Canada’s latest Nobel laureate says computational resources are foundational
Alberta-based virologist Michael Houghton received the 2020 Nobel Prize in medicine for his work — along with two American scientists — in identifying the virus that causes hepatitis C. Since coming to Canada in 2010, he now uses WestGrid and the Compute Canada Federation’s resources for work on his HCv vaccine and on other work involving collaborators on many other diseases, including recent work on COVID-19.
Houghton, director of the University of Alberta’s Li Ka Shing Institute of Applied Virology, works with many experts on tackling such diseases as cancer, Alzheimer’s, non-alcoholic fatty liver disease, and COVID-19, and says computational science is essential to all the work the Institute is doing now. Set up as the translational arm of the Li Ka Shing Institute of Virology by institute director Lorne Tyrrell, the Applied Institute has an ambitious program of developing therapies, vaccines, and diagnostics for many major human diseases.
Houghton’s Nobel, which he shares with Harvey J. Alter and Charles M. Rice, dates back to 1989 when they discovered the virus and developed blood tests.
“It’s a very great honour, obviously,” says Houghton.
Their discovery has prevented millions of infections of a chronic liver disease that kills 400,000 each year.
At the time of the discovery, Houghton, who was born in Britain, was working for Chiron Corp., a California pharmaceutical company. He was head-hunted to Alberta as part of a government effort to attract top talent to Canadian universities. He joined the University of Alberta faculty in 2010 and has since been working on a vaccine for Hepatitis C, using computational science as a key component of his research.
“In general, WestGrid and the Compute Canada Federation have been an important part of our work on HCV [hepatitis C virus],” he says.
When he started drug discovery at the University of Alberta, he realized he and his team couldn’t compete with pharmaceutical companies, which have the resources to screen hundreds of thousands of pedigreed compounds in wet-lab assays. That’s when he decided computational science could become an important part of his work. He worked with colleague Jack Tuszynski, who has pioneered this approach at the University of Alberta for many years.
“I was very impressed with his work when I talked to him about it,” Houghton says. “I decided maybe we can be competitive in computational drug discovery, building on Jack’s work for the past 20 to 30 years.”
Using computational approaches have helped his team identify target structures to prevent protein-protein interactions, he says. And he uses Tuszynski’s work to try to predict the binding of small-molecule drugs that could interfere with that protein-protein interaction.
“That whole process has given us very promising drugs for a number of diseases,” he says. “Now that Hep C is curable, the big liver disease left is non-alcoholic fatty liver disease. A lot of that is caused by obesity, but a lot of it is not. You get fatty liver disease in thin people as well as obese people and in obese people, it’s hard to treat successfully with diet alone.”
He’s been working with Kamlesh Sahu, a research associate at the Institute, to try to find drugs that can inhibit the formation of fat in the liver. Sahu, along with Tyrrell and James Nieman, head chemist in the Institute, is doing the same kind of thing to try to improve known COVID inhibitors. Having synthetic and medicinal chemistry expertise is still essential for success, but computational drug discovery is emerging as another crucial technology.
Indeed, Houghton says computational science is “ushering in a whole new era of drug discovery” as it’s being used more often in new drug discovery and drug development. He and his colleagues are also using it to perfect drugs by allowing scientists to modify the compounds to get rid of the side effects of otherwise effective drugs.
“Another major area where the Compute Canada Federation has helped us once again is in the work of other scientists,” Houghton says. “Khaled Barakat, of the pharmacy school at the University of Alberta, is using computational science to find small-molecule inhibitors of the immune checkpoints for cancer therapy.”
Barakat and the University of Calgary’s Sergei Noskov also used computational science to try to predict which small-molecule drugs might block the cardiac-iron channel known as the hERG. It turns out that a lot of drugs bind to the hERG and cause heartbeat irregularities, some serious. Both scientists have published several papers on using computational methods to predict the interaction of drugs with the cardiac-iron channel to prevent drug-induced cardiotoxicity. Together with Houghton, Tyrrell, and Tuszynski, they started a company called Achlys Inc. to deliver these sophisticated tools to the pharmaceutical industry.
“It’s been wonderful to have Compute Canada Federation facilities available to our teams — and it can yield results very quickly,” he says. “Sometimes results come back in a few weeks instead of many months or even years. The systems are so formidable; they are really superb. You have to use some sophisticated software with the hardware, but our scientists have really benefited. When I came to Canada, it was already running and a lot of our young scientists used it. It’s not just for what we’re doing on Hep-C, but it’s relevant to COVID and many other disease targets.”
His current work on COVID-19, with the help of a grant from CIHR, is figuring out the best antigen to use in a vaccine and to explore how they might make antibodies more cross-neutralizing against different strains.
“So far, we’ve been using biological or wet-lab approaches, but just yesterday I was talking to a colleague and we decided we need to bring it to Dr. Sahu to see if he can help us predict the fine structure of sub-domains of the spike protein of COVID to help us design a better vaccine antigen. We are going to use the Compute Canada Federation technology for that.”