Hutch team hunts for coronavirus antibodies

A team of scientists at Fred Hutchinson Cancer Research Center in Seattle is working to identify, from the blood of COVID-19 survivors, immune proteins that one day might be used to keep the virus from infecting others.

According to Fred Hutch immunologist Dr. Leo Stamatatos, the technique has pinpointed several immune proteins, called antibodies, that bind directly to the spike-like structures that dot the surface of the virus. In fact, those distinctive spikes, which are cone-shaped proteins coated with sugar molecules, make the virus appear under powerful microscopes to look as if it were crowned — hence the name, coronavirus.

Like certain drugs, these antibodies have the potential to grab and jam those spikes, which are key parts of the mechanism that the virus, SARS-CoV-2, uses to break into and infect human cells.

Stamatatos and his Fred Hutch colleagues, including Drs. Andrew McGuire and Marie Pancera, are now searching for antibodies to SARS-CoV-2 in blood samples being donated by recovered patients in Washington state and, in turn, hope to use them to either prevent or treat the disease.

“What we’re looking for is what a person’s effective immune response looks like in response to this infection,” McGuire said.

The hunt begins: Researchers reveal the spike structure

Their project began back in mid-February, after researchers led by Dr. Barney Graham of the National Institute of Allergy and Infectious Diseases and Dr. Jason McLellan of the University of Texas posted a model of the structure of the SARS-CoV-2 spike protein. This was an extraordinary feat, given that the virus was first sequenced in China only a month before. Their work was published March 13 in the journal Science.

Those authors, led by McLellan, shared the genetic sequences needed to generate copies of the spike with other researchers, including the Stamatatos, McGuire and Pancera team. Meanwhile, molecular biologist McGuire worked with colleagues at the University of Washington, led by Dr. David Veesler — an expert in coronavirus structures — to develop their own model of the spike. They published their results March 9 in the journal Cell.

Using the genetic sequence for the viral spike, the Fred Hutch team was able to manufacture millions of them. To these, they attached a protein, derived from luminescent jellyfish, that would glow under fluorescent light.

Next, they poured these glow-tagged spike probes into a pool of B cells harvested from patient samples. B cells are the body’s factories for antibodies. The researchers isolated every B cell in the sample and searched for those that could make antibodies against the spike.

They would know which cells could do so, because each B cell displays on its surface, like a sign on a storefront, a copy of the antibody it makes. If one of those spikes landed on a B cell with a matching antibody, it would stick to that cell. Now the researchers could know which B cells were promising: They glowed under the fluorescent light.

Out of those blood samples came about 100 candidate B cells, each making a unique antibody. The lab has since determined the genetic sequence of those antibodies and is now producing them for further testing to find those that do the best job of neutralizing the virus.

“Once you have the antibody, you can discover whether or not it neutralizes the virus, whether it cross-neutralizes another strain of coronavirus, whether it’s potent, and where it binds,” said Pancera, a structural biologist. “So, then you can do three things: You can use the antibody as a therapeutic, a prophylactic or you can use information about that antibody to help make a vaccine.”

Looking for answers, good news or bad

The efforts described are early in the phases of research needed to respond to the COVID-19 crisis. Under normal circumstances, it could take years to move from these laboratory experiments to the bedside. There is no guarantee that the antibodies will work in patients, or that they will lead to a coronavirus vaccine. Careful studies of safety and efficacy are essential.

But with the high stakes of this crisis, scientists are sifting quickly through the evidence of their experiments, looking for answers. Good news or bad, those answers will come faster than ever before.

If clinically useful neutralizing antibodies are identified and found to be effective, Stamatatos said they can be easily grown in large amounts in a laboratory, using long-established genetic engineering techniques. Eventually, the antibodies could be infused into patients infected by the virus, and the therapeutic potential of these lab-grown proteins could be assessed — perhaps in a matter of weeks.

Using manufactured antibodies, this approach would be a well-defined industrialized version of convalescent plasma, an emergency therapy being tested in China and the U.S., in which antibody-rich plasma purified from the blood of recovered COVID-19 patients is infused into those fighting it, in hopes of blocking the virus.

Another option would be to inject the lab-grown antibodies prophylactically, to protect people like health-care workers who are most at risk.

“When we get a really potent antibody, we could manufacture enough to give it to first responders and health professionals who treat COVID-19 patients,” Stamatatos said.

“The antibodies can stay in circulation in these people for many weeks, and if they get exposed to the virus during that time, the hope is the antibody will prevent them from getting infected and developing disease. A month or two later, they would get another shot of that same antibody,” he said.

Antibodies can help researchers design a coronavirus vaccine

Lab-grown antibodies can also serve as the basis for a coronavirus vaccine. Once the researchers determine which antibodies are best at neutralizing the virus, they can engineer small proteins that mimic those portions of the surface of the coronavirus spike that bind these neutralizing antibodies.

Injected into healthy people, a vaccine made from key bits of spike proteins could, in theory, stimulate a person’s immune system to make billions of identical antibodies when first exposed to the virus, blocking infection.

Stamatatos has spent much of his career designing experimental vaccines for HIV, which has been notoriously difficult because HIV mutates so rapidly, evading immune control. This coronavirus, on the other hand, is surprisingly stable for a microbe whose genetic blueprints are coded on error-prone RNA, rather than more dependable DNA. The hope is that a SARS-CoV-2 vaccine designed to elicit neutralizing antibodies might succeed where HIV vaccines have not.

“For this virus, once immunity kicks in, it will eliminate it,” he said.

Just how long that vaccine-based immunity will last is unknown. The virus will mutate and eventually evolve its way around the protective antibodies of a vaccine — just like influenza viruses do, but likely at a slower rate.

Unlike HIV and influenza, which are prone to rapid mutations that make those diseases difficult to block, coronaviruses carry a primitive gene-repair mechanism to fix errors that creep into their RNA code when they replicate. That lends the virus some genetic stability, which makes it a better target for drugs and vaccines.

Fred Hutch computational biologist Dr. Trevor Bedford has tracked the spread of the coronavirus since it first emerged. In a recent Twitter thread, he observed that SARS-CoV-2 mutates about once every 10 days, but is unlikely to evolve into a more pathogenic strain. Instead, the steady pace of these minor mutations is likely to make the current strain eventually resistant to whatever vaccine emerges, so it will have to be reformulated regularly, like the flu vaccine.

“My prediction is that we should see occasional mutations to the spike protein [of SARS-CoV-2] that allow the virus to partially escape from vaccines … but that this process will mostly take years rather than months,” Bedford said on Twitter.

Because vaccines are given to healthy people, they must first be evaluated for safety and effectiveness in a lengthy process that could take at least year.

The process for testing the use of manufactured antibodies to treat seriously ill COVID-19 patients, or to provide protection to health care workers, could be completed much sooner.