Tissue samples

Cell lines

The EAC (OE33) cell line was obtained from collaborators at the University of Michigan (Dr. David Beer). These cells were cultured in 47.5% RPMI 1640, 47.5% F-12 supplemented with 5% fetal bovine serum.

DNA and RNA extraction

Genomic DNA was extracted from biopsies and cultured cells using a DNeasy Tissue Kit (Qiagen, Valencia, CA). Total RNA was isolated cultured cells using TRIzol reagent (Invitrogen, Carlsbad, CA). DNAs and RNAs were stored at -80°C before analysis.

Bisulfite treatment and real-time quantitative methylation-specific PCR

DNA was treated with bisulfite to convert unmethylated cytosines to uracils prior to qMSP, as described previously [27]. Promoter methylation levels of CAV1 were determined with the ABI 7900 Sequence Detection System (Applied Biosystems, Foster City, CA), using primers and probes as described previously [27]. A standard curve was generated using serial dilutions of CpGenome Universal Methylated DNA (CHEMICON, Temecula, CA). The normalized methylation value (NMV) was defined as follows: NMV = (CAV1-S/CAV1-FM)/(ACTB-S/ACTB-FM), where CAV1-S and CAV1-FM represent the methylation levels of CAV1 in sample and universal methylated DNAs, respectively, while ACTB-S and ACTB-FM correspond to ß-Actin in sample and universal methylated DNAs, respectively [21].

Real-time quantitative RT-PCR

To determine CAV1 mRNA levels, one-step real-time quantitative reverse-transcriptase polymerase chain reaction (RT-PCR) was performed using a Qiagen QuantiTect Probe RT-PCR Kit (Qiagen, Hilden, Germany) and the ABI 7900 Sequence Detection System (Applied Biosystems, Foster City, CA). Primers and probes were the same as previously reported [27]. ß-Actin was used for normalization of data. A standard curve was generated using serial dilutions of qPCR Reference Total RNA (Clontech, Mountainview, CA). The normalized mRNA value (NRV) was calculated according to the following formula for relative expression of target mRNA: NRV = (TarS/TarC)/(ACTB-S/ACTB-C), where TarS and TarC represent levels of mRNA expression for the target gene in sample and control mRNAs, respectively, whereas ACTB-S and ACTB-C correspond to amplified ß-Actin levels in sample and control mRNAs, respectively [21].

5-Aza-dC treatment of esophageal cancer cell lines

To determine whether CAV1 inactivation was due to promoter hypermethylation in esophageal cancer, OE33 EAC cells were subjected to 5-Aza-dC (Sigma, St. Louis, MO) treatment as previously described [21]. Briefly, 1 × 10⁵ cells/ml were seeded onto a 100 mm dish and grown for 24 h. Then, 1 ul of 5 mM 5-Aza-dC per ml of cells was added every 24 hours for 6 days. DNA and RNA were harvested on day 4.

Data analysis and statistics

Receiver-operator characteristic (ROC) curve analysis [28] was performed using NMVs for the 67 EAC, 26 ESCC and 67 NE by Analyse-it software (Version 1.71, Analyse-it Software, Leeds, UK). Using this approach, the area under the ROC curve (AUROC) yielded optimal sensitivity and specificity to distinguish normal from malignant esophageal tissues, and corresponding NMV thresholds were calculated for CAV1. The cutoff value determined from this ROC curve was applied to determine the frequency of CAV1 methylation in each tissue type included in the present study. For all other tests, Statistica (version 6.1; StatSoft, Inc., Tulsa, OK) was used. Differences with p < 0.05 were deemed significant.

CAV1promoter hypermethylation in different esophageal tissues

Promoter hypermethylation of the CAV1 gene was analyzed in 67 NE, 60 BE, 40 D (19 LGD and 21 HGD), 67 EAC and 26 ESCC samples. All assays in this study were performed in duplicate format, and data showed reproducible and concordant results. CAV1 promoter hypermethylation showed highly discriminative ROC curve profiles, which clearly distinguished both EAC (p < 0.0001) and ESCC (p <0.0001) from NE. AUROC of EAC vs. NE and ESCC vs. NE were 0.839 and 0.920, respectively. ROC curves with corresponding AUROCs for CAV1 of EAC vs. NE and ESCC vs. NE are displayed in Figure 1.

The cutoff NMV for CAV1 (0.2) was chosen from ROC curves to maximize sensitivity and specificity. Mean NMV and frequency of CAV1 hypermethylation for each tissue type are shown in Table 1. The mean NMV of CAV1 was significantly higher in ESCC (0.326), EAC (0.294), D (0.254), HGD (0.240), LGD (0.269) and BE (0.374) than in NE (0.134; p < 0.01, Student’s t-test). The frequency of CAV1 hypermethylation was increased in BE (81.7%), D (60%), and EAC (65.7%) vs. NE (25.4%; p < 0.01, p < 0.01 and p < 0.01, respectively; Chi-square for independence test). CAV1 was hypermethylated in 21 (80.8%) of 26 ESCCs. There were no significant differences between EAC and ESCC in mean CAV1 NMV (0.294 vs. 0.326; p > 0.05) or hypermethylation frequency (65.7% vs. 80.8%, p > 0.05). Among 41 cases with matched NE and T (EAC or ESCC), CAV1 NMVs in T (mean = 0.273) were significantly higher than in corresponding NE (mean = 0.146; p < 0.01, Student’s paired t-test; Figure 2A and B). Among 15 cases with corresponding NE, BE and EAC, one (No.8) was methylated only in EAC, three (No.7, 13 and 17) were methylated only in BE, five (No.1, 6, 9, 14 and 28) were methylated in both BE and EAC, and the remaining six were methylated in NE, BE and EAC simultaneously (Figure 2C). In addition, both CAV1 mean NMV and hypermethylation frequency were significantly higher in BE (0.374, 81.7%) than in D (0.254, 60%) and EAC (0.294, 65.7%; p < 0.01 and p < 0.05, respectively; Figure 3).

No significant associations were observed between CAV1 promoter hypermethylation and patient age, survival (data not shown), smoking or alcohol consumption status, BE segment length, tumor stage or lymph node metastasis, histologic tumor differentiation, or histologic type of esophageal carcinoma (Table 1).

CAV1methylation and mRNA levels in OE33 cells after 5-Aza-dC treatment

OE33 cells were subjected to demethylation by 5-Aza-dC treatment. On day 4 after 5-Aza-dC treatment, the NMV of CAV1 was diminished, while CAV1 mRNA levels were increased (Figure 4).

CAV1 has already been previously reported to have tumor suppressor activity via. inhibiting cell proliferation and/or metastesis in several human cancers [29–32]. CAV1 down-regulation has been reported in many types of cancer, including breast, lung, oral and esophagus [9, 10, 12–14]. These results suggest that low expression of CAV1 may represent a general characteristic or even a requirement of transformed cells in many kinds of carcinogenesis. Potential mechanisms underlying this suppression of expression include posttranscriptional and epigenetic changes, such as aberrant DNA methylation [26, 27, 33]. In the current study, we systematically investigated hypermethylation of the CAV1 gene promoter in primary human esophageal lesions of differing histological types and grades. Our results demonstrate that CAV1 promoter hypermethylation occurs frequently in both human EAC and ESCC (Table 1). CAV1 NMVs in T were significantly higher than those in corresponding NE (p < 0.01, Student’s paired t-test) in 41 cases with corresponding NE and T (Figure 2). Moreover, hypermethylation of the CAV1 promoter was significantly more frequent in premalignant lesions, such as BE and D, as well as in EAC, than in NE (Table 1). There was no significant association between CAV1 promoter hypermethylation and histological subtype of esophageal carcinoma (EAC vs. ESCC). These results suggest that hypermethylation of CAV1 may represent an early epigenetic event in these subjects, that the frequency of this epigenetic event increases during esophageal carcinogenesis, and that this event is highly prevalent in human esophageal cancers.

Barrett’s carcinogenesis is a multistep process comprising genetic and epigenetic alterations in tumor suppressor genes, cell cycle-regulatory genes, and genes essential for cell–cell adhesion [34, 35]. Progressive accumulation of gene alterations is postulated to underly the transition of normal squamous epithelium to BE [36]. Many previous studies, focused on promoter hypermethylation of candidate genes for esophageal carcinomas, have shown staged growth in methylation frequency from nondysplastic esophageal squamous cell mucosa to BE and finally to EAC [15, 37, 38]. Interestingly, it has been suggested that CAV1 acts as a tumor modulator in a tissue type- and stage-dependent manner by binding several different proteins involved in different signal transduction [6, 39–42]. Recently, we reported a preponderance of hypomethylation over hypermethylation events during the epigenomic program of BE pre-progression by comparing global DNA methylation profiles of two groups of BE patients, termed ‘progressors’ and ‘non-progressors’ [16]. Although the frequency of CAV1 hypermethylation in this study increased in parallel with esophageal carcinoma progression, the mean NMV and frequency of CAV1 hypermethylation were higher in BE than in D and EAC, and these differences were significant by Student’s t test (i.e., there was an inverse correlation between CAV1 hypermethylation and Barrett’s-associated esophageal neoplastic progression) (Figure 3). Taken together, these data suggested that the CAV1 promoter is relatively hypomethylated in EAC and D vs. BE, implying that, at least in part, this event represents an early part of the temporal program of Barrett’s-associated esophageal neoplastic progression. Two previous studies demonstrated that expression of CAV1 was elevated in ESCC compared to corresponding normal tissues, and its elevation was associated with malignant progression and poor survival [43, 44]. These inconsistent results may have been due to different analytic methods used, ethnic groups studied, and smaller sample sizes in the previous studies.

In accordance with previous findings [11, 45, 46], we observed that methylation of CAV1 in EAC cell lines was associated with silenced or reduced expression of CAV1 mRNA. In this study, reversal of methylation and restoration of CAV1 expression were induced in OE33 cells by 5-Aza-dC treatment (Figure 4). Restoration of CAV1 mRNA expression due to 5-Aza-dC treatment is consistent with the interpretation that DNA hypermethylation silences the CAV1 gene. Although 5-Aza-dC or its derivatives have shown potential as therapeutic anticancer drugs [47–49], relatively hypomethylation of CAV1 in EAC and D vs. BE in the current study, and together with previous data on the re-expression of CAV1 in advance cancerf, would make CAV1 not an ideal molecular target for anti-cancer therapy involving demethylation in EAC patients.