Matthew Nayor’s research while affiliated with University of Massachusetts Chan Medical School and other places

Current genetic discovery methods are largely restricted to profiling circulating molecules or genetic architecture, limited in use of tissue-based molecular genetics to identify pathogenic and therapeutic targets. Here, we leverage a multi-level genetic discovery platform integrating population-level proteomics with functional genomic analyses based on human coronary artery tissue to reveal determinants of coronary disease susceptibility. Using aptamer-based proteomics (~7,000 aptamers) across ~3,000 individuals, we first identified the circulating proteome of prevalent and incident coronary artery calcium (CAC), a sensitive marker of subclinical coronary artery disease (CAD), with causal implication in calcified plaque formation or disease phenotypes via parallel genetic approaches (Mendelian randomization, MR) and proteome-wide association (PWAS). Identified proteins specified pathways of extracellular matrix remodeling, immune cell function, lipid metabolism, and inflammation. To resolve findings at a coronary tissue level, we performed the largest to date coronary artery-specific transcriptome-wide association study for CAC (TWAS; based on RNA-seq from 268 human coronary arteries) in >35,000 individuals, demonstrating enrichment of targets from the circulating proteome with supportive evidence by traditional MR approaches (NOTCH3, SPINK2, S100A12, RPP25, OAF, HS6ST3, TNFSF12, GPC6), several implicated in CVD-adjacent biological mechanisms. Phenome-wide association and single cell transcriptomics in human coronary arteries across atherosclerosis implicated targets in tissue-specific disease mechanisms. Finally, using coronary artery-specific functional genomic annotations of chromatin structure, conformation, and accessibility, we identified trans regulation of two of these genes (GPC6, RPP25) by CAC GWAS-significant SNPs, resolving targets for previously orphan genome-wide significant loci. These multi-level findings furnish a resource for pathobiology of CAC, atherosclerosis, and establish an adaptable framework applicable to all organ systems to parse precision targets for prevention, surveillance, and therapy of cardiovascular disease.

Background Genome‐wide association studies have identified several hundred susceptibility single nucleotide variants for coronary artery disease (CAD). Despite single nucleotide variant‐based genome‐wide association studies improving our understanding of the genetics of CAD, the contribution of structural variants (SVs) to the risk of CAD remains largely unclear. Method and Results We leveraged SVs detected from high‐coverage whole genome sequencing data in a diverse group of participants from the National Heart Lung and Blood Institute's Trans‐Omics for Precision Medicine program. Single variant tests were performed on 58 706 SVs in a study sample of 11 556 CAD cases and 42 907 controls. Additionally, aggregate tests using sliding windows were performed to examine rare SVs. One genome‐wide significant association was identified for a common biallelic intergenic duplication on chromosome 6q21 ( P =1.54E‐09, odds ratio=1.34). The sliding window‐based aggregate tests found 1 region on chromosome 17q25.3, overlapping USP36 , to be significantly associated with coronary artery disease ( P =1.03E‐10). USP36 is highly expressed in arterial and adipose tissues while broadly affecting several cardiometabolic traits. Conclusions Our results suggest that SVs, both common and rare, may influence the risk of coronary artery disease.

Background and Aims Individuals with steatotic liver disease (SLD) are at high cardiovascular disease (CVD) risk, but approaches to characterise and mitigate this risk are limited. By investigating relations, and shared metabolic pathways, of hepatic steatosis/fibrosis and cardiorespiratory fitness (CRF), we sought to identify new avenues for CVD risk reduction in SLD. Methods In Framingham Heart Study (FHS) participants ( N = 2722, age 54 ± 9 years, 53% women), vibration‐controlled transient elastography (VCTE) was performed between 2016–2019 to assess hepatic steatosis (continuous attenuation parameter [CAP]) and fibrosis (liver fibrosis measure [LSM]). Concurrently, participants underwent maximum effort cardiopulmonary exercise testing (CPET), and metabolomic profiling (201 circulating metabolites) was performed in a subsample ( N = 1268). Results Mean BMI was 28.0 ± 5.3, 27% had hepatic steatosis, 7.6% had fibrosis, and peak oxygen uptake (VO 2 ) was 26.2 ± 6.8 mL/kg/min in men and 20.7 ± 6.0 mL/kg/min in women (95% predicted overall). In linear models adjusted for cardiometabolic risk factors, greater CAP and LSM were associated with lower peak VO 2 ( p ≤ 0.002 for all), and the CAP association remained significant after BMI adjustment ( p < 0.0001). We observed shared metabolic architecture of CAP, LSM, and peak VO 2 , with metabolites mediating up to 35% (for CAP) and 74% (for LSM) of the association with peak VO 2 . Metabolite mediators included amino acids and derivatives implicated in cardiometabolic risk and both protective and deleterious lipid species. Conclusions Hepatic steatosis and fibrosis are associated with CRF impairment in the community, and these relations are partly mediated by pathways of altered lipid metabolism and general cardiometabolic risk.

Hepatic steatosis is a central phenotype in multi-system metabolic dysfunction and is increasing in parallel with the obesity pandemic. We use a translational approach integrating clinical phenotyping and outcomes, circulating proteomics, and tissue transcriptomics to identify dynamic, functional biomarkers of hepatic steatosis. Using multi-modality imaging and broad proteomic profiling, we identify proteins implicated in the progression of hepatic steatosis that are largely encoded by genes enriched at the transcriptional level in the human liver. These transcripts are differentially expressed across areas of steatosis in spatial transcriptomics, and several are dynamic during stages of steatosis. Circulating multi-protein signatures of steatosis strongly associate with fatty liver disease and multi-system metabolic outcomes. Using a humanized “liver-on-a-chip” model, we induce hepatic steatosis, confirming cell-specific expression of prioritized targets. These results underscore the utility of this approach to identify a prognostic, functional, dynamic “liquid biopsy” of human liver, relevant to biomarker discovery and mechanistic research applications.

Human studies linking metabolism with organism-wide physiologic function have been challenged by confounding, adherence, and precisionHere, we united physiologic and molecular phenotypes of metabolism during controlled dietary intervention to understand integrated metabolic-physiologic responses to nutrition. In an inpatient study of individuals who underwent serial 24-hour metabolic chamber experiments (indirect calorimetry) and metabolite profiling, we mapped a human metabolome onto substrate oxidation rates and energy expenditure across up to 7 dietary conditions (energy balance, fasting, multiple 200% caloric excess overfeeding of varying fat, protein, and carbohydrate composition). Diets exhibiting greater fat oxidation (e.g., fasting, high-fat) were associated with changes in metabolites within pathways of mitochondrial β-oxidation, ketogenesis, adipose tissue fatty acid liberation, and/or multiple anapleurotic substrates for tricarboxylic acid cycle flux, with inverse associations for diets with greater carbohydrate availability. Changes in each of these metabolite classes were strongly related to 24-hour respiratory quotient (RQ) and substrate oxidation rates (e.g., acylcarnitines related to lower 24-hour RQ and higher 24-hour lipid oxidation), underscoring links between substrate availability, physiology, and metabolism in humans. Physiologic responses to diet determined by gold-standard human metabolic chambers are strongly coordinated with biologically consistent, interconnected metabolic pathways encoded in the metabolome.

Background: Maintaining cardiorespiratory fitness (CRF) is essential for healthy aging and is linked to lower risk of age-related cardiovascular diseases and frailty. However, the relations of body composition metrics and CRF levels across the adult age range are poorly understood. Hypothesis: Healthy body composition, characterized by higher lean mass and lower visceral and total adiposity, is associated with greater oxygen (O 2 ) uptake during exercise across age and sex. Aims: To investigate relations of components of body composition with O 2 uptake measures during exercise and to evaluate how age and sex affect the associations. Methods: Framingham Heart Study participants (n=2742) underwent dual-energy X-ray absorptiometry (DXA) for body composition assessment and maximal cardiopulmonary exercise testing on a cycle ergometer at the same exam cycle. Sex-stratified linear models between DXA and O 2 uptake measures were adjusted for age, height, HDL/total cholesterol, smoking, systolic blood pressure, hypertension treatment, diabetes, and physical activity index. Results: The study sample had a mean age of 54±9 years, was 54% women, 9% non-white, and had a peak O 2 uptake [VO 2 ] 95±20% of predicted. Higher lean mass was associated with higher peak VO 2 in ml/min in both men and women (β for standard deviation (SD) difference in log-transformed peak VO 2 for a 1-SD higher value of lean mass was 0.49 in men and 0.57 in women, P<0.0001 for both). Conversely, measures of total and visceral fat mass were associated with higher peak VO 2 in women (β= 0.14-0.15, P<0.0001) but not in men (P>0.05). The variance explained (adjusted R ² ) in peak VO 2 was higher for lean mass than other DXA measures: lean mass, age, and height explained 45-47% of the variance in peak VO 2 in men and women, while age, height, and other DXA measures explained ≤37%. Furthermore, individuals with higher observed lean mass than predicted based on age, sex, and BMI had higher observed peak VO 2 , consistent with the finding that higher lean mass is associated with higher peak VO 2 across age and sex (Figure). Conclusion: Maintaining lean mass across the adult age range is linked with higher peak VO 2 , which may have important implications for promoting healthy aging.

Objective To generate data-driven phenogroups of cardiac structure and function based on echocardiographic measures assessed in asymptomatic middle-aged adults free of CVD, and examine associations between these newly defined phenogroups and incident premature cardiovascular disease (CVD). Methods Data were analyzed from participants in the Coronary Artery Risk Development in Young Adults (CARDIA) cohort study free of CVD who underwent an echocardiogram at the Year 25 (2010-2011) in-person examination. Continuous echocardiographic measures of left heart structure, left ventricular systolic function (including strain) and diastolic function, right ventricular systolic function, and hemodynamic measures were included in latent class analysis to generate novel phenogroups. Associations between data-driven phenogroups and risk of premature CVD (coronary artery disease, stroke, or heart failure) were estimated using Cox proportional hazards regression adjusted for traditional CVD risk factors. Results Among 3361 participants, mean (standard deviation) age was 50.1 (3.6) years, 57% were female, and 46% were non-Hispanic Black. Three overall phenogroups were identified and labeled as: (1) optimal cardiac mechanics (36.2%); (2) suboptimal systolic function (38.2%); and (3) suboptimal diastolic function (25.6%). Over a median 8.9 years of follow-up, 121 premature CVD events occurred. Risk of CVD was higher in the suboptimal diastolic function group (unadjusted hazard ratio [HR] 4.08 [95% CI: 2.48, 6.71] and adjusted HR 1.95 [1.12, 3.40]) compared with the optimal group. The suboptimal systolic function group had a higher unadjusted risk of CVD (1.86 [1.10, 3.15]), which was attenuated after adjustment for CVD risk factors (1.36 [0.79, 2.36]). Conclusions and relevance Unbiased, data-driven clustering of echocardiographic measures in middle-aged adults identified distinct patterns of cardiac remodeling that were associated with risk of premature CVD. Premature CVD risk was highest with the pattern of suboptimal diastolic function. This suggests potential utility of a composite echocardiography-based index for prioritizing prevention strategies earlier in the life course.

Objective Our objective was to prospectively investigate prediagnostic population‐based metabolome for risk of hospitalized gout (ie, most accurate, severe, and costly cases), accounting for serum urate. Methods We conducted prediagnostic metabolome‐wide analyses among 249,677 UK Biobank participants with nuclear magnetic resonance metabolomic profiling (N = 168 metabolites, including eight amino acids) from baseline blood samples (2006–2010) without a history of gout. We calculated multivariable hazard ratios (HRs) for hospitalized incident gout, before and after adjusting for serum urate levels; we included patients with nonhospitalized incident gout in a sensitivity analysis. Potential causal effects were evaluated with two‐sample Mendelian randomization. Results Correcting for multiple testing, 107 metabolites were associated with incidence of hospitalized gout (n = 2,735) before urate adjustment, including glycine and glutamine (glutamine HR 0.64, 95% confidence interval [CI] 0.54–0.75, P = 8.3 × 10⁻⁸; glycine HR 0.69, 95% CI 0.61–0.78, P = 3.3 × 10⁻⁹ between extreme quintiles), and glycoprotein acetyls (HR 2.48, 95% CI 2.15–2.87, P = 1.96 × 10⁻³⁴). Associations remained significant and directionally consistent following urate adjustment (HR 0.83, 95% CI 0.70–0.98; HR 0.86, 95% CI 0.76–0.98; HR 1.41, 95% CI 1.21–1.63 between extreme quintiles), respectively; corresponding HRs per SD were 0.91 (95% CI 0.86–0.97), 0.94 (95% CI 0.91–0.98), and 1.10 (95% CI 1.06–1.14). Findings persisted when including patients with nonhospitalized incident gout. Mendelian randomization corroborated their potential causal role on hyperuricemia or gout risk; with change in urate levels of −0.05 mg/dL (95% CI −0.08 to −0.01) and −0.12 mg/dL (95% CI −0.22 to −0.03) per SD of glycine and glutamine, respectively, and odds ratios of 0.94 (95% CI 0.88–1.00) and 0.81 (95% CI 0.67–0.97) for gout. Conclusion These prospective findings with causal implications could lead to biomarker‐based risk prediction and potential supplementation‐based interventions with glycine or glutamine.