Neurons on cannabinoids: dead or alive?

Preparations from Cannabis sativa (marijuana) have been used for many centuries both medicinally and recreationally. However, the chemical structure of their active components – the cannabinoids – was not elucidated until the early 1960s. Among the E60 cannabinoids produced by marijuana, D-tetrahydrocannabinol (THC) is the most relevant one owing to its high potency and abundance (Gaoni & Mechoulam, 1964). Since the early 1990s, it is widely accepted that THC acts in the organism by mimicking endogenous substances – the so-called endocannabinoids (Devane et al., 1992) – that bind to and activate specific cell surface cannabinoid receptors, two of which have been cloned and characterized so far from mammalian tissues: CB1 (Matsuda et al., 1990) and CB2 (Munro et al., 1993). These recent findings have led to a remarkable expansion of basic cannabinoid research as well as to a renaissance in the study of the therapeutic value of cannabinoids. The CB1 cannabinoid receptor is highly abundant in discrete areas of the brain, and its activation lowers the release of neurotransmitters such as dopamine and GABA, thereby affecting processes such as movement and memory (Piomelli et al., 2000). Besides these well-established neuromodulatory events, cannabinoids may control the survival/death decision of neurons (Guzmán et al., 2002). Thus, cannabinoids have been shown to protect neurons from toxic insults such as excitotoxicity, traumatic injury and ischaemia both in vitro and in vivo (Mechoulam et al., 2002; van der Stelt et al., 2002). This neuroprotective action may rely on different mechanisms, including (i) inhibition of glutamatergic neurotransmission, (ii) antioxidant capacity, (iii) modulation of glial cell function and (iv) control of the microvasculature, and is supported by the observation that the brain overproduces endocannabinoids upon damage (Mechoulam et al., 2002; van der Stelt et al., 2002). In contrast, a few studies have shown that cannabinoids induce apoptosis of cultured neurons (Chan et al., 1998), but the mechanism of this neurotoxic action is largely unknown. In this issue, Downer et al. (2003) show that THC-induced apoptosis of primary rat cortical neurons relies on the activation of the c-Jun N-terminal kinase (JNK) cascade. The JNK protein kinases participate in the blockade of the cell cycle and the induction of cell death triggered by stress signals such as irradiation, heat shock, osmotic shock and proinflammatory cytokines. As selective JNK inhibitors are not available, Downer et al. (2003) used selective antisense oligonucleotides to target JNK mRNAs and therefore deplete neurons of JNKs. The results of these elegant experiments were rather clear-cut, except perhaps for the relative role of the different JNK isoforms in the apoptotic process. Cannabinoids may therefore lead to opposite effects on neuron survival/death (Figure 1). It is conceivable that different experimental factors account for this ‘ying-yang’ action, for example: (i) cannabinoid neuroprotection is usually more evident in whole-animal than in cultured-neuron models, which may result from their aforementioned impact on various brain cell types (neurons, glia, vascular endothelium); (ii) cannabinoids may exert dual effects on neural cell fate depending on signal input (e.g. agonist dose and time of exposure), high inputs usually exerting growth inhibition or death; (iii) endocannabinoids and exogenous cannabinoids may display distinct pharmacological behaviour (e.g. agonistic potency and stability); and (iv) the origin of the neural cell and its stage of differentiation may affect sensitivity to death. Modulation of neural cell fate by cannabinoids is therefore a complex and still obscure issue, and although the bulk of the experimental evidence supports that these compounds protect the brain from damage, the paper by Downer et al. (2003) adds a cautionary note to the field.