AP-TSS: A New Method for the Analysis of RNA Expression from Particular and Challenging Transcription Start Sites

Round 1

Reviewer 1 Report

Please find my assessment of “AP-TSS: a new method for the analysis of RNA 3 expression from particular and challenging transcription start sites” by Le Berre and colleagues.

Major:

1. In general this manuscript would benefit from a more extensive review of other methods that can address a similar question. For example, several other methods that can be used to identify and quantify alternate TSS. These methods for example; https://academic.oup.com/nar/article/38/21/e194/2411905, https://www.nature.com/articles/srep19420, https://academic.oup.com/nar/article/46/21/e129/5068902, these methods use similar approaches and need to be cited and compared to the approach detailed in this paper.
2. The actual method is not very well descried. It is mostly described in the figure legend but not well described in the text. Since this is a methods development the method should be better descried in the text.
3. It would be more appropriate to compare the specificity and sensitivity of this method to another method for example any detailed in point #1.
4. In Figure 3 it appears that OTC does not correspond to a specific TSS, although OT correspond to TSSi, OTC does not correspond to any TSS. If OTC corresponded to an actual TSS, the potential of this technique would be been more apparent.
5. The fact that this method relies on prior knowledge of the TSS location should be discussed.

 

Minor:

1. Line 42: please cite specific primary articles, rather than a review article.
2. Line 23: Treatment of cells with a DNA methylation inhibitor was associated with a global increase of TERRA levels and no change for HT1080 cells and only a modest increase for HeLa cells of TERRA expression from the TSS of interest. This phrasing is awkward and should bemodified.
3. Line 27: For Alu RNA, the signal obtained by AP-TSS is specific of the RNA Polymerase III-dependent Alu transcript. “specific of the RNA” should be “specific for the RNA”.

 

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Le Berre and colleagues  propose a method to quantify the use of a particular Transcription Start Site (TSS) by using a protocol based on the specific elongation of the cDNA of interest, followed by a quantification by qPCR. This method named AP-TSS (Analysis of Particular TSS) is demonstrated on two types of repetitive sequences found in the human genome : TERRA telomeric transcripts and the Alu repetitive RNAs. This method is very interesting in the field of regulation of transcriptional sites. The experiments are well conducted with adequate controls although one additional should be done. 

 

As a proof of concept of this method, the analysis of repetitive transcripts has been proposed and used to show that even with very weak amount of RNA it is possible to quantify the targeted TSS. However this demonstration is messy because it has been done on repetitive elements, which could be more easy to detect or which could add artefact because of repetitions. It would be more demonstrative to show the detection of alternative promoter and the lower requirement of RNA by comparing it on a "normal gene" (with only 1 sequence/allele) and with a middle level of expression.

Perhaps more acuratly the demonstration would be better using gene whose alternative promoters can be activated differentially.
- Utrophin has 2 alternative promoters and the one on exon2 could be sensitive to 5azadC even in HeLa. (Burton et al. PNAS 1999).
- IKBKG (divergent with G6PD) has 2 promoters and one also sensitive to DNA methylation (Fusco et al., 2006 in Gene).
Several other example can be find in litterature or database.

 

The authors state that one advantages of this method for quantifying the TSS is that it is more quantitative than the conventional methods : 5'end RACE or CAGE methods (line 73 to 81). Neverthless they did not demonstrate it nor compare their method to one of the other. It would be much more interesting to have at least one comparison on quantification and on specificity (see the minor point about the control Figure S2). Or mitigate this claim by proposing an alternative to these "semi-quantitative methods"

 

Finally, a negative control of this method is lacking. To prove the P1 primer (fig2) is not able to amplify itself the TERRA transcripts witout the cDNA extension using the OT complementary sequences, it would be proceed a PCR using P1 + P-TERRA primers on cDNA from the reverse transcription.

 

 

in details :

line 375, the text indicates "5-aza-dC  treatment  induced  a  significant increase of the total TERRA level in both cell lines (24-fold increase in HeLa cells, 39-fold  increase in HT1080 cells) (Figure 4B)."; whereas if the data presented in the Figure 4B in "Relative fold change" (as indicated on Y axis)  "  compared  to  untreated  HeLa  sample." as indicated in the legend; the untreated Hela cells is supposed to be =1, and consequently the fold change of treated HeLa cells is around 39 (and not 24). The same question is for the HT1080 because the levels of untreated cell looks slightly weaker than the HeLa-untreated level.

Please make sure the comment in the text corresponds to the figure, or present the data in a more explicite way.

 

line 377, the Figure 4B is more  the relative levels of total TERRA compared to the untreated HeLa sample as written in the legend and not the Relative fold change as the Y axis is labelled. Relative fold change is expected to be the ratio between the treated sample and the untreated sample for each cell types.

"5aza" labelling in the figures should be replaced by "5-aza-dC"

 

The expression of the TERRA transcripts originated from TSSi (Fig4A) is relativised by using GAPDH levels. This data indicates that TSSi TERRA is 7 times more expressed in HT1080 than in HeLa cells, but what is the evidence that GAPDH is equivalent between the two cell types, or that the GAPDH expression is not modified after the 5-aza-dC treatment ?

 

line 326 to 328 : some primer can escape to the digestion by RecJf, and even the Figure S2 shows that most of the primers were destroyed, it did not demonstrate the digestion is complete. Indeed, the limit detection of Et-bromide-agarose gel is quite limited, and several rest of undigested primer can explain the background of the figure 3B. This control should be conducted with radiolabelled primers, or the conclusion of this control conducted the OTC primer is that the AP-TSS as a limited specificity because of the impossibility to totally digest the primers added for the extension step.

Author Response

Please see attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

In this manuscript, the authors describe an elegant method for a quantitative PCR to amplify cDNAs using alternative promoters. This is a powerful and useful method for the people who are interested in promoter usage.

I have one following comment for this manuscript.

Please discuss the application limitation for rare manuscript whose copy number is very low.

Author Response

In this manuscript, the authors describe an elegant method for a quantitative PCR to amplify cDNAs using alternative promoters. This is a powerful and useful method for the people who are interested in promoter usage.

I have one following comment for this manuscript.

1. Please discuss the application limitation for rare transcript whose copy number is very low.

Reply: In agreement with the reviewer comment, the minimal quantity of transcripts of interest for which the AP-TSS method enables the relative quantification has been estimated from the standard curves for TERRA (figures 3D) and Alu (figure 5B) and has been specified in the result section (lines 357-359, lines 417-419) and in the discussion (lines 547-548).

Round 2

Reviewer 1 Report

I have no further comments for the authors.

Reviewer 2 Report

I agree with the revised form of the manuscript.