The possibility that AEA could be a physiological agonist of native vanilloid receptors in mammals was initially opposed due to the low potency observed with this effect as compared to the CB1-mediated pharmacological actions of AEA in heterologous expression systems (Zygmunt et al., 1999; Smart et al., 2000). Yet, there are now numerous studies where AEA actions on native vanilloid receptor are observed at concentrations (< 1 ^M) similar to those required to observe overt CB1-mediated effects. Thus, AEA induces relaxation of rat mesenteric and hepatic arteries and of electrically-stimulated mouse vas deferens, and contracts the guinea pig ileum, with ED50 ranging between 10 and 500 nM (Zygmunt et al., 1999; Ross et al., 2001; Mang et al., 2001). On the other hand, in the case of the vanilloid-receptor-mediated contraction of guinea pig bronchi, or the release of SP and CGRP from DRG slices, high ^M concentrations are necessary to observe a response. Interestingly, the very poor effect of AEA on guinea pig bronchi is not simply due to low affinity for VR1 in this species, because if one uses the guinea pig urinary bladder instead, a much higher potency and efficacy can be again observed for AEA (Harrison et al., 2003). Even when using the same type of assay, i.e., the enhancement of intracellular calcium concentrations, and recombinant VR1 from the same animal species, but different experimental procedures, the reported potencies for AEA range between 0.4 and 5 yM. There are several possible explanations for the varying potency of AEA at vanilloid receptors. First, a heterogeneous population of vanilloid receptors may exist in sensory neurons, as it was hypothesized in the past (Holzer, 1991; Szallasi and Blumberg, 1999). Second, regulatory factors affecting vanilloid receptor sensitivity to ligands have been known for several years (Holzer, 1991; Szallasi and Blumberg, 1999) and might also render AEA more efficacious or potent at VR1. Finally, varying experimental conditions might affect AEA potency and efficacy at recombinant VR1. The only molecular support that has been found to date for the first of these possibilities consists of the finding that TRPV3, a VR1-like receptor which is not directly gated by vanilloids or AEA, regulates the activity of VR1 by forming different types of heterodimers with this receptor, thus resulting in a possible molecular and pharmacological heterogeneity of VR1 (Smith et al., 2002). The other two possibilities explaining the heterogeneity of VR1 responses to AEA have been, instead, recently supported by numerous experimental data. In particular:
• AEA activity at recombinant VR1 receptors is dramatically enhanced by inhibition of FAAH (Ross et al., 2001; De Petrocellis, Bisogno, et al., 2001) and by activation of the AMT by nitric oxide (NO) donors (De Petrocellis, Bisogno, et al., 2001). Furthermore, selective AMT inhibitors block AEA activity at VR1 (De Petrocellis, Bisogno, et al., 2001). These findings were obtained both in isolated cells and tissue preparations expressing VR1 (De Petrocellis, Bisogno, et al., 2001; Andersson et al., 2002). Together with previous data suggesting that the capsaicin-binding domain on its receptor is intracellular (Jung et al., 1999), they suggest that AEA activates this receptor by acting from the cytosolic side of the cell and that the AMT plays a permissive role in the interaction of AEA with VR1. This suggestion was confirmed using both a molecular approach (Jordt and Julius, 2002) and, more recently, by demonstrating that by allowing direct access of AEA into DRG neurons, this compound becomes ten times more efficacious on cation currents in these neurons than when it is applied extracellularly (Evans et al., 2003). More importantly, they indicate that if AEA is produced intracellularly by cells expressing VR1 in local concentrations that might be significantly higher than those measured in whole tissues, it may be able to directly activate this receptor. Finally, the permissive role of the AMT in AEA effects on VR1: (1) might open the possibility that overstimulation of NO release, such as that occurring during inflammation, creates the conditions for extracellular AEA to act on vanilloid receptors, (2) holds true also for some long-chain analogs of capsaicin, but not for capsaicin itself, and (3) might explain why some tissue preparations are more representative than others of the actions of the endocannab-inoid on vanilloid receptors (Andersson et al., 2002).
• AEA activity at recombinant VR1 receptors is significantly enhanced when protein kinase C (PKC) or protein kinase A (PKA) are stimulated (Premkumar and Ahern, 2000; De Petrocellis, Harrison, et al., 2001; Vellani et al., 2001). Activation of these two enzymes by phorbol esters and cAMP, respectively, also greatly enhances those vanilloid-receptor-mediated actions of AEA that are otherwise observed only with high micromolar concentrations of the compound, i.e., guinea pig bronchoconstriction and SP release from DRG slices (De Petrocellis, Harrison, et al., 2001). Interestingly, PKC, by promoting their internalization, also downregulates CBj receptors (Garcia et al., 1998). On the other hand, activation of PKA enhances the biosynthesis of AEA phospholipid precursor, the N-arachi-donoyl-phosphatidyl-ethanolamine (Cadas et al., 1996; Di Marzo et al., 1994). These findings indicate, therefore, that conditions and stimuli leading to PKC and PKA activation, such as those occurring during neurodegeneration and inflammation, might, together with NO release (see the preceding text), "redirect" AEA from CE^ to VR1 receptors
• Palmitoylethanolamide, an AEA congener coreleased with AEA by cells and proposed to act as an endogenous enhancer of AEA activity (the "entourage" effect; Lambert and Di Marzo, 1999), significantly potentiates AEA activity at recombinant human VR1 (De Petrocellis, Davis, et al., 2001) when coadministered at a molar ratio with AEA similar to that found to occur in mammalian tissues and cells. Other saturated or monounsaturated AEA congeners, and particularly lauroylethanolamide, exert a similar action (Smart et al., 2002), and this effect has been recently the subject of a detailed structure-activity relationship study (Vandevoorde et al., 2003). Palmitoylethanolamide was found to enhance also the activity on human VR1 of low concentrations of capsaicin and, particularly, RTX (De Petrocellis, Davis, et al., 2001), and to potentiate the antiproliferative effects of these two compounds on human breast cancer cells (De Petrocellis et al., 2002). It is possible that palmitoylethanolamide acts as an allosteric activator of VR1 because it produces a leftward shift in the dose-dependent displacement by AEA of [ !I I |RT\ binding to membranes containing human VR1 (De Petrocellis, Davis, et al., 2001).
• AEA bronchoconstricting activity in guinea pigs was found to be inhibited by lipoxygenase blockers, thus suggesting that these enzymes act on intracellular AEA to catalyze the formation of AEA products capable of activating vanilloid receptors, and that co-expression of lipoxygenases is necessary to observe stronger VR1-mediated effects with exogenous AEA (Craib et al., 2001). It has also been suggested that, in the same preparation, AEA might activate phospholipase A2, AA release, and the formation of hydroperoxy-derivatives of A A (Kagaya etal., 2002), which are known to activate VR1 receptors (Hwang etal.,
2000). The possibility that AEA lipoxygenase products mediate AEA activity at VR1 has been addressed by testing three different hydroperoxy-ethanolamides on Ca2+ influx into intact HEK cells overexpressing the human VR1 receptor (De Petrocellis, Bisogno, et al.,
2001). None of the compounds was a potent or efficacious VR1 agonist under these conditions, although it is possible that exogenous hydroperoxy-ethanolamides are not recognized by the AMT and hence cannot easily interact with VR1 in intact cells. It remains therefore to be investigated if indeed lipoxygenase plays a role in regulating AEA activity at VR1 receptors. Interestingly, a recent study demonstrated that AEA, following its hydrolysis to AA and the formation of 5,6-epoxyeicosatrienoic acid, activates TRPV4 channels, which are related to VR1 receptors but are not gated by capsaicin or AEA directly (Watanabe etal., 2003).
• Acidic pH (5.5) also enhances AEA potency and efficacy at both recombinant and native vanilloid receptors (Olah, Karai, et al., 2001), whereas an increase in the assay temperature from 22 to between 37 and 50° C enhances AEA efficacy but reduces its potency at recombinant VR1 (Sprague etal., 2001). Therefore, it can be speculated that AEA might act as an endo vanilloid during pathological states, such as inflammation and fever, which enhance temperature and increase the pH.
• Bovine serum albumin (BSA), used by most researchers in assay buffers to prevent AEA from sticking to the plastic ware, dramatically reduces AEA capability to activate recombinant VR1 in intact cells (De Petrocellis, Davis, et al., 2001). This phenomenon is possibly due to prevention of AEA cellular uptake, which is necessary to the compound to interact with the intracellular site on vanilloid receptors (see preceding text), and it explains why AEA potency at recombinant human VR1 is five- to tenfold higher in the absence (DePetrocellis etal., 2000; De Petrocellis, Bisogno, etal., 2001) than in the presence (Zygmunt et al., 1999; Smart et al., 2000; Ross et al., 2001; Olah, Karai, et al., 2001) of BSA.
These findings indicate that there are at least four separate ways of regulating AEA activity at VR1 receptors: (1) by regulating the activity of the AMT, (2) by PK-catalyzed phosphorylation of
VR1 and its subsequent sensitization (or inhibition of desensitization; see paper by Rathee et al., 2002), (3) by low pH, and (4) by coproduction of AEA with its congeners. Separately or together, these events may make of AEA a physiological agonist of vanilloid receptors. It is possible that, for example during inflammation or cell damage, when PKC is activated, pH is decreased and palmitoylethanolamide is biosynthesized by cells (Bisogno et al., 1997; Hansen et al., 1995), AEA becomes more active at vanilloid receptors than at CB, or CB2 receptors, thus possibly producing different effects on pain perception, inflammation, and cell survival, depending on whether it only activates or also immediately desensitizes VR1. In addition, it has been shown that the levels of AEA can be enhanced following stimulation of VR1, possibly as a consequence of the intracellular calcium rise induced by this receptor (Di Marzo, Lastres-Becker, et al., 2001; Ahluwalia, Urban, etal., 2003). Hence, AEA produced during VRl-mediated inflammatory hyperalgesia may feedback on nociception either negatively or positively by activating CB, or VR1 receptors, respectively.
Was this article helpful?