Screening procedures for selecting rhizobacteria with biocontrol effects upon Fusarium verticillioides growth and fumonisin B1 production

Research in Microbiology 155 (2004) 747–754 www.elsevier.com/locate/resmic

Screening procedures for selecting rhizobacteria with biocontrol effects upon Fusarium verticillioides growth and fumonisin B1 production Lilia Cavaglieri ∗ , A. Passone, Miriam Etcheverry Departamento de Microbiología e Inmunología, Facultad de Ciencias Exactas, Físico-Químicas y Naturales, Universidad Nacional de Río Cuarto, Ruta Nacional 36 Km, 601, 5800 Río Cuarto, Córdoba, Argentina Received 4 February 2004; accepted 3 June 2004 Available online 6 August 2004

Abstract Screening is a critical step in the discovery of microbial agents that can exert biological control of Fusarium verticillioides at the root level. The objectives of this research were to determine the utility of a niche overlap index to realise the first screening of maize rhizobacterial isolates during different water activities. Studies were conducted to evaluate various methods for second screening with different modes of action. The antifungal activity of bacterial isolates through antibiosis assay was checked and the influence of different isolates on Fusarium verticilliodes growth and fumonisin B1 was studied. Eleven competitive rhizobacterial isolates (Arthrobacter globiformis RC1, Azotobacter armeniacus RC2, A. armeniacus RC3, A. globiformis RC4, A. globiformis RC5, A. armeniacus RC6, Pseudomonas solanacearum RC7, Bacillus subtilis RC8, B. subtilis RC9, P. solanacearum RC10, B. subtilis RC11) were selected for the studies which followed. All bacteria were able to utilise the widest range of carbon sources and showed the highest niche overlap indices at the water activities tested. All bacterial antagonists reduced fumonisin B1 production at all levels tested. Isolates belonging to Pseudomonas and Bacillus genera significantly inhibited fumonisin B1 production, which ranged between 70 and 100%. Also, A. armeniacus RC2 caused important fumonisin B1 reduction. The results of the present work suggest that A. armeniacus RC2, A. armeniacus RC3, B. subtilis RC8, B. subtilis RC9, B. subtilis RC11, P. solanacearum RC7, and P. solanacearum RC10 could have practical value in the control of F. verticillioides root colonisation. This paper is part of an on-going study to determine their application at the field level.  2004 Elsevier SAS. All rights reserved. Keywords: Fusarium verticillioides; Growth; Fumonisin B1 ; Rhizobacteria; Biocontrol

1. Introduction Maize is the host of a number of fungi that can produce mycotoxins. Among these fungi, Fusarium verticillioides produces toxins associated with harmful effects on animal and human health [30,40]. Fumonisins, mainly fumonisin B1 (FB1 ), have been associated with several mycotoxicoses in animals, such as leukoencephalomalacia in horses [15], pulmonary oedema syndrome (PES) in pigs [42] and liver cancer in rats [15]. Fumonisin has been detected in corn and corn-based foods marketed in several countries [29]. F. verticillioides is the most prevalent Fusarium in freshly harvested maize in Argentina and the occurrence of very high levels of fumonisins in some samples was correlated * Corresponding author.

E-mail address: [email protected] (L. Cavaglieri). 0923-2508/$ – see front matter  2004 Elsevier SAS. All rights reserved. doi:10.1016/j.resmic.2004.06.001

with the presence of strains producing abundant fumonisin in the laboratory [7,16]. Biological control has potential for the management of this disease. Over the past one hundred years, research has repeatedly demonstrated that diverse microorganisms can act as natural antagonists of plant pathogens [1,5]. Pseudomonas, Burkholderia and Bacillus spp., rhizosphere colonisers, have been used to reduce disease caused by a variety of soil-borne plant pathogens [13,26,35] including Fusarium spp. [22,28]. Although many different biocontrol strains have shown some degree of control of Fusarium diseases, highly effective strains that are active against F. verticillioides growth and fumonisin production and that have potential for effective implementation in agriculture have not been identified. Furthermore, the influence of rhizobacteria on F. verticillioides strains over a range of environmental conditions, and their impact on fumonisin production, have not been studied.

748

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

Screening is a critical step in the development of biocontrol agents. The success of all subsequent stages depends on the ability of a screening procedure to identify an appropriate candidate. Microorganisms isolated from the root or rhizosphere of a specific crop may be better adapted to that crop and may provide better control of diseases than organisms originally isolated from other plant species [19]. Furthermore, many environmental variables can result in multiple interactions among microorganisms and their environment, several of which might contribute to effective biological control [11,43]. Intensive screens have yielded numerous candidate organisms for commercial development. Some of the microbial taxa that have been successfully commercialised as biopesticides include bacteria belonging to the genera Agrobacterium, Bacillus, Pseudomonas and Streptomyces [8,22]. The screening of locally adapted strains has yielded improved biocontrol in some cases. However, this finding does not provide the mechanism of action involved. The ability of F. verticillioides strains to grow in the rhizosphere and at the root surface may be an important prelude to the infection of host plants and may be related to their competitive ability to use carbon sources [47]. The first few days of seed germination are a critical time period for certain plant pathogens. Many Fusarium species germinate within a few hours of planting in response to spermosphere exudates [38]. Maize releases substantially more carbohydrates and amino acids than other seeds [2]. Wilson and Lindow [49] suggested that competition for limiting environmental resources, not antibiosis, was the primary mechanism of antagonism in the phyllosphere. They estimated the ecological similarity of epiphytic bacterial strains with niche overlap indices (NOIs). Ecological similarity was inversely correlated with the level of coexistence [23]. Other researchers used NOIs to study the competition existing between toxigenic fungi and other colonisers in vitro at different water activities. To become established, F. verticillioides needs to compete with other colonisers, including a range of fungal and bacterial antagonists. Previous studies have detailed the important influence of water activity (aW ) on the ability of Fusarium species to germinate, grow and produce fumonisins in vitro and on maize grains [3,34]. The selection of appropriate rhizobacteria able to grow under a wide range of water activities is a pre-condition for obtaining proper biological control. These bacteria should be capable of preemptively excluding F. verticillioides fumonisin B1 producers under field conditions. Other screening methods are needed to identify biocontrol agents with different modes of action such as antibiosis or some forms of competition. The objectives of this research were to: (i) determine the utility of the NOI to perform the first screening of maize rhizobacterial isolates during different water activities; (ii) check the antifungal activity of bacterial isolates through antibiosis assay; (iii) determine the impact of bacterial isolates on fungal growth and fumonisin B1 accumulation.

2. Materials and methods 2.1. Microorganisms Samples from a commercial maize field (Morgan M401, Argentina SA, Buenos Aires, Argentina) were collected during the seedling stage (15 days) and at harvest time (120 days). The field soil was loamy sand. Twenty plants were randomly chosen and removed. Plants were lifted together with adherent soil into plastic bags, transported to the laboratory within 12 h and analysed the same day. Roots were washed and dried between sheets of tissue, weighed and surface-sterilised by gently shaking in 70% ethanol (1 min), 20% household bleach (5 min) and thiosulphate Ringer solution (5 min) (Oxoid Ltd., London, UK) to quantify endorhizosphere bacterial populations. Samples were macerated in 90 ml of phosphate-buffered saline (PBS) with a mortar pestle. Threefold dilutions of the homogenates were plated on tryptic soy broth plus 2% agar (TSBA). Petri plates were incubated at 28 ◦ C for 24–48 h. Total counts and counts per colony type were made from each medium. One colony per colony type was isolated and purified on TSBA. F. verticillioides strains were isolated from the same roots as used for bacterial recovery. Threefold dilutions of the homogenates were plated on Nash–Snyder agar. Plates were incubated at 28 ◦ C for 7 days. Total counts and counts per colony type were made from each medium. One colony per colony type was isolated and purified on carnation leaf agar (CLA) [39]. Bacterial identification was performed according to Bergey’s Manual of Systematic Bacteriology [18,20,46]. Fusarium isolates were classified according to Nelson et al. [39] and the toxigenic ability of Fusarium section Liseola strains was determined according to Warfield and Gilchrist [48]. Thirteen F. verticillioides fumonisin B1 producers were selected [4]. 2.2. Determination of NOIS between bacterial antagonists and Fusarium strains of Liseola isolated from the endorhizosphere of corn Seventy-four bacterial antagonists belonging to the genera Arthrobacter, Azotobacter, Pseudomonas, Bacillus, Lactobacillus, Micrococcus, Listeria, Agromyces and 13 F. verticillioides isolates were tested for their ability to utilise 17 different compounds present in maize as sole carbon sources (dextrin, D-fructose; D-galactose; α-D-glucose; D-raffinose; D -melobiose; sucrose; L -aspartic acid; L -glutamic acid; L -histidine; L -phenylalanine; L -leucine; L -proline; L -threonine; L-alanine; L-serine; L-arginine) [2,31]. The water activity (aW ) of 2% water agar was modified with glycerol to 0.982, 0.955 and 0.937 aW according to Dallyn and Fox [10]. The carbon sources were incorporated individually into glycerolmodified water agar at a concentration of 10 mM [21]. The number of carbon sources represents the niche size. An aliquot of 0.1 ml of each bacterial antagonist growing in

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

nutrient broth to 1.5–1.8 optical density 620 nm was inoculated separately onto one plate of glycerol-modified agar containing each carbon source by water activity type. On the other hand each toxigenic strain of F. verticillioides obtained from monosporic culture on CLA [39] was inoculated onto each carbon source medium at different water activities. All plates were incubated in polyethylene bags at 25 ◦ C for up to 14 days before being scored for the presence or absence of growth. All experiments were carried out in triplicate. The NOI was defined as the number of carbon sources in common by both strains (bacterial antagonist–F. verticillioides) as a proportion of the total number of carbon sources utilized by F. verticillioides.

749

were determined by using three replicates for each test. The radial growth rate (mm h−1 ) was subsequently calculated by linear regression of linear growth. The time at which the line intercepted the x-axis was used to calculate the lag phase in relation to bacterial antagonists and water activity. The experiments were carried out three times for single and paired cultures. 2.5. Determination of inhibition of fumonisin B1 production

The NOIs were estimated for each strain in a pair. NOI values of >0.9 represent competence between species while scores of <0.9 represent occupation of separate niches [49,50].

Toxins were extracted with acetonitrile-water (1:1, v/v) by shaking the culture media and micelia from co-inoculated and control cultures with the solvent for 30 min on an orbital shaker (150 rev min−1 ) and then filtering the extracts through filter paper (no. 4 Whatman) [14]. The extracts were frozen and stored at −20 ◦ C until analysed. An aliquot of the extract (1000 µl) was taken and diluted with acetonitrile-water as necessary for high performance liquid chromatography (HPLC analysis). The fumonisin content of the orthophthaldehyde derivative (OPA) was evaluated by HPLC as described by Shephard et al. [45] and modified by Doko et al. [12].

2.3. Assessment of antibiosis

2.6. Statistical analysis

Culture medium was 2% maize meal extract agar (MMEA). The water activity (aW ) of the basic medium was adjusted with glycerol to 0.982, 0.955 and 0.937. Thirteen F. verticillioides strains were inoculated with 10 µl of 109 spores ml−1 in the center of each plate separately. Ten µl of the different bacterial antagonist suspensions (1011 CFU ml−1 ) were inoculated at 2.5 mm distance from the edge of the F. verticillioides droplet. Treatments were incubated at 25 ◦ C for up to 10 days in polyethylene bags. The antibiosis zone was measured to the edge of the growth inhibition halo of F. verticillioides (mm). Treatments were compared with the diameter of each F. verticillioides control. This assessment was carried out with at least three separate replicates per treatment.

Antibiosis and growth rate data analyses were performed by analysis of variance. Means of treatments were compared using Duncan’s multiple range test. The fumonisin B1 data obtained were transformed using a logarithmical function (log(x + 1)) before applying the analysis of variance. Then, the Scheffé test was used to determine the significant differences between the control and co-inoculated cultures [41].

NOI = (n0 of C-sources in common between bacterial antagonist–F. verticillioides) × (total n0 of C-sources utilized by F. verticillioides)−1 .

2.4. Rhizobacteria antifungal effect on F. verticillioides growth under different aW conditions in vitro Before cooling, MMEA (10 ml) was adjusted to different water activities (0.982, 0.955 and 0.937 aW ), with 100 µl (1011 cells ml−1 ) of each antagonist and poured into the Petri dishes (60 × 10 mm). Each strain of F. verticillioides was inoculated in the center of the plate with a spore from a monosporic culture. Cultures were incubated at 25 ◦ C for 20 days in polyethylene bags. The diameter of F. verticillioides colonies was measured daily with a ruler. The growing radius of the cultures containing both microorganisms was compared with the control cultures. For each colony, two radii, measured at right angles to one another, were averaged to find the mean radius for that colony. All colony radii

3. Results 3.1. NOIs Table 1 shows the levels of competence and coexistence between the 13 F. verticillioides isolates that were paired with 74 bacterial antagonists at different water activities and assayed for the ability to utilise the 17 carbon sources tested. At the lowest aW levels examined (0.937aW), 58% of the antagonistic isolates were able to utilise all of the carbon sources assayed (NOI > 0.9) whereas only 20% of the antagonistic isolates were able to utilise the same carbon sources at 0.955aW. Seventy five percent of the bacteria were able to utilise the majority of the carbon sources assayed at the 0.982aW (NOI > 0.9). A change in aW altered the number of carbon compounds utilised by F. verticillioides isolates paired with bacterial antagonists examined. According to the NOIs at 0.982aW and 0.937aW a high percentage of isolates had the ability to compete for the same carbon sources, in disagreement with 0.955aW where there was a low competence level indicative of occupation

750

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

Table 1 Percentage of rhizobacterial isolates that show ecological similarity (competence) or co-existence paired with 13 F. verticilliodes isolates Water activity

Competence1

Co-existence2

Competent isolates %

Genera

0.982

75

25

0.955

20

80

0.937

58

42

35.7 16.3 16 8.9 7.1 7.1 5.3 27.3 27.3 20 20 5.4 32 27.6 12.7 10 8.5 6.3 2.9

Arthrobacter Azotobacter Listeria Pseudomonas Micrococcus Bacillus Agromyces Arthrobacter Azotobacter Pseudomonas Bacillus Listeria Arthrobacter Azotobacter Pseudomonas Listeria Micrococcus Bacillus Lactobacillus

1 % paired strains with niche overlap index > 0.9. 2 % paired strains with niche overlap index < 0.9.

Table 2 NOI for pre-selected rhizobacterial isolates paired with 13 F. verticillioides isolates derived from carbon source utilization data at different water activities (aW ) NOI

3.2. Influence of interspecific microbial interactions on antifungal activity, growth rate and lag phase of F. verticillioides All the bacterial antagonists produced a significant decrease in the growth rate and an increase in the lag phase (P < 0.001). B. subtilis RC8 and B. subtilis RC9 isolates exerted the strongest influence on growth rate reduction, whereas the lag phase was mainly increased by B. subtilis RC8, B. subtilis RC9 and B. subtilis RC11 isolates (Figs. 1a and 1b). All antagonistic bacteria were able to significantly reduce F. verticillioides growth (antifungal activity). A. globiformis RC4 and B. subtilis RC8 isolates showed the most effective fungal growth inhibition (P < 0.001) (Fig. 1c). 3.3. Influence of interspecific microbial interactions on fumonisin B1 production All bacterial antagonists reduced fumonisin B1 production at all aW levels assayed. The isolates belonging to Pseudomonas and Bacillus genera significantly inhibited the fumonisin B1 production which ranged between 70 and 100%. Also, A. armeniacus RC2 exerted substantial FB1 reduction. At 0.955aW Arthrobacter and Azotobacter isolates showed a higher percentage of inhibition than 0.982aW.

Bacterial isolates

4. Discussion

aW 0.982

0.955

0.937

1 1 1 1 1 0.94 0.94 0.94 1 1 1

1 0.94 0.94 1 0.88 0.94 0.94 1 0.94 1 1

1 1 1 0.94 0.94 0.94 0.94 0.94 1 1 1

Arthrobacter globiformis RC1 Azotobacter armeniacus RC2 Azotobacter armeniacus RC3 Arthrobacter globiformis RC4 Arthrobacter globiformis RC5 Azotobacter armeniacus RC6 Pseudomomas solanacearum RC7 Bacillus subtilis RC8 Bacillus subtilis RC9 Pseudomonas solanacearum RC10 Bacillus subtilis RC11

of different niches. Arthrobacter strains were the most competent at all water activities assayed while most of the other strains tested (Azotobacter armeniacus, Pseudomonas solanacearum and Bacillus subtilis spp.) were competent especially at 0.955aW and at 0.937aW. Eleven antagonistic rhizobacterial competitive isolates (A. globiformis RC1, A. armeniacus RC2, Azotobacter armeniacus RC3, A. globiformis RC4, A. globiformis RC5, A. armeniacus RC6, P. solanacearum RC7, B. subtilis RC8, B. subtilis RC9, and P. solanacearum RC10; B. subtilis RC11) were selected for the next studies. All of them were able to utilise the widest range of carbon sources. They belonged to the genera with the highest NOIs at the water availabilities tested (Table 2).

Results presented here give a general impression of the manner in which different screening techniques may select a rhizobacterial antagonist on F. verticillioides, and they enable the establishment of their potential competitiveness. Eleven isolates from a wide spectrum of potential antagonists through NOIs were selected. Ecological similarity and coexistence between microbial species have been examined in relation to the similarity of biological control agents for control of pathogens on plant surfaces [49,50]. Marin et al. [33] studied the values of NOIs among F. moniliforme, F. proliferatum and species of Aspergillus, Penicillium and F. graminearum, demonstrating that the sources of carbon used by each species interact, and those used in common were greatly influenced by water activity and temperature. Lee and Magan [23] studied the utilisation of C-sources found in maize by A. ochraceus and other spoilage fungi. In this sense, La Penna et al. [21] used the niche overlap indices to establish the grade of ecological similarity between yeast antagonists and Aspergillus section Flavi strains in water agar supplemented with different C-components present in maize grain. Maize seed and root exudates have been extensively studied. These exudates include carbohydrates, amino acids and other organic acids [9,34]. The first few days of seed germination are a critical period for certain plant pathogens. Many Fusarium species

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

751

Table 3 Fumonisin B1 production by 13 F. verticillioides isolates paired with bacterial antagonists in vitro Control (F)a , interaction (F-B)b F F-B1 F-B2 F-B3 F-B4 F-B5 F-B6 F-B7 F-B8 F-B9 F-B10 F-B11

Fumonisin B1 production (ng g−1 )

Fumonisin B1 reduction percentage (%)

0.982aW

0.955aW

0.982aW

0.955aW

2411.85 ± 670.21 a 1929.48 ± 337.07 b 964.74 ± 294.77 cd 1205.92 ± 191.93 c 2074.19 ± 484.14 ab 1929.48 ± 346.94 b 2050.07 ± 354.39 ab NDc 241.20 ± 118.88 e 723.50 ± 225.67 de ND 242.02 ± 129.45 e

5925.15 ± 523.41 4206.86 ± 374.66 3258.83 ± 331.91 2073.80 ± 142.99 5095.63 ± 456.88 4740.1 ± 441.02 5036.37 ± 246.73 ND ND ND ND ND

0 20 60 50 14 20 15 100 90 70 100 90

0 29 45 65 48 40 47 100 100 100 100 100

Letters in common are not significantly different according to Scheffe test (P < 0.001). a F: mean values of fumonisin B production by 13 F. verticillioides isolates. 1 b F-B: mean values of fumonisin B production by 13 F. verticillioides isolates paired with bacterial antagonists (B : A. globiformis RC1; B : A. armeniacus 1 1 2 RC2; B3 : A. armeniacus RC3; B4 : A. globiformis RC3; B5 : A. globiformis RC5; B6 : A. globiformis RC6; B7 : P. solanacearum RC7; B8 : B. subtilis RC8; B9 : B. subtilis RC9; B10 : P. solanacearum RC10; B11 : B. subtilis RC11). c ND: not detected.

germinate within a few hours of planting in response to spermosphere exudate [38]. It is assumed that many biocontrol agents first colonise the spermosphere and rhizosphere to protect the plant. If the pathogen germinates and colonises the root within a few days of planting, the rhizobacteria can be metabolically active during that time period. The biocontrol agent must compete for C-sources in the rhizosphere. This fact may be critical for colonisation and suppression of plant pathogens. This study has shown that the antagonistic rhizobacterial isolates selected according to niche overlap indices are able to compete and dominate fumonisin B1 producers over a wide range of water availability conditions. This procedure could be useful for identifying agents that may be effective in the preemptive exclusion of a target plant pathogen. A high NOI with respect to the pathogen indicates a high degree of ecological similarity; therefore, the potential biocontrol agent becomes effective in the exclusion of F. verticillioides, by usurping a high percentage of the resources that would otherwise be available to the fungal pathogen. Once the antagonist has colonised a favorable niche it must be able to maintain its position on the root system, either by the production of siderophores [13,25] or by the production of antibiotics that suppress growth of competing microorganisms [6,36,44]. In this study 11 preselected high NOI rhizobacterial antagonists appear to be useful potential biocontrol agents, as indicated by observing zones of fungal inhibition in petri plates. All of them were able to significantly inhibit the growth of 13 F. verticillioides strains at different aW . The present work reveals how different antagonists at different aW conditions can affect the growth of F. verticillioides isolates using direct measures of growth (colony radius) with lag phase and growth rate as the estimated pa-

rameters. Magan and Lacey [27] suggested that lag time for growth, rate of germination and elongation rate were good criteria for comparing the capabilities of different fungal spores for colonising surfaces under various aW conditions. There has been great interest in the potential use of a range of ecological traits such as lag phases and growth rates in the development of hurdle technology approaches for predicting spoilage [24]. Marín et al. [32] studied the colonisation and production of FB1 and FB2 at different moisture contents as a measure of the range of conditions over which growth and mycotoxin production may be naturally initiated. There are few data about inhibition of F. verticillioides by bacterial isolates. Hinton and Bacon [17] reported that one isolate of Enterobacter cloacae associated as an endophyte with corn roots, stems and leaves is antagonistic to F. moniliforme and other toxic fungi associated with corn. More recently Motomura et al. [37] obtained Gram-positive bacilli with anti-Fusarium moniliforme activity from soil and corn but studies involving F. verticillioides and bacterial antagonists in fumonisin production are very limited [26]. In this study we detailed the impact of interactions between F. verticillioides isolates and bacterial antagonists in F. verticillioides growth and fumonisin B1 production at different aW conditions as second screening methodology to show the potential inhibition of pathogen colonisation. NOI, antibiosis, lag time for growth, growth rate and fumonisin B1 production on paired cultures between pathogens and antagonists are critical data. They serve as a base line for any studies involving preservative screenings for control of F. verticillioides fumonisin B1 producers in agricultural products. The results of the present work suggest that antagonistic rhizobacterial isolates can become candidates as biological

752

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

Fig. 1. Mean values of (a) growth rate, (b) lag phase and (c) antifungal activity of 13 F. verticillioides isolates paired with different bacterial antagonists. Bars represent standard deviation of means. Histograms with different letters are significantly different (P < 0.001) according to Duncan multiple range test.

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

control agents due to their capacity for developing different strategies to compete against F. verticilliodes. A. armeniacus RC2, A. armeniacus RC3, B. subtilis RC8, B. subtilis RC9, B. subtilis RC11, P. solanacearum RC7, and P. solanacearum RC10 could have practical value in the control of F. verticillioides root colonisation. Further stages of this on-going study will determine their use at the field level.

Acknowledgements This work was carried out through grants from the Secretaría de Ciencia y Técnica de la Universidad Nacional de Río Cuarto (Res. 241/99) and the Agencia Nacional de Promoción Científica y Tecnológica (PICT99/08-07197) throughout the years 2001 and 2002.

References [1] C.W. Bacon, I.E. Yates, D.M. Hinton, F. Meredith, Biological control of Fusarium moniliforme in maize, Environ. Health Perspect. 109 (2) (2001) 325–332. [2] J.S. Buyer, L.E. Drinkwater, Comparison of substrate utilization assay and fatty acid analysis of soil microbial communities, J. Microbiol. Methods 30 (1997) 3–11. [3] B. Cahagnier, D. Melcion, R. Richard-Molard, Growth of Fusarium moniliforme and its biosynthesis of fumonisin B1 on maize grain as a function of different water activities, Lett. Appl. Microbiol. 20 (1995) 247–251. [4] L. Cavaglieri, M. Etcheverry, Influence of Arthrobacter spp. on Fusarium verticillioides growth and fumonisin B1 production isolated from maize rhizosphere, in: III Latin American Congress, 2000, p. 43. [5] I. Chet, J. Inbar, Biological control of fungal pathogens, Appl. Biochem. Biotechnol. 48 (1) (1994) 37–43. [6] L. Chernin, Z. Ismailov, S. Haran, I. Chet, Chitinolytic Enterobacter agglomerans antagonistic to fungal plant pathogens, Appl. Environ. Microbiol. 61 (5) (1995) 1720–1726. [7] S.N. Chulze, M.L. Ramírez, M.C. Farnochi, M. Pascale, A. Visconti, G. March, Fusarium and fumonisins occurrence in Argentinian corn at different ear maturity stages, J. Agric. Food Chem. 44 (1996) 2797– 2801. [8] R.J. Cook, Making greater use of introduced microorganisms for biological control of plant pathogens, Annu. Rev. Phytophatol. 31 (1993) 53–80. [9] E.A. Curl, B. Truelove (Eds.), The Rhizosphere, Springer-Verlag, New York, 1986. [10] H. Dallyn, A. Fox, Spoilage of material of reduced water activity by xerophilic fungi, in: G.H. Gould, J.E.L. Carry (Eds.), Society of Applied Bacteriology Technical Series, vol. 15, Academic Press, London, 1980, pp. 129–139. [11] G.A. Dickie, C.R. Bell, A full factorial analysis of nine factors influencing in vitro antagonistic screens for potential biocontrol agents, Can. J. Microbiol. 41 (1995) 284–293. [12] B. Doko, S. Rapior, A. Visconti, J. Schjoth, Incidence and levels of fumonisin contamination in maize genotypes grown in Europe and Africa, J. Agric. Food. Chem. 43 (1995) 429–434. [13] B.J. Duijff, G. Recorbet, P.A.M. Bakker, J.E. Loper, P. Lemanceau, Microbial antagonism at the root level in involved in the suppression of Fusarium wilt by the combination of nonpathogenic Fusarium oxysporum Fo47 and Pseudomonas putida WCS358, Biol. Control 89 (11) (1999) 1073–1079.

753

[14] M. Etcheverry, A. Torres, M.L. Ramírez, S. Chulze, N. Magan, In vitro control of growth and fumonisin production by F. verticillioides and F. proliferatum using antioxidants under different water availability and temperature regimes, J. Appl. Microbiol. 92 (2002) 624–632. [15] W.C.A. Gelderblom, N.P.J. Kriek, W.F.O. Marasas, P.G. Thiel, Toxicity and carcinogenicity of the Fusarium moniliforme metabolite, fumonisin B1 in rats, Carcinogenesis 12 (1992) 1247–1251. [16] H.H.L. González, S.L. Resnik, Fumonisins in Argentinian field-trial corn, J. Agric. Food Chem. 41 (1993) 891–895. [17] D.M. Hinton, C.W. Bacon, Enterobacter cloacae is an endophytic symbiont of corn, Mycopathologia 129 (1995) 117–125. [18] J.G. Holt (Ed.), Bergey’s Manual of Systematic Bacteriology, ninth ed., Williams & Wilkins, Baltimore, 1993. [19] B.R. Kerry, Rhizosphere interactions and the exploitation of microbial agents for the biological control of plant-pathogenic fungi, Annu. Rev. Phytopathol. 38 (2000) 423–441. [20] N.R. Krieg (Ed.), Bergey’s Manual of Systematic Bacteriology, vol. 1, Williams & Wilkins, Baltimore, 1984. [21] M. La Penna, A. Nesci, M. Etcheverry, In vitro studies on the potential for biological control on Aspergillus section Flavi by Kluyveromyces spp., Lett. Appl. Microbiol. 38 (2004) 257–264. [22] R.P. Larkin, R. Fravel, Efficacy of various fungal and bacterial biocontrol organisms for control of Fusarium wilt of tomato, Plant Dis. 82 (9) (1998) 1022–1028. [23] H.B. Lee, N. Magan, Environmental factors and nutritional utilization patterns affect niche overlap indices between Aspergillus ochraceus and other spoilage fungi, Lett. Appl. Microbiol. 28 (1999) 300–304. [24] L. Leistner, Principles and application of hurdle technology, in: G.W. Gould (Ed.), New Methods of Food Preservation, Blackie, Glasgow, UK, 1995, pp. 1–21. [25] P. Lemanceau, C. Alabouvette, Suppression of Fusarium-wilts by fluorescent pseudomonads: Mechanisms and applications, Biocontrol Sci. Tech. 3 (1993) 219–243. [26] J.M. Ligon, D.S. Hill, P.E. Hammer, N.R. Torkewitz, D. Hofmann, H.J. Kempf, K.H. van Pee, Natural products with antifungal activity from Pseudomonas biocontrol bacteria, Pest Man. Sci. 56 (2000) 688–695. [27] N. Magan, J. Lacey, Ecological determinants of mould growth in stored grain, Intern. J. Food Microbiol. 7 (1988) 245–256. [28] W. Mao, J.A. Lewis, P.K. Hebbar, R.D. Lumsden, Seed treatment with a fungal or a bacterial antagonist for reducing corn damping-off caused by species of Pythium and Fusarium, Plant Dis. 81 (5) (1996) 450– 454. [29] W.F.O. Marasas, Fumonisins: Their implications for human and animal health, Nat. Toxins 3 (1995) 193–198. [30] W.F.O. Marasas, P.E. Nelson, T.A. Tousson (Eds.), Toxigenic Fusarium Species, The Pennsylvania State Univ. Press, University Park, 1984. [31] S. Marín, V. Sanchis, A.J. Ramos, I. Vinas, N. Magan, Environmental factors, in vitro interactions, and niche overlap between Fusarium moniliforme, Fusarium proliferatum and Fusarium graminearum, Aspergillus and Penicillium species from maize grain, Mycol. Res. 102 (7) (1998) 831–837. [32] S. Marín, V. Sanchis, I. Vinas, R. Canela, N. Magan, Effect of water activity and temperature on growth and fumonisin B1 and B2 production by Fusarium proliferatum and Fusarium moniliforme on maize grain, Lett. Appl. Microbiol. 21 (1995) 298–301. [33] S. Marín, V. Sanchis, F. Rull, A.J. Ramos, N. Magan, Colonization of maize grain by Fusarium moniliforme and Fusarium proliferatum in the presence of competing fungi and their impact on fumonisin production, J. Food Protect. 61 (11) (1998) 1489–1496. [34] J.K. Martin, The variation with plant age of root carbon available to soil microflora, in: M.W. Loutit, J.A.R. Miles (Eds.), Microbial Ecology, Springer-Verlag, Berlin, 1978, pp. 299–302. [35] E.A. Milus, C.S. Rothrock, Efficacy of bacterial seed treatments for controlling Phytium root rot of winter wheat, Plant Dis. 81 (1996) 180– 184.

754

L. Cavaglieri et al. / Research in Microbiology 155 (2004) 747–754

[36] S. Mischke, A quantitative bioassay for extracellular metabolites that antagonize growth of filamentous fungi, and its use with biocontrol fungi, Mycopathologia 137 (1997) 45–52. [37] M. Motomura, C.E. Lourenço, D. Venturini, Y. Ueno, E.Y. Hirooka, Screening and isolation of anti-Fusarium moniliforme compounds producing microorganisms from soil and corn, Rev. Microbiol. 27 (1996) 213–217. [38] E.D. Nelson, Exudate molecules initiating fungal responses to seeds and roots, in: D.L. Keister, O.B. Cregan (Eds.), The Rhizosphere and Plant Growth, Beltsville Symposia in Agricultural Research, vol. 14, Kluwer Academic, Dordrecht, 1991, pp. 197–209. [39] P.E. Nelson, T.A. Toussoun, W.F.O. Marasas (Eds.), Fusarium Species: An Illustrated Manual for Identification, The Pennsylvania State Univ. Press, University Park, PA/London, 1983. [40] W.P. Norred, Fumonisins–mycotoxins produced by Fusarium moniliforme, J. Toxicol. Environ. Health 38 (1993) 309–328. [41] G.P. Quinn, M.J. Keough (Eds.), Experimental Design Data analysis for biologists, Cambridge Univ. Press, Cambridge, 2002, pp. 72–76. [42] P.F. Ross, P.E. Nelson, J.L. Richard, G.D. Osweiler, L.G. Rice, R.D. Plattner, T.M. Wilson, Production of fumonisins by Fusarium moniliforme and Fusarium proliferatum isolated associated with equine leukoencephalomalacia and pulmonary edema syndrome in swine, Appl. Environ. Microbiol. 56 (1990) 3225–3226. [43] A.D. Ryan, L.L. Kinkel, Inoculum density and population dynamics of suppressive and pathogenic Streptomyces strains and their rela-

[44]

[45]

[46] [47]

[48]

[49]

[50]

tionship to biological control of potato scab, Biol. Control 10 (1997) 180–186. B.M. Sharga, G.D. Lyon, Bacillus subtilis BS 107 as an antagonist of potato blackleg and soft rot bacteria, Can. J. Microbiol. 44 (1998) 777–783. G.S. Shephard, E.W. Sydenham, P.G. Thiel, W.C.A. Gelderblom, Quantitative determination of fumonisin B1 and B2 by highperformance liquid chromatography with fluorescence detection, J. Liq. Chromatogr. 13 (1990) 2077–2087. P.H.A. Sneath (Ed.), Bergey’s Manual of Systematic Bacteriology, vol. 2, Williams & Wilkins, Baltimore, 1986. C. Steinberg, J.M. Whipps, D.A. Wood, J. Fenlon, C. Alabouvette, Effects of nutritional resources on growth of one non-pathogenic strain and four strains of Fusarium oxysporum pathogenic on tomato, Mycol. Res. 103 (9) (1999) 1210–1216. C.Y. Warfield, D.G. Gilchrist, Influence of kernel age on fumonisin B1 production in maize by Fusarium moniliforme, Appl. Environ. Microbiol. 65 (7) (1999) 2853–2856. M. Wilson, S.E. Lindow, Coexistence among epiphytic bacterial populations mediated through nutricional resource partitioning, Appl. Environ. Microbiol. 60 (1994) 4468–4477. M. Wilson, S.E. Lindow, Ecological similarity and coexistence of epiphytic ice-nucleating (Ice+ ) Pseudomonas syringae strains and non-nucleating (Ice− ) biological control agent, Appl. Environ. Microbiol. 60 (1994) 3128–3137.