Cocultivation of phytopathogenic Fusarium and Alternaria strains affects fungal growth and mycotoxin production

Authors


Correspondence

Marina Elsa Herta Müller, Leibniz-Centre for Agricultural Landscape Research (ZALF), Institute of Landscape Biogeochemistry, Eberswalder Str. 84, 15374 Müncheberg, Germany. E-mail: mmueller@zalf.de

Abstract

Aims

A laboratory study was conducted to evaluate the influence of cocultivation of toxigenic Fusarium (F.) and Alternaria (A.) fungi with respect to growth and mycotoxin production.

Methods and Results

Fusarium culmorum Fc13, Fusarium graminearum Fg23 and two Alternaria tenuissima isolates (At18 and At220) were simultaneously or consecutively co-incubated on wheat kernels in an in vitro test system. Fungal biomass was quantified by determining ergosterol content. Three Fusarium toxins (DON, NIV and ZON) and three Alternaria toxins (AOH, AME and ALT) were analysed by a newly developed HPLC/MS/MS method. In simultaneous cocultures, the fungal biomass was enhanced up to 460% compared with individual cultures; Alternaria toxins were considerably depressed down to <5%. Combining At18 and At220 with Fg23 inhibited the toxin production of both fungal partners. In contrast, Fc13 increased its DON and ZON production in competitive interaction with both A. strains.

Conclusions

The interfungal competitive effects aid the understanding of the processes of competition of both fungi in natural environments and the involvement of mycotoxins as antifungal factors.

Significance and Impact of Study

Cocultivation significantly affects fungal growth and mycotoxin production of phytopathogenic Alternaria and Fusarium strains. The impact of mycotoxins on the interfungal competition is highlighted.

Introduction

Fusarium head blight (FHB) is considered to be one of the most economically relevant diseases of wheat in many cereal growing regions around the world (Windels 2000; Wu 2004) and is caused by a group of up to 19 Fusarium species (Parry et al. 1995; Champeil et al. 2004). The main causal species in the cooler temperate climates of Central Europe include Fusarium (F.) graminearum [teleomorph Gibberella zeae (Schwein.) Petch] and Fusarium culmorum (W.G. Smith) Sacc (teleomorph unknown) (Bottalico and Perrone 2002). In addition to limiting yield and deteriorating grain quality, both species are capable of producing a number of mycotoxins. The most frequently reported toxic metabolites are deoxynivalenol (DON), nivalenol (NIV) and zearalenone (ZON), which are commonly found on wheat, maize, rice, barley, rye and mixed feed (Visconti et al. 1992; Bottalico 1998). Cereals severely contaminated with mycotoxins are unsuited for animal and human consumption. Fusarium toxins, for instance, may attack the digestive tract, the liver, the endocrine and circulatory systems, the skin and the blood and can cause several acute and chronic diseases in humans and animals (Miller and Trenholm 1994; Rotter et al. 1996). Owing to these health risks, the public authorities established maximum levels of DON, NIV and ZON in food and agricultural commodities (e.g. the European Union: Commission Regulation (EC) No. 1881/2006, EC 2006) and recommended limits for animal feed (EFSA 2004).

Mycotoxin accumulation by Fusarium species occurs mainly before harvest. The infection, colonization and growth of fungi in wheat leaves and ears as well as the DON, NIV and/or ZON production in the infected kernels is greatly dictated by climatic, agricultural and topographic factors (Dill-Macky and Jones 2000; Maiorano et al. 2008; Müller et al. 2010). Although numerous studies demonstrated the important influence of minimum tillage, crop rotation, susceptibility of wheat cultivars as well as humidity conditions during anthesis and ripening, a secure and reliable prognosis of FHB incidence and mycotoxin accumulation for practical purposes is still difficult and inconsistent. The few existent predictive models are usually site-specific and do not provide acceptable accuracy in different environments (Hooker et al. 2002; Schaafsma and Hooker 2007; http://www.isip.de/coremedia/generator/isip/Start,Region=flaeche0000114.html; http://www.lfl.bayern.de/ips/pflanzenschutzhinweise/12410/index.php). The hit-to-miss ratio of such a prediction is still so deficient – among other things – because the forecasting models do not include biotic driving factors. But, inter- and intraspecific microbial interactions, for instance, are essential to understand the process of infection and subsequent colonization of fusaria and their toxin production capability in planta in a natural environment. Until now, investigations of such interactions have focused on biocontrol (Laitila et al. 2002; Diamond and Cooke 2003; Singh Lakhesar et al. 2010). There are several studies of bacterial and fungal antagonists against the FHB pathogens, but control has often proved to be successful only under laboratory or climate chamber conditions (Laitila et al. 2002; Naef et al. 2006; He et al. 2009) and failed completely under field conditions (Harman 2000). Mechanisms such as competition, symbiosis or coexistence of different pathogens have been mostly neglected. Many plant pathogenic fungi compete with each other for nutrient uptake and for infection sites. Some of them can coexist in ecological niches or are separated in time and space (Xu and Nicholson 2009; Singh Lakhesar et al. 2010). But, it is generally agreed that the interspecific competition in an ecosystem is the normal case rather than the exception (Fitt et al. 2006; Xu and Nicholson 2009).

The fungi of the genus Alternaria are considered as another group of ubiquitous pathogens on wheat leaves and ears. Various Alternaria species, especially Alternaria (A.) alternata (Fr.) Keissler and Alternaria tenuissima ([Kunze ex Nees et T. Nees: Fries] Wiltshire), are frequently associated with several plant diseases of small-grain cereals, for example, black point, kernel and leaf blight (Patriarca et al. 2007; Ostry 2008; Logrieco et al. 2009). Both species are capable to produce a variety of mycotoxins belonging to several classes of chemical compounds and possessing various biological activities (cytotoxic, antiviral, antimicrobial, teratogenic, mutagenic etc.). The best-researched Alternaria metabolites are alternariol (AOH), alternariol monomethyl ether (AME), altenuen (ALT), altertoxin and tenuazonic acid (King and Schade 1984; Chełkowski and Visconti 1992; Bottalico and Logrieco 1998). The occurrence of these mycotoxins on cereals as well as on fruits and vegetables has been reported repeatedly (Webley et al. 1997; Li and Yoshizawa 2000; Müller et al. 2002; Azcarate et al. 2008; Logrieco et al. 2009; Siegel et al. 2009).

Fungi of the genera Fusarium and Alternaria attack and infect their cereal host plants in the field, mostly in the period from the anthesis to the mid-grain-filling stage (Chełkowski and Visconti 1992; Champeil et al. 2004). It is well documented that the infection and colonization of plants by these pathogens are strongly favoured by high humidity and that the species-specific mycotoxins can play an important role in their plant pathogenesis (Strandberg 1992; Bateman 2005; Xu et al. 2007). Therefore, a co-occurrence of Fusarium and Alternaria in wheat ears is very likely as is the co-occurrence of different mycotoxins. This implies a potential risk of additional or even synergistic toxic effects on human and animals after consumption of contaminated cereal or cereal products. The few studies dealing with the co-occurrence of these two field fungi genera suggest a competitive interaction between the pathogens rather than an undisturbed coexistence (González et al. 1999; Kosiak et al. 2004; Saß et al. 2007). However, quite generally, our knowledge about the processes of antagonism, competition and/or coexistence of fungi in natural environments and the involvement of mycotoxins as pathogenicity/virulence factors is very limited. Our study aimed at filling this gap. We examined the growth and the mycotoxin production of two Fusarium and two Alternaria isolates after inoculation of wheat kernels in an in vitro model system. The strains were inoculated either simultaneously or consecutively. A newly developed multimycotoxin detection method enabled us to simultaneously quantify three Fusarium toxins (DON, NIV and ZON) and three Alternaria toxins (AOH, AME and ALT). Furthermore, the growth of fungi was quantified by determining ergosterol production.

Materials and methods

Fungal isolates used as inoculum

The Fusarium and Alternaria isolates used as inoculum in this study originated from wheat plants: F. culmorum 13 (Fc13), F. graminearum 23 (Fg23), A. tenuissima 18 (At18) and A. tenuissima 220 (At220). They are stored in the culture collection of micro-organisms of the Institute of Landscape Biogeochemistry at the Leibniz-Centre for Agricultural Landscape Research Müncheberg, Germany. Single-spore stock cultures of these isolates are maintained in sterile soil mixtures at 8°C. These four isolates were selected for this study because of the fact that they are proven high mycotoxin producers with a pronounced aggressive potential screened in a biotest (Holz 2010; Korn et al. 2011).

Inoculum preparation

Single conidia of the isolates were grown in Petri dishes containing Synthetic Nutrient Agar (SNA; Nirenberg 1976) for the Fusarium isolates and Potato Carrot Agar (PCA; Simmons 1992) for the Alternaria isolates. Fungi grew at 25°C for 10 days with a 12-h light/dark cycle for 5 days. Macroconidia were harvested as described previously (Korn et al. 2011). The spore suspension was adjusted to a density of 2 × 10conidia ml−1.

In vitro test systems A and B

Wheat kernels (0·2 g) and distilled water (250 μl) were weighed into a 15-ml polypropylene centrifuge tube (Orange Scientific Braine-l'Alleud Belgium) and sterilized at 121°C for 20 min. 50 μl of the inoculum suspension with a spore density of 2 × 105 conidia ml−1 were added. Infected kernels were incubated at 25°C in the dark for 12 days. After the individual incubation time, the tubes were closed with screw caps and deep-frozen (−20°C) until analysis.

In test A, two strains (one A. isolate and one F. isolate, respectively, in four different combinations) were inoculated simultaneously for a total of 4, 8 and 12 days, respectively. DON, ZON, NIV, AOH, AME, ALT and ergosterol were determined in the wheat-mycel mixtures in the tubes at day 4, 8 and 12. Furthermore, the four strains were incubated as individual culture for 4, 8 and 12 days and the species-specific mycotoxins were analysed after incubation. All samples (for each day) were prepared in quintuplicate.

In test B, two strains (one A. isolate and one F. isolate, respectively, in 8 different combinations of Fc13, Fg23, At18 and At220) were consecutively inoculated with a time delay of 3 days for the second strain. The total incubation time was 6, 9 and 12 days, respectively. Ergosterol and mycotoxin quantification was carried out at day 3 for the first inoculated strain (before the inoculation with the second strain) and at day 6, 9 and 12 for the cocultivated samples. All four strains were also incubated individually for 3, 6, 9 and 12 days, with the thus produced ergosterol and mycotoxin quantities serving as reference values. All samples (for each day) were prepared in quintuplicate.

Sterilized wheat kernels were stored without inoculation and analysed at day 0 and day 12 as controls in test system A as well as in test system B.

HPLC-MS/MS analyses of mycotoxins

Acetonitrile/distilled water/acetic acid solution (79/20/1, v/v/v; 5 ml) were added to the wheat-mycel mixtures in the centrifuge tubes, ultrasonicated for 4 min, shaken for 60 min (IKA HS 501 horizontal shaker, IKA, Staufen, Germany), ultrasonicated again for 5 min, shaken once more for 60 min and finally centrifugated for 5 min (3828 g, Sigma 6K15 centrifuge; Sigma Zentrifugen, Osterode am Harz, Germany). The supernatant was deep-frozen till the HPLC-MS/MS analyses. The remainder in the tubes was subsequently subjected to the ergosterol workup procedure.

Crystalline standards of NIV, DON, ZON, ALT, AOH and AME were obtained from Sigma-Aldrich (Taufkirchen, Germany). Individual stock solutions at concentrations ranging from 50 to 800 μg ml−1 (NIV, DON, ZON: 100 μg ml−1; ALT: 46 μg ml−1; AOH: 780 μg ml−1; AME: 133 μg ml−1) were prepared in pure acetonitrile. From the individual stock standard solutions, a standard mixture was prepared at following concentrations: NIV (3 μg ml−1); DON (2 μg ml1); ZON (1 μg ml−1); ALT (0·4 μg ml−1); AOH (1·6 μg ml−1) and AME (0·8 μg ml−1). All stock solutions were stored at −20 °C in the dark. Mixed working solutions (n = 10) for constructing calibration curves were prepared by serial dilution of the multicomponent stock solution with acetonitrile/water at dilution factor of 400 from the highest to the lowest level. All working standard solutions were adjusted to a final identical solvent composition of 50% acetonitrile and 50% water and were stored in the refrigerator at 4 °C until analysis.

HPLC–MS/MS analyses were performed on an API 4000 QTrap® MS/MS system (AB Sciex, Darmstadt, Germany), equipped with an ESI interface and hyphenated to an 1200 series HPLC system comprising a degasser, a binary pump, auto sampler and a column oven from Agilent Technologies (Waldbronn, Germany). The chromatographic separation of 5 μl injected sample was achieved using a Gemini C18 analytical column (100 mm × 2 mm, 3 μm particles; Phenomenex, Aschaffenburg, Germany), preceded by a Gemini C18 guard column (4 mm × 2 mm, 3 μm particles). The column oven was set at 40 °C and the flow rate of the mobile phase was 400 μl min−1. The mobile phase was a time-programmed gradient using A (H2O, 5 mmol l−1 NH4Ac, 1% acetic acid) and B (MeOH, 5 mmol l−1 NH4Ac, 1% acetic acid) channels. The mobile phase gradient consisted of 0–3 min 5% B, 3·01–5 min ramp to 60% B, 5.01–15 min hold at 60% B, 15–15·01 min ramp to 100% B, 15.01–22 min hold at 100% B, and 22–22·01 min ramp back to initial conditions (equilibration time: 8 min). Time between injections was 32 min. The column effluent was directly transferred into the ESI interface, without splitting.

The ESI interface was operated in positive ion mode for AOH and AME and in negative ion mode for DON, NIV, ALT and ZON, with the settings shown in Table S1. Analysis of the 6 mycotoxins was carried out using the multiple reaction monitoring mode (MRM) with both scanning quadrupoles (Q1 and Q3) set to unit resolution. Two transitions were monitored for each precursor ion (quantifier and qualifier) resulted in 4·0 identification points (IPs) (EC 2002). The optimized conditions for each mycotoxin are summarized in Table S2.

During the method development, limits of detection (LOD) and quantification (LOQ) for each analyte were determined on the basis of the signal-to-noise ratios of 3 : 1 and 10 : 1. Absolute values for the detection/quantification limits were found to be 40·2/134·1, 6·8/22·5, 3·3/10·9, 7·1/23·7, 7·1/23·6 and 19·9/66·2 pg for NIV, DON, ZON, ALT, AOH and AME, respectively. Recoveries and relative standard deviations were estimated within satisfactory levels as recommended by the European Commission Regulation (EC 2006).

The Analyst 1.5 software package (AB Sciex) was used to control the HPLC–MS/MS system as well as for data acquisition and processing of quantitative data obtained from standard calibration and samples.

Ergosterol analyses

The ergosterol analyses were carried out according to Zhang et al. (2008) with modifications. One millilitre propan-2-ol, 1 ml methanol and 0·5 ml 2 mol l−1 NaOH were added to the wheat-mycel mixtures in the 15-ml centrifuge tubes. After brief manual shaking, the centrifuge tubes were ultrasonicated for 5 min. Subsequently, the closed tubes were irradiated for 10 s in a Privileg 8018 GE kitchen microwave (Privileg, Stuttgart, Germany) at 800 W. After 3 min of cooling, they were shaken manually and irradiated again. In total, four irradiation cycles were carried out.

After the fourth cycle, 2 ml of 0·5 mol l−1 HCl and 3 ml pentane were added to the tubes and the mixtures were shaken on an IKA HS 501 horizontal shaker at maximum velocity for 5 min. Finally, the tubes were centrifugated in a Sigma 6K15 centrifuge at 3828 g for 5 min and a fraction of the pentane supernatant was transferred into a HPLC vial.

HPLC-UV analyses were carried out on an Agilent 1200 series HPLC tower (Agilent) equipped with an Agilent 1200 series UV detector. A Phenomenex Gemini NC C18 column (150 × 2 mm, 3 μm particles; Phenomenex) was used at a column oven temperature of 30°C. The method was isocratic using 100% acetonitrile as the eluent. The runtime was 9 min, ergosterol eluted at 4·3 min and was detected at λ = 282 nm (bandwidth: 8 nm). The injection volume was 10 μl. LOD of the overall method was 0·3 μg (per tube, corresponding to a S/N ratio of 3 : 1), LOQ was 1·0 μg (S/N ratio 10 : 1). Ergosterol was quantified in a working range of 1–300 μg by means of a nine point calibration curve (R2 = 0·997) constructed using pentane stock solutions.

Statistical analyses

Data from the in vitro studies were tested for normal distribution (Kolmogorov–Smirnov test) and homogeneity of variances (Levene's test). Differences between the individual and cocultivations of different isolates were calculated with the t-test for independent variables or the nonparametric Kruskal–Wallis test. In all tests, the probability was P = 5%. All statistical analyses were computed using the spss statistical package (ver. 15.0; SPSS, IBM, Somers, NY, USA).

Results

In vitro test system A: simultaneous inoculation and cocultivation

Wheat kernels used as nutrient medium in this study were naturally contaminated with a low concentration of DON (199 ± 114 μg kg−1), ZON (40 ± 15 μg kg−1) and ALT (15 ± 11 μg kg−1). AOH, AME and NIV were not detected.

Ergosterol as an indicator of fungal biomass was not found in the control samples during the entire test period.

The Alternaria isolate At220 showed the most remarkable fungal growth rate (mean value of 25·8 μg g−1 at day 12), whereas the further three isolates accumulated significantly less biomass during the entire incubation time (At18: 14·0 μg g−1; Fc13: 18·8 μg g−1; Fg23: 14·5 μg g−1).

Comparing the ergosterol content in the individual cultures of the four isolates to the content in cocultures might indicate interactions between the two phytopathogens. For this purpose, the sum of ergosterol in the individual cultures of both corresponding fungal partners was compared with the ergosterol content in the coculture (Fig. 1). While the growth of all strain combinations in coculture at day 4 was significantly depressed (55·3–69·5%), the fungal biomass in coculture at day 12 is greatly enhanced (171·5–460·0%). The method of determining the fungal biomass used does not allow us to differentiate which fungal strain mostly contributes to this enhanced fungal growth. However, in the cocultures of Fg23 with both A. isolates at day 12 the significantly increased growth is more pronounced (293·8–460·0%) than in cocultures of Fc13 with the both A. isolates (171·5–188·2%, nonsignificant at P level <0·05). In general, our results imply that both phytopathogenic fungi respond to a direct co-occurrence in the same nutrient medium with an increased growth rate of at least one competitor.

Figure 1.

Ergosterol concentration of fungal biomass after incubation of Fusarium culmorum Fc13, Fusarium graminearum Fg23, Alternaria tenuissima At18 and A. tenuissima At220 after simultaneous inoculation on wheat kernels at day 4, 8 and 12 in cocultures and as the sum of the corresponding individual cultures. Mean values and standard deviation (n = 5). Asterisks indicate significant differences (P < 0·05) between the mean values of ergosterol in coculture and in the two corresponding individual cultures summed up. image Co-culture; image sum of single cultures.

The mycotoxin production of both individual Alternaria isolates was qualitatively similar, but different when related to the produced ergosterol: At18 accumulated 322·7 ng AOH per μg ergosterol (301·9 ng AME per μg ergosterol) at day 8, whereas At220 produced the 54-fold amount of AOH (17 478 ng μg−1 ergosterol) and the 17-fold amount of AME (5256 ng per μg ergosterol). The contents of AOH, AME and ALT decreased after peaking at day 8 in both Alternaria isolates (Table 1).

Table 1. Mean values of alternariol (AOH), alternariol monomethyl ether (AME), altenuene (ALT), deoxynivalenol (DON), zearalenone (ZON) and nivalenol (NIV) concentrations and their standard deviations in individual and cocultures of Fusarium culmorum Fc13, Fusarium graminearum Fg23, Alternaria tenuissima At18 and A. tenuissima At220 after simultaneous inoculation and a total incubation on wheat kernels of 4, 8 and 12 days, respectively, as well as the differences (P-values) between the mycotoxin contents in individual and co-cultures
 Day 4Day 8Day 12
Fc13 + At18aFc13 + At220Fg23 +  At18Fg23 +  At220Fc13 +  At18Fc13 +  At220Fg23 +  At18Fg23 +  At220Fc13 +  At18Fc13 +  At220Fg23 +  At18Fg23 +  At220
  1. Significant differences between mean values at P level <0·05 are highlighted in bold.

  2. a

    n = 5 for each strain combination.

AOH (μg μg−1 ergosterol)
Coculture0·01 ± 0·014·7 ± 1·40·01 ± 0·021·9 ± 0·710·01 ± 0·027·4 ± 3·40·001 ± 0·0010·35 ± 0·240·01 ± 0·011·0 ± 0·80·0001 ± 0·00010·34 ± 0·43
Individual culture0·04 ± 0·033·8 ± 2·50·04 ± 0·033·8 ± 2·50·32 ± 0·1317·5 ± 14·40·32 ± 0·1317·5 ± 14·40·03 ± 0·026·6 ± 2·90·03 ± 0·026·6 ± 2·9
P-value 0·028 0·5130·0890·136 0·001 0·165 0·001 0·009 0·060 0·003 0·025 0·001
AME (μg μg−1 ergosterol)
Coculture0·01 ± 0·010·78 ± 0·380·02 ± 0·010·67 ± 0·360·01 ± 0·013·1 ± 2·40·01 ± 0·0030·16 ± 0·110·007 ± 0·0060·52 ± 0·320·005 ± 0·0040·28 ± 0·45
Individual culture0·03 ± 0·020·12 ± 0·090·03 ± 0·020·12 ± 0·090·3 ± 0·245·3 ± 1·00·3 ± 0·245·3 ± 1·00·03 ± 0·023·5 ± 1·80·03 ± 0·023·5 ± 1·8
P-value0·166 0·005 0·368 0·026 0·009 0·097 0·009 0·009 0·066 0·006 0·042 0·004
ALT (μg μg−1 ergosterol)
Coculture00·05 ± 0·040·001 ± 0·0010·01 ± 0·010·0003 ± 0·00020·47 ± 0·170·006 ± 0·0020·01 ± 0·010·001 ± 0·0010·13 ± 0·080·005 ± 0·0020·01 ± 0·01
Individual culture00·05 ± 0·0400·05 ± 0·040·03 ± 0·031·6 ± 1·20·03 ± 0·031·6 ± 1·20·005 ± 0·0031·1 ± 0·640·005 ± 0·0031·1 ± 0·64
P-value0·940 0·007 0·0560·1140·0820·177 0·047 0·047 0·009 0·952 0·005
DON (μg μg−1 ergosterol)
Coculture00000·003 ± 0·0060000·16 ± 0·220·31 ± 0·2900
Individual culture00000·02 ± 0·020·02 ± 0·02000·03 ± 0·020·03 ± 0·0200
P-value 0·039 0·030 0·209 0·009
ZON (μg μg−1 ergosterol)
Coculture000·002 ± 0·00400·23 ± 0·192·4 ± 1·30·18 ± 0·130·3 ± 0·435·3 ± 4·75·1 ± 4·85·7 ± 5·41·1 ± 0·62
Individual culture000·03 ± 0·020·03 ± 0·020·55 ± 0·30·55 ± 0·31·2 ± 0·371·2 ± 0·370·15 ± 0·150·15 ± 0·1537·3 ± 14·937·3 ± 14·9
P-value 0·022 0·020 0·076 0·034 0·001 0·010 0·009 0·009 0·007 0·006
NIV (μg μg−1 ergosterol)
not detected in any of the samples

Fusarium isolates Fc13 and Fg23 differed in their capability to produce DON: Fg23 was not able to produce DON in these experiments. Fc13 continuously increased total DON produced up to 28·0 ng per μg ergosterol at day 12. Both isolates produced zearalenone: Fc13 to a maximum level of 552·9 ng μg−1 at day 8 and Fg23 to an outstanding maximum level of 37 252 ng μg−1 at day 12.

Cocultivation significantly affected mycotoxin production. The production of AOH, AME and ALT by both Alternaria isolates was considerably depressed in coculture with Fc13 as well as with Fg23 (Table 1). Compared with the individual cultures, At220, for example, produced only 42% AOH, 59% AME and 30% ALT in coculture with Fc13 at day 8. The depression was even more pronounced in coculture with the ZON producer Fg23: in this case, At220 was able to produce only 2% AOH, 3% AME and 0·9% ALT at day 8 (Table 1). The lesser mycotoxin producing strain At18 reduced its whole toxin production in coculture with both Fusarium strains down to <5% at day 8. In contrast to the behaviour of At220, Fc13 and Fg23 caused the same degree of mycotoxin production depression when co-inoculated with At18.

The DON production of Fc13 in coculture with the Alternaria isolates was remarkably delayed until day 8: completely inhibited by At220, while At18 caused a depression down to 11%. However, during the eighth and twelfth day, the DON production of Fc13 strongly increased and exceeded the individual culture production by the six-fold (At18) and by the 11-fold (At220) (Table 1).

The ZON production of both Fusarium strains in interaction with the Alternaria was quite different: The weaker ZON producer Fc13 had already enhanced its ZON production in coculture with At220 four-fold at day 8 and 35-fold at day 12. But, in coculture with At18, its ZON production was delayed until day 8, and 36-fold increased between days 8 and 12. Surprisingly, the highly potent ZON producer Fg23 was hindered in its ZON production by both Alternaria strains right from the start of the experiment (Table 1). At day 12, Fg23 had produced ZON in coculture merely to a significantly reduced degree: 15% with At18 and 3% with At220 – compared with the individual cultures. In the combination of At18 and At220 with Fg23, respectively, the toxin production of both fungal partners was reduced. In contrast, Fc13 enhanced its DON as well as ZON production at the end of the experiments in interaction with both Alternaria strains.

In vitro test system B: consecutive inoculation and cocultivation with a time delay for the second isolate

Also in this experiment, the isolate At220 proved to grow remarkably as indicated by the ergosterol content in the wheat-mycel mixture after 12 days (mean value of 128·6 μg g−1). At18, Fc13 and Fg23 grew significantly slower in the 12-day incubation period (mean values of 20·1, 39·8 and 48·6 μg g−1, respectively).

To compare the growth rates in individual and consecutive cocultures, we summed up the ergosterol contents in the individual cultures at the respective incubation times. This comparison (displayed in Fig. 2) revealed a growth depression in the majority of cocultures. Generally, the first inoculated and well growing strain does not admit an undisturbed growth of the second strain. Microbiological analyses showed a weak growth of all consecutively inoculated isolates (data not shown). Therefore, the growth rates in consecutive cocultures do not achieve the summed up values of the growth rates in the corresponding individual cultures. If both Alternaria strains were inoculated before Fc 13, the ergosterol content in these cocultures was significantly depressed (down to 37% in the combination of At18 + Fc13 and 44% in the combination of At220 + Fc13, respectively). The same was true in reverse, if the Fusarium strains had a lead of 3 days (Fig. 2): Only 67% and 33% ergosterol were produced in cocultures in the combination of Fc13 + At18 and Fc13 + At220, respectively, compared with the sum of individual cultures. In all cases, the ergosterol concentrations in cocultures at day 9 and 12 were inhibited compared with the sum of the ergosterol content of individual cultures. However, the analyses of ergosterol content as measurement of the fungal biomass do not allow us to differentiate between the growth rates of the single fungal partners.

Figure 2.

Ergosterol concentration of fungal biomass after incubation of Fusarium culmorum Fc13, Fusarium graminearum Fg23, Alternaria tenuissima At18 and A. tenuissima At220 after consecutive inoculation with a time delay of 3 days on wheat kernels at day 6, 9 and 12 in cocultures and as the sum of the corresponding individual cultures. Mean values and standard deviation (n = 5). Asterisks indicate significant differences (P < 0·05) between the mean values of ergosterol in coculture and in the two corresponding individual cultures summed up. image Co-culture; image sum of single cultures.

Overall, the second fungal partner in cocultures was barely able to produce the species-specific mycotoxins: If Fc13 and Fg23 were inoculated before the Alternaria strains, only 1·1 and 0·003% AOH were produced at day 12 by At18 and At220, respectively, compared with the mycotoxin production in individual cultures (Table 2). AME and ALT were not detected in these cocultures at day 12. It is remarkable that the Fc13 isolate enhanced its DON production at day 12 up to 172 and 169% in competition with the consecutively inoculated At18 and At220, respectively. The ZON production by both fusaria was slightly increased in combination with At18 at day 12 (123 and 150% by Fc13 and Fg23, respectively), but slightly repressed in combination with At220 (Table 2).

Table 2. Mean values of alternariol (AOH), alternariol monomethyl ether (AME), altenuene (ALT), deoxynivalenol (DON), zearalenone (ZON) and nivalenol (NIV) concentrations and their standard deviations in individual and cocultures of Fusarium culmorum Fc13 and Fusarium graminearum Fg23 (firstly inoculated) as well as Alternaria tenuissima At18 and A. tenuissima At220 (secondly inoculated) after consecutive inoculation with a time delay of 3 days and a total incubation on wheat kernels of 6, 9 and 12 days, respectively, as well as the differences (P-values) between the mycotoxin contents in individual and cocultures
 Day 6 (6 + 3)aDay 9 (9 + 6)aDay 12 (12 + 9)a
Fc13 + At18bFc13 + At220Fg23 + At18Fg23 + At220Fc13 + At18Fc13 + At220Fg23 + At18Fg23 + At220Fc13 + At18Fc13 + At220Fg23 + At18Fg23 + At220
  1. a

    Total incubation time and (in brackets) the number of incubation days of the firstly + secondly inoculated strain.

  2. b

    n = 5 for each strain combination.

  3. Significant differences between mean values at P level <0·05 are highlighted in bold.

AOH (μg μg−1 ergosterol)
Coculture00000·001 ± 0·0010000·001 ± 0·0020·0002 ± 0·000400
Individual culture00·85 ± 0·3900·85 ± 0·390·16 ± 0·1430·6 ± 21·80·16 ± 0·1430·6 ± 21·80·12 ± 0·096·2 ± 2·80·12 ± 0·096·2 ± 2·8
P-value 0·005 0·005 0·008 0·005 0·005 0·005 0·008 0·007 0·005 0·005
AME (μg μg−1 ergosterol)
Coculture0000·003 ± 0·00300000000
Individual culture00·08 ± 0·0400·08 ± 0·040·08 ± 0·064·2 ± 3·30·08 ± 0·064·2 ± 3·30·08 ± 0·092·0 ± 1·20·08 ± 0·092·0 ± 1·2
P-value 0·005 0·016 0·005 0·005 0·005 0·005 0·005 0·005 0·005 0·005
ALT (μg μg−1 ergosterol)
Coculture0000·0002 ± 0·00040000·001 ± 0·0010000
Individual culture00·0004 ± 0·000600·0004 ± 0·000602·1 ± 1·402·1 ± 1·40·05 ± 0·092·4 ± 1·30·05 ± 0·092·4 ± 1·3
P-value0·1360·541 0·005 0·014 0·054 0·005 0·054 0·005
DON (μg μg−1 ergosterol)
Coculture0·02 ± 0·040·09 ± 0·06000·34 ± 0·430·23 ± 0·12000·64 ± 0·590·62 ± 0·7900
Individual culture0·02 ± 0·020·02 ± 0·02000·26 ± 0·330·26 ± 0·33000·37 ± 0·280·37 ± 0·2800
P-value0·837 0·027 0·7460·8840·3860·517
ZON (μg μg−1 ergosterol)
Coculture0·34 ± 0·160·37 ± 0·090·12 ± 0·110·04 ± 0·0511·5 ± 6·09·0 ± 1·51·8 ± 1·31·1 ± 0·3518·4 ± 19·78·1 ± 3·95·8 ± 4·03·0 ± 1·7
Individual culture0·23 ± 0·10·23 ± 0·10·04 ± 0·050·04 ± 0·0510·2 ± 9·810·2 ± 9·80·6 ± 0·440·6 ± 0·4415·0 ± 5·715·0 ± 5·73·9 ± 2·23·9 ± 2·2
P-value0·199 0·039 0·1730·910·810·6020·1360·2020·7210·0560·3660·51
NIV (μg μg−1 ergosterol)
Not detected in any of the samples

If both Alternaria strains were inoculated before the Fusarium strains, DON was produced only in a very small extent by Fc13 inoculated after At18 (0·5% at day 12) or not at all when inoculated after At220 (Table 3). ZON was detected at day 12 in all four combinations in a reduced rate: 22 and 1·6% from Fc13 in combination with At18 and At220, respectively, and 0·8 and 5% from Fg23 in combination with At18 and At220. The both firstly inoculated Alternaria strains remarkably increased their species-specific mycotoxin production in the competition with the consecutively inoculated fusaria (Table 3): up to 350% AOH, up to 560% AME and up to 4765% ALT (in combination of At18 + Fg23 at day 12) compared with the toxin content in the corresponding individual cultures.

Table 3. Mean values of alternariol (AOH), alternariol monomethyl ether (AME), altenuene (ALT), deoxynivalenol (DON), zearalenone (ZON) and nivalenol (NIV) concentrations and their standard deviations in individual and co-cultures of Alternaria tenuissima At18 and A. tenuissima At220 (firstly inoculated) as well as Fusarium culmorum Fc13 and Fusarium graminearum Fg23 (secondly inoculated) after consecutive inoculation with a time delay of 3 days and a total incubation on wheat kernels of 6, 9 and 12 days, respectively, as well as the differences (P-values) between the mycotoxin contents in individual and cocultures
 Day 6 (6 + 3)aDay 9 (9 + 6)aDay 12 (12 + 9)a
At18 + Fc13bAt18 + Fg23At220 + Fc13At220 + Fg23At18 + Fc13At18 + Fg23At220 + Fc13At220 + Fg23At18 + Fc13At18 + Fg23At220 + Fc13At220 + Fg23
  1. a

    Total incubation time and (in brackets) the number of incubation days of the firstly + secondly inoculated strain.

  2. b

    n = 5 for each strain combination.

  3. Significant differences between mean values at P level <0·05 are highlighted in bold.

AOH (μg μg−1 ergosterol)
Coculture00·1 ± 0·0526·1 ± 12·227·1 ± 24·20·26 ± 0·180·15 ± 0·1119·3 ± 6·711·6 ± 10·30·12 ± 0·110·3 ± 0·3511·6 ± 5·04·3 ± 2·5
Individual culture0·16 ± 0·140·16 ± 0·1430·6 ± 21·830·6 ± 21·80·12 ± 0·090·12 ± 0·096·2 ± 2·86·2 ± 2·80·09 ± 0·040·09 ± 0·046·2 ± 3·26·2 ± 3·2
P-value 0·005 0·5380·6970·8150·1470·575 0·005 0·9170·5210·3470·0750·326
AME (μg μg−1 ergosterol)
Coculture00·07 ± 0·053·1 ± 1·92·6 ± 2·00·23 ± 0·20·1 ± 0·085·1 ± 4·64·4 ± 4·20·08 ± 0·090·36 ± 0·483·5 ± 1·31·2 ± 0·6
Individual culture0·08 ± 0·060·08 ± 0·064·2 ± 3·34·2 ± 3·30·08 ± 0·090·08 ± 0·092·0 ± 1·22·0 ± 1·20·06 ± 0·070·06 ± 0·071·7 ± 1·01·7 ± 1·0
P-value 0·005 0·740·5610·4030·1750·7020·2510·6020·7230·251 0·038 0·393
ALT (μg μg−1 ergosterol)
Coculture001·8 ± 0·82·3 ± 1·50·01 ± 0·010·004 ± 0·0063·9 ± 2·42·9 ± 1·40·003 ± 0·0060·11 ± 0·234·3 ± 1·93·1 ± 1·8
Individual culture002·1 ± 1·42·1 ± 1·40·05 ± 0·090·05 ± 0·092·4 ± 1·32·4 ± 1·30·002 ± 0·0040·002 ± 0·0042·8 ± 1·72·8 ± 1·7
P-value0·6270·8910·3380·4350·2540·5750·8460·0850·2440·83
DON (μg μg−1 ergosterol)
Coculture00000·007 ± 0·0160000·001 ± 0·003000
Individual culture00000·023 ± 0·01600·023 ± 0·01600·26 ± 0·3300·26 ± 0·330
P-value0·198 0·005 0·007 0·005
ZON (μg μg−1 ergosterol)
Coculture00000·045 ± 0·0450002·3 ± 2·20·005 ± 0·0070·17 ± 0·260·03 ± 0·07
Individual culture00000·23 ± 0·10·04 ± 0·050·23 ± 0·10·04 ± 0·0510·2 ± 9·80·6 ± 0·4410·2 ± 9·80·6 ± 0·44
P-value 0·005 0·054 0·005 0·0540·076 0·013 0·009 0·021
NIV (μg μg−1 ergosterol)
Not detected in any of the samples

It should be noted that the five replicates made for each combination of fungal strains often showed a high standard deviation, thus obstructing the statistical validation of differences between the mycotoxin contents in individual and cocultures.

Discussion

Until now, very little is known about interaction mechanisms between fungi belonging to the phytopathogenic genera Fusarium and Alternaria in spite of the proof that both genera are common habitants on grain ears. The overall conclusion to be derived from this study is that Fusarium and Alternaria strains interact in in vitro cocultivation and their growth as well as their mycotoxin production are considerably influenced. Therefore, it can be assumed that the co-occurrence of both fungi in planta would lead to a strong competitive relationship and an independent coexistence of both these fungal genera seems to be very unlikely. Our findings are in line with the results of Andersen et al. (1996) and Kosiak et al. (2004), who indicated an interaction between Fusarium and Alternaria on the surface of malt barley kernels in Denmark and in grains (wheat, barley, oats) of reduced and of acceptable quality in Norway, respectively. Both studies demonstrated an inverse relationship regarding the relative abundance of Fusarium spp. and Alternaria spp. The infection level of Fusarium spp. having an effect against Alternaria spp. was fairly different in both studies. The authors discussed the influence of the F. species composition and suggested that F. graminearum had a stronger effect as competitor.

Laboratory studies evaluated the competitive effects between A. sp. and F. graminearum (Riungu et al. 2007) as well as between A. alternata and F. culmorum (Liggitt et al. 1997) on artificial media and demonstrated that the Alternaria strains considerably reduced colony diameters of mycelial growth of the Fusarium isolates. Liggitt et al. (1997) argued that this antagonism may have been due at least partly to production of nonvolatile and/or volatile substances by A. alternata. Such an antibiosis mediated by volatile production has only been detected in a closed environment, and it is very unlikely to occur under field conditions. In contrast to these results, Saß et al. (2007) hypothesized that A. alternata was significantly suppressed by F. graminearum when the fungi were growing together in liquid cultures in vitro.

In our study, the growth rate (measured by quantifying the ergosterol content) of all fungal combinations in simultaneous coculture was significantly depressed at day 4 suggesting an initial strong competition. At day 12, however, the fungal biomass in cocultures is greatly enhanced (up to 460%) compared with the growth in individual culture. A comprehensive explanation of these effects is hampered by the inability of the observer to differentiate which fungal strain mostly contributes to this enhanced fungal growth. Further experiments should involve molecular genetic techniques to discriminate between the fungal partners in cocultures. Saß et al. (2007), using a semi-quantitative real-time PCR to differentiate between F. graminearum- and A. alternata-DNA concentrations in cocultivation, emphasized a significant suppression of A. growth, but did not mention any increasing growth effects.

Burgess et al. (1988) investigated the influence of high growth rates on the distribution of different F. strains and demonstrated this property to be a conspicuous advantage in antagonistic interactions. We could not confirm this thesis because the isolate with the highest growth rate in individual culture, At220, did not yield the highest increase in ergosterol content in cocultures with Fc13 and Fg23. In contrast, the cocultures of Fg23 with both A. isolates showed the highest raise in ergosterol. This fact implies that other mechanisms than high growth rates are involved in interfungal competition. F. graminearum strains were assumed to be highly aggressive plant pathogens (Champeil et al. 2004; Korn et al. 2011), probably based on enzyme activities, ascospore production and metabolite profiling. However, to our knowledge, similar enhancement of fungal growth in cocultures has never been described before in the literature.

The presence of one well-developed fungal strain before inoculation (after 3 days) with a second fungal strain restricted the development of the second strain remarkably in all combinations. Hence, it appears as if a head start in growth allows a fungal strain to successfully suppress competitors. This might also be of relevance to the antagonism processes in planta: Liggitt et al. (1997) detected reduced severity of FHB if glasshouse grown plants were inoculated with A. alternata before inoculation with F. culmorum.

Fusarium culmorum and F. graminearum as well as A. tenuissima used in the present study were aggressive colonizers of wheat ears and well known producers of mycotoxins in grains (Korn et al. 2011). Under natural field conditions, the ripening grain ears are colonized by several saprophytic and pathogenic fungi (Dawood 1982; D'Mello et al. 1993). Many studies described the effects of biological control agents such as Trichoderma spp., yeasts or Clonostachys rosea on sporulation, disease incidences and mycotoxin production of pathogenic Fusarium strains (Harman 2000; Dawson et al. 2004; Luongo et al. 2005). Otherwise, fusaria compete with further pathogens in their environment and can suppress other mycotoxigenic fungi such as Aspergillus spp. and Penicillium spp. as well as other Fusarium spp. (Jones et al. 1997). Within the genus Fusarium, some nonpathogenic or nontoxigenic species, particularly F. equiseti, were found to be the most effective antagonists against F. culmorum and F. graminearum (Dawson et al. 2004). Overall, fungal interactions in a natural environment appear as a complex process that involves principles like cohabitation, inhibition or antagonism. The prevailing micro-organism could benefit from its higher growth rates, from its aggressiveness or from the production of toxic metabolites. Mycotoxins are considered to be involved in plant pathogenesis and have a significant effect on the aggressiveness of the producing strain (Desmond et al. 2008; Miedaner et al. 2008; Scherm et al. 2011; Sanzani et al. 2012). Furthermore, Fusarium mycotoxins as well as Alternaria mycotoxins are cytotoxic, antiviral, antibacterial and/or antifungal metabolites (Prelusky et al. 1994; Visconti and Sibilia 1994).

To our knowledge, this is the first study, in which the majority of mycotoxins produced by Fusarium strains (DON, ZON, NIV) as well as Alternaria strains (AOH, AME, ALT) were determined. Therefore, we can emphasize that the production of AOH, AME and ALT by both A. tenuissima strains was considerably depressed in simultaneous cocultures with Fc13 as well as with Fg23.

Although the isolates At220 and At18 differed in their toxin production (e.g. At18 produced 323 ng AOH μg−1 ergosterol, At220 17 478 ng AOH μg−1 ergosterol), both strains reduced their whole toxin production in simultaneous cocultures with the strong ZON producer Fg23 down to <5% at day 8. Vice versa, Fc13 (DON and ZON producer) increased its ZON production in competition against At220 as well as against At18. The results are in accordance with the findings of Saß et al. (2007), and we can emphasize that ZON might play an important role in the competitive interaction between mycotoxigenic fungi. Utermark and Karlovsky (2007) found a further evidence for this hypothesis: ZON and its derivatives suppressed the growth of different filamentous fungi (amongst others also A. alternata) on Petri dishes filled with a solid medium supplemented with ZON at a concentration of 20 μg ml−1 and, in general, could act as an antifungal but not as an antibacterial substance.

Several in vitro and in vivo studies have described that the DON production of fusaria could be affected by competing micro-organisms (González et al. 1999; Cooney et al. 2001; Dawson et al. 2004; Simpson et al. 2004; Saß et al. 2007; He et al. 2009; Siebert 2011). Most of them reported about a repression of the DON amounts produced by F. culmorum or F. graminearum when F. subglutinans, F. equiseti, Microdochium nivale, A. alternata, Paenibacillus polymyxa or Trichoderma spp., respectively, were present in bioassay tests.

González et al. (1999) summarized a general negative relationship between DON contamination and A. alternata densities on Argentinian durum wheat. Surprisingly, the DON producing isolate Fc13 used in this study enhanced its DON accumulation in the simultaneous cocultures with the competing At220 and At18: six-fold amounts of DON against At18 and 11-fold amounts against At220, compared with the individual cultures, respectively. However, until day 8, the DON production by Fc13 was still remarkably inhibited: completely when At220 was present and down to only 11% when At18 was present. An unexpectedly heavy increase in DON accumulation only occurred between the 8th and 12th day of incubation. In the above-mentioned studies, most tests already finished after 4, 6 or 7 days and this short experimental time period might explain the different results with respect to DON. Nevertheless, Cooney et al. (2001) also described an increase in DON by F. graminearum tested vs a F. avenaceum isolate.

Overall, several investigations hypothesized that ZON has a biological function in the competitive interaction and inhibits the growth and toxin production of other fungi. In contrast, DON does not play a crucial role in the life-or-death struggle. The study presented here could generally confirm these facts, showing at once strain-specific differences and time-dependent variance in the mycotoxin production during the competitive interaction between mycotoxigenic fungi.

Acknowledgements

The excellent technical assistance of Martina Peters and Grit von der Waydbrink is gratefully acknowledged.

Ancillary