Genetic dissection of the pluripotent proteome through multi-omics data integration

Abstract

Genetic background is a major driver of phenotypic variability in pluripotent stem cells (PSCs). Most studies of variation in PSCs have relied on transcript abundance as the primary molecular readout of cell state. However, little is known about how proteins, the primary functional units in the cell, vary across genetically diverse PSCs, how protein abundance relates to variation in other cell characteristics, and how genetic background confers these effects. Here we present a comprehensive genetic study characterizing the pluripotent proteome of 190 unique mouse embryonic stem cell lines (mESCs) derived from genetically heterogeneous Diversity Outbred (DO) mice. The quantitative proteome is highly variable across DO mESCs, and we identified differentially activated pluripotency-associated pathways in the proteomics data that were not evident in transcriptome data from the same cell lines. Comparisons of protein abundance to transcript levels and chromatin accessibility show broad co-variation across molecular layers and variable correlation across samples, with some lines showing high and others low correlation between these multi-omics datasets. Integration of these three molecular data types using multi-omics factor analysis revealed shared and unique drivers of quantitative variation in pluripotency-associated pathways. QTL mapping localized the genetic drivers of this quantitative variation to a number of genomic hotspots, and demonstrated that multi-omics data integration consolidates the influence of genetic signals shared across molecular traits to increase QTL detection power and overcome the limitations inherent in mapping individual molecular features. This study reveals transcriptional and post-transcriptional mechanisms and genetic interactions that underlie quantitative variability in the pluripotent proteome, and in so doing provides a regulatory map for mouse ESCs that can provide a rational basis for future mechanistic studies, including studies of human PSCs.

An in vitro neurogenetics platform for precision disease modeling in the mouse

“…S5A), the single nucleotide variants (SNV) do not affect the protein sequence, and all variants are synonymous (table S1). Examination of expression quantitative trait loci (eQTL) in mESC (15,37) revealed a local eQTL in CAST and PWD/K (fig. S5B).…”

“…Diversity outbred mESC eQTL and protein QTL analysis are previously published (15,37) and available at https://churchilllab.jax.org/qtlviewer/ DO_mESC and https://doi.org/10.6084/m9.figshare.22012850. Data for Dyrk1a were accessed through publicly available data sources.…”

Which mouse multiparental population is right for your study? The Collaborative Cross inbred strains, their F1 hybrids, or the Diversity Outbred population

“…The effects of biologically related factors other than genetic variation, such as age, on gene expression, and protein abundance have been studied in the DO mice ( Takemon et al 2021 ; Gerdes Gyuricza et al 2022 ). Omic studies have also been performed in embryonic stem cells derived from DO mice ( Skelly et al 2020 ; Aydin et al 2022 ). Genetic effects are strongly consistent between the CC and DO populations ( Keele et al 2021 ), supporting the use of both for joint analysis and validation of findings between populations.…”

Which mouse multiparental population is right for your study? The Collaborative Cross inbred strains, their F1 hybrids, or the Diversity Outbred population

“…Genetic relationships between molecular traits have been characterized in embryonic stem cells derived from DO mice (Skelly et al . 2020; Aydin et al . 2022).…”

“…Further studies have investigated how age influences gene expression and protein abundance in kidney and heart tissues from DO mice ( Takemon et al 2021; Gerdes Gyuricza et al 2022). Genetic relationships between molecular traits have been characterized in embryonic stem cells derived from DO mice (Skelly et al 2020; Aydin et al 2022). Genetic effects are strongly consistent between populations, supporting the use of both populations for joint analysis and validation of findings (Keele et al 2021).…”