Magnetic soft robots deliver and release tens of millions of probiotic bacteria with timed control, disrupt tumor spheroids in vitro, and demonstrate a potential strategy for targeted cancer treatment.
(Nanowerk Spotlight) Efforts to turn bacteria into targeted cancer treatments have tested the limits of both microbiology and medical engineering. More than a century ago, doctors observed that some microbes naturally accumulate in tumors, especially in dense regions low in oxygen where most drugs fail. This sparked long-running attempts to use bacteria as living therapies, trained to attack malignancies from the inside.
Yet several barriers stood in the way. Delivering the right number of bacteria deep into specific tumors proved unpredictable. Steering individual cells inside the body required advanced imaging and manipulation tools, which struggle to keep pace with fast-moving biological systems.
Later experiments attempted to attach magnetic nanoparticles or drug payloads to bacterial surfaces so they could be directed from outside the body. But these modifications came with a major flaw. When bacteria divide, they shed the added particles and quickly lose controllability. The more they multiply, the less influence external control methods can exert over where they go.
A study by researchers at ETH Zurich published in Advanced Intelligent Systems (“Magnetic Soft Robots for Targeted Bacterial Delivery and Enhanced Tumor Spheroid Disaggregation”) describes an alternative solution that treats bacteria as a payload inside a soft robot instead of trying to steer individual cells. The researchers developed a magnetic hydrogel robot that encapsulates a concentrated population of probiotic bacteria and can be moved by an external magnetic field. Once delivered to a target area, the bacteria gradually escape from the robot and start to grow.
Magnetic living soft robots for medical interventions. A) The robots can be used for concentrated bacteria delivery to desired tumor sides, potentially inside the gastrointestinal tract, in a targeted fashion using external alternating magnetic fields. B) A prepolymer solution is formulated by homogeneously mixing sodium alginate, barium hexaferrite (BaFe) magnetic microparticles (), and probiotic bacteria (i.e., E. coli Nissle 1917). This solution is then molded into 3D-printed structures and eventually crosslinked with a calcium chloride (CaCl2) solution to create the magnetic living soft robots. C) Magnetic soft robots with different alginate (Alg) and BaFe concentrations: i. 2.5% Alg þ 10% BaFe, ii. 5% Alg þ 5% BaFe, iii. 5% Alg þ 10% BaFe, and iv. 5% Alg þ 20% BaFe. D) Confocal imaging of the magnetized soft robot loaded with bacteria and stained via a bacterial live/dead assay, showing the rod-shaped, living bacteria in green 4 h after biofabrication. Black halos indicate MMPs’ aggregates. E) SEM image of a hydrogel that is composed of 5% w/v Alg and 20% w/v probiotic bacteria, showing bacteria’s distribution within the polymer network. (click on image to enlarge)
The robot is made from calcium alginate, a soft, water-rich polymer commonly used in biomedical applications. Alginate forms a porous network when crosslinked with calcium ions, creating a structure that supports living cells while allowing fluids to pass through. To make the robot responsive to magnetic fields, barium hexaferrite microparticles were added to the gel.
These particles are about five micrometers in diameter and retain magnetization after exposure to a strong magnetic field, enabling movement under a weaker field applied externally.
Their size is noted by the authors as unlikely to cross intestinal mucus barriers. Because barium absorbs X-rays, the robot is visible under X-ray fluoroscopy, allowing potential tracking.
The bacterial strain used in the robot is Escherichia coli Nissle 1917, a nonpathogenic bacterium commonly used in gastrointestinal research. The bacteria, magnetic particles, and alginate solution were mixed together and cast into 3D-printed molds under sterile conditions. After gelation, the resulting robots measured about 11 by 4.5 by 0.4 millimeters and contained a high concentration of bacteria based on the initial loading percentage.
One of the key tests examined the timing and viability of bacterial release. When a single robot was placed in warm nutrient media, bacteria began exponential growth after a delay of around one and a half hours compared with free bacteria added directly to the same media. This delay corresponds to the time needed for the bacteria to diffuse out of the hydrogel. The final cell concentration was similar in both cases, indicating that the robots did not inhibit long-term bacterial growth.
Further tests explored how acidity affects bacterial release and proliferation. At lower pH levels, such as 6.5 and 5.5, bacteria exited the robot sooner and grew more rapidly. At pH 5.5, the final bacterial level increased by more than 30 percent compared to neutral pH. This difference in growth rate helps define how the system might behave in regions of the body with varying acidity, though the study does not extend into in vivo conditions.
The researchers also investigated whether bacteria exiting the robot could affect tumor models. They used three-dimensional spheroids composed of human colorectal cancer cells, which mimic the structure and density of solid tumors more accurately than flat cultures. When robots containing bacteria were placed in the same media, bacteria accumulated around the spheroids, detached clusters of cells, and disrupted their structure over time.
At around 17 hours, the spheroids were fully disassembled, and staining showed increased cell death. Robots without bacteria did not affect the spheroids, while free bacteria added directly to the media produced similar effects to robot-delivered bacteria.
Motion control of the robot was achieved by imprinting a magnetic pattern into its structure. The robot was wrapped around a rod at a 45-degree angle and exposed to a strong magnetic field, creating alternating magnetic regions along its length. When later placed in a rotating magnetic field, the robot bent into a C-shape and rolled across wet surfaces in a 3D-printed colon model.
In liquid, it alternated between C and V shapes, creating a swimming motion. Rolling occurred under higher field strengths and lower frequencies, while swimming occurred under lower field strengths and higher frequencies.
To assess movement in a complex environment, the robot was tested in a colon phantom that recreates the folded surfaces of the human colon. Under moderate magnetic fields and frequencies, the robot achieved speeds of over 100 millimeters per second while maintaining directional control.
Speeds increased with frequency up to about seven hertz, then declined as the magnetic response became unsynchronized. When field strength exceeded about 25 millitesla, the robot curled too tightly, reducing movement efficiency.
This study defines a soft robotic system that can deliver a concentrated dose of millions of living bacteria, navigate complex surfaces under magnetic guidance, and release its payload on a reliable timescale.
All findings so far are limited to controlled lab environments, but the design points toward a method for directing tens of millions of bacteria to a specific site, something that is not achievable with free-swimming cells.
Future investigations will need to test navigation in real tissue and track how the magnetic components behave over time, but the core demonstration sets a technical foundation for developing targeted microbial therapies.
For authors and communications departmentsclick to open
Lay summary
Prefilled posts
Nanowerk Newsletter
Get our Nanotechnology Spotlight updates to your inbox!
Thank you!
You have successfully joined our subscriber list.
Become a Spotlight guest author! Join our large and growing group of guest contributors. Have you just published a scientific paper or have other exciting developments to share with the nanotechnology community? Here is how to publish on nanowerk.com.
