Bacteria are an attractive vehicle for delivering a treatment payload to cancer as they migrate to the low oxygen environment of tumours1,2, and microbes have been investigated and shown promise as both therapeutic and diagnostic agents in animal models and humans. An example of using bacteria as a vehicle is a clinical trial showing that faecal transplants containing bacteria were effective in reducing recurrent invasions of Clostridium difficile in the gut3. In an example of their use as a diagnostic agent Escherichia coli bacteria were engineered to release an enzyme in the presence of tumours in mice, the enzyme products being detected by the emission of light in the urine4. The rationale behind using light to activate the microbes is that it can trigger them from outside the body, allowing for fine tuning of agent/drug delivery so that it works at the right time and place, without need for invasive surgery to identify or remove the offending, diseased tissue. However, the mechanisms available for regulating microbes have limitations: for instance, injected bacteria often reside at distant sites from the target organs, so that using a chemical constituent injected into the body to trigger the bacteria lacks precision; furthermore, light will only penetrate tissue to a shallow depth, thereby rendering it unable to reach some organs.
In the November issue of Nature Chemical Biology, Piraner et al. designed a new method of regulating bacteria by genetically engineering them to become sensitive to a range of temperature changes, like tiny thermostats. The researchers identified two proteins that were activated at different temperatures; one from Salmonella typhimurium (T1pA) and one from a bacteriophage (TcI), both of which displayed rapid activation and a large increase in expression of their fluorescent reporters over a small temperature range, by binding to DNA to turn reporter genes on or off. They then mutated the T1pA and TcI to switch on and off at different temperatures representing a medically relevant range of 32-46°C. For example, T1pA was mutated to switch on at 36°C (T1pA36). These proteins were chosen especially for their ability to be active in the genetic circuit in combination, without interfering with each other or with other elements in the genome, an essential feature for controlling multiple functions at different temperatures.
The authors tailored the bacteria to achieve three therapeutic regimens. In an unprecedented experiment using ultrasound to raise the temperature in a localised spot in the back legs of mice, the activity of injected fluorescently labelled, engineered bacteria could be controlled by temperature-sensitive proteins. Fluorescence could only be seen at the area where the temperature was raised. The experiment would suggest that bacteria could be activated by temperature in areas like the gut, where light cannot penetrate.
Another useful therapeutic function is for the bacterial treatment to switch off in response to a fever, which can be a lethal side effect of anti-tumour therapy5. In an experiment where bacteria with a T1pA36 mutant were injected into the back legs of mice it produced a fluorescent signal when the room temperature was raised to 41°C showing the T1pA36 responded to an increase in body temperature. In contrast, there was no fluorescent signal in the opposite leg injected with the bacteria containing the original T1pA responsive to 43.5°C.
Finally, the researchers added a temperature-dependent kill switch, essentially fusing the bacterial circuit with a stop gene that allows the bacteria to grow only above 37°C. The stop gene works as a failsafe mechanism that ensures bacteria destruction once they leave the body. The faecal pellets of mice that had ingested this construct had reduced levels of the bacteria at an environmental temperature of 25°C compared to faecal material kept at the 37°C temperature of the gut.
The authors gave the bacteria the tools to switch between multifunctional strategies, where they can activate in response to externally mediated localised beams of ultrasound, and can switch off in response to an internally regulated fever response or at lower temperatures outside the body in order to avoid environmental contamination. The research presents temperature as a new tool in the armoury for the potential treatment of disease and diagnosis in hard to reach anatomical locations.
Piraner, D. I. et al.Tunable thermal bioswitches for in vivo control of microbial therapeutics. Nat Chem Biol. (2016).
1. Forbes, N. S. Profile of a bacterial tumor killer. Nat. Biotech. 24, 1484–1485 (2006).
2. Felfoul, O. et al. Magneto-aerotactic bacteria deliver drug-containing nanoliposomes to tumour hypoxic regions Nat. Nanotech. 11, 941–947 (2016).
3. van Nood, E, et al. Duodenal infusion of donor feces for recurrent Clostridium difficile. N Engl J Med. 368, 407–415 (2013).
4. Danino, T. et al. Programmable probiotics for detection of cancer in urine. Sci. Transl. Med. 7, 289ra84 (2015).
5. Tey, S.-K. Adoptive T-cell therapy: Adverse events and safety switches. Clin. Transl. Immunology 3, e17 (2014).