2010; 16:828-842 <http://www.molvis.org/molvis/v16/a93>
Received 5 July 2009 | Accepted 5 May 2010 | Published 11 May 2010
1Department of Ocular Pathology, Vision Research Foundation, Sankara Nethralaya, Chennai, India; 2Department of Biotechnology, Indian Institute of Technology, Madras, India; 3CeNTAB SASTRA University, Tanjore, India
Correspondence to: S. Krishnakumar, Department of Ocular Pathology, Vision Research Foundation, Sankara Nethralaya Chennai – 600 006, India; Phone: 91-44-28271616; FAX: 91-44-2825 4180; email: email@example.com
Purpose: Previously we showed that epithelial cell adhesion molecule (Ep-CAM), a cell surface molecule, was highly expressed in primary retinoblastoma tumors. In the present study, we studied the genes regulated by Ep-CAM in a retinoblastoma Y79 cell line in vitro using a combination of short interference RNA and microarray technology.
Methods: Flow cytometry, quantitative reverse transcriptase PCR (Q-RT–PCR), and immunohistochemistry were performed to confirm the Ep-CAM re-expression in the Y79 cells treated with 5′-azacytidine (AZC). Ep-CAM expression in AZC-treated Y79 cells was silenced using synthetic anti-Ep-CAM short interference RNA, and whole genome microarray was performed to determine the gene expression changes post Ep-CAM knockdown. Ep-CAM inhibition was confirmed by Q-RT–PCR, western blotting, and immunofluorescence.
Results: Ep-CAM expression was significantly restored in Y79 cells on day 5 of AZC treatment. Ep-CAM inhibition significantly affected Y79 cell proliferation. We identified 465 upregulated genes (≥1.0 fold) and 205 downregulated genes (≤0.5 fold) in response to knockdown of Ep-CAM. These genes regulate several aspects of tumor function, including cell survival/proliferation, DNA replication/transcription, apoptosis, and angiogenesis. Quantitative pathway analysis using Biointerpreter further revealed that the most pronounced effect of Ep-CAM knockdown was deregulation of pathways that include mitogen-activated protein kinase (MAP) kinase and tumor protein 53 (P53) pathways. Real-time Q-RT–PCR confirmed microarray gene expression changes for selected genes.
Conclusions: Ep-CAM silencing significantly decreases Y79 cell proliferation and revealed a wide network of deregulated pathways in vitro. Future studies targeting Ep-CAM gene expression in vivo will help to delineate the mechanisms associated with Ep-CAM gene function in neoplastic transformation and define the potential for Ep-CAM-based molecular intervention in retinoblastoma patients.
Retinoblastoma (RB) is the most common intraocular malignancy in children . For many years retinoblastoma confined to the eye has been a curable disease with local therapy, such as enucleation, external beam irradiation, brachytherapy, cryotherapy, or laser coagulation . In contrast, systemic disease is difficult to cure, although it usually responds to chemotherapy [3–5]. The advent of short interfering (si)RNA might prove a useful addition to, or a substitute for, conventional treatment modalities. Previously we demonstrated the expression of epithelial cell adhesion molecule (Ep-CAM) in RB tumor tissues, and the tumors with choroidal invasion/optic nerve invasion showed significantly higher expression of Ep-CAM . Ep-CAM is a 40,000 MW type I transmembrane glycoprotein that consists of two epidermal growth factor-like extracellular domains, a cysteine-poor region, a transmembrane domain, and a short cytoplasmic tail. Ep-CAM is overexpressed in various epithelial cancers  and is an ideal therapeutic target because of the following reasons: (a) overexpression in cancer cells versus noncancerous cells, (b) apical expression in cancer cells and basolateral expression in normal epithelial cells , and (c) not shed into the circulation . For these reasons Ep-CAM has gained interest as a potential therapeutic target and an attractive candidate tumor-associated antigen to serve as a target for antibody-based immunotherapy [10,11]. There is evidence that Ep-CAM expression levels correlate with proliferative activity and contribute to neoplastic transformation [12,13]. These data suggest that Ep-CAM is a potential target for molecular intervention and that it requires further investigation.
To define the mechanisms associated with Ep-CAM gene silencing, we investigated the effect of Ep-CAM siRNA treatment on the whole genome expression by microarray technology.
Y79 was obtained from the American Type Culture Collection (Manassas, VA). Media and fetal bovine serum (FBS) were purchased from Gibco-BRL (Rockville, MD). Y79 was cultured in Rosewell Park Memorial Institute (RPMI; Gibco-BRL) 1640 supplemented with 10% heat-inactivated fetal calf serum, 0.1% ciprofloxacin, 2 mM L-glutamine, 1 mM sodium pyruvate, and 4.5% dextrose and grown in suspension at 37 °C in a 5% CO2-humidified incubator. The study adhered to the Declaration of Helsinki. This study was conducted at the Medical Research Foundation and Vision Research Foundation, Sankara Nethralaya, India, and was approved by the Vision Research Foundation ethics boards.
Approximately 1×105 Y79 cells were incubated in culture medium with and without 5′-azacytidine (AZC) at a final concentration of 5 µM, with medium changes every day for 5 days. After day 5, the Y79 cells were withdrawn from AZC exposure and subsequently passaged for 30 days in complete medium. Cells were harvested on days 5, 8, 10, 15, and 20 for extraction of total RNA and were tested for Ep-CAM gene expression, using real-time PCR. Also, 1x105 cells were harvested on days 5, 8, 10, and 15 and analyzed for Ep-CAM protein expression, using fluorescence-activated cell sorting.
Gene silencing of Ep-CAM expression was performed essentially as described previously using sequence-specific siRNA and transfection reagents . For each well of cells, 200 nm of siRNA and 12 µl of Hi-perfect transfection reagent was mixed with 88 µl of serum free RPMI medium in a vial and incubated for 5–10 min at room temperature (RT) to allow the transfection complex formation. Then the complex was added drop-wise to the wells containing Y79 cells and the plate was gently swirled to ensure the uniform distribution of the transfection complex. The plates were incubated further for 6 h in a CO2 incubator at 37 °C. After 4 h of incubation, 400 µl of complete RPMI medium was added to the wells and incubated further for 24–72 h. Briefly, 1x105 cells were plated in each well of six-well plates and allowed to grow for 24–36 h (until they were 40%–60% confluent). siRNA was then transfected into cells at a concentration of 200 nM using Hi-Perfect transfection reagent (Qiagen, Santa Clara, CA) and serum-free medium. After 4 h of incubation, serum-rich medium was added. Human Ep-CAM siRNA (Hs_TACSTD1_10; catalog number SI04343416; Forward strand: GGA ACU CAA UGC AUA ACU Att and the reverse strand: UAG UUA UGC AUU GAG UUC Cct) and scrambled siRNA (catalog number 1022563) were used in this study.
To quantify RNA expression of Ep-CAM in untreated or AZC-treated Y79 cell lines and untreated or siRNA-treated Y79 cells, RNA was extracted by the guanidine isothiocyanate and chloroform method (TRI Reagent; Sigma-Aldrich, Bangalore, India). Cells were harvested from cultures and collected in RNase free vials. To the pellet, one ml of trizol reagent (Sigma-Aldrich) was added and vortexed for 2 min and incubated at room temperature for 5 min. Later, 0.5 ml of chloroform was added to the solution and shook well for 15 s and centrifuged. The aqueous layer that contains RNA was transferred to new vials and 0.5 ml of isopropanol was added and incubated at room temperature for 10 min. After centrifugation, supernatant was discarded and 0.5 ml of 75% ethanol was added and mixed well, centrifuged and supernatant was discarded. Pellet was air dried at RT for 2 min and reconstituted in 25 µl of RNase free water. All the centrifugations in RNA extraction was performed at 15,000× g for 10 min at 4 °C. All RNA samples were treated with DNase (Turbo; Ambion, Genetix Biotech Asia Pvt. Ltd., Chennai, India). For all samples, 1 μg of total RNA was used to synthesize first-strand cDNA with reverse transcriptase (SuperScript II; Invitrogen, Joyvel, Chennai, India) and random primers. The cDNA synthesis was performed at 37 °C for 60 min after heat inactivation at 95 °C for 10 min. Gene expression assays for Ep-CAM (Hs00158980_m1) and two endogenous controls, Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Hs99999905_ml) and HPRT (Hs99999909_m1), were obtained from Applied Biosystems (LabIndia, Chennai, India). The primer sequences used for Q-RT–PCR are as follows: Homo sapiens proliferating cell nuclear antigen (PCNA): forward primer (FP) 5′- CAG GGC TCC ATC CTC AAG AA-3′, reverse primer (RP) 5′-TCT TCA TTG CCG GCG CAT T-3′; FOS: FP 5′-CAG CGA GCA ACT GAG AAG CC-3′, RP 5′-CGC TGT GAA GCA GAG CTG G-3′; JUN: FP 5′-CAA AGT TTG GAT TGC ATC AAG TG-3′, RP 5′-TAA CAT TAT AAA TGG TCA CAG CAC ATG-3′; DRAM: FP 5′-ATG GTC ATC TCT GCC GTT TC-3′, RP 5′-AAA GGC CAC TGT CCA TTC AC-3′; CCND3: FP 5′-TGG ATG CTG GAG GTA TGT G-3′, RP 5′-CGT GGT CGG TGT AGA TGC-3′. Quantification of gene expression was performed in triplicate in a 20 μl volume in 96-well plates on a real-time PCR system (Prism 7300; ABI, Lab India Instruments, Gurgaon, India). Each reaction included 1× primer probe mix (TaqMan; ABI), 1× universal PCR master mix (TaqMan; ABI), and 100 ng of cDNA. The PCR was performed as follows: 2 min at 50 °C, 10 min at 95 °C, and 40 cycles of 15 s at 95 °C, plus 1 min at 60 °C. Commercial software (SDS ver. 1.3; ABI) was used to calculate ΔΔCt  relative expression values for Ep-CAM normalized to the GAPDH endogenous control.
Untreated or siRNA-treated Y79 cells and untreated or AZC-Y79 cells were harvested, washed, and then fixed in 4% paraformaldehyde and processed for immunofluorescence, using anti-Ep-CAM mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA) as the primary antibody at a dilution of 1:100 and incubating for 12 h at 4 °C. Fluorescein isothiocynanate (FITC)-labeled antimouse antibody was used as the secondary antibody. Incubation with the secondary antibody was for 30 min at 37 °C. Immunofluorescence staining was detected using an Axio Observer fluorescent microscope (Carl Zeiss, Berlin, Germany). Due to the lack of purified Ep-CAM protein, we could not use an adsorbed control reaction for the negative control. Instead, we used isotypic antibodies (normal mouse serum) as negative controls, both for untreated and AZC-treated Y79 cells, in the immunohistochemistry. For immunohistochemistry, Novolink Polymer kit (Leica Microsystems Inc., Bannockburn, IL) was used after primary antibody incubation for developing the reaction, and hematoxylin was used for counterstaining. Immunostaining was assessed based on intensity and was graded as 1+ (low), 2+ (moderate), and 3+ (strong).
Y79 cells were harvested, washed, and the cell number determined. Approximately 3x105 cells/ml were resuspended in ice-cold phosphate buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM sodium phosphate dibasic, 2 mM potassium phosphate monobasic and a pH of 7.4), 10% fetal calf serum, and 1% sodium azide. Mouse monoclonal primary antibody against Ep-CAM with a final dilution of 1:40 (C-10; Santa Cruz Biotechnology) was used and incubated for 2 h at 4 °C. After incubation, cells were washed three times (at 400× g for 5 min each) with ice-cold PBS. FITC-conjugated antimouse secondary antibody with a final dilution of 1:1,000 was used and incubated for 1 h at 4 °C in the dark. After incubation, cells were washed three times (centrifugation at 400× g for 5 min each) with ice-cold PBS. Cells were analyzed on a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA), using the CellQuest software program (BD Biosciences). The fluorescence of FITC-stained cells was excited with an argon laser at 488 nm, and the emissions were detected at 530 nm.
Cultures were set up in triplicate and treated or not with 5 μM AZC for 24, 48, 72, 96, and 120 h. After 5 days of AZC treatment, the cells were washed with PBS and incubated with complete medium for 72 h. Then, siRNA (Qiagen, Santa Clara, CA) was added to the plates, and the cells were allowed to grow for 24, 48, or 72 h. At each time point, cells were checked visually for growth and proliferation. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma, St Louis, MO) was then added to the wells, and the cells were incubated at 37 °C for 4 h. MTT solubilization solution (10% Triton X-100 in acidic isopropanol, 0.1 N HCl) was then added, and the cells were incubated overnight. Colorimetric measurements were made using a spectrophotometer (Beckman Coulter India Private Ltd, New Delhi, India) at 562 nm, and the background was subtracted at 650 nm.
Cells were lysed in radioimmunoprecipitation assay lysis buffer for 15 min on ice. An aliquot (100 µg) of lysate was electrophoresed by 10% sodium dodecyl sulfate-polyacrylamide gel and blotted onto a nitrocellulose membrane. Membranes were blocked in 5% fat-free milk and then incubated separately with 1:500 diluted mouse monoclonal primary antibody against Ep-CAM (C-10) overnight at 4 °C. β-actin was used as a loading control (AC-15, dilution: 1:4,000; Sigma). After washing, membranes were incubated with horseradish peroxidase-conjugated antimouse gamma immunoglobulin (IgG) antibody (diluted to 1:2,000; Santa Cruz Biotechnology) for 1 h at room temperature. The bands were visualized using an enhanced chemiluminescence kit (Amersham, Pittsburgh, PA). Each experiment was performed in triplicate.
Total RNA used for the microarray analysis was isolated from cultured cells using TRIZOL reagent (Invitrogen, Carlsbad, CA) and treated with TURBO DNase (Ambion, Genetix Biotech Asia Pvt. Ltd., New Delhi, India) to remove the DNA. The RNA samples (10 μg each) in a 50-μl reaction were treated with 1 μl of TURBO DNase (2 U) in 1× TURBO DNase buffer at 37 °C for 30 min. After incubation, the RNA sample was extracted with phenol/chloroform to inactivate TURBO DNase. Agilent's (Agilent Technologies Genotypic, Bangalore, India) Low RNA Input Linear Amplification Kit PLUS was used to generate fluorescent complementary RNA (cRNA). This method uses T7 RNA polymerase, which simultaneously amplifies target material and incorporates cyanine 3-labeled cytidine tri-phosphate (CTP). Qiagen’s RNeasy mini spin columns were used for purification of the amplified cRNA samples, and the samples were then hybridized to the Human Whole Genome 44K Oligo Microarray for 17 h at 65 °C, as recommended by the manufacturer (Agilent Technologies). Data analysis was done using Genespring GX version 10 (Agilent Technologies). Agilent Feature Extraction software (G25677AA; Agilent Technologies) was used to analyze the microarray data.
The statistical analysis included the independent t test. Statistical analysis was performed using SPSS version 12.0 software (Chicago, IL). P values less than 0.05 were considered significant.
Y79 cells were exposed to 5 μM AZC for 24, 48, 72, 96, and 120 h. We first examined whether AZC treatment affected the kinetics of cell growth by evaluating cell viability using the MTT assay. Cell viability kinetics were compared to those of control cells cultured in the absence of AZC. The data of the MTT assay are indicative of both viability and cell proliferation. Treatment with AZC greatly modified these data; around 80% of the number of metabolically active cells were observed on day 5 (Figure 1). This result is consistent with cell cycle arrest in Y79 cells.
We previously have observed increased mRNA expression of Ep-CAM in RB primary tumors  compared to non-neoplastic retina (data not shown). In contrast to increased Ep-CAM expression in primary RB tumors, Y79 showed very faint to absent Ep-CAM protein expression by immunohistochemistry and real-time quantitative reverse transcriptase PCR (Q-RT–PCR). Based on earlier studies [16–19] and the presence of CpG island in the promoter region of Ep-CAM, we subjected the cell line to a demethylating agent to find out whether low expression of Ep-CAM is due to promoter methylation. In an attempt to re-express Ep-CAM, we exposed Y79 to AZC, as described in the methods. The flow cytometry analysis showed 47.73%, 44.2%, 42.96%, and 18.27% of Y79 cells positive for Ep-CAM on days 5, 8, 10, and 15 of AZC treatment, respectively (Figure 2A). On day 5, Y79 cells demonstrated strong membrane positivity for Ep-CAM by immunohistochemistry (Figure 2G). After day 5, Y79 cells were withdrawn from AZC exposure and subsequently passaged in the complete medium up to 30 days, and Ep-CAM mRNA expression was retained up to 21 days as follows: 210 fold, 132 fold, 56 fold, and 20-fold on days 8, 10, 15, and 20, respectively (Figure 2J). Thus, AZC treatment restored adequate Ep-CAM expression, which was sufficient to perform siRNA experiments in Y79 cells.
To determine whether Ep-CAM is a potential target for RB gene therapy, the Y79 cell line was treated with Ep-CAM-specific siRNA as well as with scrambled siRNA. The scrambled siRNA served as a control for nonspecific effects of siRNA treatment on cell growth. Ep-CAM protein expression, mRNA levels, and cell viability were determined at 24, 48, and 72 h using immunofluorescence (Figure 3A-E), real-time RT–PCR (Figure 4), and MTT assay (Figure 5), respectively. The data show that Ep-CAM siRNA treatment results in 12–13 fold decrease in Ep-CAM gene expression in the Y79 cell line compared with untreated and scrambled siRNA-treated Y79 cells at 48 h. Scrambled siRNA treatment had no effect on Ep-CAM mRNA levels, thus supporting the specificity of the Ep-CAM siRNA. Similarly, western blot analysis showed marked decrease in Ep-CAM expression (3.5 fold downregulated when compared to untreated) at 48 h of Ep-CAM siRNA treatment, whereas no change was shown when treated with scrambled siRNA (Figure 3F). The Ep-CAM siRNA treatment effect was maximal at 48 h and attenuated by 72 h (consistent with the known transient effects of siRNA in mammalian cells). Ep-CAM siRNA treatment resulted in a significant decrease in cell proliferation by more than 50% at all time points.
Ep-CAM expression was decreased by 12–13 fold, which was minimum at 48 h after siRNA treatment; hence at this time point cells were harvested for microarray analysis. The genome-wide expression profile of Ep-CAM siRNA-treated AZC-Y79 cells against siRNA-untreated AZC-Y79 was analyzed to investigate the molecules involved in Ep-CAM signaling. In Y79 cells, anti-Ep-CAM siRNA treatment modulated the expression of 670 gene transcripts (probe sets), 465 (69.4%) of which were increased and 205 (30.5%) were decreased. Only genes with p≤0.05 and log ratio of at least 2.0 for upregulation and log ratio of 0.5 for downregulation, keeping a median log ratio of 1 in both biologic replicates, were considered for expression analysis. The important upregulated and downregulated genes post Ep-CAM knockdown (Figure 6) are provided in Table 1 and Table 2, respectively. The data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus  and are accessible through GEO Series accession number GSE16991.
In cells with inhibited Ep-CAM expression, upregulated categories represented gene products that were involved in apoptosis, such as lactalbumin, alpha (LALBA), retinoic acid receptor, gamma (RARG), ataxin 3 (ATXN3), mitogen-activated protein kinase 13 (MAPK13), tubulin, beta 2A (TUBB2A), activating transcription factor 3 (ATF3), insulin-like growth factor binding protein 1 (IGFBP1), Homo sapiens damage-regulated autophagy modulator (DRAM), and cytochrome c, somatic (CYCS); anti-angiogenic genes, such as lectin, galactoside-binding, soluble, 3 (LGALS3), and vasohibin 1 (VASH1); cell cycle and differentiation genes, such as LGALS3, 5-hydroxytryptamine (serotonin) receptor 5A (HTR5A), INS, tachykinin 3 [neuromedin K, neurokinin beta] (TAC3), and ATF3; anti-proliferation genes, such as spectrin repeat containing, nuclear envelope 2 (SYNE2), BTB and CNC homology 1, basic leucine zipper transcription factor 2 (BACH2), tripartite motif-containing 8 (TRIM8), Rac GTPase activating protein 1 (RACGAP1), and MEG3; and tumor suppressor gene tropomyosin 1 alpha (TPM1).
Downregulated categories represented gene products that were involved in cell cycle and cell division genes, such as MYC-associated zinc finger protein [purine-binding transcription factor] (MAZ), tubulin, epsilon 1 (TUBE1), MAD1, sprouty-related, EVH1 domain containing 1 (SPRED1), U2 small nuclear RNA auxiliary factor 1 (U2AF1), CNDP, and stress induced growth inhibitor family member 2 (SGIN2); cell migration genes, such as MMP2, cell division cycle 42 [GTP binding protein, 25 kDa] (cdc42), and MADS box transcription enhancer factor 2, polypeptide A [myocyte enhancer factor 2A] (MEF2A); anti-apoptotic genes, such as DnaJ (Hsp40) homolog, subfamily A, member 3 (DNAJA3), ornithine decarboxylase 1 (ODC1), gelsolin [amyloidosis, Finnish type] (GSN), BCL2-like 1 (BCL21), and high-mobility groupbox 1 (HMGB1); proliferation genes, such as FGF9,Homo sapiens v-fos FBJ murine osteosarcoma viral oncogene homolog (FOS), Homo sapiens jun oncogene (JUN), PCNA,Homo sapiens E2F transcription factor 3 (E2F3), malignant T cell amplified sequence 1 (MCTS1), hairy and enhancer of split 1, [Drosophila] (HES1), cyclin D3 (CCND3), growth arrest and DNA-damage-inducible, alpha (GADD45A), embryonic lethal, abnormal vision, Drosophila-like 1 [Hu antigen R] (ELAVL1), nuclear receptor coactivator 1 (NCOA1), enolase 1, [alpha] (ENO1), and insulin-like growth factor 1 [somatomedin C] (IGF1); and oncogenes, such as FOS, member of RAS oncogene family (RAP2C), member RAS oncogene family (RAB8A), JUN, sarcospan [Kras oncogene-associated gene] (SSPN), RAB1B, RAB15, and v-myc myelocytomatosis viral related oncogene, neuroblastoma derived [avian] (MCYN).
Upon Ep-CAM inhibition, MAP kinase pathway was deregulated in Y79 cells. The genes involved in MAP kinase pathway were FOS, JUN, FGF9, and GADD45A. On the other hand, P53 pathway molecules were upregulated upon Ep-CAM inhibition. The molecules involved in P53 pathway were RRM2, CYCS, and DRAM.
All the distinct gene identifications were examined for their known biologic function according to gene ontology convention and grouped in the respective functional category. The proportion of each functional category in the total number of selected identified genes (taken as 100%) is shown in Figure 7. Among the upregulated genes, (Figure 7A) the majority of genes belong to the apoptosis functional category (48%). Hence, it is interesting to speculate that Ep-CAM inhibition may promote apoptosis in Y79 cells. Other genes identified belong to proliferation, anti-proliferation, angiogenesis, anti-angiogenesis, anti-apoptotic, and tumor suppressor genes. Among the downregulated genes, (Figure 7B) one-third of the total genes belong to the proliferation functional category (33%) and cell cycle and differentiation (33%). Other downregulated genes belong to cellular invasion, anti-apoptosis, oncogenes, and angiogenic genes.
Five genes (PCNA, CCND3, FOS, JUN, and DRAM) from microarray data have been confirmed by real-time Q-RT–PCR. The results are consistent with the microarray data. The gene expression of all five genes studied was higher when measured using Q-RT–PCR as compared to microarray analysis, reflecting the better dynamic range of Q-RT–PCR. The relative mRNA expression of DRAM was significantly increased, and the relative mRNA expression of PCNA, CCND3, FOS, and JUN were significantly downregulated in siRNA-treated Y79 cells compared to untreated Y79 cells (Figure 8).
To investigate the functional relevance of Ep-CAM in RB, we proposed to transiently silence Ep-CAM expression using siRNA and study its effect on whole gene expression profiling. We chose the Y79 cell line for in vitro studies for two reasons. One reason is that Y79 cells form tumors in the murine model that closely resemble the naturally occurring tumor in anatomic sites without disrupting the choroid or sclera or involving the anterior chamber . The second reason is that as RB tumors from Indian RB patients are advanced tumors, presenting with choroidal and optic nerve invasion , the Y79 cell line is appropriate for in vitro studies and the data can be translated to the actual tumor state.
Our study shows that both mRNA and protein levels of Ep-CAM are very low to absent in Y79 cells. There is substantial evidence that DNA methylation plays an important role in silencing specific genes during development and cell differentiation [23,24]. Earlier studies have shown that Ep-CAM is regulated by the most common epigenetic mechanism—hypermethylation in oral squamous cell carcinoma and breast carcinomas [16–19]. Epigenetic silencing of a gene can be reversed by drugs, such as AZC, which acts through incorporation of the modified nucleotide in the place of cytosine into the newly synthesized strand during DNA replication. The modified nucleotides form a covalent complex with the active sites of the DNA methyltransferase, depleting methyltransferase activity, resulting in generalized demethylation . In this context, we restored Ep-CAM expression in Y79 cells by subjecting the cells to demethylation (AZC treatment) in vitro for 5 days. However, these drugs have effects other than inducing demethylation. The amount of methylation and the response obviously can vary for the same gene from one cell line to another and between different genes in the same cell line. One disadvantage of a drug-based reactivation approach is that, unlike CpG island array screening, use is essentially restricted to cell cultures and is not amenable to examination of primary tumors.
In the present study, AZC treatment restored adequate Ep-CAM expression, which was sufficient to perform siRNA experiments in Y79 cells. Silencing of the Ep-CAM gene significantly decreased the proliferative capacity of RB Y79 cells. Expression microarray analysis of Ep-CAM silencing revealed several genes related to cell proliferation, apoptosis, anti-apoptosis, cell division and differentiation, angiogenesis, anti-angiogenesis, and oncogenes.
Cells expressing Ep-CAM proliferate more rapidly, grow in an anchorage-independent manner, and have a reduced requirement for growth factors. Consistent with the growth-promoting role of Ep-CAM, knockdown of endogenous Ep-CAM in tumor cells decreases cell proliferation and migration [7,26]. The role of Ep-CAM in proliferation and its association with cancer is poorly explained by proposed cell adhesion functions. Recently, Maetzel et al.  showed that regulated intramembrane proteolysis activates Ep-CAM as a mitogenic signal transducer in vitro and in vivo. The cleaved intracellular domain Ep-CAM intracellular domain (EpICD) associates with four and a half LIM domain (FHL2), β-catenin, and lymphoid enhancer binding factor 1 (Lef-1) to form a nuclear complex that contacts DNA at Lef-1 consensus sites, induces gene transcription, and is oncogenic in immunodeficient mice. Hence, this mechanism of nuclear signaling of Ep-CAM explains how it functions in cell proliferation.
Our study showed significant downregulation of PCNA post-Ep-CAM silencing. A high PCNA index has been noted in RB in a recent study by Dimaras et al. . In gastric cancer, the PCNA index of gastric cancer tissues with high Ep-CAM expression was higher than gastric cancer tissues with low Ep-CAM expression, indicating that Ep-CAM might be associated with the proliferative activity of cancer cells . We also showed that FOS and JUN, which are involved in cellular proliferation, were downregulated post-Ep-CAM silencing.
Genes related to apoptosis, such as damage-regulated autophagy modulator DRAM, a critical effector of p53-induced autophagy , and cytochrome c CYCS, a central molecule in apoptotic signaling , were significantly upregulated. Recently, a study on mammary tissues in transgenic mice showed that Ep-CAM overexpression could induce Ki-67 upregulation and reduce apoptosis of cells, accompanied by upregulation of bcl-2 expression, which was the first report about the relation between Ep-CAM and apoptosis . Our observations indicate that Ep-CAM inhibition on cellular proliferation is by cell cycle arrest accompanied by apoptosis. However, further functional studies need to be conducted to prove these effects in RB.
These changes are coupled with downregulation of important cell cycle-related transcription factors, E2F3 and CCND3. E2Fs are transcription factors best known for their involvement in the timely regulation of gene expression required for cell cycle progression. E2F3 overexpression has been previously demonstrated in RB tumors by Gallie et al. . Cyclins are periodic regulatory proteins governing cell cycle transit from G1 phase into S phase. Overexpression of cyclins could lead to abnormal cellular proliferation, which underlies the process of tumorigenesis .
Interestingly, our study showed the downregulation of MYCN post-Ep-CAM silencing. MYCN is a well studied oncogene known to be amplified in neuroblastoma  and in a subset of RB tumors [36,37]. Previously it was demonstrated that Ep-CAM has a direct impact on the cell cycle and proliferation and the ability to rapidly upregulate the proto-oncogene c-myc and cyclin A/E . However, further studies are needed to establish the MYCN relation with Ep-CAM signaling in RB.
In addition to proliferation and apoptotic genes, we showed that genes related to tumor invasion, metastasis, and angiogenesis, such as MMPs and cdc42 are significantly downregulated post Ep-CAM silencing. Previously we demonstrated high MMP  and cdc42  expression in RB primary tumors. The downregulation of MMPs in the present study supports the hypothesis that Ep-CAM inhibition could reduce tumor cell invasion, as demonstrated in an earlier study . Ep-CAM inhibition led to downregulation of the MAP kinase pathway in Y79 cells. The genes involved in the MAP kinase pathway are FOS, JUN, FGF9, and GADD45A. On the other hand, P53 pathway molecules were upregulated upon Ep-CAM inhibition. The molecules involved in the P53 pathway are RRM2, CYCS, and DRAM.
In conclusion, we showed for the first time that Ep-CAM silencing in RB in vitro leads to a decrease in Y79 cell proliferation and deregulation of several genes related to cell survival/proliferation, DNA replication/transcription, apoptosis, and angiogenesis. Targeting Ep-CAM for molecular intervention appears to be an attractive strategy. First, Ep-CAM is substantially overexpressed in primary RB tumor samples . Furthermore, because Ep-CAM overexpression is associated with proliferation and neoplastic transformation [40,41], silencing of Ep-CAM gene expression is likely to dramatically alter the phenotype of cancer cells without significantly influencing normal or nonproliferating cells. Future studies targeting Ep-CAM gene expression in vivo will help to delineate the mechanisms associated with Ep-CAM gene function in neoplastic transformation and define the potential for Ep-CAM-based molecular intervention in RB patients.
We thank Mr. Madavan Vasudevan, Genotypic Technology Pvt. Ltd for microarray analysis, Mr. Thennarasu for statistical analysis, Mrs. Krishnakumari and Ms. Uma Maheswari for the technical assistance in flow cytometry. Department of Science and Technology (DST), India for funding. Grant Identification Number SR/SO/BB-45/2005.