Diversity of new root nodule bacteria from Erythrina velutina Willd., a native legume from the dry forest Caatinga (Northeastern Brazil)

Diversidade de novas bactérias isoladas de nódulos de Erythrina velutina Willd., uma leguminosa nativa da floresta seca “Caatinga” (Região Nordeste do Brasil)

Kelly Alexsandra Souza Menezes¹, Gersika Fakirra de Oliveira Nunes¹, Aline Araujo Sampaio¹, Aleksandro Ferreira da Silva², Layane Silva Barbosa de Souza¹, Carlos Alberto Tuão Gava³, Lindete Míria Vieira Martins¹ e Paulo Ivan Fernandes-Júnior^3,*

¹ Departamento de Tecnologia e Ciências Sociais, Universidade do Estado da Bahia, Juazeiro, BA, Brazil;

² Universidade Federal Rural de Pernambuco; Recife, PE, Brazil;

³ Embrapa Semiárido, BR 428, Km 152, Petrolina, PE, Brazil, CEP: 56302-970, CP 23. *E-mail: paulo.ivan@embrapa.br

ABSTRACT

Keywords: Biological nitrogen fixation, inoculant, rhizobia, tree legumes.

RESUMO

O objetivo desse estudo foi avaliar as características fenotípicas dos isolados de nódulos de mulungu (Erythrina velutina Willd.) e determinar seu fingerprinting molecular por meio de Box-PCR. Todas as bactérias foram avaliadas pelas seguintes características fenotípicas: tempo de crescimento, alteração de pH, cor da colónia e produção de muco. As bactérias isoladas hábeis em renodular o hospedeiro original também foram avaliadas em relação à tolerância ao aumento da salinidade e da temperatura de incubação, capacidade de crescimento utilizando diferentes fontes de carbono, resistência intrínseca aos antibióticos e biossíntese de auxina “in vitro”. O fingerprint molecular foi estabelecido usando a técnica Box-PCR e os isolados foram agrupados pelos perfis obtidos. Dos 22 isolados obtidos, oito foram hábeis em renodular o hospedeiro original. Entre eles, alguns foram tolerantes a 1% de NaCl e a 39° C e todos metabolizaram as fontes maltose, frutose, glicose, sacarose e arabinose, resistiram à rifampicina e produziram auxina. As bactérias isoladas mostraram baixa similaridade genética entre elas e as estirpes de referência, o que indica a grande variabilidade genética dos isolados. Os resultados deste trabalho constituem o primeiro relato sobre bactérias capazes de nodular esta espécie endémica. O estudo mais aprofundado destas bactérias poderá revelar a existência de isolados tolerantes a stresses ambientais e adequados a serem utilizados como futuros inoculantes de mulungu.

Palavras-chave: Fixação biológica de azoto, inoculante, leguminosa arbórea, rizóbio.

Introduction

The Fabaceae (Leguminosae) is one of the most diverse and abundant among the angiosperm families with three subfamilies: Mimosoideae, Faboideae (Papilionoideae) and Caesalpinioideae and represents approximately 18.000 species of about 650 genera (Souza and Lorenzi, 2008). In the “Caatinga”, a biome that covers most part of the Brazilian northeastern region, about 86 genera and more than 320 species of this family are found (Queiroz, 2009).

Erythrina is a pantropical genus harboring more than 100 species. The genus belongs to the Faboideae subfamily and Phaseoleae tribe (Sprent and Parsons, 2000). It is commonly used as an ornamental plant due to its attractive flowering and also because of its popular medicine (Sprent and Parsons, 2000; Lorenzi, 2002; Castro and Cavalcante, 2011).

Studies on the diversity of rhizobia that are associated with autochthonous plants from the Semiarid, mainly those endemics species as mulungu, for example, are still very scarce. The phenotypical and molecular characterization of the bacteria obtained from these plants may contribute to increase the knowledge about physiological features and taxonomic position of native rhizobia and make possible the selection of bacteria that presents adaptations to harsh edaphoclimatic conditions and diverse biotechnological applications. To our knowledge, there is no available data on the characterization of rhizobial isolates from mulungu.

To select efficient rhizobial strains for BNF, phenotypic characterization has been one of the most used methods for an initial classification (Wolde-Meskel et al., 2004; Florentino et al., 2014) and may play an important role in the identification of rhizobial isolates. Phenotypic characterization is the basis of polyphasic taxonomy and, together with molecular techniques, has contributed to an improved understanding of the physiology and genetic diversity of rhizobia (Zerhari et al., 2000; Fall et al., 2008; Rasul et al., 2012). Among the molecular characterization techniques, fingerprinting by means of Box-PCR generated profiles have been used to examine the diversity of diazotrophic bacteria associated with the cultivated species (Stocco et al., 2008; Costa et al., 2014; Torres-Júnior et al., 2014) and native species (Fernandes Júnior et al., 2013; Granada et al., 2014) from Brazil. This simple and low-cost technique allows the discrimination among several rhizobial isolates and reference strains, making possible a fast fingerprinting and selection of bacterial isolates. The aim of this study was to determine the phenotypic characteristics and molecular fingerprinting of Erythrina velutina Willd. rhizobial isolates from semiarid soils.

Material and Methods

A pot experiment was set up using 500 mL soil capacity pots. Four superficial soil samples (0-0.2 m), two from Bahia state (municipality of Juazeiro) and two from Pernambuco state (municipalities of Lagoa Grande and Araripina). The soil samples used for this trial were collected under Caatinga vegetation with different land uses. A subsample of each studied soil was examined for its chemical characteristics according to Embrapa (1997). The soil chemical characteristics and the area description are described in the Table 1. For the trap host experiment, the mulungu seeds were scarred by mechanical abrasion and superficially disinfected with ethanol at 96° GL, sodium hypochlorite at 2 % (w/v) for 5 minutes and washed 10 times with sterile distilled water (SDW). The seeds were sown, six per pot, and thinned by two plants per pot, 20 days after germination. The plants received SDW according to necessity and grown for 88 days after which they were harvested and the nodules were detached from the roots.

For isolation purposes, the nodules were superficially disinfected with ethanol at 96° for 30 seconds, with sodium hypochloride at 2% (w/v) for 5 minutes and eight washings in SDW (Vincent, 1970) and, with the aid of pincers, macerated onto Petri dishes containing YMA medium with Congo Red. The dishes were incubated at 28°C in a growth chamber and the appearance of the characteristic rhizobial colonies was monitored daily. After that, the colonies were also inoculated on dishes containing YMA medium with bromothymol blue, incubated for their appropriate growth time and purified in the same medium. The bacteria were stored in YMA medium containing glycerol at 30% (w/v) in a freezer at 20°C.

Nodulation capacity of the isolates for mulungu was examined in greenhouse conditions experiment using pots containing autoclaved sand as substrate. For this test, seeds were scarified, superficially disinfected and sown in 500 mL polyethylene pots. At the beginning three seeds per pot were sown. Thinning was carried out 20 days after emergence (DAE) and only one plant was left per pot. For the inoculation, the bacterial isolates were grown in liquid YM medium (Vincent, 1970) during the appropriate growth time for each isolate and 1 mL of broth culture was inoculated per seed at the sowing time and re-inoculated at 15 DAE. Besides the inoculation treatments with each isolate obtained from the mulungu nodules, a control treatment (not inoculated) was also included. The experiment was set with three replications in a completely randomized design up to 90 DAE when nodulation was evaluated. The characterization of all the obtained bacteria, was carried out to examine the further characteristics: alteration of the medium pH (acid, neutral or alkaline), growth time of the colonies (rapid: up to three days; intermediate: four to five days and slow: over six days), colony color, type (viscous or floculous) and quantity (low or high) of the mucus.

The isolates able to re-nodulate the host were assessed for other complementary phenotypic characteristics. Tolerance to salinity was examined by the bacterial growth in YMA medium supplemented with NaCl at 1 and 2 % (w/v) concentrations. Growth of the isolates at different temperatures was determined in solid YMA medium at incubation temperatures of 28 (control), 39 and 41°C in a growth chamber during the appropriate growth time for each isolates. The capacity of microbial isolates to metabolize different carbon sources was examined in modified YMA medium with mannitol substituted for arabinose, sodium acetate, maleic acid, malic acid, succinic acid, fructose, glucose, maltose, xylose and sucrose (Vetec Química Fina, Rio de Janeiro, Brazil) at 1% (w/v).

Intrinsic resistance to antibiotics was examined through plate diffusion method with impregnated discs (Bauer et al., 1966). The bacteria were previously cultivated at YM liquid medium during the time required for each isolate. The optic density of the culture broth was adjusted to O.D.= 0.5-0.6 in a spectrophotometer (Multiskan GO, Thermo Scientific) at 560 nm. The adjusted cell broth were inoculated at plate dishes containing YMA culture medium, spread with a Drygalsky loop and paper disks containing antibiotics were placed in the culture medium. The following antibiotics were used: streptomycin (10 µg), rifampicin (5 µg), neomycin (30 µg), erythromycin (15 µg), vancomycin (30 µg), nalidixic acid (30 µg), gentamicin (10 µg), ampicillin (10 µg) and chloramphenicol (30 µg) (Sensifar, São Paulo, Brazil). The sensibility or resistance of the bacterial isolates to the antibiotics was evaluated by the presence or the absence of a growth inhibition circle, respectively, surrounding the discs. The salt, temperature and intrinsic antibiotic resistance essays were conducted with three replications. The treatments with great differences among all replicates were repeated to be assured.

Molecular fingerprinting of the isolates was performed using the Box-PCR technique. For DNA extraction, the bacteria were grown in YM culture medium (Vincent, 1970) without the pH indicator for three days for bacteria with rapid growth and six days for bacteria with slow growth. The DNA was extracted with the commercial kit HiYield^TM Genomic DNA Mini Kit (Real Biotech Corporation, Taipei, Taiwan) following the manufacturer instructions. For comparison, the rhizobial strains Rhizobium tropici (BR 322), Bradyrhizobium sp. (BR 5609), Ensifer fredii (BR 4007) and Burkholderia sabiae (BR 3405 and BR 3407) were used as reference strains. The reference strains were obtained from the “Johanna Döbereiner Diazothrofic Bacteria Culture Collection” of Embrapa Agrobiologia (Seropédica, RJ).

For the Box-PCR reaction, the A1 Box primer (CTACGGCAAGGCGACGCTGACG) was used (Versalovic et al., 1994). The PCR reaction was dimensioned for 25 µL with a reaction buffer (1x), MgCl₂ (3 mM), dNTPs (1.5 µM of each base), 2 µM of the initiator and 1.25 U of Taq DNA Polymerase, apart from approximately 50 ng (1 µL) of the genomic DNA. The PCR was set up of an initial denaturation cycle at 94° C for 8 minutes, followed by 35 cycles of denaturation step at 94° C for 1 minute, annealing step at 53° C for 1 minute and extension step at 72° C for 8 minutes, completed with a final extension for 16 minutes (Hungria et al., 2008; Fernandes Júnior et al., 2013).

The PCR product was submitted to horizontal electrophoresis in agarose gel at 1.5% (w/v) at 120 V for six hours. The gel was stained with ethidium bromide (0.05% w/v) and visualized with a photodocumentation system with UV light. The digital image of the gel was analyzed using the Bionumerics software version 7.0 (Applied Maths, Kortrijk, Belgium). The bacterial Box-PCR profiles were clustered applying the Pearson coefficient and the UPGMA clustering method was used to construct the similarity dendrogram.

Results

The cultural characteristics of 22 isolates obtained from the mulungu nodules are described in Table 2. The bacteria showed various growth characteristics and most of them were fast growth. Besides these characteristics, there also was a predominance of isolates that acidify the medium pH to acid, produced high viscous mucus and cream-colored colonies. The 22 obtained isolates were tested in a greenhouse experiment and only eight of them (ESA 0068, ESA 0069, ESA 0070, ESA 0071, ESA 0072, ESA 0073, ESA 0074 e ESA 0075) were able to re-nodulate the mulungu plants and were selected for additional phenotypic and molecular assays. Among the eight bacteria that nodulated the original host in this study, half of them showed characteristics compatible with the isolates belonging to Bradyrhizobium sp. genus, such as increased pH and intermediate or slow growth. The other nodulating bacteria showed fast growth and acidified the culture medium, which are characteristics compatible with bacteria of the Rhizobium genus.

The eight isolates were tested for their growth in different concentrations of NaCl and incubation temperatures (Table 3). None of these isolates were able to growth in medium supplemented with 2% of NaCl. The isolates ESA 0068, ESA 0069 and ESA 0070 were able to growth in culture medium supplemented with 1% of NaCl whereas the remaining isolates (ESA 0071, ESA 0072, ESA 0073, ESA 0074 and ESA 0075) did not grow in the same NaCl concentration. In relation to growth at increased temperatures it was found that none of the tested isolates grew when incubated at 41° C. The isolates ESA 0068, ESA 0069, ESA 0070, ESA 0071, ESA 0072, ESA 0074 and ESA 0075 grew at an incubation temperature of 39° C, whereas the isolate ESA 0073 showed more sensitivity growing only in the control treatment (28° C).

All the isolates were able to metabolize maltose, fructose, glucose, sucrose and arabinose as carbon sources. The more limiting sources were malic acid, maleic acid, sodium acetate and succinic acid, which did not permit the growth of five among the eight isolates studied. Regarding the growth capacity of the bacteria in different C sources, isolates ESA 0068, ESA 0069 and ESA 0070 were the most versatile and grew in all sources tested. Isolates ESA 0071, ESA 0072 and ESA 0074, which showed distinct cultural characteristics, were not able to grow in five among ten carbon sources, which were: malic acid, maleic acid, sodium acetate, succinic acid and xylose. The evaluation of the intrinsic antibiotic resistance showed that gentamicin was found to be the antibiotic that most restricted bacterial growth. On the other hand, rifampicin did not limit growth of any of the tested bacteria. The isolate ESA 0068 was the most tolerant, being resistant to eight of the nine antibiotics used.

The ESA 0068, ESA 0069 and ESA 0070 bacteria showed the same culture characteristics in terms of growth time, colony color, production and type of mucus, and were among the most tolerant ones to heat (39°C), salt (1% NaCl) and antibiotics (six of the nine tested antibiotics) and growth in the tested carbon sources.

The analysis of the similarity dendrogram showed that all the bacteria isolated from the mulungu and the reference strains had about 40 % of similarity (Figure 2). At 60% of the similarity, the formation of distinct clusters can be observed. Clusters I encompasses the isolate ESA 0071 and the strains of Rhizobium tropici and Bradyrhizobium sp. The isolates ESA 0070 and ESA 0068 belonged to the cluster II, while cluster III was composed only by reference strains, with both Burholderia sabiae strains and Ensifer fredii. The cluster V was also formed only by two bacterial mulungu isolates (ESA 0074 and ESA 0075). Clusters IV, VI and VII were formed only by one bacteria isolated from mulungu root-nodule.

Discussion

The mulungu isolates showed different growth characteristics ranging from fast to slow. The most of fast growing bacteria acidified the pH of the culture medium and showed an increased production of exopolysaccharides whereas the isolates that showed slow growth alkalinized the pH of the medium and had low mucus production. Recently Freitas et al. (2014) have observed that all isolates obtained from Mimosa tenuiflora and M. paraibana also presented fast growth. Furthermore, bacteria that acidified or did not modify the pH of the medium and had low to average mucus production were found. Evaluating phenotypic characteristics of bacterial isolates from Gliricidia sepium in Brazil, Florentino et al. (2014) also reported that all isolates had fast growing, showed high mucus production and the majority were able to acidify the medium.

Among the 14 non-nodulating bacteria, 12 showed rapid growth and acidification of the medium. The increased percentage of bacteria which are not able to re-nodulate the original host species is commonly found in studies of bacteria isolation from nodules of species in tropical climates (Lima et al., 2012). This incapacity may occur due to the loss of plasmids that carry the symbiotic genes during the successive cultivations in the process of isolation, purification and storage (Garcia de los Santos, 1996). For tree species, isolation of non-symbiotic bacteria that co-inhabit the nodules, together with the rhizobia may also occur, as is described in the literature (Muresu et al., 2008; Shiairashi et al., 2010). These non-symbiotic organisms, because they take less time to appear in culture medium, occupy the Petri dishes more rapidly, which does not permit the observation of isolates that appear lately (Ourarhi et al., 2011).

Recent studies on phenotypic characteristics of isolates from tropical arboreal legumes have shown the existence of nodulating bacteria with characteristics that are compatible with those of the Rhizobium and Bradyrhizobium genera and equivalent frequencies between these phenotypic groups were observed (Gehlot et al., 2012), i.e., the predominance of rapid growth bacteria which, acidify the pH of the medium (Freitas et al., 2014) or the predominance of slow growth bacteria with an alkaline reaction in the culture medium (Baraúna et al., 2014).

In studies examining nodulation of Millettia pinnata, an arboreal legume that also belongs to the Faboideae subfamily, Rasul et al. (2012) obtained half of the isolates of intermediate growth while in the studies of Arpiwi et al. (2013) slow growth bacteria predominated. These characteristics seem to be common in isolates that nodulate the Faboideae tree legumes since the current study also found characteristics of intermediate growth (ESA 0072) and slow growth (ESA 0073, ESA 0074 and ESA 0075), nodulating mulungu.

Three isolates (ESA 0068, ESA 0069 and ESA 0070) showed tolerance to an increased concentration of salt in culture medium, and in addition to these, the strains ESA 0071, ESA 0072, ESA 0074 and ESA 0075 were resistant to a temperature of 39°C. Rhizobia of tree legumes that are tolerant to increased salinity and temperature have already been documented in other studies (Zerhari et al., 2000; Fterich et al., 2011; Shetta et al., 2011; Rasul et al., 2012; Arpiwi et al., 2013; Florentino et al., 2014). High tolerance to salinity and temperature is an important characteristic for the survival of rhizobia in environmentally unfavorable conditions. The fact that the majority of the bacteria from this study grow at 39°C may be related to their adaptation to edaphoclimatic conditions of the Semiarid.

All the isolates tested were able to metabolize different carbon sources, which demonstrate the increased metabolic versatility of these bacteria. Arpiwi et al. (2013) verified that in relation to the use of different carbon sources, the majority of the isolates obtained from Millettia pinnata showed growth in medium containing glucose and arabinose and a considerable number did not grow in medium containing sucrose. The use of a greater variety of carbon sources by the bacterial isolates may constitute a characteristic of the adaptation to the different environmental conditions that are influenced by the root exudates of the hosts (Shetta et al., 2011).

Furthermore, the three isolates that showed increased adaptation to high temperatures, salinity and antibiotics (ESA 0069, ESA 0070 and ESA 0068) are fast growing isolates. Other authors also confirmed the same tolerance characteristics of bacteria with the same growth characteristic (Zerhari et al., 2000; Fall et al., 2008; Rasul et al., 2012), which indicates that the studied mulungu isolates show a potential for inoculation in the semiarid regions because they are well adapted to the typical stress conditions of these soils.

The capacity of IAA synthesis of rhizobia from tree species has already been documented in the literature. Manasilla et al. (2007), who studied rhizobia obtained from the nodules of the tree legumes Acacia auriculaformis Cunn., A. angium Willd., Milletia leucantha Kurz., Pterocarpus indicus Willd. and Xylia xylocarpa Taub. from Thailand, found that only four isolates were capable of producing IAA, being two from A. auriculiformis, one from P. indicus and one from X. xilocarpa. Rasul et al. (2012), who examined IAA production of 29 rhizobia, observed that the majority did not show the capacity to produce this phytohormone because only seven synthesized IAA. The results of the present study indicate that at least one isolate of rapid growth (ESA 0071) and one isolate of slow growth (ESA 0075) produced IAA quantities that are only produced by strain BR 3262. In addition to the capacity of nodulating and fixing N, the results indicate that these bacteria show complementary mechanisms for increase the plant growth and may positively influence the development and establishment of mulungu in the field.

When examining the genetic similarity between the mulungu isolates and reference strains, it is possible to observe that these bacteria show around 40% of similarity, which indicates that mulungu isolates are not closely related to the reference strains. In this study, the isolate that showed the largest similarity with any reference strain was the isolate of rapid growth ESA 0071, which showed about 77% of similarity with the R. tropici strain BR 322. The large genetic variability of the rhizobial isolates from tree legumes from the Brazilian Semiarid was already showed to other species, such as the Mimosa spp. (Teixeira et al., 2010; Freitas et al., 2014) and the data of the present study corroborate the reports already published.

The Box-PCR technique is an easy and low cost technique that has been recently used in several studies to examine genetic variability of nitrogen-fixing bacteria obtained from different hosts. Among the cultivated species, studies conducted in Brazil have already examined the genetic variability of rhizobia of legumes such as pigeonpea (Costa et al., 2014), peanuts (Lyra et al., 2013; Torres-Júnior et al., 2014), common beans (Stocco et al., 2008), among others. This technique is also used to examine the variability of strains isolated from native plants such as bromeliad (Giongo et al., 2013), wild rice (Fernandes Júnior et al., 2013) and legumes (Granada et al., 2014). The results obtained in the present study, demonstrate its applicability and indicate that this technique can be used with rhizobia of other native tree legumes.

The results of the present study indicate that mulungu shows effective rhizobial symbionts in the Brazilian Semiarid region. Complementary studies examining the taxonomic positioning and their symbiotic efficiency will be important to demonstrate the biotechnological potential of these isolates.

Conclusions

The isolates of rhizobia obtained from mulungu (Erythrina velutina) nodules show various phenotypic characteristics and low genetic similarity with the rhizobia reference strains that were used, which indicates that we can be dealing with still unknown species.

To “Coordenação de Aperfeiçoamento de Pessoal de Nível Superior” (CAPES) and to the “Conselho Nacional de Desenvolvimento Científico e Tecnológico” (CNPq) by the granted scholarships. To “Universidade do Estado da Bahia” and to “Embrapa Semiárido” for their financial assistance and infrastructure.

References

Arpiwi, N.L.; Yan, G.; Barbour, E.L.; Plummer, J.A. and Watkin, E. (2013) - Phenotypic and genotypic characterisation of root nodule bacteria nodulating Millettia pinnata (L.) Panigrahi, a biodiesel tree. Plant and Soil, vol. 367, n. 1-2, p. 363-377. http://dx.doi.org/10.1007/s11104-012-1472-4        [ Links ]

Baraúna, A.C.; Silva, K.; Pereira, G.M.D.; Kaminski, P.E.; Perin, L. and Zilli, J.E. (2014) - Diversity and nitrogen fixation efficiency of rhizobia isolated from nodules of Centrolobium paraens. Pesquisa Agropecuária Brasileira, vol.49, n.4, p. 296-305. http://dx.doi.org/10.1590/S0100-204X2014000400008        [ Links ]

Bauer, A.W.; Kirby, W.M.M.; Sherris, J.C. and Turck, M. (1966) - Antibiotic susceptibility testing by it standardized disk method. American Journal of Clinical Pathology, vol. 45, n. 4, p. 493-496.         [ Links ]

Castro, A.S. and Cavalcante, A. (2011) - Flores da Caatinga. Instituto Nacional do Semiárido, Campina Grande, 116 p.         [ Links ]

Costa, F.M.; Schiavo, J.A.; Brasil, M.S.; Leite, J.; Xavier, G.R. and Fernandes Júnior, P.I. (2014) - Phenotypic and molecular fingerprinting of fast growing rhizobia of field-grown pigeon pea from the eastern edge of the Brazilian Pantanal. Genetics and Molecular Research, vol. 13, n. 1, p. 469-482. http://dx.doi.org/10.4238/2014.January.21.16        [ Links ]

Embrapa (1997) - Manual de Métodos de Análises de solo. 2ª ed. Rio de Janeiro, Ministério da Agricultura, Pecuária e Abastecimento, 212 p.         [ Links ]

Fall, D.; Diouf, D.; Ourarhi, M.; Faye, A.; Abdelmoumen, H.; Neyra, M.; Sylla, S.N.; Missbah, E.L. and Idrissi, M. (2008) - Phenotypic and genotypic characteristics of Acacia senegal (L.) Willd. root-nodulating bacteria isolated from soils in the dryland part of Senegal. Letters in Applied Microbiology, vol. 47, n. 2, p. 85-97. http://dx.doi.org/10.1111/j.1472-765X.2008.02389.x        [ Links ]

Fernandes Júnior, P.I.; Pereira, G.M.D.; Perin, L.; Silva, L.M.; Baraúna, A.C.; Alves, F.M.; Passos, S.R. and Zilli, J.E. (2013) - Diazotrophic bacteria isolated from wild rice Oryza glumaepatula (Poaceae) in the Brazilian Amazon. Revista de Biologia Tropical, vol. 61, n. 2, p. 991-999.         [ Links ]

Florentino, L.A.; Rezende, A.V.; Mesquita, A.C.; Lima, A.R.S.; Marques, D.J. and Miranda, J.M. (2014) - Diversidade e potencial de utilização dos rizóbios de nódulos de Gliricidia sepium. Revista de Ciências Agrárias, vol. 37, n. 3, p. 320-328.         [ Links ]

Freitas, A.D.S.; Sampaio, E.V.S.B.; Santos, C.E.R.S and Silva, A.F. (2010) - Biological nitrogen fixation in tree legumes of the brazilian semi-arid Caatinga. Journal of Arid Environments, vol. 74, p. 344-349. http://dx.doi.org/10.1016/j.jaridenv.2009.09.018        [ Links ]

Freitas, A.D.S.; Borges, W.L.; Andrade, M.M.M.; Sampaio, E.V.S.B.; Santos, C.E.R.S.; Passos, S.R.; Xavier, G.R.; Mulato, B.M. and Lyra, M.C.C.P. (2014) - Characteristics of nodule bacteria from Mimosa spp. grown in soils of the Brazilian semiarid region. African Journal of Microbiology Research, vol. 8, n. 8, p. 788-796. http://dx.doi.org/10.5897/AJMR2013.6518        [ Links ]

Fterich, A.; Mahdhi, M.; Caviedes, M.A.; Pajuelo, E.; Rivas, R.; Rodriguez-Llorente, I.D. and Mars, M. (2011) - Characterization of root-nodulating bacteria associated to Prosopis farcta growing in the arid regions of Tunisia. Archives of Microbiology, vol. 193, n. 6, p. 385-397. http://dx.doi.org/10.1007/s00203-011-0683-z        [ Links ]

García de los Santos, A.; Brom, S. and Romero, D. (1996) - Rhizobium plasmids in bacteria-legume interactions. World Journal of Microbiology & Biotechnology, vol. 12, n. 2, p. 119-125. http://dx.doi.org/10.1007/BF00364676        [ Links ]

Gehlot, H.S.; Panwar, D.; Tak, N.; Tak, A.; Sankhla, I.S.; Poonar, N.; Parihar, R.; Shekhawat, N.S.; Kumar, M.; Tiwari, R.; Ardley, J.; James, E.K. and Sprent, J.I. (2012) - Nodulation of legumes from the Thar desert of India and molecular characterization of their rhizobia. Plant and Soil, vol. 357, n. 1-2, p. 227-243. http://dx.doi.org/1007/s11104-012-1143-5        [ Links ]

Giongo, A.; Beneduzi, A.; Gano, K.A.; Vargas, L.K.; Utz, L.R.P. and Passaglia, L.M.P. (2013) - Characterization of plant growth-promoting bacteria inhabiting Vrisea gigantea Gaud, and Tillandsia aeranthos (Loiseleur) L. B. Smith (Bromeliacea). Biota Neotropica, vol. 13, n. 3, p. 80-85. http://dx.doi.org/10.1590/S1676-06032013000300010        [ Links ]

Granada, C.E.; Strochein, M.; Vargas, L.K.; Bruxel, M.; Sá, E.L.S. and Passaglia, L.M.P. (2014) - Genetic diversity and symbiotic compatibility among rhizobial strains and Desmodium incanum and Lotus spp. plants. Genetics and Molecular Biology, vol. 37, n. 2, p. 396-405. http://dx.doi.org/10.1590/S1415-47572014000300012        [ Links ]

Hungria, M.; Chueire, L.M.O.; Menna, P. and Bangel, E.V. (2008) - Caracterização genética de rizóbios e outras bactérias diazotróficas e promotoras do crescimento de plantas por BOX-PCR. Londrina, Embrapa Soja, 12 p. (Embrapa Soja. Comunicado Técnico, 290).         [ Links ]

Lima, A.A.; Fernandes Júnior, P.I.; Passos, S.R.; Paulo, F.S.; Nosoline, S.M.; Faria, S.M.; Guerra, J.G.M.; Rumjanek, N.G. and Xavier, G.R. (2012) - Diversidade e capacidade simbiótica de rizóbios isolados de nódulos de Mucuna-Cinza e Mucuna-Anã. Revista Brasileira de Ciência do Solo, vol. 36, n. 2, p. 337-348. http://dx.doi.org/10.1590/S0100-06832012000200003        [ Links ]

Lorenzi, H. (2002) - Árvores brasileiras: manual de identificação e cultivo de plantas arbóreas nativas do Brasil. 4ª ed. Plantarum, Nova Odessa, 352 p.         [ Links ]

Lyra, M.C.C.P.; Freitas, A.D.S.; Silva, T.A. and Santos, C.E.R.S. (2013) - Phenotypic and molecular characteristics of rhizobia isolated from nodules of peanut (Arachis hypogaea L.) grown in Brazilian Spodosols. African Journal Biotechnology, vol. 12, n. 17, p. 2147-2156. http://dx.doi.org/10.5897/AJB11.1574        [ Links ]

Manassila, M.; Nuntagij, A.; Kotepong, S.; Boonkerd, N. and Teaumroong, N. (2007) - Characterization and monitoring of selected rhizobial strains isolated from tree legumes in Thailand. African Journal Biotechnology, vol. 6, n. 12, p. 1393-1402.         [ Links ]

Moreira, F.M.S. and Siqueira, J.O. (2006) - Microbiologia e Bioquímica do Solo. 2ª ed. UFLA, Lavras, 729 p.         [ Links ]

Muresu, R.; Polone, E.; Sulas, L.; Baldan, B.; Tondello, A.; Delogu, G.; Cappuccinelli, P.; Alberghini, S.; Benhizia, Y.; Benhizia, H.; Benguedouar, A.; Mori, B.; Calamassi, R.; Dazzo, F.B. and Squartini, A. (2008) - Coexistence of predominantly nonculturable rhizobia with diverse, endophytic bacterial taxa within nodules of wild legumes. FEMS Microbiology Ecology, vol. 63, n. 3, p. 383-400. http://dx.doi.org/10.1111/j.1574-6941.2007.00424.x        [ Links ]

Received/recebido: 2015.04.07

Accepted/aceite: 2015.08.10