Title: PLASMA PROTEOMIC ASSOCIATIONS OF ARTERIAL STIFFNESS: THE MULTI-ETHNIC STUDY OF ATHEROSCLEROSIS

Authors: BIANCA POURMUSSA1, BA; Hamed Tavolinejad1,2, MD; Marie-Joe Dib, PhD1,2; Cameron Beeche, BS1,3; Karim Kohansal Vajargah, MD11; Joe-David Azzo, MD1,2; Jerome I. Rotter, MD4; Stephen S. Rich, PhD7; Xiuqing Guo, PhD4; Sanjiv J. Shah, MD8; Susan R. Heckbert, MD, PhD5; Alain G. Bertoni, MD, MPH6; Joao A.C. Lima, MD9; Usman A. Tahir, MD10; Julio A. Chirinos, MD, PhD1,2

Affiliations:
1 Division of Cardiovascular Medicine, Hospital of the University of Pennsylvania, Philadelphia, PA
2 University of Pennsylvania Perelman School of Medicine, Philadelphia, PA
3 Department of Bioengineering, University of Pennsylvania, Philadelphia, PA
4 Lundquist Institute at Harbor UCLA, Torrance, CA
5 Department of Epidemiology, University of Washington, Seattle, WA
6 Division of Public Health Sciences, Wake Forest University School of Medicine, Winstom-Salem, NC
7 Department of Genome Sciences, University of Virginia School of Medicine, Charlottesville, VA
8 Northwestern University Feinberg School of Medicine, Chicago, IL
9 Johns Hopkins University School of Medicine, Baltimore, MD
10 Division of Cardiovascular Medicine, Beth Israel Deaconess Medical Center, Boston, MA, USA.
11 Prevention of Metabolic Disorders Research Center, Research Institute for Metabolic and Obesity Disorders, Research Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran

Introduction: Arterial Stiffness (ARTS), results from advanced age, disease, and genetic factors. ARTS leads to detrimental hemodynamics and end organ damage. Quantification of ARTS therefore informs our understanding about susceptibility to disease. Additionally, identifying the biological pathways that contribute to ARTS may reveal targets for intervention. We aimed to assess the proteomic associations of ARTS with different ARTS phenotypic measures: pulse pressure (PP), total arterial compliance (TAC), cardio-ankle vascular index (CAVI), and cardio-femoral vascular index (CAFVI).
Methods/Design: The proteomic correlates of ARTS metrics were assessed in the Multi-Ethnic Study of Atherosclerosis (MESA) using the Olink-3k panel, which measures 2,923 proteins. We used linear regression models adjusted for cardiovascular risk and demographics factors to assess the association of each protein with ARTS metrics. The false discovery rate method was applied for multiplicity correction. Overrepresentation analysis using the Reactome database was performed in R studio using the ReactomePA package.
Results: The study population included 3,135 participants (mean age 74 years, 53% females). In models adjusted for the Framingham Risk Score, prevalent cardiovascular disease, and race, we identified 807 proteins significantly associated with PP, 693 significantly associated with TAC, 346 significantly associated with CAVI, and 1,101 significantly associated with CAFVI. Significant associations of all four outcomes included RSPO3, TNFRSF11B, and MMP12. Several significantly enriched pathways were identified in this model, including extracellular matrix organization (CAVI), membrane trafficking (CAFVI), and post-translational protein phosphorylation (PP and TAC). In analyses adjusted for age, sex, and race, we identified 5 proteins associated with TAC, 410 proteins associated with PP, with no significant associations for either CAFVI or CAVI.
Conclusions: These findings reveal multiple proteins and biologic pathways associated with ARTS, improving upon current understanding of potential underlying molecular mechanisms. This analysis also demonstrates important differences in proteomic associations of various metrics related to ARTS.

Circulating metabolite levels reflecting the state of human health and disease can be impacted by genetic effects. The NHLBI Trans-Omics for Precision Medicine (TOPMed) Program has sponsored metabolomic measures in ~100,000 samples across multiple studies to promote discovery of causal molecular pathways and therapeutic targets. We initiated a standard operating procedure (SOP) to harmonize metabolite data across TOPMed studies. In Phase 1, we catalogued 1,730 circulating metabolites from two metabolomics cores (25,058 samples; 53% females) and made them accessible through TOPMed portal.
Metabolite levels are heritable. However, their genetic architectures are not fully understood, including the generalizability of findings from European ancestry dominant studies, and the identification of sex-specific metabolic signatures. Whole genome sequencing (WGS) data were available in 16,359 samples (54% females) who had metabolite data from eight studies, including African, Asian, European, and American ancestries. We performed single variant analyses (minor allele frequency &gt 0.5%) on 1,135 circulating metabolites (missing rate &lt 50%), using sex-pooled and sex-stratified approaches by GMMAT pipeline on BioData Catalyst (BDC).
We discovered 147,160 variant-metabolite pairs of associations (1,429 independent loci across 667 metabolites with P &lt 4.4x10-11). Among the associations mapped to well-known genes, four significant loci (CPS1, ALDH1L1, PSPH, GCSH) play critical roles on glycine metabolism. We also identified potential novel loci that require further investigation, e.g., SLC22A24 was associated with 11-beta-hydroxyetiocholanolone glucuronide levels. We observed sex-specific genetic associations in ~10% metabolites. Sex-stratified analysis identified 2,414 variant-metabolite pairs involving 194 independent loci and 74 metabolites (at a Bonferroni-corrected P = 2.5x10-4). We confirmed that CPS1 has a stronger effect on glycine levels in females than males. We also identified potential novel sex-specific loci with genetic effects in only one sex group, e.g., GSPT1 for N-acetylglycine in females, ABCC1 for Glutarylcarnitine (C5-DC) in males. We also detected potential novel association on chromosome X, i.e., ARSD for ascorbic acid 3-sulfate in females. The analytical pipeline is accessible through BDC, while sex-pooled and sex-stratified summary statistics are accessible through dbGaP Exchange Area.
In summary, we created a catalog for TOPMed metabolomics data and identified potential novel sex-pooled and sex-specific genetic associations contributing to our understanding of human circulating metabolites.

The Trans-Omics for Precision Medicine (TOPMed) program expects to soon release over 90,000 samples with broad-spectrum metabolomic data, representing over a dozen studies. However, investigators using this resource face potential challenges in pre-processing and integrating data across studies. Differing metabolomic platforms and analysis centers may cause technical variation. Likewise, missing metabolite values may vary in their distribution and source between studies. Consistent protocols for pre-processing and integration are thus necessary to unlock the potential of this rich resource. We compare several strategies and offer recommendations for the TOPMed community, with the goal of guiding and facilitating future genetic and phenotype-specific analyses.
As a pilot phase, we are currently analyzing data from 25,058 participants from diverse case-control and population-based cohort studies, including 15,633 participants from 3 cohort studies on the Metabolon platform and 9,425 participants from 5 cohort studies on the Broad/BIDMC platform. This dataset includes 1730 named metabolites, including 364 metabolites measured in at least some cohorts across both platforms. With within-study rank-based inverse normal transformation, we demonstrate that estimates of age-metabolite associations are highly concordant (r > 0.999), and generally consistent with the existing literature, between pooled and inverse variance meta-analyzed data, although 36 metabolites are significant only in the meta-analysis. Most named metabolites had very low missingness in our dataset, and we found that metabolite associations with age and sex were highly consistent across all missingness imputation strategies (zero, min, half-min, k-nearest neighbors, random forest, quantile regression imputation of left censored data). We recommend replacing missing values with zero in metabolites characterized as xenobiotics. For other metabolites, we will compare imputation strategies with an analysis of metabolite quantitative trait loci (mQTLs).
In summary, we find largely consistent results in pooled and inverse variance meta-analysis. We recommend inverse-normal transformation to enable integration between studies. We recommend left-censored imputation for xenobiotics and will soon release recommendations for imputation in other metabolites. To aid investigators, we will release scripts for implementing these recommendations. Such pre-processing steps are necessary to optimize power in cross cohort metabolomic analysis, including planned QTL studies.

Circulating metabolite levels reflecting the state of human health and disease can be impacted by genetic effects. The NHLBI Trans-Omics for Precision Medicine (TOPMed) Program has sponsored metabolomic measures in ~100,000 samples across multiple studies to promote discovery of causal molecular pathways and therapeutic targets. We initiated a standard operating procedure (SOP) to harmonize metabolite data across TOPMed studies. In Phase 1, we catalogued 1,730 circulating metabolites from two metabolomics cores (25,058 samples; 53% females) and made them accessible through TOPMed portal.
Metabolite levels are heritable. However, their genetic architectures are not fully understood, including the generalizability of findings from European ancestry dominant studies, and the identification of sex-specific metabolic signatures. Whole genome sequencing (WGS) data were available in 16,359 samples (54% females) who had metabolite data from eight studies, including African, Asian, European, and American ancestries. We performed single variant analyses (minor allele frequency ≥ 0.5%) on 1,135 circulating metabolites (missing rate <50%), using sex-pooled and sex-stratified approaches by GMMAT pipeline on BioData Catalyst (BDC).
We discovered 147,160 variant-metabolite pairs of associations (1,429 independent loci across 667 metabolites with P <4.4x10-11). Among the associations mapped to well-known genes, four significant loci (CPS1, ALDH1L1, PSPH, GCSH) play critical roles on glycine metabolism. We also identified potential novel loci that require further investigation, e.g., SLC22A24 was associated with 11-beta-hydroxyetiocholanolone glucuronide levels. We observed sex-specific genetic associations in ~10% metabolites. Sex-stratified analysis identified 2,414 variant-metabolite pairs involving 194 independent loci and 74 metabolites (at a Bonferroni-corrected P = 2.5x10-4). We confirmed that CPS1 has a stronger effect on glycine levels in females than males. We also identified potential novel sex-specific loci with genetic effects in only one sex group, e.g., GSPT1 for N-acetylglycine in females, ABCC1 for Glutarylcarnitine (C5-DC) in males. We also detected potential novel association on chromosome X, i.e., ARSD for ascorbic acid 3-sulfate in females. The analytical pipeline is accessible through BDC, while sex-pooled and sex-stratified summary statistics are accessible through dbGaP Exchange Area.
In summary, we created a catalog for TOPMed metabolomics data and identified potential novel sex-pooled and sex-specific genetic associations contributing to our understanding of human circulating metabolites.