Metformin Boosts Myelin Production in Human Brain Cells via Mitochondrial Boost

Multiple sclerosis (MS) involves progressive loss of myelin — the protective sheath around nerve fibers — and the brain's ability to remyelinate declines sharply with age. A key reason is that oligodendrocyte progenitor cells (OPCs), which regenerate myelin, lose responsiveness over time. Metformin, a decades-old diabetes medication, was previously shown to reverse this aging deficit in rat OPCs. However, human oligodendrocytes differ substantially from their rodent counterparts, expressing unique genes and showing distinct myelination behavior. This study set out to determine whether metformin has comparable pro-myelination effects in human cells, and to understand the metabolic mechanism behind any such effect.

The research team generated oligodendrocyte progenitor cells from human embryonic stem cells (hESCs) and tested metformin across three culture systems: a standard 2D monolayer, brain organoids, and an in vivo chimera model in which human cells were transplanted into immunodeficient Shiverer mice. To benchmark the maturity of these cells, they performed single-cell and single-nuclear RNA sequencing, comparing the in vitro and chimera-derived human cells against published adult post-mortem human CNS tissue datasets from spinal cord and brain. Computational tools including canonical correlation analysis, machine learning-based artificial neural network classification, and cosine similarity scoring were used to map developmental stage and transcriptional identity.

In the monolayer system, seven days of metformin treatment produced a mean fold-change increase of 0.52 (±0.23 SEM) in OLIG2+MBP+ mature oligodendrocytes — comparable to the established pro-differentiation drug clemastine fumarate (0.48 ±0.17 SEM increase). Transcriptomic analysis confirmed that monolayer and organoid cells most closely resembled fetal human oligodendroglia (second to third trimester), while cells in the chimera model displayed the highest transcriptional similarity to adult human post-mortem oligodendrocytes. Despite this fetal-like character in the simpler models, metformin still increased myelin protein and sheath formation across all three systems, suggesting its pro-myelination effect does not require cellular maturity.

In the chimera model, the most translationally relevant system, metformin led to a significant increase in mitochondrial area in both the transplanted human cells and the surrounding mouse axons. This morphological change was accompanied by upregulation of transcripts related to mitochondrial function and broader metabolic processes. Crucially, post-mortem brain tissue from MS patients who had received metformin prior to death showed a strikingly similar transcriptional signature — increased expression of mitochondrial metabolism genes — compared to MS patients who had not received the drug. This convergence across experimental models and real human tissue provides strong translational confidence.

The authors conclude that metformin's pro-myelination effect in humans is not restricted to one specific brain cell type but represents a broadly acting metabolic shift, primarily mediated through mitochondrial changes. This is consistent with metformin's known mechanism of inhibiting Complex I of the mitochondrial electron transport chain, which activates AMPK and shifts energy metabolism. Because metformin crosses the blood-brain barrier and is already in multiple MS clinical trials (including NCT05349474 and ISRCTN14048364), these mechanistic findings provide important biological rationale for those trials. Caveats include the fetal-like nature of even the most advanced in vitro models and the small number of post-mortem human MS cases with metformin exposure.