Rare Deleterious Variation in GWAS-Linked Genes In Familial Multiple Sclerosis: Evidence from Whole-Exome Sequencing
Multiple sclerosis (MS) is a chronic immune-mediated disorder of the central nervous system characterized by inflammatory demyelinating lesions and progressive neurodegeneration. Large genome-wide association studies (GWAS) have mapped more than 200 common risk variants for MS, yet these common alleles—typically of modest effect size—do not fully explain the disease’s expected heritability. This “missing heritability” has motivated increasing attention to the rare variant hypothesis, which posits that low-frequency alleles with moderate-to-large functional effects can meaningfully contribute to complex disease susceptibility. Familial multiple sclerosis (FMS), defined by the presence of MS in first-degree relatives and representing a clinically important subset of cases, provides a particularly informative setting to test whether rare deleterious variation concentrates in pedigrees more than in sporadic multiple sclerosis (SMS).
Study Rationale and Primary Hypothesis
Turk and colleagues examined whether rare, predicted pathogenic variants are enriched in genes already implicated by MS GWAS, and whether such enrichment differs between familial and sporadic disease. The central hypothesis was explicitly comparative: rare predicted pathogenic (RPP) variants in GWAS-associated genes would contribute disproportionately to FMS risk relative to SMS, consistent with a model in which familial clustering is partially driven by rarer variants of larger effect. To test this, the investigators evaluated rare variation across a curated set of MS GWAS-associated genes in 87 FMS cases, 89 SMS cases, and a large control group of 3,866 individuals of similar Slavic ancestry.
Cohorts, Exome Sequencing, and Variant Definition
All participants underwent whole-exome sequencing (WES), generated on Illumina HiSeq with an average coverage of approximately 30×, with reads aligned to hg38 and variants filtered using standard quality metrics (including depth and genotype quality thresholds and GATK quality passing status). Variants were annotated using Ensembl Variant Effect Predictor (VEP), supplemented with LOFTEE to classify loss-of-function calls, and CADD to prioritize variants with higher predicted deleteriousness; population allele frequencies were assessed with gnomAD. The analytic definition of RPP variants was intentionally stringent: only rare variants (allele frequency < 1% or absent from gnomAD) were considered, and included either high-impact consequences (e.g., splice donor/acceptor, frameshift, stop gained/lost, start lost) or missense variants that were concordantly predicted deleterious by multiple algorithms and had CADD ≥ 20, while synonymous changes were excluded.
Gene Panel Construction and Burden-Testing Framework
The gene panel comprised 111 autosomal protein-coding genes nominated from the International Multiple Sclerosis Genetics Consortium (IMSGC) MS GWAS loci, including 99 genes outside the extended MHC and 12 within the MHC. Analytically, the study employed gene-based burden testing using generalized linear models, aggregating qualifying RPP variants per gene and comparing burden between cases and controls under three contrasts: (i) all MS cases versus controls, (ii) SMS versus controls, and (iii) FMS versus controls, with false discovery rate correction to control for multiple testing. The workflow schematic provided in the article (Figure 1 on page 3) concisely illustrates the pipeline from WES through gene filtering to panel- and gene-level burden analyses.
Key Findings: Strong Enrichment in Familial MS, Not Sporadic MS
The principal result is a striking divergence between familial and sporadic disease: RPP variants in the GWAS-associated gene set were significantly overrepresented in FMS compared with controls (FDR-adjusted p = 5.27 × 10⁻⁷⁴), whereas SMS showed no evidence of increased burden (p = 1.00), and the combined MS cohort showed a weaker signal consistent with FMS driving the aggregate association. At the gene level (summarized in Table 1 on page 3), six genes contributed significantly to the FMS burden—ALPK2, ANKRD55, INTS8, IQCB1, JADE2, and MALT1—whereas SMS showed a statistically significant enrichment only for LIMK2. Notably, among the variants observed in the enriched genes across analyses, a substantial proportion lacked recorded allele frequency in gnomAD, aligning with the study’s emphasis on very rare variation as a plausible explanatory component of familial aggregation.
Biological Plausibility of the Implicated Genes and Pathways
The genes highlighted in FMS converge on immunoregulatory and neurobiological processes that are mechanistically compatible with MS pathogenesis. MALT1, a key mediator of lymphocyte activation and NF-κB signaling, is especially compelling given its established role in pro-inflammatory gene expression and evidence from murine demyelination models that MALT1 protease inhibition attenuates neuroinflammation and tissue injury. ANKRD55 is already linked through common intronic GWAS signals to MS and other autoimmune disorders, and its emergence here as a rare-variant burden gene in FMS provides a coherent bridge between common-variant mapping and rare-variant contribution. The additional genes (ALPK2, INTS8, JADE2, IQCB1) are discussed by the authors in relation to pathways such as WNT/β-catenin signaling, transcriptional and RNA-processing regulation, chromatin-associated regulation of neurogenesis, ciliary biology, and immune infiltration signatures—each potentially intersecting with neuroinflammation, glial function, and immune–CNS interactions that define MS as an immune-mediated neurodegenerative condition.
Study Limitations and Implications for MS Genetic Models
Several limitations temper direct clinical translation but do not diminish the conceptual importance of the findings. The FMS cohort included only one affected individual per family, precluding segregation analyses that would strengthen causal inference for specific variants, and the modest case sample size limits power for detecting rarer alleles of smaller effect. The authors also note potential contributions from linked common variation and the absence of functional validation for candidate variants. Nevertheless, the overarching implication is clear: even when restricting analysis to genes nominated by GWAS, rare predicted pathogenic variants appear to concentrate preferentially in familial MS, supporting a composite genetic architecture in which common polygenic risk and rare, higher-impact variation may differentially contribute across MS subtypes. This framework aligns with the study’s interpretation that familial clustering may partially reflect increased rare-variant loading, and it provides a concrete gene set for follow-up studies integrating segregation, functional assays, and joint modeling with polygenic risk scores.
Disclaimer: This blog post is based on the provided research article and is intended for informational purposes only. It is not intended to provide medical advice. Please consult with a healthcare professional for any health concerns.
References:
Turk, A., Maver, A., Juvan, P. et al. Increased burden of rare variants in GWAS associated genes in familial multiple sclerosis. Sci Rep 15, 21200 (2025). https://doi.org/10.1038/s41598-025-04741-7
