versión impresa ISSN 1405-3195
Agrociencia vol.42 no.1 México ene./feb. 2008
Bacteria associated with the extraradical mycelium of an arbuscular mycorrhizal fungus in an As/Cu polluted soil
Bacterias que se asocian al micelio extraradical de un hongo arbuscular en suelo contaminado con As y Cu
Ma. del Carmen A. GonzálezChávez1, Ray Newsam2, Robert Linderman3, John Dodd4, and Jorge M. ValdezCarrasco1
1 Campus Montecillo. Colegio de Postgraduados. 56230. Carretera MéxicoTexcoco, km 36.5. Montecillo, Estado de México, México. (firstname.lastname@example.org).
2 Department of Biosciences. University of Kent, Canterbury CT2 7NJ, UK.
3 U. S. Department of Agriculture. ARS Horticultural Crops Research Laboratory. Corvallis, OR, 97330, U.S.A.
4 PlantWorks Ltd, Kent Science Park, Sittingbourne, Kent ME9 8HL, UK.
Recibido: Agosto, 2006.
Aprobado: Octubre, 2007.
Synergistic interactions between bacteria and arbuscular mycorrhizal fungi (AMF) occur under natural soil conditions; however, in polluted soils there is little information regarding these relationships. Microscopy was used to study the interaction between the hyphae of an AM fungus (Glomus claroideum BEG134 from an As/Cu polluted soil) and bacteria in polluted soil cultures. BacLightTM staining showed viable bacteria mainly in the runner hyphae of the fungus associated with plants (Holcus lanatus L.) growing in polluted soils. Transmission electron microscopy revealed that a morphologically different bacterial population was intimately associated with the extraradical mycelium (ERM). Bacteria were embedded in the mucilaginous outer layer, encrusted at the outer layer, between hyphal wall layers, and inside hyphae. Crystals, comprising precipitated metal, were observed outside the ERM. The ecological relevance of this bacteriaAMF interaction is discussed.
Key words: Glomus claroideum, bacterial interaction, endosymbiosis, mycorrhizosphere, rhizosphere synergism.
La interacción sinérgica entre bacterias y hongos micorrízicos arbusculares (HMA) ocurre bajo condiciones naturales del suelo; sin embargo, hay poca información de ésta en suelos contaminados. Se estudió microscópicamente la interacción entre las hifas de un hongo MA (Glomus claroideum BEG134, aislado de un suelo contaminado con As y Cu), y bacterias en plantas de Holcus lanatus L. en suelo contaminado. La tinción con BacLightTM mostró bacterias vivas, principalmente en las hifas corredoras del hongo asociado a plantas (Holcus lanatus L.) que crecieron en suelo contaminado. Por microscopía electrónica de transmisión se observó que una población bacteriana morfológicamente diferente se asoció íntimamente con el micelio extraradical (MER). Las bacterias estaban embebidas en la capa mucilaginosa externa, incrustadas en la pared externa, entre las capas de la pared hifal, y dentro de las hifas. Se observaron cristales, conteniendo metal precipitado, fuera del MER. Se discute la importancia ecológica de la interacción bacteriaHMA.
Palabras clave: Glomus claroideum, interacción bacteriana, endosimbiosis, micorrizosfera, sinergismo en la rizosfera.
In the rhizosphere, a dynamic soil zone influenced by roots, microorganisms interact with each other in synergistic and antagonistic processes (Lynch, 1990). Under natural conditions soil bacteria, actinomycetes and fungi may be associated with arbuscular mycorrhizal fungi (AMF) (Vancûra et al., 1989; Walley and Germida, 1996).
Studies from agricultural soils or in vitro observations have focused on the association between bacteria and AMF, especially in relation to plant growth (Paulitz and Linderman, 1989) and biological control of root pathogens (Linderman, 2001). The nature of this association has been studied; for example, the direct physical interaction between germinated spores of Gigaspora margarita with a suspension of several strains of either Rhizobium leguminosarum or Pseudomonas fluorescens was evaluated by Bianciotto et al. (1996a), who observed that these rhizobacteria interacted with spore and hyphae (from germ tubes) of Gi. margarita under sterile conditions, but the degree of interaction depended upon the strain. They also suggested that AMF are a vehicle for the colonization of plant roots by soil rhizobacteria. According to Bianciotto et al. (1996b), the cytoplasm of Gi. margarita spores harbors a live Burkholderia population; similar results were observed in Scutellospora sp., but not in Glomus mosseae or Acaulospora laevis. In germinating AMF spores the expression of nif (nitrogen fixation) genes could indicate that the Burkholderia endobacteria supply the fungus with nitrogen during its preinfection growth (Minerdi et al., 2001); however, studies by Bianciotto and Bonfante (2002) did not confirm this.
AMF are beneficial to plants in polluted soils (GonzálezChávez et al., 2004a). However, the interactions between bacteria and AMF in polluted soils and their significance for remediation practices have been poorly studied. Bacteria and AMF, adapted to metals, may increase metal tolerance in their host plants (Vivas et al., 2003, 2006). Understanding the nature of microbial interactions in polluted soils is important when developing phytoremediation technologies, as microbial species are commonly involved in soil metal transformations and degradation of organic compounds (Gadd, 1993; Anderson et al., 1993). This is especially important at the extraradical mycelium (ERM), since MansfeldGiese et al. (2002) demonstrated that the ERM has a stronger influence on bacterial population density than roots colonized by AMF.
The objective of this research was to microscopically study the interaction of bacteria associated with the ERM of Glomus claroideum (BEG134) from an As/ Cu polluted soil, simulating polluted conditions.
MATERIALS AND METHODS
A singlespore fungal culture of G. claroideum BEG134, isolated from Great Consol Mines, Devon, UK (1.6 km North of Gunnislake, 50° 31' N, 4° 12' W), was used in this research. Unrooted tillers of arsenateresistant Holcus lanatus L. were used as a host plant in the propagation of the monosporic cultures.
After verifying the purity of the G. claroideum culture (by the absence of another spore type), it was propagated for four months in 15 cmdiam pots on an attapulgite clay (Agsorb 8/16, OilDri, Wisbech, Cambs, UK) amended or not with 20 g of polluted soil that had been yirradiated (10 kGy). Thus, polluted and nonpolluted fungal AMF cultures were obtained after inoculation with 200 spores per pot. Soil was polluted with DTPATEACaCl2 available arsenic (130 mg kg1) and copper (97 mg kg1) (Lindsay and Norvell, 1978), had pH 6.4, 0.9% organic matter and 3.4 mg P kg1. This soil and the plant host, H. lanatus were obtained from the same area where G. claroideum BEG134 was isolated. The fungal cultures were grown in a glasshouse at temperatures between 12 and 25 °C. Plants were watered every 2 d to maintain a water holding capacity of 80% and nutrients were added weekly as 1.4 g L"1 Vitafeed 102 (Vitax Ltd, Leicester, UK) containing 15% N, 0% P, 36% K.
Microscopic detection of bacteria on ERM
BacLightTM staining and fluorescence microscopy
After four months, five samples of the ERM of G. claroideum BEG 134 were collected from the pot cultures using the wet sieving and decanting method (Gerdemann and Nicolson, 1963). Hyphae collected on a 43 µm nylon mesh were washed twice with sterile deionizeddistilled water and stained with the Live/Dead BacLightTM Bacteria Viability Kit (Molecular Probes Europe BV). BacLight stain is a twocolour fluorescence assay of bacterial viability. Green fluorescence is observed in bacteria with intact plasma membranes (alive) and red fluorescence in bacteria with damaged membranes (dead). The time of incubation was 15 min at room temperature in the darkness (Manufacturer's instructions). Slides of stained ERM were prepared and observed under a fluorescence microscope (Leitz DMRB). Bacteria attached to the hyphae were not quantified.
Transmission electron microscopy (TEM)
The ERM was extracted and washed twice with 0.01 M phosphate buffer solution (PBS, pH=7.3) and transferred to 2.5% glutaraldehyde in PBS. Fixation was carried out for 3 h at 4 °C and was followed by two washes in PBS for 10 min. Postfixation in 1 % (w/v) OsO4 solution for 2 h was performed before dehydration in alcohol series (30%, 50%, 70%, 80%, 90% and 100%, 15 min each step) at room temperature. Samples of ERM were embedded in Spurr's resin. Polymerization occurred overnight. Using a diamond knife, mounted on an ultratome Nova (Leica), 100 µm sections were made. Samples were stained with uranyl acetate (30 min, 37 °C) followed by lead citrate (10 min) and were observed under a TEM (Phillips 410). Ten thin sections were observed from hyphae growing in nonpolluted and polluted substrate.
RESULTS AND DISCUSSION
BacLight staining revealed living bacteria associated with the ERM of G. claroideum (BEG134) and on young spores (Figure 1). Bacterial accumulation was along the thick runner hyphae and at their bifurcations (Figure 1 A, B). Runner hyphae (>20 µm diameter) form entry points on the root surface, as well as the skeleton of the ERM in the substrate and may be an efficient bacterial transporter along the mycorrhizosphere and soil (Dodd et al., 2000).
Living bacteria (green fluorescence) were observed in the ERM extracted from the nonpolluted and polluted substrate; in this last substrate, bacteria were easily observable. High concentrations of metals can potentially alter microbial morphology (GardeaTorresdey et al., 1997), making them more detectable in the metalcontaminated soil and giving the appearance of greater abundance. In non polluted conditions, Bianciotto et al. (1996a) observed that bacterial cells were irregularly distributed, producing patches around the hyphae of germinated spores of Gi. margarita.
Using fluorescence and transmission electron microscopy, it was demonstrated that under polluted conditions, living bacteria and the ERM of G. claroideum BEG 134 were very intimately associated (Figure 2). The association of bacteria and the hyphosphere of AMF in nonpolluted conditions has been reported, but the physical and cellular interaction in the ERM was studied for the firts time in the present research under polluted soil conditions. Andrade et al. (1997) analyzed bacterial population in the hyphosphere (defined by them as soil not adhering to roots) by counting colonyforming units in a non selective medium and identifying it via fatty acid methyl ester analysis (FAME). They found qualitative changes in bacterial communities affected by different AMF in the hyphosphere. Interestingly, successful establishment of Alcaligenes eutrophus (reclassified as Ralstonia eutropha) in soil depended on the presence of AM fungal hyphae and not on the presence of host roots (Andrade et al., 1998). Using FAME analysis MansfeldGiese et al. (2002) observed that Paenobacillus was mainly associated with the hyphosphere (defined as rootfree compartment) of G. intraradices, but it was not elucidated whether these bacteria were living in the proximity, on the surface or inside the mycelium. Under unpolluted soil conditions, a Bacillus cereus Swedish strain was attached to hyphae of Glomus dussii at significantly higher levels than bacterial control strains (Artursson and Jansson, 2003). Toljander et al. (2006) also compared the attachment of five different (green fluorescent protein) gfptagged bacterial strains to the ERM of G. claroideum. These last two reports studied the superficial interaction of the ERM and bacteria, but a more detailed cellular interaction was not attempted.
Different features of the association between bacteria and the ERM under metal polluted soil conditions as well as the superficial and internal bacterial interactions in the ERM were shown in our research. Bacteria were observed outside and inside the ERM; when outside the ERM, they were mainly embedded in the mucilaginous layer on the outer hyphal wall. Using microscopic interference contrast in 3dimension, GonzálezChávez et al. (2004b) showed the mucilaginous outer cell wall and sloughed material from hyphae of the ERM. Bianciotto et al. (1996a) also observed the mucilaginous cell wall in the hyphae of germinated Gi. margarita spores, with the formation of interstices and surface irregularities, which was suggested as a preferential bacterial microniche.
In Figure 2 CD it is shown clearly a remarkable bacterial inmersion into the mucilaginous on the hyphal cell wall. Additionally, on the mucilaginous layer on hyphae extracted from the polluted substrate (Figure 2 C, E), but not from nonpolluted substrate, abundant crystals were present. These crystals occurred not only around sites where bacteria were embedded, but also in segments of hyphae of G. claroideum BEG 134 where bacteria were absent (Figure 2 E). This result confirms that the hyphae are participating in the precipitation of Cu at the cell wall (GonzálezChávez et al., 2002).
The chemical nature of the sloughed mucigel material is unknown, but a protein (glomalin) produced by AMF hyphae, seems to represent this fungal mucigel. Some of its properties are: glomalin is extracted from hyphae of all AMF tested, it is an insoluble, gluelike and hydrophobic glycoprotein, with Nlinked oligosaccharides and 0.88% iron (Wright and Upadhayaya, 1998).
As metal sequestration depends on electrostatic characteristics of the fungal wall (Morley and Gadd, 1995), glomalin properties suggest its role in metal immobilization, thus affecting metal bioavailability. GonzálezChávez et al. (2004a) showed that glomalin produced by AMF hyphae is able to sequester Cu and other metals (Cd, Pb and Zn).
By fluoresence microscopy and BacLightTM staining, bacteria were visualized on AMF hyphae from polluted and nonpolluted soil conditions; however, by using TEM, we could not observe bacteria in the hyphae from the nonpolluted fungal culture (Figure 2A). Therefore, stronger bacterial retention by the AMF hyphae may be occurring under polluted conditions, which would be explained by:
1) Hyphae from As/Cu polluted soil contained metal crystals on their surfaces, which appeared to increase retention of bacteria in the hyphae; then, during sample preparation it is more difficult to wash bacteria off of crystallized hyphal surfaces than noncrystalized surfaces in hyphae from nonpolluted cultures.
2) In a more stable bacterial binding, microbial cell components may be involved (Toljander et al., 2006), whereas in polluted conditions more mucigel (glomalin) is produced and bacteria may be embedded and retained in it. In our lab it has been shown that the concentration of glomalin is increased in the presence of metals (Cd and Pb; CuellarSánchez et al., unpublished results), and there is irregular bacterial attachment along the root, depending on the quality of mucigel and the exudates produced by the host (Wiehe et al., 1994).
3) Bacterial ability to stick to the mucigel surface. Bianciotto et al. (1996a) observed that Rhizobium leguminosarum strain B556 and Pseudomonas fluorescens strain WCS365 heavily colonized fungal surfaces of Gi. margarita, but very rare cells of P. fluorescens strains CHAO, F113G22 and F113 were found on the fungal surface. Besides, there are major differences in the bacterial strains' ability to attach to hyphae and bacterial attachment may be affected by a diminution in electrostatic attraction by washing the hyphae with strong solutions before microscopic examination (Toljander et al., 2006).
Vancûra et al. (1989) showed that the hyphosphere selected gramnegative bacteria from the rhizosphere, but no fluorescent pseudomonads were present. In contrast, in our study, preliminary biochemical identification suggested that Pseudomonas spp. were present in the ERM and were the most easily isolated from the ERM (data not shown). However, this does not adequately define which bacteria were associated in the ERM because different bacteria forms were microscopically observed in situ (Figure 2BD). Molecular studies should be used to identify culturable and nonculturable bacteria associated with the ERM. Additionally, in vitro tolerance tests showed that these bacteria were Cutolerant (data not shown). Hence, the ecological relevance of these bacteria interacting with the ERM and their effects on plants and soil under metalpolluted conditions should be studied, since bacteria and AMF adapted to metals alleviate toxicity in their host plants (Vivas et al., 2003, 2006).
Bacteria were also observed inside AMF hyphae in our study (Figure 2 EF). The association of endobacteria with mycorrhizas was first reported by McDonald and Chandler (1981). Bianciotto et al. (1996b) showed that bacteria of the genus Burkholderia were endosymbionts in all life cycle stages of Gi. margarita as well as in two Scutellospora species, but not in different isolates of Gi. rosea (Bianciotto et al., 2000). Endobacteria in fungi is not a common event because fungi contain a physically strong cell wall which prevents bacterial penetration (De Boer et al., 2005); however at hyphal tips this occasionally may occur, an event more common in damaged hyphae or when fungi are attacked by lytic bacteria. Levy et al. (2003) observed lysis of spores due to bacteria and in our work bacteria were found either encrusted at the outer layers of the hyphal wall or between the layers of the hyphal wall (Figure 2 BC), especially bacteria in the hyphal cell wall with light degradation (Figure 2B). This may be one of the first events to final penetration involving mycolytic bacterial producers of chitinases, glucanases, proteases and antibiotics acting on living hyphae (Levy et al., 2003). However, this hypothesis needs to be probed in AMF.
Synergistic activities of AMF and bacteria are potentially useful in bioremediation processes often found in heavy metalpolluted areas (Trevors and van Elsas, 1997). In addition, bacteria are important to plants due to their potential to produce siderophores, plant growthpromoting substances, antifungal compounds, and participate in the degradation of organic pollutants and in nitrogen fixation (Paulitz and Linderman, 1989; Vancûra et al., 1989).
The extraradical mycelium of G. claroideum BEG134 was intimately associated with a bacterial population when grown in a polluted substrate. This kind of association may have important ecological contributions to plant survival, metal tolerance and nutrition. However, the ecological role of AMF and soil microbial associates should be elucidated by studying microbial interactions under polluted conditions.
The critical review from Sara F. Wright and anonymous referees is greatly appreciated. CGC thanks Dra. Hilda Araceli Zavaleta Mancera and M.C. Iván Mauricio Andrade Luna for their initial help with the graphic work. This paper is part of the project research SEMARNATCONACYT C0012002739.
Anderson, T. A., E. A. Guthrie, and B. T. Walton. 1993. Bioremediation in the rhizosphere. Environ. Sci. Technol. 27: 26302636. [ Links ]
Andrade, G., K. L. Mihara, R. G. Linderman, and G. J. Bethlenfalvay. 1997. Bacteria from rhizosphere and hyphosphere soils of different arbuscular mycorrhizal fungi. Plant Soil 192: 7179. [ Links ]
Andrade, G., R. G. Linderman, and G. J. Bethlenfalvay. 1998. Bacterial associations with the mycorrhizosphere and hyphosphere of the arbuscular mycorrhizal fungus Glomus mosseae. Plant Soil 202: 7987. [ Links ]
Artursson, V., and J. K. Jansson. 2003. Use of bromodeoxyuridine immunocapture to identify active bacteria associated with arbuscular mycorrhizal hyphae. Appl. Environ. Microbiol. 69: 62086215. [ Links ]
Bianciotto, V., and P. Bonfante. 2002. Arbuscular mycorrhizal fungi: a specialized niche for rhizospheric and endocellular bacteria. Antonie van Leeuwenhoek 81: 365371. [ Links ]
Bianciotto, V., E. Lumini, L. Lanfranco, D. Minerdi, P. Bonfante, P., and S. Perotto. 2000. Detection and identification of bacterial endosymbionts in arbuscular mycorrhizal fungi belong to the family Gigasporaceae. Appl. Environ. Microbiol. 66: 45034509. [ Links ]
Bianciotto, V., C. Bandi, D. Minerdi, M. Sironi, H. Volker Tinchi, and P. Bonfante. 1996a. An obligately endosymbiotic mycorrhizal fungus itself harbors obligately intracellular bacteria. Appl. Environ. Microbiol. 62: 30053010. [ Links ]
Bianciotto, V., D. Minerdi, S. Peroto, and P. Bonfante. 1996b. Cellular interactions between arbuscular mycorrhizal fungi and rhizosphere bacteria. Protoplasma 193: 123131. [ Links ]
de Boer, W., L. B. Folman, R. C. Summerbell, and L. Body. 2005. Living in a living world: impact of fungi on soil bacterial niche development. FEMS Microbiol. Rev. 29: 795811. [ Links ]
Dodd, J. C., C. L. Boddington, A. Rodríguez, M. C. GonzálezChávez, and I. Mansur. 2000. Hyphae of AMF from different genera in planta: form, function and detection with isozymes. Plant Soil 226: 135153. [ Links ]
Gadd, G. M. 1993. Interaction of fungi with toxic metals. New Phytol. 124: 2560. [ Links ]
GardeaTorresdey, J. L., I. CanoAguilera, H. Webb, and F. GutiérrezCorona. 1997. Enhanced copper absorption and morphological alterations of cell of copper stressed Mucor rouxii. Environ. Toxicol. Chem. 16: 435441. [ Links ]
Gerdemann, J. W., and T. H. Nicolson. 1963. Spores of my corrhizal Endogone species extracted from soil by wet sieving and decanting. Trans. Br. Mycol. Soc. 46: 235244. [ Links ]
GonzálezChávez, C., J. D'Haen, J. Vangrosveld, and J. C. Dodd. 2002. Copper sorption and accumulation by the extraradical mycelium of different Glomus spp. (arbuscular mycorrhizal fungi) isolated from the same polluted soil. Plant Soil 240: 287297. [ Links ]
GonzálezChávez, M. C., R. CarrilloGonzález, K. Nichols, and S. F. Wright. 2004a. The role of glomalin, a protein produced by arbuscular mycorrhizal fungi, in sequestering potentially toxic elements. Environ. Poll. 130: 317323. [ Links ]
GonzálezChávez, M. C., M. C. Gutiérrez Castorena, y S. F. Wright. 2004b. Hongos micorrízicos arbusculares en la agregación del suelo y su estabilidad. Terra Latinoamericana 22: 507514. [ Links ]
Levy, A., B. J. Chang, L. K. Abbott, J. Kuo, G. Garnett, and T. J. Inglis. 2003. Invasion of spores of arbuscular mycorrhizal fungus Gigaspora decipiens by Burkholderia sp. Appl. Environ. Microbiol. 69: 62506256. [ Links ]
Linderman, R. G. 2001. Effects of mycorrhizas on plant tolerance to diseases. In: Kapulnik, Y., and D. D. Douds (eds). Arbuscular Mycorrhizas: Physiology and Function. Kluwer Academic Publishers. Dordrecht, The Netherlands. pp: 345365. [ Links ]
Lindsay, W. L., and W. A. Norvel. 1978. Development of a DTPA test for zinc, iron, manganese and copper. Soil Sc. Soc. Am. J. 42: 421428. [ Links ]
Lynch, J. M. 1990. The Rhizosphere. John Wiley & Sons. New York, USA. 274 p. [ Links ]
MansfeldGiese, K., J. Larsen, and L. Bødker. 2002. Bacterial populations associated with mycelium of the arbuscular mycorrhizal fungus Glomus intraradices. FEMS Microbiol. Ecol. 41: 133140. [ Links ]
McDonald, R. M., and M. R. Chandler. 1981. Bacteriumlike organelles in the vesiculararbuscular mycorrhizal fungus Glomus caledonius. New Phytol. 89: 241246. [ Links ]
Minerdi, D., R. Fani, R. Gallo, A. Boariono, and P. Bonfante. 2001. Nitrogen fixation genes in an endosymbiotic Burkholderia strain. Appl. Environ. Microbiol. 67: 725732. [ Links ]
Morley, G. F., and G. M. Gadd. 1995. Sorption of toxic metals by fungi and clay minerals. Mycol. Res. 99: 14291438. [ Links ]
Paulitz, T. C., and R. G. Linderman. 1989. Interactions between fluorescent Pseudomonads and VA mycorrhizal fungi. New Phytol. 113: 3745. [ Links ]
Toljander, J. F., V. Artursson, R. L. Paul, J. K. Jansson, and R. D. Finlay. 2006. Attachment of different soil bacteria to arbuscular mycorrhizal fungal extraradical hyphae is determined by hyphal vitality and fungal species. FEMS Microbiol. Letters 254: 3440. [ Links ]
Trevors, J. T., and K. D. van Elsas. 1997. Microbial interactions in soil. In: van Elsas, J. D., J. T. Trevors, and E. M. H. Wellington (eds). Modern Soil Microbiology. Marcel Dekker. New York, USA. pp: 215243. [ Links ]
Vancûra, V., M. O. Orozco, O. Grauova, and Z. Prikryl. 1989. Properties of bacteria in the hyphosphere of a vesiculararbuscular mycorrhizal fungus. Agric. Ecosystem Environ. 29: 421427. [ Links ]
Vivas, A., I. Vörös, B. Biró, J. M. RuizLozano, and R. Azcón. 2003. Beneficial effects of indigenous Cd tolerant and Cdsensitive Glomus mosseae associated with a Cdadapted strain of Brevibacillus brevis in improving plant tolerance to Cd contamination. Appl. Soil Ecol. 24: 177186. [ Links ]
Vivas, A., B. Biró, J. M. RuizLozano, J. M. Barea, and R. Azcón. 2006. Two bacterial strains isolated from Znpolluted soil enhanced plant growth and mycorrhizal efficiency under Zn toxicity. Chemosphere 62: 15231533. [ Links ]
Walley, F. L., and J. J. Germida. 1996. Failure to decontaminate Glomus clarus NT4 spores is due to spore wallassociated bacteria. Mycorrhiza 6: 4349. [ Links ]
Wiehe, W., C. HechtBuchholz, and G. Hoflich. 1994. Electron microscopic investigations on root colonization of Lupinus albus and Pisum sativum with two associative plant growth promoting rhizobacteria, Pseudomonas fluorescens and Rhizobium leguminosarum bv. trifolii. Symbiosis 17: 1531. [ Links ]
Wright, S. F., and A. Upadhayaya. 1998. A survey of soils for aggregate stability and glomalin, a glycoprotein produced by hyphae of arbuscular mycorrhizal fungi. Plant Soil 198: 97107. [ Links ]