Abstract:Ubiquitin ligases (E3s) confer specificity to ubiquitination by recognizing target substrates. However, the substrates of most E3s have not been extensively discovered, and new methods are needed to efficiently and comprehensively identify these substrates. Mostly, E3s specifically recognize substrates via their protein interaction domains. We developed a novel integrated strategy to identify substrates of E3s containing protein interaction domains on a proteomic scale. The binding properties of the protein interaction domains were characterized by screening a random peptide library using a yeast two-hybrid system. Artificial degrons, consisting of a preferential ubiquitination sequence and particular interaction domain-binding motifs, were tested as potential substrates by in vitro ubiquitination assays. Using this strategy, not only substrates but also nonsubstrate regulators can be discovered. The detailed substrate recognition mechanisms, which are useful for drug discovery, can also be characterized. We used the Ligand of Numb protein X (LNX) family of E3s, a group of PDZ domain-containing RING-type E3 ubiquitin ligases, to demonstrate the feasibility of this strategy. Many potential substrates of LNX E3s were identified. Eight of the nine selected candidates were ubiquitinated in vitro, and two novel endogenous substrates, PDZ-binding kinase (PBK) and breakpoint cluster region protein (BCR), were confirmed in vivo. We further revealed that the LNX1-mediated ubiquitination and degradation of PBK inhibited cell proliferation and enhanced sensitivity to doxorubicin-induced apoptosis. The substrate recognition mechanism of LNX E3s was also characterized; this process involves the recognition of substrates via their specific PDZ domains by binding to the C-termini of the target proteins. This strategy can potentially be extended to a variety of E3s that contain protein interaction domain(s), thereby serving as a powerful tool for the comprehensive identification of their substrates on a proteomic scale.

Assessing substrate scope of the cyclodehydratase LynD by mRNA display-enabled machine learning models

mRNA Display in Cell Lysates Enables Identification of Cyclic Peptides Targeting the BRD3 Extraterminal Domain

Substrate Prediction for RiPP Biosynthetic Enzymes via Masked Language Modeling and Transfer Learning

Analysis and Prediction of Chymotrypsin Substrate Preferences through Large Data Acquisition with Target‐Free mRNA Display

Discovering de novo peptide substrates for enzymes using machine learning

Engineering the Substrate Specificity of the DhbE Adenylation Domain by Yeast Cell Surface Display.

Machine learning modeling of family wide enzyme-substrate specificity screens

High-throughput profiling of sequence recognition by tyrosine kinases and SH2 domains using bacterial peptide display

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

Identification of photocrosslinking peptide ligands by mRNA display

Substrate selection influences molecular recognition in a screen for lymphoid tyrosine phosphatase inhibitors

mRNA display: ligand discovery, interaction analysis and beyond

High-throughput reprogramming of an NRPS condensation domain

Substrate interactions guide cyclase engineering and lasso peptide diversification

High‐Throughput Discovery of Substrate Peptide Sequences for E3 Ubiquitin Ligases Using a cDNA Display Method

Method of mRNA display for selecting peptides that bind with target enzyme RNA

A Proteomics Strategy to Identify Substrates of LNX, a PDZ Domain Containing E3 Ubiquitin Ligase

Directed evolution of a cyclodipeptide synthase with new activities via label-free mass spectrometric screening

Rapid Screening of Lanthipeptide Analogs Via In-Colony Removal of Leader Peptides in Escherichia Coli

Machine Learning Enables Prediction of Pyrrolysyl-tRNA Synthetase Substrate Specificity