Genetically modified bacteria enlisted in fight against disease

Engineered strains of E. coli and other microbes are being tested in people to combat a slew of illnesses.

Search for this author in:

Colourful abstract confocal micrograph of the bacteria Escherichia coli (E. coli).

The Escherichia coli bacteria is being developed as a vehicle for gene therapy in people.Credit: Fernan Federici and Jim Haseloff/Wellcome Coll./CC BY

People often take medicines to rid themselves of problem bacteria. Now, a counterintuitive approach — turning genetically modified bacteria into medicines — is gaining ground.

Several companies are testing whether engineered bacteria can treat conditions that affect the brain, liver and other organs — and even kill other, harmful microbes. But although US regulators have approved trials of several types of engineered bacteria as a form of gene therapy, questions remain about whether microbes’ ability to share DNA with one another will create long-term safety risks.

The idea of using bacteria to deliver gene therapies first surfaced in the 1990s, but early clinical trials met with mixed results. Interest in the approach has increased in recent years amid mounting evidence that the bacteria that live in the body — the microbiome — can influence human health. Researchers are looking to treat disease by modifying microbes that are normally found in people or foods they consume.

Matthew Chang, a synthetic biologist at the National University of Singapore, says that genetically modified bacteria have the potential to treat many types of disease. His group is engineering the gut bacteria Escherichia coli and Lactobacillus to recognize and destroy harmful microbes1. “It’s a rapidly growing area,” says Chang, who adds that he is in talks with regulators in Singapore about starting clinical trials.

Missing piece

One strain of research is aimed at treating the genetic disorder phenylketonuria. People with the condition are deficient in an enzyme that breaks down the amino acid phenylalanine, which causes neurological damage if it builds up in the body. At the American Society for Microbiology’s annual meeting in Atlanta, Georgia, earlier this month, researchers from the biotechnology firm Synlogic in Cambridge, Massachusetts, reported that E. coli modified to produce an enzyme that degrades phenylalanine, and a protein that moves it from blood to cells, reduced levels of the amino acid in monkeys' blood by more than half compared with animals in a control group.

The company started clinical trials in healthy human volunteers in April, and will begin testing the bacteria in people with phenylketonuria as soon as it concludes that the therapy is safe, says chief executive Aoife Brennan. In April, Synlogic began a trial of engineered E. coli that make enzymes to clear the toxic build-up of ammonia in the blood of people with metabolic liver diseases.

Another firm, Intrexon of Germantown, Maryland, has altered Lactococcus lactis, a bacterium used in cheese production, to make a protein that protects the outer layers of the skin. One ongoing clinical trial that has enrolled about 200 people with cancer is testing whether an L. lactis mouthwash can prevent oral sores that are a side effect of chemotherapy. In July, the company will begin dosing people who have diabetes with a different form of L. lactis that produces both the precursor to human insulin and an immune protein that enhances cells’ ability to respond to insulin.

Both Intrexon and Synlogic have engineered their bacteria to make it less likely that they will establish colonies in the body. Patients would have take the modified microbes regularly to ensure consistent doses of the therapeutic molecules they produce. But other companies are pursuing treatments that would create colonies of transgenic bacteria in the body.

The biotechnology firm Osel in Mountain View, California, plans to seek US government approval later this year for a strain of Lactobacillus that has been engineered to prevent HIV transmission. Studies have shown that naturally high levels of Lactobacillus in the vagina can help to protect women against HIV.2 Osel is attempting to enhance the bacteria’s protective properties by modifying it to carry a human protein that prevents HIV from infecting immune cells.

Safety concerns

Challenges remain before these engineered bacteria can enter the market. Scientists need a better understanding of how the bacteria interact with the body, Brennan says, because their effects are less straightforward than those of drugs.

Then there is the risk that the microbes could pass the human genes they carry to other bacteria in the body, with unknown consequences. Several companies have attempted to prevent this sort of exchange by altering the chromosomes of a bacterium, rather than its plasmids — tiny pieces of DNA that bacteria pass back and forth. They have also built in biological ‘kill switches’ that would prevent the microbes from surviving outside the body.

This strategy can fail, however. A group led by immunologist Simon Carding of the University of East Anglia in Norwich, UK, engineered3Bacteroides ovatus to treat colitis, an inflammation of the intestine, by modulating the immune system. The group tried to prevent its bacterium from surviving outside the body by making it dependent on a molecule, thymidine, produced by naturally occurring gut bacteria. The scientists also took care to edit the bacterium's chromosome, rather than its plasmids.

But just 72 hours after the scientists fed the bacteria to mice, they found that B. ovatus had passed its modified gene to other microbes in the animals’ guts — and acquired genes that allowed it to live without thymidine.

The experience caused Carding to abandon efforts to develop bacteria as therapies. “It’s potentially harmful if it’s not properly controlled,” he says. “If you’ve got no control over other bacteria acquiring this foreign gene, others could be producing the protein as well.”

Synlogic, Osel and other companies say they have never observed this type of gene transfer but agree that it is possible. “The microbes are extremely smart and they know how to survive,” says Chang. It remains to be seen, he adds, whether engineering bacteria to colonize the body or die out quickly is a better approach — but the answer could emerge as the current set of clinical trials wraps ups in the next few years.

Nature 558, 497-498 (2018)

doi: 10.1038/d41586-018-05476-4
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Hwang, I. Y. et al. Nature Commun. 8, 15028 (2017).

  2. 2.

    Gosmann, C. et al. Immunity 46, 29–37 (2017).

  3. 3.

    Wegmann, U., Carvalho, A. L., Stocks, M. & Carding, S. R. Sci. Rep. 7, 2294 (2017).

Download references