Association of Serum Proteomic Biomarker Profile and Brain Aging in the Swedish Good Aging in Skåne Study (GÅS)

Abstract

Introduction

Magnetic resonance imaging of the brain reveals age-related pathologies like white matter hyperintensities (WMH), cerebral microbleeds (CMB), lacunar infarcts (LAC) and grey- and white matter atrophy. The term cerebral small vessel disease (CSVD) has been used to collectively describe these changes, believed to emanate from small vessel endothelial dysfunction, although pathophysiological mechanisms are still largely unknown. We hypothesized that an explorative investigation of the plasma biomarker profile in affected subjects might shed some light on underlying mechanisms involved in CSVD and brain atrophy.

Methods

In a cross-sectional design, we investigated 401 subjects aged 70-86 from the randomized population study Good Aging in Skåne Study (GÅS) with brain MRI and OLINK immune- assay proteomics of 257 serum proteins previously associated with cardiovascular disease (CVD II and CVD III) and inflammation. The Benjamini-Yekutieli correction was used for keeping the false discovery rate (FDR) at 5%.

Results

We could see no significant difference in protein expression in the individual markers of CSVD (WMH, CMB or LAC). We observed a significant association between CSVD severity score (including white and grey matter atrophies, WMH, CMB and LAC) and the elevation of 11 serum proteins (CTSL1, PGF, NTpBNP, TNFr2, GDF15, TNFr1, IL4RA, ADM, CXCL9, TFF3, BNP). Furthermore, 11 proteins were significantly associated with Cortical Atrophy (CDH5, IL4RA, TNFr1, PGF, TF, TNFr2, CD93, CTSL1, LTBR, TNFRSF11A, TNFRSF10A). Five of the proteins were significant in both models. We found an association between moderate/severe medial temporal lobe atrophy (MTA) according to Scheltens scale and overexpression of PI3. Atrophy of presumed non-vascular origin was significantly associated with a greater abundance of IL4RA. All models were corrected for FDR and entered into a multivariable model with age and sex as covariates.

Discussion

In a general population cohort of older adults, proteomic analysis of serum identified several proteins associated with MRI-markers of CSVD and brain atrophy. Many of the identified protein biomarkers have previously been associated with hypertension, metabolic disease, or chronic kidney failure. This emphasizes the importance of systemic vascular health on cerebral pathological changes.