Cucurbitacin B Alters the Expression of Tumor-Related Genes by Epigenetic Modifications in NSCLC and Inhibits NNK-Induced Lung Tumorigenesis

Lung cancer is the leading cause of cancer-related mortality worldwide, with nearly 1.4 million deaths each year (1). Non–small cell lung cancers (NSCLC) constitute almost 85% of total diagnosed cases and are caused mainly due to the exposure to the environmental carcinogens. The early stages of lung tumorigenesis involve reversible promoter methylation and histone modification changes (2). Bioactive natural compounds can be defined as the natural compounds that can exhibit a biological activity on the living cells. These compounds possess the capability to reverse the epigenetic alterations. Bioactive natural compounds have been shown to restore the expressions of to tumor suppressor genes (TSG) altered during the course of carcinogenesis (2, 3). Therefore, the uses of bioactive natural compounds have been a major focus of interest as therapeutic options against various classes of diseases, including cancer.

The use of DNA methyltransferase (DNMT) inhibitors alone or in combination with histone deacetylase (HDAC) inhibitors has shown promising results against various cancers through active chromatin modifications (4–7). However, use of synthetic inhibitors invites several limitations, including cost, dose regime, and adverse side effects. Therefore, studies have aimed to use natural DNMT inhibitors alone or in combinations with natural HDAC inhibitors for the prevention and treatment of cancers (7, 8). We have previously shown that dietary combination of green tea polyphenol, a DNMT inhibitor, and sulforaphane, HDAC inhibitor reactivates estrogen receptor (ER)-α in ER-α–negative human breast cancer cells through active chromatin modifications (6). In addition, natural compounds have shown to sensitize the cancer cells for the subsequent treatment with available therapeutics (3, 8). In this study, we explored the epigenetic modulatory activities of cucurbitacin B against NSCLC.

Cucurbitacin B (CuB) is a natural triterpenoid isolated from Cucurbitaceae plants, and it has been shown to have anti-plasmodial, immunomodulatory, free-radical scavenging, hepato-protective, cardiovascular, anti-inflammatory, anti-helminthic, and anti-fertility activities (9–12). Recently, there has been growing interest in CuB-mediated anticancer activities, mainly mediated through induction of G₂–M phase cell-cycle arrest, inhibition of the JAK/STAT pathway, and induction of apoptosis (13–19). However, these studies have not elucidated the core epigenetic mechanisms behind the anticancer activities of CuB. Our results demonstrate for the first time that CuB, a single bioactive natural compound, exhibits both HDACs and DNMTs inhibitory activities at sub-IC₅₀ concentrations in nanomolar range in NSCLC cells. CuB-mediated alterations in histone modifications and promoter methylation lead to induction of TSGs and downregulation of tumor promoter gene (TPG), which results in the inhibition of cellular growth and induction of apoptosis in human NSCLC. Furthermore, CuB treatment significantly inhibited the tumor incidence and multiplicity in 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK)-induced lung tumorigenesis in A/J mice, which was associated with the alterations in the expressions of DNMTs and HDACs. This altered expressions of epigenetic enzymes further led to changes in the expression patterns of TSGs, TPG, and oncogenes in the mice lung tissues.

CuB was isolated from the fruits of Luffa graveolense Roxb. as described in Supplementary Table S1. The structure of CuB was elucidated on the basis of spectroscopic analysis described in Supplementary Table S1, which was in good agreement with that of previously reported (20).

The human NSCLC cells were purchased from ATCC and no authentication was done by the authors except performing a mycoplasma test using the LookOut Mycoplasma PCR-detection Kit (Sigma-Aldrich). Cells were cultured and maintained in RPMI-1640 medium supplemented with 10% FBS and 1% penicillin/streptomycin solution in a humidified atmosphere at 37°C with 5% CO₂. CuB was dissolved in ethanol and stored at a stock concentration of 10 mmol/L at 4°C. The maximum final concentration of ethanol in the complete culture medium was less than 0.1% (v/v). Cell proliferation was assayed by standard MTT assay as described previously (6).

Approximately 2 × 10⁵ cells were serum-starved overnight and treated with different concentrations of CuB. After 48 hours, cells were harvested and then stained with 50 μg/mL of propidium iodide. Cell-cycle distribution was analyzed using a FACS-Caliber (BD Biosciences). CuB-induced apoptosis in the NSCLC cells were analyzed using the Alexa Fluor 488-Annexin V/Dead Cell Apoptosis Kit (Invitrogen) as described previously (6).

RNA isolation and RT-PCR were performed as described previously (6, 8, 21, 22). The primer sequences and annealing temperatures (T_A) are given in Supplementary Table S2.

The procedures have been described previously (6, 21). Briefly, equal amounts of protein were electrophoresced and transferred onto a 0.45-μm polyvinylidene difluoride (PVDF) membrane. Membranes were probed with antibodies against CDKN2A [p16^INK4A] (SC-1661), CDKN1A [p21^CIP1/WAF1] (SC-397), DNMT1 (SC-20701), DNMT3a (SC-20703), DNMT3b (SC-10235; Santa Cruz Biotechnology); SIRT3 (ab86671), KAT3B/p300 (ab10485; Abcam); MYC (06-340; Millipore); SIRT2 (12650S), SIRT6 (2590), HDAC 1-6 (9928S), RAS (3965), CBP, GCN5L2, PCAF (8686S), and β-actin (Cell Signalling). The membranes were then incubated with respective secondary antibody. Immunoreactive bands were visualized using the Enhanced Chemiluminescence Detection System.

The HDACs and DNMTs activities were analyzed using the colorimetric assay kits (Active Motif) as described previously (21, 22).

The effect of CuB treatment on the degree of genomic methylation was determined using dot-blot analysis, as described previously (6). Briefly, 1 μg of gDNA was transferred onto a nitrocellulose membrane and fixed by baking the membrane for 1 hour at 80°C. After blocking the nonspecific binding sites, the membrane was incubated with the anti-5-methyl cytosine (5-mC) antibody (MABE146; Millipore) for 2 hours. Furthermore, the membrane was incubated with horseradish peroxidase (HRP)- conjugated secondary antibody and bands were visualized using chemiluminescence detection system.

The DNA methylation statuses of the p21^CIP1/WAF1 and hTERT promoters were assessed by MSP using the EpiTect Bisulfite Modification Kit (Qiagen) following the manufacturer's protocol. After 48 hours of CuB treatment, 2 μg of gDNA was bisulfite converted and then PCR amplified using the Platinum Taq polymerase (Invitrogen). MSP primer sequences and respective annealing temperatures are given in Supplementary Table S3.

Chromatin immunoprecipitation (ChIP) assays were performed using the EZ-ChIP Kit according to the manufacturer's protocol (Millipore) as described previously (21, 22). The antibodies used in the ChIP assays were ChIP-validated anti-acetyl histone H3 (06-599), anti-acetyl histone H4 (06-598), anti-acetyl histone H3K9 (07-352), anti-histone H3K9me3 (07-442), anti-histone H3K27me3 (07-449; Millipore) and anti-MeCP2 (ab2828; Abcam). Input DNA was used as an internal control. No antibody control was used to check ChIP efficiency. ChIP-purified DNA was amplified by PCR and the reaction was initiated at 94°C for 5 minutes followed by 25 cycles of PCR (94°C, 30 seconds; T_A°C, 30 seconds; 72°C, 30 seconds) and extended at 72°C for 5 minutes. The promoter regions of p16^INK4A were amplified using touchdown PCR. The primer sequences and conditions are given in Supplementary Table S3.

Female A/J mice, 5 to 6 weeks old, were procured and handled according to the protocol approved by the Institutional Animal Ethics Committee, CSIR-CDRI. Mice were randomized into 4 groups (n = 8); three groups were injected i.p. with NNK dissolved in saline [Sigma; 100 mg/kg body weight (b.w)] and the control group was injected with saline as vehicle once a week for 2 consecutive weeks. On third week onward, NNK-administrated mice were treated with 0.1 or 0.2 mg/kg b.w. of CuB through i.p. twice a week for 10 weeks. CuB was dissolved in ethanol at a stock concentration of 5 mg/mL and diluted in PBS on the day of use. The maximum concentration of ethanol was less than 0.1% and the control mice were treated with the vehicle. After 13th week, CuB treatment was stopped and the mice were left untreated until they were sacrificed. At 21st week, mice were sacrificed and examined for tumor incidence, multiplicity, and lung wet weight.

Formalin-fixed lung tissue sections (5 μm) were processed for hematoxylin and eosin (H&E) staining. Three H&E sections, at least 600 μm apart, were analyzed blindly by a veterinary pathologist for different histopathologic lesions and graded according to the recommendations of the Mouse Models of Human Cancer Consortium as described previously (23, 24). The histopathologic lesions were scored on the basis of severity, from 0 to 4: grade 0: absent, lesions affect less than 5% of tissue; grade 1:mild, lesions affect 6%–20% of tissue; grade 2: moderate, lesions affect 21%–40% of tissue; grade 3: severe, lesions affect 41%–75% of tissue; and grade 4: extensive, lesions affect 76%–100% of tissue. Immunohistochemical analysis of proliferating cell nuclear antigen (PCNA) and terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) assays were performed as described previously (25). The numbers of PCNA- and TUNEL-positive cells were counted in at least 6 different microscopic fields as the average number of positive cells × 100/total number of cells.

Statistical significance of differences between the values of untreated controls and CuB-treated groups was determined with one-way ANOVA using Dunnet's multiple comparison test on GraphPad Prism version 5.00 for Windows. In each case, P < 0.05 was considered statistically significant.

To determine the effective dose of CuB on human NSCLC cells, we performed cellular proliferation, cell-cycle arrest, and apoptosis analyses. We observed a dose- and time-dependent inhibition of cellular proliferation with CuB treatment in A549, H1299, and H1650 cells as shown in Supplementary Fig. S1A. Inhibition of cell growth was more prominent in the p53⁻ H1299 cells as compared to p53⁺ A549 cells and p53⁺ EGFR⁻ H1650 cells (Fig. 1A). As shown in Fig. 1A–C, CuB significantly inhibited cellular proliferation and induced cellular apoptosis as well as G₂–M phase cell-cycle arrest in H1299 cells at an IC₅₀ concentration of 60 nmol/L (P < 0.05) and onward, therefore further detailed analysis of molecular mechanism of its action was performed in this cell line with 60 nmol/L as median concentration. In addition to H1299 cells, A549 cells also showed dose-dependent G₂–M phase cell-cycle arrest and induction of cellular apoptosis with CuB treatment (Supplementary Fig. S1B and S1C).

As shown in Fig. 1D, CuB at concentrations of 60 nmol/L or higher significantly increased the mRNA and proteins expression of key cell-cycle regulatory TSGs such as p16^INK4A and p21^CIP1/WAF1 in H1299 cells. However, we found slightly decreased mRNA expressions of these TSGs at 600 nmol/L as compared with 60 nmol/L CuB concentrations, which might be due to the mRNA instability as well as higher rate of protein turnover observed in case of cell-cycle regulators (26, 27). Overall, these mRNA and protein expressions are consistently upregulated by the CuB treatment in H1299 cells irrespective of dose responses. Treatment with CuB in H1299 cells downregulated the protein expressions of oncogenes, c-MYC, and K-RAS; however, their mRNA expressions were not altered significantly (Fig. 1E). As shown in (Fig. 1F), CuB treatment significantly reduced the expression of hTERT, the catalytic subunit of telomerase, starting with a dose of 6 nmol/L onwards at 48 hours. Therefore, these results suggest that the antiproliferative and proapoptotic effects of CuB are, at least in part, associated with the altered expressions of key tumor-related genes in NSCLC cells.

Genetic alterations are accompanied by epigenetic modulations during carcinogenesis (2, 6). Hence, we analyzed the expressions of both types of DNMTs, maintenance methyltransferase DNMT1, and de novo methyltransferases, DNMT3a and DNMT3b. As shown in Fig. 2A, CuB treatment significantly inhibited the expressions of DNMT1 and DNMT3b starting from CuB concentrations of 600 nmol/L or higher in H1299 cells. Interestingly, the protein expressions of all the 3 DNMTs were drastically reduced starting from IC₅₀ concentration of CuB in H1299 cells (Fig. 2B). Furthermore, we have also observed a significant inhibition of DNMTs activity starting with the CuB concentrations of 200 nmol/L in another NSCLC A549 cell line (Supplementary Fig. S1D). In addition to DNMTs, HDACs, especially HDAC1, HDAC5, and HDAC6 were found to be considerably downregulated by CuB starting with 60 nmol/L concentrations in H1299 cells. NAD⁺-dependent class III HDACs, SIRT3, and SIRT6 were also found to be downregulated at 60 nmol/L or higher concentrations of CuB. We further analyzed the protein expressions of different histone acetyltransferases (HAT) such as CBP, GCN5L2, PCAF, and p300 after CuB treatment in H1299 cells for 48 hours. We found that CBP and PCAF expressions were marginally upregulated at sub-IC₅₀ and IC₅₀ concentrations of CuB (Fig. 2C). Furthermore, CuB not only inhibited the expressions but also dose dependently inhibited the activity of HDACs in NSCLC cells (Fig. 2D; Supplementary Fig. S1D).

As we found that the treatment of H1299 cells with CuB inhibited the expression of DNMTs, we further investigated the consequence of their inhibition at the extent of global as well as gene-specific promoter methylation. Treatment of CuB in H1299 cells resulted in a significant dose-dependent reduction in 5-mC levels, a marker for global DNA methylation (Supplementary Fig. S2A). Furthermore, as shown in Supplementary Fig. S2B, gene-specific promoter methylation of p21^CIP1/WAF1 and hTERT was also found to be significantly downregulated at 60 nmol/L of CuB concentration in H1299 cells.

ChIP assays were performed to investigate CuB-mediated changes in binding of various active and inactive chromatin marks at the p16^INK4A, p21^CIP1/WAF1, and hTERT promoters. As shown in Fig. 3A, we designed 5 sets of primers for the CpG-rich p16^INK4A promoter regions upstream of the transcription start site. There are 2 Alu sites in the amplicons of the distal primers, ChIP 1 and ChIP 2, which are densely methylated and transcriptionally inactive. Therefore, the proximal promoter regions amplified by ChIP 3-5 more actively participate in the transcriptional activation of p16^INK4A. As shown in Fig. 3B and C, we selected 3 different histone acetylation marks such as acetyl histone 3 (AcH3), acetyl histone 4 (AcH4), and acetylation at lysine 9 of histone 3 (AcH3K9) as active chromatin marks. We found significantly higher enrichment of AcH3 at both proximal and distal promoters of p16^INK4A in H1299 cells after 48 hours of CuB treatment. This observation indicates that the HDAC inhibitory activity of CuB (Fig. 2D) leads to histone hyperacetylation especially in case of histone 3. The enrichment of AcH4 was almost unaltered except for the 2-fold increase in ChIP 3. Furthermore, we also found a significant decrease in the methylation status of inactive chromatin marks such as trimethylation at lysine 9 of histone 3 (H3K9me3) and trimethylation at lysine 27 of histone 3 (H3K27me3) at proximal regions (ChIP 3-5) of the p16^INK4A promoter in H1299 cells with CuB treatment.

In case of the p21^CIP1/WAF1 promoter, we designed 2 distinct sets of primers for 2 functionally important regions of the distal promoter as shown in Fig. 4A. The primer pair amplifying ChIP 1 flanks a sp1-binding site on the p21^CIP1/WAF1 promoter. Enrichment of AcH3 was found to be significantly increased at CuB concentrations of 6 and 60 nmol/L (P < 0.05), whereas the enrichment of AcH4 and AcH3K9 showed no significant alteration at both the promoter sites (Fig. 4B and C). Binding of inactive chromatin marks such as H3K9me3 and H3K27me3 at the p21^CIP1/WAF1 ChIP 1 promoter region was decreased significantly with CuB treatment. Furthermore, our results also evidenced a p53-independent p21^CIP1/WAF1 upregulation with CuB treatment in p53⁻ H1299 cells.

Similarly, we designed 5 sets of primers for the hTERT promoter spanning the regions from distal promoter to the first exonic region (Fig. 5A). As shown in Fig. 5B and C, the relative enrichment of AcH3 was increased in a range between 1.5- and 5.5-fold (P < 0.05) in CuB-treated H1299 cells at sub-IC₅₀ and IC₅₀ concentrations, respectively, starting from distal to proximal promoters of hTERT. The other 2 active acetylation marks AcH4 and AcH3K9 showed a marginal increase in enrichment at some of the promoter regions in the hTERT. In accordance with histone acetylation, histone methylation marks were also enriched up to 1.5- to 2.5-fold (P < 0.05) at distal as well as proximal promoter regions of hTERT. The enrichment of MeCP2 was found to be increased at the distal promoter sites (ChIP 1 and ChIP 2) and first exonic region (ChIP 5) in a range lying between 2.5- and 3.6-fold (P < 0.05) higher than the untreated H1299 cells. The proximal promoter and first exonic region of hTERT have binding sites for many important transcription factors and preoccupancy of these sites with inactive chromatin marks might hinder their binding to the hTERT promoter leading to the downregulation of hTERT expression (21). Furthermore, we observed a significant disruption of the enrichment of c-MYC at the selected regions of hTERT promoter after CuB treatment in H1299 cells (Fig. 5B and C). Collectively, these results suggest that the CuB-mediated chromatin modifications are, at least in part, responsible for the altered expressions of p16^INK4A, p21^CIP1/WAF1, and hTERT in H1299 cells.

Next, we sought to determine the effect of CuB on NNK-induced lung tumorigenesis. The schematic experimental design is given in Fig. 6A. We found no significant difference in the average body weights of mice in the individual groups (Supplementary Fig. S3A). Histopathologic examination revealed that the lungs of NNK-induced mice showed a highest grade of bronchiolar and alveolar hyperplasia, adenocarcinoma, and microadenomas compared with the lungs of vehicle-treated control mice (Fig. 6B; Supplementary Fig. S3B and Supplementary Table S4). Tumor angiogenesis, which is marked by the formation of new and irregular blood vessels, was also prevalent in the NNK-induced lungs, which decreased dose dependently in CuB-treated groups (Fig. 6B; Supplementary Fig. S3B and Supplementary Table S4). As shown in Fig. 6C, NNK-administrated mice had significantly higher incidence of lung cancer (100%) and tumor multiplicity (17.75 ± 7.4 lung tumors per mouse) compared with vehicle-administered control mice. Interestingly, treatment with CuB resulted in a significantly reduced tumor incidence and tumor multiplicity compared with vehicle-treated NNK-induced lung tissues (Fig. 6C).

Furthermore, as shown in Fig. 6D, the presence of PCNA was highest in the NNK-administrated control lungs. The percentage of PCNA-positive cells was significantly (P < 0.01) lowered in 0.1 and 0.2 mg/kg b.w. CuB-treated mice lungs as compared with NNK-administrated vehicle-treated mice lungs. Next, we sought to determine the effect of CuB on cell death in NNK-induced lung cancer using TUNEL assay. Immunohistochemical analysis revealed a significantly higher number of TUNEL-positive cells at 0.1 and 0.2 mg/kg b.w. CuB-treated mice lungs compared with NNK-induced mice lungs (Fig. 6D). Overall, CuB-inhibited hyperproliferation and enhanced cell death in NNK-induced hyperproliferative lung cancer in A/J mice.

To further validate our in vitro results, we analyzed the expressions of epigenetic modulatory enzymes in the CuB-treated and untreated NNK-induced mice lungs. As shown in the Fig. 6E, CuB treatment led to decreased protein expressions of DNMTs. The DNMT inhibitory effects were more pronounced at 0.2 mg/kg b.w. CuB dose. Similarly, CuB also inhibited the protein expression of different HDACs (HDACs 1–4), with considerably higher inhibition of HDACs expression at 0.2 mg/kg b.w. CuB dose. Furthermore, we have observed that CuB-induced significant upregulation of p16^INK4A and p21^CIP1/WAF1, which is consistent with our in vitro results. However, we also found a significantly increased expression of p16^INK4A in the NNK-induced group (Fig. 6F). The expression of tumor promoter Tert was significantly inhibited in 0.1 and 0.2 mg/kg b.w. CuB-treated animal groups, further validating our in vitro results (Fig. 6F). In accordance, we have also observed a considerable downregulation of the expressions of oncogenes c-MYC and RAS in CuB-treated mice lungs compared with NNK-induced lung tissue (Fig. 6F).

The upregulation of TSGs and downregulation of TPGs have been causally linked with the inhibition of cellular proliferation and induction of cellular apoptosis (4, 8, 28). In our present study, treatment of NSCLC cells with CuB-induced upregulation of key TSGs and downregulation of oncogenes as well as TPGs. There are multiple epigenetic mechanisms, such as promoter methylation, chromatin modifications, and miRNA-mediated silencing, that are involved in the alterations of gene expressions (21, 22, 28, 29). Therefore, we evaluated the CuB-mediated epigenetic modifications, including changes in the global as well as gene-specific methylation and histone modifications in NSCLC cells. Interestingly, we found that CuB inhibited the expression of both DNMTs and HDACs in these cells. Previous studies have shown that treatment with either DNMTs inhibitor alone or in combination with HDACs inhibitor altered the expressions of TSGs and TPGs in many cancers (3, 21, 28, 30). In general, promoter regions of the TSGs are hypermethylated causing their silencing in cancer and treatment with DNMTs inhibitor has been shown to reactivate their expression by altering their methylation status (28). In contrast, hypermethylation at hTERT promoter is associated with transcriptional activation by inhibiting the binding of various methylation-sensitive transcriptional repressor proteins (21, 31). Together, our results suggest that CuB-induced downregulation of DNMTs not only facilitated the expression of TSGs but also inhibited the expression of hTERT in H1299 cells.

In addition to alterations in DNA methylation, chromatin modifications also play major role in gene regulation in cancers. In general, HDACs are frequently overexpressed in many cancers, and inhibition of HDACs induces re-expression of epigenetically silenced genes (7, 8, 28). Our results suggest that CuB-mediated HDACs inhibition altered the accessibility of various transcriptional factors to the promoters of the TSGs and TPG in H1299 cells. Furthermore, we also observed that CuB at 6 nmol/L concentration induced both class I and class II HDACs, which seems to be the cellular survival strategy to combat with the autophagic effects of CuB (32, 33). The protein expressions of HATs, PCAF, and CBP were increased after CuB treatment, which might also help in inducing the expressions of the TSGs, as reported previously (34, 35).

The changes in the expression and activities of epigenetic modulatory enzymes lead to alterations in histone modifications. In general, the acetylation at H3K9 results into gene activation, whereas methylation at H3K9 (H3K9me3) results into gene repression (2, 28, 36). It is a well-established fact that the sites and degree of histone methylation define its role in gene regulation (37). The sites with higher enrichment of histone methylation marks, H3K9me3 and H3K27me3, represent transcriptionally inactive regions of chromatin. We found significantly lowered enrichment of these histone methylation marks at the promoters of TSGs, whereas their enrichment was increased at the hTERT promoter after CuB treatment in H1299 cells. Many HDACs inhibitory compounds have shown p53-independent p21^CIP1/WAF1 upregulation (38, 39), which again validates the HDACs inhibitory nature of CuB.

Methyl-CpG–binding domain proteins (MBD) form a nuclear DNA-binding protein family comprising MBD 1-4 and MeCP2. These proteins are capable of binding specifically to methylated DNA and are known to be involved in DNA methylation–mediated transcriptional repression (2). Binding of MeCP2 to the selected promoter regions of the p21^CIP1/WAF1 was significantly disrupted following CuB treatment, whereas at the hTERT promoter sites, MeCP2 binding was found to be increased. Furthermore, we have also found a higher enrichment of active chromatin marks at the p16^INK4A and p21^CIP1/WAF1 promoters and a higher enrichment of inactive chromatin marks at the hTERT promoter, which facilitate alterations in their expressions (21–22, 28, 31). Furthermore, CuB-mediated downregulation of c-MYC expression facilitates the disruption of the binding of c-MYC, a transcription activator of hTERT, to the hTERT promoter in H1299 cells. On the other hand, c-MYC represses p21^CIP1/WAF1 expression by forming complexes with transcription factors Sp1/Sp3, which are required for its expression. Because of CuB-mediated inhibition, c-MYC becomes incapable of binding to the transcription factors Sp1/Sp3, which are required to bind to the p21^CIP1/WAF1 promoter to facilitate its activation (40). The downregulation of c-MYC, due to the CuB treatment, thus indirectly leads to the upregulation of the p21^CIP1/WAF1. On the basis of these observations, we propose that CuB-mediated upregulation of p21^CIP1/WAF1 and downregulation of hTERT is, at least in part, mediated through alterations in promoter methylation and chromatin modifications in NSCLC.

NNK has been shown to induce lung carcinogenesis through both genetic and epigenetic mechanisms. Early changes during NNK-induced lung carcinogenesis include altered expression of DNMT1 and HDACs (41, 42). Because CuB was found to alter the expression of DNMTs and HDACs in vitro, we selected an NNK-induced lung cancer mice model to assess the in vivo lung anticancer potential of CuB. In previous studies, the use of combinations of 0.1 mg/kg b.w. 5-aza-2′-deoxycytidine, a standard DNMT inhibitor, and 400 mg/kg b.w. valproic acid, a standard HDAC inhibitor, have been reported to cause a moderate reversible leukocytopenia in mice (43). In yet another study performed in rats, the combinations of 20 mg/kg b.w. SGI-110, a DNMT inhibitor along with 1 mg/kg b.w. entinostat, a HDAC inhibitor also resulted in a significant weight loss in the animals (44). In contrast, CuB doses used in this study inhibited the expressions of both DNMTs and HDACs in vivo with no visible sign of toxicity in the animals. Furthermore, alterations in the expressions of these epigenetic modulatory enzymes facilitate the upregulation of TSGs and downregulations of oncogenes as well as TPG as observed in vitro. Expression of p16^INK4A is transcriptionally silenced in H1299 cells, whereas it was highly upregulated in the NNK-induced lung tissues in A/J mice. In contrast to the in vitro results, CuB was found to downregulate NNK-induced p16^INK4A upregulation, which might be due to the dysfunctional downstream Rb phosphorylation pathway resulting in the inability of the p16^INK4A to inhibit the cellular proliferation in NNK-induced tumors (45). However, CuB-mediated upregulation of other TSGs such as p21^CIP1/WAF1 and downregulation of TPGs leads to the inhibition of NNK-induced lung tumorigenicity. Furthermore, we have also observed that CuB effectively reduced the severity of the NNK-induced lung tumorigenesis by inhibiting tumor incidence, multiplicity, and lung hyperplasia, which is well correlated with the inhibition of PCNA and induction of apoptotic cells in the CuB-treated NNK-induced lung tumors.

Our study is the first of its kind to show the efficacy of CuB against NNK-induced lung tumors. As per our knowledge, this is the first report showing CuB-mediated inhibition of both DNMTs and HDACs and its epigenetic mechanism of action both in vitro and in vivo. This finding becomes even more significant due to the strong anticancer activity exhibited by a low concentration of CuB against NSCLC. In conclusion, our results suggest that CuB could be developed as a very potent lung anticancer molecule and it could also be used in designing novel epigenetic therapeutic strategy against NSCLC in humans.

No potential conflicts of interest were disclosed.

Conception and design: S. Shukla, R. Maurya, S.M. Meeran

Development of methodology: S. Shukla, S. Khan, S. Kumar, S. Sinha, M. Farhan, H.K. Bora

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S. Shukla, S. Khan, S. Kumar, S. Sinha, M. Farhan, H.K. Bora, S.M. Meeran

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S. Shukla, S. Khan, S. Kumar, S.M. Meeran

Writing, review, and/or revision of the manuscript: S. Shukla, S. Khan, R. Maurya, S.M. Meeran

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S.M. Meeran

Study supervision: R. Maurya, S.M. Meeran

Other (isolation and characterization of CuB): R. Maurya

Senior research fellowships to S. Shukla, S. Khan, and S. Kumar from the CSIR and UGC is acknowledged. Technical assistance from Jyoti Verma and Amar Deep Lakra is also gratefully acknowledged. The authors thank A.L. Vishwakarma from the division of SAIF, CSIR-CDRI for FACS analysis. CSIR-CDRI communication # 8947.

This work was supported by funds from the CSIR-EpiHeD-Network Scheme (BSC0118), Science & Engineering Research Board (SR/FT/LS-80/2011), and Department of Biotechnology-Twinning Programme (BCIL/NER-BPMC/2013/722), New Delhi, India, to S.M. Meeran.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.