Overcoming a major hurdle in the use of microbes as medicine, researchers have introduced an implantable “living material” that contains bacteria that sense infections. It can release these therapeutic molecules on demand, while keeping them physically separated from the surrounding tissue. The findings represent a shift from passive drug depots to autonomous, responsive – and living – therapeutic systems. Engineered living cells are emerging as a new class of medicine that can autonomously sense disease and deliver treatment directly at affected sites. Unlike conventional drugs, these “living therapeutics” can sustain themselves in vivo and survive in many biological environments, including tumors, inflamed tissues, infected tissues, and even within human cells. Bacteria are particularly attractive because they can be genetically programmed to release therapeutic molecules in response to specific biological signals. However, this versatility also raises an important safety concern: therapeutic bacteria must be physically contained to prevent uncontrolled spread and potential toxicity. Previous implantable biomaterial systems, such as hydrogels and capsule-like enclosures, have shown some success in confining microbes, but only for short periods – typically no more than 2 weeks.

Tetsuhiro Harimoto and colleagues discovered that bacterial growth can be halted when the surrounding material reaches a sufficient level of stiffness, preventing the internal pressure of bacterial overgrowth from causing escape. At the same time, the material also needs to be tough enough to withstand the constant mechanical stress from surrounding tissues without cracking. To achieve this balance, Harimoto et al. created an implantable living material (ILM) consisting of a hierarchical hydrogel composed of bacteria-filled gelatin microgels embedded within a reinforced polyvinyl alcohol framework. In laboratory testing, the authors show that the material remained intact for 6 months with no detectable bacterial leakage, even under conditions designed to mimic long-term physiological stress. To evaluate the material’s clinical potential, Harimoto et al. transformed the ILM into an active therapeutic system by engineering bacteria to detect chemical signals from Pseudomonas aeruginosa, a common cause of implant-related infections. In response, the bacteria autonomously self-destructed to release an antibacterial protein that killed the pathogen. In a mouse model of joint infection, the system successfully reduced bacterial burden, demonstrating the potential of durable, programmable ILM-based therapeutics for long-term disease treatment. “Rather than treating the scaffold as a passive vehicle, Harimoto et al. treat it as an active determinant of whether contained bacteria can function safely over time,” write Kaige Chen and Quanyin Hu in a related Perspective. “This reframing brings living therapeutics closer to a model in which long-term, in vivo embedded therapeutic function replaces repeated drug administration.”

Podcast: A segment of Science's weekly podcast with Tetsuhiro Harimoto, related to this research, will be available on the Science.org podcast landing page after the embargo lifts. Reporters are free to make use of the segments for broadcast purposes and/or quote from them – with appropriate attribution (i.e., cite "Science podcast"). Please note that the file itself should not be posted to any other Web site.

A study led by a University of Louisville School of Medicine pediatrics and child neurology researcher reveals how a specific signaling mechanism in microglia, the brain immune cell, can regulate anxiety and grooming behaviors. These behaviors are core symptoms of autism and obsessive-compulsive spectrum disorders.

U.S. teens report far less anabolic steroid use than they did two decades ago, but creatine use has risen rapidly in recent years, according to a new University of Michigan study.