A new study has revealed how a natural compound in our cells can reverse brain deficits in animal models of Alzheimer’s disease. The research shows that boosting levels of a molecule called NAD+ corrects errors in the way brain cells process genetic information, leading to improved memory and function. The paper detailing this mechanism was published in the journal Science Advances.
Alzheimer’s disease is a neurodegenerative condition characterized by the gradual loss of brain cells, resulting in cognitive decline. One of the main culprits in this process is a protein called tau. In healthy neurons, tau helps maintain their internal structure, acting much like railroad ties that keep tracks stable for transporting cellular materials. In Alzheimer’s disease, tau proteins can become defective and clump together, causing this internal transport system to collapse and leading to cell damage.
Previous research indicated that a metabolite known as nicotinamide adenine dinucleotide, or NAD+, could offer some protection in animal models of the disease. NAD+ is essential for energy production and maintaining the health of cells, particularly neurons. Its levels are known to decline with age and in several neurodegenerative diseases. While supplementing with precursors to NAD+ showed promise, the specific way it protected the brain remained largely unclear.
This new work was conducted by an international team led by Evandro Fei Fang at the University of Oslo in Norway, in collaboration with Oscar Junhong Luo from Jinan University in China and Joana M. Silva from the University of Minho in Portugal. The scientists wanted to understand the molecular steps behind NAD+’s protective effects. They focused on a fundamental cellular process called alternative RNA splicing, which has recently been recognized as a contributor to Alzheimer’s risk.
Alternative RNA splicing is a mechanism that allows a single gene to produce multiple distinct versions of a protein, known as isoforms. By selectively including or excluding different segments of the genetic blueprint, a cell can create proteins with varied functions from the same gene. When this finely tuned process is disrupted, as it appears to be in Alzheimer’s, it can lead to the production of faulty proteins and cellular dysfunction.
To investigate the link between NAD+, tau, and RNA splicing, the researchers used several models. They began with a species of roundworm, Caenorhabditis elegans, that was genetically modified to produce the toxic form of human tau protein. Using a fluorescent reporter system, they could visually track splicing errors inside the worms’ neurons. They observed that worms with toxic tau had significant disruptions in their RNA splicing.
The team then moved to a mouse model genetically engineered to develop tau-related brain pathology. These mice exhibit memory problems similar to those seen in human Alzheimer’s patients. The researchers provided some of these mice with nicotinamide riboside, a precursor that the body uses to make NAD+. After several months of treatment, they analyzed the animals’ brain tissue and behavior.
The analysis revealed that the mice with tau pathology had widespread errors in RNA splicing within their hippocampal cells, a brain region vital for memory. In the mice that received the NAD+ precursor, many of these splicing mistakes were corrected. This molecular repair was accompanied by functional improvements. The treated mice performed better on memory tests, suggesting that restoring proper RNA splicing helped recover cognitive function.
A specific protein, named EVA1C, emerged as a central player in this recovery. The researchers found that in the tau-affected mice, the benefits of the NAD+ precursor disappeared when the gene for EVA1C was turned off. This result indicated that EVA1C is necessary for NAD+ to exert its neuroprotective effects on RNA splicing. Consistent with these findings, an examination of brain tissue from people in the early stages of Alzheimer’s disease showed significantly reduced levels of EVA1C in their neurons.
To understand how EVA1C works at a molecular level, the team used an advanced artificial intelligence platform. The platform analyzed vast amounts of protein data to predict the three-dimensional structures of different EVA1C isoforms and how they would interact with other proteins. This computational analysis predicted that boosting NAD+ promotes a specific version of the EVA1C protein.
This particular isoform of EVA1C was predicted to bind with high affinity to other essential proteins, including one called HSP70. HSP70 functions as a “molecular chaperone,” helping other proteins fold correctly and clearing away damaged or aggregated proteins. By influencing which isoform of EVA1C is produced, NAD+ appears to enhance the cell’s protein management system, which is critically impaired in Alzheimer’s disease. The research connects cellular metabolism, through NAD+, with the genetic processing of RNA splicing and the physical management of proteins.
The study is not without its limitations. The findings are based on animal models, and the complex pathology of Alzheimer’s in humans may involve additional factors. Future research will need to confirm these mechanisms in human cells, perhaps using brain organoids grown in the lab, and to further map the precise chain of events connecting NAD+ to the cell’s splicing machinery.
Still, the work opens new avenues for therapeutic development. By identifying the connection between NAD+ and EVA1C-regulated RNA splicing, the study provides a clearer target for potential treatments. “We propose that maintaining NAD⁺ levels could help preserve neuronal identity and delay cognitive decline, paving the way for combination treatments to enhance RNA splicing,” says Ruixue Ai, a first author of the study. This research deepens the understanding of how NAD+ supports brain health and may inform the design of more effective strategies to combat neurodegenerative diseases.
The study, “NAD+ reverses Alzheimer’s neurological deficits via regulating differential alternative RNA splicing of EVA1C,” was authored by Ruixue Ai, Lipeng Mao, Xurui Jin, Carlos Campos-Marques, Shi-qi Zhang, Junping Pan, Maria Jose Lagartos-Donate, Shu-Qin Cao, Beatriz Barros-Santos, Rita Nóbrega-Martins, Filippos Katsaitis, Guang Yang, Chenglong Xie, Xiongbin Kang, Pingjie Wang, Manuele Novello, Yang Hu, Linda Hildegard Bergersen, Jon Storm-Mathisen, Hidehito Kuroyanagi, Beatriz Escobar-Doncel, Noemí Villaseca González, Farrukh Abbas Chaudhry, Zeyuan Wang, Qiang Zhang, Guang Lu, Ioannis Sotiropoulos, Zhangming Niu, Guobing Chen, Rajeevkumar Raveendran Nair, Joana Margarida Silva, Oscar Junhong Luo, and Evandro Fei Fang.
