AI news ·
AI Uncovers Hidden Alzheimer's Trigger and Points to Promising New Treatment
AI uncovers how the enzyme PHGDH may trigger Alzheimer's by affecting gene activity in brain cells. A molecule called NCT-503 shows promise in improving memory in mice.

AI Identifies Suspected Trigger of Alzheimer's and Potential Treatment
Recent research from the University of California, San Diego (UC San Diego) has uncovered a significant link between a gene and Alzheimer's disease, using artificial intelligence (AI) to explore new treatment possibilities. This finding sheds light on what might trigger the disease and how it could be slowed down.
The Role of PHGDH in Alzheimer's
The focus is on an enzyme called phosphoglycerate dehydrogenase (PHGDH) and the gene that encodes it. Previous studies showed that this gene is more active in individuals with rapidly progressing Alzheimer's, but the cause of this connection was unclear.
AI was used to model the 3D structure of PHGDH more precisely, revealing a hidden function: the enzyme appears to switch other genes on and off. Specifically, PHGDH interacts with two genes inside astrocytes—brain cells that regulate inflammation and clear waste. This interference may be a tipping point in triggering Alzheimer's.
Targeting PHGDH for Treatment
Researchers then sought a way to inhibit the gene-regulating function of PHGDH without blocking its essential enzymatic activity. They identified a molecule called NCT-503 that fits this purpose. AI modeling helped show how NCT-503 binds to a specific pocket in PHGDH, blocking its gene-switching ability.
Animal studies demonstrated that treatment with NCT-503 improved memory and reduced anxiety in mouse models of Alzheimer's. This suggests the molecule could be the starting point for a new class of Alzheimer's therapies.
Advantages and Next Steps
- NCT-503 can cross the blood-brain barrier, reaching neurons and astrocytes effectively.
- The drug candidate could potentially be administered orally, simplifying treatment.
- Although promising, the research is in early stages; further development and clinical trials are needed.
Current treatment options for Alzheimer's remain limited, and responses are often not satisfactory. This new approach offers a fresh avenue by targeting gene regulation within brain cells rather than just symptoms.
For healthcare and research professionals interested in AI applications in medicine, this study highlights how AI tools can identify molecular targets and accelerate drug discovery. Exploring AI courses may provide deeper insights into these emerging techniques.
The full research has been published in Cell.