Cannabinoids promote human glioma cell apoptosis

Mathieu Widmer, Oliver Hanemann, John Zajicek. Peninsula Medical School, Plymouth, United Kingdom.

Cannabinoids, either plant-derived, chemically synthesised or endogenously produced, have been shown to bind to two specific G protein-coupled receptors, CB1 and CB2, expressed by neurons and cells of the immune system, respectively (Howlett et al., 2002). Intriguingly, expression of cannabinoid receptors in the central nervous system is not confined to neurons, as other cells such as astrocytes also express detectable amounts of CB1 and/or CB2 (Bouaboula et al., 1995). Our knowledge of the function of cannabinoid receptors in these cells remains poor, prompting us and other groups to investigate it further. Moreover, cannabinoid receptors are expressed in tumours of the central nervous system (glioma) where their activation might promote cell death, as suggested by studies performed in rodents (Galve-Roperh et al., 2000). Whether this also applies to humans is controversial (Held-Feindt et al., 2006); we therefore wanted to assess the effect of cannabinoids on human glioma cell viability using the human astrocytoma cell line U373MG.

Our results show that U373MG human glioma cells were sensitive to cannabinoid receptor agonists (THC or HU-210), but very high concentrations (> 5 μM) of the compounds were required to promote cell death, as measured using an MTS assay (Promega) that measures cell metabolism and therefore provides an accurate estimation of cell viability. In addition, the presence of serum (10 %) almost totally abolished cannabinoid-induced cell death. Similar concentrations of the compounds promoted a rapid activation of extracellular-regulated kinase (Erk) and the c-Jun NH2-Terminal Kinase (JNK), thereby confirming that cannabinoid receptors are functional in these cells. Cell death could be prevented by 1-(2,4-dichlorophenyl)-5-(4-iodophenyl]])-4-methyl-N-(1-piperidyl)pyrazole-3-carboxamide (AM251) (5 μM), a specific antagonist of the CB1 cannabinoid receptor, suggesting that cannabinoid-induced glioma cell death is a CB1-dependent process.

As a conclusion, human U373MG glioma cells are only sensitive to very high, pharmacologically irrelevant, doses of cannabinoid receptor agonists. Subsequent experiments should be aimed at studying whether signal transduction pathways known to be hyperactive and/or constitutively active in human glioma prevent therapeutically relevant doses of cannabinoids from being efficacious. Moreover, as gliomas are known to be genetically unstable and heterogeneous, it is tempting to speculate that some of them might be sensitive to cannabinoids and others resistant. In addition, soluble proteins present in the vicinity of tumours in vivo may also prevent cannabinoid-induced cell death, as suggested by the inhibitory effect of serum in cannabinoid-induced U373MG cell death.

Bouaboula, M., et al., J Biol Chem, 1995. 270 (23): p. 13973-13980

Held-Feindt, J., et al., J Neurochem, 2006. 98 (3): p. 886-893

Howlett, A.C., et al., Pharmacol Rev, 2002. 54(2): p. 161-202.