You are here
Screening drugs to save lives
A cross-border collaboration has yielded a computer model that will screen drugs for heart-related side effects.
One of the leading side-effects of many drugs is a dangerous cardiac arrhythmia and this test will allow researchers to screen drugs for that issue. Researchers at the University of Calgary and the University of California Davis, led by professors Sergei Noskov, Colleen E. Clancy and Igor Vorobyov, have used the drug’s chemical formula to model how drugs bind to a protein known as hERG, which conducts potassium carbon at the very end of heartbeat, thereby establishing the heartbeat’s stability.
“It’s not a huge current, but it’s very important to start a new one,” says Sergei Noskov, a professor and associate head of research in biochemistry at the University of Calgary’s Centre for Molecular Simulations. “If you’re missing normal protein function, the new beat starts at the wrong time, which is a problem, because all chambers in the heart have to conduct in a very well controlled manner. Even some of the drugs that are prescribed for heart issues cause a disturbance.”
The research, published in Circulation Research journal and co-authored with his California colleagues, provides a platform that can be used to assess the drugs preclinically to see whether or not they will cause arrhythmia.
“That’s a huge issue because right now, Health Canada and the U.S Food and Drug Administration require a lot of testing and it costs taxpayers a lot of money,” Noskov says. “Even in spite of all the efforts, now and then we hear about drugs, such as a Hep C drug that met all the safety requirements in pre-clinical stage and the company developed it and then it produced adverse side-effects and long-lasting damages in human clinical trials. It was an unintended consequence of drug development.”
His goal, therefore, was to “detoxify the development.” Next, he says, the cross-border team will try to develop a collaboration that provide some rational suggestions about how to provide safer compounds.
Noskov’s team was instrumental in developing the model, which can predict from the chemical structure of a drug whether it will impact heart rhythm. The researchers validated the model with ECGs of patients who were taking two drugs known to interact with hERG channels — one with that was shown to be safe and another known to increase arrhythmias.
Noskov’s team used high-powered computing to develop the model, using the resources of Compute Canada’s Western Canada regional partner, WestGrid.
“Compute Canada resources were used extensively to develop the drug models,” he says. “To form molecular simulations of how the drug moves, you really have to perform lots of simulations of the drug itself so Compute Canada resources were used to develop drug models to test them.”
Compute Canada and WestGrid resources provided much-needed computational support to the ground-work, illuminating drug access mechanisms and refining the protein structure itself. The project, supported by the Canadian super-computer consortium, led to two publications on hERG-drug interactions and transport mechanisms in Molecular Pharmacology and Proceedings of National Academy of Sciences — both American journals, and published in 2019 and 2020, respectively.
He also used Compute Canada resources to calculate binding energies for the drugs and populate some elements of the study, though not all of them because some were done in the U.S. thanks to the study’s multi-site nature.
He says the study could technically have been done without Compute Canada resources because the University of Calgary has good high-powered computing resources of its own, but he says those resources are always oversubscribed and it would have taken a lot longer to make these findings.
“Without Compute Canada, this work probably would have taken five to six years. Instead, it only took two years,” he says. “Compute Canada resources and the resources we get access to in the U.S. simply changed the whole scope of the study. Now we can do more, be more realistic, more accurate and we can afford to do simulations.”
In addition, waiting five to six years wouldn’t have worked because of the timeline of funding from National Institutes of Health and Canadian Institutes of Health Research.