Functional Characterization Improves Associations between Rare Non-Synonymous Variants in CHRNB4 and Smoking Behavior

Smoking is the leading cause of preventable death worldwide. Accordingly, effort has been devoted to determining the genetic variants that contribute to smoking risk. Genome-wide association studies have identified several variants in nicotinic acetylcholine receptor genes that contribute to nicotine dependence risk. We previously undertook pooled sequencing of the coding regions and flanking sequence of the CHRNA5, CHRNA3, CHRNB4, CHRNA6 and CHRNB3 genes and found that rare missense variants at conserved residues in CHRNB4 are associated with reduced risk of nicotine dependence among African Americans. We identified 10 low frequency (,5%) non-synonymous variants in CHRNB4 and investigated functional effects by co-expression with normal a3 or a4 subunits in human embryonic kidney cells. Voltage-clamp was used to obtain acetylcholine and nicotine concentration–response curves and qRT-PCR, western blots and cell-surface ELISAs were performed to assess expression levels. These results were used to functionally weight genetic variants in a gene-based association test. We find that there is a highly significant correlation between carrier status weighted by either acetylcholine EC50 (b=20.67, r 2 = 0.017, P = 261024) or by response to low nicotine (b=20.29, r2 = 0.02, P = 661025) when variants are expressed with the a3 subunit. In contrast, there is no significant association when carrier status is unweighted (b=20.04, r2 = 0.0009, P = 0.54). These results highlight the value of functional analysis of variants and the advantages t