Scientists at Rice University have produced the first complete molecular map of an Alzheimer’s brain without using dyes or labels, and what they found challenges the dominant theory of the disease. Chemical disruptions spread unevenly across the entire brain, far beyond the amyloid plaques that most treatments target.
The Mapping Technique
The team combined hyperspectral Raman imaging with machine learning to analyze brain tissue. Raman spectroscopy works by shining a laser at tissue and measuring how molecules scatter the light. Different molecules produce different spectral signatures, creating a chemical fingerprint.
The “hyperspectral” part means collecting these signatures across many wavelengths simultaneously, building a detailed picture of which molecules are present and where. Because the technique doesn’t require dyes or stains, it captures the brain’s natural chemistry without artifacts.
Machine learning algorithms then processed this data in two phases. First, unsupervised learning identified natural patterns in the chemical signals without any preset assumptions about what to look for. Then supervised models learned to distinguish between Alzheimer’s and healthy brain tissue.
What They Found
The hippocampus and cortex, the memory regions hit hardest by Alzheimer’s, showed the most dramatic chemical changes. But the disruption wasn’t limited to areas with visible amyloid plaques.
Cholesterol metabolism was altered throughout the brain. Glycogen levels, which reflect how brain cells store and use energy, were abnormal. These metabolic shifts appeared in regions far from any visible protein deposits.
“This uneven pattern helps explain why symptoms appear gradually and why treatments focusing on only one problem have had limited success,” said doctoral student Ziyang Wang, the study’s first author.
Why This Matters for Treatment
The amyloid hypothesis has dominated Alzheimer’s research for decades. It holds that the disease starts when amyloid-beta proteins clump into plaques, which then trigger tau tangles and neurodegeneration. Most major drug development has focused on clearing these plaques.
But drugs targeting amyloid have largely disappointed in clinical trials. Lecanemab and donanemab, approved in the past few years, do clear plaques and modestly slow decline, but they don’t stop the disease. Many researchers have questioned whether plaques are the cause or merely a symptom.
This study adds weight to an alternative view: Alzheimer’s involves broader disruptions in how the brain processes energy and maintains its structure. The protein buildups may be part of this larger metabolic breakdown, not its root cause.
“These findings support the idea that Alzheimer’s involves broader disruptions in brain structure and energy balance, not only protein buildup,” said Shengxi Huang, the Rice associate professor who led the research.
Limitations
This was a proof-of-concept study using tissue from a single Alzheimer’s brain compared to healthy controls. The technique needs validation across many more samples to confirm these patterns are consistent.
Postmortem tissue also captures end-stage disease. The researchers can’t yet say whether these metabolic changes appear early enough to serve as diagnostic markers or treatment targets.
The study was published in ACS Applied Materials and Interfaces and funded by the National Science Foundation, National Institutes of Health, and Welch Foundation.
The Bottom Line
The research doesn’t disprove the amyloid hypothesis, but it suggests the story is more complicated. If Alzheimer’s is fundamentally a metabolic disease, treatments may need to address energy processing and cellular maintenance alongside, or instead of, protein clearance.
That’s a significant shift in how we think about a disease that affects over 55 million people worldwide. And it came from applying machine learning to an imaging technique that’s been around for decades, finding patterns that human analysis missed.