Anandamide (AEA) is the prominent member of the endocannabinoid family and its biological action is mediated through the binding to both type-1 (CB₁) and type-2 (CB₂) cannabinoid receptors (CBR). The presence of AEA and CBR in the gastrointestinal tract highlighted their pathophysiological role in several gut diseases, including celiac disease. Here, we aimed to investigate the expression of CBR at transcriptional and translational levels in the duodenal mucosa of untreated celiac patients, celiac patients on a gluten-free diet for at least 12 months and control subjects. Also biopsies from treated celiac patients cultured ex vivo with peptic-tryptic digest of gliadin were investigated. Our data show higher levels of both CB₁ and CB₂ receptors during active disease and normal CBR levels in treated celiac patients. In conclusion, we demonstrate an up-regulation of CB₁ and CB₂mRNA and protein expression, that points to the therapeutic potential of targeting CBR in patients with celiac disease.

Ethics statement

This study was approved by the Ethical Committee of Spedali Civili of Brescia. Informed written consent to participate in this study was given by patients.

Patients and tissues

Biopsy samples were collected from the second part of the duodenum of 16 patients affected by uncomplicated UCD (9 males and 7 females, mean age 40.6 years, range 19–71). The diagnosis was based on positivity of serum antiendomysial antibodies associated with the typical histopathological lesions, namely villous atrophy, increased intraepithelial lymphocyte infiltration and crypt hyperplasia. Six patients showed a Marsh IIIc lesion and 10 showed a Marsh IIIb lesion. Duodenal biopsies were also collected from 17 patients affected by uncomplicated celiac disease on a gluten-free diet for at least 12 months (7 males and 10 females, mean age 32.4 years, range 18–81), all negative for serum antiendomysial antibodies and with a substantially normal duodenal mucosal architecture. Finally, duodenal biopsies were also collected from 19 CS (8 males and 11 females, mean age 46.7 years, range 18–69) undergoing endoscopy for functional dyspepsia, negative for antiendomysial antibodies and with normal histology. Some of the biopsies were processed for routine histology or were embedded in OCT Tissue-Tek (Sakura Finetek, Torrance, CA, USA) snap frozen and then stored at −70°C; others were used for organ culture or were homogenized for immunoblotting analysis.

Organ culture

Biopsy specimens from TCD patients, placed on grids in the central well of an organ culture dish, were cultured in an airtight container with 95% O₂/5% CO₂ at 37°C in the absence or presence of 1 mg/ml PT-gliadin (Frazer III fraction, Sigma-Aldrich) in RPMI-1640 medium (Gibco, Invitrogen, Paisley, UK) supplemented with 10% HL-1 (Lonza BioWhittaker, Verviers, Belgium), 100 U/ml penicillin and 100 mg/ml streptomycin [16]. After 24 h culture, biopsies were snap frozen and stored at −70°C.

Quantitative real-time reverse transcriptase-polymerase chain reaction (qRT-PCR) analysis

Total RNA was extracted from biopsies using the RNeasy extraction kit (Qiagen, Crawley, UK), as suggested by the manufacturer. QuantiTect Reverse Trascription kit (Qiagen, Crawley, UK) was used to produce cDNA from 1 µg of purified RNA. 40 ng of the first strand of cDNA was used for amplification (in triplicate) in 15 µl reaction solution, containing 7.5 µl QuantiFast SYBR Green PCR (Qiagen) and 10 pmol of each primer (Table 1). The following qRT-PCR program was used: 95°C for 5 min PCR initial activation step; 40 amplification cycles at 95 °C for 10 s, 60°C for 30 s. The target transcripts were amplified by means Opticon™ 2 continuous fluorescence detection system (MJ Research, San Francisco, CA, USA). β-Actin was used as housekeeping gene for quantity normalization.

Table 1

Oligonucleotide sequences of CB₁, CB₂ and β-actin used for detection by qRT-PCR analysis.

Western blotting

Western blotting was performed according to standard procedures [17]. In brief, tissue samples were lysed in ice-cold lysis buffer (10 mM EDTA, 50 mM pH 7.4 Tris-HCl, 150 mM sodium chloride, 1% Triton-X-100, 2 mM phenylmethylsulfonyl fluoride, 2 mM sodium orthovanadate, 10 mg/ml leupeptin and 2 mg/ml aprotinin) and the amount of protein was determined by the Bio-Rad Protein assay (Bio-Rad Laboratories, Hemel Hempstead, UK). Equivalent amounts of protein were loaded in each lane and run on 10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis under reducing conditions. Proteins were transferred to nitrocellulose membranes (Bio-Rad Laboratories, Hercules, California, USA), that were blocked with 10% non-fat dried milk and 5% bovine serum albumin for 2 h, and then incubated overnight at 4°C with the rabbit polyclonal antibodies specific for CB₁ (1200 dilution) or CB₂ (11000 dilution) receptors (both from Abcam, Cambridge, UK). Membranes were rinsed and incubated with the appropriate horseradish peroxidase-conjugated secondary anti-rabbit antibody (diluted 14000; Dako, High Wycombe, UK) in blocking solution. Detection was performed using ECL Plus detection reagents (Amersham, Little Chalfont, UK). Blots were stripped and analyzed for β-actin, as an internal loading control, using a rabbit anti-human β-actin antibody (Abcam). Protein expression levels were quantified by densitometric analysis, using the ImageJ software after quantity normalization with β-actin.

Confocal microscopy

Immunofluorescence staining of 4 µm thick cryostat sections of OCT-embedded biopsy specimens fixed in cold acetone was performed using the same anti-CB₁ or CB₂ antibodies used in Western blotting (1100 dilution), followed by a FITC-conjugated secondary goat anti-rabbit antibody (1500 dilution; Abcam). Nuclei were counterstained by DAPI (11000 dilution; Sigma-Aldrich, Poole, UK). Appropriate isotype control antibody was included on parallel sections. The sections were mounted with coverslips using Glycergel Mounting Medium (Dako), and were analyzed using a laser scanning confocal microscope (FluoView FV1000; Olympus, Center Valley, PA, USA). Images (1,024×1,024 pixels) were acquired using an oil immersion lens (60×1.4 NA Plan-Apochromat; Olympus).

Enzyme-linked immunosorbent assay (ELISA)

Biopsy homogenates (20 µg/well) were coated overnight and were reacted with rabbit anti-CB₁ or anti-CB₂ polyclonal antibodies (1250 dilution) (both from Cayman Chemicals, Ann Arbor, MI, USA). Goat anti-rabbit antibody conjugated to horseradish peroxidase (15000 dilution, Santa Cruz Biotechnology, Santa Cruz, CA, USA) was used as second antibody, and color development of the horseradish peroxidase reaction was followed at 405 nm using 1-Step Ultra TMB ELISA substrate (Pierce, Rockford, IL, USA).

Receptor binding assay

CBR functional activity was evaluated by using the MultiScreenHTS 96-well Plates for binding assays [18]. Briefly, biopsies were homogenated in 50 mM Tris–HCl (pH 7.4) and 20 µg of lisate were incubated with [³H]CP55.940 (2.5 nM). Incubations were performed in 0.2 ml reaction buffer (50 mM Tris-HCl, 2 mM Tris-EDTA, 3 mM MgCl₂, 5 mg/ml BSA, pH 7.4). The filters were washed and transferred to vials, containing 0.1% Triton X-100 (0.5 ml) in 3.5 ml liquid scintillation cocktail (Ultima Gold XR, Perkin Elmer Life Sciences, Boston, MA, USA). Unspecific binding was determined in the presence of cold agonists (1 µM CP55940), and all binding data were expressed as fmol per mg of protein.

Statistical analysis

Results were expressed as mean±SEM of at least triplicate experiments. Data were analyzed by means of Prism 5 program (GraphPad Software, San Diego, CA, USA) using the one-way analysis of variance (ANOVA) followed by Bonferroni’s post hoc analysis. A level of p<0.05 was considered statistically significant.

In situ mucosal CB₁ and CB₂ expression by immunofluorescence

We first determined the expression of CB₁ (Figure 1A) and CB₂ (Figure 1B) by confocal immunofluorescence in the duodenal mucosa of UCD and TCD patients, and CS. Numerous CB₁– and CB₂-positive cells were evident both in the epithelium and lamina propria of UCD patients, while in TCD patients and CS positivity for CB₁ and CB₂ was limited to a few mononuclear cells scattered in the lamina propria.

Figure 1

CB₁ and CB₂ immunofluorescence by confocal miscroscopy.

Mucosal CB₁ and CB₂ mRNA and protein levels

CB₁ and CB₂ mRNA and protein expression was investigated through qRT-PCR (Figure 2A) and ELISA (Figure 2B), respectively. These analyses demonstrated the presence of both CBR in biopsies collected from the duodenum of UCD, TCD and CS. Remarkably, CB₁ mRNA levels increased significantly in the mucosa of UCD and TCD patients compared to CS (p<0.0001), although CB₁ decreased in patients after remission following a gluten-free diet (p<0.001). Additionally, the expression of CB₂ mRNA was almost 10-fold higher in UCD patients than in healthy mucosa (p<0.0001), and was lower than that of TCD patients (p<0.0001). In keeping with the qRT-PCR results, ELISA analysis revealed an higher expression of CB₁ protein in UCD patients with respect to CS (p<0.05), and lower CB₁ protein levels in TCD patients compared with UCD subjects (p<0.01). Incidentally, no statistically significant difference was found between TCD and CS groups. CB₂ protein expression showed a similar trend compared to CB₁ protein, showing an increased level in UCD patients compared with CS (p<0.0001) and a significant reduction in TCD patients (p<0.0001), where CB₂ protein levels were comparable to those of CS.

Figure 2

CB₁ and CB₂ gene and protein expression levels.

CBR binding assay

Binding assays were performed with the synthetic agonist [³H]CP55.940, that has high affinity for both CB₁ and CB₂ receptors [4]. The results reported in Figure 2C show that biopsies obtained from the three patients’ groups were able to bind the radioligand to similar extents: 186±21 fmol per mg of protein (CS), 200±54 fmol per mg of protein (UCD) and 193±32 per mg of protein (TCD).

Ex vivo mucosal CB₁ and CB₂ expression

To investigate the effect of PT-gliadin on mucosal CB₁ and CB₂ expression, we measured by immunoblotting CB₁ and CB₂ protein levels in mucosal biopsies from TCD patients cultured ex vivo in the absence or presence of PT-gliadin. Mucosal biopsies cultured with PT-gliadin showed significantly (p<0.001) higher levels of CB₁ (up to ~4-fold) in comparison to biopsies treated with medium only (Figure 3A, B). Similarly, mucosal biopsies cultured with PT-gliadin showed significantly (p<0.001) higher levels of CB₂ in comparison to biopsies treated with medium only (Figure 3C, D).

Figure 3

Ex vivo effect of PT-gliadin on mucosal expression of CB₁ and CB₂.

The authors would like to thank all patients who agreed to give biopsies and tissue samples, without whom this study would not have been possible.