Integrating Polygenic Architecture and Metabolomic Signatures in the Pathobiology of Multiple Sclerosis

Multiple sclerosis (MS) is usually described in terms of MRI lesions and relapses, but beneath those clinical features lies a dense molecular story involving both inherited risk and dynamic biochemical changes. Large genetic studies show that MS is a classic complex trait: many common variants, each with small effects, aggregate to bias the immune system toward autoimmunity, while environment and lifestyle decide whether that bias turns into disease. In parallel, metabolomics has revealed consistent alterations in amino acids, lipids and energy-related metabolites in blood, cerebrospinal fluid (CSF) and even brain tissue, suggesting that immune cells and CNS-resident cells operate in a chronically “rewired” metabolic state. Together, genetics and metabolomics are beginning to connect who is at risk of MS with how the disease behaves in real time.

Polygenic Architecture: How DNA Loads the Dice
Family clustering and twin concordance studies already hinted that MS has a strong heritable component; genome-wide association studies (GWAS) turned that suspicion into hard numbers. The strongest susceptibility signal sits in the HLA region—especially HLA-DRB1*15:01—which modifies how antigens are presented to CD4⁺ T cells and substantially increases risk. Beyond the MHC, more than 200 independent loci have been mapped, most near genes that regulate T-cell activation (IL2RA, IL7R), B-cell signaling, microglial function and cytokine pathways. These data place peripheral immune cells and microglia, rather than neurons themselves, at the center of genetic risk. Crucially, no single variant is deterministic; it is the cumulative polygenic load, interacting with factors such as vitamin D status, Epstein–Barr virus infection, smoking and obesity, that ultimately determines whether MS emerges.

Metabolomics: Capturing the “Now” of MS Biology
While germline genetics is fixed at conception, metabolite profiles are highly dynamic, reflecting both the host’s biology and external exposures. Modern metabolomics platforms (typically high-resolution mass spectrometry or NMR) can quantify hundreds to thousands of metabolites from serum, plasma, CSF, urine or feces in a single run. In MS, these studies repeatedly highlight disturbances in energy metabolism (glycolysis and TCA-cycle intermediates), amino acids (including branched-chain and aromatic species), lipids (phospholipids, sphingolipids, acyl-carnitines) and oxidative-stress–related molecules. Because metabolites are shaped by both genes and environment, they form a mechanistic bridge between inherited susceptibility and the inflammatory milieu in which demyelination and neurodegeneration occur.

What Blood and CSF Metabolites Tell Us About MS
Cross-sectional and longitudinal studies of serum and CSF reveal that metabolite shifts are not random noise, but track with clinically relevant features. In blood, specific metabolite combinations—often involving lipids, steroid-like molecules and organic acids—associate with disability scores, lesion burden and, in some cohorts, rate of brain atrophy, suggesting that systemic metabolic state mirrors CNS damage. CSF metabolomics, which samples closer to the lesion environment, has identified changes in amino acids and fatty acids in early MS compared with controls, some of which correlate with Expanded Disability Status Scale (EDSS) progression. These profiles support the idea that mitochondrial stress, altered glial metabolism and chronic low-grade inflammation leave a detectable biochemical fingerprint in accessible fluids, opening the door to less invasive monitoring than repeated lumbar punctures or frequent MRI.

Multi-Omics: Connecting Genes, Metabolites and Disease Severity
The most informative picture comes from multi-omic designs where genetics, transcriptomics and metabolomics are measured in the same individuals. In a large study of people with MS, integration of blood metabolomics with RNA-seq and clinical measures showed that genes encoding enzymes in aromatic amino-acid metabolism (tryptophan, phenylalanine, tyrosine) were differentially expressed in immune cells, with corresponding shifts in downstream metabolites that associated with brain volume loss and disability. These data suggest a coherent pathway in which genetically influenced immune signaling reshapes metabolic programs, which in turn affect neurodegenerative processes. Similar multi-omic analyses have begun to define molecular “subtypes” of MS with distinct immune and metabolic signatures—some dominated by oxidative-stress pathways, others by lipid signaling—offering a potential explanation for heterogeneous treatment responses.

Are Metabolites Drivers or Passengers? Lessons from Mendelian Randomization
A critical question is whether altered metabolites simply reflect ongoing inflammation or actively contribute to MS pathogenesis. Mendelian randomization (MR) is a powerful tool here: by using genetic variants that influence metabolite levels as instrumental variables, MR tests whether a lifelong tendency toward higher or lower levels of a metabolite changes MS risk. A recent metabolome-wide MR screen used genetic instruments for more than 500 circulating metabolites and evaluated their causal effects on MS susceptibility using large GWAS datasets. Several candidates emerged whose genetically predicted levels were associated with higher or lower MS risk, implicating specific lipid and amino-acid–related pathways as potential drivers rather than mere biomarkers. Follow-up work is now applying similar MR frameworks to CSF metabolites and circulating proteins to map causal chains from genes → proteins → metabolites → MS outcomes.

Toward Mechanism-Guided, Metabolite-Informed MS Care
Putting these strands together, a future MS clinic may look quite different. Genetic data could provide a baseline polygenic risk score and highlight key immune pathways that are “hard-wired” in an individual, while targeted metabolomics panels in blood or CSF would report which biochemical circuits are currently active or under stress. Such combined profiles could help stratify newly diagnosed patients by likely disease course, select therapies that best match their dominant pathogenic pathways (e.g. strongly inflammatory vs metabolically stressed phenotypes), and offer sensitive, minimally invasive markers of treatment response or subclinical progression. For now, major barriers remain—metabolite levels are context-dependent, analytical methods are not fully standardized, and many reported biomarkers lack replication across diverse cohorts—but the direction of travel is clear: genetics tells us how the dice are loaded; metabolomics shows how the game is actually being played.

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:
Patsopoulos NA. Genetics of multiple sclerosis: an overview and new directions. Cold Spring Harb Perspect Med. 2018;8(7):a028951. Smusz J, et al. Metabolomics in multiple sclerosis: advances, challenges, and clinical perspectives—A systematic review. Int J Mol Sci. 2025;26(18):9207. Fitzgerald KC, et al. Multi-omic evaluation of metabolic alterations in multiple sclerosis identifies shifts in aromatic amino acid metabolism. Cell Rep Med. 2021;2(10):100424. Ge A, et al. A metabolome-wide Mendelian randomization study prioritizes potential causal circulating metabolites for multiple sclerosis. J Neuroimmunol. 2023.