Meanwhile, employees of biotech startup Neoleukin Therapeutics used their free time to invent a new molecule that blocks coronavirus from infecting cells. It can be delivered in a nasal spray, and testing in hamsters shows that it protects the animals from getting sick.
“We saw an opportunity to make a difference and make good use of our time,” said Daniel-Adriano Silva, Vice President Head of Research for Neoleukin. “We are hugely enthusiastic and really excited about what we did here.”
Neoleukin (NASDAQ:NLTX) was formed in January 2019 with a mission to create designer molecules against cancer cells. A spinout company of the University of Washington’s Institute for Protein Design, Neoleukin uses computer modeling technology to design protein fragments, called peptides, that will fold into a desired shape and also bind tightly to a chosen target.
Called “de novo protein design,” this technique represents a revolutionary shift in the way drugs are developed. Rather than looking for molecules in nature that can treat disease, and then trying to optimize them for safety, efficacy, and mass-production, de novo design allows researchers to create bespoke molecules to perform a specific function.
While the company was launched to create cancer-fighting molecules, the strategy works equally well against infectious disease. The designer antiviral was created in record time, and is the most complicated working protein ever designed and built by humans from scratch.
“Here is a potential way to change the course of this pandemic,” said Silva. “I believe that if we establish this strategy as a viable technology to deal with this pandemic, it could be used to impact the course of other outbreaks in the future.”
Protein binding relies on shape, like 3D puzzle pieces that fit neatly together, but also on chemical forces. Imagine the puzzle pieces have magnets inside. If the charges don’t align, the pieces will not click into place, even if the physical surfaces match up.
The Neoleukin scientists saw the potential to apply their protein-modeling technology to the new coronavirus. The spike protein on the surface of the virus binds to a protein on the cell surface called ACE2. To stop the virus from getting into the cell, you just need to create something that blocks the part of the spike protein that binds ACE2.
A conventional approach to deflect the virus is to make monoclonal antibodies that glom onto the viral protein. The experimental therapy given to President Trump is an example of a monoclonal antibody, as is the recently approved drug bamlanivimab. Monoclonal antibodies are great at fusing tightly to their targets, but by molecular standards, they are gigantic. They stick to more parts of the virus than just the piece that links up with ACE2.
This means that the virus can mutate in ways that make it less recognizable to the drug, but that don’t hamper its ability to hook up with ACE2 and infect the cell. This is called “mutational escape,” and it explains why some drugs that work like gangbusters at first can stop being effective after a time.
Neoleukin designed a tiny peptide that exactly matches the part of the ACE2 protein that binds the coronavirus spike protein. Any change that weakens the binding between the peptide and the virus would also weaken the virus’s ability to infect the cell. And when the peptide is present, the virus spike protein
“All the interactions that are happening are the same as the virus’s interaction with ACE2,” said Thomas Linsky, Neoleukin’s Director of Protein Design. “That makes it harder for the virus to be able to escape.”
The design process goes something like this. Based on the published protein sequence for coronavirus spike protein, the researchers use their protein design software to create mimics that will bind the appropriate part of the spike protein. “For this project we designed over 35,000 sequences on the computer,” said Linsky.
Of those thousands of candidates, 198 peptides were evaluated in the lab, to measure how well they could bind to the spike protein under biological conditions. “We put a lot of thought into how we select the best possible designs that have the best chance of working,” Linsky said. Once the top contenders are identified, a great deal of optimization takes place. Through a process called “directed evolution,” the researchers make many similar peptides, each slightly different, and select for how well they bind to the spike protein. An effective antiviral needs to be able to attract the spike protein, even in the presence of ACE2, preventing it from binding the protein on the cell surface and infecting the cell.
“Once we figured out that it actually worked, we went to the live virus and confirmed that it was a very effective way to inhibit the virus,” said Silva. Not only did it prevent the virus from infecting human cells grown in a dish, but it worked in hamsters in a nasal spray. The treated hamsters received a lethal dose of SARS-CoV-2, and all survived.
For now, the company is still evaluating the logistics of how to mass-produce the molecule. They have no plan as yet to develop it for distribution, but potentially could partner with a manufacturer, if such an arrangement were feasible. “As a company that is focused on making proteins for cancer and autoimmune disease, this is a little outside our field of focus,” said Jonathan Drachman, Neoleukin’s CEO. “It’s exciting, but we have to evaluate whether it makes sense for us to try to develop it.”