[P007] Poster

12:10 PM - 1:00 PM JST(3:10 AM - 4:00 AM UTC)

[P04-515] Identification and Functional Analysis of Amino Acid Residues Governing Substrate Specificity in Polyketide Synthase Acyltransferase Domains

○Keika Kumagai¹, Misaki Aso¹, Hiroko Ueda¹, Satoshi Yuzawa¹ (1. Keio University (Japan))

Keywords:

Polyketide synthase (PKS),Substrate specificity,Combinatorial biosynthesis,Enzyme engineering

[Purpose]Accurate structure prediction of natural products from biosynthetic gene clusters requires a detailed understanding of the substrate specificity of biosynthetic enzymes. Polyketide synthases (PKSs) are multifunctional enzymes that produce diverse bioactive compounds, including antibiotics. Engineering PKS domains can alter extender unit selection, enabling the biosynthesis of novel drug analogs. Typically, acyltransferase (AT) domain specificity is predicted based on conserved motifs; however, exceptions exist. This study focused on the chalcomycin PKS (CHMS), where the module 2 AT domain (CHMS M2 AT) exhibits a discrepancy between its motifs and the substrate inferred from the final product structure. The objective was to characterize the substrate specificity of this atypical AT domain and identify novel determinants of AT specificity. [Method]Canonical malonyl-CoA-specific AT domains contain GHS(I/V/L) and HAFH motifs, while methylmalonyl-CoA-specific domains typically contain GHSQ and YASH motifs. This study targeted CHMS M2 AT, which contains methylmalonyl-CoA motifs but is predicted to incorporate malonyl-CoA. As an in vitro assay platform, module 6 and the thioesterase domain of 6-deoxyerythronolide B synthase (DEBS M6+TE) were used. Chimeric enzymes were constructed by replacing the native AT with CHMS M2 AT, and multiple DEBS M6 AT mutants carrying CHMS M2 AT-derived substitutions were generated. All constructs were expressed and purified in Escherichia coli BAP1. Enzymatic activity was evaluated by LC-MS analysis based on the resulting polyketide products. [Results]Wild-type DEBS M6+TE selectively incorporated methylmalonyl-CoA as expected. In contrast, the chimeric enzyme harboring CHMS M2 AT accepted both malonyl-CoA and methylmalonyl-CoA, with a clear preference for malonyl-CoA. To the best of our knowledge, this is the first in vitro demonstration that an AT domain containing GHSQ and YASH motifs can preferentially incorporate malonyl-CoA. Furthermore, mutational analysis identified specific residues that either suppress methylmalonyl-CoA incorporation or enhance malonyl-CoA incorporation. [Consideration]The characterization of CHMS M2 AT challenges conventional sequence-based models for predicting AT specificity. Analysis of the chimeric and mutant enzymes suggests that specificity is determined not only by conserved motifs but also by residues outside these regions. Notably, the residues identified that enhance malonyl-CoA incorporation have not been previously reported as specificity determinants. These findings imply that substrate recognition depends on broader structural features of the active site, potentially including the shape of the pocket floor. [Conclusion]This study characterized an exceptional AT domain in vitro that accepts malonyl-CoA despite possessing methylmalonyl-CoA motifs, challenging the conventional view of AT specificity. This discovery highlights the limitations of motif-only prediction models. Furthermore, previously unrecognized residues contributing to specificity were identified outside known motifs. These results indicate that PKS AT substrate selection depends not only on conserved motifs but also on broader structural features of the active site.