Fast MAS NMR Spectroscopy Can Identify G-Quartets and Double-Stranded Structures in Aggregates Formed by GGGGCC RNA Repeats
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (Review Commons)
Abstract
The expansion of GGGGCC repeats within the C9orf72 gene has been linked to amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). In neurons from patients with expanded repeats, C9orf72 GGGGCC repeat RNA predominantly forms nuclear foci, and in vitro , repeat-containing RNA can self-aggregate. Two structural motifs have been proposed to provide the interstrand interactions that drive aggregation: G-quartet (G4) structures and double-strand interactions with GG mismatches. Using in vitro transcribed RNA with physiologically relevant number of repeats, we were able to form gel-like aggregates suitable for investingation using fast MAS NMR spectroscopy. This approach enabled us to characterize the dominant interstrand interactions within the RNA gels. Both Watson–Crick and Hoogsteen base pairs were identified in RNA gels formed by RNA with 48 GGGGCC repeats. Their relative abundance shifted upon reconstitution in the presence of different divalent cations or nuclear extracts, underscoring the dynamic equilibrium between G-quadruplex and duplex interactions in GGGGCC RNA aggregation.
Article activity feed
-
Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
Reviewer #1 (Evidence, reproducibility and clarity (Required)):
In this manuscript, the authors employed fast MAS NMR spectroscopy to investigate the gel aggregation of longer repeat (48×) RNAs, revealing inherent folding structures and interactions (i.e., G-quadruplex and duplex). The dynamic structure of the RNA gel was not resolved at high resolution, and only the structural features-namely, the coexistence of G-quadruplexes and duplexes-were inferred. The 1D and 2D NMR spectra were not assigned to specific atomic positions within the RNA, which makes it difficult to perform molecular dynamics (MD) modeling to elucidate the dynamic nature of the …
Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.
Learn more at Review Commons
Reply to the reviewers
Reviewer #1 (Evidence, reproducibility and clarity (Required)):
In this manuscript, the authors employed fast MAS NMR spectroscopy to investigate the gel aggregation of longer repeat (48×) RNAs, revealing inherent folding structures and interactions (i.e., G-quadruplex and duplex). The dynamic structure of the RNA gel was not resolved at high resolution, and only the structural features-namely, the coexistence of G-quadruplexes and duplexes-were inferred. The 1D and 2D NMR spectra were not assigned to specific atomic positions within the RNA, which makes it difficult to perform molecular dynamics (MD) modeling to elucidate the dynamic nature of the RNA gel. The following comments are provided for the authors' consideration:
Reviewer #1, Comment 1:
Figure 2E and Figure 3A: The data suggest that Ca²⁺ promotes stronger G-quadruplex formation within the RNA gel compared with Mg²⁺. This observation is somewhat puzzling, as Mg²⁺ is generally known to stabilize G-quadruplex structures. The authors should clarify this discrepancy.
__Response: __Mg2+ is also a stabilizer of double-stranded RNA. In most cases, Mg²⁺ stabilizes RNA duplexes more significantly than it stabilizes G-quadruplexes. When Mg2+ is removed and replaced for Ca2+, RNA duplex is destabilized more than G4 structures. We have added a clarification regarding that to the Conclusions section.
Reviewer #1, Comment 2:
Figures 2 and 3: The authors use the chemical shift at δN 144.1 ppm to distinguish between G-quadruplex and duplex structures. How was the reliability of this assignment evaluated? Chemical shifts of RNA atoms can be influenced by various factors such as intermolecular interactions, conformational stress, and local chemical environment, not only by higher-order structures. This point should be substantiated by citing relevant references or by analyzing additional RNA structures exhibiting δN 144.1 ppm signals using NMR spectroscopy.
Response: The assignment was made by comparing the chemical shifts with published data and by comparing the obtained spectra with existing datasets in the lab. We have added an explanation to the Results section and cited the literature. The 144.1 ppm was an illustrative value selected for guiding the discussion and we noted that it could sound too specific. We modified Figure 2 to outline the regions of chemical shifts in accordance with our interpretation of spectra.
Reviewer #1, Comment 3:
The authors state that "Our findings demonstrate that fast MAS NMR spectroscopy enables atomic-resolution monitoring of structural changes in GGGGCC repeat RNA of physiological lengths." This claim appears overstated, as no molecular model was constructed to define atomic coordinates based on NMR restraints.
Response: We agree and we have rewritten the conclusions to be more precise in wording. The new text does not mention “atomic-resolution” anymore.
Reviewer #1, Comment 4: Figure 3B: The experiment using nuclear extracts supplemented with Mg²⁺ to study RNA aggregation via 2D NMR may not accurately reflect intracellular conditions. It would be informative to perform a parallel experiment using nuclear extracts without additional Mg²⁺ to better simulate the native environment for RNA folding.
__Response: __We agree that we have not yet approached physiological conditions and that it would be interesting to obtain data for conditions at physiological Mg2+ concentrations in the range between 0.5 mM – 1 mM. The buffer of purchased nuclear extracts does not contain MgCl2, so some MgCl2 would still need to be added. In our opinion, nuclear extracts are actually not the optimal way to move forward, since they still differ from real in cell environment with the caveat that their composition is not well controlled. Full reconstitution with recombinant proteins might be a better approach because stoichiometry can be better regulated.
__Reviewer #1 (Significance (Required)): __ In this manuscript, the authors employed fast MAS NMR spectroscopy to investigate the gel aggregation of longer repeat (48×) RNAs, revealing inherent folding structures and interactions (i.e., G-quadruplex and duplex). The dynamic structure of the RNA gel was not resolved at high resolution, and only the structural features-namely, the coexistence of G-quadruplexes and duplexes-were inferred. The 1D and 2D NMR spectra were not assigned to specific atomic positions within the RNA, which makes it difficult to perform molecular dynamics (MD) modeling to elucidate the dynamic nature of the RNA gel.
Response: We agree that constraints for molecular dynamics cannot be derived from these data. The focus of this work is methodological: to demonstrate how 1H-15N 2D correlation spectra can be used to characterize G-G pairing in RNA gels directly. Such spectra could be used to study effects of small molecules or interacting proteins for example.
__Reviewer #2 (Evidence, reproducibility and clarity (Required)): __ The manuscript by Kragelj et al. has the potential to become a valuable study demonstrating the role and power of modern solid-state NMR spectroscopy in investigating molecular assemblies that are otherwise inaccessible to other structural biology techniques. However, due to poor experimental execution and incomplete data interpretation, the manuscript requires substantial revision before it can be considered for publication in any journal.
__Reviewer #2, Major Concern __Inspection of the analytical gels of the transcribed RNA clearly shows that the desired RNA product constitutes only about 10% of the total crude transcript. The RNA must therefore be purified, for example by preparative PAGE, before performing any NMR or other biophysical studies. As it stands, all spectra shown in the figures represent a combined signal of all products in the crude mixture rather than the intended 48 repeat RNA. Consequently, all analyses and conclusions currently refer to a heterogeneous mixture of transcripts rather than the specific target RNA.
Response: The estimate of 10% 48xG4C2 on the gel is an overstatement. While multiple bands are visible, they correspond to dimers or multimers of the 48xG4C2 RNA. Transcripts that are longer than 48xG4C2 cannot occur in our transcription conditions. Bands at lower masses than expected are folded RNA. The high repeat length and the presence of Mg²⁺ during transcription promote multimerization, which is not fully reversed by denaturation in urea. If shorter transcripts had arisen from early termination they would be still substantially longer than 24 repeats based of what is visible on the gel and would thus remain within the pathological length range. Therefore, the observed NMR spectra primarily report on 48 repeat lengths.
__Reviewer #2, Specific Comments 1: __The statements: "We show that a technique called NMR spectroscopy under fast Magic Angle Spinning (fast MAS NMR) can be used to obtain structural information on GGGGCC repeat RNAs of physiological lengths. Fast MAS NMR can be used to obtain structural information on biomolecules regardless of their size." on page 1 are not entirely correct. Firstly, not only fast MAS NMR but MAS NMR in general can provide structural information on biomolecules regardless of their size. Fast MAS primarily allows for ¹H-detected experiments, improves spectral resolution, and reduces the required sample amount. Conventional ¹³C-detected solid-state MAS NMR can provide very similar structural information. A more thorough review of relevant literature could help address this issue.
Response: We have clarified the distinction between MAS NMR and Fast MAS NMR in the introduction.
__Reviewer #2, Specific Comments 2: __Secondly, MAS NMR has already been applied to systems of comparable complexity - for instance, the (CUG)₉₇ repeat studied by the Goerlach group as early as 2005. That work provided a comprehensive structural characterization of a similar molecular assembly. The authors are strongly encouraged to cite these studies (e.g., Riedel et al., J. Biomol. NMR, 2005; Riedel et al., Angew. Chem., 2006).
Response: We added a mention of that study in the introduction.
Reviewer #2, Experimental Description 1: The experimental details are poorly documented and need to be described in sufficient detail for reproducibility. Specifically:
- What was the transcription scale? What was the yield (e.g., xx mg RNA per 1 mL transcription reaction)?
Response: Between 3.5 mg and 4.5 mg per 10 ml transcription reaction. We’ve added this information to the methods.
Reviewer #2, Experimental Description 2:
- Why was the transcription product not purified? Dialysis only removes small molecules, while all macromolecular impurities above the cutoff remain. What was the dialysis cutoff used?
Response: RNA was purified using dialysis and phenol-chloroform precipitation. We have added the information about molecular weight cutoff for dialysis membranes to the methods.
Reviewer #2, Experimental Description 3:
- How much RNA was used for each precipitation experiment? Were the amounts normalized? For example, if 10 mg of pellet were obtained, what fraction of that mass corresponded to RNA? Was this ratio consistent across all samples?
Response: In the test gel formations, we used 180.0 µg per condition. We used 108.0 µg of RNA for gelation test in the presence of nuclear extracts. We have not determined the water content in the gels. We added this information to methods and results section.
Reviewer #2, Experimental Description 4:
- Why is there a smaller amount of precipitate when nuclear extract (NE) or CaCl₂ is added?
Response: The apparent difference in pellet size may reflect variations in water content rather than RNA quantity. While the Figure 1 might entice to directly compare pellet weights across different ion series tests, our primary goal was to determine the minimal divalent-ion concentrations required to reproducibly obtain gels. We have added a clarification in the Results section and in the Figure 1 caption regarding the comparability of conditions
Reviewer #2, Experimental Description 5:
- The authors should describe NE addition in more detail: What is the composition of NE? What buffer was used (particularly Mg²⁺ and salt concentrations)? Was a control performed with NE buffer-type alone (without NE)?
Response: We have added the full description of NE buffer to the methods section. Its composition is: 40 mM Tris pH 8.0, 100 mM KCl, 0.2 mM EDTA, 0.5 mM PMSF, 0.5 mM DTT, 25 % glycerol. After mixing the nuclear extract with RNA, the target buffer was: 20 mM Tris pH 8.0, 90 mM KCl, 0.1 mM EDTA, 0.25 mM PMSF, 0.75 mM DTT, 12.5% glycerol, and 10 mM MgCl2.
We have not performed a control with NE buffer-type alone but we confirmed separately that glycerol does not affect gel formation.
Reviewer #2, Experimental Description 6:
- How much pellet/RNA material was actually packed into each MAS rotor?
Response: Starting with a 5 mg pellet, we packed a rotor with a volume of 3 µl. We added this information to the methods section.
Reviewer #2, Additional Clarifications: P5. What is meant by "selective" in the phrase "We recorded a selective 1D-¹H MAS NMR spectrum of 48×G₄C₂ RNA gels"?
Response: That was a typo. We meant imino-selective. It is now corrected.
__Reviewer #2, Additional Clarifications: __ There are also several contradictions between statements in the text and the corresponding figures. For example: • Page 4: The authors write that "The addition of at least 5 mM Mg²⁺ was required for significant 48×G₄C₂ aggregation." However, Figure 1E shows significant aggregation already at 3 mM MgCl₂ (NE−), and in samples containing NE, aggregation appears even at 1 mM MgCl₂. Was aggregation already present in the sample containing NE but without any added MgCl₂?
Response: We changed text in the results section to more closely align with what’s depicted on the figure. There was some aggregation present in the nuclear extracts but it was of different quantity and quality. We clarified this in the results section.
__Reviewer #2 (Significance (Required)): __ The manuscript by Kragelj et al. has the potential to become a valuable study demonstrating the role and power of modern solid-state NMR spectroscopy in investigating molecular assemblies that are otherwise inaccessible to other structural biology techniques.
In its current form, tthe manuscript has significant experimental concerns - particularly the lack of RNA purification and inadequate description of materials and methods. The data therefore cannot support the conclusions presented. I recommend extensive revision and repetition of the experiments using purified RNA material before further consideration for publication.
__Response: __We’ve addressed the concerns about RNA purification within the response to the first comment (Major concern).
__Reviewer #3 (Evidence, reproducibility and clarity (Required)): __ This is an interesting manuscript reporting evidence for formation of both hairpins and G-quadruplexes within RNA aggregates formed by ALS expansion repeats (GGGGCC)n. This is in line with literature but never directly confirmed. Given the novelty of the method (NMR magic angle) and of the data (NMR on aggregate), I believe this manuscript should be considered for publication. I also trust the methods are appropriately reported and reproducible.
Below are my main points:
Major points:
__Reviewer #3, Comment 1: __
- RNA aggregation of the GGGGCCn repeat has been reported for expansion as short as 6-8 repeats (see Raguseo et al. Nat Commun 2023), so the authors might not see aggregation under the conditions they use for these shorter repeats but this can happen under physiological conditions . The ionic strengths and the conditions used can vary heavily the phase diagram and the authors therefore should tone down significantly their conclusions. They characterise one aggregate that is likely to contain both secondary structures under the conditions used (in terms of ion and pHs). However, it has been shown in Raguseo et al that aggregates can arise by both intermolecular G4s and hairpins (or a mixture of them) depending on the ionic conditions used. This means that what the authors report might not be necessarily relevant in cells, which should be caveated in the manuscript.
__Response: __We toned down our statements regarding aggregation of shorter repeats in the introduction. We added the citation to Raguseo et al. Nat Commun 2023, which indeed provides useful insights about aggregation of GGGGCC repeats. In Supplementary Figure 1, we had data on gel formation with 8x and 24x repeats which showed these repeat lengths form gels to some extent. We oversimplified our conclusion and said there were no aggregates which needs correction, especially considering other studies reported in the literature have observed in vitro aggregation of these repeat lengths. We modified the results section to reflect this nuance.
__Reviewer #3, Comment 2: __
- It would be important to perform perturbation experiments that might promote/disrupt formation of the G4 or hairpin and see if this affect RNA aggregation, which has been already reported by Raguseo et al, and wether this can be appreciated spectroscopically in their assay. This can be done by taking advantage of some of the experiments reported in the manuscript mentioned above, such as: PDS treatment (favouring monomolecular G4s and preventing aggregation), Li vs K treatment (favouring hairpin over G4s), NMM photo-oxidation (disassembling G4s) or addition of ALS relevant RNA binding proteins (i.e. TDP-43). Not all of these controls need to be performed but it would be good to reconcile how the fraction of G4 vs hairpin reflect aggregates' properties, since the authors offer such a nice technique to measure this.
Response: We appreciate the reviewer’s suggestions and we would be eager to do the perturbation experiments in the future. However, these experiments would require additional optimization and waiting for approval and availability of measurement time on a high-field NMR spectrometer. Given that the primary goal of this manuscript is reporting on the methodological approach, we think the current data adequately demonstrate the technique’s utility.
__Reviewer #3, Comment 3: __
- I disagree with the speculation of the monomolecular G4 being formed within the condensates, as the authors have no evidence to support this. It has been shown that n=8 repeat forms multimolecular G4s that are responsible of aggregation, so the authors need to provide direct evidence to support this hypothesis if they want to keep it in the manuscript, as it would clash with previous reports (Raguseo et al Nat Commun 2023)
Response: We agree that multimolecular G4s contribute to aggregation in our 48xG4C2 gels. We also realized, after reading this comment, that the original presentation of data and schematics may have unintentionally suggested the presence of monomolecular G4 in our RNA gels. To address this, we have added a clarification to the results section, we modified Figure 2 and 3, and we included a new Supplementary Figure 4. For clarification, both multimolecular and monomolecular G4s in model oligonucleotides produce imino 1H and 15N chemical shifts in the same region and cannot be distinguished by the experiments used in our study. Based on the observations reported in the literature, we believe that G4s in 48xG4C2 form primarily intermolecularly, although direct experimental proof is not available with the present data.
Minor points:
__Reviewer #3, Comment 4: __
- An obvious omission in the literature is Raguseo et al Nat Commun 2023, extensively mentioned above. Given the relevance of the findings reported in this manuscript for this study, this should be appropriately referenced for clarity.
Response: We’ve added the citation to Raguseo et al Nat Commun 2023 to the introduction where in vitro aggregation is discussed.
__Reviewer #3, Comment 5: __
- The schematic in Figure 3 is somehow confusing and the structures reported and how they relate to aggregate formation is not clear. Given that in structural studies presentation and appearance is everything, I would strongly recommend to the authors to improve the clarity of the schematic for the benefit of the readers.
Response: We thank you for your comment. We’ve modified the figure, and we hope it is now clearer.
Providing that the authors can address the criticisms raised, I would be supportive of publication of this fine study.
Reviewer #3 (Significance (Required)):
The main strength of this paper is to provide direct evidence of DNA secondary structure formation within aggregates, which is something that has not been done before. This is important as it reconcile with the relevance of hairpin formation for the disease (reported by Disney and co-workers) and the relevance of G4-formation in the process of aggregation through multimolecular G4-formation (reported by Di Antonio and co-workers). Given the significance of the findings in this context and the novelty of the method applied to the study of RNA aggregation, this reviewer is supportive for publication of this manuscript and of its relevance to the field. I would be, however, more careful in the conclusions reported and would add additional controls to strengthen the conclusions.
Response: We thank the reviewer for the comment. In the conclusion section, we have added a statement highlighting the potential roles of both double-stranded and G4 structures in gel formation, in line with what has been reported in previous studies.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
This is an interesting manuscript reporting evidence for formation of both hairpins and G-quadruplexes within RNA aggregates formed by ALS expansion repeats (GGGGCC)n. This is in line with literature but never directly confirmed. Given the novelty of the method (NMR magic angle) and of the data (NMR on aggregate), I believe this manuscript should be considered for publication. I also trust the methods are appropriately reported and reproducible.
Below are my main points:
Major points:
RNA aggregation of the GGGGCCn repeat has been reported for expansion as short as 6-8 repeats (see Raguseo et al. Nat Commun 2023), so the authors …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
This is an interesting manuscript reporting evidence for formation of both hairpins and G-quadruplexes within RNA aggregates formed by ALS expansion repeats (GGGGCC)n. This is in line with literature but never directly confirmed. Given the novelty of the method (NMR magic angle) and of the data (NMR on aggregate), I believe this manuscript should be considered for publication. I also trust the methods are appropriately reported and reproducible.
Below are my main points:
Major points:
RNA aggregation of the GGGGCCn repeat has been reported for expansion as short as 6-8 repeats (see Raguseo et al. Nat Commun 2023), so the authors might not see aggregation under the conditions they use for these shorter repeats but this can happen under physiological conditions . The ionic strengths and the conditions used can vary heavily the phase diagram and the authors therefore should tone down significantly their conclusions. They characterise one aggregate that is likely to contain both secondary structures under the conditions used (in terms of ion and pHs). However, it has been shown in Raguseo et al that aggregates can arise by both intermolecular G4s and hairpins (or a mixture of them) depending on the ionic conditions used. This means that what the authors report might not be necessarily relevant in cells, which should be caveated in the manuscript.
It would be important to perform perturbation experiments that might promote/disrupt formation of the G4 or hairpin and see if this affect RNA aggregation, which has been already reported by Raguseo et al, and wether this can be appreciated spectroscopically in their assay. This can be done by taking advantage of some of the experiments reported in the manuscript mentioned above, such as: PDS treatment (favouring monomolecular G4s and preventing aggregation), Li vs K treatment (favouring hairpin over G4s), NMM photo-oxidation (disassembling G4s) or addition of ALS relevant RNA binding proteins (i.e. TDP-43). Not all of these controls need to be performed but it would be good to reconcile how the fraction of G4 vs hairpin reflect aggregates' properties, since the authors offer such a nice technique to measure this.
I disagree with the speculation of the monomolecular G4 being formed within the condensates, as the authors have no evidence to support this. It has been shown that n=8 repeat forms multimolecular G4s that are responsible of aggregation, so the authors need to provide direct evidence to support this hypothesis if they want to keep it in the manuscript, as it would clash with previous reports (Raguseo et al Nat Commun 2023)
Minor points:
An obvious omission in the literature is Raguseo et al Nat Commun 2023, extensively mentioned above. Given the relevance of the findings reported in this manuscript for this study, this should be appropriately referenced for clarity.
The schematic in Figure 3 is somehow confusing and the structures reported and how they relate to aggregate formation is not clear. Given that in structural studies presentation and appearance is everything, I would strongly recommend to the authors to improve the clarity of the schematic for the benefit of the readers.
Providing that the authors can address the criticisms raised, I would be supportive of publication of this fine study.
Significance
The main strength of this paper is to provide direct evidence of DNA secondary structure formation within aggregates, which is something that has not been done before. This is important as it reconcile with the relevance of hairpin formation for the disease (reported by Disney and co-workers) and the relevance of G4-formation in the process of aggregation through multimolecular G4-formation (reported by Di Antonio and co-workers). Given the significance of the findings in this context and the novelty of the method applied to the study of RNA aggregation, this reviewer is supportive for publication of this manuscript and of its relevance to the field. I would be, however, more careful in the conclusions reported and would add additional controls to strengthen the conclusions.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
The manuscript by Kragelj et al. has the potential to become a valuable study demonstrating the role and power of modern solid-state NMR spectroscopy in investigating molecular assemblies that are otherwise inaccessible to other structural biology techniques. However, due to poor experimental execution and incomplete data interpretation, the manuscript requires substantial revision before it can be considered for publication in any journal.
Major Concern
Inspection of the analytical gels of the transcribed RNA clearly shows that the desired RNA product constitutes only about 10% of the total crude transcript. The RNA must therefore …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #2
Evidence, reproducibility and clarity
The manuscript by Kragelj et al. has the potential to become a valuable study demonstrating the role and power of modern solid-state NMR spectroscopy in investigating molecular assemblies that are otherwise inaccessible to other structural biology techniques. However, due to poor experimental execution and incomplete data interpretation, the manuscript requires substantial revision before it can be considered for publication in any journal.
Major Concern
Inspection of the analytical gels of the transcribed RNA clearly shows that the desired RNA product constitutes only about 10% of the total crude transcript. The RNA must therefore be purified, for example by preparative PAGE, before performing any NMR or other biophysical studies. As it stands, all spectra shown in the figures represent a combined signal of all products in the crude mixture rather than the intended 48× repeat RNA. Consequently, all analyses and conclusions currently refer to a heterogeneous mixture of transcripts rather than the specific target RNA.
Specific Comments
The statements: "We show that a technique called NMR spectroscopy under fast Magic Angle Spinning (fast MAS NMR) can be used to obtain structural information on GGGGCC repeat RNAs of physiological lengths. Fast MAS NMR can be used to obtain structural information on biomolecules regardless of their size." on page 1 are not entirely correct. Firstly, not only fast MAS NMR but MAS NMR in general can provide structural information on biomolecules regardless of their size. Fast MAS primarily allows for ¹H-detected experiments, improves spectral resolution, and reduces the required sample amount. Conventional ¹³C-detected solid-state MAS NMR can provide very similar structural information. A more thorough review of relevant literature could help address this issue. Secondly, MAS NMR has already been applied to systems of comparable complexity - for instance, the (CUG)₉₇ repeat studied by the Goerlach group as early as 2005. That work provided a comprehensive structural characterization of a similar molecular assembly. The authors are strongly encouraged to cite these studies (e.g., Riedel et al., J. Biomol. NMR, 2005; Riedel et al., Angew. Chem., 2006).
Experimental Description
The experimental details are poorly documented and need to be described in sufficient detail for reproducibility. Specifically:
- What was the transcription scale? What was the yield (e.g., xx mg RNA per 1 mL transcription reaction)?
- Why was the transcription product not purified? Dialysis only removes small molecules, while all macromolecular impurities above the cutoff remain. What was the dialysis cutoff used?
- How much RNA was used for each precipitation experiment? Were the amounts normalized? For example, if 10 mg of pellet were obtained, what fraction of that mass corresponded to RNA? Was this ratio consistent across all samples?
- Why is there a smaller amount of precipitate when nuclear extract (NE) or CaCl₂ is added?
- The authors should describe NE addition in more detail: What is the composition of NE? What buffer was used (particularly Mg²⁺ and salt concentrations)? Was a control performed with NE buffer-type alone (without NE)?
- How much pellet/RNA material was actually packed into each MAS rotor? Additional Clarifications P5. What is meant by "selective" in the phrase "We recorded a selective 1D-¹H MAS NMR spectrum of 48×G₄C₂ RNA gels"? There are also several contradictions between statements in the text and the corresponding figures. For example:
- Page 4: The authors write that "The addition of at least 5 mM Mg²⁺ was required for significant 48×G₄C₂ aggregation." However, Figure 1E shows significant aggregation already at 3 mM MgCl₂ (NE−), and in samples containing NE, aggregation appears even at 1 mM MgCl₂. Was aggregation already present in the sample containing NE but without any added MgCl₂?
Significance
The manuscript by Kragelj et al. has the potential to become a valuable study demonstrating the role and power of modern solid-state NMR spectroscopy in investigating molecular assemblies that are otherwise inaccessible to other structural biology techniques.
In its current form, tthe manuscript has significant experimental concerns - particularly the lack of RNA purification and inadequate description of materials and methods. The data therefore cannot support the conclusions presented. I recommend extensive revision and repetition of the experiments using purified RNA material before further consideration for publication.
-
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
In this manuscript, the authors employed fast MAS NMR spectroscopy to investigate the gel aggregation of longer repeat (48×) RNAs, revealing inherent folding structures and interactions (i.e., G-quadruplex and duplex).
The dynamic structure of the RNA gel was not resolved at high resolution, and only the structural features-namely, the coexistence of G-quadruplexes and duplexes-were inferred. The 1D and 2D NMR spectra were not assigned to specific atomic positions within the RNA, which makes it difficult to perform molecular dynamics (MD) modeling to elucidate the dynamic nature of the RNA gel. The following comments are provided …
Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #1
Evidence, reproducibility and clarity
In this manuscript, the authors employed fast MAS NMR spectroscopy to investigate the gel aggregation of longer repeat (48×) RNAs, revealing inherent folding structures and interactions (i.e., G-quadruplex and duplex).
The dynamic structure of the RNA gel was not resolved at high resolution, and only the structural features-namely, the coexistence of G-quadruplexes and duplexes-were inferred. The 1D and 2D NMR spectra were not assigned to specific atomic positions within the RNA, which makes it difficult to perform molecular dynamics (MD) modeling to elucidate the dynamic nature of the RNA gel. The following comments are provided for the authors' consideration:
- Figure 2E and Figure 3A: The data suggest that Ca²⁺ promotes stronger G-quadruplex formation within the RNA gel compared with Mg²⁺. This observation is somewhat puzzling, as Mg²⁺ is generally known to stabilize G-quadruplex structures. The authors should clarify this discrepancy.
- Figures 2 and 3: The authors use the chemical shift at δN 144.1 ppm to distinguish between G-quadruplex and duplex structures. How was the reliability of this assignment evaluated? Chemical shifts of RNA atoms can be influenced by various factors such as intermolecular interactions, conformational stress, and local chemical environment, not only by higher-order structures. This point should be substantiated by citing relevant references or by analyzing additional RNA structures exhibiting δN 144.1 ppm signals using NMR spectroscopy.
- The authors state that "Our findings demonstrate that fast MAS NMR spectroscopy enables atomic-resolution monitoring of structural changes in GGGGCC repeat RNA of physiological lengths." This claim appears overstated, as no molecular model was constructed to define atomic coordinates based on NMR restraints.
- Figure 3B: The experiment using nuclear extracts supplemented with Mg²⁺ to study RNA aggregation via 2D NMR may not accurately reflect intracellular conditions. It would be informative to perform a parallel experiment using nuclear extracts without additional Mg²⁺ to better simulate the native environment for RNA folding.
Significance
In this manuscript, the authors employed fast MAS NMR spectroscopy to investigate the gel aggregation of longer repeat (48×) RNAs, revealing inherent folding structures and interactions (i.e., G-quadruplex and duplex).
The dynamic structure of the RNA gel was not resolved at high resolution, and only the structural features-namely, the coexistence of G-quadruplexes and duplexes-were inferred. The 1D and 2D NMR spectra were not assigned to specific atomic positions within the RNA, which makes it difficult to perform molecular dynamics (MD) modeling to elucidate the dynamic nature of the RNA gel.
-
