Broadening the substrate range of serine palmitoyltransferase by protein engineering and applications to 3-keto-dihydrosphingosine analogs†

Serine palmitoyltransferase produces 3-keto-dihydrosphingosine (KDS) in a single step by a Claisen-like condensation/decarboxylation reaction between l-Ser and palmitoyl-CoA (n-C₁₆-CoA). Unfortunately, the enzyme's synthetic potential is limited by its highly restricted substrate range (n-C₁₄-CoA to n-C₁₈-CoA). We previously reported that the R378K variant of Sphingomonas paucimobilis serine palmitoyltransferase (SpSPTase) preferred slightly shorter acyl chain length substrates such as n-C₁₂-CoA. While this represented an improvement, we sought to broaden the biocatalyst's substrate range further to allow the synthesis of a much wider range of KDS analogs. Starting from the R378K mutant, we prepared twenty second-generation site-saturation mutant libraries targeting residues lining the active site. Screening with l-Ser and n-C₈-CoA as substrates revealed that mutations at only one of the twenty positions yielded improved variants (Tyr 73). Both the acyl-CoA substrate range as well as the interactions with the PLP:l-Ser external aldimine were significantly altered. The best double mutant (R378K/Y73N) showed superior catalytic activity for n-C₈-CoA (k_cat = 0.44 s⁻¹) while also retaining wild-type thermostability. It even accepted n-C₆-CoA and several functionalized acyl-chains, demonstrating the substantially broadened substrate range. Finally, to demonstrate the practical utility of our best variant, we used the R378K/Y73N double mutant to synthesize a short-chain KDS analog on a preparative scale.