Uppsala chemists part of international COVID-19 collaboration

8 maj 2020

Jens Carlsson and Andreas Luttens in front of a computer model of the corona virus's protein-depleting enzyme. A drug molecule must fit perfectly with the corona virus enzyme to stop the virus.

The funding of a number of research projects has been fast-tracked in order to speed up the development of drugs and vaccines to combat COVID-19. Jens Carlsson and his doctoral student Andreas Luttens at the Department of Cell and Molecular Biology have recently received expeditious funding for their Fragment2Drug project in a collaboration with British researchers.

At the end of March, SciLifeLab, a national research infrastructure for molecular biosciences, issued a nation-wide call for proposals focusing on research initiatives to combat COVID-19, the disease caused by the novel coronavirus SARS-CoV-2. The Knut and Alice Wallenberg Foundation was behind this call, offering up to SEK 110 million in immediate funding for selected projects. Among those who applied for and received a grant were computational chemists Jens Carlsson and Andreas Luttens at the Department of Cell and Molecular Biology at Uppsala University. 

Jens Carlsson, Department of Cell
and Molecular Biology.
Photo: Mikael Wallerstedt 

“We received a grant that gives us the opportunity to collaborate with research groups from  many parts of the world with the goal of developing an effective drug faster than ever before. Without the grant and the coordination from SciLifeLab, we would never have come in contact with them,” says Jens Carlsson.

Their new research colleagues are part of the international team of scientists working in the COVID Moonshot collaboration project. The network also has access to the UK’s national synchrotron radiation light source – Diamond Light Source – which can be used to study viral proteins. There, scientists recently discovered what one of the virus’s enzymes looks like at the molecular level, which can be extremely helpful in the development of new drugs to treat COVID-19. 

Computer algorithms identify candidate molecules

Jens Carlsson’s research group is now performing computer simulations to identify molecules that will bind to the coronavirus enzyme, thereby blocking the virus from entering cells or producing new virus particles or virions. According to him, molecules that can stop the spread of the virus in this way have great potential for development into drugs.

Andreas Luttens, Department of
Cell and Molecular Biology.
Photo: Mark Harris

To identify molecules that can bind to the coronavirus’s enzymes, researchers are evaluating over 12 billion different molecules in their computer models. Access to the Swedish supercomputer centre makes it possible to search among as many molecules overnight as it would take a regular computer several years to complete. The molecules predicted by the models are then tested in experiments to see if they can stop the replication of the virus.

“Drug companies have millions of molecules in their physical libraries and they test all of them experimentally to see if any of them work. We try to do this faster by first testing them using computer models. Instead of testing millions of molecules in experiments, we can select just a few hundred to test. We hope to be able to use our models to predict which ones will work. It will speed up the drug development process,” says Jens Carlsson.

The image shows the surface of the coronavirus's protein-depleting enzyme. The colorful molecules are fragments that scientists are trying to develop to create a drug for covid-19.
Illustration: Jens Carlsson

Knowing how the molecular machinery of the novel coronavirus works, the research group can now select a suitable protein for their computer modelling.  The protein they have chosen as the target for the drug is an enzyme in the virus that breaks down proteins – a protease. Researchers have previously managed to develop drugs to fight AIDS and Hepatitis C that block the activity of proteases. However, the protease present in the novel coronavirus is of a new type, thus demanding new strategies. According to Jens Carlsson, it is also essential to do test tube experiments to determine whether a drug prototype behaves in the way that the computer modelling predicts, that is, that it binds to the protease and blocks replication machinery of the virus.

Virus tests at the Biomedical Centre

Experiments will be carried out by Professor Helena Danielson at the Department of Chemistry – BMC in collaboration with the Drug Discovery and Development platform at SciLifeLab in Uppsala. But although scientists agree that the experiments are feasible, no one in the world has ever done them yet. And if testing is not up and running within four weeks, there is the option of using the equipment at the University of Oxford. Experiments are already in progress there, and their results have also been shared with the Uppsala researchers.

Helena Danielson, Department of Chemistry –
BMC. Photo: Mikael Wallerstedt

To avoid duplication of work and thus the loss of valuable time, the research teams in Sweden and the UK are in close communication with each other.

“During our first Zoom meeting, we talked a lot about the best way of working together,” says Jens Carlsson. “We shared experiences and tried to find out what the Diamond team are testing. Because when you do real experiments, you need to identify the right conditions, and we need to get it to work as fast as possible.”

Another challenge is to find manufacturers for the molecules that Andreas Luttens has identified using supercomputers. It is hoped that a company in Ukraine will be able to rapidly assist in the manufacturing. The grant from SciLifeLab will be sufficient to purchase nearly 500 molecules – a substantial budget for the chemists, who typically use around 50 molecules per project.

Drugs are built bit by bit

Jens Carlsson’s group has not studied coronaviruses before – their focus has instead been on neurological disorders and cancer. However, the same basic strategies can be applied to research into other diseases, including those caused by new viruses. Helena Danielson has already worked with proteases from HIV and Hepatitis C, in the same early stages of development and with similar challenges as with SARS-CoV-2. The Uppsala researchers can now combine their knowledge in the pursuit of a treatment drug.

“We use a fragment-based approach in which we build the drug one step at a time. The starting point is a very small molecule that binds weakly, then we build on it so it becomes a larger molecule with optimised properties. This is something we have been working on for three years. It will be a test of whether our new methods actually work," says Jens Carlsson.

However, the development of a new drug usually takes ten years. So many research groups are trying to use drugs that are already approved – this is known as repurposing. But the odds of finding a drug that works and is safe using the repurposing strategy are very poor, according to Jens Carlsson.

“Ultimately, this joint effort to look at how the virus functions will contribute to the development of a treatment drug. And that is why a broad course of action is beneficial. We will encounter similar problems again, so we see this as an opportunity to develop strategies that we can use in other areas as well.”