Amino acid intake, plasma metabolites, and incident type 2 diabetes risk: a systematic approach in prospective cohort studies.

Abstract

Background: Amino acid (AA) intake is thought to be closely related to the etiology of type 2 diabetes (T2D), although few prospective human studies have examined the association. The study prospectively examined inter-relationships among the intake of all 20 AAs, blood metabolome, and T2D incidence.

Methods: Prospective associations between 20 individual AA intake and T2D were examined among 201,113 participants from the Health Professionals Follow-up Study (HPFS), Nurses' Health Study (NHS), and NHSII who did not have T2D, cardiovascular disease, or cancer at baseline. Next, a multi-metabolite signature responsive to AA intake was derived through two sets of 7-day diet record (7DDR) assessments, and subsequently replicated in multiple cohorts. Finally, Cox regression models were used to determine and confirm the prospective associations of AA intake, plasma metabolite signatures, with T2D incidence.

Results: During 5,037,848 person-years of follow-up, 20,619 T2D cases were documented. Sulfur AAs (SAAs) and aromatic amino acids were significantly associated with higher T2D risk independent of other AAs. Comparing the highest with lowest quintiles of intake, the multivariable adjusted pooled hazard ratios (HRs) for T2D were 1.20 (95% CI: 1.13-1.27) for total SAA (P _{for trend} < 0.0001) and 1.14 (95% CI: 1.08-1.21) for total aromatic AAs (P _{for trend} < 0.0001). A total of 73 plasma metabolites were identified as responsive markers for total SAA intake, and 87 metabolites for total aromatic AA intake. Approximately 75% of these metabolites were commonly responsive within the individual AAs within the same class. The multi-metabolite signatures for SAAs and aromatic AAs intakes were replicated in an independent study [β = 0.30 or 0.31 per SD increase of SAA and aromatic AA intake, respectively, P < 0.0001 for both associations). Moreover, the signatures were associated with T2D incidence: the HRs for per SD change were 1.14 (95%CI: 1.08-1.20, P < 0.00001) for SAA signature and 1.18 (95%CI: 1.12-1.24, P < 0.00001) for aromatic AA signature. Mediation analysis showed that the metabolite signatures explained various degree (% ranging from 10.5(95%CI: 3.8-26.1) to 29.9(95%CI: 13.5-53.9)) for the associations between dietary SAAs/aromatic AAs and T2D incidence risk.

Conclusions: This study provides novel evidence that higher intake of sulfur and aromatic AAs is independently associated with an increased risk of T2D. Multiple plasma metabolites are responsive to these dietary AAs, potentially serving as objective markers for AA intake. Collectively, the metabolite signatures significantly predict a higher T2D risk, and mediate the positive associations to various degrees, corroborating findings from long-term dietary assessments.