Cannabis sativa is one of the most widely used psychoactives and has a documented history of use going back thousands of years; however, the mechanisms of its actions are only just being elucidated. Until relatively recently, the intoxicating effect of cannabis was thought to act in a way similar to ethanol. The active principle, Δ⁹-tetrahydrocannabinol (THC), a highly lipophilic molecule, was thought to insert itself into the lipid cell membrane of nerve cells. However, it is now known that a specific receptor in the brain selectively binds this ligand. The characteristic effects of cannabis intoxication are thus generated by intracellular changes and altered signalling of the neurons.

Different subtypes of this receptor are known to be present in the body. When these receptors were first discovered, there were no naturally-occurring molecules in the body that were known to bind them. Early fringe speculation suggested that the receptor system might have co-evolved with the ancient use of cannabis, but its natural function is not to mediate the effects of the most widely distributed and used drug of plant origin, but to interact with naturally occurring, or endogenous, cannabinoids. These cannabinoids, their receptors, and their possible roles in the normal functioning of the body are the focus of intensive research. Present evidence suggests that the endocannabinoids and their receptors constitute a widespread modulatory system that fine tunes bodily responses to a number of stimuli.

This short review article outlines what is currently known about this system from experiments undertaken by scientists in a range of fields. The purpose of this article is not to provide a comprehensive review of all research and knowledge in the field of endocannabinoid research, but to give an overview of the system as it is currently known and to highlight several interesting areas. First, the cannabinoid receptors shall be discussed, followed by the molecules thought to selectively bind them (their ligands) under normal physiological conditions. The final section of this review focuses on some of the possible functions this recently discovered system could perform and the individual roles that the endocannabinoids and their receptors could play. An outline of the optimistic outlook for cannabinoid therapies is then given.



The first cannabinoid receptor to be discovered was characterized and cloned in 1990 from the mammalian brain1. Its structure and function resembles that of other known hormone receptors2. As of May 2003, two subtypes of the cannabinoid receptor, CB1 and CB2, have been distinguished and are expressed both in the nervous system and peripheral tissues and organs. Both subtypes belong to the seven transmembrane spanning receptor family with seven a-helices spanning the cell membrane. The intracellular loops of the receptor protein are involved with G-proteins responsible for the transduction of the intercellular signal. This G-protein-coupled receptor causes the inhibition of the enzymatic activity of adenylate cyclase responsible for the production of cyclic adenosine monophosphate (cAMP) in the cell. A large number of hormones act through G-protein-coupled receptors and so cAMP has been termed a 'second messenger' because it transmits signals originating at the surface of cells from a variety of 'first messengers' to the interior of cells.

The CB1 receptor is present in both the nervous system and other tissues and organs of the body. By using the imaging technique called quantitative radiography, researchers have determined that this receptor is responsible for the psychotropic actions of THC and other cannabinoids3. The primary regions where cannabinoids bind in the human brain are the basal ganglia, which control unconscious muscle movements, and the limbic system, including the hippocampus, which is involved in integrating memory. It is this last distribution that points to the reason why the most consistent effect of THC on performance is the disruption of selective aspects of short-term memory tasks, similar to patients with damage to the limbic cortical areas4.

The CB1 receptor is also present in the cerebellum, throughout the cerebral cortex and also in many parts of the body including both the male and female reproductive systems. The scarcity of receptors in the medulla oblongata, responsible for controlling respiratory and cardiovascular functions, explains the virtual absence of reports of fatal cannabis overdose in humans5.

Three years after the discovery of CB1, a second human cannabinoid receptor, CB2, was identified in the marginal zone of the spleen6. The CB2 receptor is homologous to the CB1 receptor, sharing an overall 44% homology with CB17. It is confined to the immune system with its greatest density in the region where it was first discovered8. It is this form of the receptor that is expressed on T-cells of the immune system9 but is not expressed in the central nervous system (CNS) or, like the CB1 receptor, in the liver, lungs or kidneys.

The existence of two homologous receptor subtypes, with moderate to low sequence identity, allowed for the development of both agonists and antagonists selective for either type. THC is known to act as a weak, but functional, agonist of the CB2 receptor10. Exciting research is being undertaken into the possibility of developing therapeutically useful compounds that selectively bind the CB2 receptor. These compounds could perform their beneficial function without their potentially unwanted, psychotropic side effects.

Although no further subtypes have been discovered, it is possible that other cannabinoid receptors may exist. Advances in molecular biology, including the possibility of in silico screening of complete gene libraries, may uncover CBn (that is, neither CB1, nor CB2) receptors with low amino acid sequence homology to the cloned receptors. Indirect evidence also supports the existence of as yet undiscovered receptors both in the periphery and the brain. It has been shown that certain compounds exert typical cannabimimetic actions, such as the down-regulation of mast cells, but this cannot be reproduced in cells transfected with either the CB1 or CB2 receptors11.

Although there has been no progress in finding CBn receptors, a functionally active short isoform has been characterized called CB1A12. The distribution of mRNA for both the CB1 and CB1A receptor has been found throughout the brain and in all peripheral tissues examined. The putative CB1A receptor is present in amounts of up to 20% that of CB1 and has been shown to exhibit all the known properties of CB1 to a slightly attenuated extent13.



We have seen that receptors for cannabinoids exist in the body. The presence of these receptors that selectively bind THC and other cannabinoids could only be explained by the presence of endogenous ligands that can bind them. Otherwise, it would indeed be strange that receptors exist in the body, having as their only function the binding of molecules of plant origin. Researchers thus looked for molecules in the body that utilized these orphan receptors and thereby discovered their natural functions.

In 1992, Devane et al., identified the first putative endocannabinoid from porcine brain14. This ligand was later called anandamide, which is derived from the Sanskrit word for bliss (ananda) due to its possible cannabimimetic, psychotropic properties. Anandamide, or N-arachidonylethanolamine, is a modified form of arachidonic acid. It is a polyunsaturated fatty acid that serves as a common precursor for many biologically active metabolites. Although the structure of anandamide is quite different from THC, experiments have shown that it binds to cannabinoid receptors. It has also been shown to share with THC, and other cannabinoids, most of the pharmacological properties exerted both in the CNS and peripheral system. These include the basic characteristic actions in behavioral tests on rodents15. Cross-tolerance to THC also substantiates the evidence that anandamide works through the same mechanism as THC and, like THC, anandamide also increases both the affinity and number of rat cerebellum and hippocampal receptors after chronic and acute exposure16.

Because anandamide, like THC, behaves as a weak agonist at CB2 receptors, the question arose whether there may be other endogenous cannabinoids more selective for the CB2 receptor and produced in the peripheral tissues. Investigations led to the discovery of 2-arachidonoyl-glycerol from the canine gut17. This derivative of arachidonic acid was shown to bind to both CB1 and CB2 receptors.

This putative endocannabinoid caused the typical behavioral reactions in mice, affected levels of cAMP17 and had similar effects to some actions of THC in the periphery18. It has also been shown to be present in the brain of rats, at levels higher than those of anandamide19 and also in dog spleen and pancreas20.

Palmitoyl-ethanolamide, or N-(2-Hydroxyethyl)hexadecamide, is an N-acyl-ethanolamide. It is co-synthesized with anandamide in all tissues so far examined and possibly acts as an endogenous CB2 ligand. Its proposed role is that of an autocoid, or 'local hormone', capable of negatively regulating mast cell activation and inflammation [21]. It has also been reported that palmitoyl-ethanolamide can down-regulate IgE-triggered activation of cultured mast cells through the CB2 receptor present on these cells21.

Researchers looking for further endocannabinoids reasoned that other classes of chemical mediators originating from the precursor arachidonic acid, such as prostaglandins and leukotrienes, do not exist as single entities but as large families of chemically-related substances. They therefore expected that anandamide was only the first identified representative of a class of unsaturated fatty acid-derived ethanolamides that bind to the cannabinoid receptor22. Within a short period of anandamide being identified, two analogues of anandamide -- docosatetraenylethanolamide (DTEA) and homo-g-linoenylethanolamide (HLEA) were also isolated and identified. They were found to exert similar effects to both anandamide and THC in behavioral tests on rodents and also inhibited the action of adenylate cyclase through G-proteins, the action of which could be blocked by the highly specific CB1 antagonist SR 141716A 23, 24. It was therefore proposed that these substances might function as endogenous agonists at the neuronal CB1 receptor.

Another putative endogenous cannabinoid, oleamide, or cis-9-octadecenoamide, has also been isolated and shown to have similar actions to anandamide in the behavioral rodent tests. This molecule is a long-chain fatty acid derivative that was first isolated from the cerebrospinal fluid of cats and humans deprived of sleep. This extract had a sleep-inducing action in mammals25, which has often been suggested for anandamide and THC because of their sedative and motor inhibitory properties.

The cannabimimetic actions of oleamide, however, cannot have been mediated though any of the known cannabinoid receptor types. Oleamide can only bind CB1 or CB2 receptors at very high concentrations never present under physiological conditions26. [This statement on oleamide binding has been disputed, see Comments.] An indirect way that oleamide could exert its cannabimimetic action could be through the competitive inhibition of the enzyme responsible for the degradation of anandamide27. This action would thus raise the concentration of the latter cannabinoid, causing its actions to be recorded. Other long-chain fatty acid ethanolamides, co-synthesized with anandamide in neurons, are also thought to have a similar function28.



Although the distribution of receptors in the body is becoming clearer and their putative ligands becoming more fully characterized, the correlation between pathophysiological responses and the production and activation of these ligands is by no means certain. Nevertheless, from the existing data, it is possible to suggest a widespread modulatory role for the cannabinoid system, responsible for regulating a number of tasks. This system is not limited to the central nervous system but is also concerned with peripheral processes and could act to modulate neurotransmitter release and action from autonomic and sensory nerve fibers. Functions within the control of immunological, gastrointestinal, reproductive and cardiovascular performance are also indicated.

It has been shown that, in the brain, the CB1 receptor is one of the most abundant G-protein coupled receptors present29. Activation of these CB1 receptors suppresses the release of a number of nerotransmitters including acetylcholine, noradrenaline, dopamine, serotonin, GABA, glutamate and aspartate30, 31, 32, 33 and cannabimimetic drugs are known to produce a number of behavioral effects including the impairment of memory34, 35, 36. This could be due to CB1 receptors modulating cAMP-dependent synaptic plasticity and thereby preventing the recruitment of new synapses by inhibiting the formation of cAMP37. Due to both functional and anatomical evidence suggesting that CB1 receptors are present pre-synaptically30, 38, 39, cannabinoids may therefore act at this site to inhibit new synapse formation. This is further suggested by the observation that hippocampal presynaptic boutons assemble before the postsynaptic assembly40. Synaptic plasticity is an important property involved in a number of processes and the possibility therefore exists that endocannabinoids act to modulate changes in neuronal communication associated with brain development, learning, and also pain41.

It has recently been shown that the endogenous cannabinoid system has a central function in the extinction of aversive memories42. Aversive memories are important for the survival of an organism. These memories are kept by reinforcement but if reinforcement does not occur, the resulting behavioral response to the noxious stimuli will diminish until it no longer exists. This extinction process is also important but its mechanism is not fully known. Endocannabinoids acting through the CB1 receptor in the amygdala of the limbic system (which is known to be involved in this process43) are now thought to facilitate the memory loss through an inhibitory effect on local inhibitory networks (possibly GABA-using neurons).

The actions of endocannabinoids may be mediated by cannabinoid receptors located both pre- and post- synaptically. The activation of pre-synaptic receptors could lead to such intracellular changes that modulate the release and/or actions of other neurotransmitters, such as dopamine, acetylcholine and glutamate44, 45, 46, 47 and thereby have even further-reaching effects. In such a way, THC has been found to facilitate the release of dynorphins (endogenous opiate-like molecules), which act at opioid receptors. This action may have a role to play in the pain-reducing, or analgesic, properties of both THC and anandamide.

Pain is initiated when a variety of physical stimuli activate specific pain receptors. The endogenous cannabinoid, anandamide, can inhibit the stimulation of one such pain receptor, the vanilloid receptor (VR1), which results in an analgesic effect. Anandamide and structurally-related lipids may also act as vanilloid receptor modulators in the regulation of various afferent stimuli such as pain reception and visceral reflexes and also efferent actions such as vasodilation and inflammation arising from the nervous signals. However, this research is currently in the preliminary stages and the natural occurrence in vivo has yet to be determined48.

Recent research has tentatively shown that THC does not affect the VR1 receptor. In other studies, when the CB1 receptor of mice was genetically eliminated, the CB1 knockout mice did not exhibit significant alterations of pain indicators49. These results, however, appear to contradict other studies that demonstrate anti-nociceptive activity produced by marijuana or THC. One possibility that may explain these apparently contradictive data may lie in the fact that THC has a high affinity for the CB1 receptor. Exogenously applied THC, such as when a subject smokes marijuana, may compete with other agonists of the CB1 receptor thus competing with anandamide for binding to the CB1 receptor. This would free endogenous anandamide and increase the concentration available to bind to the VR1 receptor and therefore provide the reported pain relief. Some anecdotal evidence suggests that users of medical marijuana become insensitive to the euphoric effects of marijuana after sustained use while still benefiting from its pain relieving properties. The mechanism proposed above may underlie this action, although the question will have to await further research before being fully clarified.

A large amount of anecdotal evidence and several published scientific reports describe numerous effects of cannabinoids on visual perception. This includes altered thresholds of light detection and recovery from glare. The possible positions within the brain and/or retina of the eye responsible for these changes in perception are, as yet, unknown, although research has found that CB1 receptors are found in the retina of many vertebrate species50. This report also presents strong evidence for an endogenous cannabinoid signalling system in the vertebrate retina utilizing 2-arachidonoyl-glycerol and palmitoyl-ethanolamide which may act pre-synaptically to regulate the release of the neurotransmitter glutamate across synapses.

A neuroprotective role may also exist for the acyl-ethanolamides in general and palmitoyl-ethanolamide in particular, due to their production at the sites of neuronal damage and cell death51, 52, 53, 54, 55. It is also becoming clear that CB1 receptors are present in the hypothalamus and may be responsible for the fine-tuning of pituitary hormone secretion56, 57, 58. Injection of anandamide into the ventricles of the brain led to the release of the hypothalamic hormone, corticotrophin-releasing factor-4156. This hormone ultimately leads to the production of corticosterone, a regulator of carbohydrate and protein metabolism, from the adrenal gland. Anandamide working at the hypothalamus may also inhibit the release of other hormones, such as prolactin and the luteinising, follicle stimulating and growth hormones57, 58.

In addition to modulating the release of neurotransmitters and hormones, it is becoming increasingly clear that the endocannabinoid system is intimately linked to other processes in the periphery. A system may exist where endocannabinoids mediate chemical communication between different types of immune cells and between sensory fibers and blood cells. They have also been found to play an important role in acute inflammatory reactions. The standard picture of inflammatory reactions is that binding of an allergen to IgE receptors on immune cells leads to the activation of basophil and mast cells. These cells then release histamine, serotonin and leukotrienes. Within this mixture of inflammatory mediators, palmitoyl-ethanolamide and anandamide have also been discovered59. Palmitoyl-ethanolamide is thought to act as an autocoid on the same, or neighboring, basophilic or mast cells and thereby inhibits the further release of mediators60, thereby keeping the inflammatory reaction in check.

Anandamide from basophils might also increase the production of prostaglandin E2 from macrophages, which suppresses the activity and proliferation of both lymphocytes and macrophages. Anandamide could also directly inhibit the recruitment of lymphocytes during the late phase of the allergic response and induce their cell death61. It would thus appear that both palmitoyl-ethanolamide and anandamide could help to prevent the excessive propagation of the inflammatory response. This would reduce the risk of subsequent hypersensitivity to the initial stimulus and prevent the development of allergic disease54, 62. Further research is needed to determine which receptor types are expressed in the different sub-populations of each immune cell-type. It is, at present, unclear which of the immunological actions of the endocannabinoids are mediated by which cannabinoid receptor. Research directed into giving a clearer picture of receptor expression would certainly help clarify their immunomodulatory role.

There are a number of other ideas for possible roles for the endocannabinoid system based on the expression of the ligands, and/or their receptors in the body. These include the very interesting observation that tissues of the reproductive system also contain receptors and are able to synthesize and degrade endocannabinoids.

It is conceivable that endocannabinoids in the reproductive system act as local hormones and evidence exists for an anandaminergic system in the rat testes and mouse vas deferens that controls spermatogenesis and male fertility63, 64, 65. THC and anandamide are also both thought to inhibit the acrosome reaction through cannabinoid receptors on the sperm cell membrane66, 67, 68. These receptors have been found on the sperm cell of the sea urchin, and the ovaries from the same species are known to synthesize and degrade both anandamide and palmitoyl-ethanolamide69. It is therefore conceivable that the sea urchin synthesizes anandamide during the acrosome reaction in order to prevent fertilization by more than one sperm. It is not yet known whether an analogous system also occurs in mammals although some evidence does point towards an increased infertility among chronic cannabis users.

Anandamide may also play another interesting role in the female reproductive system. CB1 and CB2 receptors are present in the embryos of mice from the very early stages of their development and also in the adult uterus70. Due to the inhibitory effect of anandamide on embryonic cell division, anandamide might act as a negative signal for embryonic development and implantation71. High levels of the synthesizing enzyme, and low levels of the degrading enzyme exist at the time when the uterus is the least receptive for embryo implantation. The uterus may therefore utilize anandamide in order to direct both the location and timing of embryo implantation.



In just over one decade, the abundance of quality research has changed our basic views of the mechanism of cannabis intoxication. It has also unveiled a new and extensive regulatory system within the body. Further multidisciplinary research must be undertaken to improve our understanding of these functions and provide more data on the expression and inactivation of the components of this system. It will then be possible to exploit this knowledge in order to make therapeutic compounds for the treatment of symptoms, and possible prevention, of a number of disorders.

Such therapies could act through the agonistic/antagonistic properties of the novel compounds acting at cannabinoid receptors, or by targeting the synthesizing, or degrading, enzymes responsible for endocannabinoids. As cannabinoids are effective at countering muscle spasms, this property could be exploited to provide relief for sufferers of multiple sclerosis and patients who suffer from chronic tremors, or other involuntary movements. Ongoing research is presently determining whether cannabinoid ligands are effective agents in the treatment of chronic pain, glaucoma, spasms, and the wasting and emesis associated with AIDS and cancer chemotherapy72, 73. This latter property is currently being exploited and a cannabinoid called Nabilone is on the market, indicated for the suppression of nausea and vomiting during cytotoxic chemotherapy. The potential therapeutic application of cannabinoids is, however, controversial and constitutes a widely debated issue with relevance in both scientific and social circles.

One of the most interesting potential therapeutic actions of cannabinoids reported to date is the ability of cannabinoids to inhibit the growth of cancerous, or transformed, cells in culture. Anandamide can inhibit breast cancer cell proliferation74 and THC can cause the programmed cell death, or apoptosis, of transformed neural cells in vitro75. In vivo research has also begun to elucidate the biochemical mechanisms involved in the anti-tumoral actions of CB1 agonists, including THC76. These experiments have shown that it is possible to completely eradicate malignant brain tumors in rats by THC administration.

Cannabinoids have also been found to protect neurons in culture from glutamate-induced excitotoxicity77, 78 and from ischaemic death (lack of oxygen)79. These ligands are currently under test as therapeutic agents in the treatment of neurodegenerative diseases such as multiple sclerosis and Parkinson's Disease. Research is also being directed into the possibility of using cannabinoids as drugs that could stop the growth and spread of cancer cells, based on the research mentioned above.

A prominent researcher in the field described the discovery of anandamide as a 'new dawn for cannabinoid pharmacology'7. Although a lot of work has been conducted, we can expect far more research in the near future that could revolutionize the way we view our bodies and the treatments we use to prevent their malfunction.

---