To our knowledge, this study is the first to address the link between dietary AAs composition and its serum profiles in relation to incidence of T2DM. To evaluate the overall quality of dietary AAs compositions, AACI was developed in this study by assessing the extent to which overall dietary AAs compositions adherence to the HRAAP. In the two prospective cohorts, the AACI was consistently and positively associated with six serum AAs and the incidence of T2DM. Among the six serum AAs, serum valine, glutamic acid and histidine consistently and partially mediated the association between the AACI and T2DM in the two cohorts.
In this study, using two prospective cohorts, a positive association between the AACI and incidence of T2DM was consistently observed, suggesting that participants with low quality of overall dietary essential AAs composition may have a higher risk of T2DM. Based on the concept of the HRAAP, biological value of individual AAs can be influenced by overall dietary AAs compositions, resulting in different absorbed-, utilized- and metabolic-rates of dietary AAs(8–11). Inadequate dietary AAs compositions therefore may play important roles in the development of T2DM. This study also demonstrated that when additionally adjusted for other known dietary risk factors for T2DM including intake of total protein (29), fiber (30), saturated fat (31), and overall diet quality (32), the relationship between the AACI and incidence of T2DM was still significant, further indicating that inadequate dietary AAs composition was likely an important dietary factor for residual risk of T2DM. These findings are supported by previous study showing that risk of pre-diabetes varied with different dietary AAs patterns (33), and they are also supported by cell and animal studies with the fact that feeding with mixture of AAs rather than a single AA alone could promote development of insulin-resistance and β-cell dysfunction in rodents (34, 35). Taken together, these accumulating evidences suggest that dietary AAs composition is likely an important risk factor in prevention and management of T2DM.
Previous studies have reported that individual dietary AAs intake cannot commonly be reflected in serum AAs levels (3, 12, 13), which makes current study of this issue lack compelling evidence for understanding the relationship between dietary AAs and T2DM. To fill this gap, this study further examined the association between the AACI and serum profiles of AAs in the two cohorts. The AACI was associated with eight serum AAs in the HPHS, and it was associated with nine serum AAs in the HDNNCDS. Although the difference of sample size and heterogeneity between the two cohorts are possible reasons for these discrepancy results, the serum valine, isoleucine, glycine, glutamic acid, phenylalanine and histidine were consistently observed to be associated with the AACI in the two cohorts, demonstrating that dietary AAs composition would influence the absorbed-, utilized- and metabolic-rates of AAs, which can be reflected in the serum AAs profiles, supporting the concept of the HRAAP for the impact of dietary AAs composition on their serum profiles. Studies regarding this issue were relative scarce, but a recent study has reported that it is overall dietary pattern rather than dietary BCAA was associated with serum BCAA, which partially support the observations in this study (36).
To further clarified whether and how the association between the AACI and incidence of T2DM mediated by serum AAs profiles for understanding the impact of dietary AAs composition on their serum profiles in relation to subsequent T2DM, mediation analyses were performed in the two cohorts. Although the AACI was consistently associated with six serum AAs in the two cohorts, only serum valine, glutamic acid and histidine consistently and partially mediated this association in the two cohorts with total mediation effects of 33.4% and 54.6%, respectively, indicating that the association between inappropriate dietary AAs composition and increased risk of T2DM may be largely mediated by increasing serum valine and glutamic acid, and by decreasing serum histidine levels. Serum valine, as one of the BCAAs, has been consistently identified to be an important metabolite associated with insulin-resistance in previous epidemiologic studies (15–19, 37), probably through inhibition of insulin receptor substrate-1, and a recent study has reported that increased serum BCAAs probably produce more catabolic intermediates including propionyl CoA and succinyl CoA, resulting in accumulation of incompletely oxidized fatty acids and glucose (38). Moreover, BCAAs can produce glutamic acids, catalyzing by branched-chain aminotransferase, and the glutamic acids has been reported to be associated with insulin resistance in the Framingham offspring study (39). The potential protective effect of histidine on glucose/insulin homeostasis has been documented in previous research, probably by suppressing inflammatory factors and hepatic glucose production through central insulin action (40). Based on the findings in this study and these above mechanisms, this study therefore speculated that an inappropriate dietary AAs composition may cause varied absorbed-, utilized- and metabolic-rate of AAs, showing disordered plasma AAs profiles, and the dysregulated valine, glutamic acid and histidine may induce insulin resistance and subsequent diabetes.
Previous studies regarding this issue have frequently focused on the association between individual AA and T2DM, few studies have considered the overall quality of dietary AAs compositions. This study demonstrated that inadequate dietary AAs composition was associated with increased incidence of T2DM, and the association between the AACI and serum AAs profiles and the potential mediation effects further strength our findings, which would improve our understanding of the pathobiology and mechanisms of T2DM, and facilitate selection of potential therapeutic and intervention strategies for T2DM. Moreover, the findings of this study also emphasized that future study regarding dietary AAs and T2DM should consider dietary AAs as a whole rather than isolating individual AAs from diet in prevention and management of T2DM.
The strength of our study is that it included two prospective cohorts with relatively large OGTT sample of nutritional and metabolic analyses in this issue. Further, this study established the AACI based on the concept of the HRAAP for evaluating the overall quality of dietary AAs composition, demonstrated and emphasized the importance of dietary AAs composition. Third, the observed association between the AACI and T2DM was robust because it was observed in the two independent cohorts and it persisted after adjustment for a wide range of available confounding factors. However, we also recognize that our study has certain limitations. First, the study was observational in nature, and we cannot rule out the influence of unmeasured confounding factors. Besides, no amount of adjustment can deal completely with confounding in an observational context. Second, this study only included Asian subjects, which is likely to limit the generalizability of our findings to other ethnic populations. However, given the roles of HRAAP, and the association between serum AAs and T2DM have been shown to be generally consistent across different ethnics. We would therefore expect that our observations would hold across other populations.