New insights into the formation of fungal aromatic polyketides

Abstract

Fungal aromatic polyketides constitute a large family of bioactive natural products and are synthesized by the non-reducing group of iterative polyketide synthases (PKSs). Their diverse structures arise from selective enzymatic modifications of reactive, enzyme-bound poly-β-keto intermediates. How iterative PKSs control starter unit selection, polyketide chain initiation and elongation, intermediate folding and cyclization, selective redox or modification reactions during assembly, and product release are central mechanistic questions underlying iterative catalysis. This Review highlights recent insights into these questions, with a particular focus on the biosynthetic programming of fungal aromatic polyketides, and draws comparisons with the allied biosynthetic processes in bacteria.

Fig. 1. Representative polyketides

a) Bioactive polyketides from bacterial and fungal sources. b) Examples of fungal aromatic polyketides. The domain architecture for the NR-PKS group, represented as balls (domains) on a string (linkers), is shown: starter unit:ACP transacylase (SAT), ketosynthase (KS), malonyl-CoA:ACP transacylase, product template (PT), acyl-carrier protein (ACP or CP), and thioesterase (TE)/Claisen cyclase (CLC).

Fig. 2. Basic fatty acid and polyketide processing elements

The growing substrates are passed back and forth between the KS and ACP via thioester transfers (Enz = enzyme: KS or ACP). The acyl-transacylase (AT) loads the ACP from starter or extender acyl-CoA substrates. Product complexity arises from the number of cycles (chain length control) and selective reductive processing that can occur during product assembly by a ketoreductase (KR), dehydrase (DH), or enoyl-reductase (ER). The TE is often used to terminate the program to release the final product. Type II systems are free-standing proteins, as depicted in the image, that associate to form a multi-enzyme complex, whereas type I systems are fused into single multidomain proteins.

Fig. 3. Model fungal aromatic polyketides and their biosynthetic outcomes

Cyclization of the ACP-tethered linear poly-β-keto intermediates (R group) by the PT domain and TE/CLC domain is shown for the production of norsolorinic acid anthrone (8, A), YWA1 napthopyrone (12, B), tetrahydroxynaphthalene (14, C), and the bikaverin precursor (20, D).

Fig. 4. PT-mediated cyclization of aromatic polyketides

A) The monomer unit of the high-resolution dimeric PT crystal structure is shown, illustrating the linear arrangement of substrate binding sites. B) Proposed mechanism for cyclization/aromatization in the PksA PT domain.

Fig. 5. TE/CLC-mediated chain termination

A) Proposed mechanism for the final Claisen/Dieckmann condensation. The reaction sequence is shown beginning from the covalent TE/CLC-oxyester intermediate, which is analogous to the classical serine protease intermediate. The ACP phosphopantetheine arm must depart before the substrate fatty acyl side chain can swing into correct position for C-C cyclization as opposed to O-C product derailment. B) The modeled tetrahedral intermediate is shown filling the active site cavity volume in the TE/CLC cyclization chamber.