Utilizing Focused Computer Modeling to Quicken the Discovery of Antiviral Drugs

There is an urgent need for effective medications against viral illnesses like COVID-19 today and in the future. The advent of viral mutations and undiscovered viruses may test the boundaries of vaccinations.

DZIF scientist and bioinformatician Andreas Dräger from the University of Tübingen is working on a computer-based method that can help to accelerate the time-consuming identification and development of antiviral agents.

The research group led by Dräger has now developed a model to identify additional host cell targets that permit suppressing SARS-CoV-2 replication. This model uses a novel analysis technique that is applicable to any virus and host cell type.

“Efficient pandemic preparedness requires new, broadly effective antiviral drugs against which the viruses cannot quickly develop resistance,” explains Dräger, junior professor at the University of Tübingen and member of the Tübingen Cluster of Excellence Controlling Microbes to Fight Infections (CMFI). “But drug development takes too much precious time, which is urgently needed in an emergency.”

Dräger wants to remedy this situation through computer modeling. The Tübingen research team in 2021 discovered a human enzyme in the guanylate kinase 1 model that is essential for virus replication but that can be turned off without harming cells.

The bioinformatician and his colleagues have now created a new model to evaluate the efficacy of their targets. “Through an improved analysis technique, we can now specifically model viral infection in many different types of tissue,” explains Nantia Leonidou, first author of the current study.

Our models could represent a paradigm shift in drug development and accelerate the preclinical phase, The methods are fully transferable to any virus and host cell type and are also commercially viable.

Nantia Leonidou

Observing host metabolism after viral infection in the model

By simulating SARS-CoV-2 infection in bronchial epithelial cells using their integrated systems biology model, the researchers can pinpoint host-based metabolic pathways that can be blocked to stop virus replication.

“If you know the composition of a virus, you can run different scenarios and see how the biochemical reactions in host cells change during viral infection,” Dräger says. The team developed high-quality software to simulate an infection in a cell-type-specific manner.

New targets identified

The study team verified the previously identified target, guanylate kinase 1, using the model on a different cell type, and found additional new biochemical targets with exceptional antiviral properties.

CTP synthase 1 was the most promising new hit. In the model, inhibiting this enzyme decreased viral multiplication by 62% while having no negative effects on the survival of the human host cells.

The structure of genetic material, which needs the same building elements in both the virus and the host cells, is closely related to both target molecules. Dräger’s team believes these findings provide a crucial basis for accelerating the development of viral inhibitors.

“Our models could represent a paradigm shift in drug development and accelerate the preclinical phase,” emphasizes Nantia Leonidou, adding, “The methods are fully transferable to any virus and host cell type and are also commercially viable.”

The study is published in PLOS Computational Biology. Dräger’s group now plans to apply their methods to other viruses. The safety, toxicity, and efficacy of the first inhibitors for their discovered enzymes will be examined in animal models.

Topic : Article