Why the immune system attacked mRNA — and what Karikó and Weissman changed

In the early 2000s, the immune system attacked mRNA only in a metaphorical sense: laboratory-made RNA could activate innate immune defenses far too readily. That was a serious obstacle to using mRNA as a medicine. Instead of simply delivering instructions for a cell to make a protein, the molecule could itself be interpreted as a warning signal and provoke inflammation.

Understanding how to reduce that response became one of the key problems addressed by Katalin Karikó and Drew Weissman. Their work provided one of the fundamental pieces that eventually made modern mRNA vaccines possible.

Why mRNA needed to avoid an immune alarm

Messenger RNA, or mRNA, is a temporary set of instructions. DNA stores genetic information, while mRNA carries a working copy of part of that information to ribosomes, the cellular machinery that builds proteins.

The therapeutic appeal is straightforward. Instead of administering a protein or permanently altering DNA, a treatment can deliver temporary instructions. The cell produces the required protein, and the mRNA is subsequently degraded.

Biology, however, adds a complication. Cells evolved in an environment full of viruses and bacteria, and they possess sensors that detect molecular patterns associated with infection. Certain forms of RNA can therefore function as danger signals.

Among these sensors are Toll-like receptors, components of the innate immune system. They do not need to identify a particular virus with high precision. Their role is to recognize suspicious molecular features quickly enough to trigger an early defense.

Laboratory-produced RNA could set off that alarm.

The chemistry of RNA mattered

In 2005, Katalin Karikó, Michael Buckstein, Houping Ni and Drew Weissman published a study in Immunity that asked a fundamental question: does the innate immune system respond simply to the presence of RNA, or does the chemical form of its building blocks matter?

RNA is often represented as a sequence written with four letters. Real cellular RNA is chemically more complicated. After RNA is produced, some of its nucleosides can be chemically modified. The genetic message may remain readable while the physical and chemical properties of the molecule change.

The researchers generated RNAs containing several modified nucleosides. One of them was pseudouridine, a naturally occurring modified form related to uridine.

They then tested how these RNAs affected innate immune receptors and dendritic cells, which are among the cells that detect signs of infection and help initiate immune responses.

The difference was striking.

Unmodified RNA could activate human TLR3, TLR7 and TLR8. Incorporating particular modified nucleosides sharply reduced or eliminated this activation. Dendritic cells exposed to modified RNA also produced fewer inflammatory signaling molecules and showed less activation than cells exposed to unmodified RNA.

Modifications involving uridine, including pseudouridine, were particularly important.

The immune system, in other words, was responding not only to the information contained in an RNA molecule. It was also sensitive to the chemistry of its components.

How this changed the prospects for mRNA medicine

That distinction had major implications.

When therapeutic RNA triggers innate immune sensors, the resulting inflammatory response can be undesirable. It can also interfere with the main purpose of the mRNA: serving as a template from which cells produce a protein.

Nucleoside modification offered a way to separate these functions. The genetic instructions could remain intact while the molecule became less conspicuous to some components of the innate immune system.

Later studies by Karikó, Weissman and other researchers extended the finding. In 2008, pseudouridine-containing mRNA was shown not only to provoke less immune activation but also to support more efficient protein production. Further refinements produced other modified nucleosides with useful properties.

The Pfizer/BioNTech and Moderna COVID-19 mRNA vaccines used N1-methylpseudouridine, a modified nucleoside related to pseudouridine. This is why the popular description that scientists “changed one letter in mRNA” can be misleading. The genetic message was not simply altered at a single position. Instead, one type of RNA building block was chemically replaced at many positions throughout the molecule.

Nucleoside modification was also only one part of the technology.

Practical mRNA vaccines required methods for protecting fragile RNA and delivering it into cells, most importantly lipid nanoparticles. They also depended on sequence optimization, purification of RNA, antigen design, scalable manufacturing and clinical development.

Karikó and Weissman therefore did not single-handedly “invent the mRNA vaccine.” Their contribution was more specific: they helped establish that modifying nucleosides could greatly reduce unwanted innate immune recognition of laboratory-produced mRNA.

The 2005 paper itself was not a human vaccine trial. It was primarily a mechanistic study using cells and receptor systems. Toll-like receptors are also only part of the cellular machinery that senses foreign RNA. Later work had to address additional pathways and problems, including double-stranded RNA contaminants that can arise during laboratory RNA production.

Drew Weissman was born on September 7, 1959. In 2023, he and Katalin Karikó received the Nobel Prize in Physiology or Medicine for discoveries concerning nucleoside base modifications that enabled the development of effective mRNA vaccines against COVID-19.

The significance of those discoveries now extends well beyond one pandemic. Researchers are developing mRNA platforms for cancer vaccines, temporary production of therapeutic proteins and other forms of RNA-based medicine. The same basic lesson remains central: when RNA is used as a medicine, what matters is not only the message encoded in its sequence. The chemistry of the molecule itself can determine how the cell responds to it.