An artificial intelligence (AI) tool that optimizes gene sequences found in mRNA vaccines could help create shots with greater potency and stability that could be deployed around the world.
The tangled history of mRNA vaccines
Developed by scientists at the California division of Baidu Research, a Beijing-based AI company, the software borrows techniques from computational linguistics to design mRNA sequences with more complex shapes and structures than those used in vaccines. current. This allows the genetic material to persist longer than usual. The more stable the mRNA delivered to a person’s cells, the more antigens are produced by the protein-making machinery in that person’s body. This, in turn, leads to an increase in protective antibodies, theoretically leaving immune individuals better equipped to fend off infectious disease.
Additionally, the increased structural complexity of mRNA provides better protection against vaccine degradation. During the COVID-19 pandemic, mRNA vaccines against the SARS-CoV-2 coronavirus had to be transported and stored at temperatures below -15°C to maintain stability. This has limited their distribution in resource-poor regions of the world that do not have access to ultra-cold storage facilities. A tougher, AI-powered product could eliminate the need for cold chain equipment to handle such knocks.
The new methodology is “remarkable,” says Dave Mauger, a computational RNA biologist who previously worked at Moderna in Cambridge, Massachusetts, a maker of mRNA vaccines. “The computational efficiency is truly impressive and more sophisticated than anything that has come before.”
Vaccine developers already routinely adjust mRNA sequences to align with cell preferences for certain genetic instructions over others. This process, known as “codon optimization”, leads to more efficient protein production. The Baidu tool goes one step further by making mRNA – usually a single-stranded molecule – loop back on itself to create stiffer double-stranded segments (see “Design optimization”).
Known as LinearDesign, the tool only takes a few minutes to run on a desktop computer. In validation tests, he has yielded vaccines that, when evaluated in mice, elicit antibody responses up to 128 times greater than those mounted after immunization with more conventional, codon-optimized vaccines. The algorithm has also helped extend the storage stability of vaccine designs by up to six times in standard test-tube trials conducted at body temperature.
“It’s a huge improvement,” says Yujian Zhang, former head of mRNA technology at StemiRNA Therapeutics in Shanghai, China, who led the experimental validation studies.
So far, Zhang and her colleagues have tested LinearDesign-enhanced vaccines against only COVID-19 and shingles in mice. But the technique should prove useful when designing mRNA vaccines against any disease, says Liang Huang, a former Baidu scientist who led the creation of the tool. It should also help in mRNA-based therapies, says Huang, who is now a computational biologist at Oregon State University in Corvallis.
The researchers reported their findings on May 2 at Nature1.
Already, the tool has been used to optimize at least one licensed vaccine: a COVID-19 vaccine from StemiRNA, called SW-BIC-213, which gained approval for emergency use in Laos late last year. ‘last year. Under a licensing agreement established in 2021, French pharmaceutical giant Sanofi also uses LinearDesign in its own mRNA investigational products.
Trial settles debate over best mRNA design in COVID vaccines
Executives from both companies point out that many design features influence the performance of their candidate vaccines. But LinearDesign is “definitely a type of algorithm that can help with that,” says Frank DeRosa of Sanofi, head of research and biomarkers at the company’s mRNA Center of Excellence.
Another was reported last year. A team led by Rhiju Das, a computational biologist at Stanford School of Medicine in California, has demonstrated that even greater protein expression can be obtained from mRNA – at least in cultured human cells – if certain models loops are removed from their strands, even when such changes loosen the overall stiffness of the molecule2.
That suggests alternative algorithms might be preferable, says theoretical chemist Hannah Wayment-Steele, a former member of Das’ team who is now at Brandeis University in Waltham, Massachusetts. Or, it suggests that manual mRNA tuning optimized by LinearDesign could lead to even better vaccine sequences.
But according to David Mathews, a computational RNA biologist at the University of Rochester Medical Center in New York, LinearDesign can do the heavy lifting. “It puts people in the right stage to start doing any optimization,” he says. Mathews helped develop the algorithm and is a co-founder, along with Huang, of Coderna.ai, a Sunnyvale, Calif.-based startup that is further developing the software. Their first task was to update the platform to account for the types of chemical modifications found in most approved and experimental mRNA vaccines; LinearDesign, in its current form, is based on an unmodified mRNA platform that has fallen out of favor among most vaccine developers.
A structured approach
But mouse studies and cell experiments are one thing. Human trials are another. Because the immune system has evolved to recognize certain RNA structures as foreign — particularly the twisted ladder shapes in many viruses that encode their genomes as double-stranded RNA — some researchers worry that an algorithm of Optimization like LinearDesign does end up creating vaccine sequences that stimulate harmful immune reactions in people.
“It’s kind of a liability,” says Anna Blakney, an RNA bioengineer at the University of British Columbia in Vancouver, Canada, who was not involved in the study.
Early results from human clinical trials involving StemiRNA’s SW-BIC-213, however, suggest that the extra structure is not an issue. In small booster trials reported to date, vaccine side effects were found to be no worse than those reported with other mRNA-based COVID-19 vaccines3. But as Blakney points out, “We’ll learn more about that in the years to come.”