Deep Learning-Driven Library Design for the De Novo Discovery of Bioactive Thiopeptides

Abstract

Broad substrate tolerance of ribosomally synthesized and post-translationally modified peptide (RiPP) biosynthetic enzymes has allowed numerous strategies for RiPP engineering. However, despite relaxed specificities, exact substrate preferences of RiPP enzymes are often difficult to pinpoint. Thus, when designing combinatorial libraries of RiPP precursors, balancing the compound diversity with the substrate fitness can be challenging. Here, we employed a deep learning model to streamline the design of mRNA display libraries. Using an in vitro reconstituted thiopeptide biosynthesis platform, we performed mRNA display-based profiling of substrate fitness for the biosynthetic pathway involving five enzymes to train an accurate deep learning model. We then utilized the model to design optimal mRNA libraries and demonstrated their utility in affinity selections against IRAK4 kinase and the TLR10 cell surface receptor. The selections led to the discovery of potent thiopeptide ligands against both target proteins (K D up to 1.3 nM for the best compound against IRAK4 and 300 nM for TLR10). The IRAK4-targeting compounds also inhibited the kinase at single-digit μM concentrations in vitro , exhibited efficient internalization into HEK293H cells, and suppressed NF-kB-mediated signaling in cells. Altogether, the developed approach streamlines the discovery of pseudonatural RiPPs with de novo designed biological activities and favorable pharmacological properties.

Genome Mining for New Enzyme Chemistry

“…Many RiPPs are macrocyclic structures that possess favorable physiochemical properties such as higher metabolic stability, cell membrane permeability, and reduced entropy cost upon target binding compared to the corresponding linear peptides. , Numerous RiPP enzymes possess broad substrate specificity, enabling catalysis on unnatural substrates to accelerate early drug discovery. ,− Consequently, libraries of RiPPs have been developed and screened against disease-relevant protein targets, yielding new-to-nature RiPPs as potential drug candidates. ,− …”

“… 27 , 139 − 142 Consequently, libraries of RiPPs have been developed and screened against disease-relevant protein targets, yielding new-to-nature RiPPs as potential drug candidates. 102 , 143 − 148 …”

Engineering tRNAs for the Ribosomal Translation of Non-proteinogenic Monomers

“…Lactazole A is a thiopeptide from Streptomyces lactacystinaeus that is biosynthesized from five enzymes and a precursor peptide . Vinogradov and co-workers combined the Laz enzymes with the FIT system to create the FIT-Laz system. , Translation of the leader peptide and treatment with the Laz enzymes resulted in a thiopeptide scaffold containing a dehydroalanine, thiazole, oxazole, and pyridine moiety. To demonstrate that npM-containing core peptides could be converted into the thiopeptide scaffold, npMs were first incorporated into the AUG codon of the Laz leader peptide using tRNA Pro1E2 CAU .…”

A Compact Reprogrammed Genetic Code for De Novo Discovery of Proteolytically Stable Thiopeptides

“…Previously, we established an mRNA display-based platform for the de novo discovery of functional natural product-like thiopeptides. − The platform leverages a reengineered in vitro biosynthesis of lactazole A, a thiopeptide from Streptomyces lactacystinaeus (Figure a) . To access large (>10 12 unique compounds) libraries of lactazole-like thiopeptides, partially randomized and mRNA-barcoded precursor peptides (LazA variants) are produced with the flexible in vitro translation ( FIT ) system, and are then treated with lactazole biosynthetic enzymes (LazBCDEF) to convert the linear precursors into macrocyclic thiopeptides for downstream affinity selections.…”