Independent evolution of betulin biosynthesis in Inonotus obliquus

Abstract

Chaga mushroom (Inonotus obliquus) is a fungal species in the family Hymenochaetaceae (Basidiomycota) and the causative agent of white rot decay in Betula species. We assembled a high-quality 50.7 Mbp genome from PacBio sequencing and identified a lineage-specific whole genome duplication event approximately 1.3 million years ago, which has contributed to a major increase in biochemical diversity in the species through preferential retention of cytochrome P450 superfamily members. Secondary metabolism has further evolved through small-scale segmental duplications, such as tandem duplications within fungal biosynthetic gene clusters. Metabolomic fingerprinting confirmed increased complexity in terpene biosynthesis chemistry compared to related species that lacked the duplication event. This metabolic diversity may have arisen from co-evolution with the primary host species, which evolved high betulin content in its bark 4-8 million years ago.

Fig. 1

Phylogenetic tree of three Ascomycetes species and 17 Basidiomycetes. Phylogeny was estimated from 4040 single-copy genes present in all species and rooted to Ascomycota species. Ascomycetes are indicated with green and grouped by phylum. Basidiomycetes orders are indicated with distinct colors and grouped by taxonomic order. Bootstrap support values report the level of confidence. The phylogenetic tree was rooted to Ascomycetes clade.

Fig. 2

Estimation of whole-genome duplication events. Histograms of log10 of the number of synonymous (Ks) substitutions in syntelogs identified from self-self alignments for I. obliquus, F. mediterranea, P. niemelaei, and S. paradoxa (top five panels) and interspecies alignments. The X-axis shows log10-scaled values of synonymous substitutions per syntelogous gene pairs (Ks values). Y-axis shows absolute counts.

Fig. 3

Bubble plot of gene ontology (GO) enrichments of genes originating from tandem duplications (top panel) and whole-genome duplication events (bottom panel) in I. obliquus. The X-axis and the bubble color correspond to the adjusted p-value (FDR; see the “FDR” color key). The bubble size indicates the number of genes belonging to each category (see legend for size guide).

Fig. 4

Venn diagram of expanded gene families (“Expanded”), genes originating from whole-genome duplication (“Syntenic”), and tandemly duplicated (“Tandem”) genes in I. obliquus genome. Each category is highlighted with a distinct color and labeled according to the category. P-value reports the p-value of Fisher’s exact test, testing the statistical significance of overlaps.

Fig. 5

Barplot of gene counts and the gene tree of cytochrome P450s predicted in I. obliquus. (A) Barplot shows the counts of P450 monooxygenase enzymes per family in I. obliquus genome. (B) Gene tree of P450 monooxygenase enzymes. Each leaf is highlighted with a point colored according to the major family of the enzyme. The sequences are labeled with gene ID, P450 families, family, and class. The heatmaps to the right of the tree indicate BLAST similarity, whether the gene is tandemly duplicated, differential expression in different RNA-seq experiments, and the motif present in conserved oxygen- and heme-binding domains. The last two columns show the log₁₀-scaled count of chaga oxygen- and heme-binding domains and their overlap with motifs found in different kingdoms and taxa.

Fig. 6

Venn diagrams of UPLC-MS mass spectrum for metabolomic fingerprints. (A) Venn diagram illustrating the overlap between peaks found in pooled mass spectra from five strains of I. obliquus and one F. mediterranea. (B) Venn diagram illustrating the overlap of the mass spectrum peaks detected from five strains of I. obliquus. (C) Multidimensional scaling (MDS) plot of metabolite abundances of five strains of I. obliquus.

Fig. 7

Concentrations of BE and BA in three strains of chaga and six Betula species assessed using HPLC–MS.

Fig. 8

Summary of all predicted cytochrome P450 (pfam ID: PF00067) amino acid sequences. (A) Heatmap of P450 families for different kingdoms and taxa. Rows indicate P450 families and columns indicate kingdoms and taxa. The color palette illustrates the log10-scaled protein sequence counts of P450 families in each of the kingdoms and two relevant divisions (see the color key below). Barplot above the columns illustrates the total number of proteins annotated to the P450 families in each of the kingdoms and divisions. (B) Upset plot of P450 families in different kingdoms and phyla. (C) Treemaps illustrating the percent of proteins annotated as CYP716 in different classes of streptophyta and CYP505 from the phyla of Ascomycota and Basidiomycota. Rectangle area reflects the percent of protein sequences predicted in the different taxa.

Fig. 9

Heatmap of BE biosynthesis concentration in yeast as a heterologous host. Columns of the heatmap indicate different biological replicates (Bio1-6). Rows indicate different expression vectors, such as the single vector construct of CYP716, double-insert vector construct of lupeol synthase and CYP716 genes, standard betulin, and CDD6 yeast strain as control. Species source for lupeol synthase and CYP716 was silver birch (B. pendula). c000016_g277 gene was extracted from chaga. Vector constructs are divided to double inserts (orange, lupeol synthase and CYP716) and single inserts (royal blue). The color palette of the heatmap illustrates the concentration (μg/mL) of BE found in yeast cells (Yeast extract) and yeast growth media (Culture media), with white color as minimum (zero) and dark red as maximum concentration of BE. Heatmap is also annotated according to the growth condition; brown (- Lupeol) is yeast growth culture without added lupeol (metabolic precursor to be potentially modified by CYP716 and CYP505 genes), gold (+ Lupeol) is yeast growth culture spiked with lupeol, cyan is CDD6 yeast strain as control, and yellow is BE standard (98% purity) obtained commercially.

Fig. 10

Heatmap of differentially expressed (DE) genes homologous to terpenoid biosynthesis enzymes that act on substrates from the mevalonate pathway (MVA). Three main enzymes are highlighted with distinct colors. The color palette of the heatmap corresponds to the log2 fold-change (log2FC) of the DE genes. The gene families are indicated by duplication origins, syntenic for genes originating from whole-genome duplication events, and tandem for tandemly duplicated genes.

Fig. 11

Detailed analysis of CYP716 and CYP505 monooxygenase families. (A) Upset plot of oxygen- and heme-binding domains in different kingdoms and taxa. (B) Total number of proteins that were annotated as CYP716 and CYP505 and the total number of species containing CYP716 and CYP505 families in different kingdoms and phyla. (C) Numbers of oxygen- and heme-binding domains found from CYP716 and CYP505 across different kingdoms and taxa.