Researchers at UC San Diego have utilized artificial intelligence to uncover a previously unknown function of a gene linked to Alzheimer’s disease, offering new insights into its causes and identifying a potential therapeutic candidate.
Researchers at the University of California San Diego have leveraged artificial intelligence to discover a new cause of Alzheimer’s disease and pinpoint a potential treatment, offering fresh hope in the battle against this devastating condition.
In a new study published in the journal Cell, the research team revealed that phosphoglycerate dehydrogenase (PHGDH), a gene recently identified as a biomarker for Alzheimer’s, is not just associated with the disease but actually contributes to its development due to a previously undiscovered secondary function.
“Unfortunately, treatment options for Alzheimer’s disease are very limited. And treatment responses are not outstanding at this moment,” senior author Sheng Zhong, a professor in the Shu Chien-Gene Lay Department of Bioengineering at UC San Diego’s Jacobs School of Engineering, said in a news release
Alzheimer’s disease is the most common cause of dementia, affecting about one in nine people aged 65 and older. While some genetic mutations are known to directly lead to Alzheimer’s, these account for only a small percentage of cases. For the majority of patients without these genetic mutations, known as “spontaneous” Alzheimer’s, the causes have remained largely unknown, making this new discovery particularly significant.
The research team previously found that the expression levels of the PHGDH gene were directly correlated with the severity of Alzheimer’s.
Their latest study confirmed that PHGDH is indeed a causal gene for spontaneous Alzheimer’s disease. Through AI, they discovered that PHGDH disrupts the gene expression process in the brain, a revelation that clarifies its role in disease progression.
“It really demanded modern AI to formulate the three-dimensional structure very precisely to make this discovery,” Zhong added.
PHGDH’s known role involves producing an enzyme essential for the amino acid serine. However, the research team discovered that the protein created by PHGDH also has a structural similarity to known DNA-binding domains, allowing it to disrupt gene regulation in brain cells — key in the early stages of Alzheimer’s.
To tackle this, the researchers explored potential inhibitors of PHGDH’s newly discovered function. Among these, a small molecule called NCT-503 showed promise. Unlike other candidates, NCT-503 did not affect the enzyme’s primary role in serine production and could cross the blood-brain barrier.
Testing NCT-503 in two mouse models of Alzheimer’s, the team noted significant improvements in memory and anxiety, symptoms that closely mirror the human experience of the disease.
Now there is a therapeutic candidate with demonstrated efficacy that has the potential of being further developed into clinical tests,” aded Zhong. “There may be entirely new classes of small molecules that can potentially be leveraged for development into future therapeutics.”
While acknowledging limitations — such as the lack of perfect animal models for spontaneous Alzheimer’s — these findings are seen as a critical step forward. The next phase will focus on optimizing the compound and pursuing FDA IND-enabling studies.