Layer upon layer! That’s what cell biology provides you with – just when you think you understand some area of it, things turn out to be more complex. I’m going on in this mode after looking over this new preprint from the Bertozzi lab at Stanford, which uncovers a new class of biomolecules that no one had seen before.
It’s been known for a long time that proteins are modified with glycan residues – it’s a major type of post-translational modification. Such groups are crucial for molecular recognition in protein-protein interactions in immunology, cellular adhesion, secretory pathways and more. But until now this sort of carbohydrate-chain labeling had never been known to intersect with the large, various, and equally vital world of RNA species. As for those, I’ve said many times that back in the 1980s no one would have believed the huge variety of RNAs that have turned out to be active species in the living cell. Short hairpin RNAs, double-stranded RNAs, long noncoding RNAs, circular RNAs, small nuclear RNAs and many others form a whole world of signaling and regulation that people used to be totally unaware of, and that story is nowhere near being fully written or (especially) fully understood. This new paper just underscores that.
This was a discovery enabled by a new method for analysis, an example of what Freeman Dyson and others have noted as a cycle of new phenomena being recognized once someone invents a technology that lets them finally be noticed. In this case, it’s yet another application of azide/alkyne click chemistry. Bertozzi’s lab has developed the use of azido-sugars that get incorporated into glycan side chains, and these can then be labeled by various alkyne-containing probes. (It’s getting to the point where it’s hard to remember chemical biology before this sort of labeling was available – you could make a solid case that it’s one of the fundamental enabling technologies that’s made the field what it is today). They’ve been using this method to track down all sorts of new information about the glycan world, but they noticed, to their surprise, that some of this click-labeling was apparently showing up in subcellular fractions that were thought to be almost entirely composed of RNA.
And so they are! It turns out that Y-RNAs (yet another of those weirdo little essential thingies) do in fact pick up glycan residues. Closer study shows that it’s the guanosines that get labeled, and this process seems to be general across a wide variety of cell types (and in different species). There are interesting differences in the abundance of the glycoRNAs, which of course are as yet unexplained, since we don’t know what they’re doing in there in the first place. The side chains have varying amounts of the different sialic acids in them as well, which also has to mean something. Known disruptors of protein glycosylation also affected the levels of glycoRNAs, suggesting that the same enzymes and pathways could be involved here as well. The glycoRNAs themselves seem to be associated with the membranes of various organelles – they’re not really found in the nucleus, for example.
Well, a new area of cell biology has opened up – not for the first time, and not for the last. Here’s the paper’s conclusion:
The framework in which glycobiology is presently understood excludes RNA as a substrate for N- glycosylation. Our discovery of glycoRNA suggest this is an incomplete view and points to a new axis of RNA glycobiology, including unprecedented enzymology, trafficking, and cell biology.
Absolutely. It’s anyone’s guess what these things are being used for and what diseases might be associated with dysfunction of their pathways. Whenever something like this happens, I think about how many other things are still out there like this that we haven’t come across yet, and what that means for our attempts to understand cell biology in general. Using deep learning and so on to derive connections in our existing knowledge is a perfectly reasonable idea, but it has to be remembered that the reasonably well-known and patterned landscape of our biology knowledge is embedded in a much larger territory of ignorance. There are mountains, lakes, and canyons out there that we know nothing of, and we are unlikely to find them by rearranging the data we already have. This paper serves as an example of finding something new, and there are still so many new things to find. . .