Investigations of the human endocannabinoid system ...

International Journal of Obesity (2011) 35, 1377–1384 & 2011 Macmillan Publishers Limited All rights reserved 0307-0565/11 www.nature.com/ijo

ORIGINAL ARTICLE Investigations of the human endocannabinoid system in two subcutaneous adipose tissue depots in lean subjects and in obese subjects before and after weight loss MF Bennetzen1, N Wellner2, SS Ahmed2, SM Ahmed2, TA Diep2, HS Hansen2, B Richelsen1 and SB Pedersen1 1

Department of Endocrinology C, Aarhus University Hospital, Aarhus C, Denmark and 2Department of Pharmacology and Pharmacotherapy, Faculty of Pharmaceutical Sciences, University of Copenhagen, Copenhagen Ø, Denmark

Context: Endocannabinoids (ECs) have a role in obesity by affecting appetite and through peripheral effects. Obesity is associated with a dysregulation of the endocannabinoid system (ECS). Objective: We aimed to determine the ECS in subcutaneous adipose tissue (AT) in obese subject and investigate the influence of diet-induced weight loss on this system. Design: The obese study participants underwent a 12 weeks diet regimen resulting in 10–12% weight loss. All study participants underwent fasting blood samples and AT biopsies from abdomen and gluteal region, the obese subjects both before and after weight loss. Setting and participants: A total of 21 healthy obese individuals (10 men/11 women, age 39.5±1.6 years, body mass index (BMI): 37.5±0.8 kg m2) and 21 age- and gender-matched lean subjects (BMI: 23.8±0.4 kg m2) were studied. Main outcome measures: The activity of ECS in AT was determined by measuring arachidonoyl glycerol (2-AG) and N-arachidonoylethanolamine/anandamide in AT by mass spectrometry and gene expressions of enzymes and receptors involved in the ECS. Results: The EC, 2-AG was reduced in obese individuals in the gluteal AT depot (Po0.01). Moreover, 2-AG increased in both depots in the obese subjects following weight loss (Po0.05). The gene expression of the CB1 was either not affected by the obese state (in the gluteal AT depot) or reduced (in the abdominal depot, Po0.05) and significantly affected by weight loss. The expression of the degrading enzymes FAAH, FAAH2, MGL and MGL2 was differently affected by obesity, AT depot and weight loss. Conclusion: We found reduced levels of 2-AG in subcutaneous AT in obesity, which increased after weight loss. In abdominal AT, the low CB1 expression was normalised after weight loss, whereas in gluteal AT the CB1 expression was reduced after weight loss. These findings support the concept of a dysregulated ECS in AT in association with obesity. International Journal of Obesity (2011) 35, 1377–1384; doi:10.1038/ijo.2011.8; published online 15 February 2011 Keywords: endocannabinoids; CB1; weight loss; adipose tissue

Introduction Obesity is associated with a general over-activity of the endocannabinoid system (ECS) when measuring levels of

Correspondence: Dr MF Bennetzen, Department of Endocrinology, Aarhus Sygehus, Tage Hansensgade 2, Building 3, 2nd floor, Aarhus C 8000, Denmark. E-mail: [email protected] Received 17 August 2010; revised 21 December 2010; accepted 23 December 2010; published online 15 February 2011

arachidonoyl glycerol (2-AG) and N-arachidonoylethanolamine/anandamide (AEA) in plasma.1–4 As endocannabinoids (ECs) are produced on demand in specific tissues,5 what is measured in the circulation is probably spillover of unknown origin. Previous studies indicate a dysregulation of the ECS in human obesity with differentially regulated levels in various AT depots.6–8 Therefore, it will be of considerable interest to study the activity of the ECS in the adipose tissue (AT) to understand the importance of this system in lean and obese subjects.

The endocannabinoid system in adipose tissue MF Bennetzen et al

1378 The ECS consists of the cannabinoid receptors, the endogenous ligands for these receptors (the ECs) and the enzymes that produce and degrade these ligands. At least two cannabinoid receptors have been characterised9 cannabinoid receptor 1 (CB1) and cannabinoid receptor 2 (CB2), but other receptors may be involved as well.10 The CB1 is extensively distributed in the central nervous system, but it is also present in many tissues with relations to energy metabolism.11 The CB2 is most common in the immune cells,12 but it is also present in several other tissues including AT.13 The two most widely studied ECs are 2-AG and AEA, which belongs to different groups of lipids termed as monoacylglycerols and N-acylethanolamines, respectively.14 The ECs function mainly as neurotransmitters, and because of their highly lipophilic nature, they cannot be stored in vesicles and have to be produced on demand.5 2-AG is synthesised by diacylglycerol lipase and degraded by monoglyceride lipase (MGL),15 whereas AEA can be produced by N-acyl phosphatidylethanolamine-specific phospholipase D (NAPE-PLD) and degraded by fatty acid amide hydrolase (FAAH).16 Several forms of this enzyme exist, both FAAH2 and N-acylethanolamine-hydrolysing acid amidase but their roles are less characterised.14 2-AG can also be degraded by FAAH,17,18 and genetic variations in the gene that encodes FAAH has been associated with overweight and obesity.4 The ECS has been found to be a considerable participant in the control of energy balance and metabolism through both central and peripheral mechanisms.19 Activation of the cannabinoid receptors is followed by increased food intake and body weight.20,21 Blockage of the ECS leads to decreased body weight and improved metabolic profile.22 The peripheral effects of the ECS have been studied less than the central effects of the system.19 Roche et al.13 have established that the CB1 receptor is expressed in both pre-adipocytes and mature adipocytes and in visceral AT (VAT) as well as in subcutaneous AT (SAT), and we have previously shown a lower level of CB1 expression in VAT compared with SAT in lean subjects and similar levels in the two AT depots in obese individuals.23 CB1 antagonism leads to increased lipolysis in white AT,24 whereas CB1 stimulation increases glucose uptake in adipocytes.25 As mentioned, there is a hypothesis that obesity is characterised by a dysregulation of the ECS. AT per se may be involved in the health complications of obesity, hence, it is of interest to characterise the ECS in human AT and its activity in AT in obesity. To advance our perspective of the implications of the ECS in human AT, we investigated SAT from two depots in lean control subjects and obese subjects before and after diet-induced weight loss. The levels of 2-AG and AEA in AT was determined by liquid chromatography mass spectrometry (LC–MS). Furthermore, the gene expression of CB1 and the synthesising and degrading enzymes of the system were investigated, and the possible impact of gender on this system in AT was determined. International Journal of Obesity

Materials and methods Study participants The body mass index (BMI) of obese group was above 35 kg m2 and age between 20 and 50 years. Exclusion criteria were diabetes (fasting plasma glucose above 7 mmol l1), malignant disease, pregnancy or medication known to affect AT metabolism within the previous 3 months. The lean group was matched to the obese group according to gender and age. BMI of the lean group should be between 22 and 27 kg m2. All study participants signed written informed consent before investigations.

Study design The obese group underwent clinical investigations at baseline and after 10 weeks (8 weeks weight loss by diet followed by 2 weeks weight stabilising period with the study participants’ normal diet adjusted to the new weight). The weight loss of 10–12% was achieved by intake of a liquid very low calorie diet with 600 calories per day containing protein 41 g, carbohydrate 29 g and fat 5.6 g per 100 g, (NUPO nutritional powder, Flex pack Industry A/S, Greve, Denmark) plus 200 g of fruit and vegetables daily, and weekly motivating sessions with a dietician. The lean participants were investigated only once. The evening before clinical investigations (blood samples on one day, liposuctions on another) they all had the same meal (calorically adjusted for gender) and they fasted over night. They were also asked not to engage in excessive physical exercise or alcohol intake the day before and the morning of the clinical investigations. The study was approved by the Central Denmark Region Committees on Biomedical Research Ethics and the Danish Data Protection Agency.

Investigations Anthropometrics (BMI, waist- and hip-circumference) and bioimpedance to measure the fat mass were determined. AT samples were obtained by liposuction as previously described26 from abdomen, subcutaneously (subcutaneous abdominal adipose tissue, SAAT) just below the level of the umbilicus in the midline, and from the gluteal region (subcutaneous gluteal adipose tissue, SGAT), between the greater trochanter and tuberositas ischii. This was performed under local anaesthesia with Lidocain 10 mg ml1 without adrenalin. Fasting blood samples were drawn from all study participants and were analysed according to standard procedure at Department of Clinical Biochemistry at Aarhus University Hospital (glucose and lipids) except serum insulin, which were measured by ELISA (DAKO K6219, Electra Box Diagnostics Aps, Rødovre, Denmark) according to manufacturer’s manual. The kit had a lower detection limit of 32.6 pmol l1 and the absorbance was measured with a spectra reader photometer. The homoeostasis model assessment was used


1379 to evaluate insulin resistance according to Matthew et al.27 by the following formula: (fasting serum insulin (mU l1) fasting plasma glucose (mmol l1)/22.5), and 6 as the conversion factor for insulin from pmol l1 (SI unit) to mU l1.28

Measurements of 2-AG and AEA by LC–MS AEA and 2-AG were extracted from the AT tissue (0.1–0.6 g) and analysed using the procedure described previously by Artmann et al.29 In short, the tissues were weighed and ethyl acetate/hexane (9:1) mixture was added to give a final volume of 40 ml per g of tissue. All samples were added to 500 pmol deuterium-labelled internal standards (2H-AEA, 2 H-2-AG; Cayman Chemicals, Ann Arbor MI, USA), before homogenisation. Lipids were purified using solid phase extraction (Strata Si-1 Silica 55 mm, 80A, 200 mg per 3 ml, Phenomenex, Birkeroed, Denmark) and the eluate was analysed by LC–MS (Hewlett Packard 1100 Series single quadrupole HPLC/MS system equipped with a (150 2.00 mm, 3 m) Phenomenex HyperClone ODS column, Phenomenex).

Real-time reverse transcription PCR for mRNA analysis AT samples (200 mg) were homogenised in TriZol reagent (Gibco BRL, Life Technologies, Roskilde, Denmark) and total RNA was extracted following the manufacturer’s protocol. RNA was quantitated by measuring absorbency at 260 and 280 nm and the ratio was 1.8 or higher. The integrity of the RNA was checked by visual inspection of the two ribosomal RNAs on an ethidium bromide stained agarose gel. Reverse transcription was performed using random hexamer primers at 42 1C for 30 min and 92 1C for 2 min followed by 4 1C as described by the manufacturer (verso cDNA Kit from ABgene, Surrey, UK). Then, PCR-mastermix containing the specific primers and Taq DNA polymerase (KAPA SYBR FAST qPCR Kit, Kapabiosystems, Boston, MA, USA) were added. The following primers were designed using the primer analysis software Oligo version 6.64 (Molecular Biology Insights, Inc., Cascade, CO, USA). b2microglobulin sense: 50 -AATGTCGGATGGATGAAACC-30 and antisense 50 -TCTCTCTTTCTGGCCTGGAG-30 , 128 bp; FAAH sense: 50 -GGGCCGTCAGCTACACTATGC-30 and antisense 50 -ATGTTCCATCTGGGCCTCGTC-30 , 101 bp; FAAH2 sense: 50 -CGCTAGGCTTTCTCATAGGC-30 and antisense: 50 -C CGAAAGCAGAAGCAATGGTT-30 , 111 bp; MGL sense: 50 -GGT GTGCGCGGAGCTAGTTTC-30 and antisense: 50 -AGCGGCGC TGCGATTCTC-30 , 117 bp; MGL2 sense: 50 -CTGGCCCGCCGCA AACGA-30 and antisense: 50 -TCTTCAGGTCCGGGGCCACG A-30 , 74 bp; NAPE-PLD sense: 50 -TGGTGACCTCCCGTCTCT-30 and antisense: 50 -CCTCAGCCTCCCAAGTACCTG-30 , 99 bp. Real-time quantitation of target gene to housekeeping gene was performed with a SYBR-Green real-time PCR assay using an ICycler from Bio-Rad (Copenhagen, Denmark). The housekeeping gene was tested for stability in the present intervention. The target and housekeeping gene were

amplified in separate tubes. The increase in fluorescence was measured in real-time during the extension step. The threshold cycle (Ct) was calculated, and the relative gene expression was calculated essentially as described in the User Bulletin #2, 1997 from Perkin Elmer (Perkin Elmer Cetus, Norwalk, CT, USA). Briefly, the target gene (X0) to b2microglobulin (R0) ratio in each sample before amplification was calculated as X0/R0 ¼ k 1/((2**DCt)), DCt is the difference between Ct-target and Ct-reference, and k is a constant, set to 1. All samples were amplified in duplicate.

Statistical analysis Data are expressed as mean±s.e.m. The three groups were compared by one-way repeated-measurements analysis of variance unless otherwise stated. If results were significant we performed post hoc test by Fishers least significant difference method between groups, with focus on lean versus obese at baseline or obese at baseline versus obese after weight loss. If variables were not normally distributed, comparisons were analysed by Kruskal–Wallis rank sum test. Relationships between EC and various enzymes were assessed by Pearson’s correlation. Po0.05 is considered statistically significant. We used the statistical package Sigmaplot 11.0 with incorporated sigmastat (Systat software, Inc., Richmond, CA, USA) for performing statistical analysis

Results Anthropometric and metabolic measures Four subjects from the obese group left the study on their own request (three men and one woman). In total, we have tissue samples from 11 obese women before and 10 after the intervention, whereas 10 male subjects were investigated before and 7 after. As expected from the matched design there were no differences between lean and obese subjects according to age or sex distribution, but a significant difference in BMI and fat mass (Table 1). In the obese group, the weight loss amounted to 14.8±5.9 kg for men and 10.1±0.9 kg for women, that is B12 and 10% of initial body weight, respectively, and correspond to about 21% reduction of the fat mass. All study participants lost weight (min: 6 kg per 7% of baseline weight, max: 19.8 kg per 17% of baseline weight). As shown in Table 1 the expected differences in glucose, insulin and lipids were observed between the lean and obese groups. After weight loss in the obese group lower levels of insulin (Po0.01), homoeostasis model assessment (Po0.05) and total cholesterol (Po0.05) were found.

Levels of 2-AG and AEA in AT These ECs were determined by LC–MS in AT taken from both the abdominal and gluteal fat depot. As compared with lean subjects the amount of 2-AG in obese individuals was International Journal of Obesity


1380 Table 1

2-AG

Anthropometric and metabolic measures

n Gender (male/female) Age PE, hours per week Height, m Weight, kg BMI, kg m2 WC, cm HC, cm W/H-ratio Lean mass, kg Fat mass, kg P-glucose, mmol l1 S-insulin, pmol l1 HOMA-IR Total cholesterol, mmol l1 HDL, mmol l1 LDL, mmol l1 Triglycerides, mmol l1

Lean

Obese, baseline

Obese, after weight loss

21 10/11 39.5±1.8 5.0±0.7 1.77±0.02 74.4±2.0** 23.8±0.4** 81.1±1.5** 98.9±1.3** 0.8±0.02** 56.1±2.0* 18.3±1.2** 5.3±0.1** 48.2±3.9** 2.2±0.2** 4.8±0.05* 1.5±0.05** 2.8±0.2* 1.0±0.08**

21 10/11 39.5±1.6 3.1±0.7 1.73±0.02 112.1±3.7 37.5±0.8 117.6±2.5 117.7±1.4 1.0±0.02 64.4±2.7 47.7±1.8 5.8±0.1 93.8±7.8 4.0±0.4 5,4±0.3 1.2±0.06 3.3±0.2 2.1±0.2

17 7/10 38.3±1.8 2.8±0.5 1.71±0.02 97.4±3.1** 33.3±0.9** 101.9±2.6** 109.8±1.4** 0.9±0.02** 60.0±2.5 37.4±1.8** 5.6±0.1 67.7±7.6* 2.9±0.3* 4.9±0.2* 1.1±0.06 2.9±0.2* 1.9±0.2

Abbreviations: BMI, body mass index; HC, hip circumference; HDL, highdensity lipoprotein cholesterol; HOMA-IR, homeostasis model assessmentinsulin resistance; PE, physical exercise; P-glucose, fasting plasma glucose; LDL, low-density lipoprotein cholesterol; S-insulin, fasting serum insulin; WC, waist circumference; W/H-ratio, waist hip ratio. Data are presented as mean±s.e.m., P-value: one-way repeated-measurements analysis of variance followed by Fischers least significant difference method post hoc on lean versus obese at baseline or obese at baseline versus obese after weight loss, *Po0.05, **Po0.001.

5000 4000 3000

*

*

*

2000 1000 0 SGAT

SAAT AEA

120

pmol/g adipose tissue

Group

pmol/g adipose tissue

6000

100 80 60 40 20

significantly lower in the gluteal depot (35% lower, Po0.05) and also a tendency to lower levels in the abdominal fat depot were found (reduced by 19%, P ¼ 0.09; Figure 1). After diet-induced weight loss in the obese subjects the level of 2-AG significantly increased in both AT depots (Po0.05, Figure 1). The level of AEA was neither affected by the obese state nor by weight loss (Figure 1). No influence of gender on 2-AG or AEA in AT was found (data not shown).

Gene expression of enzymes and CB1 in AT levels Synthesising and degrading enzymes. Similar gene expression levels of FAAH and FAAH2 (degrading enzymes of both EC’s) were found between lean and obese at baseline except from an increase in FAAH2 in SAAT of obese at baseline compared with lean (P ¼ 0.003, Figure 2). In SGAT, both FAAH and FAAH2 were decreased after weight loss (Po0.001 and Po0.05, respectively). Also the expression of MGL and MGL2 (degrading enzymes of 2-AG) were reduced after weight loss in the SGAT depot. Lower levels of MGL were observed in obese compared with lean subjects in the SAAT depot. MGL2 was similarly expressed in SAAT between all three groups (Figure 2). Concerning NAPE-PLD (synthesising enzyme for AEA), we found higher gene expression in SAAT of obese at baseline compared with lean (P ¼ 0.004), but no change after weight loss and similar levels in all three groups in SGAT (data not shown). International Journal of Obesity

0 SAAT

SGAT

Figure 1 Levels of 2-AG and AEA in SAAT and SGAT. Levels of 2 arachidonoyl glycerol (2-AG) (a) and anandamid (AEA) (b) in SAAT and SGAT in lean subjects, obese subjects at baseline and in obese subjects after weight loss. 2-AG levels (pmol) per gram adipose tissue in SAAT and SGAT of lean (open bars, n ¼ 9–16), obese at baseline (black bars, n ¼ 18–20) and obese after weight loss (grey bars, n ¼ 13–15) measured in whole adipose tissue by LC–MS. Data presented as mean±s.e.m., *Po0.05. SAAT, subcutaneous abdominal adipose tissue; SGAT, subcutaneous gluteal adipose tissue.

CB1 CB1 was highly expressed in lean compared with obese subjects in SAAT (Po0.01) and weight loss was followed by an increase in CB1 expression in obese subjects (Po0.001, Figure 3). CB1 expression in SAAT and SGAT does not behave similarly, as CB1 was expressed to the same degree in SGAT in lean and obese subjects and was reduced in SGAT after weight loss only in the obese group (Po0.01). Comparing the abdominal and gluteal AT depots, we found similar CB1 levels in lean subjects, but higher levels in the gluteal compared with the abdominal depot in obese subjects at baseline (P ¼ 0.002, paired t-test). Concerning the impact of gender, the expression of CB1 was found to be higher in both AT depots in obese men compared with obese women (SAAT Po0.05, SGAT Po0.05, data not shown), with no gender differences among the lean subjects.


1381 FAAH

FAAH2

0.0030

0.0030

0.0025

0.0025 0.0020

*

mRNA

mRNA

0.0020 0.0015 0.0010

0.0010

0.0005

0.0005 SAAT

SGAT

SAAT

MGL

0.010

SGAT MGL2

0.30 0.25

*

*

0.004

mRNA

0.008 mRNA

*

0.0000

0.0000

0.006

*

0.0015

0.20

*

0.15 0.10

0.002

0.05

0.000

0.00 SAAT

SGAT

SAAT

SGAT

Figure 2 Enzyme mRNA levels (FAAH, FAAH2, MGL and MGL2) in SAAT and SGAT. Gene expression of FAAH (a), FAAH2 (b), MGL (c) and MGL2 (d) measured by PCR in relation to b2microglobulin from SAAT and SGAT from lean subjects (open bars, n ¼ 21), obese subjects at baseline (black bars, n ¼ 21) and obese subjects after weight loss (grey bars, n ¼ 17). Data presented as mean±s.e.m. *Po0.05.

CB1

0.004

mRNA

0.003

*

**

*

0.002

(data not shown). Although weight loss enhanced 2-AG in both SAAT and SGAT there was no correlation between the weight loss or change in fat mass and the increment of 2-AG in the two depots. 2-AG was not correlated with the changes in metabolic factors associated with the weight loss (data not shown). A negative correlation existed between 2-AG and FAAH2 levels (correlation coefficient –0.3, Po0.05), but not between 2-AG and MGL or MGL2 expression.

0.001

Discussion 0.000 SAAT

SGAT

Figure 3 CBR1 mRNA levels in SAAT and SGAT. Gene expression of CBR1 in relation to b2microglobulin from SAAT and SGAT from lean (open bars, n ¼ 21), obese at baseline (black bars, n ¼ 21) and obese after weight loss (grey bars, n ¼ 17). Data presented as mean±s.e.m. *Po0.05, **Po0.001.

Correlations There was a positive correlation between expression of CB1 and 2-AG levels in the SAAT depot (correlation coefficient 0.4, Po0.001), but no association in the gluteal depot or to AEA levels. Neither 2-AG nor AEA in AT were at baseline correlated to fat mass or any of the other anthropometric measurements

In the present study, we demonstrated the presence of a complete ECS (ligands, receptors and biosynthetic/degradation enzymes) in human AT, which is in agreement with recent findings.1,30 Moreover, it was demonstrated that this ECS in AT was influenced by the obese state, by weight loss, and that SAT (abdominal versus gluteal) does not behave similar to weight loss. Several research groups find increased circulating EC levels with obesity.1–3 These groups also find 2-AG to be the EC changing (the most) with weight changes. However, because the ECs are produced on demand and tightly regulated, it is unknown whether circulating levels provide useful information on ECS status in relevant tissues. We aimed to investigate ECS status in human SAT in the obese state, and our findings are compatible with the suggestion of a dysregulated ECS in the AT. This conclusion is built on our findings of lower levels of 2-AG in AT in the obese International Journal of Obesity


1382 individuals, low levels that increase/normalise in association with weight loss, and low levels of CB1 expression in SAAT in the obese group, which also increases after weight loss. The finding of low levels of ECs in SAT from obese subjects is well in line with studies in obese rodents,31,32 demonstrating a pronounced reduction in subcutaneous EC in obese rodents, whereas the change in ECs in VAT was more modest after diet-induced obesity. Animal studies do indicate hyperactivity of the ECS in obesity in the intra-abdominal (epididymal) AT depot.30,33 Also human studies support the idea that the over-activation of the ECS is in the VAT.6–8 Recent investigations of the ECs in SAT by Annuzzi el al.34 in a small sample of obese and lean individuals did not detect significant difference between lean and obese subjects, whereas obese with type 2 diabetes had significant higher AEA and lower 2-AG levels. Concerning the lean and obese individuals their findings were in good agreement with our findings in a large sample size. Thus, the increased level of 2-AG in plasma in obesity is not caused by increased spillover from the SAT. However, it could be caused by increased activity in the intra-abdominal depot, and several studies have described an association between amount of VAT and plasma 2-AG levels.2,3 This is supported by Matias et al.,30 who performed peripheral measurements of EC in VAT. They revealed higher 2-AG levels in the VAT from obese subjects compared with lean, whereas they found higher levels of both 2-AG and AEA in serum from overweight subjects with diabetes, compared with healthy subjects. Di Marzo7 was the first to suggest that an upregulation of the ECS in VAT and a downregulation of the system in SAT in obesity and diabetes might lead to excessive accumulation of VAT at the expense of SAT possibly resulting in development of diabetes and atherosclerosis. Previous studies with 5% weight loss did not show any difference in circulating EC levels, AT CB1 or AT FAAH mRNA gene levels.1,35 A study by Di Marzo et al.36 investigated EC levels in viscerally obese men, who underwent 1 year lifestyle modifications. The intervention resulted in a weight loss of almost 7% of initial body weight, and following this life style intervention, circulating levels of 2-AG and AEA were significantly reduced (62.3 and 7%, respectively), which was associated with the decrease in VAT and not to change in SAT, which is in accordance with our results. The reduced expression of CB1 in obese subjects, at least in the SAAT depot, is in accordance with our recent study, in which the amount of CB1 was determined by both mRNA and at the protein level (western blot) and this level was also found to be reduced in obese as compared with lean subjects.23 We found a reduction in the degrading enzymes for both 2-AG (MGL, MGL2, FAAH and FAAH2) and AEA (FAAH and FAAH2) in SGAT with weight loss, but interestingly only an association between 2-AG and FAAH2. The reduction in enzyme levels could explain the increase in 2-AG levels in SGAT following weight loss, but the actions International Journal of Obesity

on the enzymes may not be as simple as previously anticipated.37 Pagano et al.25 investigated the differential levels of CB1 expression in both SAAT and SGAT. They found lower CB1 expression in SGAT of obese compared with lean individuals and higher CB1 expression in SAAT of obese compared with lean. Our results confirm these and other previous findings of increased CB1 expression in the SAAT of lean compared with obese subjects.1,2,6 Pagano et al.25 also investigated synthesising and degrading enzymes and found downregulation of FAAH in SGAT and, similar to us, upregulation of FAAH in SAAT of obese compared with lean and higher NAPE-PLD-levels in SGAT of obese in both AT compartments. The previously mentioned study by Engeli et al. also showed a reduction in CB1 and FAAH expression in SAAT with obesity, and they found a negative correlation between circulating EC’s and FAAH expression in AT.38 Some limitations in the present study should be mentioned. First, the ECS was investigated in two SAT depots and not in the potential metabolically more important VAT depot. On the other hand, CBR1 in VAT in human studies also seems to be reduced in obesity2 and in our recent study23 we found reduced CB1 protein in VAT as compared with SAT but ECs were not measured in that study. Moreover, whole AT was investigated containing several other cell types than adipocytes and accordingly we do not known the cell types of most importance for ECS in AT. Overall it appears, that the ECS is dysregulated in association with obesity, with increased formation of ECs in the visceral depot and unaltered or increased CB1 receptor expression in the same depot, whereas in the subcutaneous abdominal adipose tissue (SAAT) from obese subjects, the production of 2-AG is reduced, as is CB1. We investigated two different SAT depots, and were able to confirm our previous results of higher CB1 expression in gluteal AT compared with abdominal AT in obese subjects.23 In addition, we found a different response to weight loss in the two depots; in gluteal adipose tissue FAAH and FAAH2 was reduced after weight loss, whereas in abdominal adipose tissue the expression of these enzymes were increased. Also for CB1 expression, the response to weight loss was opposite in the two depots; in abdominal adipose tissue CB1 expression increased after weight loss, whereas it decreased in gluteal adipose tissue after weight loss. In conclusion, the amount of 2-AG was reduced in SAT from obese subjects and 2-AG increased/normalised after weight loss. Moreover, the expression of CB1 was reduced in SAAT in obese individuals. Thus, both one of the important endogenous ligands, 2-AG, and the main EC receptor were downregulated in the obese state, indicating the EC tone in SAAT is reduced in obesity, but can be increased (normalised) by a weight loss. In addition, weight loss caused opposite regulation of the ECS in SAAT compared with SGAT, which might be important for the AT metabolism in the two depots.


1383

Conflict of interest The authors declare no conflict of interest.

Acknowledgements We thank all study participants. We greatly appreciate the excellent technical assistance of Lenette Pedersen and Pia Hornbek, Department of Endocrinology, Aarhus University Hospital. The study was supported by the Danish Medical Research Council, Aarhus University, Danish Obesity Research Centre, DanORC (Danish Strategic Research Council), and by the UNIK project: Food, Fitness, and Pharma.

References 1 Engeli S, Bohnke J, Feldpausch M, Gorzelniak K, Janke J, Batkai S et al. Activation of the peripheral endocannabinoid system in human obesity. Diabetes 2005; 54: 2838–2843. 2 Bluher M, Engeli S, Kloting N, Berndt J, Fasshauer M, Batkai S et al. Dysregulation of the peripheral and adipose tissue endocannabinoid system in human abdominal obesity. Diabetes 2006; 55: 3053–3060. 3 Cote M, Matias I, Lemieux I, Petrosino S, Almeras N, Despres JP et al. Circulating endocannabinoid levels, abdominal adiposity and related cardiometabolic risk factors in obese men. Int J Obes (Lond) 2007; 31: 692–699. 4 Sipe JC, Scott TM, Murray S, Harismendy O, Simon GM, Cravatt BF et al. Biomarkers of endocannabinoid system activation in severe obesity. PLoS ONE 2010; 5: e8792. 5 Maccarrone M, Gasperi V, Catani MV, Diep TA, Dainese E, Hansen HS et al. The endocannabinoid system and its relevance for nutrition. Ann Rev Nutr 2010; 30: 423–440. 6 Sarzani R, Bordicchia M, Marcucci P, Bedetta S, Santini S, Giovagnoli A et al. Altered pattern of cannabinoid type 1 receptor expression in adipose tissue of dysmetabolic and overweight patients. Metabolism 2009; 58: 361–367. 7 Di Marzo V. The endocannabinoid system in obesity and type 2 diabetes. Diabetologia 2008; 51: 1356–1367. 8 Bordicchia M, Battistoni I, Mancinelli L, Giannini E, Refi G, Minardi D et al. Cannabinoid CB1 receptor expression in relation to visceral adipose depots, endocannabinoid levels, microvascular damage, and the presence of the Cnr1 A3813G variant in humans. Metabolism 2010; 59: 734–741. 9 Pertwee RG, Howlett AC, Abood ME, Alexander SPH, Di Marzo V, Elphick MR et al. International Union of Pharmacology. Cannabinoid Receptors and their Ligands: Beyond CB1 and CB2. Pharmacol Rev 2010; 62: 588–631. 10 Lauckner JE, Jensen JB, Chen HY, Lu HC, Hille B, Mackie K. GPR55 is a cannabinoid receptor that increases intracellular calcium and inhibits M current. Proc Natl Acad Sci USA 2008; 105: 2699–2704. 11 Matias I, Di Marzo V. Endocannabinoid synthesis and degradation, and their regulation in the framework of energy balance. J Endocrinol Invest 2006; 29(3 Suppl): 15–26. 12 Howlett AC, Barth F, Bonner TI, Cabral G, Casellas P, Devane WA et al. International Union of Pharmacology. XXVII. Classification of cannabinoid receptors. Pharmacol Rev 2002; 54: 161–202. 13 Roche R, Hoareau L, Bes-Houtmann S, Gonthier MP, Laborde C, Baron JF et al. Presence of the cannabinoid receptors, CB1 and CB2, in human omental and subcutaneous adipocytes. Histochem Cell Biol 2006; 126: 177–187.

14 Ahn K, McKinney MK, Cravatt BF. Enzymatic pathways that regulate endocannabinoid signaling in the nervous system. Chem Rev 2008; 108: 1687–1707. 15 Sugiura T, Kishimoto S, Oka S, Gokoh M. Biochemistry, pharmacology and physiology of 2-arachidonoylglycerol, an endogenous cannabinoid receptor ligand. Prog Lipid Res 2006; 45: 405–446. 16 Hansen HS, Diep TA. N-acylethanolamines, anandamide and food intake. Biochem Pharmacol 2009; 78: 553–560. 17 De Petrocellis L, Di Marzo V. An introduction to the endocannabinoid system: from the early to the latest concepts. Best Pract Res Clin Endocrinol Metab 2009; 23: 1–15. 18 Kirkham TC, Tucci SA. Endocannabinoids in appetite control and the treatment of obesity. CNS Neurol Disord Drug Targets 2006; 5: 272–292. 19 Rosenson RS. Role of the endocannabinoid system in abdominal obesity and the implications for cardiovascular risk. Cardiology 2009; 114: 212–225. 20 Greenberg I, Kuehnle J, Mendelson JH, Bernstein JG. Effects of marihuana use on body weight and caloric intake in humans. Psychopharmacology (Berl) 1976; 49: 79–84. 21 Williams CM, Kirkham TC. Anandamide induces overeating: mediation by central cannabinoid (CB1) receptors. Psychopharmacology (Berl) 1999; 143: 315–317. 22 Curioni C, Andre C. Rimonabant for overweight or obesity. Cochrane Database Syst Rev 2006; CD006162. 23 Bennetzen MF, Nielsen TS, Paulsen SK, Bendix J, Fisker S, Jessen N et al. Reduced cannabinoid receptor 1 protein in subcutaneous adipose tissue of obese. Eur J Clin Invest 2010; 40: 121–126. 24 Jbilo O, Ravinet-Trillou C, Arnone M, Buisson I, Bribes E, Peleraux A et al. The CB1 receptor antagonist rimonabant reverses the diet-induced obesity phenotype through the regulation of lipolysis and energy balance. FASEB J 2005; 19: 1567–1569. 25 Pagano C, Pilon C, Calcagno A, Urbanet R, Rossato M, Milan G et al. The endogenous cannabinoid system stimulates glucose uptake in human fat cells via phosphatidylinositol 3-kinase and calcium-dependent mechanisms. J Clin Endocrinol Metab 2007; 92: 4810–4819. 26 Christiansen T, Paulsen SK, Bruun JM, Ploug T, Pedersen SB, Richelsen B. Diet-induced weight loss and exercise alone and in combination enhance the expression of adiponectin receptors in adipose tissue and skeletal muscle, but only diet-induced weight loss enhanced circulating adiponectin. J Clin Endocrinol Metab 2010; 95: 911–919. 27 Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28: 412–419. 28 Syste´me International (SI) units for plasma, serum, or blood concentrations. Diabetes 1993; 42: 370–371. 29 Artmann A, Petersen G, Hellgren LI, Boberg J, Skonberg C, Nellemann C et al. Influence of dietary fatty acids on endocannabinoid and N-acylethanolamine levels in rat brain, liver and small intestine. Biochim Biophys Acta 2008; 1781: 200–212. 30 Matias I, Gonthier MP, Orlando P, Martiadis V, De PL, Cervino C et al. Regulation, function, and dysregulation of endocannabinoids in models of adipose and beta-pancreatic cells and in obesity and hyperglycemia. J Clin Endocrinol Metab 2006; 91: 3171–3180. 31 Starowicz KM, Cristino L, Matias I, Capasso R, Racioppi A, Izzo AA et al. Endocannabinoid dysregulation in the pancreas and adipose tissue of mice fed with a high-fat diet. Obesity (Silver Spring) 2008; 16: 553–565. 32 Izzo AA, Piscitelli F, Capasso R, Aviello G, Romano B, Borrelli F et al. Peripheral endocannabinoid dysregulation in obesity: relation to intestinal motility and energy processing induced by food deprivation and re-feeding. Br J Pharmacol 2009; 158: 451–461. 33 Bensaid M, Gary-Bobo M, Esclangon A, Maffrand JP, Le FG, Oury-Donat F et al. The cannabinoid CB1 receptor antagonist

International Journal of Obesity


1384 SR141716 increases Acrp30 mRNA expression in adipose tissue of obese fa/fa rats and in cultured adipocyte cells. Mol Pharmacol 2003; 63: 908–914. 34 Annuzzi G, Piscitelli F, Di Marino L, Patti L, Giacco R, Costabile G et al. Differential alterations of the concentrations of endocannabinoids and related lipids in the subcutaneous adipose tissue of obese diabetic patients. Lipids Health Dis 2010; 9: 43. 35 Engeli S, Heusser K, Janke J, Gorzelniak K, Batkai S, Pacher P et al. Peripheral endocannabinoid system activity in patients treated with sibutramine. Obesity (Silver Spring) 2008; 16: 1135–1137.

International Journal of Obesity

36 Di Marzo V, Cote M, Matias I, Lemieux I, Arsenault BJ, Cartier A et al. Changes in plasma endocannabinoid levels in viscerally obese men following a 1 year lifestyle modification programme and waist circumference reduction: associations with changes in metabolic risk factors. Diabetologia 2009; 52: 213–217. 37 Di Marzo V, Maccarrone M. FAAH and anandamide: is 2-AG really the odd one out? Trends Pharmacol Sci 2008; 29: 229–233. 38 Engeli S, Bohnke J, Feldpausch M, Gorzelniak K, Janke J, Batkai S et al. Activation of the peripheral endocannabinoid system in human obesity Diabetes 2005; 54: 2838–2843.