A virus hiding inside bacteria may help explain colorectal cancer

The gut bacterium Bacteroides fragilis has long presented researchers with a paradox. It has been associated with colorectal cancer, yet it also lives quite happily in most healthy people. A new study from a Danish research team offers a possible clue. When they looked beyond the bacterium itself and into its genome, they found a previously unknown virus embedded within it – one that was significantly more common in cancer patients.

Colorectal cancer is the third most common cancer worldwide and is responsible for the second highest number of cancer-related deaths. Up to 80% of colorectal cancer cases are attributed to environmental factors, with one of the most significant being the gut microbiome – the collection of bacteria, fungi and viruses that live in the human gut.

This means that colorectal cancer could – in theory – be partly preventable. But the precise link between the microbiome and colorectal cancer remains poorly understood. It is much easier to associate two things than it is to show a mechanism.

Most studies of the gut microbiome examine which species of bacteria are present and how abundant they are. But species are not homogeneous. Think of how all domestic dogs belong to the same species (Canis familiaris) yet show enormous within-species diversity – a chihuahua is not the same as a great dane. The same is true for bacteria, even if it is harder to visualise.

Just looking at which species are present may not give us the resolution we need to understand what is going on. Perhaps the answer lies not in which bacteria are in the gut, but in the finer genetic differences between strains of the same species.

Bacteroides fragilis is generally considered a harmless member of the gut microbiome and is found in most healthy people. Despite this, it has repeatedly been found to be more abundant in people with colorectal cancer. So could there be specific genetic features that set some strains of B fragilis apart from others, and could these features be linked to colorectal cancer?

Even bacteria get infections

All cellular life can be infected by viruses. Bacteria are no exception. The specific viruses that infect bacteria are called bacteriophages – from the Greek phagos, meaning to eat or devour. They selectively infect bacteria and, importantly, do not infect human cells.

But not all of these viruses kill the bacteria they infect. Some integrate their own genome within the bacterium’s genome, becoming what is known as a prophage – a hitchhiker within the bacterial cell.

Many prophages carry genes that can alter the characteristics of their bacterial host. Diseases such as cholera, botulism and diphtheria are all the result of toxins carried by prophages within otherwise (mostly) harmless bacteria. The conversion of harmless bacteria to harmful ones by prophages is well documented.

To determine whether specific genetic signatures linked B fragilis to colorectal cancer, a Danish team sequenced the genomes of B fragilis from people with and without a colorectal cancer diagnosis.

First, they looked at whether the cancer-associated bacteria came from a distinct evolutionary lineage. They did not. But not all genetic features of bacteria are passed from mother to daughter. Some are acquired sideways, through a process called horizontal gene transfer – such as infection by a prophage.

When the researchers compared the genomes more closely, they found that bacteria from cancer patients carried two previously unknown prophages that were largely absent in bacteria from people without cancer.

These prophages did not carry any obvious genes that would link the bacteria to colorectal cancer – in the way that cholera toxin genes are readily identifiable – but most prophage-carried genes are very poorly understood and we know little about what they do.

A broader test

This initial finding was based on 48 bacteria collected from patients, so the team wanted to test whether the pattern held more broadly. They screened data from faecal samples taken from 877 people across Europe, the US and Asia – 434 with colorectal cancer and 443 without.

Patients with colorectal cancer were more than twice as likely to have detectable levels of the prophages. It is important to stress that this is an association, not proof that these prophages cause or contribute to colorectal cancer. No biological mechanism by which they might do so has been proposed.

It is also possible that the gut environment in cancer patients simply suits these particular strains of B fragilis – meaning the disease could be creating conditions in which the bacteria thrive, rather than the bacteria helping to cause the disease. An alternative explanation is that the gut environment itself predisposes people both to harbour these prophage-containing strains and to develop colorectal cancer.

The study had limitations worth noting. The bacteria originally examined came from patients with bloodstream infections rather than bowel cancer itself, while the broader validation used stool samples – a different source entirely. And some of the “healthy” comparison group had not been formally confirmed to be cancer free.

Despite these limitations, the finding raises an interesting possibility for cancer screening. The most common non-invasive screening method for colorectal cancer is the “faecal immunochemical test”, which checks stool samples for traces of blood. A test that also screened for these viral traces could, in principle, be performed on the same samples.

A preliminary analysis by the researchers found that a panel based on fragments of the prophage genomes detected around 40% of colorectal cancer cases. This is a very early result and would need considerable further work, but it points to the possibility of using viral signatures alongside existing screening methods.

The broader implication of this work is a shift in how we think about the gut microbiome and its relationship to disease. It may not be enough to ask which bacteria are present. We may also need to look at what is inside those bacteria – and what those hidden passengers might be doing.

Ryan Cook does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.