Lexapeptide is a lanthipeptide previously discovered through functional genome mining of Streptomyces rochei.¹ It is an attractive synthetic target with potent antimicrobial activity (MIC <1 μg/ml) against Gram-positive bacteria, including several antibiotic resistant strains, and it contains several unusual structural motifs. These include several dehydroamino acids and an unusual aminovinylcysteine (Avi[Me]Cys) residue enclosing one of its two rings. The other ring is enclosed by a typical lanthionine bridge.  Herein we describe our progress towards the total synthesis of the proposed structure of lexapeptide by a convergent approach.

Lexapeptide is naturally produced by post-translational modification of a 38-residue precursor peptide. Two macrocycles are introduced, one enclosed by lanthionine (Ring A), and the other enclosed by Avi[Me]Cys (Ring B). Synthesis of the A-ring is routine, enabled by the incorporation of an orthogonally protected lanthionine building block during SPPS. However, the B-ring presents an interesting synthetic challenge. For this reason, we conceived of a convergent approach where a B-ring fragment would be synthesized separately, then coupled to a longer A-ring containing fragment.

To date, we have successfully synthesized a protected B-ring fragment by employing a late-stage oxidative decarboxylation to introduce the thioenamide moiety. En route to the A-ring fragment, we have developed a convenient method for the introduction of dehydroalanine and dehydrobutyrine by the oxidative elimination of phenylselenocysteine and phenylselenohomocysteine respectively. The required selenocysteine building blocks are conveniently accessed by photodecarboxylation of N-hydroxyphthalimide esters of Asp/Glu without the need for expensive photocatalysts. At the time of submission, synthesis of the A-ring fragment is ongoing.