Genome-wide association study of frailty and integrative functional analysis to elucidate its relationship with aging

Frailty is a clinically recognizable condition of vulnerability with aging, leading to adverse health outcomes. Previous GWASs on frailty have identified 59 susceptibility loci, although the underlying mechanisms remain unclear. We investigated the genetic determinants of frailty and biological relationship between frailty and aging genes in 14,664 middle-aged and elderly Korean participants. We performed cohort-based (KoGES and KNHANES) meta-analysis of GWAS and candidate gene approach focusing on the aging pathway, followed by functional analyses. Frailty was defined as three phenotypes based on the FRAIL scale, CHS index, and frailty index. A meta-analysis of the GWAS on frailty identified eight novel SNPs: rs73619247 (AKAP12), rs12672085 (SUGCT), rs10881815 (SH2D4B), rs1455725, rs118100642 (BANK1), rs141617066, rs10461403, and rs10499589 at suggestive significance of P < 1.00 × 10^-6. Associations of rs12672085 and rs10881815 with frailty were highly prominent in the population aged < 65 years. Expression of the mapped genes SH2D4B and SUGCT suggests that they were involved in partially different mechanisms between frailty and aging. AKAP12 and BANK1 expression increased in senescent cells, implying a shared biological pathway between aging and frailty. The addition of the significant meQTLs of SH2D4B, SUGCT, AKAP12, and BANK1 strengthens this functional evidence. A meta-analysis of the candidate gene approach identified associations between nine aging pathway-related SNPs (rs75157957, rs145023550, rs3743343, rs60104021, rs117824268, rs140227721, rs28399561, rs534328933, and rs117413319) and frailty (P < 1.00 × 10^-4). Thus, SNPs and mapped genes (e.g., AKAP12 and BANK1) may play crucial roles in frailty and are biologically inseparable from aging; however, aging and frailty may involve partially different pathways (e.g., SUGCT and SH2D4B).