Synthetic proteins are expected to go beyond the boundary of recombinant DNA expression systems by being flexibly installed with site-specific natural or unnatural modification structures during synthesis. To enable protein chemical synthesis, peptide ligations provide effective strategies to assemble short peptide fragments obtained from solid-phase peptide synthesis (SPPS) into long peptides and proteins. In this regard, chemoselective peptide ligation represents a simple but powerful transformation realizing selective amide formation between the C-terminus and N-terminus of two side-chainunprotected peptide fragments. These reactions are highly chemo- and regioselective to tolerate the side-chain functionalities present on the unprotected peptides, highly reactive to work with millmolar or submillimolar concentrations of the substrates, and operationally simple with mild conditions and accessible building blocks. This Account focuses on our work in the development of serine/threonine ligation (STL), which originates from a chemoselective reaction between an unprotected peptide with a C-terminal salicylaldehyde (SAL) ester and another unprotected peptide with an N-terminal serine or threonine residue. Mechanistically, STL involves imine capture, 5-endo-trig ring-chain tautomerization, O-to-N [1,5] acyl transfer to afford the N,O-benzylidene acetal-linked peptide, and acidolysis to regenerate the Xaa-Ser/Thr linkage (where Xaa is the amino acid) at the ligation site. The high abundance of serine and threonine residues (12.7%) in naturally occurring proteins and the good compatibility of STL with various C-terminal residues provide multiple choices for ligation sites. The requisite peptide C-terminal SAL esters can be prepared from the peptide fragments obtained from both Fmoc-SPPS and Boc-SPPS through four available methods (a safety-catch strategy based on phenolysis, direct coupling, ozonolysis, and the n + 1 strategy). In the synthesis of proteins (e.g., ACYP enzyme, MUC1 glycopeptide 40-mer to 80-mer, interleukin 25, and HMGA1a with variable post-translational modification patterns), both C-toN and N-to-C sequential STL strategies have been developed through selection of temporal N-terminal protecting groups and proper design of the switch-on/off C-terminal SAL ester surrogate, respectively. In the synthesis of cyclic peptide natural products (e.g., daptomycin, teixobactin, cyclomontanin B, yunnanin C) and their analogues, intramolecular head-to-tail STL has been implemented on linear peptide SAL ester precursors containing four to 10 amino acid residues with good efficiency and minimized oligomerization. As a thiol-independent chemoselective ligation complementary to native chemical ligation, STL provides an alternative tool for the chemical synthesis of homogeneous proteins with site-specific and structure-defined modifications and cyclic peptide natural products, which lays foundation for chemical biology and medicinal studies of those molecules with biological importance and therapeutic potential.
