Large-Scale Identification of Novel Protein Biomarkers and Therapeutic Targets in Heart and Brain Disease
Abstract Body: Introduction: Cardiovascular (CVD) and neurological diseases (NeuD) are growing public health challenges that often coexist, suggesting shared biological mechanisms. Hypothesis: Proteins associated with heart and brain MRI imaging-derived phenotypes (IDPs) are enriched for markers and mediators of disease in one or both organs. Methods: Utilizing UK Biobank data, plasma proteomics from 53,014 individuals were correlated with cardiac and brain MRI IDPs from 50,228 individuals. Expression and pathway analyses provided insights into likely sources of correlated proteins and the biological processes driving the associations. Protein associations with future (incident) neurological and/or cardiovascular disease were assessed. Mendelian randomization and genetic colocalization were applied to infer causality. Results: 404 plasma proteins associated with cardiac IDPs, 76 with brain IDPs, and 37 with IDPs in both organs. Identified proteins strongly correlated with incident CVD and NeuD (94% and 62% for cardiac-; 95% and 57% for brain-IDP associated proteins, respectively), suggesting they could be useful predictive biomarkers. The large majority of IDP-associated proteins correlated with incident NeuD also correlated with incident CVD emphasizing the common molecular underpinnings of disease in these two organs. Expression analysis implicated fibroblasts, vascular smooth muscle cells, and macrophages within the arterial wall as the likely source of many of these proteins. Pathway analysis highlighted cytokine signaling for cardiac-associated proteins and extracellular matrix pathways for brain-associated proteins. Genetic analyses supported causal roles for most (>63%) of the proteins in disease pathogenesis in one or both organs. Although some known (e.g. NT-proBNP; Renin) candidates were confirmed by these studies, >90% of the potential biomarkers and therapeutic targets identified are novel. Conclusions: These studies underscore the value of large-scale integrated multi-organ datasets including imaging-derived endophenotypes and plasma proteomics in unraveling complex disease interactions, highlight the close connections between heart and brain disease, and provide a catalogue of hundreds of novel candidate biomarkers and therapeutic targets in CVD and NeuD.
Wu, Chao
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Wang, Meng
( University of Michigan
, ANN ARBOR
, Michigan
, United States
)
Liang, Huiying
( Southern Medical University
, Guangzhou
, China
)
Rosenzweig, Anthony
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Li, Dantong
( Southern Medical University
, Guangzhou
, China
)
Yuan, Zixun
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Huang, Shengyuan
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Guerra, Justin
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Li, Chunyu
( University of Michigan
, ANN ARBOR
, Michigan
, United States
)
Zhou, Qiulian
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Quan, Meixi
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
He, Jiaqi
( University of Michigan Medical Center
, Ann Arbor
, Michigan
, United States
)
Author Disclosures:
Chao Wu:DO NOT have relevant financial relationships
| Meng Wang:No Answer
| Huiying Liang:No Answer
| Anthony Rosenzweig:No Answer
| Dantong Li:No Answer
| Zixun Yuan:DO NOT have relevant financial relationships
| Shengyuan Huang:No Answer
| Justin Guerra:No Answer
| Chunyu Li:DO NOT have relevant financial relationships
| Qiulian Zhou:DO NOT have relevant financial relationships
| Meixi Quan:DO NOT have relevant financial relationships
| Jiaqi He:No Answer