mRNA Works Differently Than We Thought (and That’s Good)

Scientists may have been wrong about mRNA vaccines — and they’re thrilled. While mRNA vaccines are already proven safe and effective, scientists are just now uncovering new details about how they work. A new study finds they can bring about a response by entering nonimmune cells, such as muscle cells, challenging the dogma that mRNA must enter immune cells (dendritic cells) to be effective. The findings, along with other emerging research, suggest mRNA is more versatile than scientists realized — capable of bypassing traditional immune pathways and using a variety of cells to jump-start the immune system. That understanding could open the door to more effective therapies, with potential implications for hundreds of drugs. “The assumption for 20 years of my career has been: The secret sauce of mRNA and other nucleic acid vaccines is that they get into dendritic cells,” said senior study author Brian D. Brown, PhD, a professor of genetics and genomic sciences and of immunology and immunotherapy at the Icahn School of Medicine at Mount Sinai in New York City. “That turned out not to be the case.”

Why Muscle Cells Might Do Some Heavy Lifting

Brown was studying gene therapy in the 1990s when he designed a technology to turn mRNA expression on or off in different cells. For the new mouse study, published in Nature Biotechnology , he adapted the technology to turn off mRNA expression in dendritic cells, muscle cells, or liver cells. The researchers then vaccinated the mice with each version, delivering the vaccines both intravenously and intramuscularly. “The results were pretty stunning,” Brown said. When mRNA expression was turned off in muscle cells, T-cell response went down, suggesting muscle cells play a role in immunity. When expression was turned off in liver cells, T-cell expression tripled — indicating liver cells dampen immunity. Turning off expression in dendritic cells had no effect on T-cell activation, though it did reduce the number of killer T cells by as much as half. (Interestingly, no such reduction occurred when the antigen was SARS-CoV-2 spike. Brown is now investigating why different antigens had varying effects.) Knowing all this is crucial for designing effective mRNA vaccines and therapies. That’s because different mRNA therapies require different strategies. Cancer vaccines must boost tumor-fighting killer (CD8+) T cells. For genetic disease treatments, scientists want to avoid triggering the immune system to prevent killing the very cells the mRNA is meant to modify. “Understanding the immunology is extremely important for this class of drug,” Brown said. The finding doesn’t mean dendritic cells aren’t important for mRNA vaccines to work. “It just means that the mRNA doesn’t have to get into those cells to induce an immune response,” Brown said. Instead, the antigen can be transferred to those dendritic cells.

Here’s What That Process Might Look Like

Traditional vaccines rely on cross-presentation: Dendritic cells capture a protein or inactivated virus, process it, and display it on MHC class I molecules to alert the immune system. By contrast, mRNA vaccines were thought to require direct presentation, where the mRNA enters a dendritic cell and the antigen is produced and presented by that same cell. However, Brown’s research suggests that mRNA vaccines may also use cross-presentation. While “the vast majority” of mRNA likely does enter immune cells, Brown speculates that the magic happens when those cells release the newly made antigens into the surrounding tissue. Neighboring dendritic cells then scoop up those free-floating antigens to kick-start the killer T-cell response. Or a different process entirely could be happening alongside this. Dendritic cells can also trigger a response through “cross-dressing.” That’s when a nonimmune cell (like a muscle cell) processes the mRNA, loading the antigen onto its own MHC. The dendritic cell then snatches the preloaded MHC molecule right off the cell’s surface. In the case of the SARS-CoV-2 spike protein, however, Brown suspects a simpler handoff: The spike protein likely gets cleaved from the muscle cell surface, allowing dendritic cells to vacuum it up and cross-present it. “It’s a fine finding,” said Kenneth Murphy, MD, PhD, a professor of pathology and immunology at Washington University School of Medicine in St. Louis, who was not involved in the study. “But they didn’t distinguish between what was cross-dressing or cross-presentation.” Murphy’s recent research published in Nature sheds further light.

‘Everything Is Happening’

Although mRNA cancer vaccines are already being used in clinical settings , very little scientific evaluation of the mechanism exists, Murphy said.

Kenneth Murphy, MD, PhD

Scientists have previously assumed that a specific type of dendritic cells — cDC1 — were essential for activating killer T cells to fight viruses or tumors. However, Murphy’s lab discovered that mice lacking these cells still mounted an immune response to mRNA vaccines. Even when researchers stripped all dendritic cells of their MHC class 1 (the part that alerts the immune system), the mice still produced a partial T-cell response. “You don’t even need [MHC] class 1 to be expressed on a dendritic cell,” Murphy said. “That’s where cross-dressing comes in.” In cross-dressing, nonimmune cells pick up the slack. The nonimmune cell’s MHC transfers to a dendritic cell, which then presents it to the T cell. “The subtlety in our paper is that everything’s happening,” Murphy said — direct presentation, cross-presentation, and cross-dressing.

Expanding the mRNA Landscape

Murphy also captured a gene expression signature for individual killer T cells in wild type mice and mice with only cDC1s or cDC2s. There were differences depending on which dendritic cell did the priming, with cDC1s driving a stronger killer T-cell response. “If you’re going to try to optimize a CD8+ T-cell response to a neoantigen in a cancer setting, you might want to understand how the antigens are being processed and what dendritic cell is presenting them,” Murphy said. His mouse models are available in The Jackson Laboratory database, which vaccine and therapeutic developers can use “to see whether or not their antigens are cDC1 processed or cDC2 processed or both, or cross-dressed, or something else,” Murphy said. How mRNA vaccines spur development of helper (CD4+) T cells is also under investigation. Research published in Nature Communications in January found that mRNA vaccines are better at activating helper T cells through direct presentation. The researchers’ previous flu research in 2015 suggests that some helper T cells can only be activated by direct presentation.

Laurence “Ike” Eisenlohr, VMD, PhD

“It is an important consideration,” said senior study author Laurence “Ike” Eisenlohr, VMD, PhD, a professor of pathology and laboratory medicine at Children’s Hospital of Philadelphia Research Institute in Philadelphia. “It’s safe to say that all three papers” — his, Brown’s, and Murphy’s — “point to an expanded landscape of antigen processing and presentation when it comes to mRNA-based vaccines.” Yet scientists have likely only scratched the surface of the biological foundations of mRNA. “We looked at three different types of cells,” Brown said. “We can go much, much more granular.” Brown reported being on the scientific advisory boards of Noetik, Asgard Therapeutics, and Navexio. Murphy and Eisenlohr had no relevant disclosures.