Engineered bacteria to detect gut inflammation
I wrote about engineered probiotics at the beginning of 2017, and the field continued throughout 2017 with more papers and startup news using engineered bacteria in the gut. For instance, one paper used engineered probiotics to attack Pseudomonas aeruginosa gut infection and another used engineered probiotics to treat phenylketonuria (PKU is genetic disorder that prevents breakdown of the amino phenylalanine). I was particularly interested in a pair of papers – one from Pam Silver’s lab at Harvard and one from Jeff Tabor’s lab at Rice University – described new sensors in bacteria so that they can detect inflammation in the mammalian gut. This type of work shows how probiotic-based diagnostics can provide measurements from directly inside your gut without invasive procedures.
How the inflammation sensors work
Both of these papers used two-component regulatory sensors and transcriptional regulators that respond to chemicals which are released when there is inflammation in your gut. The regulator can turn memory or output genes on or off depending on how much of its chemical there is. If the chemical goes up and down with how much inflammation there is, then you have a good record of of the inflammation happening where the bacteria goes in your gut.
Jeff Tabor’s lab’s paper in Molecular Systems Biology uses two component-systems from a Shewanella species to create a thiosulfate sensor and an improved tetrathionate sensor in a probiotic E. coli. Both of these sensors were characterized in a regular lab strain of E. coli and then transferred to a probiotic strain of E. coli called Nissle 1917. The expression of the two component sensors was tuned to have good dynamic range in E. coli Nissle. Both the thiosulfate sensor and the tetrathionate sensor had sensitivities in the submillimolar range.
For their in vivo tests, they used a chemically induced mouse model of gut inflammation called the dextran sodium sulfate (DSS)-induced colitis model, and they tested their bacteria for sfGFP fluorescence by flow cytometry. This gives measurements of individual cells and how much thiosulfate or tetrathionate that cell sensed. The paper shows good detection of thiosulfate in the mouse model but no detectable increase in tetrathionate.
Pam Silver’s lab published their paper in Nature Biotechnology (preprint here) used parts from Salmonella typhimurium in E. coli to detect tetrathionate and record the information in a memory switch that controlled a lacZ reporter.
The memory worked by borrowing the Cro-inducible CI/Cro transcriptional switch derived from lambda phage. Usually the CI/Cro tells a phage when to be in lysogenic or lytic life cycles. Instead the engineered version replaced lytic genes with a lacZ reporter that creates a color change. That way when Cro is produced it flips the switch into a state where lacZ is created and the color can be seen in bacterial colonies isolated after they pass through the mouse gut. Previous work from the Silver lab had shown that this could be used to detect a small molecule inducer like tetracycline.
To make the memory record if it saw inflammation, they put Cro under control of the PttrBCA promoter that gets regulated by the ttrR/S genes in response to tetrathionate that’s produced during inflammation. They tested this system in vivo using a genetic mouse model of inflammation instead of using DSS to induce gut inflammation.
Probiotics could measure inflammation in ways current tests cannot
Current tests for gut inflammation usually measure molecules in your stool or a biopsy of your intestine. Measuring different proteins in your stool is non-invasive but can’t distinguish duration of inflammation or where it’s localized to. Biopsies give more spatial information but require invasive techniques done by a doctor.
Since diagnostic probiotics function inside of your gut with different sensors or genetics circuits, one could potentially integrate inflammation measurements with other biosensors to create more sophisticated readouts. By measuring inflammations molecules along with other signals, you could imagine measuring where, when, and how much inflammation was in the gut. More detailed information of inflammation in the gut could help to monitor and manage diseases like inflammatory bowel disease (IBD). If coupled with therapeutic capability, a probiotic could also be designed to sense inflammation and respond with an anti-inflammatory molecule.
Some engineered probiotics are already in early clinical trials
As the research on engineered probiotics has advanced so has the push to make an impact in the clinic. Synlogic reported positive results for its early-state clinical trials trying to treat high blood ammonia levels. They also received orphan drug status for an engineered probiotics to treat phenylketonuria (PKU) similar to the PLOS paper already mentioned in first paragraph. There are also clinical trials for LACTIN-V a vaginal probiotic. But there hasn’t been strong evidence for the usefulness of probiotics in the gut on there own. Only in late 2017 were there a report of strong clinical trial data on probiotic use in which Lactobacillus plantarum with a particular sugar supplement could help prevent sepsis in newborns.
For engineered probiotic diagnostics to impact the clinic, there will need to be clear markers that are both indicative of disease and detectable using synthetic sensors. Markers of inflammation could certainly be helpful. Detecting a larger set of biomolecules like human signaling molecules or metabolic markers would open up more complex and complete diagnostic information.
Relevant papers
David T Riglar, Tobias W Giessen, Michael Baym, S Jordan Kerns, Matthew J Niederhuber, Roderick T Bronson, Jonathan W Kotula, Georg K Gerber, Jeffrey C Way & Pamela A Silver. “Engineered bacteria can function in the mammalian gut long-term as live diagnostics of inflammation” (full text, preprint)
Kristina N‐M Daeffler, Jeffrey D Galley, Ravi U Sheth, Laura C Ortiz‐Velez, Christopher O Bibb, View ORCID ProfileNoah F Shroyer, Robert A Britton, View ORCID ProfileJeffrey J Tabor. “Engineering bacterial thiosulfate and tetrathionate sensors for detecting gut inflammation” (full text)
In Young Hwang, Elvin Koh, Adison Wong, John C. March, William E. Bentley, Yung Seng Lee & Matthew Wook Chang. “Engineered probiotic Escherichia coli can eliminate and prevent Pseudomonas aeruginosa gut infection in animal models” (full text)
Katherine E. Durrer, Michael S. Allen , Ione Hunt von Herbing. “Genetically engineered probiotic for the treatment of phenylketonuria (PKU); assessment of a novel treatment in vitro and in the PAHenu2 mouse model of PKU” (full text)
Kotula JW, Kerns SJ, Shaket LA, Siraj L, Collins JJ, Way JC, Silver PA. “Programmable bacteria detect and record an environmental signal in the mammalian gut”. (full text)