Scientists uncover key proteins driving brain aging, offering critical insights into timing interventions to preserve brain health and combat neurodegenerative disorders.
Study: Plasma proteomics identify biomarkers and undulating changes of brain aging. Image Credit: MattL_Images / ShutterstockStudy: Plasma proteomics identify biomarkers and undulating changes of brain aging. Image Credit: MattL_Images / Shutterstock
In a recent study published in the journal Nature Aging, researchers identified plasma proteomic biomarkers and dynamic changes associated with brain aging, leveraging a multimodal approach combining brain age gap (BAG) and proteome-wide association analysis.
Background
The global aging population is expected to exceed 1.5 billion individuals aged 65 and above by 2050, highlighting the urgent need to address aging-associated challenges.
Aging disrupts brain homeostasis and resilience, leading to functional decline, structural abnormalities, and increased risk of neurodegenerative disorders, such as Alzheimer’s disease (AD) (a neurodegenerative disorder causing memory loss and cognitive decline) and Parkinson’s disease (PD) (a movement disorder characterized by tremors, stiffness, and slow movement).
Despite their prevalence, effective therapies for these conditions remain limited, emphasizing the importance of early identification and intervention. Biomarkers, including imaging traits and histological features, offer insights into brain aging but lack molecular depth.
Further research exploring causal relationships through advanced methods like Mendelian randomization is essential to identifying dynamic plasma proteomic changes, uncovering mechanisms, and guiding interventions.
About the Study
Nonlinear Proteomic Patterns: The study uncovered undulating plasma proteomic changes across brain aging, with distinct peaks at ages 57, 70, and 78, highlighting dynamic shifts in biological pathways, including metabolism and immune responses.
The study adhered to the ethical principles outlined in the Declaration of Helsinki, with participants providing written informed consent. Approval was granted by the North West Multi-Center Research Ethics Committee.
The study utilized data from the United Kingdom (UK) Biobank, a large-scale cohort of over 500,000 participants across 22 assessment centers in the UK. Participants' demographic, biological, and physical characteristics were recorded at recruitment, and all were registered with the UK National Health Service.
Brain imaging data were collected using Siemens Skyra 3T scanners approximately four years after recruitment. The imaging dataset included over 40,000 participants, and after quality control and screening, 1,705 unique imaging-derived phenotypes (IDPs) reflecting the brain's structural, functional, and diffusion characteristics were selected. These IDPs were incorporated into a multimodal brain age model, developed using Least Absolute Shrinkage and Selection Operator (LASSO) regression, a machine learning approach, to predict the BAG, an indicator of brain health.
Plasma proteomics data from over 2,900 proteins were analyzed using samples collected at recruitment. Rigorous quality control, normalization, and statistical adjustments identified proteins associated with BAG. Functional enrichment analyses highlighted biological processes relevant to brain health. Repeat imaging visits validated six key proteins, strengthening the reliability of findings. Mendelian randomization further established causal links between key proteins and brain aging.
Study Results
After rigorous screening, the multimodal brain age model was developed using data from 1,705 brain IDPs. The participants, totaling 10,949 after excluding those with neuropsychiatric or other disorders, were divided into training and testing sets.
Using LASSO regression, 864 IDPs were identified as contributors to brain age prediction. The model demonstrated accurate performance, with a mean absolute error (MAE) of 2.76 years, with better performance observed among females.
Broader Aging Connections: While proteins like GDF15 are significant for brain aging, their roles extend to general aging processes and age-related diseases, making them potential targets for broader aging research.
Proteome-wide association analysis identified 13 plasma proteins significantly associated with BAG, a measure of brain aging. These proteins exhibited diverse roles, with some positively associated with BAG, such as Growth Differentiation Factor 15 (GDF15) and Glial Fibrillary Acidic Protein (GFAP), indicating stress-related processes, while others like Brevican (BCAN) and Kallikrein-6 (KLK6) negatively correlated, suggesting their role in cellular regeneration and adhesion. BCAN emerged as a promising biomarker, supported by Mendelian randomization.
Validation with repeat imaging confirmed six proteins as consistently linked to BAG, emphasizing their potential as biomarkers.
Biological profiling revealed distinct functions of BAG-associated proteins. Positively associated proteins were enriched in pathways like tyrosine kinase signaling, linked to cellular stress, while negatively associated proteins related to neuron projection regeneration and synaptic functions, highlighting key aging mechanisms.
Expression analyses showed differential localization of these proteins in brain cells, with BCAN prominent in astrocytes and oligodendrocyte progenitor cells. Notably, BCAN expression was reduced in neurodegenerative disorders like AD, further reinforcing its significance. These patterns were further validated in datasets from neurodegenerative disorders, underscoring the relevance of BAG proteins, particularly BCAN, in brain aging and conditions like AD.
The BAG proteins also demonstrated significant associations with brain structures and functions. BCAN and KLK6 showed notable correlations with cortical volumes and subcortical structures, especially in regions commonly affected by aging and neurodegenerative diseases. Functional analyses linked proteins like GDF15 to movement and mental health traits, reinforcing their broader implications for brain health.
Undulating patterns in plasma proteins during brain aging revealed significant changes at ages 57, 70, and 78, offering critical time points for potential interventions. These waves correspond to shifts in biological pathways, such as metabolic processes at earlier ages and immune-related pathways at later stages. Proteins at these waves were associated with brain health traits, including cognition, mental health, and movement, providing a dynamic view of brain aging.
Conclusions
To summarize, thirteen plasma proteins were identified as associated with brain aging, and GDF15, BCAN, GFAP, and KLK6 were validated for their biological functions and implications for brain health.
Causal associations between these proteins and the BAG were tested, suggesting BCAN as a candidate biomarker for brain aging. Undulating proteomic changes were observed during brain aging, with significant peaks in the late fifth, seventh, and late seventh decades, indicating crucial periods for potential interventions. The study highlights the importance of personalized approaches to address brain aging and associated disorders.
Journal reference:
Liu, W., You, J., Chen, S., Zhang, Y., Feng, J., Xu, Y., Yu, J., & Cheng, W. (2024). Plasma proteomics identify biomarkers and undulating changes of brain aging. Nature Aging, 1-14. DOI: 10.1038/s43587-024-00753-6, https://www.nature.com/articles/s43587-024-00753-6