Brain-body Genetic Signatures for White Matter Hyperintensities and Alzheimer’s Disease

2143 

Abstract Submission 

Manpreet Singh¹, Kimia Shafighi², Sarah Taliun¹, Ikrame Housni³, Flavie Detcheverry³, Sridar Narayanan⁴, Danilo Bzdok⁵, AmanPreet Badhwar⁶

¹University of Montreal, Montréal, QC, ²McGill University, Montréal, QC, ³Université de Montréal, Montréal, Québec, ⁴Department of Neurology and Neurosurgery, McGill University, Montreal, Quebec, ⁵McConnell Brain Imaging Centre (BIC), Montreal Neurol, McGill Universityogical Institute (MNI), Montreal, Quebec, ⁶Département de pharmacologie et physiologie, Faculté de médecine, Université de Montréal, Montreal, Quebec

Manpreet Singh  
University of Montreal
Montréal, QC

Kimia Shafighi  
McGill University
Montréal, QC

Sarah Taliun  
University of Montreal
Montréal, QC

Ikrame Housni  
Université de Montréal
Montréal, Québec

Flavie Detcheverry  
Université de Montréal
Montréal, Québec

Sridar Narayanan  
Department of Neurology and Neurosurgery, McGill University
Montreal, Quebec

Danilo Bzdok  
McConnell Brain Imaging Centre (BIC), Montreal Neurol, McGill Universityogical Institute (MNI)
Montreal, Quebec

AmanPreet Badhwar  
Département de pharmacologie et physiologie, Faculté de médecine, Université de Montréal
Montreal, Quebec

White matter hyperintensities (WMHs), an established MRI-detected marker of vascular brain injury, are frequently present in Alzheimer's disease and related dementias (ADRDs), and are known to exert an independent effect on cognitive decline in these diseases (Debette and Markus 2010). Genetically, WMHs demonstrate high heritability in twin, sibship and family studies (~70%) (Turner et al. 2004; Atwood et al. 2004). While research on WMHs, as well as the ADRDs, have mostly focussed on the brain, emerging evidence points to crosstalk between brain and body (Makin et al. 2015),(Huang et al. 2022). However, the heritability of WMHs across tissue- and cell-types in the whole body (i.e., the brain-body axis) have not been characterized. Addressing this gap in knowledge, the aims of this study are to (1) characterize WMH tissue- and cell-specific partitioned heritability in the whole body, and (2) identify common partitioned heritability components between WMHs and ADRDs.

Assessment of tissue-specific heritability: We ran stratified-linkage disequilibrium score regression (sLDSC) on summary statistics from genome wide association studies (GWAS) (N=3 WMH studies; N=10 ADRD studies; Figure 1) for partitioning heritability of the phenotypes (i.e., WMHs or ADRDs) across published tissue-specific binary annotations (N=10) (Finucane et al. 2015). Per GWAS study, enrichment of phenotype-associated-single nucleotide polymorphism (SNPs) within each tissue-type was calculated. Assessment of cell-specific heritability: Using sLDSC, we partitioned heritability of phenotypes across published cell-specific continuous annotations (Cao et al. 2020) for all cell-types associated with tissues enriched for WMHs. -log10(enrichment_p) values were used to interpret the data and represent the strength of the association. Only significantly enriched cell types (Figure 2) are presented below (>-log10(0.05)). A 5% False Discovery Rate threshold was applied to correct for multiple comparisons.

We found that WMH-associated-SNPs were significantly enriched in four tissues. Cardiovascular and kidney enrichments were observed in WMH only, while CNS and liver enrichments were found common to both WMH- and AD-associated-SNPs (Figure 2A). Cell analysis within the four tissues enriched with WMH-associated-SNPs shows that 16/64 cell-types were also enriched, with vascular endothelial cells (vECs) being enriched in all four tissues. The tissue with the highest proportion of cell-types showing WMH-associated-SNPs enrichment was the liver (5/9 cell-types, 55%), followed by CNS (6/18 cell types, 33%). While WMHs and AD both showed enrichment in CNS and liver tissues, in the CNS, cell-specific analysis highlighted enrichment in distinct cerebellar cell-types, with inhibitory interneurons and Purkinje cells being enriched for AD and WMHs, respectively (Figure 2B). Cell-specific analyses on liver cells showed no AD-associated-SNPs enrichment.

Established literature highlights brain vEC dysfunction as a pathogenic mechanism of WMHs (Hassan et al. 2003). We demonstrated that SNPs associated with WMHs were enriched in vECs not only in the CNS, but also in cardiovasculature (heart, lungs), liver, and kidney tissues. In line with our findings, WMHs have also been linked to heart hypoperfusion (Berry et al. 2019), non-alcoholic fatty liver disease (Jang et al. 2019), and worse kidney function (Makin et al. 2015). Overall, our findings lend strength to the proposition that presence of MRI-detected WMHs is indicative of an underlying multi-system endothelial disorder affecting several vascular beds (Vogels et al. 2012). In addition, the enrichment of both WMH- and AD-associated SNPs in inhibitory CNS cells may suggest common mechanistic pathways. Further follow-up studies are required to enhance our understanding of the causative pathways associated with these multi-systemic genetic findings.

Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)

Genetic Modeling and Analysis Methods ²

Neuroanatomy Other ¹

Cellular

Cerebrovascular Disease

Computational Neuroscience

Informatics

Phenotype-Genotype

Statistical Methods

Other - white matter hyperintensities, Alzheimer's disease and related dementias, heritability analysis, linkage disequilibrium score regression