Tiny “living machines” made of frog cells can replicate themselves, making copies that can then go on to do the same. This newly described form of renewal offers insights into how to design biological machines that are self-perpetuating.
“This is an incredibly exciting breakthrough,” for the field of biologically based robotics, says Kirstin Petersen, an electrical and computer engineer at Cornell University who studies groups of robots. Robots that can copy themselves are an important step toward systems that don’t need humans to operate, she says.
Earlier this year, researchers described the behaviors of the lab-made living robots, called xenobots (SN: 3/31/21). Plucked out of frogs’ growing bodies, small clumps of skin stem cells from frog embryos knitted themselves into small spheres and began to move. Cellular extensions called cilia served as motors that powered the xenobots as they cruised around their lab dishes.
That cruising can have a bigger purpose, the researchers now report in the Dec. 7 Proceedings of the National Academy of Sciences. As the xenobots bumble about, they can gather loose frog cells into spheres, which then coalesce into xenobots themselves.
This type of movement-created reproduction, called kinematic self-replication by the researchers, appears to be new for living cells. Usually, reproducing organisms contribute some parental material to their offspring, says study coauthor Douglas Blackiston of Tufts University in Medford, Mass., and Harvard University. Sexual reproduction, for instance, requires parental sperm and egg cells to get started. Other types of reproduction involve cells splitting or budding off from a parent.
“Here, this is different,” Blackiston says. These xenobots are “finding loose parts, sort of like robotics parts in the environment, and cobbling them together.” Those collections then grow into “a second generation of xenobots that can move around like their parents,” Blackiston says.
Left to their own devices, spheroid xenobots could generally create only one more generation before dying out, the researchers found. But with the help of an artificial intelligence program that predicted an optimal shape for the original xenobots, the replication could be pushed to four generations.
The AI program predicted that a C shape, much like an openmouthed Pac-Man, would be a more efficient progenitor. Sure enough, when improved xenobots were let loose in a dish, they began scooping up loose cells into their gaping “mouths,” forming more sphere-shaped bots. A mobile offspring took shape once about 50 cells had glommed together in a parent’s opening, Blackiston says. A full-bodied xenobot consists of about 4,000 to 6,000 frog cells.
Xenobots’ tiny size is an advantage, Petersen says. “The fact that they were able to do this at such a small scale just makes it even better, because you can start to imagine biomedical application areas,” she says. Minuscule xenobots might be able to sculpt tissues for implantation, for instance, or go inside bodies to deliver therapeutics to specific spots.
Beyond the possible jobs for the xenobots, the research advances an important science, one that has existential importance for humans, says study coauthor Michael Levin, a developmental biologist at Tufts. That is, “the science of trying to anticipate and control the consequences of complex systems,” he says.
“Originally, no one would have predicted any of this,” Levin says. “These things are routinely doing things that surprise us.” With xenobots, researchers can push the limits of the unexpected. “This is about a safe way to explore and advance the science of being less surprised by things,” Levin says.