Silvana Gaetani, Editor

Subjects

The study was approved by the Haut-Lévêque Hospital Research Ethics Committee and all participants provided written informed consent.

The characteristics of the 24 subjects (12 normal weight and 12 non diabetic, insulin-resistant obese) participating to the meal study were as previously described [10]. The meal was designed by a hospital nutritionist to guarantee that, independently from the food items consumed, it contained an equal amount of macronutrients (35% lipids, 45% carbohydrates and 20% proteins) and calories. The meal consisted of a starter (a small portion of salad or vegetables), a main dish (roasted pork, lamb, chicken, salmon or tuna) accompanied by pasta or rice and vegetables, white bread and a cheese and fruit portion. Subjects randomly consumed meals containing different food items, so to avoid any potential effect on endocannabinoids precursors’ pools [24].

The lifestyle intervention program of the Haut-Lévêque Hospital, which is based on [25], [26], consists of 12 weekly sessions of group meetings with a dietician, a physical activity teacher, a psychologist and a medical doctor. The characteristics of the 12 obese subjects enrolled in the hospital lifestyle intervention program and whose salivary samples were used for the study are reported in Table 1. These 12 subjects included 4 patients with type 2 diabetes and 1 patient with fasting hyperglycemia. Eight of these patients were suffering from hypertension, while 5 had sleep apnea syndrome. Anthropometric measurements (weight, height, waist circumference) and blood pressure as well as blood samples for the analysis of biochemical parameters (blood glucose, lipid profile, liver enzymes) were collected at the beginning and at the end of the 12 weeks of the lifestyle intervention. Fasting salivary samples were also collected at the beginning and at the end of the 12 weeks. The subjects included in the study achieved a mean body weight loss of 5.33±0.97% at the end of the program.

Submandibular gland fragments and accessory salivary glands were obtained from 2 patients who respectively underwent maxillofacial surgery for suspicion of tumor and accessory salivary gland biopsy for suspicion of amyloidosis. No tissue collection was superimposed and leftover tissues were used. Only macroscopically intact tissues were used for subsequent PCR, endocannabinoids and N-acylethanolamines content analyses.

Blood and saliva sampling

For the meal study, subjects underwent blood and saliva sampling after an overnight fast, soon after the arrival at the hospital (09:00 h). Saliva samples were then also collected 1 h before the meal (11:00 h), immediately before the meal (12:00 h) and 1 h after the termination of the meal. For the lifestyle intervention study, salivary samples were collected before and after body weight loss and always after an overnight fast. Subjects were asked to wash their mouth with water and to not drink or eat anything at least half hour before spitting into a falcon 15-ml tube. In all the studies, subjects were forbidden to tooth brush on the day of the salivary sample collection, so to avoid blood contamination of the samples. The salivary samples were immediately placed at 4°C without centrifugation and then kept at −80°C until analysis. Blood samples were collected in heparin tubes, immediately centrifuged at 4°C at 1500 g for 20 min and then kept at −80°C until analysis.

Plasma lipids, glucose and liver enzymes measurements

Lipids, glucose and liver enzymes were determined by enzymatic and fluorometric methods, as described in [10].

Endocannabinoids and N-acylethanolamines measurement in plasma, saliva and salivary glands

Endocannabinoids and N-acylethanolamines were extracted, purified and quantified following a set of different biochemical steps [27]. Briefly, human plasma, saliva or submandibular salivary glands were homogenized and extracted with chloroform/methanol (21, v/v) containing internal deuterated standards (Cayman Chemicals, Ann Arbor, MI, USA) and then pre-purified by open bed chromatography if necessary. Tandem mass spectral analyses were performed on a TSQ Quantum Access triple quadrupole instrument (Thermo-Scientific, Waltham, MA, USA) equipped with an atmospheric pressure chemical ionization source and operating in positive ion mode. A new sensitive and specific liquid chromatography-tandem mass spectrometric analysis (LC-MS/MS) method was developed and validated [10]. Endocannabinoids and N-acylethanolamines were then quantified by isotopic dilution using a seven-point calibration curve and measurements expressed as pmol/ml of plasma or saliva, or fmol/mg or pmol/mg of tissue of salivary glands.

RNA extraction and PCR procedure

RNA was extracted with Tri Reagent following the manufacturer’s instructions (Fermentas, Villebon sur Yvette, France). After extraction and ethanol precipitation, RNA was treated with Turbo DNA-free (Ambion, Austin, TX, USA). The integrity of the RNA was checked by capillary electrophoresis using the RNA 6000 Nano Lab-on-a-Chip kit and the Bioanalyzer 2100 (Agilent Technologies, Palo Alto, CA, USA). cDNA was synthesized from 2 µg total RNA, with RevertAidTM Premium Reverse Transcriptase (Fermentas), using random hexamer primers and Oligo(dT)18 primer (Fermentas). The RT-PCR amplification of primers was carried out with a denaturation step at 95°C for 15 min, followed by 40 cycles of denaturation at 95°C for 30 s, primer annealing at 60°C (or 55°C depending on primer condition) for 30 s, and primer extension at 72°C for 1 min. Upon completion of the cycling steps, a final extension was done at 72°C for 5 min. RT-PCR was carried out using a PTC-200 Biorad Thermal Cycler. The primer (Eurogentec, Seraing, Belgium) sequences were the following: CB1, Forward (F): 5′ GAT GTC TTT GGG AAG ATG AAC AAG C 3′, Reverse (R): 5′ AGA CGT GTC TGT GGA CAC AGA CAT GG 3′; fatty acid amide hydrolase (FAAH), F: 5′ GTG GTG CT(G/A) ACC CCC ATG CTG G 3′, R: 5′ TCC ACC TCC CGC ATG AAC CGC AGA CA 3′; monoacylglycerol lipase (MAGL), F: 5′ TCT TCC TTC TGG GCC ACT CCA 3′, R: 5′ GGA TTG GCA AGA ACC AGA GG 3′; diacylglycerol lipase alpha (DAGL alpha) F: 5′ GCC ATC TTC CTC TTT CTC CTG 3′, R: 5′ GAG CAC GTA CTG CAT GGA GTC 3′; diacylglycerol lipase beta (DAGL beta) F: 5′ AAG CTG CCA GAT ACG TTT ACC 3′, R: 5′ GGA GAA GGC GTA GCA CCT GAC 3′; N-acyl phosphatidylethanolamine phospholipase D (NAPE-PLD) F: 5′ TCA AAC AGT AGA ACA GTG TGT ACG 3′, R: 5′ TTG AAT GAA AAA TGC TTA TGT ATT G 3′; N-acylethanolamine-hydrolyzing acid amidase (NAAA) F: 5′ AGC AGA CAT TTG GCC TCT AG 3′, R: TTC TCC GGT CAT CTT CCT T 3′; glyceraldehyde-phosphate dehydrogenase (GAPDH), F: 5′ CCC TTC ATT GAC CTC AAC TAC ATG GT 3′, R: 5′ GAG GGG CCA TCC ACA GTC TTC TG 3′; hypoxanthine phosphoribosyl transferase (HPRT), F: 5′ CTT GCT CGA GAT GT(C/G) ATG AAG 3′, R: 5′ TGC ATT GTT TT(A/G) CCA GTG 3′. Housekeeping genes GAPDH and HPRT were used as internal controls. PCR products were analyzed by electrophoresis on a 1.8% agarose gel.

Statistical analysis

Statistical analysis was performed using Statistica version 8.0 (StatSoft, Tulsa, OK, USA). All values are reported as means ± SEM. For the meal study, data were analyzed by 1-way and 2-way repeated measurements ANOVA or by two-tailed Student’s t-tests when appropriate. For the lifestyle intervention study, data were analyzed with paired t-test or wilcoxon test if data were not normally distributed. Significant ANOVAs were followed by Tukey post-hoc test. For correlation analysis, Pearson’s or Spearman’s were used when appropriate. p values less than 0.05 denote statistical significance.

Endocannabinoids and N-acylethanolamines can be detected in saliva

Salivary endocannabinoids and N-acylethanolamines were reliably detectable and quantifiable in human saliva (Figure 1). Fasting saliva endocannabinoids (AEA and 2-AG) and N-acylethanolamines (PEA and OEA) levels were significantly higher in obese, insulin-resistant subjects as compared to normal weight controls (Figure 1). Similar changes were also found when salivary endocannabinoid and N-acylethanolamine measurements were expressed as pmol/mg of total salivary lipids (data not shown). Conversely, in the same subjects, only fasting AEA and OEA plasma levels were higher in obese than in controls (Figure 1). Interestingly, compared to normal weight subjects, the increase was significantly higher in obese salivary than plasma samples for all the compounds analyzed but OEA (Figure 1).

There were no significant correlations between fasting salivary 2-AG and N-acylethanolamines and their respective plasma levels. However, a trend towards a significant correlation was found between salivary and plasma AEA levels (r=0.39, p=0.068).

Similarly to what has been previously reported for plasma measurements [7], [9], [10], fasting salivary AEA and OEA levels were significantly correlated with BMI, waist circumference, and fasting insulin (Figure 2). Fasting salivary 2-AG levels positively correlated with waist circumference (r=0.50, p=0.002) and insulin (r=0.48, p=0.027), while fasting salivary PEA did not correlate with any of these variables (data not shown).

The ECS is present in human salivary glands

Gene expression analysis of samples from human submandibular and salivary accessory glands revealed that MAGL, DAGL alpha and beta, FAAH and CB1 mRNAs were all expressed in saliva-producing organs (Figure 3A). Similarly, gene expression of enzymes involved in the synthesis of AEA, OEA and PEA, such as NAPE-PLD [28], and in their degradation, such as NAAA [29], were also present in salivary glands (Figure 3B). However, while MAGL, DAGL alpha and beta mRNAs were highly expressed, particularly in the accessory glands, mRNA levels of expression of CB1, FAAH, NAPE-PLD and NAAA were relatively low. AEA, 2-AG, PEA and OEA were also detectable in the gland tissues (Figure 3C).

Salivary endocannabinoids and N-acylethanolamines meal-related responses

Normal weight and obese subjects had similar energy intake (normal weight: 2851.21±37.68 kJ vsobese: 2876±87.92 kJ, p=0.8). Whatever the time point analyzed, salivary AEA, PEA and OEA levels were significantly higher in obese than in normal weight subject (Figure 4). However, salivary endocannabinoids and N-acylethanolamines levels did not change in response to the meal, neither in normal weight nor in obese (Figure 4).

Salivary endocannabinoids and N-acylethanolamines weight loss-related responses

The characteristics of the 12 obese subjects before and after the 12-weeks lifestyle intervention program are shown in Table 1. The compliance to the program led to significantly reduced body weight, BMI, waist circumference as well as systolic blood pressure (Table 1). It also significantly ameliorated fasting blood glucose, triglycerides, aspartate aminotransferase, alanine aminotransferase and gamma-glutamyl transferase (Table 1). Notably, at the end of the 12-weeks program the subjects had a significant reduction of fasting salivary AEA levels (p<0.05, Figure 5A), while salivary 2-AG, OEA and PEA did not significantly change (Figure 5B–D). Studies carried out in animal models suggest that AEA and OEA might have opposing roles in energy balance regulation [1], [2], [30], [31], thus the relationship between the ratio of salivary OEA/AEA levels and the percentage of body weight loss was also analyzed, but only a non-significant trend for a positive correlation was found (r: 0.44, p=0.151).

We thank the Gene Profiling Platform of the NeuroCentre Magendie, M. Vallée, S. Vitiello, C. Majoufre-Lefebvre and the nurses of the Endocrinology Department at the Haut-Lévêque Hospital for help with the study; we thank the subjects that have participated to the study.

This work was supported by INSERM/AVENIR (DC and GM), INSERM/ interface (DC), Aquitaine Region (DC and GM), French Society of Endocrinology (BGC), Fondation Recherche Médicale, EU-FP7 HEALTH-F2-2008-223713 and EU-FP7 ENDOFOOD ERC-2010-StG (GM). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.