Branched-chain and aromatic amino acids are predictors of insulin resistance in young adults

Abstract

Objective: Branched-chain and aromatic amino acids are associated with the risk for future type 2 diabetes; however, the underlying mechanisms remain elusive. We tested whether amino acids predict insulin resistance index in healthy young adults.

Research design and methods: Circulating isoleucine, leucine, valine, phenylalanine, tyrosine, and six additional amino acids were quantified in 1,680 individuals from the population-based Cardiovascular Risk in Young Finns Study (baseline age 32 ± 5 years; 54% women). Insulin resistance was estimated by homeostasis model assessment (HOMA) at baseline and 6-year follow-up. Amino acid associations with HOMA of insulin resistance (HOMA-IR) and glucose were assessed using regression models adjusted for established risk factors. We further examined whether amino acid profiling could augment risk assessment of insulin resistance (defined as 6-year HOMA-IR >90th percentile) in early adulthood.

Results: Isoleucine, leucine, valine, phenylalanine, and tyrosine were associated with HOMA-IR at baseline and for men at 6-year follow-up, while for women only leucine, valine, and phenylalanine predicted 6-year HOMA-IR (P < 0.05). None of the other amino acids were prospectively associated with HOMA-IR. The sum of branched-chain and aromatic amino acid concentrations was associated with 6-year insulin resistance for men (odds ratio 2.09 [95% CI 1.38-3.17]; P = 0.0005); however, including the amino acid score in prediction models did not improve risk discrimination.

Conclusions: Branched-chain and aromatic amino acids are markers of the development of insulin resistance in young, normoglycemic adults, with most pronounced associations for men. These findings suggest that the association of branched-chain and aromatic amino acids with the risk for future diabetes is at least partly mediated through insulin resistance.

Figure 1

Cross-sectional and prospective associations of amino acids with HOMA-IR in the Cardiovascular Risk in Young Finns Study. Linear regression models were adjusted for baseline age, BMI, systolic blood pressure, HDL cholesterol, triglycerides, smoking status, and physical activity index. Prospective associations (filled circles) for insulin resistance index at 6-year follow-up were further adjusted for baseline HOMA-IR. Association magnitudes are in standardized units of 1 SD HOMA-IR per 1 SD amino acid concentration. Error bars indicate 95% CIs. Amino acid score: sum of isoleucine, leucine, valine, phenylalanine, and tyrosine concentrations. *P < 0.05; ★P < 0.005.

Figure 2

Associations of amino acids with baseline and 6-year fasting glucose. Linear regression models were adjusted for baseline age, BMI, systolic blood pressure, HDL cholesterol, triglycerides, smoking status, and physical activity index. Prospective associations (filled squares) were further adjusted for baseline glucose. Association magnitudes are in standardized units of 1 SD glucose per 1 SD difference in amino acid concentration. Error bars indicate 95% CIs. *P < 0.05; ★P < 0.005.

Figure 3

ORs and 95% CIs for insulin resistance at 6-year follow-up (HOMA-IR ≥90th percentile) per 1 SD difference in baseline amino acid concentration. Logistic regression models were adjusted for baseline HOMA-IR, age, BMI, systolic blood pressure, HDL cholesterol, triglycerides, smoking status, and physical activity index. *P < 0.05; ★P < 0.005.