Machines Powered by Heart Muscles
Devices powered by living cells could lead to novel robots and new insights into how the heart works.
Researchers at Harvard University have made several small mechanical devices powered by heart muscle harvested from rats. The mechanical devices include pumps, a device that “walks,” and one that swims. The scientists made the novel machines to study the behavior of muscles and provide a platform for testing heart drugs. But one day these devices could be used as parts of new types of robots that can change shape.
The research, featured in the current issue of Science, began as an attempt to grow working muscle tissue to patch holes in congenitally defective hearts, or to replace dead tissue after a heart attack. In the course of this work, Adam Feinberg, a postdoctoral researcher in the lab of Harvard biomedical-engineering professor Kevin Kit Parker, found that if patterned correctly and applied to carefully shaped sheets of plastic, the muscle could be used to make the plastic bend and twist in various ways.
In one example, Feinberg made a rectangular strip of plastic that curls up on itself, with the diameter of the resulting tube decreasing, then increasing again, as the muscle repeatedly contracts and relaxes. The researchers say that the device could serve as a pump. Another strip of plastic opens and closes like a pair of pinchers at a rate determined by electrical signals sent to the device. A curled triangular piece of plastic walks across the bottom of a petri dish as muscle tissue repeatedly contracts, and another triangular sheet, with a different arrangement of heart-muscle cells, mimics the movement of a fish’s tail to swim through a solution.
To grow heart muscle in the lab that contracts in a regular beating rhythm, it’s necessary to arrange the cells so that they are mechanically and electrically connected. They must also be oriented correctly. To do this, the researchers created micropatterns of proteins. These proteins create “cues” for rat muscle cells deposited on the plastic, guiding their alignment. Once the cells are deposited on a surface patterned with the proteins, they orient themselves to form a working tissue, Parker says.
A thin plastic serves as both a substrate for the tissue and a way of causing the devices to spring back to a certain shape in between contractions of the tissue. Parker envisions these devices one day being incorporated into octopuslike robots that can squeeze through small openings but also grip and manipulate objects and propel themselves along.
In their present form, however, the devices will have limited usefulness in robotics. For one thing, the beating muscles only survive for a few weeks, even when fed with a constant supply of nutrients, including glucose, which serves as the fuel for the muscles. But future designs could mimic natural heart tissue in more detail to extend longevity. For example, the researchers may also try constructing a three-dimensional tissue, rather than the flat arrangement they have now. Previous experiments have suggested that three-dimensional structures may be key to the survival of the cells.
The devices could give scientists a better tool for understanding healthy and diseased heart tissue, specifically how the precise arrangement of cells is necessary for the heart to function properly, says Kenneth Chien, director of the Massachusetts General Hospital Cardiovascular Research Center. Eventually, the devices could be used to screen drugs for effectiveness and toxicity by giving researchers an easy way to measure how a drug alters heart-muscle function.
Several other researchers have used heart muscle for powering devices. Earlier this week, for example, researchers in South Korea reported engineering a tiny heart-muscle-powered crablike device that “crawled” for 10 days. That work was reported in the journal Chemical Science.
But the Harvard muscle-power machines are larger and have the potential to be used in a wide variety of devices, says Bob Dennis, a professor of biomedical engineering at the University of North Carolina and North Carolina State University. What distinguishes Parker and Feinberg’s work, he says, is the fact that it isn’t a one-off prototype but a method that can be further developed for practical uses.