The potential of ITCs to prevent melanogenesis has been documented in a number of in vitro [
12,
13,
32‐
37] and in vivo [
14,
15,
38‐
41] studies. Overall, our results showed that exposure to AITC (2.5–50 µM) reduced viability in human A375 and Hs 294T as well as murine B16-F10 melanoma cells in a concentration- and time-dependent manner. In particular, AITC significantly reduced viability of these cells (at 10 µM onwards) while human VMM1, A431 and HaCaT cells remained relatively resistant. Moreover, of all these cell lines, only A375 cells appeared to be the most sensitive to the effect of AITC thereby providing the rationale for their subsequent use. To this end, it was apparent that AITC was capable of modulating the apoptotic response by mediating the differential expression of a number of genes representative of various apoptotic cascades (e.g., intrinsic, extrinsic, p53-dependent apoptosis, etc.) upon exposure to A375 cells.
In the context of regulating gene expression, both acetylation and methylation of histone proteins have been known as important modulators primarily through changes in chromatin structure. Specifically, regarding melanoma pathogenesis, overexpression of class I and II HDACs has been associated with the disease progression and drug resistance [
42‐
45]. On the other hand, ITCs have recently been reported as potent HDAC inhibitors thus disrupting the ratio of HAT/HDAC in a manner capable of inducing cell death in various cancers [
46‐
48]. Furthermore, inhibition of these enzymes has been associated with modulation in the expression of genes involved in tumor suppressor mechanisms including those of Nrf-2-dependent-detoxification of xenobiotics, cell cycle inhibition and apoptosis-induced cancer cell death [
17,
18,
21,
49,
50]. Among the genes reported to be regulated by HDAC inhibitors, the re-activation of p21
WAF1/Cip1 resulting in cell cycle inhibition and apoptosis is the most common one [
50‐
54]. In this study, our data revealed a reduction in protein expression levels of HDACs 4 and 6 but without a significant decrease in total HDAC activity. Similarly, there was a reduction in protein expression levels of CBP and acetyl CBP/p300 but also without an accompanied decrease in the activity levels of HATs, upon AITC exposure. To this end, work by others has shown that inhibition of CBP/p300 promotes cell cycle arrest and cellular senescence, deregulates DNA/damage response and induces apoptosis in melanoma cells [
55‐
57]. Such findings suggest that AITC could act as a potent HAT inhibitor capable of suppressing melanoma cell proliferation. In addition, we evaluated the histone acetylation status on specific lysine residues, at both H3 and H4 N-terminus, and we observed a dramatic decrease on the acetylation levels of lysines 5 (H4K5Ac), 8 (H4K8Ac) and 12 (H4K12Ac) on histone H4. Of these, H4K8 and H4K5 are known to be target sites for the action of CBP/p300 as this HAT is being known to preferentially acetylate these particular lysine residues [
58]. In comparison, there were no significant changes associated with the acetylation levels of lysines 9 (H3K9Ac), 14 (H3K14Ac), 18 (H3K18Ac), and 27 (H3K27Ac) of histone H3 upon exposure to AITC (data not shown). Our data, also revealed that combined exposure of AITC with panobinostat (known as an HDAC inhibitor [
59‐
61]) increases the acetylation status of H4K5, H4K8 and H4K12 which, in turn, suggests that inhibition of HDACs could lead to a higher turnover of HATs (perhaps as a compensation mechanism) leading to higher acetylation levels in these lysine residues. Furthermore, co-exposure of AITC with anacardic acid (known as a HAT inhibitor [
62‐
64]) abrogates the effect of AITC on the de-acetylation status of H4K5, H4K8 and H4K12, and, in such case, it restores the acetylation status of these lysine residues back to their control levels. Finally, it is worth mentioning that inhibition of total DNA methylation by decitabine [
65,
66] did not show any impact in the context of rescuing A375 cells from the observed AITC-induced cytotoxicity suggesting that such cytotoxicity is not linked to increased DNA methylation.
On another note, the extent of histone methylation (mono-, di-, and tri-) has also been shown to influence the extent of acetylation on H3. It is noteworthy that we have observed SET7-9 to be downregulated in this study. This histone methyltransferase (HMT) enzyme is known to catalyze the mono-methylation of H3K4 and is also associated with the methylation of non-histone proteins including p53. The role of this HMT in carcinogenesis is controversial as some studies report its tumor suppressor function [
67,
68] while others associate its activity with increased proliferation [
69]. Overall, among all di- and tri-methylated lysines on histone H3 that we examined (K36me2/me3, K4me2/me3, K79me2/me3, K27me2/me3 and K9me2/me3), it was observed that only the expression levels of H3K4me3 were significantly reduced upon exposure to AITC. Specifically, this is an epigenetic modification capable of regulating gene expression by means of activating the transcriptional process. Although one of the least abundant histone modifications, it is used as an epigenetic mark in order to identify active gene promoters [
70,
71].
To conclude, we have shown a significant involvement of AITC in regulating the epigenetic response by modulating specific lysine acetylation(s) and/or methylation(s) on histone proteins H3 and H4 as well as the expression of enzymes capable of catalyzing such epigenetic modifications (Fig.
6). In principle, such a response can impact on transcriptional activation and/or repression and consequently alter the outcome of gene expression. To our knowledge, this is the first report documenting a detailed characterization of the interaction of AITC with the epigenome, in human malignant melanoma, a finding that highlights the importance of dietary interventions in regulating the epigenome as a result of their action against various types of cancer.