Identification of a biosynthetic gene cluster for a red pigment cristazarin produced by a lichen-forming fungus Cladonia metacorallifera

Abstract

Lichens are known to produce many novel bioactive metabolites. To date, approximately 1,000 secondary metabolites have been discovered, which are predominantly produced by the lichen mycobionts. However, despite the extensive studies on production of lichen secondary metabolites, little is known about the responsible biosynthetic gene clusters (BGCs). Here, we identified a putative BGC that is implicated in production of a red pigment, cristazarin (a naphthazarin derivative), in Cladonia metacorallifera. Previously, cristazarin was shown to be specifically induced in growth media containing fructose as a sole carbon source. Thus, we performed transcriptome analysis of C. metacorallifera growing on different carbon sources including fructose to identify the BGC for cristazarin. Among 39 polyketide synthase (PKS) genes found in the genome of C. metacorallifera, a non-reducing PKS (coined crz7) was highly expressed in growth media containing either fructose or glucose. The borders of a cristazarin gene cluster were delimited by co-expression patterns of neighboring genes of the crz7. BGCs highly conserved to the cristazarin BGC were also found in C. borealis and C. macilenta, indicating that these related species also have metabolic potentials to produce cristazarin. Phylogenetic analysis revealed that the Crz7 is sister to fungal PKSs that biosynthesize an acetylated tetrahydoxynaphthalene as a precursor of melanin pigment. Based on the phylogenetic placement of the Crz7 and putative functions of its neighboring genes, we proposed a plausible biosynthetic route for cristazarin. In this study, we identified a lichen-specific BGC that is likely involved in the biosynthesis of a naphthazarin derivative, cristazarin, and confirmed that transcriptome profiling under inducing and non-inducing conditions is an effective strategy for linking metabolites of interest to biosynthetic genes.

Fig 1. Identification of polyketide synthase for the biosynthesis of cristazarin.

PKS gene expression profiles of the Cladonia metacorallifera mycobiont grown on culture media containing either fructose, glucose, ribitol or sorbitol as a single carbon source and on malt extract agar (MYA). Expression values (Log2-transformed (RPKM+1)) are shown as heat maps for 39 PKS genes in C. metacorallifera.

Fig 2. Demarcation of the cristazarin BGC boundaries by mapped reads of RNA-seq.

(A) Mapped reads of five RNA-seq samples of the C. metacorallifera mycobiont growing on different carbon sources were shown for a genomic locus harboring the PKS22 (crz7; gene ID: Cmt_01711). RNA-seq reads mapped on the C. metacorallifera reference genome were subsampled to 60 million reads for visual comparison of expression levels between samples. Arrows on the x-axis indicate genes (Cmt_01704–Cmt_01717). The numbers on the y-axis are per-base coverage of mapped reads. (B) Synteny of the cristazarin BGCs in Cladonia species: C. metacorallifera (top), C. macilenta (middle), and C. borealis (bottom). Arrows indicate open reading frames (ORFs) found in the BGCs, and homologous genes were represented with different colors. Numbers above the arrows indicate crz1–crz9.

Fig 3. Phylogenetic dereplication of polyketide synthases related to melanin production.

A maximum likelihood phylogenetic tree of non-reducing polyketide synthases (NR-PKSs). Colored strips on the right side of leaves indicate polyketide backbones produced by the functionally characterized NR-PKSs. Clades of the previously described NR-PKS groups (Kim et al., 2021; Mosunova et al., 2022) were shaded with different colors (see inset). Arrows indicate clades of the Cladonia PKS families, PKS13, PKS14, PKS15, and PKS22, that belong to group II NR-PKS (Kim et al., 2021). Leaves highlighted in red are lichen NR-PKSs. Crz7 in C. metacorallifera was denoted by a red asterisk. An NR-PKS in Aspergillus fumigatus (EDP55264) that produce melanin pigment using YWA as a precursor was set as an outgroup. Bootstrap values of greater than 75% were shown. Branch lengths are proportional to the inferred amount of evolutionary change, and the scale represents 0.1 amino acid sequence substitutions per site. T4HN, 1,3,6,8-tetrahydroxynaphthalene; AT4HN, 2-acetyl-1,3,6,8-tetrahydroxynaphthalene.

Fig 4. Divergent biosynthetic routes of cristazarin.

A proposed biosynthetic pathway of cristazarin from a polyketide precursor, 2-acetyl-1,3,6,8-tetrahydoxynaphthalene (AT4HN), can be deduced from putative function of tailoring enzymes in the cristazarin gene cluster (red). Note that a biosynthetic pathway for 6-O-methylasparvenone (a naphthalenone) is analogous to that for cristazarin, involving several biosynthetic genes in the naphthalenone BGC (blue) in Aspergillus parvulus (Mosunova et al. 2022). ACP, acyl carrier protein domain.