This study was approved by Derbyshire Regional Ethics Committee (08/H0401/94) and Royal Derby Hospital Trust (DHRD/2008/081), and recruited patients undergoing elective laparoscopic bariatric or cholecystectomy surgery at Royal Derby Hospital. Informed written consent was obtained in accordance with Good Clinical Practice and the Declaration of Helsinki. The animals were used in accordance with the Home Office Code of Practice for the Housing and Care of Animals used in Scientific Procedures and were killed by an appropriate humane Schedule 1 method (cervical dislocation).
Three strains of male Zucker rat were used; the lean Zucker control (289 ± 6 g, 6.6 ± 0.3 mmol.L-1 glucose, n = 6), Zucker Fatty (obese, 363 ± 7 g, 10.2 ± 0.6 mmol.L-1, n = 8) and Zucker Diabetic Fatty (ZDF, obese diabetic, 325 ± 7 g, 18.3 ± 1.0 mmol.L-1, n = 8). After sacrifice, adipose tissue was immediately dissected from the subcutaneous abdominal (inguinal), visceral (perirenal) and epididymal adipose depots and immediately stored at -80°C. Blood glucose concentration was measured using a glucometer (Optium Xceed, Abbott Laboratories Ltd., UK).
Patients (23 women and 5 men, this split is representative of the fact that more women were undergoing bariatric surgery at the time) were recruited into one of three groups; those diagnosed with T2M (diabetic, n = 10); patients without diabetes, but with at least three markers of metabolic syndrome (metabolic syndrome, n = 12) ; and those without diabetes and with fewer than two markers of metabolic syndrome (‘healthy’, n = 6). Markers of metabolic syndrome included waist circumference ≥94 cm (male) or ≥80 cm (female); serum triglyceride ≥1.7 mmol.L-1; serum HDL-cholesterol <1 mmol.L-1 (male) or <1.3 mmol.L-1 (female); systolic blood pressure ≥130 mm Hg and/or diastolic blood pressure ≥85 mm Hg; and fasting serum glucose ≥5.6 mmol.L-1. Omental (n = 14) and subcutaneous adipose (n = 28) biopsies approximately 2 cm by 3 cm by 0.5 cm in size were taken atraumatically without heat coagulation. The samples were stored in ice-cold physiological salt solution before immediate transfer to the laboratory and stored within one hour at -80°C. Blood for serum glucose, insulin, triglycerides and cholesterol was processed and analysed routinely at the Royal Derby Hospital Pathology Laboratories.
Blood pressure was measured with subjects rested and supine. Anthropometric measurements performed by one trained person while the patient was standing. Waist circumference was measured at the midpoint between the iliac crest and costal margin, and hip circumference was taken at the widest point around the hips. Neck circumference was measured at the level of the cricothyroid cartilage and arm circumference was measured at the midpoint between the shoulder and elbow. Skinfold thickness was measured at 7 anatomical sites using Harpenden calipers. The 7 sites were: tricep (posterior, level with circumference), bicep (anterior, level with circumference), subscapular (parallel with inferior angle of scapular), suprailiac (immediately superior to iliac crest), abdominal (2 cm to the side of umbilicus), chest (as high as possible between anterior axillary line and nipple) and midaxillary (on midaxillary line, level of xiphoid process of sternum) .
Isolation of mature adipocytes
The method used to obtain mature adipocytes was adapted from Rodbell  as we have previously published  for both human and rat adipose samples. Adipose samples were thawed on ice, added to an equal volume of type II collagenase in phosphate buffered saline (PBS, 1 mg.ml-1, Sigma-Aldrich, UK) and digested at 37°C for 45 minutes. The samples were washed twice in PBS using centrifugation (500 x g, 2 minutes) to separate the mature adipocytes which formed a floating layer. The isolated mature adipocytes were stored at -80°C until homogenisation. Isolated mature adipocytes were homogenised in TE buffer (50 mM Tris, 1 mM EDTA, pH 7.4) using a hand-held glass homogeniser on ice. The homogenates were centrifuged (18,000 x g, 10 minutes) and the supernatant removed and spun again (20,000 x g, 30 minutes). The supernatant layer from this step was then stored at -80°C as the cytosolic fraction. The cellular pellet was homogenised in PBS (10 mM phosphate, 2.7 mM potassium chloride, 137 mM sodium chloride, pH 7.4), centrifuged (20,000 x g, 30 minutes), resuspended and stored at -80°C as the total particulate fraction.
Enzyme activity assays
Enzyme assays were carried out with minor modifications of the method of Boldrup et al.. Samples of the total cell particulate or cytosolic fraction (in duplicate) were diluted in TE buffer (fatty acid free albumin 1 mg.ml-1, pH 7.4) and at 37°C for 10 min with the FAAH inhibitor URB597 (1 μM, Sigma Chemical Company, UK) or vehicle. [3H]-AEA (American Radiolabelled Chemicals, USA) was added to a final concentration of 2 μM, and the samples were incubated at 37°C for 30 min. Activated charcoal (400 μl, 10% w/v in 0.5 M HCl) was used to stop the reaction. After brief centrifugation at approximately 13 000 x g, an aliquot of each supernatant layer was taken for scintillation counting. Tubes without homogenate were run in parallel to establish blank values. The total cell particulate or cytosolic fraction was assayed for MGL activity using the same method as above, substituting a MGL inhibitor, methylarachidonylfluorophosphonate (MAFP, 1 μM, Sigma Chemical Company, UK), and 2-oleoyl-[3H]-glycerol (2-OG, American Radiolabelled Chemicals, USA) to a final concentration of 100 μM and incubated at 37°C for 15 min. Concentrations of substrate (3H-AEA and 3H-2OG) were used at sub-saturating levels in order to allow visualisation of any changes in either substrate affinity or maximal hydrolysis rate.
GraphPad Prism software was used to analyse the data. A D’Agostino & Pearson omnibus normality test was carried out on all data and non-parametric analysis applied where appropriate. Where linear regression was used, the Pearson or Spearman correlation coefficient is reported. When three groups were compared, one-way analysis of variance (ANOVA) was used with Bonferoni multiple comparison (or Krustal Wallis for non-parametric data). For comparison of two groups, either paired or unpaired Student’s t test (or Wilcoxon signed rank test) was used as appropriate.