TOKYO —

Researchers at Carnegie Mellon University’s School of Computer Science, the University of Pittsburgh School of Medicine, and the University of Washington have made a significant breakthrough in understanding Alzheimer’s disease, identifying a previously under-explored facet of the devastating neurological condition that could pave the way for new therapeutic strategies. Their groundbreaking study, published in the prestigious journal Science, reveals that the three-dimensional architecture of the genome within specific brain cells exhibits distinct alterations in individuals affected by Alzheimer’s. This discovery connects changes in genome folding to shifts in gene activity and the structural organization of brain tissue, offering a more holistic view of the disease’s molecular underpinnings.

The collaborative effort, involving scientists from Carnegie Mellon’s Ray and Stephanie Lane Computational Biology Department, the University of Pittsburgh’s Department of Neurobiology, and other contributing institutions, leveraged cutting-edge single-cell technology, sophisticated spatial mapping of brain tissue, and a newly developed deep learning model to construct this intricate picture.

Moving Beyond Established Hallmarks

For decades, research into Alzheimer’s disease has primarily focused on two key pathological hallmarks: the accumulation of amyloid-beta plaques and the formation of tau tangles. While these remain critical areas of study, the new findings suggest that a deeper, more fundamental layer of cellular organization—the three-dimensional structure of DNA within the cell nucleus—plays a crucial role and is significantly impacted by the disease.

"Alzheimer’s disease cannot be understood one layer at a time," stated Jian Ma, the Ray and Stephanie Lane Professor of Computational Biology, who spearheaded and oversaw the study. "The genome’s 3D structure is a fundamental regulatory layer that helps to connect DNA sequence to gene activity. By integrating genome folding, cell state, and tissue context, we can move beyond cataloging disease-associated changes toward understanding how they fit together and which mechanisms to test next."

DNA, the blueprint of life, is not a passive, linear molecule within the cell. Instead, it undergoes complex folding into a three-dimensional architecture. This intricate arrangement is critical for regulating gene expression, determining which genes are accessible for activation and which remain silenced. Consequently, alterations in this physical organization can profoundly influence cellular function, leading to the dysregulation observed in disease states.

The research team meticulously examined postmortem samples of the prefrontal cortex, a region of the brain crucial for higher-level cognitive functions such as planning, decision-making, and memory, which is often affected by Alzheimer’s. These tissue samples were obtained from individuals who had participated in long-term dementia studies and subsequently donated their brains for scientific research, allowing for a direct comparison between individuals with and without Alzheimer’s disease.

A Multi-Omic Approach to Unraveling Complexity

To achieve an unprecedented level of detail, the researchers employed GAGE-seq, a sophisticated technique capable of simultaneously measuring both gene expression levels and three-dimensional genome contacts within individual cells. This single-cell resolution data was then integrated with spatial transcriptomic maps, which preserve crucial information about the precise location of gene activity within intact brain tissue.

By synergistically combining these diverse datasets, the scientists were able to establish direct links between the physical organization of the genome and gene regulation. Crucially, this integrated approach also allowed them to pinpoint where Alzheimer’s-related molecular and cellular changes manifested within the surrounding tissue microenvironment.

Illuminating a New Dimension of Alzheimer’s Pathology

Hansruedi Mathys, an assistant professor of neurobiology at the University of Pittsburgh’s Department of Neurobiology and the director of the Pitt arm of the study, emphasized the study’s significance. "Our study represents a major advance in understanding what goes wrong in Alzheimer’s disease," he commented. "We know the classic hallmarks of Alzheimer’s disease—accumulation of amyloid-beta plaques and tau tangles—but our results establish higher-order chromatin alterations as a component of the molecular pathology associated with the disease, which currently affects seven million Americans, a number that continues to grow."

The implications of this finding are substantial. While amyloid-beta plaques and tau tangles have long been considered the primary culprits, the new research posits that changes in chromatin—the complex of DNA and associated proteins that package the genome—should now be recognized as a fundamental aspect of the disease’s molecular landscape. This broadens the scope of potential therapeutic targets significantly.

The Role of Artificial Intelligence in Deciphering Genome Structure

A pivotal component of this research was the application of Hicformer, an advanced artificial intelligence model specifically designed to explore the intricate relationship between genome structure and cellular behavior. This AI model integrates diverse data streams, including DNA sequence information, broad patterns of genome folding, and detailed maps of physical DNA contacts, to predict gene activity across various cell types.

Xinyue Lu, a doctoral student in Computational Biology and a co-leader of the research, described Hicformer as an invaluable computational testbed. "Measuring gene activity and genome folding in the same cell allows us to directly connect chromosome structure with disease-related gene programs," she explained. "Across several kinds of brain cells, this paired view revealed a consistent signature of 3D genome reorganization in Alzheimer’s disease and helped us prioritize regulatory regions for future mechanistic and therapeutic investigation."

Yang Zhang, a project scientist in the Computational Biology Department and another co-leader of the study, further elaborated on the AI’s contribution. "Across several kinds of brain cells, this paired view revealed a consistent signature of 3D genome reorganization in Alzheimer’s disease and helped us prioritize regulatory regions for future mechanistic and therapeutic investigation."

Characterizing Structural Deviations in Alzheimer’s Brain Cells

The research identified several consistent and notable differences in the genome architecture of brain cells from individuals with Alzheimer’s disease. Typically, large segments of the genome are organized into distinct active and inactive regions, referred to as compartments. However, in Alzheimer’s-affected cells, these boundaries appeared to be less sharply defined, a phenomenon the researchers characterized as "increased compartment mingling."

Furthermore, several types of brain cells exhibited a reduction in interactions between nearby genomic regions and an increase in contacts between more distant DNA segments. The study observed a correlation between greater compartment mingling and lower overall levels of gene activity.

The team also noted weakened interactions between genes and their proximal regulatory elements, which are crucial for controlling gene expression. Concurrently, some contacts across intermediate distances within the genome became more pronounced.

These observed structural alterations were associated with a decrease in the activity of neuronal and synaptic programs, essential for brain cell communication and function. Additionally, changes were detected in metabolic pathways and cellular stress responses. Notably, the researchers also uncovered links to senescence-related programs in microglia, the brain’s resident immune cells, which play a vital role in maintaining brain health and responding to injury and disease.

Implications for Future Therapeutic Interventions

When these molecular changes were mapped onto intact brain tissue, the researchers discovered that the observed genome reorganization was not only linked to altered gene activity but also to discernible differences in the spatial arrangement of brain cells within the tissue.

The findings firmly establish three-dimensional genome organization as a critical, yet previously underappreciated, layer of Alzheimer’s disease biology. This work provides a robust framework for scientists to systematically investigate which specific structural changes in genome architecture might directly contribute to the progression of the disease.

The study’s revelations open new avenues for future research. Scientists can now delve deeper into whether particular genomic structural anomalies are drivers of Alzheimer’s progression. Moreover, the identification of affected regulatory regions presents promising opportunities for the development of novel therapeutic interventions targeting these specific genomic sites.

The research was generously supported by grants from the National Institutes of Health, underscoring the federal government’s commitment to advancing Alzheimer’s research. Additional authors from Carnegie Mellon University included doctoral students Shahul Alam and Shike Wang, and postdoctoral research associate Junjie Tang. Contributing authors from the University of Pittsburgh included doctoral students Alexander K. Kunisky and Jude Baroudi, post-baccalaureate research fellows Sahar and Sahel Ghorbanikalateh, and visiting scholar Shihan Wang. The collaborative network extended to researchers from the Broad Institute of MIT and Harvard, the University of California, Los Angeles, the University of Washington, and the Rush Alzheimer’s Disease Center, highlighting the broad and interdisciplinary nature of this significant scientific endeavor.