All cells in the body contain the same DNA, but different cell types express different genes; skin cells express genes for the skin, liver cells express liver genes, and so on. This coordination is crucial to help cells differentiate into their assigned roles, but a new study from researchers at the University of Chicago shows how cells can randomly "shake up" regions of the genome to express genes normally reserved for other cell types.
The study, published this week in Nature, suggests that randomness or variability in the way DNA is packaged can create a kind of "epigenetic noise," enabling cells to take on the identify of different cell types. This flexibility plays an important role in tissue repair and the immune system but can also be exploited for the development of tumors.
"We believe that this capacity to change a cell's identity is underappreciated, and we wanted to investigate the mechanisms underlying how cells are able to change their fates," said Andrew Koh, PhD, Assistant Professor of Pathology at UChicago and senior author of the new study.
Open, noisy, and jiggly
Koh and recent PhD graduate Noah Gamble, the study's lead author, worked with an incredibly resourceful group of cells called medullary thymic epithelial cells (mTECs). These cells are found in the thymus, a small, specialized organ of the immune system located just above the heart. They are one of the few cell types in the body that can express a wide variety of genes and alter their identity to mirror cell types from other tissues.
mTECs play an important role in training the immune system to prevent autoimmunity. They present proteins that are normally expressed only in specialized tissues and organs to T cells developing in the thymus. Then, the T cells that react too strongly to molecules from the body's own cells are purged so they don't later trigger an autoimmune response.
The capability to express almost any gene and alter their identities made mTECs a great candidate for studying how cells can change their fates. "Each individual cell does not express the entire genome. Instead, they express only a unique subset of the tissue-specific genes at any given snapshot," Koh said. "There's a great deal of heterogeneity, so we thought that it was really important to look cell-by-cell to uncover the mechanisms that allow the activation of each subset of tissue-specific genes."
Since such heterogeneity is important, Gamble used a series of single cell sequencing techniques to study gene expression and chromatin structure in individual mTECs, instead of using traditional bulk sequencing tools that average the results over thousands of cells.
Chromatin is the complex of DNA and proteins in the nucleus that packages long stretches of DNA into more compact structures. When chromatin is more loosely packed, or open, genes are more poised to be activated than if it's tightly coiled.
When Gamble analyzed the data, he did not find links between peak levels of chromatin accessibility and the expression of tissue-specific genes. Instead, he saw a lot of accessibility "noise" that gave cells the potential to activate genes solely expressed in other specialized tissues. This "ectopic expression" in turn helped train T cells to discriminate between self and non-self.
"Chromatin is usually tightly regulated to sequester regions that encode other cell fates and focus accessibility for regions pertinent for the established cell identity," Gamble said. "In our context, we found the genomic regions that should be tightly packed were more labile or 'jiggly', allowing more opportunities for factors to access and activate genes specific to different cell types."
Avoiding the 'guardian of the genome'
The team then tried to understand how this "chromatin noise" is amplified in cells. They found that the activity of the tumor suppressor protein p53, known as "the guardian of the genome," is repressed by mTECs prior to their genome becoming noisy. p53 is usually activated when DNA is damaged and can trigger cell death or stop tumor cell growth. So, it made sense to Gamble and Koh that it would be implicated in a process where epithelial cells promiscuously express genes dedicated to other tissues and organs.
When the researchers genetically engineered p53 activity to be enhanced in mTECs, their chromatin became more stable, epigenetic noise was turned down, and the cells could no longer activate tissue-specific genes. This ultimately resulted in the escape of self-reactive T cells from the thymus to cause multi-organ autoimmune disease.
"This suggests that thymic epithelial cells adopt deviant states that should normally trigger p53 activation and cell death," Koh said. "But because p53 is downregulated, the cells survive and facilitate this ectopic gene expression to promote the self/non-self discrimination."
It's a fascinating idea to think that cells are programmed to loosen their grip on genes to give them more freedom to get creative and solve problems like preventing T cells from attacking their own tissues. Koh and Gamble extended their studies and found that epigenetic noise also allows lung cancer to sample more of the genome once p53 is deleted. This activates programs specific to other tissues to develop into more aggressive, malignant states. They hope to continue studying whether other cancer types exploit similar mechanisms for tumorigenesis.
The team also wants to see if epigenetic noise is amplified for wound healing and tissue repair, and whether or not it can be leveraged to reprogram cells to alternate phenotypes for various clinical contexts, including cancer immunotherapy and treating autoimmunity.
"It makes sense that to empower an immune system that uses a random process to recognize virtually any entity in the universe, thymic epithelial cells amplify random noise in the genome to ensure the immune system is focused on pathogens and cancers and not its own tissues. It's fighting fire with fire," Gamble said. "The moral of the story is that sometimes the random background noise can be just as important as the signal."
The study, "Thymic epithelial cells amplify epigenetic noise to promote immune tolerance," was supported by the National Institute of Health, the National Science Foundation, and the Chan Zuckerberg Biohub. Additional authors include Jason A. Caldwell, Joshua McKeever, Caroline Kaiser, Alexandra Bradu, Peyton J. Dooley, Narutoshi Hibino, and Aaron R. Dinner from UChicago; and Sandy Klemm, William J. Greenleaf from Stanford University.
