What’s happening in the gut at any given moment is a bit of a mystery. A dietary log and a stool sample can provide some general information, but neither reveals the varying, instantaneous responses of different intestinal segments as food, pathogens, and antibiotics pass through. Without that level of granularity, it’s difficult to pinpoint the cause of intestinal symptoms or evaluate the response to treatment.
To give people a real-time peek into their digestive tracts, a team of researchers led by scientists at the Massachusetts Institute of Technology (MIT) created a device that collects and wirelessly transmits data while traversing the intestine. The device is an ingestible capsule that houses bacteria engineered to sense specific molecules and electronic circuitry designed to relay information from the bacteria to the outside world.
Timothy Lu combines biotechnology and electronics to expand scientific understanding of the body’s inner workings.
credit: Yves Shahar
In a recent Nature paper describing their invention, the researchers focused on detecting indicators for inflammatory bowel disease (IBD) (1). Their device has the potential to give people all sorts of information about their inner workings that could help diagnose and treat gut-linked diseases. “What we’re trying to do here is illuminate this dark matter of the gut,” said Timothy Lu, a bioengineer at MIT and coauthor of the study.
To develop the capsule’s living, unicellular sensors, Lu’s team set out to transform ordinary bacteria into chemical detecting machines. “Bacteria have evolved over trillions of years to live in the human body, and so we think bacteria are a really good platform [for] biosensing,” said Lu. Specifically, the scientists used the Escherichia coli strain Nissle 1917 (EcN), which was first isolated from the stool sample of a healthy soldier fighting in World War I and has since been established as a safe probiotic for human consumption (2).
EcN bacteria naturally build their own chemical sensors from genetic blueprints either passed down to them by their ancestors or collected from the scraps of DNA they encounter in their environment. Genes on these DNA strands encode proteins that alert bacteria to initiate certain responses if they encounter specific molecules. For instance, bacteria need to prepare defenses if toxic reactive oxygen species (ROS) and reactive nitrogen species (RNS) reach damagingly high levels, which may occur in the human body during stressful situations such as a flare up of IBD symptoms.
Because bacteria readily internalize strands of DNA that they encounter in their environments, engineering EcN bacteria required Lu’s team to simply modify existing genes and introduce them to the plates where the bacteria grew. The biologists took the DNA for four different ROS and RNS sensors and paired each gene with the genetic code for green fluorescent protein (GFP). They combined the sensor and GFP genes so that when the sensor detected its specific ROS or RNS molecule, the bacteria expressed GFP, which emits light upon exposure to UV radiation.
A population of bacteria can go through several generations in the time it takes for material to pass through the digestive tract, so the scientists needed to find a way for the record of a molecular encounter to pass from one generation to the next. To do so, they engineered the GFP gene so that it remained active even after the DNA was replicated and passed down to a bacterium’s offspring.
The biologists optimized the genetic circuits so that the GFP genes turned on when the bacteria encountered relatively high levels of specific ROS and RNS molecules. Then they validated the diagnostic capabilities of their bacteria by assessing the sensors’ detection of intestinal inflammation, which produces high levels of ROS and RNS species. They administered the bacteria to mice and pigs with and without inflammation and, after giving the bacteria time to colonize the animals’ guts, collected the sensors from the stool samples of the mice and directly from the intestines of the pigs. When they exposed their samples to UV light, they found that the bacteria collected from animals with inflamed intestines showed higher GFP expression, confirming that the bacteria sensed high levels of ROS and RNS.
While this experiment yielded an exciting proof of concept, the device didn’t yet achieve the ultimate goal of providing real-time data from the intestine. Rather than analyzing the bacterial sensors with UV exposure after excretion, the researchers turned to a sensing method they had used in a previous study (3).
They built genetic circuits that combined ROS and RNS sensor genes with a gene encoding a luciferase protein. Luciferases catalyze chemical reactions that produce visible light, and while they’re found in a variety of organisms, they’re most famous for lighting the fire in fireflies. In the genetic circuit that Lu’s team built, the luciferase gene turned on when an ROS or RNS molecule was detected and turned off when it wasn’t. In other words, the bacteria lit up like fireflies in an inflamed portion of the gut but stayed dark in a healthy one. Because there’s no other source of light in the gut, the researchers could monitor the visible light from the bacteria with a photodetector connected to a wireless transmitter that continuously signals to an external receiver for real-time detection.
Still, problems could arise with this form of signal transmission. “Is that photodetector going to get corroded or wear out?” asked Arthur Prindle, a chemical and biological engineer at Northwestern University who was not involved in Lu’s study. A potential alternative is to bypass the photodetector and engineer bacteria that can integrate themselves directly into the electronic circuitry leading to the wireless transmitter. “There are bacteria that can induce currents on electrodes,” Prindle said. “It just seems cleaner to have a direct, wired connection there [rather] than using light.”
For now, the researchers focused on designing an ingestible capsule for their bacterial sensing technology. They miniaturized the circuitry so that it fit inside a pill small enough to swallow. They also created a casing with distinct chambers that protect the engineered bacteria from the inhospitable elements of the gastrointestinal tract while they sample chemicals from their surroundings and communicate with the photodetectors. Having multiple chambers enabled the researchers to load the capsule with several lines of EcN bacteria engineered to test for different ROS and RNS molecules.
The Fitbit for the gut is what you want, and so this is a huge leap forward.
– Arthur Prindle, Northwestern University
To evaluate their device, the researchers exposed it to harsh conditions that simulate ingestion and passage through the stomach and confirmed that it continued to function. They also surgically inserted the device into the intestines of living pigs and confirmed that they could measure real-time changes in the concentration of specific RNS and ROS molecules. They monitored wireless signals from the device through a commercial receiver compatible with laptops and smartphones.
Although they confirmed that their device could communicate information from within a living animal, the researchers could not let the pigs swallow and pass the pill due to limitations in resources and the testing for which they were authorized to use the animals. Still, the results were promising.
“The next big step for us is really trying to figure out how do we get this into clinical studies,” Lu said. At the same time, he is anticipating how to adapt the device for additional applications. “What we’re trying to do here is basically build a modular platform so we can try to plug in any biosensor of interest in the future,” he said.
Prindle is excited about the potential of Lu’s modular real-time sensing platform as his team recently developed an EcN bacterial sensor for calprotectin, a protein found in stool samples that clinicians often use to diagnose IBD (4). Prindle said, “The Fitbit for the gut is what you want, and so this is a huge leap forward.”
References
- Inda-Webb, M. et al. Sub-1.4 cm3 capsule for detecting labile inflammatory biomarkers in situ. Nature 620, 386-392 (2023).
- Sonnenborn, U. Escherichia coli strain Nissle 1917—from bench to bedside and back: history of a special Escherichia coli strain with probiotic properties. FEMS Microbiology Letters 363, fnw212 (2016).
- Mimee, M. et al. An ingestible bacterial-electronic system to monitor gastrointestinal health. Science 360, 915-918 (2018).
- Xia, J. et al. Engineered calprotectin-sensing probiotics for IBD surveillance in humans. PNAS 120, e2221121120 (2023).
