Blood Metabolomics Uncovers Glycolysis as a Potential Therapeutic Target in Multiple Sclerosis
Multiple sclerosis (MS) is a chronic inflammatory and demyelinating disorder of the central nervous system in which immune cells attack myelin and contribute to neuronal and axonal damage. The relapsing-remitting form of the disease (RRMS) is characterized by recurrent episodes of neurological deterioration followed by partial or complete recovery. Although modern disease-modifying therapies can reduce relapse frequency and inflammatory activity, they do not provide a definitive cure, making the identification of new biological targets an important research objective. In the 2022 Proceedings of the National Academy of Sciences study by Zahoor and colleagues, the investigators approached this problem through untargeted metabolomics, a technology capable of simultaneously measuring hundreds of small molecules generated by cellular metabolism. Because metabolites represent the downstream products of gene expression, protein activity, diet, cellular signaling, and environmental influences, they can provide a particularly sensitive picture of an individual's current physiological state. The study therefore asked whether the circulating metabolic profile of patients with RRMS could reveal not merely biomarkers of disease, but metabolic processes that could themselves be therapeutically manipulated.
A Distinct Metabolic Signature in the Blood of Patients with RRMS
The researchers recruited 35 patients with RRMS overall, while the serum metabolomic analysis described in the results included 33 untreated patients and 14 age- and sex-matched healthy subjects. Using comprehensive mass-spectrometry-based metabolic profiling, they detected 632 known serum metabolites, of which 60 differed significantly between RRMS and healthy controls, with a false-discovery rate below 0.10. Remarkably, 53 of these metabolites were increased and only seven were decreased in the RRMS group. Multivariate partial least-squares discriminant analysis demonstrated a clear separation between the metabolic profiles of patients and controls, while the heat map presented in Figure 1 on page 3 visually showed distinct clustering of the altered metabolites. Lipids represented approximately 46% of the metabolic alterations, followed by xenobiotics, peptides, amino acids, carbohydrates, vitamins and cofactors, energy-associated metabolites, and nucleotides. These results indicate that RRMS is associated with a broad systemic metabolic disturbance rather than a change in one isolated biochemical reaction. Importantly, causal-network analysis also implicated sphingosine-1-phosphate receptor 2 and transforming growth factor-β1 as possible regulatory nodes, reinforcing the connection between the observed circulating metabolites and established inflammatory signaling mechanisms in MS.
Four Metabolic Pathways Converge on Glycolysis
To determine whether the altered metabolites belonged to biologically connected pathways, the investigators integrated several bioinformatics platforms, including Metscape, MetaboAnalyst, the Kyoto Encyclopedia of Genes and Genomes pathway library, and Ingenuity Pathway Analysis. Four pathways emerged as particularly important: glycerophospholipid metabolism, the tricarboxylic acid or citrate cycle, sphingolipid metabolism, and pyruvate metabolism. Figure 2 on page 4 illustrates the central observation of the paper: all four pathways can be metabolically linked to glycolysis, the biochemical process through which glucose is converted into pyruvate while producing ATP and biosynthetic intermediates. Pyruvate may subsequently become lactate or enter the mitochondrial TCA cycle through acetyl-CoA, while other glycolytic intermediates can contribute to glycerol and lipid biosynthesis. Consequently, glycolysis does considerably more than supply cellular energy; it provides the molecular building blocks required for membrane synthesis, proliferation, signaling, and inflammatory effector functions. The convergence of several apparently different metabolic abnormalities on a common upstream pathway therefore led the investigators to propose that enhanced glycolytic activity might be a central feature of immune dysfunction in RRMS. Interestingly, analysis of glycolysis-associated gene expression showed relatively few transcriptional differences, suggesting that the abnormal metabolic state may depend more strongly on enzyme activity, substrate availability, or inflammatory regulation than on large changes in gene expression.
Immune Cells from Patients with RRMS Exhibit Increased Glycolytic Activity
The investigators next tested whether the serum metabolic signature reflected a functional change in circulating immune cells. Peripheral blood mononuclear cells (PBMCs) obtained from patients with RRMS and healthy subjects were examined using a Seahorse extracellular flux analyzer, which measures extracellular acidification rate as an indicator of glycolytic activity. PBMCs from patients with RRMS showed significantly greater basal glycolysis than cells obtained from healthy individuals. A comparable increase was detected in PBMCs from mice with experimental autoimmune encephalomyelitis (EAE), a widely used animal model of MS. These findings provided an important bridge between observational metabolomics and functional immunology: the circulating metabolite abnormalities were not simply passive biochemical markers but corresponded to a measurable metabolic alteration within immune cells. This observation is biologically plausible because activated T cells, monocytes, macrophages, and other immune populations commonly increase their dependence on glycolysis during inflammatory responses. Rapid glucose metabolism supplies both energy and metabolic intermediates required for proliferation, cytokine synthesis, membrane production, and migration. Thus, the heightened glycolytic state detected in RRMS PBMCs could help sustain the inflammatory phenotype responsible for immune-mediated damage within the central nervous system.
Blocking Glycolysis Suppressed Disease in Experimental Models
Having identified glycolysis as a potentially dysregulated pathway, the authors tested whether suppressing it could influence disease progression. They used 2-deoxy-D-glucose (2DG), a glucose analogue that is phosphorylated by hexokinase but cannot proceed normally through glycolysis, thereby restricting glycolytic flux. Treatment with 2DG produced substantial improvements across several EAE models. In relapsing-remitting SJL mice, it significantly reduced peak disease severity and almost completely suppressed relapse; similar protective effects were observed in C57BL/6 and 2D2 T-cell-receptor-transgenic models. Oral administration through drinking water also delayed disease development and reduced neurological severity. Histological findings shown in Figure 4 on page 5 supported the clinical results: treated animals exhibited fewer inflammatory cell infiltrates, smaller spinal-cord lesions, and markedly reduced demyelination. In addition, 2DG decreased myelin-specific lymphocyte proliferation and reduced the pro-inflammatory cytokines IFN-γ, IL-17, and GM-CSF while increasing the anti-inflammatory cytokine IL-10. Flow-cytometric analysis further demonstrated reductions in CNS-infiltrating CD4-positive T cells and in the number of T cells producing inflammatory cytokines. Collectively, these experiments provided proof of principle that the metabolic abnormality identified through human blood profiling could be converted into a testable therapeutic hypothesis and successfully manipulated in animal disease models.
Reprogramming Macrophages and Restricting Neuroinflammation
One of the study's most mechanistically significant findings concerned monocytes and macrophages, immune cells that contribute extensively to inflammatory tissue injury and myelin degradation in MS and EAE. Treatment with 2DG decreased macrophage infiltration into the central nervous system and reduced glucose consumption, lactate production, and glycolytic activity in isolated monocytes. At the same time, the treated cells displayed reduced expression of glycolytic genes and shifted away from a strongly pro-inflammatory state. Markers such as inducible nitric oxide synthase and IL-1β decreased, whereas markers associated with an anti-inflammatory macrophage phenotype, including arginase-1 and Ym1/2, increased. Most strikingly, transferring monocytes obtained from 2DG-treated animals into mice with active EAE reduced subsequent disease severity, supporting the conclusion that metabolic intervention had functionally altered the immune cells rather than merely suppressing their numbers. The study also demonstrated that a phosphorylated form of 2DG accumulated in spinal-cord tissue, indicating access to the central nervous system. In cultured brain glial cells, both pharmacological glycolysis inhibition and reduced glucose availability suppressed inflammatory mediators such as nitric oxide, iNOS, IL-1β, IL-6, TNF-α, and MCP-1. The experiments summarized in Figures 6 and 7 therefore suggest that inhibition of glycolysis may act at several levels simultaneously—restricting inflammatory immune-cell activity, promoting a more regulatory macrophage phenotype, and reducing inflammatory activation within the nervous system itself.
Therapeutic Promise, Scientific Limitations, and Future Directions
The principal importance of this study lies in its metabolomics-to-therapy workflow: researchers began with an unbiased survey of metabolites in patient blood, used pathway analysis to identify a central biochemical abnormality, confirmed that abnormality directly in patient immune cells, and finally tested its therapeutic relevance in multiple experimental models. Nevertheless, the findings should not be interpreted as evidence that 2DG is currently an established treatment for people with MS. The human metabolomic cohort was relatively small, the therapeutic experiments were performed primarily in EAE rather than patients, and EAE does not reproduce every feature of human multiple sclerosis. Furthermore, glycolysis is fundamental to many healthy tissues, meaning that chronic systemic inhibition could produce unwanted physiological effects. The authors themselves emphasize the poor drug-like characteristics and short half-life of 2DG and present it principally as a proof-of-concept compound. Future research will therefore need to determine whether more selective inhibition of glycolysis in particular immune-cell populations, specific glycolytic enzymes, or disease-active periods can achieve therapeutic benefit while preserving normal metabolism. The study nevertheless demonstrates the considerable potential of blood-based metabolomics in precision neurology: metabolic signatures may eventually contribute not only to disease classification and biomarker development but also to the discovery of biologically rational treatment targets. By identifying heightened glycolysis as a functional component of inflammatory immune-cell behavior in RRMS, Zahoor and colleagues provide a compelling example of how systems-level metabolic analysis can connect molecular observations in patients with experimentally testable strategies for treating autoimmune neurological disease.
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:
Zahoor, I., Suhail, H., Datta, I., Ahmed, M. E., Poisson, L. M., Waters, J., ... & Giri, S. (2022). Blood-based untargeted metabolomics in relapsing-remitting multiple sclerosis revealed the testable therapeutic target. Proceedings of the National Academy of Sciences, 119(25), e2123265119.
