The action was innocent: While cleaning the Veterans Affairs hospital in Boston this past January, a contractor knocked a loose freezer plug from its socket. This simple mistake resulted in the loss of nearly 2,000 doses of Moderna's COVID-19 vaccine that had been chilling inside the unplugged appliance. While that's a small hit on the grand scale of worldwide vaccination, it's emblematic of a much larger issue for many COVID-19 vaccines; they have to be kept frozen.
Two of the major coronavirus shots authorized for emergency use in the United States—the Moderna and the Pfizer/BioNTech vaccines—rely on a costly series of temperature-controlled shipments and storage, known as the cold chain, to get vaccines from manufacturers to muscle injection. Such stringent temperature requirements also pose an obstacle for equitable vaccine distribution, increasing the cost and difficulty of shipments and cutting off access to remote communities without reliable electricity or refrigeration.
The reason for these frigid conditions is that the key vaccine ingredient—a molecule called messenger RNA (mRNA)—is extremely fragile and storage at cold temperatures slows down the chemical reactions that can tear it apart. But promising efforts to reduce this frosty burden are already in the works—from tinkering with the mRNA structure to shipping the vaccine in solid form with a sugary protectant.
Such efforts are not just important for halting the current pandemic. Scientists see promise in mRNA vaccines for treating a wide variety of other diseases, since they can be readily tweaked for different viral variants as well as rapidly developed for new viruses.
"All those steps that are taken now will be really important in the coming years," says Rein Verbeke, a pharmaceutical scientist specializing in mRNA vaccines at Ghent University in Belgium.
The necessity for cold storage lies at the heart of how these vaccines work, which is mRNA. These strands of genetic code in the COVID-19 vaccine carry instructions that the human cell uses to manufacture the characteristic spike protein, which sits on the surface of SARS-CoV-2. This preview of the protein familiarizes the body's immune system with the virus so that it can recognize and fight future coronavirus invaders.
Messenger RNA is similar to a single strand of DNA, but its backbone carries one crucial difference: an additional chemical group made up of oxygen and hydrogen, known as hydroxyl.
If the RNA strand bends in just the right way, this hydroxyl group can interact with another part of the backbone sparking a reaction that severs the genetic chain, explains Hannah Wayment-Steele, a PhD student studying RNA structure at Stanford University.
"It cuts the message off," she says. And these shortened messages cannot build a complete protein. "Only one cut in your mRNA strand can be enough to lose your function," says Verbeke, the pharmaceutical scientist.
To slow degradation, companies keep the vaccines at low temperatures. The lower the temperatures, the slower the molecular movements—and the lower the chance of damaging reactions, Verbeke explains. The Pfizer/BioNtech vaccine must be shipped at temperatures colder than nearly 80 degrees below zero. It can be stored for up to two weeks in a standard freezer, up to five days in a fridge, and only six hours at room temperature. Moderna's vaccine is slightly more forgiving. It is stable for up to six months in a standard freezer, up to 30 days if refrigerated, and 12 hours at room temperature.
Vaccine storage is further complicated by another key component: fat. In both the Pfizer/BioNTech and Moderna vaccines the mRNA is encased in fat bubbles known as lipid nanoparticles. They serve as a delivery vehicle to shuttle the mRNA into cells where the cellular machinery can get to work producing the encoded spike protein.
Lipid nanoparticles also help with mRNA stability by shielding it from RNA-degrading enzymes that are abundant both within our bodies and throughout the environment. Yet over time, the lipid nanoparticles themselves can degrade or aggregate, and for a vaccine to work the structure of both fats and mRNA must be injected intact. "It's a difficult thing to accomplish," Verbeke says.
Some natural forms of RNA can survive within our bodies for more than 12 hours, says Rhiju Das, a computational biochemist at Stanford University. "They're these proofs of concept that the RNA should be able to last longer than it does in those vaccines," he says. And one thing these robust RNA molecules have in common are intricate structures that constrict the strand and prevent it from bending in a way in which it can cut itself in two.
"Industry folks had tried using a lot of other things," Das says. They tried tweaking the lipid formulas. They shifted the acidity of solutions. "They couldn't find a way to solve it," he says. But one avenue that was largely unexplored was these intricate folded RNA structures.
This is a potentially useful strategy for vaccine development because multiple mRNA sequences can code for the same protein—and each crumples up in a different way. So if scientists can identify the sequence that folds into the most stable shape, they can produce a vaccine with less stringent temperature requirements for shipping and storage.
The trick, however, is identifying the best genetic origami. "You have these astronomical numbers of possible sequences," says Wayment-Steele, which leads to “whole galaxies of structures that a molecule could take.” To narrow the possibilities, Wayment-Steele and her colleagues turned to an online game known as Eterna, which harnesses the power of crowds to assist in RNA design through puzzles.
Das and his colleague Adrien Treuille of Carnegie Mellon developed the game about a decade ago when they kept running into problems that AI couldn't solve. "Almost out of desperation we decided to try this sort of crowdsourcing approach," says Das, who is Wayment-Steele's graduate advisor. "Eterna’s ended up solving hard problem after hard problem."
Eterna users switch out units of the genetic code, called bases, and the game predicts the folded shape and estimates its stability. "Sometimes it will cause the whole [mRNA] structure to change by changing just that one base," says Amy Barish, a retired chemist and an Eterna player in Cumming, GA. The scientists then work with the players to develop AI, using their structures as examples to train a computer to predict the most stable RNA forms.
Through their work with Eterna players, the team developed a series of mRNA sequences that encode for the spike protein of SARS-CoV-2 variants B.1.351, P.1, and B.1.1.7, first identified in South Africa, Brazil, and the United Kingdom, respectively, that are potentially twice as stable as conventionally designed sequences. They are freely available online for vaccine developers, Das notes.
"It’s just great that we can work on this fun, challenging game but yet we’re potentially helping the world," says Barish, who worked on some of the spike protein puzzles.
Much more work is required, however, before these so-called “superfolder” mRNAs can be injected into arms. One previous concern is that their structure would hinder cellular machinery, known as ribosomes, from reading and translating the mRNA instructions into proteins, explains Maria Barna, a geneticist at Stanford University. She teamed up with Das' lab to test superfolders’ translation using mRNA that codes for a set of easily analyzed proteins, including one that fluoresces green. They were surprised and delighted to find that ribosomes could not only unwind the superfolder structures to produce lots of protein, but the superfolders actually generated more protein than the less stable RNA structures.
"These superfolder mRNAs are not just a dream, they can actually work, and they work well—more than we would have expected," Barna says.
Exactly how this will translate to COVID-19 vaccine stability remains uncertain, but Barna says they hope to produce vaccines that can be stored at room temperature for weeks at a time, if not longer. The team is now collaborating with a pharmaceutical company to test the superfolder spike protein structures in real world applications.
Drying it down
Another possibility for stabilizing the vaccines is drying or freeze drying so they can be stored at room temperature in solid form. But removing the water while keeping the RNA structure intact is no small feat. As the liquid freezes, the crystallizing ice can crush the molecule while whisking away water that can lead to structural collapse.
One way to avoid this damage is through the addition of sugar. Carlos Filipe, chemical engineer at McMaster University, and his colleagues have been testing sugary recipes for drying vaccines, and their current formulation relies on two different types of sugar—trehalose and pullulan.
Trehalose helps fill the voids in the molecule as the water dries away, acting like scaffolding to prop up the structure. The sugar pullulan, which is the base of Listerine strips, encapsulates the molecule to keep it from twisting, which prevents the backbone from cutting itself apart.
"It’s like Hans Solo when he was in the carbonite," says Filipe, posing frozen like the fictional Star Wars character with his hands held up, mouth agape.
Before the COVID-19 pandemic the team demonstrated the efficacy of this sugar treatment to dry out vaccines for the Herpes Simplex type 2 virus and the Influenza A virus and then tested the reconstituted vaccines in mice. Along with his colleague Robert DeWitte, Filipe co-founded the company Elarex to bring this technology to market. They're now working to test the mixture for drying mRNA encapsulated in lipid nanoparticles.
There are multiple different sugar combinations that might work, notes Daan Crommelin, a pharmaceutical scientist at Utrecht University, Netherlands. Yet even with sugar, drying may still have its challenges. For one, drying vaccines could increase the time and cost of production, Crommelin notes. But such costs could be greatly offset by elimination of the cold chain, says DeWitte, who is CEO of Elarex.
Most importantly, there are many options to investigate, or as Crommelin says, "There are several ways that lead to Rome." But he notes the old adage needs a tweak in this case since it's likely not just one road or another. A combination of efforts will be required to distribute COVID-19 vaccines to people no matter where they are in the world.
The next wave of vaccines
Versions of a more stable mRNA vaccine for COVID-19 seem to be on the horizon. Pfizer and BioNTech are currently recruiting participants for a phase 3 trial that will evaluate a freeze-dried version of their SARS-CoV-2 vaccine. They hope for results in the second half of 2021, after which they can submit the results to regulatory agencies for review.
Other companies also have new versions of a liquid mRNA COVID-19 vaccine that may be refrigerator rather than freezer stable. But scant details are available on the reasons behind the stability. Moderna initiated a Phase 1 trial for a version of their next-gen COVID-19 vaccine that they say is refrigerator stable. But after repeated requests the company did not answer questions about reasons behind stability of the new formulation.
The German company Curevac also claims its vaccine is stable in a refrigerator for up to six months and at room temperature for 24 hours. Similar to other vaccines on the market, Curevac’s is encapsulated in lipid nanoparticles (LNPs) and must be protected from cutting itself apart. “We think we might have achieved this by having the mRNA tightly packed within the LNP,” company spokesperson Thorsten Schüller wrote in an emailed statement to National Geographic. “Our theory is that the more compactly the mRNA is packaged, the less attack surface there is." When pressed for details the company responded: "it is hard to pin down differences in stability to just one aspect."
Still, the diversity of possibilities is an encouraging sign of potential improvements to mRNA vaccines already on the market. "This feat was tremendous," Verbeke says of the speedy delivery of a safe and effective vaccine against COVID-19. But he adds, "I’m quite sure there’s still a lot of room for improvement."