A new article published in Science offers an unprecedented glimpse into how multiple sclerosis (MS) develops and spreads in the brain. Using a combination of live magnetic resonance imaging (MRI) and advanced genetic analysis in marmosets, a primate species with brains and immune systems extremely similar to humans, researchers traced the molecular and cellular evolution of MS-like lesions over both time and space. Their findings could pave the way for more precise imaging techniques and targeted therapies for early-stage MS.
“One of the big questions in MS is that the immune system attacks the myelin of the brain—the insulating layers around the axons,” said Daniel Reich, senior author of the study, a senior investigator at the National Institute of Neurological Disorders and Stroke, and adjunct professor of neurology at Yale School of Medicine. “It’s a highly inflammatory attack. During most of these attacks, which can be seen on MRI, the patient experiences nothing, but about ten percent of these attacks cause a loss of function.”
Reich and his team, including lead author Jing-Ping Lin, a research scientist at Yale School of Medicine, used a model called experimental autoimmune encephalomyelitis, where they triggered an MS-like response in marmosets by injecting a piece of the brain’s own myelin protein under the skin. The immune system then began to target the brain, mimicking the attacks seen in MS patients. Serial MRI scans captured the formation and growth of lesions over several weeks. The researchers found that early inflammation often began in small blood vessels and expanded outward in a pattern resembling a shock wave.
At the outer edge of new lesions—the most recent zone of inflammation—they discovered astrocytes, a type of non-neuronal brain cell, clustering together to form a protective barrier. These astrocytes showed high activity of a gene called SERPINE1, which has also been implicated in wound healing and blood clot regulation. “We weren’t surprised to find astrocytes involved in this way,” Reich said. “The analysis we’ve done suggests this might be an attempt to form a barrier and protect the rest of the brain.”
Crucially, the study showed that repair and destruction occur at the same time, often within the same lesion. This discovery, along with the identification of specific subtypes of brain cells and early MRI biomarkers, could guide new treatments that promote repair instead of just reducing inflammation.
Perhaps most surprisingly, molecular and cellular analyses revealed signs of inflammation even in brain regions that looked normal on MRI scans. “We weren’t surprised. It’s been known for a long time that areas that look normal on MRI aren’t actually normal,” Reich said. “It’s been termed in the MRI world ‘normal-appearing tissue.’ It looks fine at first glance, but it’s not.” These subtle abnormalities could represent either early stages of inflammation or downstream effects from damage elsewhere.
Beyond understanding MS, Reich hopes the study will inspire new ways to design human clinical trials. “We’re excited that the cells, molecules, and pathways we’ve uncovered give us new directions for developing therapeutics,” Reich said. “That’s true not just for MS, but potentially for virtually every brain disease—tumors, strokes, Alzheimer’s. There’s increasing evidence that these pathways are strongly overlapping.”
While Reich’s work has mostly used animal models, he’s now pivoting increasingly toward human applications. “We’ve been doing more and more experiments in patients,” Reich said. “The insights we’ve gotten from imaging in marmosets have given us a powerful lens for designing clinical trials. Even when trials don’t work, we can learn why, where, and how a drug failed. And that’s an insight too.”