PREreview del A novel class of tiny box C/D-like RNAs reveals hidden complexity in the mouse box C/D RNA repertoire

Summary:

In this preprint, Tomas et al. describe a new class of box C/D-like RNAs in mice which they named tiny SNORDs (tiSNORDs). Using deep sequencing of RNAs associated with two box C/D snoRNP (small nucleolar ribonucleoprotein) proteins and systematic mining of public small RNA-sequencing datasets, the authors identify a large number of candidate tiSNORDs across mouse tissues and cell types. They characterized representative candidates and showed that one candidate is stable in cells, and another can support a methylation-like RT stop in a transiently expressed minigene assay, which was consistent with retained guide activity associated with canonical C/D box snoRNAs. Overall, the descriptive and computational aspects of this study are thorough, and expand the catalog of box C/D-like RNAs in mice. The bioinformatics pipeline presented here to identify and classify tiSNORDs can be adapted to examine the presence of tiSNORDs in the human context as well, following clearer delineation of the criteria for this classification. The functional significance of tiSNORDs in the endogenous cellular context, however, is still preliminary and the manuscript would benefit from reframing the current functional claims or strengthening them with additional validation. This work is significant for the field of RNA biology, as future studies can expand upon the 2’-O-methylation guiding role of tiSNORDs in cells alongside investigating potential scaffolding or other non-canonical roles, providing greater understanding into the roles these RNAs play in biology.

Major points:

1. The authors undertook a detailed investigation of the tiSNORDs in a variety of mouse tissues and cell lines in Figures 1-3, and the tiSNORDs are well characterized in length, expression, and structure. The authors also conducted preliminary functional characterization for a few candidate tiSNORDs via a stability assay (Figure 3G), RNA IP to detect association of candidate tiSNORDs with FBL (Figure 4C), and a primer extension assay to detect potential 2’-O-methylation guided by tiSNORDs (Figure 4D). These assays provide partial support to their claim that tiSNORDs can retain 2’-O-methylation guide activity. This can be improved by introducing a condition to their primer extension assay in which the candidate tiSNORD is perturbed, either by knockdown, steric block, or point mutation to disrupt the predicted base pairing region. Loss of the 2’-O-methylation modification upon perturbation would show the modification depends specifically on the tiSNORD.

2. The claim that these tiSNORDs retain the modification guide function of C/D box snoRNAs is important for the significance of this study, given the low abundance and limited evolutionary conservation of these RNAs, and for distinguishing them from snoRNA-derived RNAs. However, the current data in Figure 4 support guide activity for a tiSNORD in a reporter assay, not against endogenous targets. If additional functional validation is beyond the scope of this preprint, the manuscript could still be strengthened by computational analysis identifying which tiSNORDs could be structurally compatible with canonical guide activity, for example based on spacer length and the presence of a plausible antisense element upstream of the D-box. Predicted endogenous targets, perhaps based on complementarity of the base pairing region, could be compared with published 2’-O-methylation datasets. This could clarify whether a subset of tiSNORDs are reasonable endogenous guide RNAs and set a foundation of candidates for future investigations.

3. The discussion acknowledges the low abundance and limited evolutionary conservation of this class of tiSNORDs, and touches on other speculative hypotheses for what roles these tiSNORDs may serve. I would suggest that the authors also include a discussion of sdRNAs (snoRNA-derived RNAs or snoRNA fragments), and how the tiSNORDs differ from this previously characterized class of snoRNA fragments. It would strengthen the claim that tiSNORDs are a previously uncharacterized class of RNAs if they are conceptually distinguished from these snoRNA-derived fragments. For example, criteria such as the presence of C and D boxes at the proper positions, association with the box C/D snoRNP, similarity to known C/D box snoRNAs, exclusion of annotated sdRNAs, and the presence of predicted stem/k-turn features can be included as criteria. This would help to better define this class of tiSNORDs against the context of similar, already characterized RNAs.

Minor points:

1. Experiments in Figure 3G and Figure 4 are shown without clear information on biological replication. The manuscript can be strengthened by reporting the number of independent experiments performed for each presented figure and, where possible, adding replicate data or quantification for Figure 3G showing stability of the tiSNORD and SNORD22, the methylation assay in Figure 4D, and the RIP/Northern blot in Figure 4C.

2. The approximately 20,000 candidates presented may depend in part on the low stringency screen applied during computational analyses, potentially over-representing the scale of the tiSNORD repertoire. It would be helpful if the authors could define a subset of “high confidence” tiSNORDs within this broader set, as well as a discussion of how the candidates are defined and how that may impact the scale of the tiSNORD repertoire. This would be important for future studies to consider when investigating tiSNORDs in the human or another context.

3. As the authors caveat, the primer extension assay in Figure 4D does not provide quantitative support that the tiSNORD is responsible for guiding the modification. To strengthen support for the authors’ claim that tiSNORDs can retain modification guide activity, I would suggest a quantitative assay, such as Nm-VAQ (DOI: 10.1016/j.crmeth.2024.100721) or a similar approach. This would enable the authors to quantify modification efficiency of the ectopically expressed target in an experimental set-up similar to what they already have.

I am one of 2 participants from the same research lab that reviewed this preprint as part of the Transparent and Accountable Peer Review training program. This report was written with the support of my research mentor.

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they did not use generative AI to come up with new ideas for their review.