Proteomics and Metabolomics for Prevention, Diagnosis, and Mechanistic Insights in Cardiovascular Disease

Cardiovascular disease is multifactorial, which means it is influenced by genetics, behaviors, socioeconomics, and environment. Preventive cardiology by addressing modifiable risk factors has advanced significantly in recent years. Despite the current risk prediction algorithms and intervention strategies, cardiovascular disease remains the leading cause of mortality. Therefore, the use of additional biomarkers is being advocated to improve risk prediction accuracy. Biomarkers based on the proteins in an organism (proteomic biomarkers) or derived from metabolism (metabolomic biomarkers) are useful tools for more accurate phenotyping, reflecting molecular characteristics and enabling personalized interventions. The use of proteomics and metabolomics can thus improve the comprehension of underlying biological mechanisms beyond apparent clinical abnormalities. This PhD thesis aimed to a) investigate proteomic signatures for vascular health and their prognostic value for cardiovascular outcomes in the general population; b) use urinary proteomic biomarkers to screen for long-term complications after heart transplantation and understand the potential disease mechanisms; c) apply metabolomics to study lipid metabolism in upper body obesity and the metabolic changes in individuals seemingly healthy.

Blood vessel health (vascular health) gets worse with aging. In this PhD thesis (Chapter I), we studied the urinary proteomic signature for three vascular changes: arterial stiffness, vascular calcification, and coronary atherosclerosis. Arterial stiffness is a hallmark of vascular aging and a known risk factor for cardiovascular disease. Pulse wave velocity is a standard measurement for arterial stiffness. We identified a urinary proteomic profile highly correlated with pulse wave velocity and this may be an alternative approach to quantify arterial stiffness (Chapter I, Part I). The urinary proteomic signature included diverse proteins involved in collagen turnover, cell adhesion, inflammation, and lipid metabolism. The urinary proteomic also had predictive value for cardiovascular disease and mortality.

Vascular calcification is a basic pathological change associated with arterial stiffness and atherosclerosis. Matrix Gla protein is an important inhibitor of vascular calcification. Chapter I, Part II found that increased urinary matrix Gla protein correlates with increased risk of overall and cardiovascular mortality. The risk stratification was more accurate when urinary matrix Gla protein was added to the risk calculation. Factors affecting urinary matrix Gla protein excretion were circulating matrix Gla protein (desphospho-uncarboxylated, dp-ucMGP), sex, age, urine microalbumin, smoking, and total cholesterol. 

Along the same line, the urinary proteomic signature associated with coronary artery disease was studied (Chapter I, Part III). We developed a new proteomic classifier consisting of 160 urinary peptides and validated its predictive value in an independent cohort of 893 participants. This classifier was able to improve the risk classification on top of the known cardiovascular disease prediction formulas (the Framingham risk score and SCORE2). The new proteomic classifier maintained its predictive power for coronary artery disease after accounting for clinical risk factors. Particularly, peptides derived from collagen degradation, inflammatory responses, and lipid metabolism were found to be abundant in our urinary proteomic classifier.

We also used urinary proteome analysis to detect two common complications after heart transplantation that can impair the function of the transplanted heart and increases the risk of death (Chapter II). Vasculopathy of the transplanted heart is a progressive thickening of the donor's coronary arteries. Its incidence increases with time after heart transplantation, and screening is completed by invasive coronary artery examination. In Chapter II, Part I, we identified a urinary proteomic signature for the detection of transplanted heart vasculopathy and validated it in an internal cohort. The peptides involved in fibrosis, platelet aggregation, and coagulation might play a role in the development of transplanted heart vasculopathy. 

The use of anti-rejection medications increases the risk of cancer after heart transplantation. Screening cancer in patients who underwent heart transplantation involves multidisciplinary efforts. Thus, we sought urinary proteomic markers specific to cancer after heart transplantation in Chapter II, Part II. Our findings showed that urinary proteomic analysis can detect cancer patients and distinguish solid organ cancer from skin cancer. The urinary proteomic signature of solid organ cancer was related to collagen degradation, cell adhesion and apoptosis, and tumor formation. Moreover, it was associated with an increased risk of death.

In Chapter III, we applied metabolomics to characterize the metabolic changes in obesity. We showed that VLDL particle number is the lipoprotein component most strongly associated with waist-to-hip ratio and that LDL might underestimate the lipid abnormality in people with obesity (Chapter III, Part I). The association between waist-to-hip ratio and VLDL particles was significantly determined by insulin sensitivity. This study highlighted the importance of waist-to-hip ratio and the treatment targeting VLDL particles and insulin sensitivity in abdominal obesity-related dyslipidemia.

People with obesity have diverse cardiometabolic risks. In Chapter III, Part II, we conducted a large prospective population-based metabolomic study. We found that there were unfavorable metabolic changes in individuals with metabolically healthy obesity, although their cardiovascular risk might remain low for a short period of time. Our findings supported that early intervention should be recommended for any form of obesity to prevent potential long-term cardiovascular risk.

In summary, our findings clearly suggested that urinary proteomic analysis and circulating metabolomic profiling allow molecular phenotyping of cardiovascular disease and additional refinement of risk calculations based on clinical characteristics. These omics signatures can be used to improve the risk stratification of cardiovascular disease, develop new diagnostic tools for detecting complications after heart transplantation, and provide insights into pathological changes that thereby enable the development of personalized treatment strategies.