There was an increase in the relative expression of FXR in the sequence from normal esophagus (1 ± 0.83) to esophagitis (44 ± 15, p = 7·10^-5) and to BE in which this receptor resulted most elevated (84 ± 36, p = 8·10^-4). In AC, the expression of FXR was inferior to that measured in esophagitis and BE, but it was significantly superior to that of normal esophagus (16 ± 9, p = 0.01) (Figure 1A).

The relative quantity of VDR mRNA was not significantly different in normal esophagus (1 ± 0.67) compared to esophagitis (0.78 ± 0.89, p = ns), BE (2.23 ± 1.62, p = ns) and AC (0.97 ± 0.99, p = ns), suggesting a constitutive expression of this receptor in these different esophageal conditions (Figure 1B).

The expression of FXR was significantly higher in BE-derived cells compared to AC-derived cells (1 ± 0.05 vs 9·10^-4 ± 1·10^-4, p = 3·10^-6) (Figure 2A). This difference suggests an almost complete loss of expression of FXR in esophageal AC cells.

VDR expression was inferior in AC compared to BE cells, although this difference did not reach statistical significance (0.36 ± 0.02 vs 1 ± 0.28, p = 0.06), (Figure 2B).

In tissue samples of normal esophagus (Figure 3A) there was a weak positive signal of the majority of cellular nuclei, which was limited to the basal layer of the squamous epithelium. The superficial epithelial layers and the lamina propria were characterized by a negative staining. In Barrett's esophagus (Figure 3B), there was a strongly positive nuclear staining in some areas, while other metaplastic regions remained negative. The focally positive areas were apparently randomly distributed in the crypts and the positive cells were morphologically indistinguishable from those which did not stain. Tissue from dysplasia (Figure 3C) or esophageal adenocarcinoma (Figure 3D) resulted completely negative.

Treatment of the BE-derived cells with the FXR agonist GW 4064 did not significantly affect the percentage of apoptotic cells compared to untreated cells (9.5% ± 1.6% vs. 6.9% ± 0.8%, p = ns; Figure 4A). On the other hand, treatment with the FXR antagonist guggulsterone resulted in an increased proportion of cells undergoing apoptosis (15.1% ± 3.5% vs 6.9% ± 0.8%, p = 0.02; Figure 4A), without any induction of necrosis, as assessed by flow cytometry (Figure 5).

VDR stimulation or inhibition by lithocholic acid or ZK 168281, respectively, did not significantly alter the proportion of apoptotic cells (Figure 4B).

The significant proapoptotic effect of guggulsterone on BE-derived cells (CP-18821) previously demonstrated by flow cytometry was additionally verified by measuring the caspase 3 activity. Guggulsterone treatment was associated with a significant induction of apoptosis compared to untreated cells (177 ± 15 vs 53 ± 7; p = 7·10^-4) as measured fluorometrically (Figure 6).

Finally, the induction of apoptosis was qualitatively assessed by observing the morphological changes after treatment of cells with guggulsterone (Figure 7). A considerable proportion of these cells showed shrinkage and nuclear condensation (Figure 7B). Staurosporin induced similar, but more pronounced changes (Figure 7C).

Esophageal tissue biopsies were obtained during upper digestive endoscopy from 24 patients with a normal esophagus, reflux esophagitis, BE or AC (n = 6 for each group). Tissue samples were taken during endoscopy for histological examination and additional tissue samples were stored in RNAlater™ (Qiagen, Hombrechtikon, Switzerland) and frozen at -70°C. For each patient, the diagnosis was confirmed histologically. Biopsies of BE presented no or low-grade dysplasia and all AC were moderately to poorly differentiated. The research protocol has been approved by the Ethics Committee of Geneva University Hospital, and written informed consent was obtained from all patients.

Patient's characteristics including the Reflux Symptom Index Score 26 and intake of proton pump inhibitors during the 7 days preceding endoscopy are shown in Table 1.

Primary cell cultures from BE (CP-18821) transduced with a retrovirus containing the human catalytic subunit of telomerase reverse transcriptase 27 were maintained at 37°C in 5% CO₂ atmosphere in MCDB 153 medium supplemented with 5% fetal bovine serum, 0.4 μg/mL hydrocortisone, 10^-10 mol/L cholera toxin, 140 μg/mL bovine pituitary extract, 20 μg/mL adenine, 5 μg/mL insulin, 5 μg/mL transferrin, 5 μg/mL selenium (all from Sigma, Buchs, Switzerland), 4 mmol/L glutamine, 20 ng/mL epidermal growth factor, 100 U/mL penicillin, 100 μg/mL streptomycin and 0.25 μg/mL amphotericin B (all from Gibco, Basel, Switzerland).

The human gastro-esophageal junction AC-derived cell line OACP 4C 28 and the epithelial cell line derived from human colorectal adenocarcinoma HT-29 (ECACC, Salisbury, UK) were maintained at 37°C in 5% CO₂ atmosphere in RPMI-1640 (Sigma) medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin and 100 μg/mL streptomycin (Gibco). HT-29 cells served as a positive control for the quantitative real time PCR 12.

Total RNA was extracted from tissue biopsies weighing 5-10 mg using the RNeasy Mini Kit (Sigma) according to the manufacturer's instructions. cDNA was synthesized from 1 μg total RNA with the Omniscript reverse transcriptase (Qiagen), an RNAse inhibitor and random hexamers (Promega, Wallisellen, Switzerland). A LightCycler system (Roche, Basel, Switzerland) was used for real time PCR. The relative quantification of gene expression was normalized to the endogenous ribosomal 18S RNA. Amplification was performed in a total volume of 10 μL containing 5 μL of kit-supplied QuantiTect™ Master mix 2×, 0.5 μL of specific primer mix, 0.5 μL of QuantiProbe, 3 μL RNase-free water and 1 μL of cDNA sample.

The following primers were used for FXR gene amplification: F 5'-GGACCATGAAGACCAGATT-3' and R 5'-ATGCCCAGACGGAAGTTT-3', with a 5'-GCTATGTTCCTTCGTT-3' probe. The expression of VDR was measured using the following primers: F 5'-TGACCCTGGAGACTTTGA-3' and R 5'-TCGCCTGAAGAAGCCTTT-3', with a 5'-CTGTGGGGTGTGTGGA-3' probe. PCR amplification curves were analyzed with the LightCycler Software version 3.5.28.

Resection specimens of 8 patients with Barrett's carcinoma, all containing normal esophagus, metaplasia, dysplasia and adenocarcinoma were stained. Four-micrometer paraffin sections were dewaxed, and antigen retrieval was performed in 10 mmol/L Tris-EDTA (pH 9) buffer in a microwave oven for 20 minutes at 100°C. Monoclonal mouse anti-human FXR/NR1H4 antibody (R&D Systems, Oxon UK; clone A9033A; dilution: 1:50) was applied for 1 hour at room temperature.

After washing, immunoreactivity was visualized with Envision kit (Dako B.V., Heverlee, Belgium). The sections were subsequently counterstained with Mayer hematoxylin and evaluated under a light microscopy. As positive control normal human colon was used and as negative controls the primary antibody was omitted.

BE-derived cell cultures CP-18821 (6·10⁴ cells/well) were plated for 12 hours and incubated in the presence of specific ligands for 48 hours prior to flow cytometric analysis to quantify the percentage of cells which had undergone apoptosis. The following ligands were used: FXR agonist GW4064 (GlaxoSmithKline, Research Triangle Park, USA) at a final concentration of 5 μM, FXR antagonist guggulsterone (Steraloids, Newport, USA) at 20 μM, VDR agonist lithocholic acid (Sigma) at 30 μM and VDR antagonist ZK168281 (Schering, Berlin, Germany) at 1 μM.

Following treatment with ligands, BE cells were labelled with 0.25 μg 7-amino-actinomycin D (7-AAD, BD Biosciences, Switzerland) according to the manufacturer's instructions. Flow cytometry acquisition of events was performed using the CellQuest software with a FACScan equipment (Becton Dickinson, Heidelberg, Germany). In each cell suspension, acquisition was terminated when 10,000 events were analyzed. Debris were discriminated from nonviable cells by a forward scatter (FSC)/FL-3 dot plot in which FL-3 corresponded to 7-AAD-associated fluorescence. In this plot (forward scatter [FSC; x axis] versus fluorescence [FL-3; y axis]), the combination of size and DNA fluorescence criteria provides a rationale for discrimination between debris and apoptotic cells. Fragments with very low 7-AAD fluorescence were considered to be cell debris with either no or very little DNA and were excluded from cell gates 29.

Cell cultures treated with 2 μM staurosporine (Calbiochem, Luzern, Switzerland) for 24 hours served as positive controls for apoptosis. The control for necrosis was induced in cell cultures with 0.001% sodium dodecyl sulfate for 8 min before the assay (data not shown). Additionally, the 4',6-diamino-2-phenylindole dihydrochloride (DAPI, Sigma) staining was used to confirm the induction of apoptosis: cells were fixed in 4% paraformaldehyde for 20 min, incubated with DAPI 1 μg/mL for 5 min and finally observed with a fluorescence microscope.

To confirm the data on apoptosis issued from the FACS analysis, the activity of caspase 3 was measured in cell cultures exposed to guggulsterone for 8 hours with a fluorometric immunosorbent enzyme assay kit (Roche) according to the manufacturer's instructions.

Data are presented as mean ± SD. All measurements were performed in triplicate. Comparisons between groups were performed using the Student's t-test. A p value ≤ 0.05 was considered statistically significant.