Genome-wide study of DNA methylation in Amyotrophic Lateral Sclerosis identifies differentially methylated loci and implicates metabolic, inflammatory and cholesterol pathways

Abstract

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease with an estimated heritability of around 50%. DNA methylation patterns can serve as biomarkers of (past) exposures and disease progression, as well as providing a potential mechanism that mediates genetic or environmental risk. Here, we present a blood-based epigenome-wide association study (EWAS) meta-analysis in 10,462 samples (7,344 ALS patients and 3,118 controls), representing the largest case-control study of DNA methylation for any disease to date. We identified a total of 45 differentially methylated positions (DMPs) annotated to 42 genes, which are enriched for pathways and traits related to metabolism, cholesterol biosynthesis, and immunity. We show that DNA-methylation-based proxies for HDL-cholesterol, BMI, white blood cell (WBC) proportions and alcohol intake were independently associated with ALS. Integration of these results with our latest GWAS showed that cholesterol biosynthesis was causally related to ALS. Finally, we found that DNA methylation levels at several DMPs and blood cell proportion estimates derived from DNA methylation data, are associated with survival rate in patients, and could represent indicators of underlying disease processes.

Mitochondrial function determines severity but not risk of amyotrophic lateral sclerosis

“…MtCN is cell-specific 18 and ALS is known to alter the proportions of blood leukocytes 3 . To overcome this limitation we used a subset of our whole blood WGS cohort for whom we could derive proportions of WBC based on DNA methylation 27 ( Methods ); this subset included 3,549 ALS patients and 1,529 controls. As expected, specific mitochondrial haplotypes were associated with mtCN in our dataset after adjusting for WBC proportions (0_0_2_2_2_2 p=0.007, beta=6.12, se=2.26; 0_2_2_2_2_2, p=0.05, beta=9.01, se=4.65, multivariable linear regression, Figure 2a, Supplementary Table 1, Methods ); our findings match a previous study of the same haplotypes in peripheral blood 19 .…”

“…White blood cell (WBC) proportions including granulocytes, monocytes, natural killer (NK) cells, T-cells (CD4+ and CD8+) and B cells were estimated from methylation data in the Project MinE cohort as previously described 27 . Platelets counts were not available as they cannot be inferred from methylation data which may impact on the estimation of mtCN because platelets contain mitochondrial DNA but no autosomal DNA 16 .…”

“…MtCN is cell-specific 18 and ALS is known to alter the proportions of blood leukocytes 3 . To overcome this limitation we used a subset of our whole blood WGS cohort for whom we could derive proportions of WBC based on DNA methylation 27 Indeed, this protective haplotype was overrepresented in this cohort compared to the original test cohort (Supplementary Figure 1b). Direction of effect and effect size for all mitochondrial haplotypes are similar in both cohorts (Figure 2c-d).…”

Telomere length analysis in amyotrophic lateral sclerosis using large-scale whole genome sequence data

“…The last decade has seen substantial advances in our understanding of the genomic basis of ALS ( van Rheenen et al, 2021 ; Hop et al, 2022 ) but a significant proportion of the genetic contribution to risk remains unexplained. This hidden heritability may be harbored in other types of genomic variation as well as in rare variants that may be unique to an affected individual or family ( Al-Chalabi et al, 2010 ; McLaughlin et al, 2015 ).…”