Codon-specific isotope labeling for NMR resonance assignment

Project summary It’s estimated that more than 30% of cancer-associated pathways involve intrinsically disordered proteins (IDPs) whose lack of structural information poses a major obstacle to rational drug discovery. An example is the fusion protein EWS-FLI1 that is the driver of the childhood cancer Ewing Sarcoma. Although there are some small molecules that engage EWS-FLI1 with moderate potency, the lack of structural validation has recently called their mechanism of action into question. NMR is the ideal technology for characterizing IDPs; however EWS-FLI1’s size, structural disorder, and prevalence of repeat sequences yields extensive peak overlap that challenges conventional resonance assignment strategies. Technologies exist to reduce this spectra crowding, but they either lack selectivity (labeling/unlabeling all residues of a given type) or require mutagenesis which alters protein structure and spectra. This proposal focuses on developing a new technology, termed codon-specific isotope labeling, that overcomes these limitations by segregating isotope labels at the codon rather than the residue level. Each isotopologue (14N, 15N, 15N13Cα) produces a distinct signal (no peak, singlet, or doublet, respectively) in the 2D HSQC spectra that facilitates rapid resonance assignment from a small number of samples. And, as amino acid isotopologues rather than mutants, they do not modify the protein structure or spectra. Our published work shows that the six leucine codons can be split to encode five different isotopologues on small-scale, but significant barriers remain to applying this technology for NMR-scale protein production. The overall goal of this proposal is to establish this technology using EWS- FLI1 as a model IDP cancer target. Aim 1 focuses on the leucine codons. Using a commercial extract-based in vitro translation system, we will first optimize the yields and incorporation efficacy of green fluorescent protein (GFP) using fluorescence and ESI-MS. Following optimization, the technology will be applied to EWS-FLI1. The focus of Aim 2 is to expand our technology to two amino acids overrepresented in IDPs, serine and glycine, by dividing their codons into three isotopologues each. We hypothesize these residues will cluster by type in the 2D HSQC spectrum to enable multiplex codon-specific isotope labeling. This proposal will culminate in the comprehensive structural characterization of EWS-FLI1 in complex with each of the putative two small molecule inhibitors YK-4-279 and TK216. This proposal will establish codon-selective labeling as a new powerful strategy for deconvoluting NMR resonance assignments of IDPs, opening the door to drug discovery for the many IDPs involved in cancer.