PITTSBURGH — In a laboratory overlooking Fifth Ave in the heart of Oakland, unassuming machines as expensive as cars divide brain tissue from deceased patients with Alzheimer’s disease down to their individual cells.

The murmur of small talk among technicians mingles with the whir of these machines and the clink of glass instruments. It’s here where a collaboration between University of Pittsburgh and Carnegie Mellon researchers has led to the discovery of something new about the chromosomes of people with Alzheimer’s disease: They’re organized differently compared to healthy patients.

The researchers were able to discover this thanks to a multi-pronged approach that looked not only at the gene activity in single cells but also information from chromatin. Chromosomes — where all our genetic information is stored — are made up of chromatin, a combination of our DNA and proteins, tightly wound up. It’s like the difference between words on a page (the DNA sequence or genetic code), and how the paper is crumpled up (chromatin), which influences not only which words or genes are visible, but how close they are to each other.

This allowed them to look at Alzheimer’s disease from a different view and see things they couldn’t spot from reading DNA alone.

Not only is this the first time scientists have harnessed these multiple technologies to view the Alzheimer’s disease state, but the two Pittsburgh institutions did so together. The results were published July 23 in the journal Science.

“We’re not doing that enough in Pittsburgh,” said Jian Ma, senior author on the paper and a CMU professor and director of its Center for AI-Driven Biomedical Research. “You don’t really find two institutions so close to each other in the same city that are so complementary.”

The partnership began when Ma was searching journal articles for opportunities to apply a technology he’d recently built, along with scientist Zhijun Duan, at the University of Washington in Seattle. Ma’s lab, which constructs artificial intelligence and machine learning models to learn more about how cells operate, is especially interested in the brain and neurodegenerative diseases.

He came across a paper that interested him. It was written in 2019, and it spelled out the first time scientists were able to split tissue from people with Alzheimer’s disease into individual cells and sequence that RNA (ribonucleic acid) — a single-stranded temporary copy of our genetic code used to make proteins and learn about the genome.

Who was the paper’s first author? A Pitt assistant professor of neurobiology and Alzheimer’s researcher, Hansruedi Mathys.

“I realized he’s just a 15-minute walk from my lab at CMU,” Ma said. He emailed Mathys and asked if he wanted to collaborate on a project using his technology to learn more about the brain cells and chromatin in the Alzheimer’s disease state.

Mathys was quick to accept: “I was very enthusiastic about his proposal,” he said. “I immediately agreed to collaborate.”

The team then requested postmortem brain tissue from David Bennett, director of the Rush Alzheimer’s Disease Center at the Rush University Medical Center in Chicago. Bennett has been conducting a long-term cohort study called the Religious Orders Study and the Rush Memory and Aging Project, or ROSMAP, which now has more than 3,000 participants and has performed more than 1,500 autopsies.

He agreed to donate postmortem brain tissue samples of 20 individuals: 10 with Alzheimer’s disease and 10 age-matched individuals without, to act as a control group. They then got to work using Ma and Duan’s technology, called GAGE-Seq, a single cell technology that can probe both gene activity and chromosome structure from the same cell.

Ma and Mathys also used an imaging technology, Xenium, which measures the gene activity of brain cells, maps them spatially and informs researchers how certain brain cell types congregate in the layers of the cortex. This helps them understand how similar certain brain cells are in terms of their gene expression, which can hint at how the cells interact with each other.

It’s not common for studies to employ these multiple “layers” of imaging, from the tiny, microscopic environment of our genes inside a cell nucleus to how cells gather within layers of cortex: “We are really going across all levels of resolution,” said Mathys.

“They put everything together to capture Alzheimer’s disease from every perspective they can,” said Ghada Abdelhady, a postdoctoral student in professor Andreas Pfenning’s neuroscience lab at CMU, specializing in genomics and identifying therapeutic targets for Alzheimer’s disease. She’s used many of the same technologies Ma and Mathys did to study the disease — but never at the same time. Abdelhady was not involved in the research but worked in Mathys’ lab as a research fellow in 2021.

“This is definitely something no one has done before,” she said. “It’s a very cool paper, and it’s definitely adding to the field of Alzheimer’s disease. People should be excited about it.”

While understanding someone’s gene sequence yields plenty of information, the 3D space of the gene environment also influences the identity and health of a cell. This is also relevant in Alzheimer’s disease.

The researchers found that certain regions of Alzheimer’s patients’ genomes were less separate than in the healthy patients, and different “compartments” — that is, regions of the genome that are active, and those that are not — mingled more.

When they sequenced the genes from these samples, they found that this conformation of increasing mingling was associated with altered activity in genes linked to brain cell activity and the body’s metabolism and energy, such as how it handles insulin.

Notably, this finding cannot be seen by researchers just looking at the gene sequence or gene technology, which means, too, that Alzheimer’s disease cannot be understood with just one lens.

“You really have to have a very holistic and systematic view of things to study this disease,” said Ma.

The study’s findings don’t tell us what caused the different chromatin structure in Alzheimer’s disease patients or what this structural change might mean for Alzheimer’s symptoms, behaviors or progression of the disease.

Similar to how researchers are still working to understand the chicken-or-the-egg problem of the distinct amyloid-beta plagues and tau protein tangles in Alzheimer’s — are the plaques and tangles causing Alzheimer’s behaviors or arising as a result of something else? — the identified chromatin changes are simply a snapshot scientists can use to ask more questions, and, over time, get more answers.

“We do not really know what the root cause of Alzheimer's disease is, and I think it's really important to better understand the disease as a first step towards therapeutics,” said Mathys. “We need to understand what is going wrong in the disease, and what this study provides is one piece of this puzzle.”

The study also looked at the most extreme form of Alzheimer’s, not more mild forms, or disease in early stages of progression.

“I would really love to see different time points, more donors and different stages of Alzheimer’s to see whether you would come to the same conclusion,” Abdelhady said.

Ma and Mathys hope to continue collaborating, and that their teamwork encourages other scientists at Pitt and CMU to think about pairing up.

“These types of collaborations is where I think the science can be advanced more rapidly,” Ma said.

© 2026 the Pittsburgh Post-Gazette. Visit www.post-gazette.com. Distributed by Tribune Content Agency, LLC.