We appreciate the comments by Doctor Vale on our article. His comments are appropiate in regard to the possibilty that epigenetic therapy can act as double-edged sword. Although not mentioned in Doctor Vale´s comment, the possibility that the transcriptional effects of epigenetic agents could not only down-regulate tumor suppressor genes but also up-regulate tumor promoting genes or oncogenes has been pointed out several times before. However, this apparently unspecific way to modifiy the expression of hundreds of genes is not so. Repeatedly, global expression analyses employing DNA demethylating and HDAC inhibitors show that no more than 2% of the whole transcriptome is indeed transcriptionally changed. This suggests that the reduced transcriptional effects could be explained on the basis that these agents may preferably target those “abnormaly silenced genes” regardless whether they are tumor suppressor genes or oncogenes. However, this apparently selectivity for abnormally silenced genes needs to be proven in experimental systems.

On the other hand, it is widely known that anticancer agents, in spite of of their mechanism of action, commonly have differential effects regarding their antiproliferative and antiapoptotic actions mostly depending on the models investigated, which can only reflect the intrinsic and specific genetic and epigenetic abnormalities that tumors have, which also can be referred as “tumor heterogeneity”.
Therefore, it is not unexpected that epigenetic drugs may activate (at transcriptional or protein levels) molecules and/or pathways with antiapoptotic actions. However, the overall effect may still be growth inhibitory upon tumors. The data provided by Pan et al [1] is a clear example where valproic acid exerts antiapoptotic neuroprotection in the SH-SY5Y neuroblastoma cells against oxidative stress and endoplasmic reticulum stress but not from against thapsigargin-evoked cell death and caspase-3 activation. This is not the only example where valproic acid exerts cell protection. Although by other mechanisms, this drug also offers considerable cardiac protection in a hemorrhagic shock model [2]. Regarding the theoretical risk of stimulate hematopoietic stem cell proliferation and self-renewal with this drug as shown in experimental systems [3] the overall effect goes in the other direction as not only valproic acid but other HDAC inhibitors have alone or in combination with other agents, strong antileukemia activity [4-6].

With respect to the PI3K/Akt system, a known and critical antiapoptotic pathway frequently active in cancer cells, indeed, trichostatin A, a prototype HDAC inhibitor has been shown to transiently activate this pathway and stimulate survivin expression in Caov-3 ovarian cancer cells, yet this transient action still is accompanied by growth inhibitory effects in that system. Nevertheless, if this transient activation of PI3K/Akt system is avoided by LY294002 –a PI3K inhibitor- the growth inhibitory effects of trichostatin A are increased [7]. These observations clearly agree with Doctor Vale´s suggestions on adding these kind of inhibitors to increase the antitumor effects of hydralazine and valproate. To further support these observations, we have also observed in microarray analyses that hydralazine and valproate in at least two tumor systems up-regulate PI3K despite the overall cellular effect is inhibitory. This could happen at least partly because valproic acid –as well as butyrate- inhibit the expression and activation of AKT and robustly activate caspase-9 [8]. In ongoing laboratory work from our group, we have also observed AKT inhibition at the phosphorylation level with valproic acid.

The precise mechanism by which epigenetic cancer drugs work is still unclear. It is possible that the activation and concurrent deactivation of specific genes and pathways are the responsible for their antitumor effects but is also possible that their widespread changes at the transcriptome (regardless of specif genes and pathways) just disrupt the tight gene regulation of the malignant phenotype in a direction that facilitates tumor cell death. This view is supported by the “gene addiction” concept where cells need to have a “certain amount” of individual genes (either suppressors or oncogenes) expressed as cells are extremely dependant on that degree of expression [9]. If this tune is disrupted, cells could die. In conclusion, more research is clearly needed on cancer epigenetics, and how this type of drugs work to optimize their use in cancer therapy, particularly in a combinatorial approach using them along classical cytotoxic drugs or with another molecular-targeted therapies.

1. Pan T, Li X, Xie W, Jankovic J, Le W (2005) Valproic acid-mediated Hsp70 induction and anti-apoptotic neuroprotection in SH-SY5Y cells. FEBS Lett 579:6716-6720.

2. Gonzales E, Chen H, Munuve R, Mehrani T, Britten-Webb J, et al (2006) Valproic acid prevents hemorrhage-associated lethality and affects the acetylation pattern of cardiac histones. Shock 25:395-401.

3. Bug G, Gul H, Schwarz K, Pfeifer H, Kampfmann M, et al (2005) Valproic acid stimulates proliferation and self-renewal of hematopoietic stem cells. Cancer Res 65:2537-2541.

4. Cimino G, Lo-Coco F, Fenu S, Travaglini L, Finolezzi E, et al (2006) Sequential Valproic Acid/All-trans Retinoic Acid Treatment Reprograms Differentiation in Refractory and High-Risk Acute Myeloid Leukemia. Cancer Res 2006 66:8903-8911.

5. Garcia-Manero G, Kantarjian HM, Sanchez-Gonzalez B, Yang H, Rosner G, et al (2006) Phase 1/2 study of the combination of 5-aza-2'-deoxycytidine with valproic acid in patients with leukemia. Blood 108:3271-3279.

6. Ten Cate B, Samplonius DF, Bijma T, de Leij LF, Helfrich W et al (2007) The histone deacetylase inhibitor valproic acid potently augments gemtuzumab ozogamicin-induced apoptosis in acute myeloid leukemic cells. Leukemia 21:248-252.

7. Zhour C, Qiu L, sun Y, Healey S, Wanebo H, et al (2006) Inhibition of EGFR/PI3K/AKT cell survival pathway promotes TSA's effect on cell death and migration in human ovarian cancer cells. Int J Oncol 29:269-278.

8. Chen J, Ghazawi FM, Bakkar W, Li Q (2006) Valproic acid and butyrate induce apoptosis in human cancer cells through inhibition of gene expression of Akt/protein kinase B. Mol Cancer 5:71.

9. Weinstein IB, Joe AK (2006) Mechanisms of disease: Oncogene addiction--a rationale for molecular targeting in cancer therapy. Nat Clin Pract Oncol 3:448-57.