*3.11. Epigenetics and Mirnas*

Epigenetics refers to heritable phenotype changes, in which DNA sequence alteration are not included. The most known epigenetic mechanisms involve DNA methylation and histone modification [184]. Extensively, epigenetics concerns all events "outside" DNA sequence modification; indeed, this paragraph wants to describe those mechanisms characteristic of DNA methylation and histone modification involving EGCG. Numerous works have been reported the significant impact of EGCG in epigenetic mechanism. For example, in 2003, EGCG was first studied as an inhibitor of DNA hypermethylation of CpG islands by acting on 5-cytosine DNA methyltransferase (DNMT). EGCG was used at 20 and 50 μM with concentration-dependent e ffects in human esophageal cancer cells (KYSE 510), colon cancer HT-29, and prostate cancer PC3 cells [185]. The authors showed that EGCG could block DMNT and reactivate methylated-silenced genes, as p16(INK4a), retinoic acid receptor β (RARβ), O-(6)-methylguanine methyltransferase (MGMT), and human mutL homologue 1 (hMLH1). Similar results were obtained in human breast cancer cell lines (MCF-7 and MDA-MB-231), whereas EGCG was able to block prokaryotic SssI DNA methyltransferase (DNMT) and human DNMT1 [186]. A recent work used the same cell lines (MCF-7 and MDA-MB-231) investigating RARβ expression and methylation, as mentioned in the first paragraph [52]. EGCG combined with 2-chloro-2'-fluoro-2'-deoxyarabinosyladenine (a deoxyadenosine analogue) reduced RARβ methylation with consequent increase of its expression, together with PTEN and CDKN1A, and with apoptosis induction and cell proliferation inhibition. EGCG acted on inhibitor of matrix metalloproteinases (TIMPs) and MMPs, being highly involved in tumorigenesis, in human prostate cancer cells (DUPRO and LNCaP) [187]. The authors showed that EGCG restored TIMP-3 expression and concomitantly increased histone H3K9/18 acetylation and decreased class I histone deacetylase, impacting on cell invasion and migration. Interestingly in serum of patients undergoing prostatectomy consuming 800 mg EGCG up to 6 weeks (participating to clinical trial NCT01340599), TIMP-3 levels were increased, suggesting a potential mechanism of action of EGCG in prostate cancer [187]. However, the trial failed to show a significant benefit of EGCG-treated prostate cancer patients [188].

Moreover, in a SKH-1 hairless mouse model, EGCG added to a hydrophilic cream and applied to approximately 1 mg/cm<sup>2</sup> skin area could protect against photocarcinogenesis through the reduction of tumor incidence, tumor multiplicity, and tumor size [189]. Mechanisms at the basis of these results are related to the inhibition of DNA hypomethylation induced by ultraviolet B treatment. Likewise, in human skin cancer cells (A431) and squamous cell carcinoma (SCC 13) cells, EGCG could reactivate p16INK4a and Cip1/p21, impacting not only on DNA methylation and DNMT activity, but also on histone deacetylase (HDAC) activity and acetylated histones [190]. Again, in CD4+ Jurkat T cells, EGCG 50 μM could inhibit DNMT and induce a concomitant Foxp3 (a factor regulating the development and function of regulatory T cells) and IL-10 expression. Moreover, in mice that were treated with EGCG (injected intraperitoneally daily per mouse 50 mg/kg), there was an increase of regulatory T cells frequencies and numbers [191]. In human malignant lymphoma cells (CA46), EGCG alone (6 μg/mL) or combined with trichostatin A (inhibitor of HDACs) diminished p16INK4a gene methylation (thus increasing its expression) and provoked a cell proliferation decrement [192]. The effects were more evident with the combination of the compounds, rather than EGCG alone, being suggestive an additive or synergist e ffect. In human breast tumor tissues, expression of DNMTs was found increased by Mirza et al., consequently MCF-7 and MDA MB 231 cells were used as models for testing EGCG and other compounds [193]. The authors showed that EGCG decreased DNMT1, HDAC1, and methyl-CpG binding protein 2 (MeCP2 essential for nerve function), with a reversal of epigenetic changes present in cancer cells. On the same line, another work investigated a co-treatment of 5-aza-2'-deoxycytidine and EGCG in breast cancer cells (MCF-10A) and confirmed the e ffects on DNA methylation and histone modifications [194]. In a mouse lung tumor model that was induced by the tobacco-specific carcinogen 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), EGCG (0.5% in diet *ad libitum*) was implicated in DMNT1 reduction with simultaneous decrease of p-AKT, phospho-histone H2AX and tumor growth, again emphasizing the epigenetic e ffects of the catechin, potentially due the antioxidant activity of EGCG, as suggested by the authors. [195]. Furthermore EGCG was studied in the methylation status of the reversion-inducing cysteine-rich protein with Kazal motifs (RECK) gene, a tumour suppressor gene downregulating matrix metalloproteinases and blocking metastatic potential [196]. It has been shown in a salivary adenoid cystic carcinoma cells (SACC83) that the catechin was able to demethylate RACK gene and thus impact on invasion, migration, and metastasis. EGCG has been also implicated in autophagic process, as seen in Paragraph 3.1. It is known that in macrophages from aged mice Atg5 and LC3B genes are normally hypermethylated (genes regulating autophagy). The use of EGCG (probably a high dose, even if the authors did not report any cell viability change) in macrophages restored these genes expressions both in vivo and in vitro [197]. Similarly, in macrophages cells of cystic fibrosis (CF) mice, EGCG (administered intratracheally for three days with 25 mg/kg) restored Atg12 gene expression, improving autophagy in such a disease, where autophagic process highly impaired [198]. EGCG (10 μM given diabetic pregnan<sup>t</sup> mice at embryonic day 5.5 in drinking water) inhibited DNA hypermethylation in maternal diabetes-induced neural tube defects, decreasing the congenital defects and opposing the high glucose levels in embryo [199]. In another animal models, C57BL/6J male mice thatwere fed with high-fat diet (HFD) or control diet (CD), EGCG administered *ad libitum* at 25 mg/kg body weight per day in water modulated DNMT1 and MutL homologue 1 genes methylation, ameliorating inflammatory status in HFD mice [200]. In addition, EGCG enhanced gu<sup>t</sup> microbiota, as evidenced by DNA damage in the *Firmicutes*/*Bacteroidetes* ratio and the total number of bacteria in the same work. In HeLa cells, EGCG abrogated DNMT3B and HDAC1 activity, while only the expression of DNMT3B was reduced, also affecting expression of retinoic acid receptor-β, cadherin 1, and death-associated protein kinase-1 [201]. A genome-wide study of DNA methylation reported the e ffects of EGCG in oral squamous cell carcinoma (CAL-27) [202]. The results showed that the main di fferentially expressed genes that were derived from MAPK, Wnt, and cell cycle signaling pathways, cell processes that are strongly associated cancer proliferation.

MicroRNA (miRNA) is a small non-coding RNA molecule that contains about 22 nucleotides that can post-transcriptionally regulate gene expression [203]. miRNAs are involved in numerous cell processes, including growth, proliferation, apoptosis, autophagy, metabolism, etc., and play a fundamental part in cancer. In prostate cancer cells (PC3) and PC3 mouse xenografts, EGCG blocked nuclear translocation and protein expression of androgen receptor (implicated in prostate cancer etiology), and these effects were associated with the down-regulation of androgen-regulated miRNA-21 and up-regulation of miRNA-330, a tumor suppressor [204]. Another target of EGCG was miRNA-210, being upregulated by the catechin in lung cancer cell lines (CL13, H1299, H460, A549), with induced accumulation of HIF-1<sup>α</sup>, which is a known miRNA-210 activator [205]. Moreover, in tobacco carcinogen-induced lung tumors in A/J mice, miRNAs profile was analysed after EGCG treatment (mice fed with purified diet containing 0.4% EGCG) [206]. It has been revealed that these miRNAs were strictly connected with MAPK, AKT, NF-κB pathways, and cell cycle regulation. Rat hepatoma cell line (FAO) were used to test different extracts and polyphenols, including EGCG [207]. The catechin reduced miR-33a and miR-122 levels (both being involved in dyslipidemia, cholesterol homeostasis, insulin resistance), with direct binding to miRNAs, as demonstrated by 1H NMR spectroscopy and, thus, could potentially impact cell metabolism. In non-small cell lung cancer cells (A549) mouse xenografts, EGCG that was associated with cisplatin was able to reduce tumor size, while in NCI-H460 cells (another non-small cell lung cancer cell line) was not [208]. This difference was imputable to different expression of miRNA-98-5p and miRNA-125a-3p after EGCG treatment, consequently, the authors suggested that EGCG and cisplatin could have a therapeutic role in some subtype of non-small cell lung cancer and could conveniently be associated. In nasopharyngeal carcinoma cell line (CNE2) differential miRNAs expression was observed, in particular 16 miRNAs with 20 μM EGCG-treated and 32 miRNAs with 40 μM EGCG-treated were modulated (>2-fold expression changes when compared with the control) [209]. Furthermore, while using an in vitro subarachnoid hemorrhage model (oxyhemoglobin added to PC12 cells), EGCG at 1 and 50 μM altered miRNAs expression and mainly impacted on the MAPK pathway (p38, Ca2+, autophagic activation), suggesting that the differential expression of miRNAs could be at the basis of the therapeutic efficacy of different concentrations of EGCG [210]. EGCG in a dose- and time-dependent manner inhibited the proliferation and regulated miRNAs expression of miR-210, miR-29a, miR-203, and miR-125b in different cervical carcinoma cells in multiple human cervical cancer cell lines (HeLa (HPV16/18+), SiHa (HPV16+), CaSki (HPV16+), and C33A (HPV-)) [211]. In osteosarcoma cell lines (MG-63 and U-2OS), EGCG induced miRNA-1 upregulation and inhibited c-MET expression, while its combination with c-MET inhibitor enhanced inhibitory effects on tumor cell growth, heading to cell proliferation reduction, cell cycle arrest, and apoptosis induction [212]. EGCG downregulated Ca2+ (CRAC) channel (Orai1), its regulator stromal cell-interaction molecule 2 (STIM2), and store operated Ca2+ entry (SOCE), all being implicated in Ca2+ regulation, in murine CD4+ T cells and human Jurkat T cells. The work showed that EGCG repressed Ca2+ entry by up-regulation of miR-15b, which ruled STIM2, Orai1, SOCE, and acted on AKT/PTEN pathway [213]. In anaplastic thyroid carcinoma cells (SW1736 and 8505C), EGCG and other natural compounds had heterogeneous effects, with curcumin being the most effective [214]. Nonetheless, EGCG provoked a significant reduction of miRNA-221 only, even if five important miRNAs that were implicated in thyroid cancer progression were studied (miRNA-221, miRNA-222, miRNA-21, miRNA-146b, and miRNA-204). In multiple myeloma cells (MM1.s), EGCG (1 and 5 μM) was shown to inhibit miRNA-25, miRNA-92, miRNA-141, and miRNA-200a, all being implicated in p53 targeting reduction, a hallmark of multiple myeloma [215]. The suppression of these miRNAs could have a potential effect in the treatment of such disease, as the authors claimed that the EGCG doses were physiologically relevant and within the range of human consumption. In pancreatic ductal adenocarcinoma cell lines (BxPc-3, MIA-PaCa2, and CRL-1097), EGCG and other catechins (also associated with quercetin and sulforaphane) were used and showed to reduce apoptosis, migration, and the matrix metalloproteinases MMP-9 and MMP-2, with the activation of miR-let-7 leading to the inhibition of its target gene K-Ras [216]. Of note that EGCG was not among the most effective compounds, whereas the combination of sulforaphane, quercetin, and catechins extract seemed to have a strong therapeutic action.
