(LB04-04-25) Identification of Cinnamoylglycine-Isoleucine Ratio as a Biomarker Linking Dietary Quality, Microbial Metabolism, and Diabetes Risk: The CARDIA study

Objectives: Diet and microbiota homeostasis are known risk factors of diabetes pathogenesis. Microbial metabolism has rarely been linked to dietary quality and diabetes risk. We explored the associations among dietary intake, plasma metabolites, and diabetes parameters in adults enrolled in the Coronary Artery Risk Development in Young Adults (CARDIA) study.

Methods: A total of 2296 non-diabetic CARDIA participants at Year 7 were included in this analysis of diet quality (DQ), plasma metabolome, glucose metabolism, and incident diabetes. DQ is defined as higher intake of fruit, vegetables, and whole grains and lower intake of red meat, processed meat, refined grains, sugar-sweetened beverages, and eggs. Spearman correlation analysis assessed the associations of plasma metabolites with DQ, fasting glucose, insulin, and HOMA index. Subsequent propensity matching of 131 incident diabetes cases diagnosed between Year 7 and 25 with controls yielded a paired dataset when examined by logistic regression. The predictive markers from these analyses were validated by Cox proportional hazard models on three sets of random samples of 2296 participants. Models were adjusted for age, sex, body mass index, and fasting glucose. To account for multiple comparisons, a false discovery rate of q < 0.10 was applied.

Results: Among 611 plasma metabolites, cinnamoylglycine, a metabolite of joint metabolism from phenylalanine gut microbiome fermentation and hepatic glycine conjugation, was positively correlated with DQ. Cinnamoylglycine was inversely associated with incident diabetes risk (odds ratio [OR] =0.66; 95% C.I. 0.49, 0.87), with variable importance in projection (VIP) score ranked top 2 in the total 611 plasma metabolites. Isoleucine was inversely correlated with DQ and positively associated with incident diabetes risk (OR=1.98; 95% C.I. 1.36, 2.87), with VIP score ranked 63. This apparent contrast between cinnamoylglycine and isoleucine led to the designation of cinnamoylglycine/isoleucine (C/I) ratio as a predictive indicator of DQ and incident diabetes risk (OR= 0.61; 95% CI 0.46, 0.82). The ratio remained robust in randomly selected sample validation (hazard ratio [HR] 0.75; 95% C.I. 0.59, 0.96).

Conclusions: The C/I ratio was associated with incident diabetes risk.

Funding Sources: NIH/NHLBI 1R01HL150053