Summary: A landmark neurobiology study delivers the first direct, data-driven comparison of the two most widely used preclinical models for multiple sclerosis (MS). Focusing on the loss and recovery of myelin—the fatty insulating layer that surrounds nerve fibers—the research team mapped cellular and genetic responses in the cuprizone (CPZ) and lysophosphatidylcholine (LPC) models and compared those findings with human MS tissue.
Using single-cell RNA sequencing to chart how individual cells and genes respond during demyelination and remyelination, the study demonstrates that CPZ and LPC are not interchangeable. Instead, each model produces distinct spatial, temporal, cellular, and genetic signatures. The findings create a practical roadmap to guide researchers in selecting the most appropriate model for specific therapeutic goals in MS drug development.
Key Facts
- The role of myelin: Multiple sclerosis affects over one million people in the United States and involves the immune-driven destruction of myelin, which insulates axons much like plastic insulating electrical wires.
- Why models matter: Because obtaining fresh brain and spinal cord tissue from people with progressive MS is difficult, scientists depend on animal models to study injury, repair, and therapeutic strategies.
- Differences between CPZ and LPC: Both paradigms lead to myelin loss, but they operate on very different scales. CPZ produces widespread, gradual demyelination over weeks, while LPC creates a single, focused lesion within days.
- Model selection guidance: CPZ is better suited for studying oligodendrocyte stress, death, and intrinsic repair mechanisms due to its slower timeline. LPC is preferable when examining acute, aggressive immune responses at lesion sites.
- Translation to human disease: The researchers generated single-cell genetic maps from both mouse models and matched them against human MS tissue to verify which model features most closely reflect patient lesions.
- Therapeutic gap: Current MS treatments largely aim to suppress harmful immune activity. Regenerating lost myelin within established lesions remains an important but as-yet-unrealized therapeutic objective.
Source: University of Notre Dame
More than 1 million people across the United States live with multiple sclerosis (MS), a disease that affects the brain, optic nerves, and spinal cord.
MS is a variable and unpredictable disorder. Symptoms such as extreme fatigue, muscle spasms, and vision impairment can flare and then remit over diverse intervals. To develop new therapies, researchers must first understand the cellular damage at the heart of the disease.
Katrina Adams, a neurobiologist at the University of Notre Dame, directs research into how myelin loss and recovery influence MS progression. Myelin is a lipid-rich coating that wraps axons and enables efficient transmission of electrical signals across the nervous system. When myelin is lost in MS, the resulting lesions vary in size, location, and cellular composition, which complicates both study and treatment.
Because collecting high-quality tissue from patients with progressive MS is difficult, preclinical models are essential for testing hypotheses about myelin injury and repair. Adams’ team set out to evaluate two commonly used models—cuprizone (CPZ) and lysophosphatidylcholine (LPC)—and determine which aspects of human disease each best replicates.
Published in Nature Communications, the study applies single-cell transcriptomics to compare the cellular states and gene expression changes across the two models and in human MS lesions. The researchers mapped which cell types and genes are activated during demyelination and remyelination phases and identified where models align with or diverge from patient tissue.
“Our analysis of these two models of myelin loss and regeneration provides a road map based on robust scientific evidence that we hope will advance the study of MS and related diseases,” said Adams, who is the Gallagher Assistant Professor in the Department of Biological Sciences.
The research clarifies practical choices for investigators. CPZ’s gradual, widespread myelin loss creates a context that reveals stress responses and progression in oligodendrocytes—the myelin-producing cells—making it ideal to study cell-intrinsic damage and repair. LPC, with its rapid, localized lesion and a stronger immune activation, is better suited for exploring acute immune-mediated events at lesion sites.
Importantly, the team compared the cellular and genetic signatures from each mouse model directly with data from human MS tissue. Single-cell RNA sequencing allowed them to determine which gene expression shifts observed in mice also occur in patient lesions. This cross-species matching increases confidence that discoveries in a given model will translate to human disease.
The study also uncovered model-specific genetic differences. Certain oligodendrocyte states and gene expression patterns were prominent in CPZ or LPC but not identical to those in human tissue. Some genetic changes were unexpected, and their implications—whether they promote or impede remyelination—remain to be determined. These differences highlight avenues for further study.
Clinically, most approved MS therapies work by dampening the immune system to limit new attacks. By contrast, interventions that actively rebuild lost myelin inside established lesions are not yet available. The authors argue that selecting the right preclinical model is crucial for advancing therapeutic strategies that aim not only to prevent damage but to restore nerve function.
“The strategic use of these two preclinical models is essential for translating insights into therapies that might restore lost myelin,” Adams said. “We need to better understand the process of demyelination and repair to address one of the underlying causes of this debilitating disorder.”
Key Questions Answered:
A: Although CPZ and LPC both produce myelin loss, this study shows they create very different cellular and genetic environments. Using an inappropriate model risks testing therapies against biological mechanisms that do not match the human disease, reducing the likelihood of clinical success.
A: Present treatments are primarily immunomodulatory; they limit new autoimmune attacks but do not directly regenerate myelin inside established lesions. Repairing existing damage—remyelination—remains a key unresolved therapeutic aim, and this study offers a framework to pursue that goal.
A: Single-cell sequencing reveals which genes individual cells express during injury and recovery. By applying this approach to both mouse models and human MS lesions, researchers can identify shared genetic changes in the same cell types, indicating validated targets for further drug discovery.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- The journal paper was reviewed in full.
- Additional context was provided by our editorial staff.
About this multiple sclerosis research news
Author: Brandi Wampler
Source: University of Notre Dame
Contact: Brandi Wampler – University of Notre Dame
Image: The image is credited to Neuroscience News
Original Research: Open access.
“A comparative transcriptomic analysis of mouse demyelination models and multiple sclerosis lesions” by Erin L. Aboelnour, Veronica R. Vanoverbeke, Elizabeth A. Maupin, Madelyn M. Hatfield & Katrina L. Adams. Nature Communications
DOI: 10.1038/s41467-026-72383-y
Abstract
A comparative transcriptomic analysis of mouse demyelination models and multiple sclerosis lesions
Demyelinating diseases, including multiple sclerosis (MS), are defined by the loss of myelin and progressive neurodegeneration. It has been unclear whether commonly used mouse models of demyelination, such as cuprizone (CPZ) and lysophosphatidylcholine (LPC), trigger distinct cellular responses or adequately mirror human disease.
This study integrates new and published single-cell transcriptomic datasets from CPZ- and LPC-induced demyelination and compares them with human MS lesion data. The analysis reveals that CPZ induces a stressed oligodendrocyte state—marked by genes such as Cdkn1a and Nupr1—that resembles phenotypes found in MS lesions. Both models converge on an immune-responsive oligodendrocyte state expressing Socs3, B2m, and interferon-response genes during remyelination. Mouse microglia exhibit a conserved activation program, with LPC producing a stronger and more prolonged response. Nonetheless, neither model fully captures the oligodendrocyte progenitor and microglial heterogeneity observed in human MS tissue.
These results create a cross-model, cross-species atlas of glial states and provide a structured framework to use mouse models strategically for studying myelin injury and repair.