Pyrimidine Triones as Potential Activators of p53 Mutants

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

 

Jebril Fallatah and co-workers describe the initial computational and subsequent experimental identification of the pyrimidine triones UCI-1001 and UCI-1014, which are able to restore DNA-binding activity to p53 missense mutants and inhibit cell proliferation and apoptosis. Using various experimental approaches they provide evidence that UCI-1001 could directly bind to p53 and stabilize it in the wt conformation. This is a very interesting study that identifies a potential novel small molecule to restore p53 activity in cancer.

There are some additional experiments or controls that the authors could consider including in order to further strengthen their manuscript:

 

Minor points:

 

Fig. 1B. I understand that the Saos2-derived cell lines are genetically almost identical. In addition to the Saos-2 p53 null, and in order to exclude that potential alterations arising during clonal selection might contribute to the sensitivity to the compounds, the authors might consider including data from the Saos2-p53-G245S cell line in the absence of doxycyclin-induced mutant p53 expression.

 

Fig. 2B: As above: To exclude clonal effects, an additional control would be to show that Saos-2 p53-R175H behaves like wt cells in the absence of doxycyclin-induced p53 expression.

 

Fig. 2A,C: As only three cell lines of very different origin are compared, these data may be over-interpreted. Of note, 1uM doxorubicin is also causes more cell death to TOV-112 cells than in MCF-7 cells (Fig2C), suggesting that TOV-112D cells may be more vulnerable by various insults. This would also be consistent with the observation in Fig. 2A – what would a DXR dose curve look like for these cell  lines?

 

Fig. 4A: Is the thermal stabilization specific to p53-mutant, or is it also observed for p53-wt? This control could be included.

 

Fig. 4B: Does UCI-1001 also alter the IF of p53-wt?

 

Fig 5C: The specificity and p53 dependence of the expression profiles could be further supported by inclusion of a related, but “inactive” compound, such as UCI-1002, which does not bind to p53.

 

Author Response

Point-by point response to reviewer’s comments

 

We would like to thank the reviewers for their insightful and productive comments. We have addressed the points raised by the reviewers as outlined below.

Please note that all changes/updates in the manuscript are highlighted in yellow.

We thank the reviewers for their time and critically reading our manuscript. We believe the changes further improve our manuscript. Note that some of the figure numbers have changed due to insertion of an additional figure (Fig. 2).

 

Reviewer 1: (reviewer comments in blue, authors’ response in black)

 

Jebril Fallatah and co-workers describe the initial computational and subsequent experimental identification of the pyrimidine triones UCI-1001 and UCI-1014, which are able to restore DNA-binding activity to p53 missense mutants and inhibit cell proliferation and apoptosis. Using various experimental approaches they provide evidence that UCI-1001 could directly bind to p53 and stabilize it in the wt conformation. This is a very interesting study that identifies a potential novel small molecule to restore p53 activity in cancer. 

There are some additional experiments or controls that the authors could consider including in order to further strengthen their manuscript:

 

Minor points:

Fig. 1B. I understand that the Saos2-derived cell lines are genetically almost identical. In addition to the Saos-2 p53 null, and in order to exclude that potential alterations arising during clonal selection might contribute to the sensitivity to the compounds, the authors might consider including data from the Saos2-p53-G245S cell line in the absence of doxycyclin-induced mutant p53 expression. 

Fig. 2B: As above: To exclude clonal effects, an additional control would be to show that Saos-2 p53-R175H behaves like wt cells in the absence of doxycyclin-induced p53 expression.

We thank the reviewer for this comment. We would like to mention that these cell lines are not clonally selected but used as a pool of hundreds of pooled clones (please also see PMID: 23360998). We note that Saos-2 cells (p53-null) were compared in the presence of DOX, which we did not mention in the manuscript. This was added. Unfortunately we cannot use Saos-2 p53-R175H cells in the absence of DOX because the TET-controlled is not completely repressed in the absence of DOX and shows some expression of p53.

Fig. 2A,C: As only three cell lines of very different origin are compared, these data may be over-interpreted. Of note, 1uM doxorubicin is also causes more cell death to TOV-112 cells than in MCF-7 cells (Fig2C), suggesting that TOV-112D cells may be more vulnerable by various insults. This would also be consistent with the observation in Fig. 2A – what would a DXR dose curve look like for these cell  lines?

The reviewer raised a good point. We evaluated cell survival at Doxorubicin concentrations (0, 0.01, 0.10, and 1 μM) for TOV-112D (p53R-175H) and MCF7 (p53-WT) cell lines. Cell numbers were compared after 3 days using the CellTiter-Glo® reagent. The survival curves show similar Doxorubicin IC50 values: 0.071 ± 0.0014 μM for TOV-112D and 0.081 ± 0.0158 μM for MCF-7. We added these results to the text and as Figure S1B.

Fig. 4A: Is the thermal stabilization specific to p53-mutant, or is it also observed for p53-wt? This control could be included.

This is another good suggestion from the reviewer. We performed differential scanning fluorimetry (DSF) and observed no change in the thermal stability of recombinant WT p53 after treatment with UCI-1001. This result was added to figure 5D.

Fig. 4B: Does UCI-1001 also alter the IF of p53-wt? 

We performed immunofluorescence (IF) staining using PAb1620, which recognizes wild-type p53. HCT-116 cells expressing WT-p53. We observed a subtle but significant increase in WT-p53 signals after treatment with UCI-1001 compared to vehicle treatment (DMSO). This may suggest that UCI-1001 could slightly stabilize the wt-fold of the, at 37˚C, typically metastable wild-type p53. It is also possible that there is a slight increase in protein level that is not detected by immunoblotting. We note that DSF experiments do not suggest stabilization of p53-WT by UCI-1001 (Fig. 5D). However, in vitro DSF is performed on monomeric p53 DNA-binding domains and p53 forms higher order structures such as dimers and tetramers in vivo. An effect of UCI-1001 on wt-p53 can thus be not excluded. We discussed this topic in the revised manuscript and the new experiment was added as Figure S1C.

Fig 5C: The specificity and p53 dependence of the expression profiles could be further supported by inclusion of a related, but “inactive” compound, such as UCI-1002, which does not bind to p53. 

As suggested by the reviewer we tested the effect of the inactive compound UCI-1002 on p53-dependent expression. We treated TOV-112D cell lines with 1 μM UCI-1002 (approximately 2.5 times the IC50 to compare with the equivalent compound concentration used for UCI-1001 in figure 4B) and observed no increase in the gene expression levels of the p53 target genes CDKN1A (p21) and PMAIP1 (NOXA), as shown in the new Figure S4.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors present the identification and characterization of activators of p53 mutants. The work is very complete using different techniques to characterize the activators and will be of interest to those working in cancer drug design. Just there are few points that deserve attention.

1.- What was the reason to choose a binding score of -7.1 kcal/mol as cut off? Please include the answer in the text.

2.- Please include in supplementary material a scheme explaining how the 450 studied were selected, the description in methodology section is not so clear.

3.- Please perform a structural analysis of the complex p53-UCI-1001 based on molecular dynamics simulation studies. 

Author Response

Point-by point response to reviewer’s comments

We would like to thank the reviewers for their insightful and productive comments. We have addressed the points raised by the reviewers as outlined below.

Please note that all changes/updates in the manuscript are highlighted in yellow.

We thank the reviewers for their time and critically reading our manuscript. We believe the changes further improve our manuscript. Note that some of the figure numbers have changed due to insertion of an additional figure (Fig. 2).

Reviewer 2: (reviewer comments in blue, authors’ response in black)

The authors present the identification and characterization of activators of p53 mutants. The work is very complete using different techniques to characterize the activators and will be of interest to those working in cancer drug design. Just there are few points that deserve attention.

1.- What was the reason to choose a binding score of -7.1 kcal/mol as cut off? Please include the answer in the text.

The best docking score for the entire library was -8.4 kcal/mol. An arbitrary docking score threshold of -7.1 kcal/mol was used to select 87 highest-scoring compounds which correspond to ~20% of the finally tested 450 compounds. The idea here was not to miss any hits among the highest-scoring compounds. This information is now in the Methods section 2.1 as detailed below in reply to comment 2.

2.- Please include in supplementary material a scheme explaining how the 450 studied were selected, the description in methodology section is not so clear.

Thank you for the feedback. We have added a schematic of our workflow as Figure S1A that explains the compound selection and modified the text about compound selection in Methods section 2.1.

3.- Please perform a structural analysis of the complex p53-UCI-1001 based on molecular dynamics simulation studies. 

In response to this comment, we have performed an additional MD simulation of the p53-UCI-1001 complex as suggested by the reviewer. We have added Figure 2 describing the results and added relevant text to the Results and discussion section as well as methods section.

 

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have addressed all points raised in my previous review and included substantial additional data, which further improved the manuscript and further supported the key hypothesis of the authors.