Profiling and structural elucidation of secondary metabolites produced by the filamentous fungus Penicillium chrysogenum and derived deletion strains were used to identify the various metabolites and enzymatic steps belonging to the roquefortine/meleagrin pathway. Major abundant metabolites of this pathway were identified as histidyltryptophanyldiketopiperazine (HTD), dehydrohistidyltryptophanyldi-ketopiperazine (DHTD), roquefortine D, roquefortine C, glandicoline A, glandicoline B and meleagrin. Specific genes could be assigned to each enzymatic reaction step. The nonribosomal peptide synthetase RoqA accepts L-histidine and L-tryptophan as substrates leading to the production of the diketopiperazine HTD. DHTD, previously suggested to be a degradation product of roquefortine C, was found to be derived from HTD involving the cytochrome P450 oxidoreductase RoqR. The dimethylallyltryptophan synthetase RoqD prenylates both HTD and DHTD yielding directly the products roquefortine D and roquefortine C without the synthesis of a previously suggested intermediate and the involvement of RoqM. This leads to a branch in the otherwise linear pathway. Roquefortine C is subsequently converted into glandicoline B with glandicoline A as intermediates, involving two monooxygenases (RoqM and RoqO) which were mixed up in an earlier attempt to elucidate the biosynthetic pathway. Eventually, meleagrin is produced from glandicoline B involving a methyltransferase (RoqN). It is concluded that roquefortine C and meleagrin are derived from a branched biosynthetic pathway.

BackgroundPenicillium chrysogenum is a filamentous fungus that is employed as an industrial producer of β–lactams. The high β–lactam titers of current strains is the result of a classical strain improvement program (CSI) starting with a wild-type like strain more than six decades ago. This involved extensive mutagenesis and strain selection for improved β–lactam titers and growth characteristics. However, the impact of the CSI on the secondary metabolism in general remains unknown.ResultsTo examine the impact of CSI on secondary metabolism, a comparative genomic analysis of β-lactam producing strains was carried out by genome sequencing of three P. chrysogenum strains that are part of a lineage of the CSI, i.e., strains NRRL1951, Wisconsin 54-1255, DS17690, and the derived penicillin biosynthesis cluster free strain DS68530. CSI has resulted in a wide spread of mutations, that statistically did not result in an over- or underrepresentation of specific gene classes. However, in this set of mutations, 8 out of 31 secondary metabolite genes (20 polyketide synthases and 11 non-ribosomal peptide synthetases) were targeted with a corresponding and progressive loss in the production of a range of secondary metabolites unrelated to β–lactam production. Additionally, key Velvet complex proteins (LeaA and VelA) involved in global regulation of secondary metabolism have been repeatedly targeted for mutagenesis during CSI. Using comparative metabolic profiling, the polyketide synthetase gene cluster was identified that is responsible for sorbicillinoid biosynthesis, a group of yellow-colored metabolites that are abundantly produced by early production strains of P. chrysogenum.ConclusionsThe classical industrial strain improvement of P. chrysogenum has had a broad mutagenic impact on metabolism and has resulted in silencing of specific secondary metabolite genes with the concomitant diversion of metabolism towards the production of β–lactams.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-015-2154-4) contains supplementary material, which is available to authorized users.

Secondary metabolism in Penicillium chrysogenum was intensively subjected to classical strain improvement (CSI), the resulting industrial strains producing high levels of ␤-lactams. During this process, the production of yellow pigments, including sorbicillinoids, was eliminated as part of a strategy to enable the rapid purification of ␤-lactams. Here we report the identification of the polyketide synthase (PKS) gene essential for sorbicillinoid biosynthesis in P. chrysogenum. We demonstrate that the production of polyketide precursors like sorbicillinol and dihydrosorbicillinol as well as their derivatives bisorbicillinoids requires the function of a highly reducing PKS encoded by the gene Pc21g05080 (pks13). This gene belongs to the cluster that was mutated and transcriptionally silenced during the strain improvement program. Using an improved ␤-lactam-producing strain, repair of the mutation in pks13 led to the restoration of sorbicillinoid production. This now enables genetic studies on the mechanism of sorbicillinoid biosynthesis in P. chrysogenum and opens new perspectives for pathway engineering. IMPORTANCESorbicillinoids are secondary metabolites with antiviral, anti-inflammatory, and antimicrobial activities produced by filamentous fungi. This study identified the gene cluster responsible for sorbicillinoid formation in Penicillium chrysogenum, which now allows engineering of this diverse group of compounds.