Circular DNA Throws Gene Regulation for a Loop

Circular tracks of DNA found inside cancer cells commonly contain large numbers of replicate oncogenes, and the uneven inheritance pattern of this DNA, located outside of chromosomes, enables tumor cells to rapidly adapt to selective pressures such as chemotherapy. However, the impact of this so-called extrachromosomal DNA (ecDNA) goes beyond genomic amplifications, regulatory flexibility, and accelerated tumor evolution.

Two new studies reveal how the topology of ecDNA affects gene regulation to promote cancer growth. A third report details how ecDNA can reintegrate itself into chromosomes, disrupting cancer-related genes. And a preprint article demonstrates the broad clinical relevance of the phenomenon by linking ecDNA-based amplifications to shorter survival times across a range of tumor types.

The findings collectively fill in key details about how oncogene-containing ecDNA boosts tumor aggressiveness and fuels resistance to therapies—insights that have helped jump-start at least one new company, Boundless Bio. The company, which launched in September 2019, is focused on “understanding and finding the vulnerabilities that are created by having this really different kind of DNA,” says cofounder Paul Mischel, MD, of the University of California, San Diego (UCSD).

With the growing appreciation of ecDNA's critical role in cancer development, Mischel adds, “I think we're at the dawning of a new field.” Late in January, researchers gathered in Berlin, Germany, for the first-ever meeting dedicated to circular DNA's role in disease and normal development.

Scientists first noticed ecDNA under the microscope more than a half-century ago, but it was only in the past 5 years or so that the research community began to appreciate the contribution of these loops to cancer cell behavior. A study from Mischel and UCSD's Vineet Bafna, PhD, also a Boundless Bio cofounder, demonstrated that ecDNA amplifications can rapidly increase oncogene copy number and accelerate intratumor heterogeneity, enabling cancers to adapt more effectively to variable environmental conditions (Nature 2017;543:122–5). Yet questions loomed about how the structure and spatial architecture of ecDNA contributed to pathogenesis.

In their recent study, Mischel and Bafna—along with their UCSD colleague Bing Ren, PhD, and Howard Chang, MD, PhD, another Boundless Bio cofounder from Stanford University School of Medicine in California—showed that the genomic packaging of ecDNA in several tumor types is unlike that of normal chromosomes (Nature 2019;575:699–703). Although ecDNA is wound into condensed complexes, it is not as compact as typical chromatin.

This structure makes the encoded genes more accessible for active transcription, which, along with the increased copy number, helps explain why oncogenes encoded on ecDNA are among the most highly expressed genes in tumors. The circularity of ecDNA also creates more long-distance interactions within the active chromatin that could further promote oncogene expression. “The shape is actually changing the regulation of the genes,” Mischel says.

In a separate report, another team discovered that noncoding enhancer elements are frequently coamplified alongside oncogenes on ecDNA (Cell 2019;179:1330–41). In glioblastomas with EGFR amplifications, the researchers showed, having the additional regulatory elements brought about a molecular rewiring of gene regulation, with the many enhancers combining forces to bolster oncogene expression.

“These amplifications are scooping up as many regulatory elements as they can get their hands on,” says Peter Scacheri, PhD, of Case Western Reserve University in Cleveland, OH, “and the vast majority of [the enhancers], if not all of them, are providing some proliferative advantage to the cancer cells.” Scacheri co-led the study with UCSD's Jeremy Rich, MD.

A third paper from Anton Henssen, MD, PhD, of Charité University Hospital Berlin in Germany, and Richard Koche, PhD, of Memorial Sloan Kettering Cancer Center in New York, NY, explored the impact of ecDNA slipping back into chromosomes in neuroblastomas (Nat Genet 2020;52:29–34). In these pediatric brain cancers, genomic reintegration of ecDNA into tumor-suppressor genes and nearby proto-oncogenes caused their inactivation and activation, respectively. As a result, patients with ecDNA-derived rearrangements had worse clinical outcomes than patients without such rearrangements.

“It's a striking clinical finding,” Henssen says. “These circular DNAs serve as substrates for further oncogenic genome remodeling.”

Additional support for the idea that ecDNA-based amplification events result in increased tumor aggressiveness and less favorable patient prognoses comes from a manuscript posted to the preprint server bioRxiv by Mischel, Bafna, and their fellow Boundless Bio cofounder Roel Verhaak, PhD, of The Jackson Laboratory for Genomic Medicine in Farmington, CT (bioRxiv 2019 Nov 28). By looking at whole-genome sequencing data from nearly 2,000 tumor samples, the researchers showed that amplified ecDNA occurred in at least 26% of cancers analyzed—and across a wide variety of histologies to boot.

What's more, patients whose tumors harbored ecDNA had worse outcomes than patients whose tumors had either noncircular amplifications or no amplifications at all. Yet perhaps the same genetic trick that gives these intractable cancers their selective edge will ultimately lead to the tumors' demise, Mischel says. At Boundless Bio, he and his colleagues now hope to therapeutically exploit any vulnerability created by hosting ecDNA in the cell.

It is “not an unreasonable premise,” says Anindya Dutta, MD, PhD, a molecular biologist at the University of Virginia School of Medicine in Charlottesville, who is unaffiliated with the company. “Clearing out the circles may be a possibility for therapy.” –Elie Dolgin