Mammalian Y RNAs are modified at discrete guanosine residues with N-glycans

Ryan A. Flynn, Benjamin A. H. Smith, Alex G. Johnson, Kayvon Pedram, Benson M. George, Stacy A. Malaker, Karim Majzoub, Jan E. Carette, Carolyn R. Bertozzi

Preprint posted on September 30, 2019

The Discovery of Glycosylated RNA

Selected by Connor Rosen

Categories: biochemistry, cell biology


Glycosylation is a key modification of biological macromolecules. The modification of proteins by glycans can play important roles in folding, signaling, binding partner recognition, immunogenicity, and more. Similarly, glycolipid conjugates are key determinants of cell-cell interactions and immunogenicity. However, the last major branch of biological macromolecules, nucleic acids, has not been previously described to be modified by glycosylation. The study of glycosylation is complicated by the non-genetically encoded nature of complex branched glycans, the diversity of biological sugars and their linkages, and the difficulty of synthesizing and characterizing complex glycan chemistries. New chemical technologies are enabling more complete and sophisticated probing of the biology and chemistry of glycosylation, and in this preprint, those techniques are used to identify glycosylation of small non-coding RNAs in mammalian cells.


Key findings:

  • Small non-coding RNAs are glycosylated

The authors describe that the use of azide-sugar labeling of cells enables isolation of apparently glycosylated RNAs (glycoRNAs). Stringent purifications and replication in multiple cell lines and in mice established that there is in fact a population of RNAs that are glycosylated. Fractionation and sequencing results revealed that these were predominately small (<200 nt) RNAs, including a major representation of Y RNAs. Experiments with chemical inhibitors, cells deficient in glycosyltransferases, and recombinant glycosidases suggested the glycosylation was N-linked glycosylation involving canonical enzymes such as OST, with terminal sialic acids. Finally, subcellular fractionation experiments suggested that at least some fraction of glycoRNAs reside in membrane-enclosed compartments, such as the ER (which is where N-glycosylation of proteins canonically occurs).

  • This glycosylation appears to be focused on specific guanosine residues

The authors identified sites of glycosylation by mapping sites where reverse transcription of RNAs was “stopped” during library preparation due to a modified base pair blocking reverse transcription. These stops were enriched at guanosine sites in particular glycoRNAs, suggesting the glycosylation occurred specifically at those sites. This was specific to the glycoRNA enriched pool of RNAs, suggesting it is not a basal modification of those RNAs. This specificity provides an important handle for future studies on glycoRNAs.



These findings represent fundamental advances in cell biology and biochemistry that open huge avenues for future investigation. At a fundamental level, this raises questions about the regulation and specificity of RNA glycosylation, the exact biogenesis of glycoRNAs, the consequences of glycosylation on RNA biochemistry (e.g. folding and half-life), and the trafficking components for RNA transport. More specifically, the biology of Y RNA glycosylation (and other small RNAs they detected, such as snoRNAs) is a new field itself – how does glycosylation modify the canonical functions of Y RNAs? Does it impact association with Ro60, or other client molecules? Does it impact immunogenicity (the authors mention briefly that many of their glycoRNAs are targets of autoantibodies in human autoimmune diseases)?


Moving Forward / Questions for Authors:

  • There are a few potential “killer experiments” that should convince any remaining doubters (it has been suggested that there could be residual glycoproteins contaminating the RNA preparations, despite stringent purification protocols, for example). One might be mass spectrometry showing definitive linkage between the sugars and RNA backbone, which would also potentially elucidate the structure of the glycan linkage and any post-transcriptional modifications on the guanosine residues that are necessary for the linkage. Another possibility would be chemical reconstitution of glycosylation with purified enzymes (e.g. OST) and defined input RNA (possibly in vitro synthesized) – however, this is complicated by the question of prior post-transcriptional modifications required for glycosylation and an incomplete knowledge of the linkages and sugars used and therefore the necessary enzymes.
  • Deletion of Y5 reduced total glycoRNA levels substantially – is point mutation of the guanosine target sites feasible? That is, would total Y5 levels be comparable between WT and G35/64 double-mutant cells? If those mutations likewise reduce glycoRNA levels and Y5 sequencing reads in glycoRNA-enriched fractions, that would also be strongly suggestive of specific glycosylation at those sites, as validation beyond the RT-stop method.
  • The evidence for luminal RNA is limited to evidence that some glycoRNA is protected by membranes. RNA, including small non-coding RNA, is known to be a component of exosomes, which already suggests one membrane-protected compartment accessible to RNAs. Do the methods used by the authors distinguish ER/Golgi from endosomes and exosomes (or other similar vesicular bodies / organelles)? Can glycoRNA be found outside the cell (does it complete passage through the secretory pathway), or is it predominately “retrotransported” back into the cytosol? As mentioned above, the future dissection of regulation of glycoRNA (or pre-glycoRNA) transport is an exciting opportunity to discover new cell biology, and the precise definition of the compartment(s) where glycoRNA resides will be critical to inform those studies.
  • How conserved are the glycosylation sites described in this paper, particularly in the well-defined Y5 RNA? The paper only examines mammalian cells but Y RNAs are evolutionarily ancient, stretching back to bacteria – the point at which glycosylation emerges will be interesting to uncover.


Posted on: 7th October 2019

Read preprint (No Ratings Yet)

  • Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here