When immunologist Janelle Arthur was training as a postdoc at UNC nearly a decade ago, she proposed a project looking at how a Western diet – rich in salt, fat, and red meat – might increase the risk of developing colorectal cancer. Numerous studies had suggested a link between diet and cancer, and Arthur wondered if those links could be traced back to the trillions of bacteria that call the gut home.

To test her hypothesis, Arthur planned to feed a Western diet to mice that were prone to cancer, and then track how their gut bacteria — collectively known as the microbiome — changed as they developed tumors. The experiment sounded straightforward enough, but quickly went off the rails. "The challenge with science is it never goes how you think it's going to go," says Arthur, now an assistant professor of microbiology and immunology at UNC and member of the NC Translational and Clinical Sciences (NC TraCS) Institute.

The problem was Arthur had to sterilize every item on the rodent menu in a special oven called an autoclave to ensure that she didn't introduce any microbes with each meal. Under the high temperatures, the high-fat foods dissolved into a gelatinous mess that had to be served in tiny little ramekins to the mice. The creatures turned up their noses at their new diets, and Arthur couldn't say she blamed them. So when another postdoc left the lab and asked her to take over his project, she didn't hesitate.

That project eschewed the diet and instead directly manipulated the composition of microbes in the mouse gut to determine its impact on inflammation-associated cancer development. Arthur found that whereas mice whose guts contained a bacteria called E. faecalis developed only inflammation, those harboring E. coli developed invasive colorectal tumors. That discovery led to a 2012 paper in the top-tier journal Science, and a new line of inquiry into the microbial origins of cancer that Arthur continues to this day.

"There are several well-established hallmarks of cancer that every normal cell must acquire as it evolves into a cancer cell," says Arthur. "We believe that our resident microbes likely influence most, if not all, of those hallmarks, either tipping the balance toward stability or setting the course for cancer."

In her seminal 2012 paper, Arthur investigated why E. coli strains promoted cancer whereas E. faecalis strains did not. She sequenced both bacterial strains, and found that the cancer-causing strains contained a chunk of genetic material known as the PKS gene island. This island produces a poisonous chemical called colibactin that damages DNA. Importantly, when she profiled the microbiomes of human subjects, she showed that a high percentage of patients with inflammatory bowel diseases (IBD) and colon cancer possessed this dangerous strain of E. coli.

Since then, Arthur has been trying to figure out what other aspects of this E. coli strain might be contributing to chronic inflammation, IBD, and cancer. She found that microbial gene islands similar to PKS produce a bunch of different molecules, some of which bind and transport metals like zinc or iron across cell membranes. Arthur says these metals are under-recognized but absolutely essential.

"There's so much that we don't think about that just happens in the background of our body that is dependent upon these metals," she says. "You need zinc just for your cells to be alive. These metals are normally bound to proteins, but when they're not they can create reactive oxygen species or free radicals that can damage cells."

In recent years, a scientific controversy has arisen over whether or not one metal in particular — iron — is good for patients with IBD. Some say that iron supplements can help patients deal with the anemia wrought by blood loss in their intestines; others argue that there may be bad microbes in the gut that will feed off the iron, putting patients at further risk. "There is this constant battle for iron between the host and the microbe, and you can't tell who is going to win," says Arthur.

To explore this controversy, Arthur deleted a receptor responsible for taking up iron in E. coli and used it to colonize the guts of mouse models of IBD. She found that the mice developed fibrosis, a common and potentially serious complication for which the only treatment is removal of a portion of the bowel. "This is a huge problem and nobody knows why it occurs," says Arthur. Follow-up experiments suggest that when bacteria lose the ability to compete for iron, it throws off the equilibrium or "local metal homeostasis" in cells. "You need these metals for many cellular processes, but if they're just hanging out, not doing what they should be doing, they could be damaging cells," Arthur says.

But Arthur still isn't sure iron is the culprit. The iron-binding molecule produced by E. coli can also interact with other metals like copper, zinc, cobalt, and aluminum. So she will need to tweak more receptors in bacteria and assess the health of more mice before she will gain a clearer picture of what is causing this severe form of IBD. Ultimately, such work may lead to better diagnostics and treatments for IBD, which affects as many as 1.6 million Americans and significantly increases the risk of colorectal cancer. For example, scientists might design a microbiome test that screens for specific microbial genes — such as those that produce DNA-damaging poisons or screw up the composition of metals — in order to identify patients who might benefit from tailored therapies.

For now, Arthur is focused on finding more answers. She secured a TraCS $2K pilot award to work with a collection of about 50 E. coli strains cultured from the guts of IBD, cancer, and healthy patients. In the pilot project, one of her postdocs developed a system where they can tag each of the E. coli strains with molecular barcodes, colonize the guts of mouse models, and then assess the relative abundance of each strain by high-throughput sequencing. "The TraCS grant was really instrumental because it enabled us to establish this barcoding system, which we used to complete K01 studies and generate preliminary data for an R01 application."

Ironically, Arthur is using some of the new funding to put rodents on a diet, this time testing various concentrations of metals like iron or zinc. She says that when the metal-heavy foods are stuck in the autoclave, they come out solid like bricks. Luckily, she isn't responsible for the cooking anymore, as a staff scientist in the lab has taken on the task.