Ribosomal Synthesis of Thioether-Bridged Bicyclic Peptides

Abstract

Many biologically active peptides found in nature exhibit a bicyclic structure wherein a head-to-tail cyclic backbone is further constrained by an intramolecular linkage connecting two side chains of the peptide. Accordingly, methods to access macrocyclic peptides sharing this overall topology could be of significant value toward the discovery of new functional entities and bioactive compounds. With this goal in mind, we recently developed a strategy for enabling the biosynthesis of thioether-bridged bicyclic peptides in living bacterial cells. This method involves a split intein-catalyzed head-to-tail cyclization of a ribosomally produced precursor peptide, combined with inter-sidechain cross-linking through a genetically encoded cysteine-reactive amino acid. This approach can be applied to direct the formation of structurally diverse bicyclic peptides with high efficiency and selectivity in living Escherichia coli cells and provides a platform for the generation of combinatorial libraries of genetically encoded bicyclic peptides for screening purposes.

Keywords: Amber stop codon suppression; Bicyclic peptides; Split intein; Thioether-linked peptide macrocycles; Unnatural amino acid.

Figure 1

Overall strategy for the ribosomal synthesis of thioether-bridged bicyclic peptides in E. coli. From the N- to C-terminus, the precursor protein comprises: (a) the C-terminal domain of DnaE split intein (Int_C); (b) a variable target peptide sequence; (c) the N-terminal domain of DnaE split intein fused to a chitin binding domain (Int_N-CBD). The target peptide sequence contains an initial Ser or Cys residue (dotted circle) at the ‘Int_C+1’ site for mediating split intein-catalyzed head-to-tail cyclization and comprises the unnatural amino acid O-(2-bromoethyl)-tyrosine (O2beY, circle with vertical lines) and the reactive cysteine (white circle) for inter-side-chain crosslinking via thioether bond formation.

Figure 2

Mechanism of split intein-catalyzed peptide circularization. Int_N and Int_C correspond to the N-domain and C-domain respectively, of DnaE split intein from Synechocystis sp. PCC6803. The Int_N+1 cysteine and Int_C+1 cysteine (or serine) residues are indicated.

Figure 3

Synthetic route for preparation of O-(2-bromoethyl)tyrosine (O2beY).

Figure 4

DNA and protein sequence for the precursor protein Z3C_O2beY (Table 1). The target peptide sequence is highlighed in pink and is preceeded by the Int_C domain of DnaE split intein (green) and C-terminally fused to DnaE Int_N domain (yellow) and the chitin binding domain affinity tag (grey).

Figure 5

Analysis of DnaE trans splicing products. A) SDS-PAGE analysis of the protein construct Z3C_O2beY (lane 1) and Z8C_O2beY (lane 2) isolated using chitin beads. The bands corresponding to Int_N-CBD and full-length precursor protein are labeled. The latter is not visible due to complete splicing of the precursor protein. B) Deconvoluted MS spectrum of chitin-affinity purified Int_N-CBD/Int_C complex from Z3C_O2beY sample. The position of the full-length precursor protein (if present) is also indicated. (C) Deconvoluted MS spectrum corresponding to the dissociated Int_N-CBD fragment upon treatment of the chitin-affinity purified Int_N-CBD/Int_C complex (from Z3C_O2beY sample) with 6 M guanidine hydrochloride.

Figure 6

LC–MS analysis of macrocyclic products obtained from in vivo cyclization of Z8C_O2beY precursor protein. (A) LC-MS extracted-ion chromatogram (left panel), MS/MS fragmentation spectrum (middle panel), and chemical structure (right panel) of bicyclic peptide isolated via streptavidin affinity from cells expressing the contruct Z8C_O2beY. (B) Same as (A) for the head-to-tail cyclic peptide byproduct. Peaks labeled with * correspond to unrelated multicharged ions from adventitious proteins.