Genetic evidence of a causal effect of insulin resistance on branched-chain amino acid levels

Fasting plasma levels of branched-chain amino acids (BCAAs) are associated with insulin resistance, but it remains unclear whether there is a causal relation between the two. We aimed to disentangle the causal relations by performing a Mendelian randomisation study using genetic variants associated with circulating BCAA levels and insulin resistance as instrumental variables. We measured circulating BCAA levels in blood plasma by NMR spectroscopy in 1,321 individuals from the ADDITION-PRO cohort. We complemented our analyses by using previously published genome-wide association study (GWAS) results from the Meta-Analyses of Glucose and Insulin-related traits Consortium (MAGIC) (n = 46,186) and from a GWAS of serum BCAA levels (n = 24,925). We used a genetic risk score (GRS), calculated using ten established fasting serum insulin associated variants, as an instrumental variable for insulin resistance. A GRS of three variants increasing circulating BCAA levels was used as an instrumental variable for circulating BCAA levels. Fasting plasma BCAA levels were associated with higher HOMA-IR in ADDITION-PRO (beta 0.137 [95% CI 0.08, 0.19] p = 6 x 10(-7)). However, the GRS for circulating BCAA levels was not associated with fasting insulin levels or HOMA-IR in ADDITION-PRO (beta -0.011 [95% CI -0.053, 0.032] p = 0.6 and beta -0.011 [95% CI -0.054, 0.031] p = 0.6, respectively) or in GWAS results for HOMA-IR from MAGIC (beta for valine-increasing GRS -0.012 [95% CI -0.069, 0.045] p = 0.7). By contrast, the insulin-resistance-increasing GRS was significantly associated with increased BCAA levels in ADDITION-PRO (beta 0.027 [95% CI 0.005, 0.048] p = 0.01) and in GWAS results for serum BCAA levels (beta 1.22 [95% CI 0.71, 1.73] p = 4 x 10(-6), beta 0.96 [95% CI 0.45, 1.47] p = 3 x 10(-4), and beta 0.67 [95% CI 0.16, 1.18] p = 0.01 for isoleucine, leucine and valine levels, respectively) and instrumental variable analyses in ADDITION-PRO indicated that HOMA-IR is causally related to higher circulating fasting BCAA levels (beta 0.73 [95% CI 0.26, 1.19] p = 0.002). Our results suggest that higher BCAA levels do not have a causal effect on insulin resistance while increased insulin resistance drives higher circulating fasting BCAA levels.