SEARCH

SEARCH BY CITATION

Keywords:

  • biological control;
  • ecological fitness;
  • Fusarium oxysporum Fo47;
  • microbial community;
  • risk assessment;
  • terminal restriction fragment length polymorphism (T-RFLP)

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Some nonpathogenic strains of Fusarium oxysporum can control Fusarium diseases responsible for severe damages in many crops. Success of biological control provided by protective strains requires their establishment in the soil. The strain Fo47 has proved its efficacy under experimental conditions, but its ecological fitness has not been carefully studied. In a series of microcosm studies, the ability of a benomyl-resistant mutant Fo47b10 to establish in two different soils was demonstrated. One year after its introduction at two concentrations in the disinfected soils, the biocontrol agent (BCA) established at similar high population densities, whereas in the nondisinfected soils it survived at lower densities, related to the initial concentrations at which it was introduced. The BCA behaved similarly in the two soils at temperatures ranging from 5 to 25 °C and soil water potentials between −0.01 and −1.5 MPa. In addition, terminal restriction fragment length polymorphism analysis of 16S and 18S rRNA showed that the structures of the bacterial and fungal communities evolved with time but were not significantly affected by the introduction of the BCA. Overall, the results showed that Fo47 is potentially a good BCA, able to establish in different soil environments without perturbing the investigated microbial structures.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The fungal species Fusarium oxysporum is ubiquitous in soils worldwide. Plant pathogenic strains of this species induce tracheomycosis (wilt) or crown and root rots in many crops of economical importance. Besides these pathogenic strains, the soil harbors a large diversity of nonpathogenic strains of F. oxysporum, some of them can protect the plant against infection by the pathogenic strains. The protective capacity of the nonpathogenic strains of F. oxysporum has been established many years ago, when studying the mechanisms of soil suppressiveness to Fusarium wilts (Rouxel et al., 1979). Suppressive soils are soils that naturally limit the incidence of Fusarium wilts. It was shown by Toussoun (1975) that these soils supported large populations of Fusarium spp. It is the case of the suppressive soils of Châteaurenard, described by Louvet et al. (1976), from which the protective strain Fo47 has been isolated. The mechanisms by which Fo47 and other nonpathogenic strains protect the plant are not fully understood. They are based on direct microbial antagonism, mostly competition for nutrients, and induced resistance in the plant (Fravel et al., 2003). Whatever the mechanisms involved, these protective strains offer a unique opportunity to develop biological methods to control Fusarium-induced diseases (Alabouvette et al., 2007).

Soil-borne Fusarium diseases are among the most difficult to control because it is not possible to directly apply fungicides to the roots and soil treatments are usually not effective. Until recently, growers could only eliminate the pathogens by biocidal treatments such as methyl bromide fumigation. This practice, which destroys both pathogenic and beneficial organisms, has been banned because it was dangerous to man and the environment. This led to a renewed interest in biological control and more specifically in the use of nonpathogenic strains of F. oxysporum to control Fusarium wilts. Two methods of application of protective strains of F. oxysporum have to be considered: seed coating or direct incorporation in the soil or, in the case of soilless cultures, in the growing substrate. Until now, the direct incorporation in the soil or potting mixture has been mostly considered. In that case, to be effective, the biocontrol agent (BCA) must establish in soil and in the rhizosphere of the plant. Therefore, it has to be compatible with the type, temperature and water content of the soil in which it is introduced. Indeed failure of biological control has often been attributed to a lack of compatibility between the BCA selected in vitro and the soil environmental conditions in which it has to be effective.

After its establishment in the soil, the BCA will interact not only with the pathogen to be controlled but also with all the biotic components of the soil. There is a fear that a successful BCA might displace the microbial balance of the soil and have some unexpected effects on the nontarget organisms. Therefore, there is a need to study the side effects of an introduced antagonist on the native microbial communities. In Europe, application of BCA is subjected to the Directive 91/414-ECC, which imposes this type of study. Until recently, there were no practical methods available to detect the impacts of an introduced BCA on the whole soil microbial community. Thanks to the development of molecular approaches based on extraction of total DNA from the soil, it is possible to overcome this limitation today. Several methods are available to assess microbial community structures by molecular fingerprinting. Among them, terminal restriction fragment length polymorphism (T-RFLP) has already been used to address the impact of cultural practices on the structure of bacterial and fungal communities (Edel-Hermann et al., 2004; Pérez-Piqueres et al., 2006).

The first objective of this study was to analyze the population dynamics of the biocontrol strain Fo47 in two soils of different physico-chemical properties and in a range of temperature and water potential more realistic with field applications than the usual conditions of laboratory experiments. The second objective of this study was to detect the side effects of the inoculation of Fo47 into these two soils on the microbial balance, using T-RFLP targeting both the bacterial and the fungal communities. Both the population dynamics of Fo47 and the side effects of the soil inoculation on the indigenous microbial communities were monitored for 1 year.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Fungal inoculum

The BCA F. oxysporum Fo47 was previously isolated from the natural Fusarium wilt-suppressive soil from Châteaurenard (France) (Alabouvette et al., 1993). The study was conducted with Fo47b10, a benomyl-resistant mutant of Fo47, which showed the same saprophytic ability in soil and the same biocontrol efficacy as the wild-type strain (Eparvier et al., 1991). The strain is stored by cryopreservation at −80 °C in the Collection ‘Microorganisms of Interest for Agriculture and Environment’: MIAE, INRA Dijon, France.

The fungus was grown in malt broth (10 g L−1 malt extract, Biokar Diagnostic, Beauvais, France) at 25 °C on a rotary shaker (250 r.p.m.). After 5 days of culture, the mycelial mats were removed by filtration through a 40-μm mesh. The conidia in the filtrate were washed three times by centrifugation at 10 000 g for 20 min and resuspended in sterile distilled water. The density of the conidial suspension was determined by direct observation with a haemocytometer.

Soils and treatments

Two soils with different physico-chemical properties were used. The soil from Epoisses was a silt loamy soil (9% sand, 50% silt, 41% clay, 2.6% organic matter, pH 8.2). The soil from Morvan was a sandy soil (68.5% sand, 16.5% silt, 15% clay and 4.4% organic matter, pH 4.3). The soils were passed through a 6-mm sieve. In some experiments, the soils used were disinfected. They were autoclaved three times on 3 consecutive days for 1 h at 110 °C and stored at room temperature for 1 week before inoculation.

Soil microcosms were prepared in glass jars of 1 L, containing 400 g of equivalent dried soil and protected by a non-tied lead. The 12 treatments performed corresponded to all combinations of the following conditions: (1) soil of Epoisses or soil of Morvan, (2) disinfected or nondisinfected and (3) noninoculated or inoculated with (4) 103 or 106 conidia of Fo47b10 g−1 dry soil. Soil inoculation was performed by adding the volume of conidial suspension needed to reach both the inoculum density and the soil moisture chosen. Each treatment was carried out in triplicate in three independent jars with independent inoculum. For the population dynamics study conducted at 25 °C for 1 year, the soil moisture was adjusted with sterile water to 18% and 15% of water content for the soil of Epoisses and the soil of Morvan, respectively. In this first experiment, the relationship between water potential and water content was not known. In order to obtain similar water potential in both soils, we supposed that less water was needed in the sandy soil of Morvan than in the soil of Epoisses. Glass jars were incubated at 25 °C for 12 months. At each sampling time, the soil moisture was checked and adjusted if needed by addition of sterile distilled water. After 7 days, 1 month, 2.5 months, 6 months and 12 months of incubation, 5-g soil samples were aseptically taken and immediately used to assess the population density of Fo47b10 as described below. In the case of the nondisinfected soils inoculated with 106 conidia g−1 dry soil, soil samples were also taken after 2 days, 1 month, 6 months and 12 months of incubation, passed through a 2-mm sieve and stored at −80 °C for further DNA extraction and molecular analysis.

Similar microcosms were prepared to assess the effect of temperature on the population dynamics of Fo47b10. The 24 treatments performed corresponded to all combinations of the following conditions: (1) soil of Epoisses or soil of Morvan, (2) disinfected or nondisinfected, (3) noninoculated or inoculated with (4) 103 or 106 conidia of Fo47b10 g−1 dry soil and (5) incubated at 5 or 15 °C. Each treatment was carried out in triplicate in three independent jars. The soil moisture was adjusted with sterile water to the same water potential of −0.1 MPa corresponding to 24% and 17% of water content for the soil of Epoisses and the soil of Morvan, respectively. After 15 days and 3 months of incubation, soil samples were taken to determine the population density of Fo47b10.

Finally, similar microcosms were also prepared to assess the effect of water potential on the population dynamics of Fo47b10. The 24 treatments performed corresponded to all combinations of the following conditions: (1) soil of Epoisses or soil of Morvan, (2) disinfected or nondisinfected soil, (3) noninoculated or inoculated with (4) 103 or 106 conidia of Fo47b10 g−1 dry soil and (5) adjusted to a water potential of −0.01 or −1.5 MPa. Each treatment was carried out in triplicate in three independent jars. The corresponding water contents for the soil of Epoisses and the soil of Morvan were 27% and 21% for a water potential of −0.01 MPa, and 20% and 9% for a water potential of −1.5 MPa, respectively. After 15 days and 3 months of incubation at 25 °C, soil samples were taken to determine the density of Fo47b10.

Assessment of population density of Fo47b10

The population density of Fo47b10 in the soil samples was measured using a plate count method. Five grams of soil were suspended into 45 mL of sterile distilled water. Flasks were placed on a three dimensions shaker (900 oscillations min−1) for 20 min. Tenfold serial dilutions were performed by mixing 5 mL of soil suspension with 45 mL of sterile distilled water in a flask, until the dilution 10−4. Five repetitions of each dilution were performed. Ten aliquots of 1 mL of the dilutions 10−2, 10−3 and 10−4 were each incorporated in a plate into molten malt medium (malt extract 10 g, agar 15 g, deionized water 1 L) supplemented postautoclaving with citric acid (250 mg L−1), antibacterial antibiotics (streptomycin 100 mg L−1 and chlortetracycline 50 mg L−1), pentachloronitrobenzene (0.94 g L−1) and benomyl (10 mg L−1). Plates were incubated in the dark at 25 °C. The CFU on the plates were counted twice after 3 and 7 days of incubation. Results were compared by anova of log-transformed CFU numbers and Tukey's tests using xlstat-pro version 7.1 (Addinsoft).

Assessment of bacterial and fungal community structures

Nucleic acids were extracted from 1 g of soil as described previously (Edel-Hermann et al., 2004). Briefly, a physical disruption (bead-beater) and a chemical extractant (sodium dodecyl sulfate) at 70 °C were used. The crude nucleic acid extracts were purified twice using a polyvinylpolypyrrolidone spin column to remove coextracted humic acids and once using a Geneclean® Turbot kit (Q-BIOgene). DNA was extracted in triplicate from independent soil samples. Purified DNA extracts were quantified by electrophoresis in agarose gels using dilutions of calf thymus DNA and stored at −20 °C.

Bacterial and fungal community structures were assessed by T-RFLP analysis of the 16S rRNA gene and the 18S rRNA gene, respectively. The bacterial 16S rRNA gene was amplified by PCR, using the primer 27F (AGAGTTTGATCCTGGCTCAG) (Edwards et al., 1989) labelled with the fluorescent dye D3 (Beckman Coulter, Fullerton, CA) and the primer 1392R (ACGGGCGGTGTGTACA) (Braker et al., 2001), and PCR products were digested using the restriction enzyme HaeIII. The fungal 18S rRNA gene was amplified by PCR, using the primer nu-SSU-0817-5′ (TTAGCATGGAATAATRRAATAGGA) labelled with the fluorescent dye D3 and the primer nu-SSU-1536-3′ (ATTGCAATGCYCTATCCCCA) (Borneman & Hartin, 2000), and PCR products were digested using the restriction enzyme MspI. The labelled and unlabelled primers were synthesized by Proligo (Paris, France) and MWG Biotech (Courtaboeuf, France), respectively. PCR amplifications and T-RFLP analyses were performed using the procedure described by Edel-Hermann et al. (2008). Fluorescently labelled terminal restriction fragments (TRF) were separated and detected using a capillary electrophoresis sequencer CEQ 2000XL (Beckman Coulter). The sizes of the TRF were determined by comparison with Size Standard-600 (Beckman Coulter). Bacterial and fungal community structures were characterized by the size and fluorescence intensity of TRF. For each PCR product, the T-RFLP analyses were performed in triplicate. Mean values for the intensity of peaks found in at least two of the three analyses were considered for further statistical analyses of microbial community structures. The comparison of the TRF sizes between samples was automated by assigning them to discrete categories using the program lis with an interval of 1.25 bp (Mougel et al., 2002). The communities, characterized by the sizes of the TRF and their intensity measured by the height of the peaks, were compared by principal component analysis using the ade-4 software (Thioulouse et al., 1997). This ordination method summarizes multivariate data to a few variables or dimensions and provides an arrangement of the communities in a two-dimensional diagram based on their scores on the first two dimensions. The significance of the resulting structures was checked using the Monte-Carlo tests with 1000 random permutations of the data.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Population dynamics of Fo47b10 in disinfected and nondisinfected soils

The strain Fo47b10 established well in the two disinfected soils (Fig. 1a). Whatever the initial inoculum concentration, the fungal population reached a density >3.8 × 106 CFU g−1 dry soil after 2.5 months of incubation and stabilized around 106 CFU g−1 dry soil. One year after the inoculation, there was no significant difference among the population densities in disinfected soils, neither between the two soils nor between the two initial inoculum concentrations.

image

Figure 1.  Population dynamics of the BCA Fusarium oxysporum Fo47b10 inoculated at 103 conidia g−1 dry soil and 106 conidia g−1 dry soil in the disinfected soils (a) and the nondisinfected soils (b) of Epoisses and Morvan. The soils were regulated at 18% and 15% of water content, respectively, and incubated at 25°C. Values with similar letters are not significantly different (P=0.05).

Download figure to PowerPoint

Fo47b10 was also able to maintain in nondisinfected soils (Fig. 1b). However, the population density tended to decrease with time. One year after the inoculation of 106 conidia g−1 dry soil, the density of Fo47b10 remained at 8.2 × 104 and 2.1 × 104 CFU g−1 in the soils of Epoisses and Morvan, respectively. Similarly, when inoculated at the lower concentration Fo47b10 established in the two soils but at a low density. Finally, for each of the two inoculum concentrations, and after 1 year of incubation, the density of Fo47b10 was significantly higher in the nondisinfected soil of Epoisses than in the nondisinfected soil of Morvan.

Effect of the temperature on the population dynamics of Fo47b10

After 15 days of incubation in the disinfected soils, and at the lower inoculum concentration, the population density was higher at 15 °C than at 5 °C; on the contrary, at the higher concentration of Fo47b10, the population density was not affected by the temperature whatever the soil (Fig. 2a). After 3 months of incubation, in the disinfected soils and inoculation at the lower concentration, the population density of Fo47b10 was significantly different in the two disinfected soils (Fig. 2b). In the soil of Epoisses, it established at a high level whatever the temperature of incubation; however, in the soil of Morvan, Fo47b10 was no more detected at 5 °C although it reached a high density at 15 °C. After inoculation at 1 × 106 conidia g−1 dry soil, the situation was less contrasted; the population density was always significantly higher in the soils incubated at 15 °C than in the soils incubated at 5 °C; the lowest density was registered in the soil of Morvan incubated at 5 °C.

image

Figure 2.  Effect of the temperature on the densities of Fusarium oxysporum Fo47b10 in the soils of Epoisses (E) and Morvan (M), disinfected (a, b) or nondisinfected (c, d). The soils were inoculated at 103 conidia g−1 dry soil (E+103, M+103) or 106 conidia g−1 dry soil (E+106, M+106), adjusted at a water potential of −0.1 MPa and incubated at 5 or 15°C. The densities of Fo47b10 were measured 15 days (a, c) and 3 months (b, d) after the inoculation. Values are means of CFU g−1 dry soil ±SE. For each sampling time, values with similar letters are not significantly different (P=0.05).

Download figure to PowerPoint

In the nondisinfected soils, after 15 days, the temperature of incubation did not significantly affect the population density of Fo47, which established at low and high levels following inoculation at the lower or greater inoculum concentration, respectively (Fig. 2c). Similarly, after 3 months of incubation, the population density of Fo47b10 was not affected by the temperature with the exception of the soil of Morvan inoculated at 106 conidia g−1 dry soil, where the density was significantly lower at 15 °C than at 5 °C (Fig. 2d).

Effect of the water potential on the population dynamics of Fo47b10

In both disinfected soils, after 15 days of incubation and despite slight differences, the population density of Fo47b10 established at a density >106 CFU g−1 dry soil, whatever the soil water potential and the initial inoculum level (Fig. 3a). After 3 months of incubation, the population density of Fo47b10 has only slightly evolved and there was not a clear effect of soil water potential on the population dynamics of Fo47b10 (Fig. 3b). When differences were significant, the population density was slightly lower at the lower water potential.

image

Figure 3.  Effect of the soil water potential on the densities of Fusarium oxysporum Fo47b10 in the soils of Epoisses (E) and Morvan (M), disinfected (a, b) or nondisinfected (c, d). The soils were inoculated at 103 conidia g−1 dry soil (E+103, M+103) or 106 conidia g−1 dry soil (E+106, M+106), adjusted at a water potential of −0.01 or −1.5 MPa and incubated at 25°C. The densities of Fo47b10 were measured 15 days (a, c) and 3 months (b, d) after the inoculation. Values are means of CFU g−1 dry soil±SE. For each sampling time, values with similar letters are not significantly different (P=0.05).

Download figure to PowerPoint

In the nondisinfected soils after 15 days of incubation, the only clear difference in the population density was in relation to the inoculation level, but there was no difference in relation to water potential (Fig. 3c). After 3 months of incubation in the nondisinfected soils, the population density of Fo47b10 was clearly different according to the initial level of inoculation; the effect of water potential was only significant after inoculation at the higher density (Fig. 3d). In that case the population density was higher at the lower water potential in both soils.

Impact of the soil inoculation with Fo47b10 on the microbial community structures

In both soils, 2 days after the introduction of the fungal strain, the structure of the bacterial communities diverged slightly between the soil inoculated with Fo47b10 and the noninoculated soil (Fig. 4). But this differentiation was not confirmed later on, until 1 year of monitoring. Permutation tests confirmed that the introduction of Fo47b10 did not significantly affect the bacterial community structures in both soils. On the contrary, the bacterial community structures evolved significantly with time in both soils, but always in a similar way in the inoculated and the control soil. After 1 year of incubation, the bacterial community structure was again similar to the initial community structure in the soil of Epoisses. On the contrary, this was not the case in the sandy acidic soil of Morvan.

image

Figure 4.  Bacterial community structures: principal component analysis (PCA) of the 16S T-RFLP data sets from the soils of Epoisses (a) and Morvan (b), noninoculated (○) or inoculated (•) with Fusarium oxysporum Fo47b10 at four sampling times: 2 days (2d, 2dF), 1 month (1 m, 1 mF), 6 months (6 m, 6 mF) and 12 months (12 m, 12 mF). The soils of Epoisses and Morvan were regulated at 18% and 15% of water content, respectively, and incubated at 25°C. Ellipses represent 90% confidence limits.

Download figure to PowerPoint

Similar to the bacterial communities, the fungal community structures tended to diverge slightly between the inoculated soil and the noninoculated soil immediately after the inoculation, in both soils (Fig. 5). The main difference between the two treatments was observed in the soil of Morvan 1 month after the inoculation. But from 6 months after the inoculation, the community structures of the noninoculated and inoculated soils were overlapping. Permutation tests also confirmed that the introduction of Fo47b10 did not significantly modify the fungal community structures in both soils. As for bacteria, the fungal community structures changed with time in both soils, but always in a similar way in the inoculated soil and the control. Finally, in both soils, after 1 year of incubation, the fungal community structure remained different from the initial community structure.

image

Figure 5.  Fungal community structures: principal component analysis (PCA) of the 18S T-RFLP data sets from the soils of Epoisses (a) and Morvan (b), noninoculated (○) or inoculated (•) with Fusarium oxysporum Fo47b10 at four sampling times: 2 days (2d, 2dF), 1 month (1 m, 1 mF), 6 months (6 m, 6 mF) and 12 months (12 m, 12 mF). The soils of Epoisses and Morvan were regulated at 18% and 15% of water content, respectively, and incubated at 25°C. Ellipses represent 90% confidence limits.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Success of biological control against soil-borne diseases requires establishment of the BCA in the soil in which it has to protect plant roots. Because the main mode of application of the protective strain Fo47b10, effective in controlling Fusarium wilts, will be soil application, the first aim of the present study was to assess the capacity of the BCA to establish in soil. Knowing that the soil physico-chemical properties influence the biological balance considerably and thus the capacity of a microorganism to become established, two soils presenting contrasted physico-chemical properties were chosen for this experiment.

A population dynamics study was conducted for a 1-year period at 25 °C, and at a water content of 18% and 15% in the soil of Epoisses and Morvan, respectively. Results showed that in both disinfected soils Fo47b10 grew well and reached the plateau at a high population density, whatever the initial inoculum concentration. These results are consistent with the theory that in a disinfected soil a microbial population will reach a density corresponding to the carrying capacity of the soil (Alabouvette & Steinberg, 1995). The population density was higher in the disinfected soil of Morvan, which has a higher level of organic matter than the soil of Epoisses. Under axenic conditions, the soil organic matter supported the saprophytic growth of Fo47b10, which reached a higher population density in the soil with a higher content of organic matter.

In the nondisinfected soils, Fo47b10 established at levels relative to the initial inoculum concentration. Indeed, after the introduction at the higher concentration (106 conidia g−1 dry soil), the population density of Fo47b10 stabilized in both soils at a density >2 × 104 CFU g−1 dry soil. After inoculation at the lower concentration (103 conidia g−1 dry soil), the BCA behaved differently in the two soils. Indeed it stabilized at a higher density in the soil of Epoisses than in the soil of Morvan in which it was still detectable. Because in the disinfected soils, the population of Fo47b10 stabilized at a higher level in the soil of Morvan than in the soil of Epoisses, the results obtained in the raw soils showed that the natural microbiota of the soil of Morvan is more suppressive towards the establishment of Fo47b10 than that of the soil of Epoisses. However, even in this less favourable environment, Fo47b10 became established after 1 year of incubation. Whether the established population of Fo47b10 is high enough to control the disease remains to be demonstrated.

This first experiment was performed at 25 °C, which is not a common soil temperature throughout the year in France. Therefore, the effect of more realistic soil temperatures on the establishment of the biocontrol strain in soil was studied. The experiment on the effect of temperature on the saprophytic behaviour of Fo47b10 was performed at a water potential of −0.1 MPa. Results showed that, in the disinfected soils, with a single exception (soil of Morvan inoculated at the lower concentration), the behaviour of Fo47b10 was quite similar at 5 and 15 °C and comparable to the behaviour described above when the microcosms were incubated at 25 °C. In the disinfected soil of Morvan inoculated at the lower concentration and incubated at 5 °C, the population of Fo47b10 decreased below the detectable level. This is difficult to explain in the absence of any specific observation related, for example, to an accidental contamination in these microcosms. The most interesting results were obtained in nondisinfected soils. They showed that, whatever the temperature, Fo47b10 established in both soils at levels lower but not too different from the concentrations at which it was introduced.

Water potential is another parameter that controls the establishment of microorganisms in soil. This parameter is not taken into consideration in laboratory experiments very often. Because the temperature experiment was conducted in soils regulated at −0.1 MPa, we chose to study the effect of two contrasted levels of water potential, −0.01 and −1.5 MPa, the latter value being considered as the wilting point for most of the plants. Globally, results showed that the fungus established well in both disinfected soils at both water potentials. However, population densities were significantly higher in the soils with a high moisture content (−0.01 MPa). In the nondisinfected soils, the population densities clearly decreased, and after 3 months of incubation, there were no significant differences according to the water potential after introduction of the lower inoculum concentration. In contrast, after inoculation at 106 conidia g−1 dry soil, the population densities were higher in the soils with a low water content (−1.5 MPa). These results showed that depending on the presence or absence of the indigenous microbial communities, the soil water potential affected the population kinetics of Fo47b10 differently. Indeed the development of Fo47b10 was favoured by a high water content in the disinfected soil but not in the drier soils. These results indicating that the water potential controls the competitive interactions between the BCA and the indigenous microbial communities, are in agreement with data obtained by Toyota et al. (1996), who showed that a pathogenic strain of F. oxysporum grew better in disinfected soil aggregates than in aggregates recolonized by ‘challenging’ organisms.

Altogether, these results showed that Fo47, a strain isolated from a Fusarium wilt suppressive soil, is well adapted to saprophytic growth and survival in the two different soil environments. Because the characteristics of the soil of Châteaurenard, from which Fo47 originated, is close to that of the soil of Epoisses, it may not be surprising to observe a strong capacity of Fo47 to establish in this soil. The soil of Morvan was chosen because its characteristics are quite different and we might have expected a different behaviour of Fo47 in this sandy acidic soil. This ability of a strain of F. oxysporum to establish in different soil environments fits with the results of a survey carried out on 42 sites showing that F. oxysporum is one of the Fusarium species that are less influenced by climatic conditions (Sangalang et al., 1995).

Besides the fitness of a BCA to various environmental conditions, another aspect that must be taken into consideration before application of a BCA is the potential impacts of the inoculated strain on the indigenous soil microbial communities. To address this question, we used a molecular fingerprinting method, enabling one to monitor the bacterial and fungal community structures. T-RFLP analysis has already proved its usefulness in revealing the impact of different agricultural practices and human activities on microbial communities (Pérez-Piqueres et al., 2006; Bordenave et al., 2007; Edel-Hermann et al., 2008; Ulrich et al., 2008). Thus, it was used to follow the bacterial and fungal community dynamics in soils over 1 year following the introduction of the BCA. Two different soils were used to demonstrate reproducibility. Results showed that soil inoculation with Fo47b10 did not significantly affect the bacterial and fungal fingerprints, either in the soil of Epoisses or in the soil of Morvan. Only slight and transient modifications of the community structures were observed immediately after the inoculation of the BCA. Similar minor and temporary effects on soil microbial communities have also been reported for bacterial BCA (Timms-Wilson et al., 2004; Winding et al., 2004). On the contrary, the microbial communities evolved with time in both soils, although the conditions of temperature and moisture of the soil microcosms were kept constant. At the beginning of the experiment, changes in the soil microbial communities might be explained by the perturbation induced by the addition of liquid, either water in the control or conidial suspension in the inoculated soil, and following shaking of the soil microcosms. After 1 year of incubation, the bacterial community structure of the soil of Epoisses became again similar to that observed at the beginning of the experiment. On the contrary, neither the bacterial community structure in the soil of Morvan nor the fungal community structure in both soils returned to the initial stage. These changes might be related to the decrease in nutriments in soil microcosms after several months of incubation, favouring the development of the microbial components better adapted to the modified soil conditions. After 1 year of incubation, the bacterial and fungal community structures were similar in the control and inoculated soil of Epoisses or Morvan. These results indicate that the indigenous soil microbial communities were not affected by the introduction of the BCA in these two soils. This conclusion is important in regard to the registration procedure and environmental protection. Indeed, it would be dangerous to introduce a BCA that would durably modify the microbial balance of the soil and thus could modify the soil quality. Our results clearly show that a biocontrol fungal strain such as this nonpathogenic strain of F. oxysporum can establish durably in two soils of different physico-chemical properties without modifying the microbial balance.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work was supported by the EU project 2E-BCAs in Crops: Enhancement and Exploitation of Soil Biocontrol Agents for Bio-Constraint Management in Crops; specific targeted research project FOOD-CT-2003-001687 in the 6th Framework Programme.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Alabouvette C & Steinberg C (1995) Suppressiveness of soils to invading microorganisms. Biological Control Benefits and Risks (HokkanenH & LynchJM, eds), pp. 312. Cambridge University Press, Cambridge, Saariselka, Finland.
  • Alabouvette C, Lemanceau P & Steinberg C (1993) Recent advances in biological control of Fusarium wilts. Pestic Sci 37: 365373.
  • Alabouvette C, Olivain C, L'Haridon F, Aimé S & Steinberg C (2007) Using strains of Fusarium oxysporum to control Fusarium wilts: dream or reality. Novel Biotechnologies for Biocontrol Agents Enhancement and Management (VurroM & GresselJ, eds), pp. 157177. Springer, Dordrecht, The Netherlands.
  • Bordenave S, Goni-Urriza MS, Caumette P & Duran R (2007) Effects of heavy fuel oil on the bacterial community structure of a pristine microbial mat. Appl Environ Microb 73: 60896097.
  • Borneman J & Hartin RJ (2000) PCR primers that amplify fungal rRNA genes from environmental samples. Appl Environ Microb 66: 43564360.
  • Braker G, Ayala-Del-Rio HL, Devol AH, Fesefeldt A & Tiedje JM (2001) Community structure of denitrifiers, Bacteria, and Archaea along redox gradients in Pacific Northwest marine sediments by terminal restriction fragment length polymorphism analysis of amplified nitrite reductase (nirS) and 16S rRNA genes. Appl Environ Microb 67: 18931901.
  • Edel-Hermann V, Dreumont C, Pérez-Piquerez A & Steinberg C (2004) Terminal restriction fragment length polymorphism analysis of ribosomal RNA genes to assess changes in fungal community structure in soils. FEMS Microbiol Ecol 47: 397404.
  • Edel-Hermann V, Gautheron N, Alabouvette C & Steinberg C (2008) Fingerprinting methods to approach multitrophic interactions among microflora and microfauna communities in soil. Biol Fert Soils 44: 975984.
  • Edwards U, Rogall T, Blöcker H, Emde M & Böttger EC (1989) Isolation and direct complete nucleotide determination of entire genes. Characterization of a gene coding for 16S ribosomal DNA. Nucleic Acids Res 17: 78437853.
  • Eparvier A, Lemanceau P & Alabouvette C (1991) Population dynamics of non-pathogenic Fusarium and fluorescent Pseudomonas strains in rockwool, a substratum for soilless culture. FEMS Microbiol Ecol 86: 177184.
  • Fravel D, Olivain C & Alabouvette C (2003) Fusarium oxysporum and its biocontrol. New Phytol 157: 493502.
  • Louvet J, Rouxel F & Alabouvette C (1976) Recherches sur la résistance des sols aux maladies. I – Mise en évidence de la nature microbiologique de la résistance d'un sol au développement de la fusariose vasculaire du melon. Ann Phytopathol 8: 425436.
  • Mougel C, Thioulouse J, Perriere G & Nesme X (2002) A mathematical method for determining genome divergence and species delineation using AFLP. Int J Syst Evol Micr 52: 573586.
  • Pérez-Piqueres A, Edel-Hermann V, Alabouvette C & Steinberg C (2006) Response of soil microbial communities to compost amendments. Soil Biol Biochem 38: 460470.
  • Rouxel F, Alabouvette C & Louvet J (1979) Recherches sur la résistance des sols aux maladies. IV – Mise en évidence du rôle des Fusarium autochtones dans la résistance d'un sol au développement de la fusariose vasculaire du melon. Ann Phytopathol 11: 199207.
  • Sangalang AE, Burgess LW, Backhouse D, Duff J & Wurst M (1995) Mycogeography of Fusarium species in soils from tropical, arid and Mediterranean regions of Australia. Mycol Res 99: 523528.
  • Thioulouse J, Chessel D, Dolédec S & Olivier JM (1997) ADE-4: a multivariate analysis and graphical display software. Stat Comput 7: 7583.
  • Timms-Wilson T, Kilshaw K & Bailey MJ (2004) Risk assessment for engineered bacteria used in biocontrol of fungal disease in agricultural crops. Plant Soil 266: 5767.
  • Toussoun TA (1975) Fusarium-suppressive soils. Biology and Control of Soil-Borne Plant Pathogens (BruehlGW, ed), pp. 145151. American Phytopathological Society, St Paul.
  • Toyota K, Young IM & Ritz K (1996) Effects of soil matric potential and bulk density on the growth of Fusarium oxysporum f. sp. raphani. Soil Biol Biochem 28: 11391145.
  • Ulrich A, Klimke G & Wirth S (2008) Diversity and activity of cellulose-decomposing bacteria, isolated from a sandy and a loamy soil after long-term manure application. Microb Ecol 55: 512522.
  • Winding A, Binnerup SJ & Pritchard H (2004) Non-target effects of bacterial biological control agents suppressing root pathogenic fungi. FEMS Microbiol Ecol 47: 129141.