The cells were placed in a replica shoulder joint that was moved around to stimulate growth.

By James Vincent

A new method of tissue engineering is only a proof of concept for now.

The science of tissue engineering — or growing human cells for use in medicine — is very much in its infancy, with only the simplest lab-grown cells able to be used in experimental treatments today. But researchers say a new method of tissue engineering could potentially improve the quality of this work: growing the cells on a moving robot skeleton. 

Typically, cells used in this sort of regenerative medicine are grown in static environments. Think: petri dishes and miniature 3D scaffolds. A few experiments in the past have shown that cells can be grown on moving structures like hinges, but these have only stretched or bent the tissue in a single direction. But researchers from the University of Oxford and robotics firm Devanthro thought that, if you want to grow matter designed to move and flex like tendons or muscles, it’d be better to recreate their natural growing environment as accurately as possible. So they decided to approximate a mobile human body. THE THEORY IS THAT MOVING CELLS AS THEY WOULD IN YOUR BODY WILL HELP THEM GROW

Growing cells in an actual person creates all sorts of difficulties, of course, so the cross-disciplinary team decided to approximate the human musculoskeletal system as best they could using a robot. As described in a paper published in Communications Engineering, they adapted an open-source robot skeleton designed by the engineers at Devanthro and created a custom growing environment for the cells that can be fitted into the skeleton to bend and flex as required. (Such growing environments are known as bioreactors.) 

The site they choose for this tissue agriculture was the robot’s shoulder joint, which had to be upgraded to more accurately approximate our own movements. Then, they created a bioreactor that could be fitted into the robot’s shoulder, consisting of strings of biodegradable filaments stretched between two anchor points, like a hank of hair, with the entire structure enclosed within a balloon-like outer membrane.

The skeleton was adapted from the open-source Roboy model. 

The hair-like filaments were then seeded with human cells and the chamber flooded with a nutrient-rich liquid designed to encourage growth. The cells were grown over a two-week period during which they enjoyed a daily workout routine. For 30 minutes each day, the bioreactor was slotted into the shoulder and, for want of a better term, waggled about (though in a very scientific manner).

Here’s the big caveat though: while the team observed changes in the exercising cells that were different from those grown in a static environment, they aren’t actually sure yet if those changes were any good. 

The lead researcher on the project, Pierre-Alexis Mouthuy of the University of Oxford’s Botnar Institute of Musculoskeletal Sciences, told The Verge that the differences he and his colleagues observed in the cells grown this way — which were based on measuring the activity and growth of certain genes — were, at best, ambiguous in terms of future medical applications. “WE’RE JUST SHOWING FEASIBILITY.”

“We do get differences out of the loading regime [the movement of the bioreactor in the robot shoulder joint] but whether those differences mean better cells? We don’t know yet,” says Mouthuy. “We’re not saying this system is better than the others. Or there’s a particular motion that is better than the others. We’re just showing feasibility.”

So: the team has shown that growing cells in a robot skeleton is certainly possible. Now, they just need to find out if it’s worth the time. In the paper, though, the researchers enjoyed some optimistic speculation about the potential of this line of work. They reason that, in the future, detailed scans of patients could be used to create joint-perfect replications of their bodies, allowing tissue like tendons to be grown for surgeries in a human simulacra. 

For now, though, it’s back to the drawing board — or, rather, the robot skeleton. As Mouthuy says, “We need to do much more work to understand what’s really going on.”