Aging is a complex and inevitable biological process involving changes at multiple levels, including molecular, cellular, tissue, and organ levels. It is often accompanied by the degeneration of bodily functions and an increased susceptibility to diseases. The loss of proteostasis in the body is a significant hallmark of aging; therefore, the discovery of novel protein biomarkers related to aging has important strategic significance.

Although cross-sectional studies have revealed a correlation between blood proteins and age-related changes, there have been no effective blood protein biomarkers to predict aging and healthy aging status early. It is against this backdrop that the research team at the School of Medicine, Westlake University, conducted an in-depth exploration.

This study was jointly completed by the Zheng Jusheng team and Guo Tiannan team from the Westlake Institute for Advanced Study/School of Medicine, Westlake University/Future Industries Research Center, and the Chen Yuming team from the School of Public Health, Sun Yat-sen University. The study explored and discovered aging biomarkers based on a cohort of elderly Chinese Han people (Guangzhou Nutrition and Health Study, GNHS).

During the 9-year follow-up of the GNHS cohort, the research team collected a total of 7565 serum samples from 3796 elderly participants. The study further divided them into two sub-cohorts: the GNHS discovery cohort (including 1939 participants and 4637 serum samples) and the GNHS validation cohort (including 1857 participants and 2928 serum samples).

In addition, this study included an external validation cohort of 124 participants. The research team used high-throughput mass spectrometry to determine the serum proteome of the participants and, through in-depth mining and analysis, discovered 86 protein biomarkers associated with aging. These proteins showed characteristics related to 32 clinical characteristics and the incidence of 14 major chronic diseases associated with aging.

Researchers further used machine learning algorithms to identify the 22 most critical blood proteins and constructed a "Proteomic healthy aging score" (PHAS). This score can predict and warn early about the status of healthy aging in the elderly population and the risk of related cardiovascular and metabolic diseases.

Specifically, the researchers used a decision tree aggregation algorithm and a random forest model to generate the probability of each participant being judged as healthy, which serves as the PHAS. This model showed high accuracy in the GNHS training set, GNHS validation set, and external validation cohort. The study found that higher PHAS values were longitudinally associated with improved anthropometric parameters, lipid and glucose metabolism biomarkers, and improved liver and kidney biomarkers.

Researchers also investigated the potential association between baseline levels of 86 aging-related proteins and the incidence of 14 chronic diseases. The results showed that 67 aging-related proteins had 131 nominally significant associations, with more than ten proteins related to dyslipidemia, hypertension, type 2 diabetes (T2D), fatty liver, and hepatitis.

Among the 131 observed protein-disease associations, 35 remained significant after multiple testing correction. In addition, eight proteins, including α-1-antitrypsin (A1AT), leucine-rich α-2-glycoprotein (A2GL), A2MG, adiponectin (ADIPO), zinc finger protein Gfi-1 (GFI1), ITIH3, RAIN, and vitronectin (VTNC), were associated with two or more aging-related metabolic diseases.

This longitudinal study expands our understanding of the serum proteomic profile in the context of aging and its impact on human health. The study not only identified serum proteomic biomarkers associated with aging but also provided valuable insights into the potential mechanisms of human aging from a proteomic perspective.

These discovered proteomic biomarkers are expected to become valuable tools for monitoring and predicting age-related cardiometabolic diseases. In the future, with further research, these biomarkers may be applied clinically to provide new strategies and methods for the prevention and intervention of aging and related diseases.

References:

Tang, J., Yue, L., Xu, Y. et al. Longitudinal serum proteome mapping reveals biomarkers for healthy ageing and related cardiometabolic diseases. Nat Metab 7, 166–181 (2025). DOI: 10.1038/s42255-024-01185-7.