Sponsored by:
Background. In average, Americans consume more than 3,400 milligrams of sodium per day, an amount much higher than recommendations in the US 2015-2020 Dietary Guidelines. Studies have shown that dietary intake high in sodium, fat and sucrose are strongly related to an increased risk for cardiovascular disease (CVD). However, the molecular mechanisms by which these nutrient impacts CVD are not fully understood. The aim of this analysis is to conduct an unbiased transcriptome analysis in African Americans (AA), a population with disproportionate burden of CVD. Methods The analysis is a cross-sectional study of whole blood transcriptome mRNA sequencing (mRNA-seq) of 472 AA from the MHGRID and GENE-FORECAST studies. Random Forests (RF) was used to define a transcriptome profile empirically related to salt intake, and evaluate the relationships between that profile and other diet variable (fat, sucrose and calories) and cardiovascular traits including hypertension status (HTN), systolic and diastolic blood pressure (SBP and DBP) and kidney function (eGFR). Weighted Gene Co-expression Network Analysis (WGCNA) was subsequently conducted to identify clusters of co-expressed genes within the transcriptome profile defined by RF. Results. From the 13,359 protein coding transcripts that passed QC filters, RF defined a transcriptome profile of 816 mRNA that classified subjects with high vs low sodium intake with an area under the curve (AUC) of 0.82. That transcriptome profile classified fat, sucrose, calorie intake and the cardiovascular traits with AUC=76 (fat), AUC=67 (sucrose), AUC=0.77 (calorie), AUC=0.72 (HTN), AUC=0.71 (SBP), AUC=0.67 (DBP), AUC=0.71 (eGFR). WGCNA analysis identified 3 network modules that include genes such as ALMS1, RAP1, CXCR4, TRAF3, and IL16 involved in pathways that regulate vascular tone, flow, remodeling and sustained high blood pressure. Conclusions. This study employed machine learning predictive models to examine the blood transcriptome and define molecular signatures of mRNA transcripts associated with sodium intake and cardiovascular traits, in one of the largest mRNA-seq sample set of AA to date. The findings provide new insights into biomarkers and molecular pathways that may mediate the effects of sodium, fat and sucrose on cardiovascular health.
Amadou Gaye, National Human Genome Research Institute, National Institutes of Health
Malak Abbas, National Human Genome Research Institute, National Institutes of Health
Paule V. Joseph, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health
Gabriel Goodney, National Human Genome Research Institute, National Institutes of Health
Henry Gong, National Human Genome Research Institute, National Institutes of Health
Lisa DeRoo, National Human Genome Research Institute, National Institutes of Health
Sharon K. Davis, National Human Genome Research Institute, National Institutes of Health
Gary H. Gibbons, National Human Genome Research Institute, National Institutes of Health