Verticillin A is likely not produced by Verticillium sp.

Verticillin A is a fungal epipolythiodioxopiperazine (ETP) metabolite with antibiotic properties, which was originally reported to be produced from Verticillium sp.¹ It is structurally very similar to Chaetocin from Chaetomium minutum, differing only in the position of two hydroxyl groups (Figure 1). Chaetocin is the first described histone methyltransferase inhibitor,² and may therefore influence chromatin remodeling. As many Verticillium sp. are known phytopathogens,³ and Cochliobolus carbonum has been shown to produce the histone deacetylase inhibitor HC-toxin as a virulence factor,⁴ it is possible that Verticillin A might constitute another fungal means to interfere with the host's chromatin remodeling machinery for enhancing pathogenesis. In fact, two other ETPs, Gliotoxin (from Aspergillus fumigatus) and Sirodesmin (from Leptosphaeria maculans), have been identified as virulence factors.^{5, 6} Thus, a search for Verticillium strains producing Verticillin A was initiated to disrupt the genes for toxin production and assess its impact on pathogenicity. Genes involved in ETP biosynthesis are reported to be organized in clusters,⁷ so degenerate primers were designed to amplify genes encoding a aminocyclopropane carboxylic acid synthase, thioredoxin reductase, glutathione S-transferase and methyl transferase from genomic DNA of 39 Verticillium strains. No results were obtained from this approach, and also Southern blot analysis of these DNAs with a heterologous thioredoxin reductase probe derived from the corresponding Leptoshaeria maculans gene did not result in detection of any signal, except in the positive control. A parallel approach to isolate Verticillin A from the Verticillium dahliae strain ST37.01 grown in different media and under different temperatures was also not successful, whereas medium spiked with commercially available Verticillin A (produced from a fungus of the genus Paecilomyces; Iris Biotech, Germany) before purification served as internal positive control in our LC–MS analysis. After the release of the genome sequence of the two Verticillium strains V. dahliae VdLs17 and V. albo-altrum VaMs102 by the Broad Institute, a search for the ETP cluster was performed in silico. However, such a gene cluster could not be detected, and hence corroborates the failure to amplify/clone such genes from 39 independent samples described above. This questions the ability of Verticillium sp. to produce Verticillin A, because they seem to lack the necessary genes. As the genus Verticillium was known to be heterogenous,⁸ it is possible that the fungal strain reported to produce Verticillin A might belong actually to a different genus. Unfortunately we were unable to obtain this original strain to test this hypothesis. Interestingly, Verticillin A has been shown to be produced by Clonostachys rosea, formerly known as Gliocladium roseum, and this fungus is not only morphologically similar to, but also a mycoparasite of Verticillium.^{9, 10} Taken together, it is conceivable that the original Verticillium sample was either confused with another fungus or contaminated with Clonostachys rosea. Thus, the designation of the isolated ETP as Verticillin A might have been not appropriate.

Figure 1

Structural comparison of four epipolythiodioxopiperazine (ETP) secondary metabolites. (a) Chaetocin from Chaetomium minutum, (b) Verticillin A from Gliocladium roseum, (c) Sirodesmin from Leptosphaeria maculans and (d) Gliotoxin from Aspergillus fumigatus.