Antibody Cocktails Against Future COVID Variants, Thanks to Global Consortium CoVIC

“Give us your antibodies” might be the mantra of the The CoVIC Consortium, a global group of eclectic experts who introduce a “framework for antibody cocktail selection” in the journal Science. They haven’t just predicted which antibodies, alone or in pairs, can “neutralize” viral variants, including some that haven’t even evolved yet, but have actually tested the tango between antibodies and their targets. From 56 labs on 4 continents, CoVIC has amassed more than 350 monoclonal antibodies against the spike protein with which SARS-CoV-2 latches onto and enters our cells.

As I read the paper, I envisioned a war room, where strategists scrutinize giant, detailed maps as they move symbols of troops and weapons into position, planning assaults from different directions.

Antibodies 101

Antibodies are proteins that the B cells of the immune system produce upon encountering something foreign, like a bacterium or virus. They bind, highly specifically, to molecules on the pathogen’s surface, called antigens.

A natural antibody response is “polyclonal.” Like the Indian parable of the blind men and the elephant in which each man touches a different part of the animal and then describes it based on limited information, each antibody of a polyclonal response binds to a different part of the pathogen.

Upon infection, B cells rapidly mature into plasma cells, which synthesize and release about 2,000 antibodies per second. I have a special fondness for the plasma cell because it was my “unknown” in histology lab in grad school, easily recognizable by its gigantic Golgi body, a clear oval in the cell where those antibodies build up. Our B cells are able to muster a defense against nearly anything we encounter thanks to clusters of immunity genes that mix and match to an astonishing degree.

But not all antibodies are equal, in terms of protection. Only some attack the pathogen in a way that keeps it out of our cells – those are the neutralizing antibodies. For SARS-CoV-2, neutralizing antibodies target the spikes. Vaccines program our cells to produce and release viral spikes, and the B cells go to work. People who’ve had COVID have a broader antibody response than those who haven’t because their natural antibodies bind many parts of the virus, not just the spikes that vaccines target.

If the body’s deployment of polyclonal antibodies is like attacking a city from the ground, sea, and air, then a taking a monoclonal antibody (mAb)-based drug is more like a drone, zeroing in on a specific target. For many years, separating individual antibody types from a natural polyclonal response was daunting. Then in 1975, British researchers Cesar Milstein and George Köhler devised a way to mass-produce single, aka monoclonal, antibodies (mAbs), for which they won the Nobel Prize nine years later.

The original recipe for monoclonals was complex: injecting a mouse with a sheep’s red blood cells to activate immunity, isolating a single B cell from the mouse’s spleen, then fusing it with a cancerous white blood cell, also from a mouse. The fused cell spewed a single antibody type, continuously. The doctored cell, called a hybridoma, brilliantly teamed the specificity of the B cell with the immortality of the cancer cell. Monoclonals have found their way into dozens of products, from tests to detect pregnancy and turf grass disease to treating cancers.

Today, mAbs are collected from “humanized” mice or derived from the blood plasma of convalescent patients. Artificial intelligence and machine learning complement these approaches by predicting possible antibody structures that nature might not yet have coaxed into existence.

To find out if an antibody disables a virus enough to keep it out of our cells, researchers use a “pseudovirus neutralization assay.” The test inserts genetic instructions for the viral spike into a safe-to-work-with lentivirus, replacing part of its surface to resemble that of SARS-CoV-2. It’s a viral sheep-in-wolf’s-clothing. Then the altered viruses bathe human cells growing in culture that bear ACE2 receptors. The neutralization assay is also used to evaluate vaccines and antivirals.

Roots in the Ebola Fight

CoVIC is modeled after the Viral Hemorrhagic Fever Immunotherapeutic Consortium (VIC), which NIH began in 2014 to develop antibody-based treatments for Ebola virus disease and Lassa fever. Researchers from around the world collaborated to identify exactly how specific antibodies, and pairs of antibodies, neutralize the viruses – which can rapidly reduce a human body to a pool of blood. They compiled a panel of nearly 250 mAbs against the Ebola virus, some of which are now anti-Ebola drugs.

According to the CoVIC statement of purpose:

“The VIC showed the power of collaboration to advance therapeutics to the clinic, and offered a framework by which individual investigators could race to therapeutic discovery while simultaneously contributing to the greater knowledge base for the future. We can apply the same approach to finding effective antibody-based treatments for COVID-19.”

Like in the Ebola effort, CoVIC members include structural biologists who probe molecular interactions; virologists; immunologists; clinicians; and public health practitioners, coming from academic institutions, government agencies, and industry settings.

COVID, with its vast geographic reach, presents an even more compelling target than Ebola. Researchers must identify which mAbs are neutralizing, but also which neutralizing antibodies induce the most lasting effects, called “durability,” in the host. An antibody level that signals a long-lasting effect is said to provide a “correlate of protection.” That can take years to figure out; in the meantime, high and sustained neutralizing antibody levels are stand-ins for durability.

The consortium is using the neutralization assay to identify antibodies, and pairs of antibodies, which disarm the viral spikes. They’re also identifying the assays that best predict which antibodies signal correlates of protection.

Knowing the Enemy

Like snippets of intelligence that military minds use to plan an invasion, the more-than-a-million sequenced SARS-CoV-2 genomes reveal how the virus is evolving. Our terminology can barely keep up.

At first we identified new variants
https://dnascience.plos.org/2021/08/05/scary-variants-and-vaccine-hesitancy-set-up-a-perfect-storm-for-the-virus/ simply by where they were first detected: the UK, South Africa, Brazil. By the time that delta debuted in India, researchers realized that overlapping sets of mutations comprise the geographically-defined variants. Now we know that certain mutations stand out as responsible for https://dnascience.plos.org/2021/01/21/mutants-come-to-saratoga-covid-new-variants-explained/ increased transmissibility from person-to-person, promoting transfer from other animals to us (“spillovers,” like from bats), and from us to them (“spillbacks,” like to minks).

The names of the mutations can sound like mumbo jumbo jargon to those unfamiliar with flitting from the languages of genes to proteins. For example, mutation V367F designates the amino acid in the spike that’s been replaced. V stands for the amino acid valine and F for phenylalanine, at position #367 in the chain of 1273 amino acids of the spike protein (the inspiration for the original name of the Moderna vaccine, mRNA-1273). The rules of the genetic code dictate the exact change of a single DNA base that switches the mRNA instructions for valine to phenylalanine. And that alters the spike in a way that makes it bind more tenaciously to our ACE2 receptors.

Only mutations that sound like a word sneak into the media, like “eek.” Technically “E484K,” eek is an “escape” mutation that can reinfect people and first showed up in South Africa and Brazil, then flitted its way into variants all over.

Engineering Antibody Cocktails

The antibodies and analytical tools pouring in from the consortium members have built “a framework for antibody cocktail selection.” Most target the “receptor binding domain,” or RBD, the exact place where a spike meets the ACE2 receptors, especially in the lungs. Each spike consists of an S1 portion that attaches and a smaller S2 portion that drags the virus inside the cell. And the spikes poke out of the virus in trios.

The consortium is focusing on 186 mAbs that hit the RBD, using the pseudovirus neutralization assay to figure out how the antibodies disable the virus. Artificial intelligence groups the 186 into 7 “RBD communities,” and within them, into functionally relevant subgroups, or “bins,” like war strategists analyzing the enemy’s weapons and protections.

And gradually, a battle map is emerging of how mAbs can be deployed to attack the viral Achilles heel, its RBD.

Individual antibodies bind the RBD from the top or from the bottom, from the inside or outside. Pairs of antibodies disable single spikes, fuse two spikes of the triplets, and target and take out multiple mutations, like bombing buildings along a street. Many neutralizing antibodies crosslink the spikes, tilting and then toppling them, blocking viruses from entering our cells.

But it’s a race. As mutations accrue, our antibody weapons lose their oomph. “Constellations of multiple mutations in the RBM” lead to an “almost complete loss of neutralization activity,” according to the consortium. However, the researchers are devising variant-resistant cocktails that will not only protect against SARS-CoV-2, but by reaching beyond the spike to more shared areas of the viral anatomy, provide weapons against future coronaviruses.

Conclude the consortium members in the Science article:

“Taken together, the analysis presented here, made possible by broad participation of a few hundred therapeutic candidates in a global study, offers a detailed structural and competitive landscape of key antibody binding sites on Spike. The results of this effort can be used to predict and interpret effects of VOCs (variants of concern), and for strategic selection of durable therapeutics and cocktails against emerging variants.”