Objective 1: Conduct and analyze dietary intervention studies to validate gene-diet interactions and identify the underlying mechanisms using omic technologies. Sub-Objective 1A: To characterize the response of cardiometabolic, epigenetics and other age-related biomarkers and the microbiome to diets differing in saturated fat and prebiotics content (animal-based diet versus plant-based diet) in individuals carrying CC and TT genotypes at the common APOA2 -265T>C (rs5082) SNP using a short-term crossover, randomized feeding study, and to elucidate the physiological mechanism(s) by which diet impinges on metabolic pathways through APOA2 genotypes. Sub-Objective 1B: To characterize the TCF7L2-by-diet interaction with respect to those type 2 diabetes (T2D) and cardiovascular disease (CVD) risk factors identified in observational studies for validation in the context of a short-term randomized controlled feeding study (low-fat diet versus Mediterranean diet), and to elucidate the molecular mechanisms responsible for these GxD interactions using epigenetics and metabolomics. Sub-Objective 1C: To develop polygenic risk scores (PRS) predicting the changes in and relationships between cardiovascular disease (CVD) risk factors and disease incidence in response to long-term (>=1 y) dietary interventions [Mediterranean diet (MedDiet) or Low-fat control diet]. Objective 2: Identify genomic, epigenomic, metabolomic, and microbiome-related biomarkers that sustain healthy aging, and define specific personalized dietary, physical activity, and other lifestyle factors associated with optimal health of older adults. Subobjective 2A: To identify genetic and dietary factors that modify CPT1A methylation and cardio-metabolic traits. Subobjective 2B: To identify interactions between the genome, epigenome and diet and lifestyle on lipid profiles that signify CMD risk.
Promoting healthy aging by tailoring nutritional guidance based on a person's genetic makeup is an emerging science that has great promise. The Nutrition and Genomics lab is a pioneer in this area and focuses its research on the role of precision nutrition and cardiometabolic diseases – the leading cause of death in the United States. Our approach harnesses the availability of tremendous computing power and huge datasets from existing cohorts to study the crosstalk between habitual diets and the genome to identify gene-by-diet interactions that sustain individual optimal health for older adults. This objective will be accomplished using Big Data analytics of omics data (i.e., genome-wide datasets on gene and protein expression, genetic variation, methylation, and metabolite levels). We also conduct short-term feeding studies in people preselected based on particular genotypes to validate gene-by-diet interactions revealed by previous observational studies and, using multi-omic data integration (i.e., genomics, epigenomics, microbiomics, and metabolomics) methods, identifying the mechanisms underlying such interactions. This research will generate new knowledge on how non-modifiable and modifiable factors interact to prevent cardiovascular diseases and type 2 diabetes. Further, it will contribute much-needed evidence and tools to define and implement personalized nutrition as a common practice for the benefit of all stakeholders.
Progress was made on the two objectives, both of which fall under National Program 107. The National Program in Human Nutrition is designed to improve the health of all Americans throughout their lifespan. Under Objective 1, our goal is to conduct and analyze dietary intervention studies to validate gene-diet interactions and identify the underlying mechanisms using omic technologies. In support of this objective, and in collaboration with investigators in the U.K. and other U.S. sites, we examined the genetic, metabolic, microbiome, and meal composition/ context contributions to postprandial metabolic responses in the PREDICT (Personalised REsponses to DIetary Composition Trial) 1 Study. This study has enrolled 1,102 twins and unrelated healthy adults in the U.K. and U.S. For this report, we will focus on postprandial responses and microbiome characteristics. Regarding postprandial response, meal-induced metabolic changes trigger an acute inflammatory response, contributing to chronic inflammation and associated diseases. Therefore, we aimed to characterize variability in postprandial inflammatory responses using traditional (IL-6) and novel [glycoprotein acetylation (GlycA)] biomarkers of inflammation and dissect their biological determinants. We measured postprandial glucose, triglyceride (TG), IL-6, and GlycA responses at multiple intervals after sequential mixed-nutrient meals (0 h and 4 h) in PREDICT1 participants. Our results show that the postprandial changes in GlycA and IL-6 concentrations were highly variable between individuals. Peak postprandial TG and glucose concentrations were significantly associated with 6-h GlycA (both P < 0.001) but not with 6-h IL-6 (both P > 0.26). A random forest model revealed that the maximum TG concentration was the strongest postprandial TG predictor of postprandial GlycA. Structural equation modeling showed that visceral fat mass (VFM) and fasting TG were most strongly associated with fasting and postprandial GlycA. Network Mendelian randomization demonstrated a causal link between VFM and fasting GlycA, mediated by fasting TG. Individuals eliciting enhanced GlycA responses had higher predicted cardiovascular disease (CVD) risk than the cohort. In summary, GlycA and its associations with TG metabolism highlight the importance of its modulation in concert with obesity to reduce GlycA and associated low-grade inflammation-related diseases. Our progress related to nutrition-microbiome-health involved PREDICT1 and the Coronary Diet Intervention with Olive Oil and Cardiovascular Prevention (CORDIOPREV) Studies. First, we performed metagenomic sequencing of gut microbiomes from PREDICT1 participants. We found significant associations between microbes and nutrients, foods, food groups, and general dietary indices, driven primarily by the presence and diversity of healthy and plant-based foods. Overall microbiome composition was predictive for a large panel of cardiometabolic markers, including fasting and postprandial glycemic, lipemic and inflammatory indices. The panel of microbes associated with healthy dietary habits overlapped with those associated with favorable cardiometabolic markers, indicating that we can potentially stratify the gut microbiome into generalizable health levels in individuals without clinically manifest disease. Moreover, we investigated the association between dietary type 2 diabetes (T2D) prevention and remission and the microbiome. For this purpose, we examined T2D dietary prevention on all CORDIOPREV patients without T2D at baseline (n=462). The risk of T2D was assessed, after a five-year follow-up, by Cox analysis. Linear discriminant analysis effect size (LEfSe) analysis showed a different baseline gut microbiota in patients who developed T2D consuming low-fat (LF) and Mediterranean (Med) diets. Higher Paraprevotella and lower Gammaproteobacteria and B. uniformis were associated with T2D risk when an LF-diet was consumed. Conversely, higher Saccharibacteria, Betaproteobacteria, and Prevotella were associated with T2D risk when a Med-diet was consumed, suggesting that different interactions between the microbiome and dietary patterns may partially determine T2D incidence. For T2D remission, we included 110 newly diagnosed T2D CORDIOPREV patients. We evaluated whether baseline gut microbiota composition improves the identification of patients undergoing T2D remission while consuming the two dietary models for 5 years. Using LEfSe, we showed that the responder group's gut microbiota was characterized by the Ruminococcus genus of the Lachnospiraceae family. Conversely, base-line gut microbiota in the non-responder group was enriched in the Porphyromonadaceae family and Parabacteroides genus. Therefore, our results reveal a gut microbiota profile associated with T2D remission and provide evidence of a role of the microbiome as a predictive factor for response to diet-induced T2D remission. For Objective 2, we have made substantial progress, specifically in developing computational models for cardiometabolic disease, responsible for decreased longevity and poorer cardiovascular outcomes. Our objective was to define a molecular basis for cardiometabolic stress and assess its association with cardiovascular prognosis. For this purpose, we conducted a prospective observational cohort study in a population-based setting across two centers Boston Puerto Rican Health Study (BPRHS) with a Hispanic population and Atherosclerosis Risk in Communities (ARIC) Study with White and African American populations. The primary exposure was metabolite profiles across both cohorts. Outcomes included associations with multisystem cardiometabolic stress and all-cause mortality and incident CHD (in ARIC). BPRHS participants had higher prevalent cardiometabolic risk relative to those in ARIC. Multisystem cardiometabolic stress was defined for the BPRHS as a composite of hypothalamic-adrenal axis activity, sympathetic activation, blood pressure, dyslipidemia, insulin resistance, visceral adiposity, and inflammation. Two hundred six metabolites associated with cardiometabolic stress were identified in the BPRHS. A parsimonious metabolite-based score was created and associated with cardiometabolic stress in the BPRHS; this score was applied to shared metabolites in the ARIC study, demonstrating significant associations with coronary heart disease (CHD) all-cause mortality after multivariable adjustment at a 30-year horizon. These results underscore the shared molecular pathophysiology of metabolic dysfunction, CVD, and longevity and suggest pathways for modification to improve prognosis across all linked conditions. In addition to biomarkers, clinical practice guidelines recommend assessing subclinical atherosclerosis (SA) using imaging techniques. Therefore, we aimed to develop a machine-learning model based on routine, quantitative and easily measured variables to predict the presence and extent of SA in young, asymptomatic individuals participating in the Progression of Early Subclinical Atherosclerosis [PESA] Study. The Elastic Net (EN) model was built to predict SA extent. The performance of the model for the prediction of SA was compared with traditional CVD risk scores. An external cohort was used for validation. The EN-PESA yielded a c-statistic of 0.88 for the prediction of generalized SA. Moreover, EN-PESA was found to be a predictor of 3-year progression, independent of the baseline extension of SA. In summary, the EN-PESA model uses age, systolic blood pressure, and 10 commonly used blood/urine tests and dietary intake values to identify young, asymptomatic individuals with an increased risk of CVD based on their extension and progression of SA. These individuals are most likely to benefit from dietary and pharmacological treatments.
1. Intake of carbohydrates and fats influences the risk of metabolic diseases. The specific role of a signaling mechanism, known as methylation, that controls genes associated with the risk of metabolic diseases such as hypertriglyceridemia, obesity, type 2 diabetes, hypertension, and metabolic syndrome remains unknown. ARS-funded researchers in Boston, Massachusetts, examined whether carbohydrate and fat intakes influenced methylation and the risk of metabolic diseases (i.e., high blood lipids, obesity, type 2 diabetes, and hypertension) in 3,954 people representing Hispanic, Black and White populations. The analyses demonstrated strong associations of a specific methylation marker with metabolic characteristics such as body mass index, triglyceride, glucose, and hypertension in each population and all three populations combined. The results demonstrated that carbohydrate intake induces a specific methylation site that reduces the risk of all metabolic diseases examined. In contrast, fat intake inhibits a specific methylation site and increases the risk of such metabolic diseases. These findings identify how balancing carbohydrate and fat intake can have a causal effect on the risk of metabolic diseases that currently affects millions of Americans.
Lai, C., Parnell, L.D., Smith, C.E., Guo, T., Sayols-Baixeras, S., Aslibekyan, S., Tiwari, H.K., Irvin, M.R., Bender, C., Fei, D., Hidalgo, B., Hopkins, P., Absher, D.M., Province, M., Elosua, R., Arnett, D.K., Ordovas, J.M. 2020. Carbohydrate and fat intake associated with risk of metabolic diseases through epigenetics of CPT1A. American Journal of Clinical Nutrition. 112(5):1200–1211. https://doi.org/10.1093/ajcn/nqaa233.
Smith, C., Parnell, L.D., Lai, C., Rush, J.E., Freeman, L.M. 2021. Investigation of diets associated with dilated cardiomyopathy in dogs using foodomics analysis. Scientific Reports. 11:15881. https://doi.org/10.1038/s41598-021-94464-2.