It is still not fully understood how, despite having the same set of genes, cells turn into neurons, bones, skin, heart, or roughly 200 other kinds of cells, and then exhibit stable cellular behavior over a human life span which can last for more than a century – or why aging degrades this process.
Using advanced imaging, sensing, and analysis techniques, researchers discovered that transcriptional memory is encoded in 3D structural elements called nanoscale chromatin packing domains, which function as a computational device.
Researchers can now begin to develop strategies to rationally design cellular memory states to regulate cell behavior for applications in health, disease and aging.
Professor Vadim Backman; Professor Igal Szleifer; Luay Almassalha, Department of Gastroenterology and Hepatology, Feinberg School of Medicine
It’s a marvel how the human genome — two meters of DNA — fits into the tiny nucleus of a cell while orchestrating the complex processes that sustain life. A new study from Northwestern Engineering’s Center for Physical Genomics and Engineering (CPGE) reveals the genome is far more dynamic than previously thought, operating less like a static instruction manual and more like a sophisticated, physically encoded computer that integrates signals to form 3D structures essential for cell function. And like a computer, scientists may soon be able to ‘reprogram’ these cellular memories to create radically new strategies for human health, disease and longevity.
Using advanced imaging and analysis techniques developed at CPGE at the McCormick School of Engineering, the research shows how these genome structures can store and create transcriptional memories, allowing cells to perform their roles effectively across an entire lifetime. These transcriptional memories not only regulate cellular activity but also maintain stability across different cell types, ensuring tissues work as cohesive units.
“Instead of the genome acting as a storage bin for information, we found that it operates more like a dynamic computer processor that integrates the intended signals with what the available space can allow,” the McCormick School of Engineering’s Vadim Backman said.
Scientists have long viewed chromatin — the genetic material that makes up chromosomes — as existing in two opposing states: heterochromatin, which is compacted and inactive; and euchromatin, which is open and active. This study challenges that model, showing that heterochromatin and euchromatin work together as unified computational units. This shows why inhibiting heterochromatin doesn’t turn on every gene – it’s needed to create the right space for the activation of the genes involved in making proteins.
This coupling allows cells to optimize the limited space within the nucleus while still carrying out essential processes, a balancing act crucial for tissue development and stability. In effect, heterochromatin is crucial for turning on the system in a well-organized manner.
“By closing parts of the genome that are not generally transcribed, it can create the right configuration to support the segments that get transcribed,” Backman said. “Each cell in a tissue can essentially trust that their neighbor will behave predictably, and each cell will respond to signals in a coherent manner.”
Backman is the Sachs Family Professor of Biomedical Engineering and Medicine at Northwestern Engineering and the Feinberg School of Medicine. The work is a three-way collaboration between the Backman lab and Igal Szleifer, Christina Enroth-Cugell Professor of Biomedical Engineering at the McCormick School of Engineering; and Luay Almassalha, of the Department of Gastroenterology and Hepatology within the Feinberg School of Medicine. The team revealed their findings in the paper “Chromatin Conformation, Gene Transcription, and Nucleosome Remodeling as an Emergent System,” published January 10 in Science Advances.
This insight into the genome has especially significant implications for aging and disease. The study highlights how cells use transcriptional memories to establish predictable, stable behaviors within tissues. As we age, these memories degrade, and they could become replaced by ‘mistaken’ memories. Across every tissue in the human body, this process is associated with diseases of aging like cancer, Alzheimer’s, atherosclerosis, cognitive decline, and muscle loss. The discovery of how transcriptional memories are encoded potentially explains how and where to reverse these processes and could lead to entirely new kinds of therapies targeting cancer, enhancing tissue regeneration, and promoting longevity.
The research also emphasizes the importance of advanced tools and multidisciplinary collaboration. This research was made possible by cutting-edge high-resolution imaging and modeling techniques developed at Northwestern, which provided the resolution necessary to observe nanoscale structures in the genome.
These tools allowed the team – including experts in imaging, mathematical modeling, biophysics, molecular biology, and medicine – to move beyond outdated models and uncover how chromatin states cooperate to create the optimal conditions for gene transcription.
“The model reproduces the images that are observed experimentally, while suggesting a novel theoretical interpretation of chromatin structure. In turn, these results served as the inspiration for a new global experimental and theoretical interpretation of chromatin organization,” Szleifer said.
“It was only with the development of these new imaging and modeling tools that we could understand the limits of prior studies,” Backman added.
The next steps of the research involve translating these discoveries into practical applications for health and disease, and exploring the development of powerful new computing techniques inspired by genomic geometric computation mechanisms. The findings could also inspire efforts in synthetic biology, potentially enabling the design of artificial organisms with custom-built transcriptional memories.
Looking further, the findings also raise intriguing questions about evolution and biology across species. The researchers plan to investigate whether similar physically encoded computations exist in other multicellular organisms, such as plants and fungi, and explore the evolutionary implications.
“This system would likely have an evolutionary benefit to create complex structures, so it’s worth examining if other multicellular species like plants, fungus, and animals use similar physically encoded computations,” Almassalha said.
“This indicates a fundamentally new approach to understanding how our genome is organized,” Backman said. “When viewed as cooperative properties necessary to generate a coherent configuration for optimal transcription to occur, this immediately generates the capacity for chromatin domains to act as a computational structure.”