There is significant sequence homology and anatomical co-distribution between cannabinoid (CB1) and sphingosine-1-phosphate (S1P) receptors in the CNS, but potential functional relationships between these lysolipid receptors have not been examined. Therefore, to investigate possible relationships between these two systems at the level of G-protein activation, agonist-stimulated [35S]GTPγS binding and autoradiography were conducted. Autoradiographic studies were first performed to localize receptor-mediated G-protein activation in mouse brain. Coronal brain slices were processed for stimulation of [35S]GTPγS binding using the synthetic cannabinoid agonist WIN 55,212-2 (WIN) or SIP. High levels of WIN- and S1P-stimulated [35S]GTPγS binding were observed in the caudate putamen, hippocampus, substantia nigra, and cerebellum. To further characterize the relationship between S1P-and CB1-mediated G-protein activation, spinal cords from adult male CB1 receptor knockout mice, CNS-deleted S1Pl receptor knockout mice and wild type C57 mice were collected, and assessed using agonist-stimulated [35S]GTPγS binding. Results from this experiment revealed that the S1Pl receptor is predominant in mouse spinal cord. To further investigate potential CBl and SIP receptor interactions spinal cords were collected from adult male ICR mice. Additivity studies were preformed using agonist-stimulated [35S]GTPγs binding. Results showed significantly less than additive stimulation when spinal cord tissue was treated with both WIN and SIP. These results suggest an interaction between the CB1 and S1P receptors in the mouse spinal cord. The effect of cannabinoid antagonists, SR141716A (CB1) and SR144528 (CB2) on S1P-and WIN-stimulated [35S]GTPγS binding were also examined in mouse spinal cord homogenates. These results showed that there was no significant difference between S1P-stimulated [35S]GTPγS binding in the presence of SR141716A or SR144528 compared to vehicle control. This shows that S1P produced stimulation independent of the CBl or CB2receptor. In addition WIN-stimulated [35S]GTPγS binding was not affected by SR144528, but was inhibited by SR141716A, confirming that this action is due to the CB1 receptor. The combined results of this project demonstrate an interaction between CB1 and S1P receptors in certain CNS regions where they are co-distributed, such as the caudate putamen, hippocampus, substantia nigra, cerebellum and spinal cord. These results may be due to convergence on a common pool of G-proteins via dimerization or co-localization in lipid rafts, or a possible direct ligand-receptor interaction.

There is good pharmacological evidence that cannabinoids caused cellular changes by interacting with specific cannabinoid receptors (CBR) (Klein et al., 2000). To date, two CBRs have been identified in the human body, designated Cannabinoid Receptor 1 (CB1) and Cannabinoid Receptor 2 (CB2) (Begg et al., 2005). Endogenously occurring compounds with action at the CBRs also exist and they are called endocannabinoids. One of the four known endocannabinoids is anandamide (AEA). The endocannabinoid system, present in the human body, plays a significant role in altering the physiology of the immune system. Enhancement of this system’s anti-inflammatory effect could possibly present a vital therapeutic target for central and peripheral inflammatory disorders. A number of synthetic CB1 or CB2 specific antagonists have been developed including the highly specific CB1 receptor antagonist/reverse agonist named Rimonabant/ SR141716A. SR compounds are considered unique because these compounds not only inhibit the binding and function of cannabimimetic agents, but also act as inverse agonists. Activation of CB1 receptors produces inappropriate CNS side effects including psychoactivity, dependence and sedation (Clayton et al., 2002) whereas CB1 receptor antagonists/inverse agonists avoid or prevent these side effects. Taking the above information into consideration, Rimonabant has the potential to offer an effective long term treatment of chronic inflammatory disorders without the serious side effects of commonly used treatments. The main aim of this study is to investigate the in vitro effects of Rimonabant alone and in combination with anandamide on inflammatory associated cytokine production by human umbilical vein endothelial cells (HUVEC) and macrophage cultures. After careful consideration of the evidence stating that endothelial cells produce several important molecules vital to the inflammatory response of the body and the confirmation that CB1 receptor mRNA is generally present in endothelial cells, the use of HUVEC was deemed to be satisfactory for this study. The first phase of the study was dedicated to establishing the technique to isolate HUVEC from fresh human umbilical cord within the local laboratory and to maintain these in culture for further use during experimental procedures to test the effects of CB1 ligands. The isolation procedure, trypsinising, freezing away and thawing methods used during this experiment produced healthy HUVEC in sufficient numbers for further use. The next step was to determine the maximum in vitro concentrations at which Rimonabant and anandamide had insignificant cytotoxic effect on selected human cells and in doing so, determine suitable concentrations for further experimentation. Both compounds had a dose related anti-proliferative response when tested on HUVEC. The same dose related response was observed during the Rimonabant exposure to human lymphocytes, but no decrease in lymphocyte viability was observed when treated with anandamide at the concentrations tested. It is evident from the results that there was an almost ten times difference in the IC50 value of the two different products (14.3 μM for Rimonabant and 124.2 μM for anandamide) which was statistically significant. Flow cytometry was used to determine the effects of Rimonabant and anandamide on the surface expression of the CR3 complement receptor by human neutrophils. Neither Rimonabant nor anandamide significantly affect CR3 expression on the surface of freshly isolated human neutrophils and would exclude the CR3 expression pathways as a potential mechanism of action for the anti-inflammatory effects of these compounds. The in vitro effect of Rimonabant and anandamide alone and in combination on the production of cytokines by human macrophages and by HUVEC was determined. Anandamide was shown to inhibit the production of all the detectable cytokines (IL-8, IL-1β and IL-6 in both cell types and IL-10 and TNF-α in macrophages). Furthermore this inhibitory effect was attenuated by pre-treatment Rimonabant. These results would suggest that anandamide could induce anti-inflammatory effects observed in macrophages and HUVEC, through cannabinoid receptors. Rimonabant also inhibited the production of all the detectable cytokines following treatment with 0.5 μM and 3 μM respectively. The anti-inflammatory effects of anandamide were attenuated when combined with 1 μM of Rimonabant. Throughout the various cytokine responses, the dose-response relationship appeared to follow a bell-shaped dose-response. This occurrence proposes that Rimonabant displaces anandamide and blocks the anti-inflammatory effects of the agonist. Flow cytometry was used to determine the effects of Rimonabant and anandamide alone and in combination on the extracellular surface expression of ICAM-1 by HUVEC. Neither Rimonabant nor anandamide had any significant inhibitory effect on the expression of ICAM-1 by HUVEC. Considering the low levels for ICAM-1 expressed by the HUVEC during this experiment and the literature supporting more effective methods of activating the ICAM-1 gene and subsequent up-regulation of ICAM-1 proteins, TNF-α stimulation of HUVEC might produce a different result compared to IL-1β stimulation. The final phase of the project was to determine the effects of Rimonabant and anandamide on the adhesion of human neutrophils to HUVEC. There was no significant difference with relation to the neutrophil adhesion to HUVEC following the treatment with various combination concentrations of the compounds, and also no significant effect following treatment with either test compound individually. Although a specific mechanism of action for Rimonabant could not be uncovered during this study, there is evidence that several possible mechanisms can be excluded. The results support observations made by other researchers and the hypothesis that Rimonabant has anti-inflammatory effects. The results provide motivation for further experimentation to better understand these anti-inflammatory actions of Rimonabant. / Dissertation (MSc)--University of Pretoria, 2012. / Pharmacology / unrestricted