SEARCH

SEARCH BY CITATION

Abstract

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

Interaction of Fusarium oxysporum and Paenibacillus polymyxa starts with polar attachment of bacteria to the fungal hyphae followed by the formation of a large cluster of non-motile cells embedded in an extracellular matrix in which the bacteria develop endospores. Enumeration of fungal viable counts showed that less than one of 36 000 colony-forming units survived in paired cultures for 71 h. Effective antagonism was not observed below pH 5 and was specific for the bacterial species. Development of F. oxysporum was inhibited in cell-free filtrates derived from cultures of P. polymyxa, but was much more strongly repressed in the presence of living bacteria. Furthermore, recovery of fungal growth started immediately after addition of antibiotics to paired cultures. Restoration of fungal growth was enhanced in filtrates that were supplemented with MgCl2, which suggests that anti-fungal compounds produced by the bacteria were counteracted by magnesium ions. In paired cultures, fungal counts remained very low, even in the presence of the magnesium salt.This study clearly showed that P. polymyxa antagonizes the plant pathogenic fungus F. oxysporum in liquid medium by means of an interaction process in which the presence of living bacteria is a prerequisite for continuous suppression of fungal growth.

The soil-inhabiting bacterium Paenibacillus polymyxa (previously Bacillus polymyxa but reclassified; Ash et al. 1994) can act as a plant growth-promoting bacterium in association with several plant species (Holl et al. 1988; Holl & Chanway 1992; Shishido et al. 1996). Initially, this beneficial effect may result from the capability of P. polymyxa to fix molecular nitrogen (Heulin et al. 1994). Additionally, P. polymyxa indirectly activates Rhizobium etli populations as a co-resident of the rhizosphere of Phaseolus vulgaris (Petersen et al. 1996). Other plant growth-promoting activities may include phosphate solubilization, as suggested by Jisha & Alagawadi (1996), and the production of indolic compounds as auxin (Lebuhn et al. 1997).

As a member of the rhizosphere community, P. polymyxa may antagonize plant pathogenic micro-organisms and therefore, minimize damage to the roots. The bacterium has long been known for its ability to produce antibacterial compounds, i.e. polymyxins, which were described for the first time in 1947 and since then have been extensively studied (for review see Storm et al. 1977). In addition, P. polymyxa produces other antimicrobial substances active against bacteria and fungi (Kurusu & Ohba 1987; Rosado & Seldin 1993; Pichard et al. 1995; Kajimura & Kaneda 1996). Paenibacillus polymyxa isolated from wheat rhizosphere exhibited a clear in vitro antagonistic activity against the fungus Gaeumannomyces graminis var. tritici that causes take-all of wheat (Heulin et al. 1994). Bacilli isolated from rhizosphere or non-rhizosphere habitats showed more heterogeneous chitinase activities than P. polymyxa originating from the rhizoplane (Mavingui & Heulin 1994). Therefore, the in vitro activity against G. graminis var. tritici was not correlated with the activity of chitinase, which suggests that additional mechanisms may be important. Recently, in vitro antagonism against the fungi Aphanomyces cochleoides, Pythium ultimum and Rhizoctonia solani was reported (Nielsen & Sorensen 1997). Seed inoculation of Pisum sativum with B. subtilis and P. polymyxa increased seedling survival and shoot dry weight in soil infested with Pythium spp. (Hwang et al. 1996). However, a similar treatment of safflower seeds, with isolates of P. polymyxa from rhizospheres in Southern Alberta was not effective against damping-off caused by Pythium sp. ‘group G’ (Liang et al. 1996).

Smid et al. (1993) isolated from soil six strains of Bacillus with in vitro antagonism towards Penicillium hirsutum, the causative agent of storage-rot on flower bulbs. Among these, five isolates were identified as P. polymyxa and one as P. macerans. Compounds produced by these organisms in a chemically defined medium inhibited the development of the fungi P. hirsutum and Fusarium oxysporum f. sp. narcissi. Pilot studies with one of these P. polymyxa isolates showed that motile bacterial cells were attracted towards conidia of P. hirsutum during the processes of swelling and germination. This prompted a study of the nature of the direct interaction between a filamentous fungus and B. polymyxa in more detail. As a model organism, F. oxysporum (Schlecht) f. sp. tulipae, a plant pathogenic fungus that can easily be cultivated in liquid media, was chosen.

Materials and methods

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

Organisms and growth conditions

Paenibacillus polymyxa T129 and P. macerans G22 were isolated from different soils (Smid et al. 1993) and maintained at −80 °C as frozen endospore suspensions containing 15% (v/v) glycerol. The endospore suspensions were obtained from agar plate cultures of the bacterial species on Potato Dextrose Agar (PDA, Oxoid, CM139) or Tryptone Soya Agar (TSA, Oxoid, CM 131). On the latter medium, P. polymyxa exhibited less slimy growth.

Fusarium oxysporum Schlecht f.sp. tulipae CBS 195·65 was kept on 2% malt agar at 4 °C. Initially, batch media were inoculated by means of agar pieces containing fungal material. Routinely, liquid cultures (stored at 4 °C) that had formed many microconidia were used for inoculation of the paired cultures.

Both the bacterium and the fungus were cultured on a chemically defined minimal medium containing 1% (v/v) glycerol as a carbon and energy source. Further components (g l−1) were: (NH4)2SO4 (1·0), NaCl (2·0), MgSO4 (0·25), arginine (0·7) and glutamine (0·1). The medium was buffered with 75 mmol l−1 potassium phosphate or HEPES (for the experiments with higher concentrations of magnesium) at pH 7·2. In the latter case, 7·5 mmol l−1 phosphate buffer was supplied as a phosphorus source for the micro-organisms. Vischniac trace elements (1 ml l−1) and biotine/thiamine solution (1 mg l−1 for each vitamin) were also added. The vitamins, MgSO4 and amino acids were filter-sterilized (0·22 μm) and added to heat-sterilized mineral medium.

Escherichia coli K12 was pre-cultured on TSB (Tryptone Soya Broth, Oxoid, CM 129) medium.

Interaction studies

Bacteria were pre-cultured in mineral medium supplemented with 1% glycerol that was inoculated with an endospore suspension (resulting in a 25-fold dilution) and cultivated in an orbital shaker (25 °C, 100 rev min−1, Gallenkamp, Loughbrough, UK). The presence of a tryptone yeast solution (Oxoid, 25 times diluted) enhanced the numbers of motile cells.

Fungal cultures were inoculated by addition of a stationary culture (approximately 107 cfu ml−1, stored at 4 °C) 25 times diluted into fresh medium, and after 24–48 h of cultivation (25 °C, 100 rev min−1), diluted for a second time and inoculated 15–20 h prior to the addition of motile cells of P. polymyxa (approximately 106 cfu ml−1). At regular time intervals, paired cultures were sampled for viable counts, determination of antifungal activity, and optical and fluorescence microscopy.

Viable counts were enumerated on plate count agar (for bacteria) or oxytetracycline glucose yeast agar (for fungi) to which antibiotics (gentamycin, 50 mg l−1; oxytetracycline, 100 mg l−1) were added. Antifungal activity was determined (in duplo) in a microtitre plate assay with the yeast Saccharomyces cerevisiae (CBS 1782) as an indicator organism. Samples (1–2 ml) of the paired cultures were centrifuged (12 000 g, 5 min), filter-sterilized (0·2 μm) and kept at −20 °C. Serial dilutions of the samples were made in Yeast Nitrogen Base (YNB, Difco), to which 0·5% dextrose (w/v) was added, in microtitre plates and 10 μl of an exponential growing yeast culture were added to each vial. The plates were cultivated in a rotary shaker at 30 °C, 140 rev min−1 at an angle of 45°. After 15–20 h optical density was measured at 665 nm using a microtitre plate-reader. The activity of the filtrate was expressed as the maximal dilution for which growth of the test organism was inhibited by more than 50% with respect to cells that were incubated in YNB (dextrose) without filtrate.

For optical microscopy, fast growing cells of F. oxysporum were mixed with P. polymyxa in a Petri dish and observed at different time intervals with a Zeiss (Oberkochen, Germany) inverted microscope (for minimal disturbance of the interacting organisms) and a Zeiss Axiophot (for detailed investigations). In addition, shaking cultures were sampled and investigated in all experiments.

For fluorescence microscopy, hyphal floccules (diameter approximately 1 mm) were briefly washed in 10 mmol l−1 HEPES buffer (pH 7·2) and transferred into 500 μl of this buffer with 2% glycerol by means of a glass needle. Before the addition of hyphae, 1 μl 10 mmol l−1 FUN-1 and 2·5 μl 5 mmol l−1 Calcofluor White M2R (LIVE/DEAD®FungolightTM Yeast Viability Kit, Molecular Probes Europe, Amsterdam, The Netherlands) were added to the droplet, mixed well, kept for 30–45 min in the dark and subsequently investigated by means of a Zeiss Axiophot.

Influence of culture filtrates on fungal development

In order to evaluate the role of antibiosis in the antagonistic interaction, the development of F. oxysporum in filtrates of cultures of P. polymyxa, which were prepared by centrifugation (15 000 g, 10 min) and filter sterilization (0·45 μm) was studied. To the filtrates, 1% glycerol (v/v) was added for supplementation of the carbon source before fungal cells were added. As preliminary experiments had shown that the presence of magnesium ions counteracted the effect of antifungal compounds on S. cerevisiae in microtitre assays, additional experiments were conducted in which P. polymyxa was cultured on HEPES-buffered media with or without magnesium ions (75 mmol l−1 MgCl2).

Results

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

Axenic growth of F. oxysporum and P. polymyxa in mineral medium

Mineral medium supplemented with glycerol as a carbon and energy source supported good growth of P. polymyxa and F. oxysporum, which is a prerequisite for a well considered investigation of the interaction between both micro-organisms.

During axenic growth of F. oxysporum in this medium, many microconidia germinated to form short hyphae on which phialides and new conidia appeared (Fig. 1a). Hyphae also continued to develop, resulting in the formation of floccules. Sub-cellularly, lipid droplets and elongated mitochondria were observed in actively growing apical cells. Hyphae aged during prolonged cultivation, as was judged by the appearance of vacuoles and the disappearance of lipid droplets. After prolonged cultivation, F. oxysporum caused considerable acidification of the growth medium.

image

Figure 1. (a) Morphology of Fusarium oxysporum cells in an actively growing culture. Note the swelling of the two cells that forms the microconidium (arrow) and gives rise to two hyphae. In this medium, numerous microconidia are formed; (b–g) progress of interaction between F. oxysporum and Paenibacillus polymyxa; (b,c) polar attachment of one bacterium to a hyphal cell during initial stages; (d) small cluster of bacteria around the hypha 2·5 h after addition of P. polymyxa; (e) many bacteria clustering around a hypha (arrow) in which cell death occurs. Note the actively moving cells at the periphery of the nidus; (f) many bacterial clusters (arrow) in a fungal floccule; (g) large bacterial cluster 26 h after start of the interaction. Inside the massive nidus, remnants of dead hyphae are visible. A number of bacterial cells have started to form an endospore (arrow). Bar = 6 μm except in Fig. 1(f) where it equals 30 μm

Download figure to PowerPoint

Germination of P. polymyxa spores typically occurred after a lag phase of 1–2 d. Addition of 20 × diluted TSB medium accelerated this process and within 20 h, motile cells were observed. After 48 h of cultivation, many cells became immobile and appeared to be gathered in a mucous secretion. Subsequently, the bacteria swelled and formed one centrally-located endospore. Growth and development of bacilli were strongly influenced by acidification of the culture fluid as a result of metabolic activity of the organisms. A 30 mmol l−1 phosphate buffer was ineffective in preventing a decrease in the pH below 5. Under these conditions, bacterial cells became non-motile and died, as judged by optical microscopy and viable count measurements. In addition, spore development was hampered; occasionally, cells contained more than one aborted spore.

Course of the interaction process

Interaction experiments were started by addition of motile bacterial cells to fast growing fungal cells in a ratio of approximately 1:1, assessed by means of viable count enumeration. Apart from the shaking cultures, mixtures of the organisms were prepared in Petri dishes for an investigation of the initial stages of the interaction by means of optical microscopy studies. Following their addition, bacteria adhered to fungal cells by polar attachment (Fig. 1b,c). There was no preference for hyphal tips, branch initials or phialides. Occasionally, bacteria became detached again, probably as a result of their movements. After prolonged adhesion, the number of bacteria gradually increased at a specific site (Fig. 1d) in a process which included attraction/adhesion of new cells. During enlargement of this bacterial nidus, the cells near to the fungal cell wall became non-motile, but at the periphery, numerous actively-moving cells were still observed (Fig. 1e).

Fluorescence microscopy showed that unaffected fungal cells accumulated the stain FUN-1 inside vacuoles as a red light-emitting crystal, which indicates active metabolic activity. Fungal cells inside a floccule apparently differed in metabolic activity and the staining was most prominent at hyphal tips. Furthermore, fungal cell walls were clearly stained with the fluorescent dye Calcofluor White (Fig. 2a). After 2·5 h from the beginning of the interaction experiment, metabolic activity was still evident in a number of fungal cells. Additionally, cell wall-associated specks of intensive Calcofluor staining were observed. After 4·5 h, the sites of Calcofluor staining were larger and sometimes a hyphal cell was obstructed by the material (Fig. 2b). Within 6 h, damage to the sub-cellular morphology of hyphal cells was observed with phase contrast optical microscopy (Fig. 1e). Within the floccule, many bacterial nidi were observed (Fig. 1f).

image

Figure 2. Staining of fungal cells with Calcofluor White. (a) Control cells show staining of the fungal cell wall (arrow); (b) 4 h after addition of the bacteria, several specks of intense staining are visible; (c) hyphae and conidia amidst Escherichia coli without signs of morphological change. Bar = 6 μm

Download figure to PowerPoint

After 26 h, massive disintegration of the fungal cells was observed (Fig. 1g), although the number of Calcofluor-coloured structures had not increased above the level observed after 4·5 h.

pH dependency of the interaction process

In paired cultures containing a 30 mmol l−1 phosphate buffer, antagonism of bacteria towards fungal cells was confined to the very early stages of interaction, as was revealed by optical microscopy. Viable counts of fungal cells decreased only by 10–50% during the course of the interaction. In media with higher buffering capacities (50 and 100 mmol l−1 potassium phosphate buffers), acidification was limited, which allowed undisturbed development of the bacilli. Under these conditions, an extreme reduction of the fungal viable count took place. Less than one of 36 000 colony-forming units survived co-cultivation for 71 h when compared with the viable count of axenic cultures (Fig. 3a). The same characteristics of the interaction process were observed in non-agitated Erlenmeyer flasks (data not shown).

image

Figure 3. Enumeration of viable counts during cultivation of Fusarium oxysporum and two species of Paenibacillus which can form antifungal compounds. (a) Extensive death of fungal cells (▪) in the presence of Paenibacillus polymyxa (●) compared with axenic fungal cultures (□); (b) no effects on F. oxysporum are observed in paired cultures with P. macerans (▪) compared with controls (□, ○). The bacterial species P. macerans shows a sharp decline in viable count (●) in paired cultures compared with controls (axenic cultures)

Download figure to PowerPoint

Paired cultures of F. oxysporum with E. coli or P. macerans

In order to determine whether the antagonistic effect of P. polymyxa resulted from competition for nutrients, development of F. oxysporum was studied in the presence of the fast growing bacterial species Escherichia coli. Viability of the fungus was not affected during at least 160 h of cultivation, with more than 1010 bacterial cells ml−1 culture fluid present (Fig. 2c, viable count experiments not shown). Clearly, the presence of a large number of living bacteria around fungal cells does not necessarily account for the antagonistic effect. Paired cultures with the related species Paenibacillus macerans, which produces antifungal compounds (data not shown) on the mineral medium, were investigated to reveal the specific nature of the antagonistic interaction between F. oxysporum and P. polymyxa. While viable counts of the fungus dropped below 1% within 120 h with P. polymyxa, no inhibitory effect was observed in the presence of P. macerans (Fig. 3b). On the other hand, after 120 h, hardly any bacteria were observed by optical microscopy and viable count measurements.

Culture filtrates and fungal development

During the course of the experiments, very little growth inhibitory activity was detected in the culture fluid of paired cultures of P. polymyxa and F. oxysporum by means of the microtitre plate assay. Batch cultures of P. macerans on glycerol-containing medium, however, produced compounds that clearly inhibited growth of S. cerevisiae (with an antibiotic activity of >16, see Materials and Methods). This suggests that P. macerans, under the prevailing conditions, produces a compound(s) that is specific for the yeast but which cannot inhibit F. oxysporum, and illustrates the specificity of the interaction between P. polymyxa and F. oxysporum.

For an evaluation of the effects of freely-diffusing antifungal compounds produced by P. polymyxa on the development of F. oxysporum, fungal cells were transferred to filtrates derived from cultures of P. polymyxa that had started to form clusters and endospores. Both culture filtrates and paired cultures caused extensive damage to hyphal cells and microconidia in 24 h (Fig. 4a). However, 78 and 150 h after addition of F. oxysporum to the filtrates, the viable count of the fungal cells was markedly higher compared with that of the paired cultures (Fig. 3a). Paired cultures to which antibiotics specific for bacteria (the same mixture as used for agar surfaces, see Materials and Methods) were added 24 h after the onset of the interaction, showed that P. polymyxa cells had stopped clustering around hyphae. After 141 h, the viable count of F. oxysporum cultures treated with antibiotics was nearly 5–6 orders of magnitude higher than the controls (Fig. 4b). These experiments clearly illustrate that the presence of living bacteria is important for long lasting suppression of fungal growth.

image

Figure 4. The role of the presence of living bacterial cells of Paenibacillus polymyxa in paired cultures. (a) Development of Fusarium oxysporum in paired cultures (□) or filtrates derived from P. polymyxa cultures (▪). For the preparation of filtrates 24–48 h old cultures (when many motile cells are starting to form clusters and spores) were used. Subsequently, the filtrates were inoculated with a concentrated suspension of hyphal cells and microconidia of F. oxysporum (to prevent considerable dilution of the filtrates and antifungal compounds). Therefore, fast-growing fungal cultures were centrifuged at 5000 g for 10 min; (b) addition of antibacterial antibiotics (gentamycin 50 mg l−1; oxy-tetracycline 100 mg l−1) to paired cultures (arrow, ▪) that show restoration of fungal growth while control flasks contain very low fungal viable counts (□) after 141 h

Download figure to PowerPoint

Fluorescence microscopy was conducted on living cells in the presence of culture filtrates of P. polymyxa diluted × 4 in mineral growth medium. In this case, cell wall appositions were observed which stained intensely with Calcofluor White, as in paired cultures. This indicates that the response of the fungal cells is independent of the presence of bacterial cells. Depositions emitting red staining of the fluorescent dye FUN-1 could be observed inside vacuoles. Control cells cultivated in the absence of the filtrate showed more intense staining, indicating a higher metabolic activity at that stage of cultivation.

Magnesium ions which counteract the activity of antifungal compounds produced by P. polymyxa (see Materials and Methods) were used to emphasize that the antagonism of the bacterium towards F. oxysporum is not solely determined by production and excretion of an antifungal factor. In filtrates containing 75 mmol l−1 MgCl2, cell death of F. oxysporum was markedly reduced compared with controls which contained standard amounts of 1 mmol l−1 MgCl2. Within 48 h, the fungal cells had recovered well and showed distinct growth (Fig. 5a). In culture fluids with a low concentration of magnesium ions, recovery also occurred but at a much later stage of the process. Both F. oxysporum and P. polymyxa grew well in media that contained magnesium ions and/or HEPES. In paired cultures with 75 mmol l−1 MgCl2, no damage to fungal cells was observed 24 h after the start of the experiment and there was no clustering of motile bacteria around hyphae. After 48 h, extensive cell death had occurred which was associated with the appearance of large clusters of bacteria around the hyphae (Fig. 5b) This might indicate that bacteria react to the presence of 75 mmol l−1 MgCl2 with a delay in development; during later stages of interaction, no major deviations from the pattern observed in other paired cultures were noted. Thus, although compounds that counteract antifungal activity do not result in a major alteration of the antagonistic interaction between P. polymyxa and F. oxysporum, culture fluids show a markedly reduced activity.

image

Figure 5. Magnesium ions counteract antifungal activity of culture-derived filtrates but do not influence fungal viable counts in paired cultures. (a) Filtrates in the presence of 75 mmol l−1 MgCl2 show considerably higher counts of Fusarium oxysporum (▪) during every stage of cultivation, compared with filtrates without the salt (□); (b) in paired cultures with added magnesium ions, fungal counts invariably remain low (▪) amidst numerous bacterial cells and spores (●).The growth media and filtrates in these experiments were buffered with HEPES (67·5 mmol l−1; pH 7·2) to prevent precipitation of Mg2PO4 and contained 7·5 mmol l−1 Kpi buffer (pH 7·2) as phosphor source for the organisms

Download figure to PowerPoint

Discussion

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

Observations on the role of direct contact between organisms in biocontrol situations are relatively scarce. Backhouse & Stewart (1989) describe a strain of P. polymyxa that was weakly antagonistic to Sclerotium cepivorum, the causal agent of Allium white rot. Faull & Campbell (1979) studied the interaction between G. graminis and B. mycoides by means of electron microscopy and concluded that hyphae were lysed in the presence of bacteria.

Polar attachment, the onset of the interaction between P. polymyxa and F. oxysporum, is also observed during the initial stages of lysis of cyanobacteria by a gliding myxobacterium (Daft & Stewart 1973). Subsequent stages differed markedly from the interaction described here as the cyanobacteria were lysed before accumulation of the antagonistic bacteria had occurred. Attachment also plays a role in the Enterobacter cloacaePythium ultimum model system (Nelson & Maloney 1992). Addition of living bacterial cells to hyphae resulted in firm attachment, more or less randomly dispersed over the hyphal surface, with evidence of polar attachment (Nelson et al. 1986). Different sugars markedly blocked the binding, suggesting a role of sugar residues during the association. In the P. polymyxaF. oxysporum interaction, attachment is initially polar, but during enlargment of the bacterial nidus, most cells are not in direct contact with the fungal cell wall.

In the present study, the formation of a bacterial nidus around hyphae was found to play an important role in the interaction. Although filtrates of a P. polymyxa culture had an inhibitory effect on the development of the fungus, antagonism was clearly enhanced when the bacteria clustered around the hyphae. Effective antagonism was (i) prevented below pH 5, (ii) was not the result of competition for nutrients and (iii) was specific for P. polymyxa. During cultivation, P. polymyxa formed notable amounts of slimy extracellular material which may have a function in glueing the organisms to surfaces, e.g. plant roots in their natural habitats. The clustering behaviour of bacteria may therefore be part of the normal life cycle of P. polymyxa because the phenomenon also occurred in axenic culture. Localization of the organism directly near the hyphae may, however, be the result of this property and may enhance the effectiveness of any antifungal compounds produced. These compounds may include fusaricidin A, a depsipeptide characterized by Kajimura & Kaneda (1996) which was produced by strains of P. polymyxa isolated from the rhizosphere of garlic plants suffering from basal rot caused by F. oxysporum.

Cell wall-degrading enzymes such as chitinase and glycanase (Mavingui & Heulin 1994; Nielsen & Sorensen 1997) may also play a role in antagonism towards fungi. Production of these enzymes is also observed in Erwinia agglomerans (also known as Erwinia herbicola) which shows antifungal activity by protecting cotton seedlings from Rhizoctonia solani damping-off (Chernin et al. 1995). With E. cloacae, inhibitory effects on spore germination and germ tube elongation of Botrytis cinerea, F. solani and Uncinula necator were synergistically increased by mixing the bacterial species with fungal chitinolytic enzymes (Lorito et al. 1993). The bacterium was able to adhere to fungal cells even in the presence of sucrose. Lorito and coworkers observed that localization of bacteria near hyphae is important for effective antagonism but, under the experimental conditions, is not the result of chitinolytic activities.

The different stages of fungal cell death are similar to those observed in the fungus Arthrobotrys oligospora opposed by Drechmeria coniospora inside nematode corpses (Dijksterhuis et al. 1994); they resemble the damage described by Lorito et al. (1993) when several fungi are treated with E. cloacae in combination with enzymes produced by the fungus Trichoderma harzianum. These stages include (i) formation of cell-wall appositions, (ii) shrivelling of the cell wall and (iii) disintegration of the cell contents.

Our results show that the presence of living bacteria is a prerequisite for continued suppression of fungal growth. This may be due to the combined effect of continuous production of antifungal compounds and effective localization of bacterial cells towards fungal hyphae so that antifungal compounds are kept at the site of action. However, there might be additional factors which add to the antagonistic interaction between P. polymyxa and F. oxysporum including (i) extreme densities of bacteria around hyphal cells that may act as a nutrient sink, resulting in a weaker condition of the fungal cells, or (ii) effects of the extracellular material produced by the bacteria. During development, P. polymyxa clearly produces freely diffusible compounds that do damage to the hyphae and microconidia of F. oxysporum. This activity is (i) unable to influence the development of S. cerevisiae in contrast to compounds produced by the related bacterial species P. macerans, (ii) counteracted by the presence of magnesium ions and (iii) unable to inhibit fungal growth for a prolonged time. It is not clear if the latter is the result of active degradation (by the fungus) or instability of the compounds produced.

This study clearly shows that P. polymyxa is able to antagonize cells of the important plant pathogenic fungus F. oxysporum in a liquid medium by means of an interaction process in which the presence of the living bacterium is a prerequisite for continuous suppression of fungal growth.

Acknowledgements

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

The work presented here was made possible by a grant of the PTBS (project BIO 93041). The authors are indebted to Dr Roy Moezelaar at the ATO for critical reading of the manuscript and valuable suggestions, and to Ir. Marleen Visker for help during experiments.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Ash, C., Priest, F.G. & Collins, M.D. 1994 Molecular indentification of rRNA group 3 bacilli (Ash, Farrow, Wallbanks and Collins) using a PCR probe test. Antonie Van Leeuwenhoek, 64, 253260.
  • Backhouse, D. & Stewart, A. 1989 Interactions between Bacillus species and sclerotia of Sclerotium cepivorum. Soil Biology and Biochemistry, 21, 173176.
  • Chernin, L.S., Ismailov, Z.F., Haran, S. & Chet, I. 1995 Chitinolytic Enterobacter agglomerans antagonistic to fungal plant pathogens. Applied Environmental Microbiology, 61, 17201726.
  • Daft, M.J. & Stewart, W.D.P. 1973 Light and electron microscope observations on algal lysis by bacterium CP-1. New Phytology, 72, 799808.
  • Dijksterhuis, J., Sjollema, K.A., Veenhuis, M. & Harder, W. 1994 Competitive interactions between two nematophagous fungi during infection and digestion of the nematode Panagrellus redivivus. Mycological Research, 98, 14581462.
  • Faull, J.L. & Campbell, R. 1979 Ultrastructure of the interaction between the take-all fungus and antagonistic bacteria. Canadian Journal of Botany, 57, 18001808.
  • Heulin, T., Berge, O., Mavingui, P., Gouzou, L., Hebbar, K.P. & Balandreau, J. 1994 Bacillus polymyxa and Rahnella aquatilis, the dominant N2-fixing bacteria associated with wheat rhizosphere in French soils. European Journal of Soil Biology, 30, 3542.
  • Holl, F.B. & Chanway, C.P. 1992 Rhizosphere colonization and seedling growth promotion of lodgepole pine by Bacillus polymyxa. Canadian Journal of Microbiology, 38, 303308.
  • Holl, F.B., Chanway, C.P., Turkington, R. & Radley, R.A. 1988 Response of crested wheatgrass (Agropyron cristatum L.), perennial ryegrass (Lolium perenne) and white clover (Trifolium repens L.) to inoculation with Bacillus polymyxa. Soil Biology and Biochemistry, 20, 1924.
  • Hwang, S.F., Chang, K.F., Howard, R.J., Deneka, B.A. & Turnbull, G.D. 1996 Decrease in incidence of Pythium damping-off of field pea by seed treatment with Bacillus spp. & metalaxyl. Journal of Plant Disease and Protection, 103, 3141.
  • Jisha, M.S. & Alagawadi, A.R. 1996 Nutrient uptake and yield of sorghum (Sorghum bicolor L. Moench) inoculated with phosphate solubilizing bacteria and cellulolytic fungus in a cotton stalk amended vertisol. Microbiological Research, 151, 213217.
  • Kajimura, Y. & Kaneda, M. 1996 Fusaricidin A, a new depsipeptide antibiotic produced by Bacillus polymyxa KT-8 taxonomy, fermentation, isolation, structure elucidation and biological activity. Journal of Antibiotics, 49, 129135.
  • Kurusu, K. & Ohba, K. 1987 New peptide antibiotics LI-F03, F04, F05, F07 and F08, produced by Bacillus polymyxa. I. Isolation and characterization. Journal of Antibiotics, 40, 15061514.
  • Lebuhn, M., Heulin, T. & Hartmann, A. 1997 Production of auxin and other indolic and phenolic compounds by Paenibacillus polymyxa strains isolated from different proximity to plant roots. FEMS Microbiology and Ecology, 22, 325334.
  • Liang, X.Y., Huang, H.C., Yanke, L.J. & Kozub, G.C. 1996 Control of damping-off of safflower by bacterial seed treatment. Canadian Journal of Plant Pathology, 18, 4349.
  • Lorito, M., Di Pietro, A., Hayes, C.K., Woo, S.L. & Harman, G.E. 1993. Antifungal, synergistic interaction between chitinolytic enzymes from Trichoderma harzianum and Enterobacter cloacae. Phytopathology, 83, 721728.
  • Mavingui, P. & Heulin, T. 1994 In vitro chitinase and antifungal activity of a soil, rhizosphere and rhizoplane population of Bacillus polymyxa. Soil Biology and Biochemistry, 26, 801803.
  • Nelson, E.B., Chao, W.L., Norton, J.M., Nash, G.T. & Harman, G.E. 1986 Attachment of Enterobacter cloacae to hyphae of Pythium ultimum: possible role in the biological control of Pythium preemergence damping-off. Phytopathology, 76, 327335.
  • Nelson, E.B. & Maloney, A.P. 1992 Molecular approaches for understanding biological control mechanisms in bacteria: studies of the interaction of Enterobacter cloacae with Pythium ultimum. Canadian Journal of Plant Pathology, 14, 106114.
  • Nielsen, P. & Sorensen, J. 1997 Multi-target and medium independent fungal antagonism by hydrolytic enzymes in Paenibacillus polymyxa and Bacillus pumilus strains from barley rhizosphere. FEMS Microbiology and Ecology, 22, 183192.
  • Petersen, D.J., Srinivasan, M. & Chanway, C.P. 1996. Bacillus polymyxa stimulates increased Rhizobium etli populations and nodulation when co-resident in the rhizosphere of Phaseolus vulgaris. FEMS Microbiology Letters, 142, 271276.
  • Pichard, B., Larue, J.P. & Thouvenot, D. 1995 Gavaserin and saltavalin, new peptide antibiotics produced by Bacillus polymyxa. FEMS Microbiology Letters, 133, 215218.
  • Rosado, A.S. & Seldin, L. 1993 Production of a potentially novel anti-microbial substance by Bacillus polymyxa. World Journal of Microbiological Biotechnology, 9, 521528.
  • Shishido, M., Masicotte, H.B. & Chanway, C.P. 1996 Effect of plant growth promoting Bacillus strains on pine and spruce seedling growth and mycorrhizal infection. Annals of Botany, 77, 433441.
  • Smid, E.J., Beek, J.A.M. & Van And Gorris, L.G.M. 1993 Biological control of Penicillium hirsutum by antagonistic soil bacteria. In: Biological Control of Foliar and Post-Harvest Diseases, Vol. 16, eds. Fokkema, N. ., Köhl, J. & Elad, Y, pp. 190193. IOBC/WPRS Bulletin. Wageningen, The Netherlands.
  • Storm, D.R., Rosenthal, K.S. & Swanson, P.E. 1977 Polymyxin and related peptide antibiotics. Annual Review of Biochemistry, 46, 723763.