Secondary Logo

Journal Logo

Original Article: Gastroenterology: Celiac Disease

The Role of Cannabinoid Receptor Type 2 in the Bone Loss Associated With Pediatric Celiac Disease

Tortora, Chiara; Punzo, Francesca; Argenziano, Maura; Di Paola, Alessandra; Tolone, Carlo; Strisciuglio, Caterina; Rossi, Francesca

Author Information
Journal of Pediatric Gastroenterology and Nutrition: November 2020 - Volume 71 - Issue 5 - p 633-640
doi: 10.1097/MPG.0000000000002863

Abstract

See “Endocannabinoid System and Bone Loss in Celiac Disease: Towards a Demanding Research Agenda" by Mengoli and Di Stefano on page 583.

What Is Known/What Is New

What Is Known

  • Celiac disease predisposes to low bone mineral density and osteoporosis.
  • The cannabinoid receptor type 2 has a protective role on bone metabolism.
  • The gluten-free diet is not always enough to normalize bone mineral density.

What Is New

  • Celiac disease children have an osteoclast hyperactivation.
  • Osteoclasts from celiac disease children have a reduced cannabinoid receptor type 2 expression.
  • The cannabinoid receptor type 2 agonist, JWH-133, reduces the osteoclast hyperactivation and could be used in celiac disease patients who need active drugs on bone, in addition to the gluten-free diet.

Celiac disease (CD) is a chronic immune-mediated enteropathy that arise after gluten ingestion, in genetically predisposed individuals (1,2).

CD results in many extra-intestinal manifestations (3,4), such as low bone mineral density (BMD), osteopenia and osteoporosis (OP) with bone fractures (5–7), even though the immune response mainly affects the intestinal mucosa. More than 50% of CD children and adolescents show low BMD at diagnosis (8). Interestingly, low BMD has been also observed in asymptomatic patients (9–11).

These pathologic bone alterations have a multifactorial etiology (12); however, the pro-inflammatory cytokines released by the immune cells of the intestinal mucosa could represent the principal pathophysiological mechanism for bone alteration in CD patients (9,13). Accordingly, higher levels of serum cytokines that stimulate osteoclasts (OCs) (IL-1, IL-6, and TNF-α) and low levels of inhibitory cytokines (IL-18 and IL-12) have been detected in patients with untreated CD (14–18). Moreover, increased levels of IL-6 in active CD inversely correlated with lumbar BMD values (14). This altered cytokine release compromises bone homeostasis that is normally maintained by a dynamic equilibrium between the receptor activator of nuclear factor kappa-B ligand (RANKL) and its decoy receptor osteoprotegerin (OPG) that inhibits osteoclastogenesis and bone resorption (19). OPG/RANKL ratio is significantly lower in patients with CD than in healthy controls and positively correlates with low BMD (20).

Also, the intestinal malabsorption of calcium and vitamin D, however, appears to be related to bone alteration in CD patients (6,21,22). In particular, studies suggest the vitamin D deficiency in CD as a significant cause of low BMD (8,23–25).

To the present, a rigorous lifelong gluten-free diet (GFD) is the only efficacious treatment for CD (26–28), nevertheless, it may not effectively normalize BMD levels and reduce the risk of fracture (29,30). Considering that bone alterations in CD children could have a strong impact on their physical health and growth (31), certainly an early diagnosis of the disease is extremely important to counteract long-term complications related to bad bone health. In addition, the identification of molecular markers related to bone alteration could allow to select CD children who need not only GFD but also drugs active on bone metabolism to prevent bone damage.

In a previous study, we demonstrated a significant association between the cannabinoid receptor type 2 (CB2) Q63R variant (rs35761398) and CD, suggesting the receptor as a possible biomarker for the diagnosis of CD (32). CB2, a G protein-coupled receptor, is a member of the endocannabinoid system (33,34) predominantly expressed in peripheral tissues, such as bone, where it regulates osteoclasts and osteoblasts differentiation, survival and function (35,36).

We demonstrated that CB2-reduced expression or function results in a lower bone density and that its pharmacological modulation reduces osteoclast activity in several diseases characterized by OP (postmenopausal OP, glucocorticoid-induced OP, iron-overload-induced OP) (37–43).

Starting from these evidences, we investigated the expression of osteoclast biomarkers and CB2 receptor in OCs isolated from CD patients at diagnosis and after 1 year of GFD to highlight a possible role of the CB2 receptor in CD-related bone alterations. Moreover, we evaluated the effects of the pharmacological modulation of the CB2 receptor on osteoclast activity and compared them with those induced by 25-hydroxy vitamin D3 [25(OH)D3], to suggest the use of a CB2 agonist to reduce bone mass loss in CD patients.

METHODS

Patients

This study was conducted using OCs isolated from the peripheral blood mononucleated cells (PBMCs) of 10 CD children (median age 5 ± 3 years; 50% boys) at diagnosis, 5 CD children (4,8 ± 3,4 years; 50% boys) after 1 year of GFD and 10 healthy donors (median age 6 ± 3 years; 55% boys) age- and gender-matched. Patients and donors were recruited at the Department of Women, Child and General and Specialist Surgery of University of Campania Luigi Vanvitelli. The inclusion criteria for CD patients were: confirmed diagnosis of CD according to Espghan guidelines (44), absence of associated autoimmune diseases and the abstention from using nutritional supplements; for healthy donors were: absence of diagnosed chronic disease and gastrointestinal disease; not assumption of nutritional integration. The clinical data of CD patients are listed in Supplemental Table 1 (Supplemental Digital Content, https://links.lww.com/MPG/B893). All proceedings executed in this study were in agreement with the Helsinki Declaration of Principles, the Italian National Legislation and the Ethics Committee of the University of Campania Luigi Vanvitelli, which formally approved the study (Identification code 499, September 12, 2017). Written informed consent was obtained from parents and approval was acquired from children before any proceedings.

Osteoclasts Cell Cultures

Primary cultures of OCs were differentiated from PBMCs as previously described (45). PBMCs were isolated by centrifugation over Histopaque 1077 density gradient (Sigma Chemical, St Louis, MO), diluted at 1 × 106 cells/mL in α-Minimal Essential Medium (α-MEM) (Gibco, Uxbridge, UK) and supplemented with 10% fetal bovine serum (FBS) (Euroclone, Siziano, Italy), 100 IU/mL penicillin, 100 g/mL streptomycin (Gibco, Uxbridge, UK) and L-glutamine. To obtain differentiated OCs, the PBMCs were then cultured in 24-well plates for 21 days in the presence of 25 ng/mL recombinant human macrophage colony-stimulating factor (rh-MCSF) (Peprotech, London, UK) and 50 ng/mL RANK-L (Peprotech, London, UK). Culture medium was changed every 3 days with fresh and complete medium.

RNA Isolation, Real-time Polymerase Chain Reaction

Total RNA from OCs was isolated using Quiazol (Quiagen, Hilden, Germany) following the manufacturer's instructions and as previously described (45). SensiFAST cDNA Synthesis Kit (Bioline Meridian Bioscence, River Hills Drive Cincinnati, Ohio; Bio-Rad, Hercules, California, United States) was used to obtain from 500 ng mRNA, the first strand cDNA. The transcript levels of TRAP, Cathepsin K, and CB2 were detected by RT-qPCR using a CFX96 Real-Time PCR system (Bio-Rad, Hercules, CA) and using I-Taq Universal SYBR Green Master Mix (Bio-Rad). The cycling conditions were 10 min at 95 °C (initial denaturation) followed by 40 cycles of 15 seconds at 94 °C (denaturation) and 1 min at 68 °C (annealing/extension/data collection). The β-Actin gene was used for the normalization of the RT-qPCR products. The linearity and efficiency of the assays were analyzed through cDNA dilutions including 5 orders of magnitude. Experiments were performed in technical triplicate. The dissociation curve analysis of amplification products was performed for each PCR reaction to validate the specificity of the amplification. The 2-ΔΔCt method was used to analyze the data and to obtain the relative gene expression levels compared with the controls or the untreated samples.

Protein Isolation, Western Blot

Proteins were isolated from OCs cultures through RIPA lysis and extraction buffer (Millipore, Italia), according to the manufacturer's instructions. The Bradford dye-binding method (Bio-Rad) was used to quantify the total protein concentrations. TRAP, Cathepsin K, and CB2 expression in total lysates from OCs cultures were determined by Western blot. Twenty micrograms of denatured protein were loaded. Membranes were incubated overnight at 4 °C with rabbit polyclonal anti-TRAP (1 : 200 dilution; Santa Cruz), rabbit monoclonal anti-Cathepsin K (1 : 1000 dilution; Abcam, Cambridge, United Kingdom), rabbit polyclonal anti-CB2 (1 : 500 dilution; Elabscience) and then with the relative secondary antibody for 1 h. Reactive bands were detected by chemiluminescence (Immobilion Western Millipore) on a C-DiGit blot scanner (LI-COR Bioscences, Lincoln, Nebraska, United States). A mouse monoclonal anti-β-Tubulin antibody (1 : 5000 dilution; Elabscience) was used as housekeeping protein to verify the protein loading. Images were captured, stored and analyzed using “Image studio Digits ver. 5.0” software.

Tartrate-resistant Acid Phosphatase Assay

Tartrate-resistant acid phosphatase (TRAP) was evaluated as specific OCs activity marker and quantified through the ACP method (Takara Bio, Japan) as previously described (45). Cells were first fixed for 5 minutes at room temperature using citrate buffer pH 5.4 containing 60% acetone and 10% methanol. Then 50 μL of chromogen substrate solution (naphtol-AS-BI-phosphate substrate/fast red violet LB), mixed with 0.1 volume of sodium tartrate, was added to each well. The TRAP enzyme, cleaving the substrate, produces a red azoic dye with purplish red color that can be revealed with an optical microscope (Nikon Eclipse TS100, Nikon Instruments, Badhoevedorp, The Netherlands). TRAP(+) and multinucleated-OCs were counted in at least 3 different wells in each group of patients and treatment. To guarantee the functionality of the assay in each experiment, a positive and a negative control were included.

Bone Resorption Assay

The bone resorption assay was performed through a commercially available kit (CosMo Bio, Tokyo, Japan). OCs were differentiated from PBMC isolated from CD patients (n = 6) in calcium phosphate-coated 24 multiwell. RANK-L was used at 100 ng/mL. JWH-133 (100 nmol/L), AM630 (10 μmol/L) and 25-hydroxy vitamin D3 [25(OH)D3] (10 nmol/L) were added at day 12 for 48 h. At day 14, cells were removed using 5% sodium hypochlorite in order to visualize and count the reabsorption pits with an optical microscope (Nikon Eclipse TS100, Nikon Instruments, Badhoevedorp, The Netherlands).

Drugs and Treatments

JWH-133, AM630, and 25-hydroxy vitamin D3 [25(OH)D3] (Tocris, Avonmouth, UK) were dissolved in PBS containing DMSO. DMSO final concentration on cultures was 0.01%. OCs were treated with JWH-133 (100 nmol/L), AM630 (10 μmol/L) and 25-hydroxy vitamin D3 [25(OH)D3] (10 nmol/L) for 48 h. Nontreated cultured cells were preserved in incubation media during the same treatment time with or without vehicle (DMSO 0.01%). The concentrations of drugs were determined following concentration-response experiments and were those inducing the strongest effect without altering cells viability.

Statistical Analysis

Results are expressed as means ± standard error of the mean (SEM). The experiments were conducted in triplicate. Statistical analyses on molecular, biochemical, and cellular data were performed using the nonparametric Wilcoxon test (StatGraphics Centurion XV.II Software. Adalta, Arezzo, Italy; Statpoint Technologies Inc., VA) to calculate differences between set of paired groups. A P value ≤ 0.05 (∗) (^) was considered statistically significant.

RESULTS

Characterization of Osteoclasts Derived From Celiac Disease Patients

To evaluate mRNA and protein expression levels of specific osteoclast biomarkers, TRAP and Cathepsin K, in OCs derived from CD patients, we performed a real-time PCR (qPCR) and a western blot (WB). As expected, TRAP and Cathepsin K were highly expressed in OCs from CD patients (Fig. 1 A and C). Moreover, we quantified TRAP using the colorimetric “TRAP assay,” which identify TRAP (+) and multinucleated (n≥3) OCs. Accordingly, the assay revealed an increase in both number and size of OCs derived from CD patients (Fig. 1B). Then we compared the CB2 receptor expression of CD-OCs with CTR-OCs. Interestingly, in CD-OCs CB2 was significantly lower than CTR-OCs as demonstrated by both molecular and biochemical analysis (Fig. 1D).

F1
FIGURE 1:
Tartrate-resistant acid phosphatase (A), cathepsin K (CTK) (C) and cannabinoid receptor type 2 (D) mRNA and protein expression in osteoclasts (OCs) from 10 celiac disease patients (CD) compared with OCs from 10 healthy donors (CTR). mRNA levels were analyzed by qPCR. Five hundred nanograms of total mRNA have been used for the RT reaction. Results, normalized for the housekeeping gene β-actin, were showed as mean ± standard error of the mean (SEM) of independent experiments on each individual sample. Protein levels were determined by western blot loading 20 μg of total lysate. The most representative images are displayed. The protein bands were detected through Image Studio Digits software and the intensity ratios of immunoblots compared with CTR, taken as 1 were quantified after normalizing with respective controls. The relative quantification for TRAP, CTK, and CB2 expression, normalized for the housekeeping protein β-Tubulin, is represented in histogram as mean ± SEM of independent experiments on each individual sample. Statistical differences in TRAP, CTK, and CB2 expression were evaluated with the nonparametric Wilcoxon test. Asterisk () indicates P ≤ 0.05 compared with CTR. (B) TRAP assay on OCs from 10 CD and 10 CTR. The most representative images are displayed. TRAP (+) multinucleated (n ≥ 3) OCs, stained in purple, were counted through an AE2000 inverted microscope at 10× magnification, in at least 3 different wells for each individual sample. The percentage number of TRAP (+) cells respect to the total cell number for each sample is presented in histogram as mean ± SEM. Statistical differences were evaluated with the nonparametric Wilcoxon test. Asterisk () indicates P ≤ 0.05 compared with CTR.

Effects of JWH-133 and AM630 on Osteoclast Activity

To highlight a possible role of CB2 receptor in the pathogenesis of CD-related bone resorption, we treated in vitro CD-derived OCs with the CB2 agonist JWH-133 (100 nmol/L), and the CB2 inverse agonist AM630 (10 μmol/L), analysing the effects on OCs function and morphology. JWH-133 (100 nmol/L) induced a significant decrease of TRAP and Cathepsin K expression as demonstrated by qPCR and WB (Fig. 2A and B). Conversely the treatment with AM630 (10 μmol/L) induced an increase of both osteoclast biomarkers TRAP and Cathepsin K. According to molecular and biochemical data, also TRAP assay showed a significant decrease in OC number and activity in CD-OCs treated with the CB2 agonist JWH-133 (100 nmol/L), and an increase, although not statistically significant, after CB2 blockade with AM630 (10 μmol/L)] (Fig. 3A).

F2
FIGURE 2:
Tartrate-resistant acid phosphatase (A) and cathepsin K (CTK) (B) mRNA and protein expression in osteoclasts (OCs) from 10 celiac disease (CD) patients after 48 hours of exposure with JWH-133 (100 nmol/L) and AM630 (10 μmol/L). mRNA levels were analyzed by qPCR. Five hundred nanograms of total mRNA have been used for the RT reaction. Results, normalized for the housekeeping gene β-Actin were showed as mean ± standard error of the mean (SEM) of independent experiments on each individual sample. Protein levels were determined by western blot, loading 20 μg of total lysates. The most representative images are displayed. The protein bands were detected through Image Studio Digits software and the intensity ratios of immunoblots compared with the untreated control (NT), taken as 1, were quantified after normalizing with respective controls. The relative quantification for TRAP and CTK expression, normalized for the housekeeping protein β-Tubulin, is presented in histogram as mean ± SEM of independent experiments on each individual sample. Statistical differences in TRAP and CTK expression were evaluated with the non-parametric Wilcoxon test. indicates p ≤ 0.05 compared to NT.
F3
FIGURE 3:
(A) Tartrate-resistant acid phosphatase assay on osteoclasts (OCs) from 10 celiac disease patients (CD) after 48 h of exposure with JWH-133 (100 nmol/L) and AM630 [10 μM]. The most representative images are displayed. TRAP (+) multinucleated (n ≥ 3) OCs, stained in purple, were counted through an AE2000 inverted microscope at 10x magnification, in at least 3 different wells for each individual sample. The percentage number of TRAP (+) cells respect to the total cell number for each sample is represented in histogram as mean ± SEM. Statistical differences were evaluated with the nonparametric Wilcoxon test. Asterisk (∗) indicates P ≤ 0.05 compared with the untreated control (NT). (B) Bone resorption assay on OCs from 6 CD patients after 48 h of exposure with JWH-133 (100 nmol/L) and AM630 (10 μmol/L). The most representative images are displayed. The reabsorbed areas on the plate were visualized through an AE2000 inverted microscope at 10× magnification, in at least 3 different wells for each individual sample. The percentage number of the reabsorption pits is presented in histogram as mean ± SEM. Statistical differences were evaluated with the nonparametric Wilcoxon test. Asterisk (∗) indicates P ≤ 0.05 compared with NT.

Effects of JWH-133 and AM630 on Bone Resorption Induced by RANK-L

We also performed a Bone Resorption assay to evaluate the effects of the CB2 agonist JWH-133 (100 nmol/L) and the CB2 inverse agonist AM630 (10 μmol/L) on RANK-L-induced bone resorption, evaluating the area reabsorbed by OCs. We found that the pharmacological stimulation of CB2 with JWH-133 (100 nmol/L) exerted an inhibitory effect on RANK-L-induced bone resorption, reducing the pits area and confirming its well known anti-osteoporotic effect. The inverse agonist AM630 (10 μmol/L) increased bone resorption, even worsening the basal condition (Fig. 3B).

Effects of Gluten-free Diet on Osteoclast Activity and Cannabinoid Receptor Type 2 Expression

We analyzed the effects of a GFD on osteoclast activity, evaluating TRAP expression, number of active OCs and CB2 expression. OCs derived from CD patients on GFD, showed a significant reduction of the osteoclast marker TRAP, as demonstrated by qPCR and WB (Fig. 4A), and a reduction of OCs number and size (Fig. 4B) compared with OCs derived from CD patients at diagnosis. In accordance, CB2 receptor expression was higher in OCs from CD patients on GFD compared with OCs from CD patients at diagnosis (Fig. 4C). The basal conditions, however, were not restored as shown by the comparison with OCs derived from healthy subjects.

F4
FIGURE 4:
(A and C) Tartrate-resistant acid phosphatase and cannabinoid receptor type 2 mRNA and protein expression in osteoclasts (OCs) from 5 celiac disease patients (CD) at diagnosis and 5 CD patients on gluten-free diet (GFD) compared with OCs from 5 healthy donors (CTR). mRNA levels were analyzed by qPCR. Five hundred nanograms of total mRNA have been used for the RT reaction. Results, normalized for the housekeeping gene β-Actin were showed as mean ± standard error of the mean (SEM) of independent experiments on each individual sample. Protein levels were determined by western blot, loading 20 μg of total lysates. The most representative images are displayed. The protein bands were detected though Image Studio Digits software and the intensity ratios of immunoblots compared with CTR, taken as 1 were quantified after normalizing with respective controls. The relative quantification for TRAP expression, normalized for the housekeeping protein β-Tubulin, is represented in histogram as mean ± SEM of independent experiments on each individual sample. Statistical differences in TRAP expression were evaluated with the nonparametric Wilcoxon test. Asterisk (∗) indicates P ≤ 0.05 compared with CTR, (^) indicates P ≤ 0.05 compared with CD Diagnosis. (B) TRAP assay on OCs from 5 CD patients at diagnosis and 5 CD patients on GFD compared with OCs from 5 CTR. The most representative images are displayed. TRAP (+) multinucleated (n ≥ 3) OCs, stained in purple, were counted though an AE2000 inverted microscope at 10× magnification, in at least 3 different wells for each individual sample. The percentage number of TRAP (+) cells respect to the total cell number for each sample is represented in histogram as mean ± SEM. Statistical differences were evaluated with the nonparametric Wilcoxon test. Asterisk () indicates P ≤ 0.05 compared with CTR, (^) indicates P ≤ 0.05 compared with CD diagnosis.

Effects of JWH-133 Alone or in Combination With Vitamin D on Osteoclast activity and Bone Resorption Induced by RANK-L

We compared the effects of the Pharmacological modulation of the CB2 receptor on osteoclast activity and resorption in CD, with those induced by the 25-hydroxy vitamin D3 [25(OH)D3].

As shown, vitamin D was able to reduce osteoclast size and number (Supplemental Figure 1 A and C, Supplemental Digital Content, https://links.lww.com/MPG/B893) and the resorption pits area (Supplemental Figure 1 B and D, Supplemental Digital Content, https://links.lww.com/MPG/B893). Nevertheless, JWH-133 was more effective than vitamin D in inhibiting osteoclast activity and resorption. The anti-osteoporotic effect of JWH-133 significantly decreased when used in co-treatment with vitamin D, indicating that the presence of vitamin D might inhibit the anti-osteoporotic activity of the CB2 agonist.

DISCUSSION

Celiac disease (CD), in addition to its gastrointestinal manifestation, results in a wide range of complications (3,4). Certainly, the reduction of bone mineral density (BMD), osteopenia, osteoporosis (OP) with bone fractures have a high impact on patient's health (5–7). More than 50% of CD children and adolescents show low BMD at diagnosis (8). Malabsorption and chronic inflammation have been identified as the main causes of CD-induced bone loss (13–18,21,23,24); however, the exact mechanism remains unclear (30).

GFD is the only treatment for CD (26–28); however, GFD alone is not always sufficient to normalize BMD (12,29). There are still few available biomarkers to predict the bone response to GFD. Studies have demonstrated that OPG and propeptide of type I procollagen (PICP) serum levels could be useful to predict the post-treatment bone mass gain in adult celiac disease patients (13,46). In children, the management of bone complications is still controversial and more complex, considering that BMD influences the skeletal growth (31,47).

In 2012, we associated a CB2 functional variant with CD, suggesting the receptor as a possible biomarker for the diagnosis of the disease (32). Moreover, we widely showed that CB2 stimulation inhibits osteoclastogenesis and bone resorption in primary and secondary OP, demonstrating a protective role of the receptor in bone metabolism and suggesting it as a potential target for OP therapy (37–40,43).

Starting from these solid evidences, we hypothesized a role for CB2 in the bone loss associated with CD; therefore, we investigated the expression of osteoclast biomarkers and CB2 receptor in OCs obtained from CD patients at diagnosis and after 1 year of GFD.

Moreover, we evaluated the effects of the pharmacological modulation of the CB2 receptor on CD osteoclast activity and resorption.

Accordingly, with an increased resorption activity in CD patients, which well explains the reduction of bone mass associated with the disease, we found in CD patients an osteoclast hyperactivation and high levels of bone resorption markers, TRAP and Cathepsin K.

Interestingly, in OCs derived from CD patients, CB2 receptor was significantly lower than in healthy OCs. This result is coherent with our previous findings and also with the well-known protective role of CB2 receptor on bone metabolism (37–40), providing new insights into the pathogenesis of CD-related bone resorption. Moreover, it suggests CB2 as a possible molecular marker to predict the risk of pathological bone alterations in CD children at the moment of diagnosis.

The pharmacological modulation of CB2 with the selective agonist JWH-133 reduced the osteoclastic hyperactivation, decreasing bone marker expression, number and size of active OCs and the bone resorption area as demonstrated through 2 different in vitro assays, confirming the receptor as an important inhibitor of bone resorption. In addition, the CB2 blockade with its inverse agonist AM630 resulted in a marked increase of osteoclast activity.

In order to analyze the effects of a GFD on osteoclast activity and on CB2, we also investigated the expression of the osteoclast biomarkers and CB2 receptor in OCs obtained from CD after GFD for at least 1 year. As expected, GFD has a beneficial effect on bone health, as demonstrated by the reduction of osteoclast activity. Moreover, we found a significant CB2 expression increase in CD patients on GFD. Nevertheless, the basal condition was not restored. Although it would be interesting to evaluate this data after a longer period considering that GFD could be have a dose-dependent effect on BMD improvement, this result certainly suggests that osteoprotective treatments in patients who have a persistent increased osteoclast activity after GFD, may be necessary. Several evidences suggest that vitamin D plays a direct action on OCs reducing their resorptive capacity (48–50). Therefore, to strengthen our data, we compared the effects of the pharmacological modulation of CB2 receptor on osteoclast activity and resorption in CD, with those induced by the 25-hydroxy vitamin D3 [25(OH)D3]. According to literature, we demonstrated that vitamin D reduces osteoclast activity.

The pharmacological stimulation of the receptor CB2 with JWH-133 was significantly more effective than vitamin D. Moreover, unexpectedly, the anti-osteoporotic effect of JWH-133 significantly decreased when used in co-treatment with vitamin D, suggesting that this combination cannot prevent the bone resorption in CD. The active vitamin D metabolite 1α,25-dihydroxyvitamin D3 (1α,25(OH)2D3) is an important regulator of gene expression (51) and transcriptome-wide analysis indicated that numerous genes change their mRNA expression after Vitamin D stimulation (52,53). Moreover, Vukic et al reported the expression of Vitamin D target genes also in PBMCs (54). Therefore, it cannot be excluded that Vitamin D could decrease the expression of CB2 receptor reducing the agonist anti-osteoporotic activity. Future investigations are, however, required to clarify this interaction.

Even though different studies suggest that vitamin D deficiency in CD may be responsible of low BMD (8,23–25,55), only few clinical studies report benefits after its supplementation in CD patients (11), suggesting that probably other mechanisms are related to bone loss in pediatric CD.

Moreover, in literature, there is 1study on CD children and adolescents, which reports an increase of BMD after 2 years of vitamin D supplementation, without reaching the values of control population (11).

This study has some limitations. First, we have a relative small sample size. Nevertheless, it is recommended to have a little number of patients when investigating a new treatment approach, certainly, our results will be validated in a larger cohort. Moreover, we do not have data on the presence of bone loss at the diagnosis. No hypocalcaemia was detected in CD patients, although it is known that the impairment of bone mineral metabolism in CD could be related to several factors other than calcium and vitamin D malabsorption (56). Despite these limitations, for the first time, our study suggest, in a cohort of paediatric patients, a new disease biomarker. In clinical practice, the evaluation of CB2 receptor expression in osteoclasts for patients at risk of osteoporosis, could be more practical, less expensive, and safer compared with other methods. This data, however, needs to be confirmed in larger comparative studies.

In conclusion, our study confirmed the well-known protective role of CB2 receptor on bone metabolism, providing new insights into the pathogenesis of CD-related bone resorption. Collectively, our data suggest the necessity of an early diagnosis of the disease to counteract long-term complications related to poor bone health and surprisingly highlight CB2 as a new disease biomarker to identify CD patients at risk of pathological bone alterations. Moreover, CB2 could be a good pharmacological target to reduce bone mass loss in patients who need a direct intervention on bone metabolism, in addition to the GFD.

REFERENCES

1. Caio G, Volta U, Sapone A, et al. Celiac disease: a comprehensive current review. BMC Med 2019; 17:142.
2. Serena G, Lima R, Fasano A. Genetic and environmental contributors for celiac disease. Curr Allergy Asthma Rep 2019; 19:40.
3. Nardecchia S, Auricchio R, Discepolo V, et al. Extra-intestinal manifestations of coeliac disease in children: clinical features and mechanisms. Front Pediatr 2019; 7:56.
4. Jericho H, Guandalini S. Extra-intestinal manifestation of celiac disease in children. Nutrients 2018; 10:755.
5. Fedewa MV, Bentley JL, Higgins S, et al. Celiac disease and bone health in children and adolescents: a systematic review and meta-analysis. J Clin Densitom 2019; 23:200–211.
6. Micic D, Rao VL, Semrad CE. Celiac disease and its role in the development of metabolic bone disease. J Clin Densitom 2019; 23:190–199.
7. Webster J, Vajravelu ME, Choi C, et al. Prevalence of and risk factors for low bone mineral density in children with celiac disease. Clin Gastroenterol Hepatol 2019; 17:1509–1514.
8. Zanchetta MB, Longobardi V, Bai JC. Bone and celiac disease. Curr Osteoporos Rep 2016; 14:43–48.
9. Pantaleoni S, Luchino M, Adriani A, et al. Bone mineral density at diagnosis of celiac disease and after 1 year of gluten-free diet. ScientificWorldJ 2014; 2014:173082.
10. Drummond FJ, Annis P, O'Sullivan K, et al. Screening for asymptomatic celiac disease among patients referred for bone densitometry measurement. Bone 2003; 33:970–974.
11. Zingone F, Ciacci C. The value and significance of 25(OH) and 1,25(OH) vitamin D serum levels in adult coeliac patients: a review of the literature. Dig Liver Dis 2018; 50:757–760.
12. Krupa-Kozak U. Pathologic bone alterations in celiac disease: etiology, epidemiology, and treatment. Nutrition 2014; 30:16–24.
13. Di Stefano M, Bergonzi M, Benedetti I, et al. Alterations of inflammatory and matrix production indices in celiac disease with low bone mass on long-term gluten-free diet. J Clin Gastroenterol 2019; 53:e221–e226.
14. Fornari MC, Pedreira S, Niveloni S, et al. Pre- and post-treatment serum levels of cytokines IL-1beta, IL-6, and IL-1 receptor antagonist in celiac disease. Are they related to the associated osteopenia? Am J Gastroenterol 1998; 93:413–418.
15. Taranta A, Fortunati D, Longo M, et al. Imbalance of osteoclastogenesis-regulating factors in patients with celiac disease. J Bone Miner Res 2004; 19:1112–1121.
16. Wei S, Kitaura H, Zhou P, et al. IL-1 mediates TNF-induced osteoclastogenesis. J Clin Invest 2005; 115:282–290.
17. Garrote JA, Gomez-Gonzalez E, Bernardo D, et al. Celiac disease pathogenesis: the proinflammatory cytokine network. J Pediatr Gastroenterol Nutr 2008; 47 Suppl 1:S27–S32.
18. Tilg H, Moschen AR, Kaser A, et al. Gut, inflammation and osteoporosis: basic and clinical concepts. Gut 2008; 57:684–694.
19. Silva I, Branco JC. Rank/Rankl/opg: literature review. Acta Reumatol Port 2011; 36:209–218.
20. Fiore CE, Pennisi P, Ferro G, et al. Altered osteoprotegerin/RANKL ratio and low bone mineral density in celiac patients on long-term treatment with gluten-free diet. Horm Metab Res 2006; 38:417–422.
21. Pazianas M, Butcher GP, Subhani JM, et al. Calcium absorption and bone mineral density in celiacs after long term treatment with gluten-free diet and adequate calcium intake. Osteoporos Int 2005; 16:56–63.
22. Krupa-Kozak U, Drabinska N. Calcium in gluten-free life: health-related and nutritional implications. Foods 2016; 5:
23. Duerksen D, Pinto-Sanchez MI, Anca A, et al. Management of bone health in patients with celiac disease: practical guide for clinicians. Can Fam Physician 2018; 64:433–438.
24. Ahlawat R, Weinstein T, Pettei MJ. Vitamin D in pediatric gastrointestinal disease. Curr Opin Pediatr 2017; 29:122–127.
25. Margulies SL, Kurian D, Elliott MS, et al. Vitamin D deficiency in patients with intestinal malabsorption syndromes--think in and outside the gut. J Dig Dis 2015; 16:617–633.
26. Chellan D, Muktesh G, Vaiphei K, et al. Effect of gluten-free diet and compliance on quality of life in pediatric celiac disease patients. JGH Open 2019; 3:388–393.
27. Choung RS, Lamba A, Marietta EV, et al. Effect of a gluten-free diet on quality of life in patients with nonclassical versus classical presentations of celiac disease. J Clin Gastroenterol 2019; 54:620–625.
28. Czaja-Bulsa G, Bulsa M. Adherence to gluten-free diet in children with celiac disease. Nutrients 2018; 10:
29. Ludvigsson JF, Michaelsson K, Ekbom A, et al. Coeliac disease and the risk of fractures - a general population-based cohort study. Aliment Pharmacol Ther 2007; 25:273–285.
30. Zylberberg HM, Lebwohl B, RoyChoudhury A, et al. Predictors of improvement in bone mineral density after celiac disease diagnosis. Endocrine 2018; 59:311–318.
31. Wood CL, Stenson C, Embleton N. The developmental origins of osteoporosis. Curr Genomics 2015; 16:411–418.
32. Rossi F, Bellini G, Tolone C, et al. The cannabinoid receptor type 2 Q63R variant increases the risk of celiac disease: implication for a novel molecular biomarker and future therapeutic intervention. Pharmacol Res 2012; 66:88–94.
33. Lu HC, Mackie K. An introduction to the endogenous cannabinoid system. Biol Psychiatry 2016; 79:516–525.
34. Ligresti A, De Petrocellis L, Di Marzo V. From phytocannabinoids to cannabinoid receptors and endocannabinoids: pleiotropic physiological and pathological roles through complex pharmacology. Physiol Rev 2016; 96:1593–1659.
35. Ofek O, Karsak M, Leclerc N, et al. Peripheral cannabinoid receptor, CB2, regulates bone mass. Proc Natl Acad Sci U S A 2006; 103:696–701.
36. Zimmer A. A collaboration investigating endocannabinoid signalling in brain and bone. J Basic Clin Physiol Pharmacol 2016; 27:229–235.
37. Rossi F, Tortora C, Punzo F, et al. The endocannabinoid/endovanilloid system in bone: from osteoporosis to osteosarcoma. Int J Mol Sci 2019; 20:
38. Rossi F, Siniscalco D, Luongo L, et al. The endovanilloid/endocannabinoid system in human osteoclasts: possible involvement in bone formation and resorption. Bone 2009; 44:476–484.
39. Rossi F, Bellini G, Luongo L, et al. Endocannabinoid Research Group (Italy). The endovanilloid/endocannabinoid system: a new potential target for osteoporosis therapy. Bone 2011; 48:997–1007.
40. Rossi F, Bellini G, Tortora C, et al. CB(2) and TRPV(1) receptors oppositely modulate in vitro human osteoblast activity. Pharmacol Res 2015; 99:194–201.
41. Bellini G, Torella M, Manzo I, et al. PKCbetaII-mediated cross-talk of TRPV1/CB2 modulates the glucocorticoid-induced osteoclast overactivity. Pharmacol Res 2017; 115:267–274.
42. Rossi F, Perrotta S, Bellini G, et al. Iron overload causes osteoporosis in thalassemia major patients through interaction with transient receptor potential vanilloid type 1 (TRPV1) channels. Haematologica 2014; 99:1876–1884.
43. Rossi F, Bellini G, Luongo L, et al. The 17-beta-oestradiol inhibits osteoclast activity by increasing the cannabinoid CB2 receptor expression. Pharmacol Res 2013; 68:7–15.
44. Husby S, Koletzko S, Korponay-Szabo I, et al. European Society Paediatric Gastroenterology, Hepatology and Nutrition Guidelines for Diagnosing Coeliac Disease 2020. J Pediatr Gastroenterol Nutr 2020; 70:141–156.
45. Punzo F, Tortora C, Argenziano M, et al. Iron chelating properties of Eltrombopag: investigating its role in thalassemia-induced osteoporosis. PLoS One 2018; 13:e0208102.
46. Corazza GR, Di Stefano M, Jorizzo RA, et al. Propeptide of type I procollagen is predictive of posttreatment bone mass gain in adult celiac disease. Gastroenterology 1997; 113:67–71.
47. Di Stefano M, Mengoli C, Bergonzi M, et al. Bone mass and mineral metabolism alterations in adult celiac disease: pathophysiology and clinical approach. Nutrients 2013; 5:4786–4799.
48. Kogawa M, Findlay DM, Anderson PH, et al. Osteoclastic metabolism of 25(OH)-vitamin D3: a potential mechanism for optimization of bone resorption. Endocrinology 2010; 151:4613–4625.
49. Verlinden L, Janssens I, Doms S, et al. Vdr expression in osteoclast precursors is not critical in bone homeostasis. J Steroid Biochem Mol Biol 2019; 195:105478.
50. Zarei A, Morovat A, Javaid K, et al. Vitamin D receptor expression in human bone tissue and dose-dependent activation in resorbing osteoclasts. Bone Res 2016; 4:16030.
51. Pike JW, Christakos S. Biology and mechanisms of action of the vitamin D hormone. Endocrinol Metab Clin North Am 2017; 46:815–843.
52. Ramagopalan SV, Heger A, Berlanga AJ, et al. A ChIP-seq defined genome-wide map of vitamin D receptor binding: associations with disease and evolution. Genome Res 2010; 20:1352–1360.
53. Hossein-nezhad A, Spira A, Holick MF. Influence of vitamin D status and vitamin D3 supplementation on genome wide expression of white blood cells: a randomized double-blind clinical trial. PLoS One 2013; 8:e58725.
54. Vukic M, Neme A, Seuter S, et al. Relevance of vitamin D receptor target genes for monitoring the vitamin D responsiveness of primary human cells. PLoS One 2015; 10:e0124339.
55. Di Nardo G, Villa MP, Conti L, et al. Nutritional deficiencies in children with celiac disease resulting from a gluten-free diet: a systematic review. Nutrients 2019; 11:
56. Volkan B, Fettah A, Islek A, et al. Bone mineral density and vitamin K status in children with celiac disease: is there a relation? Turk J Gastroenterol 2018; 29:215–220.
Keywords:

cannabinoid receptor type 2; celiac disease; children; osteoclasts; osteoporosis

Supplemental Digital Content

Copyright © 2020 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition