Search

Background: Experimental disruption of both gene copies (“knockout”) in model organisms has proven useful to understand gene function. Identification of “human knockouts” (i.e., individuals in which null mutations disrupt both copies of a given gene) should therefore provide valuable insights to a gene function that could be of direct relevance to humans. Objective and Methods: We leverage naturally-occurring null mutations and consanguinity in the human population in order to: 1) identify humans knocked out for genes; and 2) characterize the phenotypic consequences of complete gene disruption. We performed whole-exome sequencing in 7,078 individuals living in Pakistan, a region of the world with high levels of consanguinity, and tested whether knockouts differed from wild-type participants across a range of >200 cardiometabolic traits. Results: Median length of genome-wide homozygosity among Pakistani participants was 6-7 times higher than participants of European (CEU, TSI), East Asian (CHB, JPT, CHD) and African ancestries (YRI, MKK), respectively. In Pakistanis, we were able to enumerate 47,656 mutations predicted to be null (nonsense, frameshift, or canonical splice-site); variants were further filtered based on MAF, transcript position, splice sites, and conservation to identify 36,850 “high-confidence” null mutations across 12,131 autosomal genes. Across all participants, 961 distinct genes were completely disrupted by homozygous null mutations. 1,306 participants (18.4%) had at least one gene knocked out. In a phenotypic screen, homozygosity for null mutations at APOC3 was associated with absent plasma apolipoprotein C-III levels; at PLAG27 , with absent enzymatic activity of lipoprotein-associated phospholipase A2 in the blood; at CYP2F1 , with higher plasma interleukin-8 concentrations; and at either A3GALT2 or NRG4 , with markedly reduced plasma insulin C-peptide concentrations. After physiologic challenge with oral fat, APOC3 knockouts, when compared to wild-type family members, displayed marked blunting of the usual post-prandial rise in plasma triglycerides. Conclusion: These observations provide a roadmap to understand the consequences of complete disruption of a large fraction of genes in the human genome.