Linking Lichen Metabolites to Genes: Emerging Concepts and Lessons from Molecular Biology and Metagenomics

Abstract

Lichen secondary metabolites have tremendous pharmaceutical and industrial potential. Although more than 1000 metabolites have been reported from lichens, less than 10 have been linked to the genes coding them. The current biosynthetic research focuses strongly on linking molecules to genes as this is fundamental to adapting the molecule for industrial application. Metagenomic-based gene discovery, which bypasses the challenges associated with culturing an organism, is a promising way forward to link secondary metabolites to genes in non-model, difficult-to-culture organisms. This approach is based on the amalgamation of the knowledge of the evolutionary relationships of the biosynthetic genes, the structure of the target molecule, and the biosynthetic machinery required for its synthesis. So far, metagenomic-based gene discovery is the predominant approach by which lichen metabolites have been linked to their genes. Although the structures of most of the lichen secondary metabolites are well-documented, a comprehensive review of the metabolites linked to their genes, strategies implemented to establish this link, and crucial takeaways from these studies is not available. In this review, I address the following knowledge gaps and, additionally, provide critical insights into the results of these studies, elaborating on the direct and serendipitous lessons that we have learned from them.

Keywords: atranorin; biosynthetic gene clusters; grayanic acid; gyrophoric acid; lecanoric acid; lichen metabolites; physodic/olivetoric acid; secondary metabolites; usnic acid.

Figure 1

Biosynthetic gene clusters, types of PKSs (based on domains), and the evolutionary relationships of non-reducing PKSs. (A) Different kinds of biosynthetic gene clusters, depending upon the core genes: (i) a cluster containing a single core gene—this could be a type-I NRPKS (usually 5000–8000 bp) or a type-III NRPKS (~800–1300 bp), an NRPS (~3000–4000 bp), terpene (~800 bp), indol (800–1300 bp), or ribosomally synthesized and post-translationally modified peptides (RiPP, 600–900 bp) BGC; (ii) a cluster containing two core genes—this could be an NRPKS and a reducing PKS, or both reducing PKSs, or both NRPKSs; (iii) a BGC with two core genes in the opposite orientation; (iv) a hybrid BGC with both an NRPS and an NRPKS. (B) Domain composition of minimal PKSs, highly-reducing PKSs, partially-reducing PKSs, and non-reducing PKSs. Dotted borders denote that the domain is facultative. (C) A representative 1000-bootstrap maximum-likelihood NRPKS tree, showing the nine NRPKS groups. Colored clades represent the groups containing the PKSs, linked to a molecule. The structure of the molecules coded by these clades and the domains of the corresponding PKS are shown in the figure (R = C_nH_n). (adapted from Singh et al. 2021 [15]; dataset downloaded from supplemental material Table S2, for details on the dataset, alignment and tree annotation please refer to Singh et al. 2021 [15]). The final data set consisted of amino acid sequences of 229 NR-PKSs from 18 species belonging to five LFF genera-Dermatocarpon, Cladonia, Pseudevernia, Stereocaulon, and Umbilicaria.

Figure 2

The molecular structures of four classes of lichen compounds linked to the genes. The colored circles denote the characteristic bond of a compound-ester bond present in a depside (A), the ester bond of a ß-orcinol depside (B), the ester and ether bond present in a depsidone (C), and the characteristic dibenzofurane bond (D).

Figure 3

The lichen compounds that are linked to a PKS. The compounds highlighted in green are the ones in which the link between the molecule and the PKS was also verified by heterologous expression. (A) lecanoric acid, (B) atranorin, (C) olivetoric acid (D) gyrophoric acid, (E) physodic acid, (F) grayanic acid and (G) usnic acid.