The effects of obesity, diabetes and metabolic syndrome on the hydrolytic enzymes of the endocannabinoid system in animal and human adipocytes

Abstract

Background: Circulating endocannabinoid levels are increased in obesity and diabetes. We have shown that fatty acid amide hydrolase (FAAH, an endocannabinoid hydrolysing enzyme) in subcutaneous adipose tissue positively correlates with BMI in healthy volunteers. The aim of the present study was to investigate whether the hydrolytic enzymes of the endocannabinoid system are affected by diabetes or metabolic syndrome in obesity.

Methods: Using radiolabelled substrates, FAAH and monoacylglycerol lipase (MGL) activities were assessed in adipocytes from various adipose depots in Zucker rats (n = 22, subcutaneous abdominal, visceral and epididymal) and bariatric patients (n = 28, subcutaneous abdominal and omental).

Results: FAAH activity was significantly increased in adipocytes of obese (Zucker Fatty) compared to Zucker lean rats (P < 0.05) but was not raised in the Zucker Diabetic Fatty rats (ZDF). MGL activity was raised in both Zucker Fatty (P < 0.001-0.01) and ZDF rats (P < 0.05) and was positively correlated with body weight and plasma glucose levels (P < 0.01). In bariatric patients (BMI range 37-58 kg.m²), there was a trend for MGL activity to correlate positively with BMI, reaching significance when type 2 diabetic patients were removed. FAAH and MGL activities in obese humans were not correlated with blood pressure, skinfold thicknesses, fasting glucose, insulin, HbA1c, triglycerides or cholesterol levels.

Conclusions: FAAH in adipocytes is differentially altered in animal models of obesity and diabetes, while MGL activity is increased by both. However, in obese humans, FAAH or MGL activity in adipocytes is not affected by diabetes, dyslipidaemia or other markers of metabolic dysfunction. This suggests increased circulating levels of endocannabinoids are not a result of altered degradation in adipose tissue.

Figure 1

Endocannabinoid degradation enzyme assay validation. FAAH activity was detected in the total particulate and not cytosolic fraction of adipocyte homogenates of Zucker rats (A, n = 48), and was significantly inhibited by URB597 (1 μM, C, n = 48). MGL activity was detected in the cytosolic and not total particulate fraction of adipocyte homogenates (B, n = 33), and was significantly inhibited by MAFP (1 μM, D, n = 66). Data are given as means with error bars representing S.E.M and were analysed by a Mann–Whitney test (*** P < 0.001, *** P < 0.001).

Figure 2

Endocannabinoid degradation enzymes in lean and obese Zucker rats. FAAH and MGL activities in mature adipocytes isolated from subcutaneous (A,D), abdominal (B,E) and epididymal (C,F) adipose tissue depots in lean (n = 6), obese (n = 8) and obese diabetic Zucker rats (n = 8). Data are given as means, with error bars representing S.D., and were analysed by Krustal Wallis with comparisons between the means of all data (* P < 0.05, ** P < 0.01, *** P < 0.001).

Figure 3

Correlation studies in Zucker rats. Correlative studies between body weight and enzyme activity in mature adipocytes isolated from each adipose tissue depot in abdominal (A,D), subcutaneous (B,E), and epididymal (C,F) adipocytes. When the Zucker diabetic rats were omitted from analysis, a significant correlation was observed between FAAH activity and plasma glucose levels in the abdominal (G) and subcutaneous (H) but not epididymal adipocytes (I). Stronger correlations were also seen between MGL activity and plasma glucose levels in all adipose depots (J,K,L). The Spearman correlation coefficient is reported. Key: green, lean rats; red, obese rats; black, obese diabetic rats.

Figure 4

Endocannabinoid degradation enzymes in different adipose depots. MGL activity in paired samples of adipose tissue from lean Zucker rats (A). FAAH (B) and MGL (C) activities in paired samples of subcutaneous and visceral mature adipocytes from obese humans (n = 14). Data are presented as scatterplots and were analysed using Krustal Wallis with comparisons between the means of all data (A) or by Wilcoxon matched-pairs signed rank test (B and C, *P < 0.05).

Figure 5

Endocannabinoid degradation enzymes in obese humans. The relationship between FAAH and MGL activity and BMI in the whole population (A) and in the non-diabetic patients (B), and between FAAH and MGL activity and neck circumference (C) and abdominal skinfold thickness in the female only population (D).

Figure 6

Endocannabinoid degradation enzymes in metabolic syndrome and diabetes. The effects of fasting serum glucose concentration (A,D), HbA1c (B,E) and insulin concentration (C,F) on FAAH and MGL activities in subcutaneous adipocytes from obese humans. Data are presented as scatterplots, with error bars representing S.D., and were analysed using Student’s t test or linear regression. FAAH (G) and MGL (H) activities in subcutaneous adipocytes from obese humans assigned to one of three groups based on the following criteria: healthy <2 components of metabolic syndrome (n = 6, triangles); metabolic syndrome ≥3 components of metabolic syndrome (n = 11, squares); diagnosed type 2 diabetes with or without metabolic syndrome (n = 10, circles). Data are presented as scatterplots, with error bars representing S.D., and were analysed using one way ANOVA and Bonferroni post hoc test.

Figure 7

Endocannabinoid degradation enzymes and blood lipids. The effects of fasting serum triglyceride concentration (A,D), total cholesterol concentration (C,F), and HDL cholesterol concentration (B,E) on FAAH and MGL activities in subcutaneous adipocytes from obese humans. Data are presented as scatterplots, with error bars representing S.D., and were analysed using Student’s t test.