Reader Comments

Post a new comment on this article

Clarification requested for Figure 3 G-H

Posted by ppryciak on 10 Oct 2013 at 21:54 GMT

In this interesting paper, there is insufficient description of the experiments in Fig 3 G-H, which creates substantial ambiguity about what is actually being compared. In addition, there may be an error in the description of related data in Fig S3E, which adds to the uncertainty. A minor clarification or correction from the authors would help resolve these ambiguities.

1. In Fig 3G, it is unclear what is being compared between the grey/black versus the blue plots. The possibilities are:
(a) a comparison of SIC1 vs sic1∆ in the genetic background indicated above the graph (e.g., swi4∆ SIC1 vs swi4∆ sic1∆ for the top left graph); or
(b) a comparison of cells with vs without the deletion indicated above the graph (e.g., SWI4 vs swi4∆ for the top left graph). If this is correct, then is the comparison performed in SIC1 or sic1∆ background?

2. A clue about the previous question can be obtained by comparing the bottom right plot of Fig 3G (labelled cln1∆ cln2∆ CLB5pr) with related data shown in Supplemental Fig S3E. That is, it appears that the the grey/black lines in Fig S3E match what is plotted in Fig 3G: the grey/black lines in Fig 3G match the grey/black lines in Fig S3E-top, while the blue lines in Fig 3G match the grey/black lines in Fig S3E-bottom. This suggests that Fig 3G is a comparison of CLN1 CLN2 vs cln1∆ cln2∆ cells, or option (b) noted above.

3. If the deduction in point 2 is correct, then the next question is whether the comparisons in Fig 3G were performed in SIC1 or sic1∆ background. The legend to Fig S3E suggests the SIC1 background, because it states that "Black and red represent cells with and without SIC1, respectively"; i.e., black = SIC1. However, I suspect that this description might be an error, and that the correct labels are the reverse: black = sic1∆, red = SIC1. Otherwise, as currently labelled, the data in Fig S3E-bottom imply that CLB5pr transcription is MORE noisy in SIC1, and LESS noisy in sic1∆. The reverse would seem to make more sense and fit better with other findings in the paper.

4. A similar uncertainty exists regarding Fig 3H. It is clear that each plot compares cells with vs without the stress indicated above each graph. But in what cells is this being tested? Is this performed in SIC1 cells (and hence would correspond to measurements in the LEFT halves of Figs 3 E-F), or is this performed in sic1∆ (or sic1-0P) cells (and hence would correspond to measurements in the RIGHT halves of Figs 3 E-F)?

Clarification of these ambiguities would be appreciated.

No competing interests declared.

RE: Clarification requested for Figure 3 G-H

xiaojing replied to ppryciak on 13 Oct 2013 at 03:00 GMT

Thank you very much for your comments, and we are very sorry that "unstressed cells" in Fig3G and Fig3H leads to some confusion.

1. As for the Fig3G and 3H, we consider both genetic and environmental perturbations are stress, thus "unstressed cells" (grey/black) in Fig3G,3H are all WT cells under normal condition. We hope Fig3H and 3G can give a sense on how much noise that different perturbations introduce to the G1/S transcription.

2. As for the Fig S3E, the legend is correct. Upper: WT(black) vs. Sic1D(red) Lower: cln1Dcln2D(black) vs. sic1Dcln1Dcln2D(red). Yes, that is the same type of plot as Fig.3, but it is a supplement for Fig.4, as we noted in the paper. Fig S3E is to demonstrate that Clb5/6 contributes to the first feedback loop when Cln1/2-Cdk is absent.

---Deleting SIC1 rescues cln1Dcln2D, but the transcription in sic1Dcln1Dcln2D strain is still much nosier than that in WT (compare black/grey lines in upper panel to red lines in lower panel), which can be the reason of increased Sic1 half-life variability in sic1Dcln1Dcln2D. Note that, in principle, in terms of transcription, deleting SIC1 should rescue all perturbations on the first feedback loop in certain level, including swi4D, mbp1D, whi5D, cln2D and environmental stress as well. It is because that Clb5/6 contributes to the first feedback loop as SIC1 is deleted, thus increases the strength of the first feedback loop. Given a fact that same perturbation leads to less noise on transcription in sic1D background compared to the WT background, we observed a more dramatic effect on Sic1 half-life in sic1D background suggests the DNFBL functions as a noise filter even strongly. Of course, perturbations affect the cell might be in a more complex way, leading to noisy transcription might be just one of them as we discussed in the paper.

No competing interests declared.