HMN 2026: How Completing DNA replication triggers genomic instability in bacteria

If you are anything like us, whenever you plan a journey, you spend a remarkable amount of time thinking about the start and the middle. Is everything packed? What time should we leave? Will there be traffic? Is there a faster route? We rarely think about the arrival; we just—arrive.

For decades, molecular biologists thought about DNA replication in a similar way. Entire careers have been built on understanding how cells start to duplicate their chromosomes and how the molecular machines that copy DNA race along the chromosome. But the final step, when two of these machines meet, was largely treated as a nonevent. The proteins disassemble. The strands are stitched together. Job done.

But what if arrival is harder than we thought?

If we scale the gap between the two strands of DNA up to the gauge of a railroad track, the Escherichia coli chromosome becomes a single circular line running from London to Edinburgh and back. At realistic replication speeds, two trains leaving Edinburgh in opposite directions would meet head-on in London after roughly 45 minutes. But what actually happens at that meeting point? Do the two trains speed into one another, slow down for a gentle approach or never quite meet?

Our latest study, published in Nucleic Acids Research, shows that the answer is stranger than any of those options. In bacteria, the very act of two replication machineries meeting is an inevitable event in every cell cycle—and we show that it is inherently destabilizing.

To test this, we engineered E. coli strains in which we could switch fork fusion events on and off at specific chromosomal locations. At the same locations, we placed a genetic recombination reporter cassette, essentially a sensor that detects local DNA rearrangements. We then asked a simple question: Do we see more genome rearrangements when replication forks fuse than when they do not?

The answer was clear. Wherever replication forks met, recombination increased significantly, and when we moved the meeting point elsewhere, the elevated signal moved with it. The completion of DNA replication directly generated genome instability. Crucially, this was observed in healthy bacteria with all of their machinery intact. The final handshake of two replication machineries is itself risky.

Cells, of course, are not helpless. Bacteria have evolved an arsenal of proteins that defuse problematic DNA structures. Among the key players are an enzyme called RecG helicase and a family of enzymes known as 3′ exonucleases. Removing either RecG or the exonucleases caused modest overreplication near the fusion point. Removing both at once had dramatic consequences: Cells generated more new DNA at the fork fusion site than at the chromosomal origin where replication had begun. Two safety nets, both essential, both working in parallel.

This raises a genuine puzzle. Many bacteria, including E. coli, also possess a dedicated “replication fork trap” system. It is a set of DNA roadblocks bound by a protein called Tus that corrals the two converging machines into a confined region before they meet. Yet when you eliminate Tus, cells are barely affected. For a system so precisely built, and independently evolved more than once across bacteria, that is oddly unremarkable. Why guard so carefully against something that seems to cost so little to lose?

Part of the answer comes from the story taking a rather unexpected turn.

For separate reasons, we set out to measure levels of R-loops—structures in which an RNA strand folds back into the DNA double helix, increasingly recognized as sources of genome instability across all life because they obstruct the orderly duplication of the genome. As part of establishing the method, we included cells lacking Tus as a control we expected to behave normally: Tus binds DNA but has no known role in R-loop biology. To our surprise, cells lacking Tus accumulated markedly more R-loops than normal. We initially treated this as a curiosity. Across the many stages of peer review, it emerged as one of the findings that drew great attention.

We do not yet understand the mechanism. But it raises a tantalizing possibility: that one reason the cell maintains its arrival-management system so carefully is that, without it, the balance of RNA and DNA across the whole chromosome is disturbed—far beyond the meeting point itself.

Why does this matter? In bacteria, genome instability is a major driver of adaptation and antibiotic resistance, so its routine sources are worth understanding. But the findings also speak to a longstanding evolutionary question. Most bacteria replicate their chromosome from a single origin, even though additional origins can usually be tolerated. If every arrival carries an inherent risk, evolution had a strong incentive to keep the number of arrivals low—which is exactly what we see across bacteria.

The implications may extend further still: Human cells replicate from hundreds or thousands of origins, and so complete thousands of these termination events every cell cycle. Surprisingly little is known about whether those events, too, threaten the genome.

When we began, the prevailing view was that termination was little more than a mechanical endpoint: forks meet, synthesis stops, replication is complete. Our work shows that the routine completion of DNA replication is itself a source of instability, requiring active management by the cell. The proteins that defuse these intermediates are not insurance against rare accidents. They are essential countermeasures against routine danger.

Completing replication, it turns out, is as difficult as anything else the cell has to do.

This story is part of Science X Dialog, where researchers can report findings from their published research articles. Visit this page for information about Science X Dialog and how to participate.

The content is provided for information purposes only.