Metabolic Signatures of Multiple Sclerosis: Uncovering Molecular Clues Through Multi-Omic Analysis
Multiple sclerosis (MS) is a chronic autoimmune disorder of the central nervous system characterized by inflammation, demyelination, failure of effective remyelination, and progressive neurodegeneration. Although disease-modifying therapies can reduce long-term neurological disability, their benefit depends partly on timely diagnosis and treatment initiation. Current diagnostic procedures rely substantially on neurological manifestations and radiological evidence of lesions disseminated in space and time, which can contribute to diagnostic delays. For this reason, a reliable blood-based biomarker would have considerable clinical value. In the article “Metabolome-based signature of disease pathology in MS,” Andersen and colleagues investigated whether circulating metabolites could provide such a molecular signature while simultaneously revealing biological mechanisms underlying MS pathology. The study is particularly notable because it did not examine metabolites in isolation; instead, it integrated metabolomics with gene-expression and genetic data to investigate connections between altered metabolism, immune activity, and inherited MS susceptibility.
A Multi-Platform and Multi-Omic Experimental Strategy
The investigators studied serum from 12 male patients with MS and 13 male controls. Participants were non-Hispanic White, non-smokers, and controls were frequency matched to cases for age and body mass index; importantly, the MS participants had received no disease-modifying treatment for at least three months before biospecimen collection. The researchers combined untargeted two-dimensional gas chromatography–time-of-flight mass spectrometry (GC×GC-TOFMS) with the targeted Biocrates AbsoluteIDQ p150 platform, enabling simultaneous examination of amino acids, acylcarnitines, phospholipids, sphingolipids, and additional metabolites. Of approximately 400 metabolite variables initially measured, 325 passed quality-control criteria. Batch effects were addressed with ComBat, after which random-forest machine learning was used to identify metabolites with the greatest importance for classifying MS status. The random forest contained 5,000 trees, and candidate metabolites were subsequently evaluated using logistic regression and receiver operating characteristic analysis. An area under the ROC curve exceeding 80% was selected as the threshold for the strongest candidates. This metabolomic analysis was then complemented by whole-genome expression measurements and genotyping of established MS-associated genetic variants, creating a genuinely multi-omic framework for interpreting the metabolic observations.
Six Metabolites Form a Promising MS-Associated Signature
Random-forest analysis initially identified 12 metabolites as informative for classification of MS, eight of which were significantly associated with disease status in logistic regression. Six achieved an AUC greater than 80% and therefore emerged as the study's strongest candidate biomarkers: pyroglutamate, laurate, phosphatidylcholine PC ae C42:5, tetradecenoyl-L-carnitine (acylcarnitine C14:1), phosphatidylcholine PC ae C40:5, and N-methylmaleimide. Pyroglutamate and laurate produced AUC values of approximately 0.85 and 0.86, respectively, while the two phosphatidylcholines, acylcarnitine C14:1, and N-methylmaleimide produced AUC values of approximately 0.81–0.82. Importantly, all six metabolites were present at higher levels in the MS group than in controls, a pattern illustrated by the boxplots presented in Figure 1 on page 5 of the article. These compounds are biologically diverse rather than members of a single metabolic pathway: they represent glutathione metabolism, fatty-acid metabolism and oxidation, membrane phospholipid biology, mitochondrial energy metabolism, and electrophilic/TRPA1-related signaling. Consequently, the findings suggest that MS is associated not with one isolated metabolic defect but with a broader disturbance involving oxidative balance, lipid handling, cellular membranes, immunity, and mitochondrial function.
Pyroglutamate and Laurate Connect Metabolism with Oxidative and Immune Stress
Pyroglutamate, also called 5-oxoproline, ranked first in the random-forest analysis and provides a particularly compelling link between the metabolome and oxidative pathology. It is produced during glutathione metabolism, and glutathione is one of the principal intracellular antioxidant systems. The elevated pyroglutamate observed in MS may therefore indicate disturbed antioxidant regulation and increased oxidative stress, processes already relevant to neuronal and mitochondrial injury. The authors further identified pyroglutamate-associated enrichment of glutathione biosynthesis, iron-homeostasis, ceramide, and sphingosine-related pathways, connecting this metabolite with oxidative damage, apoptosis, and myelin biology. Laurate, or lauric acid, offers a different but complementary mechanistic perspective. This medium-chain saturated fatty acid has previously been associated experimentally with the differentiation of pro-inflammatory Th1 and Th17 cells and reduced regulatory T-cell differentiation. In experimental autoimmune encephalomyelitis, increased dietary laurate has also been linked with more severe disease. Thus, its elevation in the MS participants may reflect an interaction between lipid metabolism and immune activation. Associated signaling pathways—including inflammasome-related, lymphotoxin, IL-8, and immune-cell signaling processes—further support the possibility that alterations in circulating fatty acids participate in the inflammatory environment characteristic of MS rather than functioning merely as passive metabolic by-products.
Phospholipids and Acylcarnitines Point Toward Membrane and Mitochondrial Dysfunction
The lipid-associated candidates provide additional insight into the structural and energetic abnormalities of MS. PC ae C40:5 and PC ae C42:5 are phosphatidylcholines, major constituents of cellular membranes and myelin. Their altered circulating concentrations are therefore biologically relevant in a disease characterized by extensive myelin injury. The researchers also found that expression of PLA2G4C, a phospholipase A2-family gene with potential roles in mitochondrial function and immune-cell biology, was positively associated with both phosphatidylcholines. Acylcarnitine C14:1 provides an even more direct connection with mitochondrial energy metabolism because acylcarnitines are intermediates generated during fatty-acid transport and oxidation. Elevated concentrations can indicate incomplete or disturbed fatty-acid oxidation and altered mitochondrial function. Consistent with this interpretation, genes associated with C14:1 were enriched for antigen-presentation and immune pathways, while several metabolite-associated gene sets implicated oxidative phosphorylation and sirtuin signaling. N-methylmaleimide, the sixth candidate, was associated with pathways involving mitochondrial dysfunction, oxidative phosphorylation, sirtuin signaling, apoptosis, and the mevalonate pathway. Although its precise biological role in MS remains less well characterized, its relationship with TRPA1 signaling is noteworthy because TRPA1 has been implicated experimentally in neurogenic inflammation and oligodendrocyte injury. Collectively, these observations place altered mitochondrial bioenergetics and membrane lipid metabolism near the center of the metabolic phenotype identified by the study.
Genetics and Gene Expression Strengthen the Biological Interpretation
A major strength of the investigation was its attempt to connect metabolic abnormalities with transcriptional and inherited genetic variation. Expression data from 9,067 quality-controlled genes were examined in relation to each of the six leading metabolites. Although HLA-DRB1 expression itself was not significantly related to the metabolites, expression of several other HLA class II genes—including HLA-DMA, HLA-DMB, HLA-DOA, HLA-DPA1, HLA-DPB1, HLA-DRA, HLA-DRB3, and HLA-DRB6—was associated with acylcarnitine C14:1. Pathway analysis further linked C14:1 with antigen presentation, while PC ae C40:5 and N-methylmaleimide were prominently related to mitochondrial dysfunction. Across metabolites, recurrent enrichment was observed for apoptosis-associated pathways, oxidative phosphorylation, mitochondrial dysfunction, and sirtuin signaling, suggesting convergence on a relatively small group of cellular processes. Genetic analyses added another layer of evidence: the major MS susceptibility allele HLA-DRB1*15:01 was significantly associated with acylcarnitine C14:1, whereas HLA-A*02 was not. Multiple non-MHC MS risk variants were also related to metabolite concentrations, including variants involving ETS1, IL2RA, and AFF1 that were associated with both phosphatidylcholines. These relationships do not demonstrate that the genetic variants directly cause the metabolic abnormalities, but they provide an important mechanistic bridge linking inherited MS susceptibility with immune regulation, lipid metabolism, and cellular energy pathways.
Significance, Limitations, and the Future of Metabolomic Biomarkers
The study presents an important proof of concept for integrating metabolomics, transcriptomics, and genetics to characterize MS as a systemic biochemical disorder as well as an inflammatory neurological disease. Its methodological strengths include the combination of targeted and untargeted metabolomics, careful matching of participants, exclusion of recent disease-modifying treatment as a major confounder, and the integration of metabolic findings with gene-expression and genetic information. Nevertheless, the results must be interpreted as exploratory rather than as validation of a clinically applicable diagnostic test. The cohort consisted of only 25 participants, no independent replication population was available, and the intentionally homogeneous sample consisted exclusively of non-Hispanic White, non-smoking men. Furthermore, the study examined individuals with established MS rather than patients at the moment of disease onset, meaning that the metabolites could represent consequences of ongoing disease pathology rather than factors that precede or predict MS development. Future investigations therefore require substantially larger and more diverse populations, independent validation cohorts, inclusion of different sexes and MS subtypes, longitudinal sampling, and comparison with other inflammatory and neurodegenerative diseases to determine disease specificity. Despite these limitations, the six-metabolite signature provides a biologically coherent framework in which altered glutathione metabolism, lipid and fatty-acid processing, mitochondrial dysfunction, apoptosis, immune activation, and genetic susceptibility intersect. Rather than establishing a finished diagnostic biomarker, the work provides a foundation for understanding how the circulating metabolome may reveal both measurable markers and mechanistic features of multiple sclerosis.
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:
Villoslada, P., & Baranzini, S. (2012). Data integration and systems biology approaches for biomarker discovery: challenges and opportunities for multiple sclerosis. Journal of neuroimmunology, 248(1-2), 58-65.
