Science Spotlight

Oncohistones: A Vicious Cycle

From the Henikoff lab, Basic Science Division, and the Olson lab, Clinical Research Division

“The processes of life are like a drama, and I am studying the actors, not the plot. There are many actors, and it is their characters which make this drama. I seek to understand their habits, their peculiarities.”

— Albrecht Kossel, The New York Times interview

In his quest to understand the chemical composition of the cell nucleus, German biochemist Ludwig Karl Martin Leonhard Albrecht Kossel isolated and described for the first time the nucleic acids that form the DNA molecule. Kossel was duly awarded the Nobel Prize in Physiology or Medicine in 1910. Kossel’s research showed the components of the nucleus or “nucleins” consisted of a protein portion (nucleoproteins) and a nonprotein portion (nucleic acids). He later went on to show that much of the nucleoproteins are made up of highly basic proteins, which he named “histones”.

Eukaryotic DNA is tightly wrapped around histone proteins into a complex structure called chromatin. One subunit of chromatin consists of 147 bp of DNA tightly wrapped around a histone octamer. Histones come in different types and are categorized by number; histone one (H1) through four (H4). Histones not only provide structural support to chromosomes; they also organize the process of chromatin opening and closing to regulate DNA replication and gene expression via modifications such as methylation and acetylation deposited on their tails. For example, trimethylation of histone 3 lysine 27 (H3K27me3) marks closed chromatin and is a signal for gene silencing. Disruption of H3K27me3 deposition can lead to aberrant gene activation, which can drive tumorigenesis.

Lysine 27-to-methionine (H3K27M) mutations in histone variants H3.1 or H3.3 are common in diffuse midline glioma (DMGs), a lethal pediatric brain tumor. Dr. Jay Sarthy, a Damon Runyon/Sohn Pediatric Cancer Research Fellow in the Henikoff lab (Basic Sciences Division), studies brain tumors in children. Working with Kami Ahmad, PhD, a Drosophila geneticist, and Sarthy’s mentors Dr. Steven Henikoff and pediatric neuro-oncologist Dr. Jim Olson, Sarthy conducted a study to understand how H3.1 and H3.3 oncohistones predispose cells to tumorigenesis. The study was recently published in Elife.

Sarthy and colleagues took advantage of the genetically amenable fruit fly model to study the role of oncohistones in disrupting gene silencing. “We found that mutant histones inhibit chromatin silencing only if the histone is deposited onto DNA, and only if the cells are actively proliferating.” Said Sarthy. These results helped inform observations the researchers made in DMG patient-derived cells, where they identified robust H3K27M oncohistone deposition in chromatin. “These results explained the specific genomic patterns we observed in the tumors.” He added. 

A graphical abstract showing how nucleosomes  are assembled from new histones by replication-coupled and replication-independent pathways.
A graphical abstract: Nucleosomes (yellow) are assembled from new histones by replication-coupled (blue arrows) and replication-independent (green arrow) pathways. Deposition of H3.1 is exclusively replication coupled. Image from article.

Although H3K27M “oncohistones” comprise only ~5–15% of the total H3 histone within DMG cells, they inhibit the H3K27 methyltransferase Enhancer of Zeste Homologue-2 (EZH2) activity, dominantly reducing the global H3K27 methylation. Most DMG patients carry oncohistone K27M substitutions in the genes encoding the histone variant H3.3 and fewer are seen in the histone H3.1 gene. This is a surprising observation since the human genome contains 12 genes that encode for H3.1 but only two genes encoding for H3.3. Sarthy explained: “it has been mysterious why mutations in one certain histone are more severe than another histone despite the two histones having nearly identical amino acid sequences”. Sarthy et al., found that one histone is much more toxic to the silencing enzyme because it is deposited at higher levels near where this enzyme would normally act.

The authors used the careful regulation of cell cycle progression in the fly larval retina as a tool to dissect the contribution of the cell cycle to oncohistone activity. They found H3K27M inhibited H3K27 trimethylation exclusively in cycling cells during S phase, and that this inhibition was dependent on oncohistone incorporation into chromatin. The use of the developing fruit fly retina to probe the relationship of mutant histones to the cell cycle was a major breakthrough as explains Sarthy: “this was not possible to do using mammalian systems.” 

The authors then used CUT&RUN –a chromatin profiling technology developed in the Henikoff lab– in a panel of DMG patient-derived cell lines and found that H3.1- and H3.3-mutated cell lines exhibited distinct K27M distribution patterns. “The application of our CUT&RUN method to these gliomas was a major breakthrough as it allowed us to precisely map the locations of mutant histones in patient cells.” Said Sarthy. “Because the assay has an outstanding signal to noise ratio, we were able to identify low but important signal throughout the genome that had been missed by other assays.” He added. Importantly, the broader distribution of H3.1K27M throughout the genome was associated with lower activity of the silencing enzyme than the levels seen in H3.3K27M-DMG cells.

“This particular histone mutation is found in children with a certain type of brain cancer, but this mechanism for chromatin silencing plays a role in many other cancers” said Sarthy, who hopes his work will help clarify how these toxic histones cripple gene silencing during gliomagenesis. A therapeutic corollary to this work is how these oncohistones may predispose certain cancers to anticancer drugs that inhibit chromatin modifiers such as the EZH2 enzyme.

Sarthy JF, Meers MP, Janssens DH, Henikoff JG, Feldman H, Paddison PJ, Lockwood CM, Vitanza NA, Olson JM, Ahmad K, Henikoff S. 2020. Histone deposition pathways determine the chromatin landscapes of H3.1 and H3.3 K27M oncohistones. Elife  doi: 10.7554/eLife.61090

This work was funded by the Howard Hughes Medical Institute, the National Institutes of Health, the Alex’s Lemonade Stand Foundation for Childhood Cancer, and the Damon Runyon Cancer Research Foundation.

UW/Fred Hutch Cancer Consortium members James Olson, Patrick Paddison, and Steve Henikoff contributed to this work.