Identification of gene clusters associated with fusaric acid, fusarin, and perithecial pigment production in Fusarium verticillioides

Abstract

The genus Fusarium is of concern to agricultural production and food/feed safety because of its ability to cause crop disease and to produce mycotoxins. Understanding the genetic basis for production of mycotoxins and other secondary metabolites (SMs) has the potential to limit crop disease and mycotoxin contamination. In fungi, SM biosynthetic genes are typically located adjacent to one another in clusters of co-expressed genes. Such clusters typically include a core gene, responsible for synthesis of an initial chemical, and several genes responsible for chemical modifications, transport, and/or regulation. Fusarium verticillioides is one of the most common pathogens of maize and produces a variety of SMs of concern. Here, we employed whole genome expression analysis and utilized existing knowledge of polyketide synthase (PKS) genes, a common cluster core gene, to identify three novel clusters of co-expressed genes in F. verticillioides. Functional analysis of the PKS genes linked the clusters to production of three known Fusarium SMs, a violet pigment in sexual fruiting bodies (perithecia) and the mycotoxins fusarin C and fusaric acid. The results indicate that microarray analysis of RNA derived from culture conditions that induce differential gene expression can be an effective tool for identifying SM biosynthetic gene clusters.

Introduction

Fungal secondary metabolite (SM) biosynthetic genes are typically located adjacent to each other (e.g. in a cluster) and tend to exhibit similar patterns of expression (Keller et al., 2005, Shwab and Keller, 2008). This tightly controlled regulation has been suggested as a main selective force for gene clustering (Khaldi and Wolfe, 2011). SM gene clusters usually include a core synthase gene responsible for synthesis of the SM parent chemical, genes responsible for chemical modifications (e.g. oxidoreduction, acyl or amino transfer, and dehydrogenation), one or more genes involved in transport of the SMs, and one or more genes that regulate cluster gene transcription. Most core synthase genes encode either a polyketide synthase (PKS), non-ribosomal peptide synthetase (NRPS), or terpene cyclase (TC). Multiple methods have been used to identify SM biosynthetic gene clusters including over expression of transcriptional regulators located proximal to core genes, heterologous expression of the core synthase gene, altering cluster transcription by modifying chromatin structure (Bok et al., 2009, Chiang et al., 2009, Watanabe et al., 1999) and the bioinformatic (e.g. SMURF or Secondary Metabolite Unknown Regions Finder) identification of multiple SM genes located adjacent to a core gene by homology (Khaldi et al., 2010). The number of core synthase genes per fungal genome (Ma et al., 2010, Yoder and Turgeon, 2001) suggest that many fungal species can produce up to 10 fold more SMs than indicated by chemical analysis. In accordance with this genetic potential, fungi in the genus Fusarium have been found to produce a wide variety of SMs that are diverse in structure and biological activity. The most thoroughly studied Fusarium SMs are the mycotoxins, fumonisins, trichothecenes, and zearalenone. These SMs pose a threat to human and animal health because consumption of contaminated grain or grain-based food and feed has been associated with a variety of diseases (Glenn, 2007, Morgavi and Riley, 2007). The economic impact of these mycotoxins on health costs and effect on international trade is estimated to be in the 100 s of millions of dollars each year (Wu, 2007).

Fusarium verticillioides is one of the most common maize-associated fungi worldwide and can infect and colonize maize as an endophyte (i.e. without causing disease symptoms) or as a pathogen (i.e. causing symptoms). During growth on maize, the fungus can produce fumonisins and a number of other polyketide-derived SMs, including the toxins fusaric acid (5-butylpicolinic acid) and fusarins, the pigment bikaverin, and an uncharacterized violet pigment present in the walls of sexual fruiting bodies (perithecia). Production of these metabolites has been reported in other, but not all, Fusarium species as well as in other fungal genera. For example, fumonisin production has been reported in several Fusarium species within the Gibberella fujikuroi species complex (GFSC), in rare isolates of the closely related species Fusarium oxysporum (Proctor et al., 2004), and in Aspergillus niger (Frisvad et al., 2007) and Tolypocladium (Mogensen et al., 2011). In contrast, fusarin production occurs across a genetically broad range of Fusarium species including multiple GFSC and trichothecene-producing species (Desjardins and Proctor, 2001; Proctor et al., unpublished).

The role of these SMs in the ecology of Fusarium is not clear. However, they do exhibit varying levels of toxicity to plants and animals. For example, fumonisins affected maize sphingolipid synthesis and pathogenesis of F. verticillioides in some maize seedling assays (Glenn et al., 2008, Myung et al., 2011). However, they did not affect pathogenesis in other maize seedling assays or in an ear rot assay (Desjardins et al., 2007, Desjardins et al., 2002). Fumonisin consumption can cause a variety of diseases in animals and is epidemiologically associated with esophageal cancer and neural tube defects in humans (Gonzalez et al., 1999, Marasas et al., 2004). Fusaric acid is a broad-spectrum plant toxin and is thought to contribute to the severity of F. oxysporum-induced vascular wilt, damping-off and root rot diseases of numerous vegetable crops (Capasso et al., 1996). Fusaric acid is considered a mild toxin to mammals (Bacon et al., 1996). In contrast, fusarins have been reported to be mutagenic in the Ames Salmonella mutagenicity assay (Gelderblom et al., 1983, Gelderblom et al., 1984) but not carcinogenic in mouse and rat model systems (Gelderblom et al., 1986). Recently, it has been reported that fusarin C stimulates growth of a breast cancer cell suggesting that it acts as an estrogenic mimic (Sondergaard et al., 2011). Initial studies with fusarin-nonproducing F. verticillioides strains in maize seedling and ear rot assays indicate that they are not required for pathogenesis (Brown and Proctor, unpublished). Fusarium pigments (e.g. the naphthoquinones bikaverin and fusarubin) are best known for the protective role they provide fungi against environmental stresses such as bacteria and plants (Medentsev and Akimenko, 1998). Overall, fumonisins, fusarins, fusaric acid and bikaverin may contaminate maize and maize-based food and feed and as a result, pose a risk to human and livestock health.

Examination of the genetics of Fusarium SM biosynthesis is motivated by the need to limit mycotoxin contamination of agricultural products and to understand how SMs impact Fusarium-induced plant disease. Analysis of F. verticillioides genome sequence identified 16 genes predicted to encode PKSs (Kroken et al., 2003, Ma et al., 2010). The functions of two of these genes have been assigned based on studies in F. verticillioides: FUM1 (PKS11) is required for fumonisin production (Proctor et al., 1999) and PGL1 (PKS3) is required for production of the violet perithecial pigment (Proctor et al., 2007). The function of two more have been assigned based on studies in closely related Fusarium species: PKS4 is required for bikaverin production (Fusarium fujikuroi) (Wiemann et al., 2009) and PKS10 is required for fusarin C production (Fusarium moniliformis and Fusarium venenatum) (Song et al., 2004) and Fusarium graminearum (Gaffoor et al., 2005). To date, associated gene clusters have been identified for two of the PKS genes: FUM1 (Proctor et al., 2003) and BIK1 (Brown et al., 2008) based on gene expression and/or gene inactivation analysis.

The identification of multiple SMs that are likely derived from polyketides combined with the identification of the remaining PKS genes in the F. verticillioides genome (Desjardins and Proctor, 2007) provides an excellent opportunity to identify additional PKS gene-containing clusters and link them to production of specific SMs. The overall goal is to better understand the role of SMs in the life of F. verticillioides; in particular, with regards to the maize disease process. Thus, in the current study, we employed microarray analysis to identify clusters of co-expressed genes that include a PKS gene. The analysis confirmed the existence of the two previously described biosynthetic gene clusters and provided evidence for three heretofore uncharacterized clusters. In addition, results of functional analysis conducted here and previous studies linked the three novel clusters to production of the SMs fusarins, the violet perithecial pigment, and fusaric acid. The identification of these clusters will facilitate future analyses of whether the SMs have a role in plant pathogenesis.