How modified RNA is read by innate immune machinery
The same sequence is read differently by innate immune sensors depending on what is attached to it: the cap it carries, the methylation on its ribose, and whether its uridines are uridine at all.
01What this article answers
Innate immune sensors respond to features of an RNA rather than to its sequence alone. A cap structure, a methyl group on the 2′-O position of the ribose and the isomerisation of uridine to pseudouridine each change how a transcript is handled by the machinery that decides whether an RNA belongs to the cell.
For modification work this adds a second axis. The modification readout describes what is on the RNA; the immune readout describes how the RNA is responded to. A study that wants to make both statements has to plan for both, and the two halves are most useful when they come from the same material.
The sections below set out which modified features are read this way, how the two readouts are paired, and where the boundary of a modification service sits in a question that is partly immunological.
02Which features the sensors read
Three features are the ones that appear in this literature. They differ in what the sensor is actually reading, and in what a modification method can report about them.
| Feature on the RNA | What changes for the sensor | What the modification side can measure |
|---|---|---|
| The cap and the cap-proximal position | Whether the end of the transcript is read as self | The cap-proximal adenosine and its methylation state, through a route with an exonuclease step that keeps it apart from internal m6A |
| Methylation at the 2′-O position of the ribose | Ribose methylation is one of the features that separates foreign RNA from self RNA for some sensors | The position and the level of the mark, at single-base resolution |
| Pseudouridine in place of uridine | One base position is read as a different nucleotide | Which positions carry the mark, and at what level |
What sits behind these rows is work on viral and synthetic RNA in cell systems. Methylation at the 2′-O position of exogenous RNA has been linked to reduced recognition by innate sensors, and in HIV the methylation deposited on the viral transcript has been read as part of how it avoids a sensor that would otherwise restrict it. In synthetic RNA, substituting pseudouridine for uridine has been reported to lower innate immune activation.
For a study outside virology the point is narrower. If a treatment moves the modification of the RNA and the immune response moves with it, that is two variables measured in one system, and separating them needs the modification to be perturbed on its own.
The cap side carries one extra requirement, because the mark of interest sits next to marks that share its chemistry. Cap-proximal m6Am and the cap m7G form part of what is read as a self cap, and a readout of that position has to keep internal m6A out of the measurement — which is what the exonuclease step is for.
03How to run the two readouts together
- Decide which half carries the claim. “This RNA carries the mark” and “this RNA is responded to differently” are answered by different readouts, and only one of them is a modification measurement.
- Match the modification route to the position being discussed. A claim about the cap end needs the cap-proximal route; a claim about a nucleotide needs a single-base method; a claim about the transcript as a whole can be carried by an enrichment profile.
- Plan the immune side as its own measurement. Interferon and cytokine readouts, sensor activation or the cellular response being studied are run by the group working on that side; they are not a deliverable of a modification service.
- Share the material. Both readouts move with how the cells or tissue were grown, treated and harvested, so a modification result and an immune result from different batches are two findings rather than one comparison.
- Add a perturbation where the argument is causal. A change in a writer or eraser, or a construct with and without the mark, is what separates a change the modification drives from one it accompanies.
- Sample more than one point in time. A modification change and an immune response can follow each other in either order, and a single time point cannot put them in sequence.
04Limits and common misreadings
- A modification readout is not an immune measurement. Sequencing reports what is on the RNA. It does not report what a cell does about it, and the two are never substituted for one another.
- An immune response is not evidence of a modification. Plenty of signals move interferon and cytokine readouts, so the mark has to be measured rather than inferred from the response.
- Cap-proximal methylation is not internal m6A. The two share a chemical mark and sit in different places, so a cap-proximal conclusion cannot be built on an internal m6A profile.
- Pseudouridine substitution changes more than one thing. A study that reports only the immune side has described one axis of a change that also reaches translation and structure.
- The page describes readouts, not outcomes. No statement here concerns therapy, efficacy or the development of a product, and no platform is credited with lowering immunogenicity.
06FAQ
Which marks come into an innate immune question?
Three features are read this way in the literature: the cap structure together with the cap-proximal methylation, methylation at the 2′-O position of the ribose, and pseudouridine in place of uridine. The modification side of a study is designed around whichever of the three the RNA actually carries.
Can one sample set answer both halves?
It can, if both readouts are planned from the start. The modification side runs on total RNA, cells, tissue or blood; the immune readout is a cellular measurement belonging to the group running that side, and the two are most useful when the material is shared.
Why does the cap-proximal route work without an antibody?
Because the mark of interest sits at the very end of the transcript, next to the cap. An exonuclease step digests inward from the free end while the cap protects the few nucleotides behind it, so the cap-proximal position is read on its own and internal m6A does not blur the measurement.
Does a lower immune response from a modified RNA mean the modification is protective?
It is one reading from one measurement in a defined system. The modification side reports what is on the RNA; what a cell or an organism does about it is a separate measurement, so protection in an organism, and any statement about a product, sits outside what a modification readout can support.