For years, Dr Robert A Holt, a genomics researcher at the British Columbia Cancer Agency, wrestled with a question about colon cancer. Might it be caused, or pushed along, by a bacterial infection?
Cancers of the liver, stomach and cervix have all been linked to microbes, he knew. And if there is one place in the body with a lot of microbes, it is the colon. Microbial cells outnumber human cells there by a ratio of at least 9-to-1.
The new tools of genomic analysis offered an opportunity to look for a connection. What Holt and another group of researchers, working independently, have found is completely unexpected and puzzling. One particular species of bacterium never particularly prevalent in the colon seems to have a disturbing affinity for colon cancers.
The two research groups discovered the link by analysing genetic material in tumour samples. They then subtracted human genes from the mix. What remained were microbe genes.
More bacteria in tumours
An analysis of these microbial genes showed that a type of bacterium, fusobacterium, was abundant in the tumours although it normally is not among the more prominent species in the gut. Not only were the bacteria lurking around the cancer cells, but Holt found in subsequent experiments that they actually were burrowing into tumour cells – “which is kind of creepy,” he said. An ability to invade cells, he said, is often what distinguishes a disease-causing microbe from one that is harmless. Of course, that doesn’t prove that Fusobacteria are causing tumours. They might just find the cancer cells a good place to live.
As Holt and his colleagues investigated further, they found the bacteria were especially prevalent in patients whose cancer had spread beyond their colons.
The finding could have been an anomaly. But, with no knowledge of Holt’s results, Dr Matthew Meyerson and his colleagues at the Dana-Farber Cancer Institute in Boston found the same thing. And while Holt’s patients were from Canada, Meyerson’s were from people in the United States, Vietnam and Barcelona, Spain. All had the bacteria in far greater abundance in their tumors than in normal colon cells.
“That, to me, was a real eye-opener,” Meyerson said. He expected lots of different bacteria in the tumour tissue, he said. “It turned out not to be that way.”
In their study, Holt and his colleagues began by looking at RNA, which reflects active genes, from 11 colon cancer patients. The colon cancer cells had an average of 79 times as many Fusobacteria as normal cells.
The investigators then looked for the bacteria in 88 more tumours and corresponding adjacent noncancerous colon cells, using probes for Fusobacteria genes. With that more sensitive method, they found an average of 415 times as many Fusobacteria in the tumor cells as in the normal cells.
Meyerson and his colleagues did similar experiments but looked at DNA, the gene sequences, instead of RNA. They began with nine patients, finding Fusobacteria DNA sequences mostly in the cancer tissue. Then they looked at cells from an additional 95 patients, searching specifically for Fusobacteria gene sequences. Again, the researchers found the bacteria in the cancer cells. “I don’t know what to make of it,” Meyerson said. “The bacteria are hanging around the tumors, but I have no idea if they spur or cause cancer.”
If Fusobacteria actually do predispose humans to colon cancer, one day researchers may be able to devise a colon cancer vaccine, much like the HPV vaccine that protects against cervical cancer.
Intriguing findings
Fusobacteria were known before this, of course, but were thought of as microbes that mostly live in the mouth – they are often in plaque and are associated with periodontal disease. But there are also recent reports associating them with ulcerative colitis and Crohn’s disease. Both of these diseases, especially ulcerative colitis, increase the risk of colon cancer.
But if the bacteria are linked to colon cancer, the question is how? One possibility, researchers say, is inflammation. Fusobacteria elicit inflammation, and cancer is linked to inflammation.
That does not necessarily mean that Fusobacteria cause cancer, Relman noted. Tumours themselves can cause inflammation, and some bacteria are quick to take advantage and invade inflamed and damaged tissues.
What remains is an intriguing finding and a lot more work to figure out what is going on. Holt will be looking at polyps, tiny lumps on the colon wall. Colon cancers develop from polyps, though most polyps are harmless.
“The idea is that if we see infections in the early stage lesions, maybe that is one of the factors that allows them to progress,” Holt said. “It does not provide a mechanism or prove anything, but it is consistent.”
