Epigenetic heterogeneity helps cancer cells change their minds

From the Henikoff Lab, Basic Sciences Division and the Meshinchi Lab, Clinical Research Division

Individuals don’t evolve. That’s among the first and most fundamental lessons of Evolution 101. Natural selection may act on individuals, but evolution is a population-level phenomenon.  At its most basic, evolution results from genetic heterogeneity between individuals within a population. Selective environmental pressures serve to increase or decrease the fitness of individuals based on these heterogeneities, ultimately leading to more fit individuals becoming more prevalent in the population. A classic example of this process is the evolution of antibiotic resistance: treat a population of bacteria with antibiotics – a very strong selective pressure – and they’ll die; unless, that is, a small fraction of those bacteria contain a mutation conferring resistance. In this case, those few resistant bacteria will survive and reproduce, and the population will transition to one full of antibiotic-resistant individuals.

Given these basic evolutionary principles, it may be surprising to hear that evolution can, and does, occur within our bodies. To understand this perspective, think of yourself not as an individual, but a population of cells, each with its own genetic material, within which exist heterogeneities that selection can act upon. This insight has had profound impacts on our understanding of development and disease, perhaps most notably cancer. It is now well-understood that, as a tumor grows, its cells acquire distinct mutations. This heterogeneity is a major reason why so many cancer treatments eventually stop working: much like in the case of antibiotic resistance, while a treatment may kill almost all the cells in a tumor, if even a few have a mutation that confers resistance, the shrunken and seemingly defeated tumor can surge back, leaving doctors struggling to adjust treatment regimens to keep up with their evolving foe. In a new paper in Nature Genetics, Drs. Derek Janssen and Steve Henikoff from the Fred Hutch Basic Sciences Division, in collaboration with Dr. Soheil Meshinchi from the Fred Hutch Clinical Research Division, reveal a new manner of cancer heterogeneity.

While this story became one of heterogeneity, it did not begin as such. Rather, the group initially set out to understand the role of the epigenetic regulator KMT2A in acute leukemia. “Ten percent of acute leukemias harbor chromosomal translocations involving the KMT2A gene,” they wrote. “More than 80 translocation partners have been identified in KMT2A-rearranged [leukemias].” It is likely that each of these translocations has a different effect on the epigenetic landscape and, therefore, on gene expression, but identifying genomic KMT2A binding sites in these conditions has been technically difficult. To overcome this challenge, the group used a new method they developed, called Automated CUT&RUN, to efficiently identify KMT2A genomic binding sites in tumor samples containing several different translocations. They identified some shared targets, including many genes known to be frequently dysregulated in leukemia, as well as many differences based on which gene had coupled with KMT2A.

Next, the authors wanted to know how KMT2A fusion proteins were affecting chromatin marks at these sites. For this they developed a new, automated version of the CUT&Tag chromatin profiling technology (AutoCUT&Tag) and examined eight activating and repressive chromatin modifications in their tumor samples. From the vast data that this experiment generated, one pattern stood out: “In our collection of leukemia samples, we observed both [the active mark] H3K4me3 and [the repressive mark] H3K27me3 at some promoters that were called as KMT2A fusion protein targets,” they wrote. The presence of both activating and repressive marks – called bivalency – is difficult to interpret in genomic data. Does it indicate that these conflicting marks are simultaneously present at these promoters? Or does it mean that some cells in the tumor sample have one mark, and some cells have the other? The only way to distinguish between these possibilities is to assay these chromatin marks in individual cells. Therefore, the authors performed single-cell CUT&Tag and found that indeed different cells contain different chromatin marks, revealing a surprising heterogeneity within the tumor samples. The consequence of this heterogeneity, explained Dr. Janssens, is that “the mutant protein in these leukemias activates distinct gene networks in cells of the same leukemia.” This finding, he said, fits with a known quality of these tumors: “These leukemias are associated with the unusual capacity to switch from a lymphoid to a myeloid-like blood cell type… [the heterogeneity we observed] likely contributes to this unusual lineage plasticity.”

The basis of this cell type plasticity remains unclear. Is it guided by some underlying genetic heterogeneity? Or is it a purely epigenetic phenomenon, perhaps driven by environmental conditions? Either way, said Dr. Janssens, this flexibility “can make [these cancers] extremely challenging to treat,” and better understanding it is a key focus of his plans: “In the future we are interested in determining what causes cells that have activated one cancerous gene program to switch and use an alternative program. Is there a distinct group of cells that maintain this potential to switch programs?” He is also excited to explore how the methods developed for this work can be deployed to directly help patients: “we are planning to incorporate these methods into ongoing clinical trials to further investigate the diagnostic potential of chromatin profiling,” he said.

Finally, this work is also a story the type of collaborative progress that is at the core of the Fred Hutch/UW Cancer Consortium’s mission. While Dr. Henikoff’s lab provided the basic science expertise to precisely profile the chromatin landscape in these tumors, “Our collaboration with the Meshinchi lab [in the Clinical Research Division] here at Fred Hutch was essential to all phases of this study, from planning the chromatin profiling approach, to acquiring the primary patient samples and eventually interpretting our results and following up on clinically relevant observations,” Dr. Janssens said.

KMT2A function in leukemia
Oncogenic KMT2A fusion proteins can activate distinct gene networks to drive cell type switching in leukemia. Image Provided by Derek Janssens.

This work was supported by the National Institutes of Health, the Howard Hughes Medical Institute, the Chan-Zuckerberg Initiative, The Damon Runyon Cancer Research Foundation, and the Alex’s Lemonade Stand Foundation.

Fred Hutch/UW Cancer Consortium members Steve Henikoff and Soheil Meshinchi contributed to this work

Janssens DH, Meers MP, Wu SJ, Babaeva E, Meshinchi S, Sarthy JF, Ahmad K, Henikoff S. Automated CUT&Tag profiling of chromatin heterogeneity in mixed-lineage leukemia. Nat Genet. 2021 Oct 18. doi: 10.1038/s41588-021-00941-9.