As the COVID-19 pandemic progresses, the hunt for antibodies (the protective proteins produced by the human body in response to an infection) has reached a fever pitch. Researchers continue to struggle to reliably detect SARS-CoV-2 antibodies in patients who have recovered from COVID-19; they remain uncertain how long those antibodies last; and, most importantly, they are working around the clock to artificially produce those antibodies in patients who have not yet been infected by COVID-19. Now, a researcher from the Catholic University of America is looking at reengineering existing antibodies to fight the novel coronavirus.

“Antibodies are important because they prevent infection and heal patients affected by diseases,” said Victor Padilla-Sanchez, a researcher at The Catholic University of America, in an interview with Faith Singer-Villalobos at the Texas Advanced Computing Center (TACC). “If we have antibodies, we are immune to disease, as long as they are in your system, you’re protected. If you don’t have antibodies, then infection proceeds and the pandemic continues.”

Padilla-Sanchez is looking to the past – specifically, SARS-CoV – to inform the future. Both SARS-CoV and its current successor share the notorious spike protein that enables the viruses to infect human cells – so Padilla-Sanchez decided to examine whether the specific antibodies that could fight off SARS-CoV (named 80R and m396) could be modified to do the same for the current pandemic.

Padilla-Sanchez conducted detailed simulations of 80R and m396 using the Rosetta molecular docking software suite on two supercomputers: Bridges at the Pittsburgh Supercomputer Center (PSC) and Stampede2 (10.7 Linpack petaflops) at TACC. Finally, he generated high-resolution visualizations of the results on TACC’s Frontera supercomputer (23.5 Linpack petaflops), which currently ranks as the 8th most powerful publicly ranked supercomputer on the Top500 list. 

“XSEDE resources were essential to this research,” Padilla-Sanchez said. “Frontera has great performance when importing a lot of big data. We’re usually able to look at just protein interactions, but with Frontera and Bridges, we were able to study full infection processes in the computer.”

The simulations showed that certain key sequences prevent 80R and m396 from binding to the virus – but that amino acid substitutions might be able to improve binding interactions.

“We’re at the initial steps of this now, and this is where I’m hoping my work might help,” Padilla-Sanchez said. “Understanding why 80R and m396 did not bind to the SARS-CoV-2 spike protein could pave the way to engineering new antibodies that are effective. Mutated versions of the 80r and m396 antibodies can be produced and administered as a therapeutic to fight the disease and prevent infection.”

“Now,” he said, “I need to prove it in the lab.”

To read the article from TACC’s Faith Singer-Villalobos discussing this research, click here.