In vitro cytotoxicity of fungi spoiling maize silage

Abstract

Penicillium roqueforti, Penicillium paneum, Monascus ruber, Alternaria tenuissima, Fusarium graminearum, Fusarium avenaceum, Byssochlamys nivea and Aspergillus fumigatus have previously been identified as major fungal contaminants of Danish maize silage. In the present study their metabolite production and in vitro cytotoxicity have been determined for fungal agar and silage extracts. All 8 fungal species significantly affected Caco-2 cell viability in the resazurin assay, with large variations for each species and growth medium. The 50% inhibition concentrations (IC₅₀) of the major P. roqueforti metabolites roquefortine C (48 μg/mL), andrastin A (>50 μg/mL), mycophenolic acid (>100 μg/mL) and 1-hydroxyeremophil-7(11),9(10)-dien-8-one (>280 μg/mL) were high. Fractionating of agar extracts identified PR-toxin as an important cytotoxic P. roqueforti metabolite, also detectable in maize silage. The strongly cytotoxic B. nivea and P. paneum agar extracts contained patulin above the IC₅₀ of 0.6 μg/mL, however inoculated onto maize silage B. nivea and P. paneum did not produce patulin (>371 μg/kg). Still B. nivea infected maize silage containing mycophenolic acid (∼50 mg/kg), byssochlamic acid and other metabolites, was cytotoxic. In contrast hot-spots of P. roqueforti, P. paneum, M. ruber and A. fumigatus were not more cytotoxic than uninoculated silage.

Introduction

Farmers all over the world produce maize silage to feed dairy cows (Wilkinson and Toivonen, 2003). Maize silage may constitute 50–75% of the diet (Driehuis et al., 2008b) for a dairy cow consuming approximately 26 kg dry matter/day (Eastridge, 2006). Maize plants are converted into maize silage as a result of many naturally occurring enzymatic and microbiological processes taking place when chopped plant material is compressed and packed airtight. A natural lactic acid fermentation of maize sugars into organic acids enables anaerobic and acidic storage of the maize silage. Long-term storage is possible as a well-managed maize silage stack is a very hostile growth environment to most microorganisms (Storm et al., 2010). Nevertheless specific filamentous fungi are known to spoil maize plants in the field or silage during the storage period. The fungi are able to produce many secondary metabolites including mycotoxins. Mycotoxin exposure can affect dairy cows health (Korosteleva et al., 2009) and productivity (Fink-Gremmels, 2008b). The most important toxigenic genera associated with maize and silage are Aspergillus, Fusarium, Alternaria, Penicillium, Byssochlamys and Monascus (Storm et al., 2008).

The fungal metabolite production depends on fungal species (Frisvad et al., 2008), isolate (O’Brien et al., 2006, Andersen et al., 2008, Frisvad et al., 2009), growth medium and environmental factors (Frank, 1998, Furtado et al., 2002). The pre-harvest secondary fungal metabolites in maize silage includes; aflatoxin B₁, alternariol, alternariol monomethyl ether, beauvericin, deoxynivalenol, 15-acetyl-deoxynivalenol, enniatin B and B₁, fumonisin B₁, nivalenol and zearalenone. From post-harvest spoilage of maize silage andrastin A, citreoisocoumarin, citrinin, cyclopiazonic acid, fumigaclavine A, gliotoxin, marcfortine A and B, mycophenolic acid, patulin, PR-toxin, roquefortine A and C have been detected (Müller and Amend, 1997, Garon et al., 2006, Richard et al., 2007, Driehuis et al., 2008a; Mansfield et al., 2008, Sørensen et al., 2008, Rasmussen et al., 2010). Many of those metabolites are mycotoxins, which can elicit carcinogenic, mutagenic, neurotoxic, hepatotoxic, nephrotoxic, oestrogenic, immunosuppressive, antimicrobial (Scudamore and Livesey, 1998) or acute toxic effects (Chen et al., 1982). The symptoms identified in animal trails include feed refusal, kidney, liver or lung damages, birth defects, abortion and death (Scudamore and Livesey, 1998). A chronic exposure to low levels of mycotoxins typically gives non-specific symptoms such as impaired immune system and increased infections or metabolic and hormonal imbalances (Morgavi and Riley, 2007, Fink-Gremmels, 2008b). To protect animal health some countries have recommendations for deoxynivalenol, ochratoxin A, fumonisins, zearalenone content in feed (European Commission, 2006). The transfer of toxins to dairy and meat products is a potential risk for humans (Miller, 2008, Fink-Gremmels, 2008a) and the regulation on aflatoxins B₁ in feed (European Commission, 2003) owes to transfer of the carcinogenic aflatoxin M₁ metabolite to milk (IARC, 1993).

Compared to other animals ruminants are more robust to many mycotoxins (EFSA, 2004a, EFSA, 2004c, EFSA, 2005), partly due to biotransformation by the rumen microorganisms (He et al., 1992). The rumen microbiota inactivates and degrades some mycotoxins, but not all types whereas others are metabolised to the even more potent compounds in the rumen. For example, ochratoxin A is extensively degraded to the less toxic ochratoxin α (EFSA, 2004b), fumonisin B₁ is unaffected in the rumen (EFSA, 2005) whereas zearalenone is metabolised to α-zearalenol which has stronger oestrogenic effect (EFSA, 2004c). Antimicrobial fungal metabolites such as patulin (Tapia et al., 2002), mycophenolic acid (Bentley, 2000), citrinin (Wang et al. 2004) and roquefortine C (Kopp and Rehm, 1979) can affect rumen microorganisms (Tapia et al., 2002). An impaired rumen function cause severe metabolic disorders, which can reduce feed utilization (Chiquette, 2009) and may increase the mycotoxin uptake (Fink-Gremmels, 2008a). Cases of ill-thrift, disease and death in livestock have been related to the presence of mycotoxins in silage and the issue is much debated (Storm et al., 2008). Especially high-yielding dairy cows may be susceptible to diseases caused by mycotoxins, due to a high level of stress (Jouany and Diaz, 2005) but acute intoxications causing death are rare (Yiannikouris and Jouany, 2002). Actually was the occurrence of 20 mycotoxins in feedstuffs for dairy cows low compared to the effect concentrations of the individual toxins in a maize silage based diet based (Driehuis et al., 2008a). However fungi are capable of producing many bioactives (Samson et al., 2002) and simultaneous exposure to several toxins could elicit synergism (Bouslimi et al., 2008).

In vitro testing systems are a good screening tool for toxicological effects (Gutleb et al., 2002). Cell cultures of yeast, mammalian cells or bacteria are typically applied. Compared to animal studies in vitro assays are fast and cheap, though they may indeed give different results than animal studies, due to lack of an integrated organism response (Gad, 2000). Cytotoxicity assays can to some extent be used as a screening test for acute toxicity in animals and humans (Binderup et al., 2002). The human intestinal epithelial cell line (Caco-2) is widely used and well validated (Videmann et al., 2008). Metabolic conversion of dye by viable Caco-2 cells in vitro can determine the general cytotoxicity with similar sensitivity as many other cell lines (Cetin and Bullerman, 2005).

In the present study an in vitro cytotoxicity assay is used in combination with chemical analysis and bio-directed fractionation to identify important toxic mycotoxins in mixtures of unknown composition. For this purpose concentration–response curves were made for a range of known mycotoxins. In vitro cytotoxicity tests of fungal agar extracts and silage extracts have been carried out along with chemical identification using liquid chromatography with diode array and mass spectrometry detection. The viability of Caco-2 cells was determined from their metabolic conversion of resazurin dye (Binderup et al., 2002). Filamentous fungi often isolated from Danish maize (Fusarium graminearum, Fusarium avenaceum and Alternaria tenuissima) and maize silage (Aspergillus fumigatus, Monascus ruber, Penicillium roqueforti, Penicillium paneum and Byssochlamys nivea) were included.

The aims were: (i) to relate the cytotoxicity of well known mycotoxins with their presence in toxic fungal agar extracts, (ii) to determine the toxicity and presence of metabolites of inoculated maize silage, (iii) to identify the most cytotoxic compound in a crude P. roqueforti agar extract. The present study is a part of a large Danish collaborative project aiming to determine if mycotoxins in maize silage cause disease and poor performance in dairy cattle (Kristensen et al., 2007, Sørensen, 2009, Storm, 2009).