Plasma Proteome Variation and its Genetic Determinants in Children and Adolescents

Abstract

The levels of specific proteins in human blood are the most commonly used indicators of potential health-related problems1. Understanding the genetic and other determinants of the human plasma proteome can aid in biomarker research and drug development. Diverse factors including genetics, age, sex, body mass index (BMI), growth and development including puberty can affect the circulating levels of proteins2-5. Affinity-based proteomics can infer the relationship between blood protein levels and these factors at a large scale6-10. Compared to these methods, mass spectrometry (MS)-based proteomics provides much higher specificity of identification and quantification11-13, but existing studies are limited by small sample sizes or low numbers of quantified proteins14-17. Here we aim to elucidate to which extent genomic variation affects plasma protein levels across diverse age ranges and cohort characteristics. Employing a streamlined and highly quantitative MS-based plasma proteomics workflow, we measured the plasma proteome of 2,147 children and adolescents. Levels of 90% of these proteins were significantly associated with age, sex, BMI or genetics. More than 1,000 protein quantitative trait loci (pQTLs) - a third of which were novel - regulated protein levels between a few percent and up to 30-fold. These replicated excellently in an independent cohort of 558 adults, with highly concordant effect sizes (Pearsons r > 0.97). We developed a framework to eliminate artefactual pQTLs due to protein-altering variants, paving the way for large-scale interrogation of pQTLs using MS-based proteomics. Our data reveal unexpectedly extensive genetic impacts on plasma protein levels, consistent from childhood into adulthood. These findings have implications for biomarker research and drug development. HighlightsO_LIFirst large-scale proteome-wide and genome-wide association study in children and adolescents C_LIO_LIMS-based proteomics achieves very high specificity and quantitative accuracy C_LIO_LIRobust plasma protein trajectories during development predict age and body mass index C_LIO_LILargest set of pQTLs for plasma proteome by MS-based proteomics C_LIO_LIpQTLs are highly replicable between children and adults C_LIO_LILarge-scale pQTL identification enables generic drug target validation C_LI