Introduction:
Metabolic syndrome (MetS) is a prominent risk factor for both cardiovascular disease (CVD) and type 2 diabetes (T2D). MicroRNAs (miRs) are small noncoding RNA molecules that target messenger RNAs to alter gene expression. Circulating miRs have been studied as potential clinically meaningful biomarkers of risk for MetS as they are readily measured from blood. Though associations among MetS components and social constructs of race, ethnicity, and gender have previously been established, grouping according to these constructs alone may conflate genetic ancestry with the environmental/behavioral effects of being characterized as a particular gender, race, or ethnicity. The purpose of this study was to identify differences in circulating miRs associated with demographic factors and MetS components to better illuminate the nuanced health impacts of race, ethnicity and gender.
Methods:
This was a secondary analysis of a subset of participants (N=1000) from the Diabetes Prevention Program (DPP). A custom Fireplex assay was used to quantify miRs from banked plasma collected at baseline. Correlations between miRs and metabolic syndrome components were assessed by Pearson’s correlation coefficient. Multivariable linear models adjusted for age and weight were used to analyze associations between gender, race, ethnicity, and miR expression. The Benjamini-Hochberg false discovery rate (FDR) method was applied.
Results:
The sample was 68% female, 19% Black, 15% Hispanic, and mean age was 52 ± 10 years. After adjusting for multiple comparisons (FDR<0.05), female gender was positively associated with miRs 320c and 320a, and negatively associated with miR 122. Among 17 miRs associated with race and ethnicity, five (23a, 24, 27a, 92a, and 146a) had different directions of the race and ethnicity-miR associations across Black, Hispanic, and Other race and ethnic groups. Of the miRs significantly associated with gender, race, and ethnicity, 12 were significantly correlated with one or more component of MetS. Two miRs were significantly correlated with fasting glucose (22, 15a), four with HDL cholesterol (122, 23a, 22, 192), four with triglycerides (122, 106b, 22, 192), six with systolic blood pressure (122, 106b, 23a, 27a, 146a, 221), two with diastolic blood pressure (24, 27a), and seven with waist circumference (320c, 320a, 106b, 23a, 24, 27).
Conclusions:
Significant differences in the enrichment of MetS component-associated miRs were identified across gender, race, and ethnic groups. Obtaining a greater understanding of demographic differences in miRs may aid in measuring the relationship between constructs of race, ethnicity, and gender and clinical parameters, and potentially inform future studies of circulating miRs as quantifiable diagnostic, prognostic, and/or therapeutic biomarkers for the impacts of social determinants within the context of underlying genetic ancestry.