Penn/CHOP research helps to better understand cause of pediatric and treatment-related leukemias

Jun 16 2017

Certain pediatric leukemias share a common underlying cause with treatment-related secondary leukemias. Both diseases involve translocations in the KMT2A gene, in which a portion of this gene is swapped out with DNA from a "partner" gene on a separate chromosome. The resulting recombination causes an abnormal genetic rearrangement called a translocation that leads to leukemia, which is cancer of the blood cells. Patients with these types of leukemias tend to have poor outcomes.

As a step toward better understanding these forms of cancer, a joint effort by University of Pennsylvania and Children's Hospital of Philadelphia researchers has applied an innovative new genome sequencing technique to catalog the sites of DNA cleavage by the enzyme topoisomerase II, called TOP2.

The work was led by Brian D. Gregory, an associate professor in Penn's Department of Biology in the School of Arts & Sciences; Xiang Yu, a postdoctoral researcher in Gregory's lab; Carolyn A. Felix, the Joshua Kahan Endowed Chair in Pediatric Leukemia Research and an attending physician at CHOP and a professor of pediatrics in Penn's Perelman School of Medicine, and James W. Davenport, a research associate in the Felix lab. They reported their findings in the journal Genome Research.

"This tool opens new possibilities to better understand and eventually be able to manipulate TOP2 cutting to prevent the rearrangements that give rise to leukemias," said Felix, study co-leader.

The translocations that lead to infant leukemias and treatment-related secondary leukemias involve the action of the enzyme TOP2. This is because TOP2 plays a helping hand during DNA transcription and replication by cleaving the two DNA strands and easing the tangles and torsion that occurs during these processes, thus allowing them to untwist or pass through one another, then repairing the break. Translocations arise when the "repairs" result in mismatching and joining DNA from two locations in the genome.

Certain chemotherapeutic agents are TOP2 poisons and, while they can effectively kill cancer cells, they sometimes lead to abnormal DNA rejoining and cause translocations, which are the hallmark of the treatment-related secondary leukemias. Felix and colleagues have also previously reported that babies exposed in utero to TOP2 poisons, found in some foods in maternal diet, are at an increased risk of developing infant leukemia.

Felix's lab has identified several specific leukemia-causing translocations between the KMT2A gene and partner genes, but wanted a more efficient way to identify all of the points in the genome that are cleaved by TOP2 to be able to confirm the role of TOP2 cleavage in the DNA damage that is repaired incorrectly to create translocations.

That's where Gregory's group came in. Over several years they had developed a technique to perform genome-wide sequencing of locations where an enzyme makes a covalent bond to DNA, which is what TOP2 does while snipping strands of DNA.

"We designed a way to pull down the DNA bound to TOP2, then break that bond so only the DNA covalently attached to TOP2 is free to be sequenced to single base-pair precision," Gregory said. "This enabled us to map, for the first time, topoisomerase II cleavage on a genome-wide scale."

The team performed the analysis on a human leukemia cell line derived from a patient with leukemia, obtaining all of the cleavage sites, then repeated the technique on the same cells treated with either chemotherapy drugs or other TOP2 poisons found in food or the environment.

Examining the patterns they amassed, one of the key findings was that the cleavage events clustered in certain areas of the genome. They found hundreds of thousands of these clusters, mostly in gene introns, the non-coding portion of genic DNA, or in long non-coding RNAs, which play important roles in regulating gene expression. The clusters also tended to occur toward what is known as the 3-prime end of a gene, or the tail-end that is the last to be synthesized or transcribed.

"We think that TOP2 cleavage clusters in this region perhaps to decrease the torsion on the end of the DNA and permit elongation of the transcript," Yu said. "It lets stress out of DNA that is highly expressed."

The chromosome on which KMT2A is located has a higher density of cleavage clusters than other regions, and cleavage clusters are more prevalent in KMT2A's known partner genes and in other genes that are translocated in leukemia. "But even more surprising," said Felix, "is that genes involved translocations in many forms of cancer, not just in leukemia, are more likely to show TOP2 cleavage."

Several of the findings in the cells treated with chemotherapy agents or dietary or environmental TOP2 poisons matched with patterns already identified in regions of genes involved in translocations in treatment-related leukemias and in infant leukemias, supporting the importance of disturbances in TOP2 cleavage in causing these diseases.

The researchers compared their findings with functional genomic information from the National Human Genome Research Institute's ENCODE project to show that TOP2A cleavage clusters occurred in areas of the human genome that tended to be less variant, or conserved, across the human population.

"TOP2A cleavage regions seem to be quite conserved in humans," Yu said.

The current work elucidated many patterns of interest in TOP2A cleavage, but the researchers hope to move their work from the arena of basic science into findings that will benefit patients.

"One of our future questions is: Why do these TOP2 poisons both kill cancer cells and also lead to the formation of leukemia-causing translocations in normal blood cells?" Gregory said. "Perhaps it is because of different patterns of TOP2 cleavage in the normal and cancer cell populations."

Felix said that the findings open possibilities for new clinical approaches.

"The better we identify where cleavage occurs, the better we can understand how the drugs act and how the translocations happen," she said. "We could use that knowledge to design smarter drugs to target the TOP2 enzyme that don't have such a high risk of causing translocations or drugs to protect sequences in the genome from unwanted cutting."