Cancer Cells Use Built-In DNA Anchors to Hijack Cell Division and Spread Oncogenes
Researchers discovered 'retention elements' — CpG-rich gene promoters that tether cancer's extrachromosomal DNA to chromosomes, ensuring oncogene survival across generations.
Summary
Scientists at Stanford identified a new class of human genomic elements called retention elements that help extrachromosomal DNA (ecDNA) — circular, oncogene-carrying DNA fragments common in aggressive cancers — hitch rides on chromosomes during cell division. Using a novel genome-scale screen called Retain-seq, they found thousands of CpG-rich gene promoters that tether ecDNA to mitotic chromosomes, ensuring it passes to daughter cells rather than being lost. These elements are co-amplified with oncogenes on ecDNA in human cancers and are focally hypomethylated. Critically, artificially methylating these elements abolished their retention activity and caused ecDNA loss, pointing toward a potential therapeutic vulnerability in ecDNA-driven cancers.
Detailed Summary
Extrachromosomal DNA (ecDNA) is a megabase-sized circular form of amplified oncogene found in roughly 14% of human cancers and associated with poor prognosis. Because ecDNA lacks centromeres, it cannot attach directly to the mitotic spindle, raising a fundamental question: how does it reliably pass to daughter cells across many generations? This study provides the first systematic answer to that four-decade-old puzzle.
The researchers developed Retain-seq, a genome-scale shotgun functional screen in which pools of bacterial plasmids carrying random human genomic inserts were transfected into cancer cell lines and serially passaged. Plasmid DNA retained across generations was isolated and sequenced to identify enriched inserts. As validation, the screen correctly identified the EBV oriP repeat sequence as a retention element only in EBNA1-expressing cells, mirroring known viral episome biology. Applied across ecDNA-positive (COLO320DM, GBM39) and ecDNA-negative (K562) lines, Retain-seq revealed thousands of human retention elements — predominantly CpG-rich gene promoters — that confer heritable episomal persistence to heterologous plasmids.
Live-cell imaging and IF–DNA-FISH confirmed that ecDNA co-segregates with chromosomes in 97–98% of mitotic events across multiple cancer cell lines harboring different oncogenes (EGFR, MYC, FGFR2). Retention elements act additively, with multiple elements on a single episome increasing retention probability. Hi-C chromatin conformation capture demonstrated that retention elements physically interact with mitotic chromosomes at regions bookmarked by transcription factors and chromatin proteins such as BRD4, essentially recapitulating promoter–enhancer-type interactions in trans during mitosis.
Analysis of ecDNA sequences from human cancer datasets showed that retention elements are significantly co-amplified with oncogenes on individual ecDNA molecules, and their number and distribution correlate with ecDNA size and structural complexity. Critically, retention elements are focally CpG-hypomethylated relative to surrounding genomic regions. Targeted cytosine methylation via dCas9-DNMT3A abolished retention activity and led to measurable ecDNA loss from cancer cells, establishing that methylation-sensitive chromatin interactions underlie this mechanism.
These findings reframe ecDNA as a selectively assembled genetic vehicle requiring three co-evolved components — an oncogene (fitness), an origin of replication (copying), and retention elements (segregation). The discovery that retention elements are methylation-sensitive opens a potential therapeutic angle: epigenetic reprogramming of ecDNA could destabilize oncogene amplification and sensitize tumors to treatment. Caveats include the use of heterologous plasmid models that may not fully recapitulate native megabase ecDNA behavior, and the need for in vivo validation of targeted methylation as a therapeutic strategy.
Key Findings
- Retain-seq screen identified thousands of CpG-rich gene promoters as human retention elements conferring episomal persistence across cell generations.
- ecDNA co-segregates with chromosomes in 97–98% of mitotic events across multiple cancer cell lines with distinct oncogene amplifications.
- Retention elements physically tether to mitotically bookmarked chromosomal regions via transcription factor and BRD4-mediated chromatin interactions.
- Retention elements are co-amplified with oncogenes on ecDNA in human cancers and are focally CpG-hypomethylated.
- Targeted cytosine methylation of retention elements abolishes their activity and causes measurable ecDNA loss from cancer cells.
Methodology
The study used Retain-seq, a novel genome-scale functional screen deploying pooled human genomic insert plasmid libraries in serial-passaged cancer cell lines, combined with live-cell imaging, IF–DNA-FISH, Hi-C chromatin conformation capture, and targeted dCas9-DNMT3A cytosine methylation experiments across multiple ecDNA-positive cancer cell lines and human cancer genomic datasets.
Study Limitations
Retain-seq uses heterologous bacterial plasmids (kilobase-scale) as episome surrogates, which may not fully replicate the behavior of native megabase-sized ecDNA in terms of chromatin structure or retention efficiency. In vivo validation of targeted methylation as a therapeutic approach in animal models and patient-derived tumors is still needed. The study does not fully resolve whether retention element activity is primarily promoter-sequence-specific or depends on associated transcription factor occupancy.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: