Early in the pandemic, vaccination or a bout with Covid-19 seemed to ward off the risk of another infection. But now, new viral variants are increasingly able to dodge that hard-earned protection. Keeping track of those variants and how they escape immune protection is an exhausting game, one that scientists would like to squelch with a new type of vaccine the virus hasn’t managed to out-evolve.
Scientists have tried several routes to attack the problem. The narrowest starts with the existing Covid mRNA vaccines and seeks to create updated boosters that target the virus’s most recent variants, an effort that drugmakers Moderna and Pfizer are attempting with Omicron’s progeny. The broadest, most ambitious route is to invent a vaccine that would target the entire coronavirus family, including the merbecoviruses that cause MERS, the embecoviruses responsible for ordinary colds, and the sarbecovirus subgenus that gave rise to both Covid and the original SARS virus that broke out in 2002.
But there’s a middle path: a vaccine that would attack just the sarbecoviruses, meaning the Covid virus and all of its future offspring, as well as any new SARS-CoV siblings that might appear in the future. This pipeline already has several candidates; some have been tested in primates or in mice, and one is undergoing a small clinical trial for people. All exploit commonalities shared by sarbecoviruses that could be used to combat their entire lineage.
“If you have a way to target these parts that are very conserved, you might have a way of targeting all of these sarbecoviruses,” says Alex Cohen, a postdoctoral researcher at Caltech who is developing this kind of vaccine. Ideally, he adds, this encompassing protection could be achieved with “one type of vaccination, or one type of immunization.”
Here’s a look at some of the candidates that are being developed.
Mosaic Nanoparticle Vaccines
Cohen works in Pamela Bjorkman’s laboratory in Caltech’s department of biology and biological engineering, which recently published a paper in Science on their candidate, showing that it demonstrated protection in monkeys and mice against multiple sarbecovirus strains. Theirs is a mosaic nanoparticle-based vaccine, which means it is built on a tiny, cage-like protein ball.
Their idea is to train the immune system to attack a target that many sarbecoviruses have in common. The Caltech lab chose a part of Covid’s famous spike protein called a receptor binding domain (RBD), which helps the virus enter and infect a host cell. RBDs are often evolutionarily conserved among different sarbecoviruses, meaning that although some regions of the binding site may mutate as new variants emerge, others stay the same. (As a hypothetical example, the Delta and Omicron variants would have similar RBDs, but a few differences as well.) This similarity creates an opportunity: If you can encourage the body to generate the antibodies that target those shared regions, they can protect against many different variants instead of just one.
Bjorkman’s team came up with this plan by studying antibodies from patients who were previously infected with Covid and analyzing where those antibodies would bind on the spike protein’s RBD. Bjorkman pulls out a model of the spike protein that is about the size of her head (in other words: very not to scale). “Early on, there were all these potent neutralizing antibodies that people isolated from infected people, and they blocked receptor binding,” she says, pointing to a region at the tips of the RBD. “But as variants came, they no longer worked.”
Most PopularThe End of Airbnb in New YorkBusiness
Her team realized that those early antibodies that had once seemed so powerful would bind to the outermost region of the RBD. These sites were effective targets for attacking the earliest versions of the virus. But those areas mutated over time. Once they did, it was harder for the antibodies to grab on to them and neutralize the virus.
Other rarer antibodies, however, could bind to a harder-to-reach area that was not as easily mutated. Bjorkman points to a part of the RBD that is closer to the middle of the spike protein than the tips, indicating where those special antibodies bind. “These are the antibodies we really want, because the RBDs should stay conserved among the sarbecoviruses and among any variant that could ever arise of SARS-CoV-2,” she says. The task for their vaccine would be to prompt the immune system to create antibodies that could latch on to those shared sites.
The team’s first step was to turn their nanoparticle into a kind of template that would train the immune system to make those antibodies. They dunked a protein nanoparticle shell into a mixture of eight different RBDs, which stuck to its surface—kind of like coating a sticky candy apple with different nuts. Because “there’s no reason for them to go to any particular place,” Bjorkman says, the end product was a nanoparticle with a random assortment of different RBDs on its surface. (Hence the “mosaic” in “mosaic nanoparticle vaccine.”)
When injected into an animal, the B cells in the animal’s immune system, which are in charge of producing protective antibodies, would start to make ones that attack these binding sites. Should the animal later encounter the actual version of the virus, its antibodies would know to glom onto these sites, stopping the virus from entering cells.
You might think this eight-RBD approach would result in antibodies that are designed to target only eight different kinds of binding sites. But the researchers took advantage of a quirk in the shape of the antibodies: They are two-armed and shaped like the letter Y. Instead of binding with one arm to a region specific to one RBD type, they can be designed to bind with both arms to conserved regions of two adjacent sites. That means that instead of gumming up only eight specific sarbecoviruses RBDs, they can theoretically attach to any with those conserved regions.
Most PopularThe End of Airbnb in New YorkBusiness
First the scientists tested their vaccine in mice, which were split into groups of six. Two of those groups were immunized with the mosaic nanoparticle, then each group was exposed to either Covid’s Beta variant or SARS-CoV-1, the first SARS virus from 2002. All 12 of the vaccinated mice survived. In contrast, most of the unvaccinated mice exposed to either virus lost weight and died.
Next, the team ran a similar experiment with macaque monkeys, divided into groups of four. Two of the groups were immunized by being injected three times with the mosaic nanoparticle. Then, about a month after the third dose, the animals were exposed to either Covid’s Delta variant or the original SARS virus. None of the vaccinated monkeys became infected with either sarbecovirus type, although three out of four monkeys in the Delta control group showed infection, and all the monkeys in the SARS control group did.
What’s significant is that in the experiment with monkeys, neither the original SARS nor Delta RBDs were included on the mosaic nanoparticle. To the team, this indicated that the antibodies generated after the inoculation targeted viral versions that the vaccine had not explicitly been designed to immunize against—and that it was useful against an array of sarbecoviruses. “The animals elicited a pretty consistent response where their antibodies were pretty much cross-reactive to every coronavirus we tested against, including those that were not present on the particle,” Cohen says.
Other Nanoparticle Contenders
These findings add the mosaic nanoparticle to a running list of RBD—or more broadly, spike protein-based—vaccines that have been created by different academic groups around the world. One candidate being developed by scientists at the University of Washington has been tested in mice, and another is currently in Phase 1 clinical trials at the Walter Reed Army Institute of Research. Another vaccine poised to enter human clinical trials is being developed by biologist Kevin Saunders and colleagues at the Duke Human Vaccine Institute, who published a paper describing their work in Nature in June 2021, and circulated an additional preprint in January 2022.
Like Bjorkman’s group, Saunders’ had noticed that the antibodies that were protective against multiple strains of sarbecoviruses targeted the innermost end of the RBD—and that these antibodies, among others, could be generated through immunization with their nanoparticle. But unlike the eight-RBD mosaic nanoparticle from the Caltech team, this version relies on just one RBD type from the original Covid virus. The nanoparticle is different too; it is based on a ferritin (a protein that stores iron) shell derived from Helicobacter pylori bacteria. (Saunders points out that ferritin nanoparticles are already used in flu vaccines, making it a “nanoparticle platform with some clinical experience.”)
In their 2021 paper, they also tested on monkeys. They found that in macaques, their vaccine generated antibodies that could protect against the original Covid virus. Then in the 2022 preprint, which has not yet been published or peer-reviewed, the scientists challenged more immunized macaques with the Beta and Delta Covid variants. They split the monkeys into several groups of five. One immunized group and one unvaccinated control group were exposed to the Beta variant, while another immunized group and control group were exposed to Delta. The immunized monkeys showed little to no detectable levels of virus—indicating that the vaccine protected them against infection—while most control monkeys did.
Most PopularThe End of Airbnb in New YorkBusiness
Even though the researchers only used a RBD from one version of Covid, their vaccine generated a robust polyclonal response—meaning it created multiple antibody types, rather than just one. To Saunders, this is part of the approach’s charm: Creating many antibody types is beneficial, he says, because one that is extremely effective against a certain variant might not be as effective against another. Or vice versa: A previously weak antibody could better neutralize a newer variant. “Some of those antibodies are going to be great at responding to Omicron, some will be great at responding to Alpha, some will be great at responding to Delta,” he says. And some, ideally, will be great at responding to variants that don’t even exist yet.
Jumpstarting the Vaccine
David Martinez, a postdoctoral scholar at the University of North Carolina at Chapel Hill who was a coauthor on several RBD-nanoparticle papers, has studied whether these kinds of vaccines can be boosted by an adjuvant: a substance that “jumpstarts” the immune system and is delivered along with the vaccine. “If you were asleep in bed, your alarm went off, you didn’t get up, and someone threw an ice-cold bucket of water on you—that’s what an adjuvant can do to the immune system,” he says.
Adjuvants can be made from lipids, salts, or other kinds of oils. One kind even contains oil from a shark. They are often used in vaccines; the first mRNA Covid vaccines, for example, used lipid nanoparticles as their adjuvant.
In the January preprint with Saunders’ lab, the team tested their RBD nanoparticle vaccine with three different kinds of adjuvants. They found that in comparison to the standalone vaccine, those with any of the three adjuvants produced higher concentrations of antibodies.
One particular adjuvant, called 3M-052-AF, produced the highest number of antibodies that cross-neutralized different sarbecovirus strains. While its exact recipe is proprietary, the adjuvant contains something called a TLR7/8 agonist: small molecules that stimulate immune cells to activate an immune response. These types of molecules can “essentially talk to the immune system and hyperactivate the immune system to counteract whatever external insult it’s seeing,” Martinez says.
Scientists are also exploring other nano-based methods for variant-proof vaccination. One of these, called a “nanotrap,” was originally described in Matter in June 2021 as a treatment for those who have already been infected rather than as a vaccine. A nanotrap is a mechanism to get rid of Covid viruses through phagocytosis, meaning that a macrophage or other immune cell eats it. Nanotraps work a little like bait—they essentially trick the body into chomping up the invading virus.
The idea could work on a variety of viruses, but bioengineer Jun Huang from the University of Chicago and his team created one that is specific to sarbecoviruses because it has a polymeric nanoparticle shell studded with ACE2 receptors, which are the receptors on human cells that the Covid virus binds to. Because of the high density of ACE2 receptors on the nanotrap’s surface, Covid viruses are attracted to it and get stuck. But here’s where the trap comes in: Sprinkled amid the ACE2 receptors are ligands, little molecules that can bind to a cell receptor and, in this case, induce phagocytosis. The body’s macrophages recognize the ligand and eat up the rest of the virus-flecked nanotrap, thus getting rid of the virus. “We first catch the virus, and then clear the virus,” Huang says.
Most PopularThe End of Airbnb in New YorkBusiness
Now, Huang is curious about how these nanotraps can be leveraged as vaccine candidates. When the macrophages swoop in, they not only eat viruses but can stimulate the rest of the immune system to start creating antibodies against them. Creating a nanotrap with ACE2 receptors would kickstart the immune system into making antibodies that fight Covid-like viruses. “Then we can basically tackle all the variants,” Huang says. “If the virus loses the ability to bind to ACE2, then it cannot infect cells.”
Huang’s nanotrap version is the least tested of all these candidates—he has applied for a patent and demonstrated successful infection clearance in human lung tissue taken from donated organs, but not yet in animals infected with Covid. The others have demonstrated efficacy in Covid animal models, but entering human clinical trials could take another year or two. The vaccine developed by Saunders and colleagues is projected to go into human clinical trials in 2023; the same for the one at the University of Washington. Bjorkman’s group estimates that clinical trials would start in 2024. (“I wish it could be earlier, but there is regulatory stuff we have to go through,” she says.)
A representative for Walter Reed said they were not able to give information about their Phase 1 clinical trial, pending the release of a study.
In the meantime, researchers are already thinking ahead to the next pandemic and how these candidates might be expanded to target even more coronavirus types. “We’ve been working to really expand our vaccine so that it’s effective against MERS coronaviruses as well,” says Saunders, noting that MERS has around a 30 percent mortality rate—a “high mortality rate for a respiratory virus.”
But given the time it will take to carry out human testing, their future utility may come from fighting sarbecoviruses we haven’t even imagined yet. Cohen is optimistic that the lessons learned from these experiments can be helpful in dealing with future zoonotic infections, meaning ones that cross from other animals to humans, as the Covid virus spilled over from bats. “It’s not really far-fetched to think that there will be more animal spillovers in the future,” he says. “So having something that targets this entire category of viruses might be useful for preventing, or at least mitigating, any future outbreaks.”