Reader Comments

Post a new comment on this article

Referee Comments: Referee 1

Posted by PLOS_ONE_Group on 12 May 2008 at 18:12 GMT

Referee 1's review:

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.

Bren and colleagues present studies to support their hypothesis that expression of HIV protease during viral replication results in procaspase 8 cleavage, resulting in the production Casp8p41. They hypothesize that casp8p41 then leads to NF-κB-mediated stimulation of HIV transcription and HIV production.

In the first experiment, Jurkat T cells are infected with HIV-IIIb and transfected with a TAR-deleted HIV LTR reporter expressing luciferase. Details of transfection efficiency and MOI of infection are not clear, and it seems likely that most cells will not be dually transfected and infected. Expression of the reporter plasmid is reported as fold relative to a mock control rather than in absolute terms, and does not increase until day 4 after infection/transfection. This is of interest, given that the reporter is TAR-deleted and may express poorly. Throughout the paper, information about the absolute level of LTR expression, as well as it relative change, would be informative.

Overall cell viability does not decline until day 4, although this does not appear to be compared to a control culture. Casp8p41 expression is not reported prior to infection/transfection, but increases 3-fold from day 1 to day 5, to 14%. Therefore, it is not clear that protease-mediated caspase cleavage in the first cells infected is leading to increased LTR reporter expression, or if this is due to an effect induced by the 2nd round of infection on cells previously transfected. It is a bit surprising that protease production in the 1rst Jurkat cells infected, which should rapidly produce virus, did not increase luciferase production before day 4. This experiment could be strengthened in various ways, including selection of transfected/infected cells, the use of protease inhibitors or caspase inhibitors, or HIV entry inhibitors.

The authors then report that HIV protease expression induces a 3-to-5-fold increase in LTR-reporter expression. This effect is also seen in caspase-deficient cells that are provided with caspase by transfection. The details of this experiment would be helpful, as again the absolute levels of activation are unclear in this system that lacks Tat activation. Again, this experiment might benefit by the use of HIV protease inhibitors as a control.
Caspase is then tested by overexpression in 293T cells of either FL or p41 caspase. This experiment would be strengthened by the demonstration of the levels of expression of the FL and p41 caspases, the absolute level of LTR expression, and a direct quantitation of increase of LTR-driven RNA expression rather than a measure of luciferase activity. The role of NF-κB is explored through the use of a reporter that lacks NF-κB sites. It seems that this is the same reporter that also lack a TAR region. Such crippled reporters generally express at extremely low levels, and thus the negative result reported might simply be due to the weakened state of this dually mutated promoter. Although there is an increase in NF-κB binding activity seen (Fig. 4B), the p41 caspase appears to induce more LTR expression without an increase in NF-κB compared to FL caspase. Further, while the p50 subunit of NF-κB is detected, the activating p65 subunit does not appear. p50 homodimers of NF-κB can be repressive (Williams EMBO 2007), although NF-κB binding activity may be due to factors other than p50 or p65.

Finally, the authors attempt a very difficult experiment involving transfection of HIV+ patients' cells. Infected cells in the circulation will be relatively rare, and transfection inefficient. Not unexpectedly, levels of p24 expression are near what would be the background of the assay (at least in our laboratory) in all but one patient.

Overall, I would ask the authors to consider the timing of protease expression and viral replication. Protease is a late gene; Tat will have been made in signficant quantities by the time protease is expressed. Is there a way to demonstrate that the hypothesized effect adds significantly to Tat activation?