWA that share the R. etli 16S rRNA allele. The dendrogram is derived from analysis of ERIC-PCR by using the computer-assisted system of analysis GelCompar 4.1. Isolates NOAP186N1 and NOAP186N2, obtained from the two nodules of the same plant, show a high degree of similarity. Isolates NOAP184N1 and NOAP191N1, obtained from two plants collected in sites located about 2 km apart, were very similar.
formed in plants inoculated with the more diluted suspension were isolated and identified by using DNA REP-fingerprinting. Isolates from both groups that share the fingerprint were considered identical genotypes and selected for further studies. The rationale underlying this selection is that bean rhizobial genotypes that were found in the nodules of plants inoculated with the extreme soil suspensions should represent the more competitive and abundant rhizobial populations in the soil sample. In the following step, this group of
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
186
Table 1 Effectiveness of bean isolates from NWA under greenhouse conditions Isolate
16S rRNA genotype
Relative dry weight of aerial part
Standard deviation
Sc15 T51N5PP Camp11 T44N22P 321 NOAP195N5 NOAP205N9 NOAP194N8 152 NOAP207N1 T33N2P T34N2P 77 CFN42 NOAP194N15
R. R. R. R. R. R. R. R. R. R. R. R. R. R. R.
2.80 2.50 1.77 1.60 1.59 1.54 1.45 1.43 1.43 1.41 1.35 1.27 1.25 1.23 1.11
0.06 0.02 0.08 0.11 0.03 0.06 0.12 0.12 0.07 0.15 0.08 0.00 0.07 0.08 0.01
etli etli etli tropici B etli etli etli etli etli etli etli etli etli etli etli
The dry weight of the aerial part of five plants was determined and effectiveness was calculated as weight of inoculated plants/weight non-inoculated plants. The average weight was 0.30 g per plant. The 16S rRNA genotype was established by RFLP-analysis as described in Section 2.
isolates was tested for efficiency in plant by determining the dry matter accumulation under greenhouse conditions. The result of the selection is shown in Table 1. Isolates Sc15, T51N5PP and Camp11 obtained from soil samples from Santa Clara, Cerrillos and Campichuelo, respectively, which were found to have the 16S rRNA allele of species R. etli, were more efficient in dry matter accumulation. A subgroup from the more efficient isolates together with other bean nodulating genotypes were selected for further assessment under field environmental conditions in NWA. As isolates Sc15 and T51N5PP were isolated from the two sites with similar soil characteristics, only Sc15 was selected for the field assay.
about 106 per seed. In the first year of experiments inoculation improved yields and in the case of strains R. tropici CIAT899, R. etli Sc15 and R. tropici A NOAP25N1 statistically significant increase between 30 and 20% was obtained, respectively, as compared with the non-inoculated
3.3. Field assays Field experiments in Cerrillos in the province of Salta (24°55%S, 65°29%W) where beans were cultivated at least during the last 4 years, five bean nodulating isolates representing different genotypes were assessed for grain yield. Plant dilution estimates of the bean nodulating rhizobia initially present were high by about 104 cells per gram of dry soil. Levels of viable rhizobia that adhered to the seed after treatment with the inoculant were
Fig. 3. Bean yields in field inoculation assays. Field assays in the NWA were performed during two successive bean-cropping seasons in years 1999 and 2000. Bars indicate yield in kilogram per hectare. Asterisks denote values with no significant difference (Turkey, PB 0.05) for assay year 1999. Values of assay year 2000 did not show statistical differences.
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
treatment (Fig. 3). The increase in grain yield was similar to that found with chemically fertilized treatment. As R. tropici CIAT899 and NOAP25N1 can be distinguished from the R. etli-like native population by using nifH primers and PCR as described earlier by Aguilar et al. (1998a), nodule occupancy 40 days after sowing was estimated in plants inoculated with these two strains, respectively. A number of 367 and 229 nodules picked, respectively, from five plants inoculated with strains CIAT899 and NOAP25N1 were examined. Our previous investigations demonstrated that the native bean nodulating rhizobial populations other than species R. etli were not found in nodules of common beans planted in NWA. The result of the occupancy analysis of plants inoculated with strains CIAT899 and NOAP25N1 resulted in 18 and 16%, respectively. These low values and the spontaneous nodulation that we observed in the non-inoculated treatment agree with the previous findings in which the problem in overcoming the dominance of indigenous rhizobia was evidenced (Streeter, 1994). The conclusion is that although it is very difficult to displace the resident population of bean rhizobia, yet a low occupancy by the inoculant strain may lead to a measurable effect on yield. The strain R. tropici CIAT899 which was described as having good symbiotic performance under adverse environmental conditions such as soil acidity and high level of aluminum is being used in bean inoculants in Brazil and other countries (Hungrı´a et al., 2000). The good symbiotic performance of CIAT899 was also confirmed in our assays where the pH of soil is near neutral although the level of the resident bean nodulating rhizobia is rather high. Other strains tested in our field experiment, selected from the diversity of bean nodulating rhizobia found in NWA, had symbiotic performances comparable to CIAT899. This result supports the validity of exploring the diversity to approach the improvement of bean yields. The field assay performed the following year in the same location but at a different plot, did not result in a significant yield response. Unfortunately, adverse climate conditions such as copious rainfall soon after sowing might have washed out the rhizobia from the inoculated seeds, while low
187
numbers of sunny days and cloudiness during the cropping season might have limited the yield in general. Nodule occupancy was found to be lower than 3%; however, mean yields were found to be higher compared with the yield in the previous season (Fig. 3). We were unable to identify the basis of this difference. Definitely, further field assays are essential to validate the results. However, it is possible to conclude that response to inoculation can render yield improvement, and therefore among other potentials a programme on strain selection and bean inoculation still represents a valid approach to revert poor yields and depletion of soil N content (Buttery et al., 1992). In developing countries where the gap between actual and potential crop yield is large, this improvement could have an important social impact.
Acknowledgements We are grateful to Juan Grassano for technical assistance and Susana Garcı´a Medina (INTA, Salta) for advice and help in field assays. This work was supported by grants of Secyt-CONICET, Argentina (PID No. 08-07072 BID 1201 OC/AR) and of the Commission of Economic Communities (ERBIC18 CT98 0321). O.M.A. is a member of the Argentine Research Council (CONICET). M.V.L. is the recipient of a scholarship from CONICET. P.M.R. is supported by CIC (Buenos Aires).
References Aguilar, O.M., Grasso, D.H., Riccillo, P.M., Lo´ pez, M.V., Szafer, E., 1998a. Rapid identification of bean rhizobial isolates type I by a nifH-PCR assay. Soil Biol. Biochem. 30, 1655 – 1661. Aguilar, O.M., Lo´ pez, M.V., Riccillo, P.M., Gonza´ lez, R.A., Pagano, M., Grasso, D.H., Pu¨ hler, A., Favelukes, G., 1998b. Prevalence of the Rhizobium etli-like allele in genes coding for 16S rRNA among the indigenous rhizobial populations found associated with wild beans from the Southern Andes in Argentina. Appl. Environ. Microbiol. 64, 3520 – 3524. Amarger, N., Bours, M., Revoy, F., Allard, M.R., Laguerrre, G., 1994. Rhizobium tropici nodulates field-grown Phaseolus 6ulgaris in France. Plant Soil 161, 147 – 156.
188
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
Amarger, N., Macheret, V., Laguerre, G., 1997. Rhizobium gallicum sp.nov. and Rhizobium gardinii sp.nov., from Phaseolus 6ulgaris nodules. Int. J. Syst. Bacteriol. 47, 996 – 1006. Anyango, B., Wilson, K.L., Beynon, J.L., Giller, K.E., 1995. Diversity of rhizobia nodulating Phaseolus 6ulgaris L. in two Kenyan soils with contrasting pHs. Appl. Environ. Microbiol. 61, 4016 – 4021. Burkart, A., 1952. Las leguminosas argentinas silvestres y cultivadas. Ediciones Acme Agency. Buttery, B.R., Park, S.J., Hume, D.J., 1992. Potential for increasing nitrogen fixation in grain legumes. Can. J. Plant Sci. 72, 323 – 349. de Bruijn, F.J., 1992. Use of repetitive (repetitive extragenic palindromic and enterobacterial repetitive intergenic consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Appl. Environ. Microbiol. 58, 2180 – 2187. Gepts, P., Bliss, F.A., 1988. Dissemination pathways of common bean (Phaseolus 6ulgaris, Fabaceae) deduced from phaseolin electrophoretic variability. II. Europe and Africa. Econ. Bot. 42, 86 –104. Gepts, P., 1990. Biochemical evidence bearing on the domestication of Phaseolus (Fabacea) beans. Econ. Bot. 44, 28 –38. Herrera-Cervera, J.A., Caballero-Mellado, J., Laguerre, G., Tichy, H.V., Requena, N., Amarger, N., MartinezRomero, E., Olivares, J., Sanjua´ n, J., 1999. At least five rhizobial species nodulate Phaseolus 6ulgaris in a Spanish soil. FEMS Microbiol. Ecol. 30, 87 –97. Hungrı´a, M., Andrade, S.D., Chuceire, L.M. de O., Probanza, A., Guitierrez-Man˜ ero, F.J., Megı´as, M., 2000. Isolation and characterization of new efficient and competitive bean
(Phaseolus 6ulgaris L.) rhizobia from Brazil. Soil Biol. Biochem. 21, 1515 – 1528. Kami, J., Becerra-Velasquez, Debouck, D.G., Gepts, P., 1995. Identification of presumed ancestral DNA sequences of phaseolin in Phaseolus 6ulgaris. Proc. Natl. Acad. Sci. USA 92, 1101 – 1104. Laguerre, G., Allard, R.M., Revoy, F., Amarger, N., 1994. Rapid identification of rhizobia by restriction fragment length polymorphism analysis of PCR-amplified 16S rRNA genes. Appl. Environ. Microbiol. 60, 56 – 63. Laguerre, G., Nour, S.M., Macheret, V., Sanjuan, J., Drouin, P., Amarger, N., 2001. Classification of rhizobia based on nodC and nifH gene analysis reveals a close phylogenetic relationship among Phaseolus 6ulgaris symbionts. Microbiology 147, 981 – 993. Lie, T.A., Goktan, M., Engin, M., Pijnenborg, J., Anlarsal, E., 1987. Co-evolution of the legume – rhizobium association. Plant Soil 100, 171 – 181. Martı´nez-Romero, E., Segovia, L., Mercante, F.M., Franco, A.A., Graham, P., Pardo, M.A., 1991. Rhizobium tropici, a novel species nodulating Phaseolus 6ulgaris L. beans and Leucaena sp. trees. Int. J. Syst. Bacteriol. 41, 417 – 426. Segovia, L., Young, J.P.W., Martı´nez-Romero, E., 1993. Reclassification of American Rhizobium leguminosarum biovar phaseoli type I strains as Rhizobium etli sp.nov. Int. J. Syst. Bacteriol. 43, 374 – 377. Streeter, J.G., 1994. Failure of inoculant rhizobia to overcome the dominance of indigenous strains for nodule formation. Can. J. Microbiol. 40, 513 – 522. Vincent, J.M., 1970. A Manual for the Practical Study of the Root-nodule Bacteria. IBP Handbook No. 15. Blackwell, Oxford, UK.
The diversity of rhizobia nodulating beans in Northwest Argentina as a source of more efficient inoculant strains O. Mario Aguilar *, M. Vero´nica Lo´pez, Pablo M. Riccillo Instituto de Bioquı´mica y Biologı´a Molecular, Facultad de Ciencias Exactas, Uni6ersidad Nacional de La Plata, Calles 47 y 115, 1900 La Plata, Argentina Received 24 November 2000; received in revised form 30 March 2001; accepted 4 April 2001
Abstract The common bean (Phaseolus 6ulgaris L.) is cultivated widely in Central and South America and particularly in the Northwest of Argentina. In order to describe the diversity of the common bean nodulating rhizobial population from the bean producing area in Northwest Argentina (NWA), a collection of about 400 isolates of common beans recovered from nodules and soil samples from NWA were characterized by using nifH-PCR, analysis of genes coding for 16S rRNA and nodC, and REP-fingerprinting, respectively. It was found that species Rhizobium etli is predominant in common bean nodules although a high degree of diversity was found within the species. Other bean nodulating genotypes recovered from soils by using Leucaena sp. as the trapping host was found to have the 16S rDNA alleles of species such as Sinorhizobium fredii, Sinorhizobium saheli, Sinorhizobium teranga, Mesorhizobium loti, and Rhizobium tropici. Some of the bean genotypes that were found to be more efficient in green house experiments were selected and assayed in two successive bean-cropping seasons in the field environment in NWA, and an increase in yields with inoculation was found. The performance of strains isolated from the region indicates potential for exploiting the diversity. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Bean rhizobia; Diversity; Inoculation; Yields
1. Introduction Common beans (Phaseolus 6ulgaris L.) are an important crop in South America and in Africa, supplying in some cases up to 20% of the protein intake per person. Beans are grown in a great variety of farming systems, ranging from a highly mechanized cropping in Northwest Argentina * Corresponding author. Tel.: + 54-221-4250497. E-mail address: [email protected] Aguilar).
(O.M.
(NWA) to the low-input subsistence cultivation in Africa and Central America. Generally, the nitrogen requirements of legumes can be met either by the inorganic N from the soil or from the chemical fertilizers applied as well as by biological nitrogen fixation (BNF) which results from the symbiotic association with the soil bacteria rhizobia. The rhizobial isolates that are able to nodulate bean roots form a very heterogeneous group. Until now five species have been described; however some other isolates, all of which are also able to nodulate common beans with different degrees
0168-1656/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 5 6 ( 0 1 ) 0 0 3 3 6 - 4
182
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
of effectiveness, still appear to represent other distinct phylogenetic lineages (Martı´nez-Romero et al., 1991; Segovia et al., 1993; Amarger et al., 1997; Herrera-Cervera et al., 1999). Species Rhizobium tropici, which is able to nodulate a broad range of tropical trees including the Leucaena spp., was first isolated in South America but, has also been associated with P. 6ulgaris in France by Amarger et al. (1994) and in acid soils in Kenya by Anyango et al. (1995). The predominance of the Rhizobium etli-like 16S rRNA allele in the native rhizobial populations associated to wild beans in NWA was reported by Aguilar et al. (1998b). It is generally accepted that P. 6ulgaris L. is native to the Americas. Domestication of wild beans took place independently in the Mesoamerican center of origin (Mexico, Central America and Colombia) and in the Andean center in South America (Gepts and Bliss, 1988; Gepts, 1990; Kami et al., 1995). The southernmost region of domestication in the Southern Andes in the NWA is also the main region of Argentine common bean production. With the discovery of America, the primitive variety of wild beans were collected by explorers and taken to Europe where after breeding bigger sized grain cultivars were selected. Later on at the beginning of century XX, European immigrants brought these agriculturally convenient bean varieties back to NWA where settlers found the region to be appropriate for bean cultivation. Thus, currently in this region there exist areas of virgin land that supports the growth of wild beans as well as areas of cultivated land that account for the annual common bean production of about 300 000 ton (Burkart, 1952). Lie et al. (1987) showed that the center of diversity for peas also contained a great diversity of pea rhizobia; therefore it seems likely that a rather small proportion of the genetic diversity in bean nodulating rhizobia has been examined. In order to increase grain yields the inoculation of beans with appropriate rhizobia strains represents an agriculturally sustainable approach to meet needs. However, common beans are often considered as rather poor nitrogen fixers, although there are reports indicating high levels of fixation as well as the isolation of more efficient
bean rhizobia (Buttery et al., 1992; Hungrı´a et al., 2000). As bean is a short-season crop any environmental constraint affecting symbiosis may result in a low seasonal level of fixed nitrogen (Buttery et al., 1992). Other factors such as the low competitiveness of the introduced strains against native rhizobia may result in the lack of response to inoculation (Buttery et al., 1992; Streeter, 1994). In this work, our objective was to examine the diversity of rhizobia nodulating common bean in NWA to select more efficient strains in the field environment in the NWA cropping region.
2. Material and methods
2.1. Rhizobial collection All rhizobia were isolated from common beans collected in NWA or were retrieved in the laboratory from field soils. The procedure applied for isolation was that described for the isolation of rhizobia from wild beans by Aguilar et al. (1998b). Soil samples from Cerrillos (24°55%S, 65°29%W), Meta´ n (24°55%S, 65°27%W), Santa Clara (23°55%S, 64°29%W), Campichuelo (23°15%S, 65°29%W) and Pichanal (23°32%S, 64°58%W) were transported from the field to the laboratory for the recovery of rhizobial isolates by using the trapping host Leucaena leucocephala.
2.2. nifH-PCR The isolates were assigned as either type I or type II bean rhizobia by applying the test described by Aguilar et al. (1998a) in which R. etli and Rhizobium leguminosarum bv. phaseoli type I strains but not type II strains, yield a nifH PCR amplification product of 570 bp.
2.3. Analysis of 16S rRNA genes The procedure described by Laguerre et al. (1994) to identify the restriction sites in a PCRamplified 16S rDNA region of about 1.5 kb that encompasses conserved and variable regions to permit identification of the individual species, was
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
used. The endonucleases assayed were HaeIII, MspI, HinfI, and CfoI.
183
incubated in a dry oven at 60 °C until constant weight. Five plants per treatment were pooled and subjected to dry weight determination.
2.4. REP-PCR 2.7. Field experiments The DNA sequences of the repetitive extragenic palindromic (REP) and enterobacterial repetitive intergeneric consensus (ERIC) primers have been reported by de Bruijn (1992). The amplification reactions using the REP and ERIC primers were performed as described by de Bruijn (1992). The electrophoretic patterns were ordered by similarity by the unweighted pair group (UPGMA) method of clustering using the GELCOMPAR 4.1 software package from Applied Maths, Kortrijk, Belgium.
2.5. nodC RFLP The nodC sequence from rhizobial isolates was PCR amplified by using nodC-specific primers (Laguerre et al., 2001). The sequence of the primers is as follows: NodCF (5%-AYGTHGTYGAYGACGGTTC-3%) and NodCI (5%-CGYGACAGCCANTCKCTATTG-3%). The restriction fragment length polymorphism analysis was performed by electrophoresis in 2% agarose gel after restriction with endonucleases HinfI, CfoI, MspI, NdeII, RsaI, HaeIII, respectively.
2.6. Selection of strains under greenhouse conditions Bean seeds (P. 6ulgaris L. cultivar Negro Camilo, Instituto Nacional de Tecnologı´a Agropecuaria — INTA Estacio´ n Experimental Salta, Argentina) were surface sterilized for 3 min in 96% ethanol, followed by 15 min in 20% commercial bleach, and then washed five times with sterile water. Seedlings were germinated on top of 1.5% agar in water, and then individually transferred into 500 ml plastic pots containing washed sterilized vermiculite. After 5 days, seedlings were inoculated with rhizobial suspensions and watered twice with N-free mineral nutrient solution and with sterile distilled water as described earlier by Aguilar et al. (1998b). After 6 weeks of growth, the aerial part of the plants were harvested and
Two field experiments were performed during the bean seasons in years 1999 and 2000 in Cerrillos, province of Salta. The main chemical characteristics of the soil were as follows: pH 6.4; nitrogen content, 0.1%; organic carbon, 1.03%; organic matter, 1.78%; phosphorus, 10 ppm. The numbers of rhizobia nodulating beans were estimated by the soil dilution-plant infection most probable number method (Vincent, 1970). The experimental design consisted of randomized block design with four replicates with plots 5× 2 m2, with 0.5 m between lines. Inoculation was carried out by mixing 50 g of a peat-based inoculant (109 rhizobia per gram), 50 ml of rhizobial broth (109 rhizobia per ml) and 720 g (about 3000 seeds) of common bean seeds cultivar Negro Camilo (INTA, Estacio´ n Experimental Salta, Salta, Argentina). Sucrose 0.3% (w/v) was added to promote adherence. Sowing was carried out with 16 seeds per meter. The experiments also included the non-inoculated controls with or without N-fertilizer (25 kg of N per hectare as urea at sowing time and 25 kg of N per hectare at preflowering time). About 100 days after sowing the pod fill was completed and yield was evaluated as grain weight (11.5% humidity). Data were subjected to statistical analysis of variance.
3. Results and discussion
3.1. Characterization of isolates from common beans A collection of 347 isolates were obtained from common bean nodules collected in the NWA from five sites located in the bean producing area in NWA. In an attempt to obtain isolates that represented a full spectrum of genetic diversity in the soil populations in NWA, we used common beans and leucaena which were grown in the laboratory after inoculation with soil suspensions
184
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
prepared with samples brought from the fields. From these experiments, 250 isolates were retrieved from beans and 107 isolates were recovered from leucaena which added to our collection. The bean nodulating population density in soil at several sites in NWA was found to be high between 104 and 105 rhizobia per gram of dry soil, whereas the leucaena nodulating rhizobia were found to be lower (102 per gram of dry soil). The two populations isolated from beans and leucaena, respectively, were examined by applying the same criteria of genomic analysis. In order to determine whether the isolates belong to the type I group of bean nodulating rhizobia, we used a nifH PCR probe as it was described earlier by Aguilar et al. (1998a). It was found that the 570 bp nifH fragment could be amplified for all the bean isolates investigated indicating that most likely the isolates are related to R. etli or R. leguminosarum bv. phaseoli. On the contrary, no amplification with nifH primers was observed in the case of isolates of leucaena (data not shown). In order to attempt species assignment of our isolates, further investigation by RFLP analysis of the 16S rDNA region was performed by using the procedure described by Laguerre et al. (1994), and in a few cases by DNA sequencing of a 260 bp fragment as described earlier (Aguilar et al., 1998b). It was found that all the isolates from common beans had RFLP patterns identical to that of the reference strain R. etli CFN42. A similar prevalence of the R. etli allele among rhizobia isolated from wild bean nodules was observed by Aguilar et al. (1998b). On the other hand the result of the 16S rDNARFLP analysis of the isolates (n = 107) recovered from soils under laboratory conditions, by using leucaena as the trapping host was found to be more diverse as alleles of 16S rRNA which appear identical to the strains Sinorhizobium fredii USDA191 (n= 58), Sinorhizobium saheli USDA4893 (n=11), Sinorhizobium teranga USDA4894 (n =15), R. leguminosarum USDA2671 (n=3), R. tropici CIAT899 (n= 16) and Mesorhizobium loti USDA3455 (n =4), were identified in this population. As it was assessed by comparing the aspect of bean plants inoculated
with these genotypes and non-inoculated control plants, we concluded that all of them were able to induce effective nodules in beans under greenhouse inoculation conditions. This indicated that in the rhizobial soil complexity in NWA, species other than R. etli may also establish symbiosis with beans, and that multiple trap hosts are required to estimate diversity in soils. However, our results also show that P. 6ulgaris is highly discriminating in the microsymbiont to which it associates in nature. In order to assess the population for the intraspecies diversity a genomic analysis comprising REP-PCR DNA fingerprinting, and RFLP of the nodC sequence was performed as follows. The total genomic DNA from each of the isolates was used as the template for PCR with either Rep or ERIC primers to produce DNA fingerprints. The patterns were found highly heterogeneous, i.e. no two isolates of the same 16S rRNA allele gave identical profiles of fragments produced by PCR. The maximum similarity between isolates was at a degree of relatedness of about 90%. To illustrate this diversity, a representative pattern of 18 bean nodulating isolates from Cerrillo, the place where the field experiments were performed, is shown in Fig. 1. The nodC sequence was characterized in a collection of 80 bean rhizobial isolates and reference strains having the 16S rRNA allele of R. etli and R. leguminosarum, respectively. RFLP analysis of the PCR products resulted in three groups which for convenience were identified here as A, B, and C. The different patterns obtained are shown in Fig. 2. Strains CFN42 and Viking 1 which represent two lineages of species R. etli originating from Mexico and Belize, respectively, were distinguished in patterns A and B, whereas the large majority of NWA isolates shared the pattern C. R. leguminosarum strains were found to have patterns A and B. By putting these results together it is possible to conclude that the bean nodulating rhizobia present in the NWA soils are genetically diverse; however, the population represented by species R. etli is prevalent in nodulation of common beans under the natural conditions in NWA.
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
185
3.2. Strain selection under greenhouse conditions To select more efficient bean strains from the NWA diversity, a procedure based on the recovery of rhizobia from bean plants that were inoculated in the laboratory by using serial dilutions of soil suspensions from NWA was applied. Four weeks after inoculation, plants were examined for nodulation. Rhizobia from 52 nodules that formed in the plants inoculated with the undiluted soil suspension as well as those from 50 nodules
Fig. 2. RFLP analysis of the nodC sequence of common bean nodulating rhizobia from NWA. Isolates harboring the R. etli 16S rRNA allele were examined for polymorphism in the PCR-amplified nodC sequence by using endonucleases. The figure shows the result of digestion with endonuclease HinfI. Characteristic patterns were obtained for R. etli strains CFN42 (A) and Viking 1 (B) and for most of the isolates from NWA (C).
Fig. 1. Diversity in the bean nodulating rhizobia from NWA. ERIC-PCR patterns for 18 isolates of common beans from Cerrillos at NWA that share the R. etli 16S rRNA allele. The dendrogram is derived from analysis of ERIC-PCR by using the computer-assisted system of analysis GelCompar 4.1. Isolates NOAP186N1 and NOAP186N2, obtained from the two nodules of the same plant, show a high degree of similarity. Isolates NOAP184N1 and NOAP191N1, obtained from two plants collected in sites located about 2 km apart, were very similar.
formed in plants inoculated with the more diluted suspension were isolated and identified by using DNA REP-fingerprinting. Isolates from both groups that share the fingerprint were considered identical genotypes and selected for further studies. The rationale underlying this selection is that bean rhizobial genotypes that were found in the nodules of plants inoculated with the extreme soil suspensions should represent the more competitive and abundant rhizobial populations in the soil sample. In the following step, this group of
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
186
Table 1 Effectiveness of bean isolates from NWA under greenhouse conditions Isolate
16S rRNA genotype
Relative dry weight of aerial part
Standard deviation
Sc15 T51N5PP Camp11 T44N22P 321 NOAP195N5 NOAP205N9 NOAP194N8 152 NOAP207N1 T33N2P T34N2P 77 CFN42 NOAP194N15
R. R. R. R. R. R. R. R. R. R. R. R. R. R. R.
2.80 2.50 1.77 1.60 1.59 1.54 1.45 1.43 1.43 1.41 1.35 1.27 1.25 1.23 1.11
0.06 0.02 0.08 0.11 0.03 0.06 0.12 0.12 0.07 0.15 0.08 0.00 0.07 0.08 0.01
etli etli etli tropici B etli etli etli etli etli etli etli etli etli etli etli
The dry weight of the aerial part of five plants was determined and effectiveness was calculated as weight of inoculated plants/weight non-inoculated plants. The average weight was 0.30 g per plant. The 16S rRNA genotype was established by RFLP-analysis as described in Section 2.
isolates was tested for efficiency in plant by determining the dry matter accumulation under greenhouse conditions. The result of the selection is shown in Table 1. Isolates Sc15, T51N5PP and Camp11 obtained from soil samples from Santa Clara, Cerrillos and Campichuelo, respectively, which were found to have the 16S rRNA allele of species R. etli, were more efficient in dry matter accumulation. A subgroup from the more efficient isolates together with other bean nodulating genotypes were selected for further assessment under field environmental conditions in NWA. As isolates Sc15 and T51N5PP were isolated from the two sites with similar soil characteristics, only Sc15 was selected for the field assay.
about 106 per seed. In the first year of experiments inoculation improved yields and in the case of strains R. tropici CIAT899, R. etli Sc15 and R. tropici A NOAP25N1 statistically significant increase between 30 and 20% was obtained, respectively, as compared with the non-inoculated
3.3. Field assays Field experiments in Cerrillos in the province of Salta (24°55%S, 65°29%W) where beans were cultivated at least during the last 4 years, five bean nodulating isolates representing different genotypes were assessed for grain yield. Plant dilution estimates of the bean nodulating rhizobia initially present were high by about 104 cells per gram of dry soil. Levels of viable rhizobia that adhered to the seed after treatment with the inoculant were
Fig. 3. Bean yields in field inoculation assays. Field assays in the NWA were performed during two successive bean-cropping seasons in years 1999 and 2000. Bars indicate yield in kilogram per hectare. Asterisks denote values with no significant difference (Turkey, PB 0.05) for assay year 1999. Values of assay year 2000 did not show statistical differences.
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
treatment (Fig. 3). The increase in grain yield was similar to that found with chemically fertilized treatment. As R. tropici CIAT899 and NOAP25N1 can be distinguished from the R. etli-like native population by using nifH primers and PCR as described earlier by Aguilar et al. (1998a), nodule occupancy 40 days after sowing was estimated in plants inoculated with these two strains, respectively. A number of 367 and 229 nodules picked, respectively, from five plants inoculated with strains CIAT899 and NOAP25N1 were examined. Our previous investigations demonstrated that the native bean nodulating rhizobial populations other than species R. etli were not found in nodules of common beans planted in NWA. The result of the occupancy analysis of plants inoculated with strains CIAT899 and NOAP25N1 resulted in 18 and 16%, respectively. These low values and the spontaneous nodulation that we observed in the non-inoculated treatment agree with the previous findings in which the problem in overcoming the dominance of indigenous rhizobia was evidenced (Streeter, 1994). The conclusion is that although it is very difficult to displace the resident population of bean rhizobia, yet a low occupancy by the inoculant strain may lead to a measurable effect on yield. The strain R. tropici CIAT899 which was described as having good symbiotic performance under adverse environmental conditions such as soil acidity and high level of aluminum is being used in bean inoculants in Brazil and other countries (Hungrı´a et al., 2000). The good symbiotic performance of CIAT899 was also confirmed in our assays where the pH of soil is near neutral although the level of the resident bean nodulating rhizobia is rather high. Other strains tested in our field experiment, selected from the diversity of bean nodulating rhizobia found in NWA, had symbiotic performances comparable to CIAT899. This result supports the validity of exploring the diversity to approach the improvement of bean yields. The field assay performed the following year in the same location but at a different plot, did not result in a significant yield response. Unfortunately, adverse climate conditions such as copious rainfall soon after sowing might have washed out the rhizobia from the inoculated seeds, while low
187
numbers of sunny days and cloudiness during the cropping season might have limited the yield in general. Nodule occupancy was found to be lower than 3%; however, mean yields were found to be higher compared with the yield in the previous season (Fig. 3). We were unable to identify the basis of this difference. Definitely, further field assays are essential to validate the results. However, it is possible to conclude that response to inoculation can render yield improvement, and therefore among other potentials a programme on strain selection and bean inoculation still represents a valid approach to revert poor yields and depletion of soil N content (Buttery et al., 1992). In developing countries where the gap between actual and potential crop yield is large, this improvement could have an important social impact.
Acknowledgements We are grateful to Juan Grassano for technical assistance and Susana Garcı´a Medina (INTA, Salta) for advice and help in field assays. This work was supported by grants of Secyt-CONICET, Argentina (PID No. 08-07072 BID 1201 OC/AR) and of the Commission of Economic Communities (ERBIC18 CT98 0321). O.M.A. is a member of the Argentine Research Council (CONICET). M.V.L. is the recipient of a scholarship from CONICET. P.M.R. is supported by CIC (Buenos Aires).
References Aguilar, O.M., Grasso, D.H., Riccillo, P.M., Lo´ pez, M.V., Szafer, E., 1998a. Rapid identification of bean rhizobial isolates type I by a nifH-PCR assay. Soil Biol. Biochem. 30, 1655 – 1661. Aguilar, O.M., Lo´ pez, M.V., Riccillo, P.M., Gonza´ lez, R.A., Pagano, M., Grasso, D.H., Pu¨ hler, A., Favelukes, G., 1998b. Prevalence of the Rhizobium etli-like allele in genes coding for 16S rRNA among the indigenous rhizobial populations found associated with wild beans from the Southern Andes in Argentina. Appl. Environ. Microbiol. 64, 3520 – 3524. Amarger, N., Bours, M., Revoy, F., Allard, M.R., Laguerrre, G., 1994. Rhizobium tropici nodulates field-grown Phaseolus 6ulgaris in France. Plant Soil 161, 147 – 156.
188
O.M. Aguilar et al. / Journal of Biotechnology 91 (2001) 181–188
Amarger, N., Macheret, V., Laguerre, G., 1997. Rhizobium gallicum sp.nov. and Rhizobium gardinii sp.nov., from Phaseolus 6ulgaris nodules. Int. J. Syst. Bacteriol. 47, 996 – 1006. Anyango, B., Wilson, K.L., Beynon, J.L., Giller, K.E., 1995. Diversity of rhizobia nodulating Phaseolus 6ulgaris L. in two Kenyan soils with contrasting pHs. Appl. Environ. Microbiol. 61, 4016 – 4021. Burkart, A., 1952. Las leguminosas argentinas silvestres y cultivadas. Ediciones Acme Agency. Buttery, B.R., Park, S.J., Hume, D.J., 1992. Potential for increasing nitrogen fixation in grain legumes. Can. J. Plant Sci. 72, 323 – 349. de Bruijn, F.J., 1992. Use of repetitive (repetitive extragenic palindromic and enterobacterial repetitive intergenic consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Appl. Environ. Microbiol. 58, 2180 – 2187. Gepts, P., Bliss, F.A., 1988. Dissemination pathways of common bean (Phaseolus 6ulgaris, Fabaceae) deduced from phaseolin electrophoretic variability. II. Europe and Africa. Econ. Bot. 42, 86 –104. Gepts, P., 1990. Biochemical evidence bearing on the domestication of Phaseolus (Fabacea) beans. Econ. Bot. 44, 28 –38. Herrera-Cervera, J.A., Caballero-Mellado, J., Laguerre, G., Tichy, H.V., Requena, N., Amarger, N., MartinezRomero, E., Olivares, J., Sanjua´ n, J., 1999. At least five rhizobial species nodulate Phaseolus 6ulgaris in a Spanish soil. FEMS Microbiol. Ecol. 30, 87 –97. Hungrı´a, M., Andrade, S.D., Chuceire, L.M. de O., Probanza, A., Guitierrez-Man˜ ero, F.J., Megı´as, M., 2000. Isolation and characterization of new efficient and competitive bean
(Phaseolus 6ulgaris L.) rhizobia from Brazil. Soil Biol. Biochem. 21, 1515 – 1528. Kami, J., Becerra-Velasquez, Debouck, D.G., Gepts, P., 1995. Identification of presumed ancestral DNA sequences of phaseolin in Phaseolus 6ulgaris. Proc. Natl. Acad. Sci. USA 92, 1101 – 1104. Laguerre, G., Allard, R.M., Revoy, F., Amarger, N., 1994. Rapid identification of rhizobia by restriction fragment length polymorphism analysis of PCR-amplified 16S rRNA genes. Appl. Environ. Microbiol. 60, 56 – 63. Laguerre, G., Nour, S.M., Macheret, V., Sanjuan, J., Drouin, P., Amarger, N., 2001. Classification of rhizobia based on nodC and nifH gene analysis reveals a close phylogenetic relationship among Phaseolus 6ulgaris symbionts. Microbiology 147, 981 – 993. Lie, T.A., Goktan, M., Engin, M., Pijnenborg, J., Anlarsal, E., 1987. Co-evolution of the legume – rhizobium association. Plant Soil 100, 171 – 181. Martı´nez-Romero, E., Segovia, L., Mercante, F.M., Franco, A.A., Graham, P., Pardo, M.A., 1991. Rhizobium tropici, a novel species nodulating Phaseolus 6ulgaris L. beans and Leucaena sp. trees. Int. J. Syst. Bacteriol. 41, 417 – 426. Segovia, L., Young, J.P.W., Martı´nez-Romero, E., 1993. Reclassification of American Rhizobium leguminosarum biovar phaseoli type I strains as Rhizobium etli sp.nov. Int. J. Syst. Bacteriol. 43, 374 – 377. Streeter, J.G., 1994. Failure of inoculant rhizobia to overcome the dominance of indigenous strains for nodule formation. Can. J. Microbiol. 40, 513 – 522. Vincent, J.M., 1970. A Manual for the Practical Study of the Root-nodule Bacteria. IBP Handbook No. 15. Blackwell, Oxford, UK.