This is your brain on a chip. This is your liver on a slide. This is your body in a supercomputer. Any questions?
It’s a bit more complicated than that, but recently scientists have provided a sneak preview of the future of biomedicine with a range of projects seeking to assemble virtual humans — or parts of them — on computers and “labs on a chip.” Someday, the descendants of these sophisticated new programs and devices could serve as our stand-ins for clinical tests on drugs, cosmetics and toxic compounds.
“I would predict that this century is going to be dominated by our ability to handle biomedical problems in a computational domain,” said Peter Coveney, director of the Centre for Computational Science at University College London.
The increasing ability of computers and biochips to mimic brain chemistry, internal organs, and the interactions between drugs and viruses such as HIV could help reduce the reliance on animal testing to understand the potency and side effects of pharmaceuticals. A more informed leap between experiments on dish-grown cells and lab animals, in turn, could lead to a better drug development process. And eventually, the technology could usher in a new era of personalized medicine in which rapid tests tell doctors which treatments have the best chances of success for individual patients.
Billions of neurons
But first, researchers have more than a little tinkering to do, especially given that the brain boasts tens of billions of neurons and that copying its chemistry would require replicating the thousands of connections emanating from every cell. Andre Levchenko, an associate professor of biomedical engineering and an affiliated researcher with the Institute for NanoBioTechnology at Johns Hopkins University in Baltimore, said the monumental task is nonetheless important for understanding Alzheimer’s, Parkinson’s, and other conditions that affect the brain.
“After a stroke, a huge part of the brain tissue may become disabled,” Levchenko said. “If one understands how this network is put together in the first place, it’s possible to predict what should be done to put the tissue back into place after the trauma.”
He and colleagues at Johns Hopkins have begun tackling the problem by placing neurons onto a plastic-like “lab on a chip” and then introducing two signals in the form of liquid chemicals. By controlling how the fluids flow through the multilayered chip with tiny valves and channels, the researchers can adjust how and when the chemical cues reach the cells.
The neurons haven’t always moved in expected ways after receiving their cues. But Levchenko said the unexpected results, published last month in the British journal Lab on a Chip, allowed his lab to test new possibilities. “If we don’t get surprised, then it means that we already know everything,” he said.
Although the device was initially designed for more basic research, he said it could provide a “wonderful tool” for screening how potential drugs affect the cells’ responses. Similarly, he said scientists could use the chips to engineer basic brain tissue or begin exploring more complicated interactions among different cell types such as neurons and muscle cells.
“I think there’s a quantum leap here by changing not only the technology but also the philosophy of how these experiments are done,” Levchenko said. “It may be difficult to control what happens inside the cells, but in this day and age, we should be able to control exceedingly well what happens outside of the cell.”
Mimicking the body's environment
Jonathan Dordick, a professor of chemical and biological engineering at Rensselaer Polytechnic Institute in Troy, N.Y., has taken a different tack by arranging human cells into thousands of three-dimensional spots on a microscope slide, or chip. “Within those spots, we’re essentially mimicking what the environment is within the human body,” he said.
One device, known as MetaChip, mimics how the liver processes toxic compounds. A newer technology, called the DataChip, provides more realistic arrangements of how liver, kidney and breast tissue cells grow, and it is being expanded to include the cardiac and central nervous systems. “Obviously it’s a long way from the cell to what’s happening within the human body, but you’ve got to start somewhere,” Dordick said. “If something kills the cell or makes it unhappy, it’s likely to be bad for the body.”
The liver, for instance, rids the body of toxins by metabolizing, or processing, them. But sometimes, the organ breaks down a compound into a subunit that can be equally bad or even worse. “So it’s important to know what the body is likely to do with a compound,” Dordick said. For the MetaChip, his team included about a dozen of the most important processing enzymes.
Instead of applying the pharmaceutical industry’s one-size-fits-all strategy of the past, Dordick said the approach he describes in a recent issue of the Proceedings of the National Academy of Sciences recognizes that humans can process drugs quite differently. “We’re trying to predict what the human response would be to any molecule that’s out there,” he said, whether for existing drugs or ones still in the pipeline.
The technology that could help make such predictions possible has been licensed to a small start-up company co-founded by Dordick, called Solidus Biosciences Inc. By mid-2009, he aims to have chips available commercially to other companies as self-contained testing kits.
In the pharmaceutical industry, he said, “obviously we’re not going to get rid of animal testing entirely.” Devices that look at what happens within the body as a whole could eventually take over many roles played by the lab rat and other animals, however. And with a sizeable percentage of clinically tested compounds ultimately failing because of their toxicity, Dordick said a new approach could help weed out faulty candidates much earlier in the development process. “No matter if you want to replace animal testing for ethical reasons or the expense, there is a scientifically valid concern that animal testing isn’t predictive of what happens in a human,” he said.
Of course, the same could be said for a collection of cells on a chip, but as the technology improves, Dordick is convinced the range of simulations being worked on now will increasingly approach reality. The innovation is being pushed by other looming concerns: European Union members face a 2009 ban on animal testing for all cosmetic ingredients. And a separate law being phased in over the next decade requires EU manufacturers and importers to assess the toxicity of chemicals used in any significant quantity — a regulation that Dordick said is leaving many companies scrambling for cost-effective tests.
Tapping national power grids
University College London’s Coveney has adopted a third tactic toward mimicking the body by tapping the national power grids of both the United Kingdom and the United States. His team’s simulation of how the anti-HIV drug saquinavir blocks the protein HIV-1 protease, published earlier this year in the Journal of the American Chemical Society, predicts how the medicine might bind to three drug-resistant mutants encountered by doctors of AIDS patients.
The high-tech exercise allowed his team to rank the viral mutants in terms of their drug vulnerability, an analysis that could be crucial for picking the right medicine for desperately ill patients. “You also need to be able to do it fast enough so it’s going to be relevant to a clinic and a patient,” Coveney said, noting that his lab’s exercise took only two weeks. For those infected with HIV, doctors already can gather some specific genetic information about the rapidly mutating virus, but choosing which of nine possible protease inhibitor drugs to prescribe is still done mainly by trial and error.
As for computing power, Coveney’s project required the not-inconsiderable equivalent of what would be needed for a long-range weather forecast.
“Once you open the door, there are quite a few questions that people would have to address,” he conceded. With the restricted availability of supercomputers, for example, who would decide when the technology could be used for specific life-or-death medical cases? “And what level of demand would there be? We’re literally just now thinking it through,” he said. “We’re at, as they say, ground zero, in that debate.”
In the meantime, as part of the joint U.K.-U.S. GENIUS project accessing a portion of the two countries’ national grids, Coveney is also using the geometry of blood vessels taken from MRI scans to visualize how blood flows through the human brain — a simulation that could be incredibly useful for a neurosurgeon planning her next surgery.
Coveney’s research has likewise provided an early glimpse at the European Commission’s Virtual Physiological Human, a broad initiative to simulate the workings of the human body all the way down to the molecular level. Digitally stitching it together will be no small feat, he said. But the final product could be well worth the trouble.