Methylxanthines Inhibit Primary Amine Oxidase and Monoamine Oxidase Activities of Human Adipose Tissue

Abstract

Background: Methylxanthines including caffeine and theobromine are widely consumed compounds and were recently shown to interact with bovine copper-containing amine oxidase. To the best of our knowledge, no direct demonstration of any interplay between these phytochemicals and human primary amine oxidase (PrAO) has been reported to date. We took advantage of the coexistence of PrAO and monoamine oxidase (MAO) activities in human subcutaneous adipose tissue (hScAT) to test the interaction between several methylxanthines and these enzymes, which are involved in many key pathophysiological processes. Methods: Benzylamine, methylamine, and tyramine were used as substrates for PrAO and MAO in homogenates of subcutaneous adipose depots obtained from overweight women undergoing plastic surgery. Methylxanthines were tested as substrates or inhibitors by fluorimetric determination of hydrogen peroxide, an end-product of amine oxidation. Results: Semicarbazide-sensitive PrAO activity was inhibited by theobromine, caffeine, and isobutylmethylxanthine (IBMX) while theophylline, paraxanthine, and 7-methylxanthine had little effect. Theobromine inhibited PrAO activity by 54% at 2.5 mM. Overall, the relationship between methylxanthine structure and the degree of inhibition was similar to that seen with bovine PrAO, although higher concentrations (mM) were required for inhibition. Theobromine also inhibited oxidation of tyramine by MAO, at the limits of its solubility in a DMSO vehicle. At doses higher than 12 % v/v, DMSO impaired MAO activity. MAO was also inhibited by millimolar doses of IBMX, caffeine and by other methylxanthines to a lesser extent. Conclusions: This preclinical study extrapolates previous findings with bovine PrAO to human tissues. Given that PrAO is a potential target for anti-inflammatory drugs, it indicates that alongside phosphodiesterase inhibition and adenosine receptor antagonism, PrAO and MAO inhibition could contribute to the health benefits of methylxanthines, especially their anti-inflammatory effects.

Keywords: DMSO; adipose tissue; caffeine; copper-containing amine oxidase; hydrogen peroxide; methylxanthine; monoamine oxidase; semicarbazide; theobromine.

Figure 1

Influence of DMSO, ethanol, and semicarbazide on 1 mM benzylamine oxidation by human adipose tissue homogenates. Assays were performed at 37 °C as described in Materials and Methods with hSCAT homogenates and 1 mM benzylamine as the control and the indicated final concentration of: DMSO: dimethyl sulfoxide; EtOH: ethanol; Semi: semicarbazide. The highest tested dose of DMSO (3700 mM) corresponds to its final concentration when used in pure form as a vehicle to solubilize a xanthine at 10 mM, then testing it at 2.5 mM final. The amount of H₂O₂ generated during 30-min oxidation of benzylamine (1.40 ± 0.20 nmoles/mg protein/min) was set at 100% (dotted line), while baseline with only hScAT was set at 0 %. Mean ± SEM of n = 7–27 different individual preparations. Different from control at: * p < 0.05; *** p < 0.001 by one-way ANOVA test.

Figure 2

Dose-dependent inhibition of benzylamine oxidation by methylxanthines. The control consisted of a standard assay of PrAO-dependent oxidation of 1 mM benzylamine (BZA), performed in triplicate for each of the 27 individual tissue samples, and set at 100%. Mean ± SEM of the number of observations indicated for each agent, added over a concentration range from 1 µM to 2.5 mM: 7-methylxanthine: 8–20; Paraxanthine: 8–10; Theophylline: 8–20; Caffeine: 8–24; Theobromine: 8–20; IBMX: 8–14. The differences from control (plotted as 0 on the X-axis) at: * p < 0.05; ** p < 0.01; *** p < 0.001 were analyzed by one-way ANOVA. In most cases, the SEM bars lie within the symbol.

Figure 3

Inhibition by methylxanthines of the methylamine oxidation by human PrAO activity. The control condition consisted of the oxidation of 0.5 mM methylamine by hScAT homogenates during 30 min at 37 °C and was set at 100% with the baseline of hydrogen peroxide release set at 0%. Each methylxanthine was added at 1 mM final and preincubated for 10 min before the addition of methylamine. Each column is the mean + SEM of seven individuals, each performed in duplicate. Different from control (dark column) at: * p < 0.05; ** p < 0.01; *** p < 0.001 by one-way ANOVA.

Figure 4

Inhibition of hScAT PrAO by theobromine. The concentration of benzylamine varied between 0.1 and 1.0 mM in the presence of increasing concentrations of theobromine: 0, 0.1, 1.0 and 2.5 mM. The concentration of theobromine used were: 0 mM (diamonds), 0.1 mM (squares), 1.0 mM (diamonds), and 2.5 mM (circles).

Figure 5

Interaction between DMSO vehicle and tyramine oxidation by human abdominal ScAT. Assays were performed at 37 °C with homogenates and 1.0 mM tyramine as a control plus the indicated final concentration of: DMSO, Parg: pargyline; Semi: semicarbazide. The highest tested dose of DMSO (3700 mM) corresponds to its final concentration (25% v/v) when used as the vehicle for solubilization of 2.5 mM theobromine. The amount of H₂O₂ generated during 30-min oxidation of tyramine (averaging 0.42 nmoles/mg protein/min) was set at 100% (dotted line), with the baseline set at 0%. Mean ± SEM of n = 8–12 different individual preparations. Different from control at: *** p < 0.001, determined by one-way ANOVA.

Figure 6

Dose-dependent inhibition of tyramine oxidation by methylxanthines. The control consisted of a standard assay of MAO-dependent oxidation of 1.0 mM tyramine (TYR) by hScAT homogenates, performed in triplicate for each of the 25 individuals and set at 100%. Mean ± SEM of the number of observations indicated for each agent: 7-methylxanthine: 14–16; Paraxanthine: 8; Theophylline: 20; Caffeine: 8–18; Theobromine: 8–19; IBMX: 7. For theobromine, the (open) symbol at 2.5 mM is different from those of lower doses to denote the extensive influence of 25% v/v DMSO vehicle needed for this xanthine at this dose, which could not be accurately estimated. The difference from control (plotted as 0 on the X-axis) at: * p < 0.05; ** p < 0.01; *** p < 0.001 was determined by one-way ANOVA.