Many consider enzymes the workhorses of biochemistry (move over, mitochondria)—catalyzing reactions, breaking down substrates, keeping the machinery of life humming along. But a growing number of researchers are re-envisioning what enzymes can do. Instead of facilitating chemistry, what if enzymes could steer and even guide tiny robots to a tumor? 

That’s exactly what’s happening in the rapidly expanding field of enzyme-powered microscopic robots (a.k.a “microrobots”). Microrobots are tiny, engineered devices—often smaller than the width of a human hair—built to perform tasks inside the body that would be difficult or impossible at a larger scale, like delivering drugs to a specific tissue. A recent paper published in Nature Nanotechnology by a team of researchers at California Institute of Technology and the University of Southern California offers a particularly elegant example that we highlight below¹.

Part 1: Building the Bubble Engine

The Caltech and USC team built microrobots out of something deceptively straightforward: protein-shelled microscopic bubbles. By sonicating a solution of bovine serum albumin (BSA) with an ultrasound probe, they generated thousands of tiny bubbles with protein shells that are both structural and functional, providing chemical groups like amines that act as attachment points for other molecules.

The researchers took advantage of this “sticky” structure by chemically attaching the enzyme urease to the bubble surfaces.  Urease catalyzes the hydrolysis of urea (a waste product abundant in biological fluids) into ammonia and carbon dioxide. Because the urease isn’t uniformly distributed across the bubble surface, reaction products accumulate asymmetrically. Think of releasing air from a balloon: a balloon has only one opening, so the escaping gas propels it in one direction. Similarly, that chemical imbalance generates a net directional push, effectively turning each bubble into a self-propelled microrobot fueled by urea, a chemical already present in the human body. 

Part 2: Designing the Chemical Compass 

With propulsion solved, the team turned to a second engineering challenge: directing the microrobots to a specific target. For the next iteration of the bubble bots, the team added a second enzyme to the surface: catalase. Catalase breaks down hydrogen peroxide into water and oxygen, and tumors are known to produce elevated concentrations of hydrogen peroxide compared to healthy tissue. That difference in concentration creates a chemical gradient, with more hydrogen peroxide near the tumor and less in the surrounding healthy tissue. This means the catalase-equipped microbubbles move up the hydrogen peroxide gradient toward tumor sites on their own through chemotaxis—a process typically associated with living cells navigating chemical gradients, but here engineered into a synthetic device. In other words, the catalase enzyme steers toward the tumor site for drug delivery. 

Putting all of this together, the researchers deployed the bubble robots into mouse models. When tested in a mouse model of bladder cancer, the enzyme-powered bubble bots were loaded with anti-tumor therapeutics and delivered to the tumor site. Once there, the researchers applied focused ultrasound to burst the bubbles. When the bubbles pop, they simultaneously release the drug and generate mechanical forces that drive the drug deeper into the tumor tissue than it would penetrate on its own. Over 21 days, this approach reduced bladder tumor weight by roughly 60% compared to drug treatment alone. 

Enzymes As Machines: A Broader Trend 

The “bubble bots” are part of a broader shift in how scientists think about enzymes. Across the field of micro- and nanorobotics, enzymes are increasingly deployed as active mechanical components in these robots, creating tiny engines, sensors and guidance systems built from proteins.

Urease, in particular, has become a popular molecular motor. In 2024, a separate team published work in Nature Nanotechnology demonstrating urease-powered nanobots made from mesoporous silica that could carry radioactive iodine directly to bladder tumors in mice, achieving roughly 90% tumor reduction². Other groups have used catalase to propel polymer-based nanomotors through hydrogen peroxide gradients, glucose oxidase to power motion in glucose-rich environments, and even lipase to drive particles through lipid-containing fluids³.

What makes enzyme-powered systems particularly attractive for biomedical applications is their fuel source: many of these designs run on substrates already present in the body. Urea is everywhere in biological fluids. Glucose is abundant in the bloodstream. The fuel is already there and researchers are just getting started figuring out what enzymes can do with it. 

References

1. Tang, S. et al. Enzymatic microbubble robots. Nat. Nanotechnol. (2026). https://doi.org/10.1038/s41565-025-02109-6   ↩︎
2. Simó, C. et al. Urease-powered nanobots for radionuclide bladder cancer therapy. Nat. Nanotechnol. 19, 554–564 (2024). https://doi.org/10.1038/s41565-023-01577-y   ↩︎
3. Chen, S., Prado-Morales, C., Sánchez-DeAlcázar, D. & Sánchez, S. Enzymatic micro/nanomotors in biomedicine: from single motors to swarms. J. Mater. Chem. B 12, 2711–2719 (2024). https://doi.org/10.1039/D3TB02457A ↩︎

The following two tabs change content below.

• Bio
• Latest Posts

Anna Bennett

Anna earned her PhD in microbiology at the University of Minnesota in 2022 where she studied the microbial communities in hot springs. She joined Promega in 2023 as science writer within the Marketing Services department. When she's not writing, she enjoys being outdoors with her dog, Calvin.

Latest posts by Anna Bennett (see all)

• Down the Rabbit Hole: The Search for New England’s Disappearing Cottontail - March 31, 2026
• How Enzymes Are Powering A New Generation of Micro-Robots - February 24, 2026
• A New View of Protein Degradation with HiBiT and Live Cell Imaging - September 23, 2025