Reader Comments

Post a new comment on this article

Referee comments: Referee 1 (Francois Morle)

Posted by PLOS_ONE_Group on 17 Apr 2008 at 16:01 GMT

Referee 1's review (Francois Morle):

N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication the manuscript has been revised in light of these comments and to address other editorial requirements.

The same group already reported that EWS/FLI activates NKX2.2 and that NKX2.2 protein critically contributes to the oncogenic phenotype of Ewing's sarcoma (R Smith et al, 2006, Cancer Cell 9: 405-416). In this previous study, the authors nicely showed that A673 cells loose their ability to generate colonies in soft-agar following endogenous NKX2.2 knockdown using constitutive shRNA expression and that this transformed phenotype can be rescued by exogenous NKX2.2 expression. In the present study, the authors use this very elegant "knockdown-rescue" approach, to demonstrate that the known transactivation domain of NKX2.2 is completely dispensable to rescue the transformed phenotype but that a mutant NKX2.2 protein containing only the DNA binding and repressor domains is sufficient for that. New microarrays analyses of A673 cells expressing or not NKX2.2 or EWS/FLI are presented which indicate that NKX2.2 only contributes to the downregulation of a few number of genes (159), the half of which (72) overlap with 2% of the 3075 genes which are downregulated by EWS/FLI. Further experiments indicate that the transformed phenotype of A673 cells can be suppressed by inhibiting the activity of two known classes of NKX2.2 corepressors: the TLE corepressors (through the expression of a TLE dominant negative protein) or HDAC (through the addition of the HDAC inhibitor SAHA). Finally, statistical analysis of microarrays data from A673 cells grown in the presence or absence of SAHA indicates that a significant proportion of the 159 genes downregulated by NKX2.2 are upregulated in the presence of SAHA. Based on these data, the authors conclude that "the transcriptional repressive function of NKX2.2 is necessary and sufficient for the oncogenic phenotype of Ewing's sarcoma".

Major comment
All experiments, statistics and other analyses are performed to high technical standard, the results are new, interesting and are worth to be published. However, my main concern is the ambiguity in the conclusions made by the authors.
In one hand, the results presented in this study convincingly demonstrate that the known contribution of NKX2.2 to the transformed phenotype of Ewing's sarcoma cells does not require its known transactivation domain and that NKX2.2 is mainly (if not exclusively) involved in the repression of a few number of genes. This in turn strongly suggests that the contribution of NKX2.2 to the transformed phenotype of Ewing's sarcoma is through transcriptional repression.
On the other hand, the written manuscript makes the reader understand that the genes repressed by NKX2.2 are direct target which are repressed through the direct recruitment of NKX2.2 with TLE and HDAC corepressors. However, the data presented in the manuscript are clearly not sufficient to drive this conclusion. I would suggest to the authors either to clarify this point in the redaction of the manuscript or to directly prove (for example by chromatin immunoprecipitation) that NKX2.2 is indeed present on the promoter of some of the repressed genes. Obviously the second option would significantly enhance the interest of the study.

Specific points
1- Figure 4G indicates that several of the 159 genes repressed by NKX2.2 seem to be downregulated by SAHA. This in turn would suggest that the repression of at least these particular genes does not depend on HDAC activity. It should be interesting to see if the 72 genes which are repressed by both EWS/FLI and NKX2.2 are dependent or not on HDAC activity.
2- The authors should discuss the possibility that NKX2.2 may repress transcription by competing with the recruitment of other positive transcription factor. I'm not sure that the present data completely exclude this possibility.
3- Figure 2C track TN-HD : what is the strong signal revealed by the flag antibody and which appears on the top of the gel ?
4- A Table (given eventually as a supplementary document) identifying the 72 genes which are downregulated by both EWS/FLI and NKX2.2 could be useful.