Branched-chain and aromatic amino acid profiles and diabetes risk in Chinese populations

Abstract

Recent studies revealed strong evidence that branched-chain and aromatic amino acids (BCAAs and AAAs) are closely associated with the risk of developing type 2 diabetes in several Western countries. The aim of this study was to evaluate the potential role of BCAAs and AAAs in predicting the diabetes development in Chinese populations. The serum levels of valine, leucine, isoleucine, tyrosine, and phenylalanine were measured in a longitudinal and a cross sectional studies with a total of 429 Chinese participants at different stages of diabetes development, using an ultra-performance liquid chromatography triple quadruple mass spectrometry platform. The alterations of the five AAs in Chinese populations are well in accordance with previous reports. Early elevation of the five AAs and their combined score was closely associated with future development of diabetes, suggesting an important role of these metabolites as early markers of diabetes. On the other hand, the five AAs were not as good as existing clinical markers in differentiating diabetic patients from their healthy counterparts. Our findings verified the close correlation of BCAAs and AAAs with insulin resistance and future development of diabetes in Chinese populations and highlighted the predictive value of these markers for future development of diabetes.

Figure 1

Heat map of AA and metabolic marker levels (a) and scatter plot of combined score (b) in individuals of healthy control (HC) and diabetes (DM) from longitudinal study. Abbreviations used: Score, the first decomposed principal component derived from the abundance of the five AAs; DM, future diabetes; DP, diastolic blood pressure; Glucose0, fasting glucose; Glucose120, 2 h glucose; HC, future healthy controls; HDL, high-density lipoprotein-cholesterol; HOMA-Beta = 20*INS0/(Glucose0-3.5); HOMA-IR = Glucose0*Glucose120/22.5; INS0, fasting insulin; INS120, 2 h insulin; Isoleu, isoleucine; LDL, low-density lipoprotein-cholesterol; Leu, leucine; Matsuda index = 10000/(Glucose0*Glucose120*INS0*INS120)1/2; Phenyl, phenylalanine; SP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; Tyro, tyrosine; Val, valine.

Figure 2

Heat map of Spearman correlation coefficients between AAs and metabolic markers (a), relationship between combined score and HOMA-IR (b), Scatter plot of combined score (c) and HOMA-IR (d) in HL, OW/OB, and DM individuals from cross sectional study. Abbreviations used: ALT, alanine aminotransferase; AST, aspartate aminotransferase; Score, the first decomposed principal component derived from the abundance of the five AAs; DM, overweight or obesity with type 2 diabetes; DP, diastolic blood pressure; γ-GT, gamma-glutamyl trans-supeptidase; Glucose0, fasting glucose; Glucose120, 2 h glucose; HDL, high-density lipoprotein-cholesterol; HL, healthy lean; HOMA-Beta = 20*INS0/(Glucose0-3.5); HOMA-IR = Glucose0*Glucose120/22.5; INS0, fasting insulin; INS120, 2 h insulin; Isoleu, isoleucine; LDL, low-density lipoprotein-cholesterol; Leu, leucine; Matsuda index = 10000/(Glucose0*Glucose120*INS0*INS120)1/2; OW/OB, healthy overweight or obesity; Phenyl, phenylalanine; SP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; Tyro, tyrosine; Val, valine.