Prioritizing disease-associated missense variants with chemoproteomic-detected amino acids

TL;DR

This study shows that chemoproteomic data can help identify disease-related protein variants, improving variant interpretation and drug development.

Contribution

The study introduces the use of chemoproteomic amino acid reactivity to prioritize disease-associated missense variants.

Findings

Genes with chemoproteomic-detected amino acids are enriched for monogenic-disease phenotypes.

CpDAAs are enriched near known pathogenic variants in 1D and 3D protein structures.

Regions around CpDAAs in FH are enriched for VUSs and pathogenic variants, validated by altered oligomerization.

Abstract

Missense variants are the most common type of protein-altering genetic variation. Due to their wide-ranging potential functional consequences, missense variants are challenging to interpret and, as a result, are often classified as unknown pathogenicity or as variants of uncertain significance (VUSs). Genomic-based predictive tools have made significant inroads into the challenge of accurately pinpointing functional missense variants by providing genome-wide assessments of deleteriousness or potential pathogenicity. Complementary to these tools, here we provide an initial study into the utility of harnessing protein-based measures of amino acid reactivity to delineate functionally significant missense variants. These reactivity measurements, which are generated using mass spectrometry-based chemoproteomic methods, have already proved capable of pinpointing functional sites on proteins, which provide the added value of delineating potential sites suitable for drug-development efforts. Here, using published chemoproteomic datasets for three specific privileged amino acids, cysteine, lysine, and tyrosine, we assessed the utility of proteomic reactivity measurements to identify clinically important variants and regions within monogenic-disease-associated genes. We found that genes where amino acids are detected via chemoproteomics are enriched for monogenic-disease phenotypes, indicative of functional importance. Chemoproteomic-detected amino acids (CpDAAs) are enriched at and around sites with known pathogenic missense variants when assessed with either one- or three-dimensional protein structures. To further illustrate the utility of our findings, we found that regions at or around CpDAAs in fumarate hydratase (FH) were enriched for VUSs and pathogenic variants, which we validate through demonstration of an altered FH oligomerization state. Collectively, our study highlights the potential of chemoproteomic and genetic data integration for enhancing the identification of functional genetic variants and protein sites with potential value for drug-development efforts.

Mass spectrometry-based chemoproteomics can provide a high-throughput assessment of the likely functionality of specific sites on proteins. Here, we show that chemoproteomic measures of protein-based amino acid reactivity effectively highlight clinically important variants in disease-associated genes. These findings should have relevance in enhancing variant interpretation and drug development.

Figures