Expanding Extender Substrate Selection for Unnatural Polyketide Biosynthesis by Acyltransferase Domain Exchange within a Modular Polyketide Synthase

Abstract

Modular polyketide synthases (PKSs) are polymerases that employ α-carboxyacyl-CoAs as extender substrates. This enzyme family contains several catalytic modules, where each module is responsible for a single round of polyketide chain extension. Although PKS modules typically use malonyl-CoA or methylmalonyl-CoA for chain elongation, many other malonyl-CoA analogues are used to diversify polyketide structures in nature. Previously, we developed a method to alter an extension substrate of a given module by exchanging an acyltransferase (AT) domain while maintaining protein folding. Here, we report in vitro polyketide biosynthesis by 13 PKSs (the wild-type PKS and 12 AT-exchanged PKSs with unusual ATs) and 14 extender substrates. Our ∼200 in vitro reactions resulted in 13 structurally different polyketides, including several polyketides that have not been reported. In some cases, AT-exchanged PKSs produced target polyketides by >100-fold compared to the wild-type PKS. These data also indicate that most unusual AT domains do not incorporate malonyl-CoA and methylmalonyl-CoA but incorporate various rare extender substrates that are equal to in size or slightly larger than natural substrates. We developed a computational workflow to predict the approximate AT substrate range based on active site volumes to support the selection of ATs. These results greatly enhance our understanding of rare AT domains and demonstrate the benefit of using the proposed PKS engineering strategy to produce novel chemicals in vitro.

Discrete Acyltransferases and Thioesterases in Iso-Migrastatin and Lactimidomycin Biosynthesis

“… Four motifs that have been implicated in the catalytic mechanism of type I PKS ATs were also conserved among the five sequences (Figure S2A). − AlphaFold models of the AT domains of MgsB and MgsH were compared to the model of DszD built from a crystal structure (PDB 3RGI), and the overall structures were conserved along with the positioning of the four motifs (Figure S2C). A sequence alignment of TE domains of MgsB and LtmB with RifR, a functionally and structurally characterized type II TE, revealed that the catalytic triad Ser-Asp-His was conserved among the three sequences (Figure S2D,E).…”

“…By sequence alignment, all four motifs identified in the AT regions of MgsB and MgsH were present and nearly identical, and AlphaFold models of both AT regions overlaid well with one another (1.09 Å RMSD; Figure S2). ,− Therefore, given the flexible nature of ATs, 50 ns MD simulations were carried out for the AT regions of both proteins to determine if the missing AT activity in MgsB could be rationalized (Figure S5A). No obvious differences in the catalytic Ser-His dyads (Figure S5B) or the large and small AT subunits of MgsB and MgsH could be observed (Figure S5C).…”

Evolution-guided engineering of trans -acyltransferase polyketide synthases

“…S13). This suggested that the LPTYPFx 5 W motif, which also occurs in cis -AT PKSs downstream of the AT domain and has been successfully used in AT swapping experiments in cis -AT PKSs ( 43 , 44 ), separates evolutionarily autonomous parts in trans -AT PKSs. In line with terminology from NRPS engineering ( 45 ), we use the term “exchange units” for these evolutionarily autonomous parts that contain various domains ( 25 , 34 , 46 ), whereas “module” refers to biochemically functional KS-to-ACP sections (Fig.…”

Current State‐of‐the‐Art Toward Chemoenzymatic Synthesis of Polyketide Natural Products

“…[71] More broadly, recent work has shown that exchanging AT domains with those from orthogonal PKSs enables the incorporation of alternative groups, from small (H or Me) to large (Ph, n hex, etc). [72] Leveraging this approach, a large panel of TKLs was quantified, enabling robust structural diversification.…”