Sites of Action

Cannabinoid receptors are particularly abundant in some areas of the brain. The normal biology and behavior associated with these brain areas are consistent with the behavioral effects produced by cannabinoids (table 2.5 and figure 2.5). The highest receptor density is found in cells of the basal ganglia that project locally and to other brain regions. These cells include the substantia nigra pars reticulata, entopeduncular nucleus, and globus pallidus, regions that are generally involved in coordinating body movements. Patients with Parkinson and Huntington disease tend to have impaired functions in these regions.

CB₁ receptors are also abundant in the putamen, part of the relay system within the basal ganglia that regulates body movements, the cerebellum, which coordinates body movements; the hippocampus, which is involved in learning, memory, and response to stress; and the cerebral cortex, which is concerned with the integration of higher cognitive functions.

CB₁ receptors are found on various parts of neurons, including the axon, cell bodies, terminals, and dendrites.^{57, 165} Dendrites are generally the "receiving" part of a neuron, and receptors on axons or cell bodies generally modulate other signals. Axon terminals are the "sending" part of the neuron.

Cannabinoids like the inhibitory neurotransmitter -aminobutyric acid (GABA) -tend to inhibit neurotransmission, although the results are somewhat variable. In some cases, cannabinoids diminish the effects of the inhibitory neurotransmitter, -aminobutyric acid (GABA),¹⁴⁴ in other cases, cannabinoids can augment the effects of GABA.¹²⁰ The effect of activating a receptor depends on where it is found on the neuron: if cannabinoid receptors are presynaptic (on the "sending" side of the synapse) and inhibit the release Of GABA, cannabinoids would diminish GABA effects; the net effect would be stimulation. However, if cannabinoid receptors are postsynaptic (on the "receiving" side of the synapse) and on the same cell as GABA receptors, they will probably mimic the effects of GABA; in that case, the net effect would be inhibition.^{120, 144, 160}

CB₁ is the predominant brain cannabinoid receptor. CB₂ receptors have not generally been found in the brain, but there is one isolated report suggesting some in mouse cerebellum.¹⁵⁰ CB₂ is found primarily on cells of the immune system. CB₁ receptors are also found in immune cells, but CB₂ is considerably more abundant there (table 2.6) (reviewed by Kaminski⁸⁰ in 1998).

As can be appreciated in the next section, the presence of cannabinoid systems in key brain regions is strongly tied to the functions and pathology associated with those regions. The clinical value of cannabinoid systems is best understood in the context of the biology of these brain regions.

Table 2.5 Brain regions in which cannabinoid receptors are abundant^e

^e Based on reviews by Pertwee 1997 ¹²⁴ and Herkenham 1995⁵⁷ This table will be accompanied by a figure.

Figure 2.5. Location of brain regions in which cannabinoid receptors are abundant.

See table 2.5 for summary of functions associated with those regions.

Table 2.6 Summary table of cannabinoid receptors

.	CB1	CB2
Effects of various cannabinoids
9-THC	Agonist	Weak antagonist
Anandamide	Agonist	Agonist
Cannabinol (CBN)	Weak agonist	Agonist; greater affinity for CB2 than for CB1
Cannabidiol (CBD)	Does not bind to receptor	Does not bind to receptor
Receptor distribution
Areas of greatest abundance	Brain	Immune system, especially B cells and natural killer cells

Cannabinoid Receptors and Brain Functions

Motor effects

Marijuana affects psychomotor performance in humans. The effects depend both on the nature of the task and the experience with marijuana. In general, effects are clearest in steadiness (body sway and hand steadiness) and in motor tasks that require attention. The results of testing Cannabinoids in rodents are much clearer.

Cannabinoids clearly affect movement in rodents, but the effects depend on the dose: low doses stimulate and higher doses inhibit locomotion.^{111, 159} Cannabinoids mainly inhibit the transmission of neural signals, and they inhibit movement through their actions on the basal ganglia and cerebellum, where cannabinoid receptors are particularly abundant (figure 2.6a and 2.6b). Cannabinoid receptors are also found in the neurons that project from the striatum and subthalamic nucleus, which inhibit and stimulate movement, respectively.^{58, 101}

Cannabinoids decrease both the inhibitory and stimulatory inputs to the substantia nigra, and therefore might provide dual regulation of movement at this nucleus. In the substantia nigra, Cannabinoids decrease transmission from both the striatum and the subthalamic nucleus.141 The globus pallidus has been implicated in mediating the cataleptic effects of large doses of Cannabinoids in rats.¹²⁶ (Catalepsy is a condition of diminished responsiveness usually characterized by trancelike states and waxy rigidity of the muscles.) Several other brain regions - the cortex, the cerebellum, and the neural pathway from cortex to striatum - are also involved in the control of movement and contain abundant cannabinoid receptors.^{52, 59, 101} They are, therefore, possible additional sites that might underlie the effects of Cannabinoids on movement.

Figure 2. 6a & b Diagrams showing motor regions of the brain

Figure 2.6. Basal ganglia are a group of three brain regions, or nuclei - caudate, putamen, and globus pallidus. Figure 2.6a is a 3-dimensional view showing the location of those nuclei in the brain. Figure 2.6b shows those structures in a vertical cross-sectional view The major output pathways of the basal ganglia arise from the globus pallidus and pars reticula of the substantia nigra. Their main target is the thalamus.

Memory effects

One of the primary effects of marijuana in humans is disruption of short-term memory.⁶⁸ That is consistent with the abundance of CB₁ receptors in the hippocampus, the brain region most closely associated with memory. The effects of THC resemble a temporary hippocampal lesion.⁶³ Deadwyler and colleagues have demonstrated that cannabinoids decrease neuronal activity in the hippocampus and its inputs ^{23, 24, 83} In vitro, several cannabinoid ligands and endogenous cannabinoids can block the cellular processes associated with memory formation.^{29, 30, 116, 157, 163} Furthermore, cannabinoid agonists inhibit release of several neurotransmitters: acetylcholine from the hippocampus,^49,50,51 norepinephrine from human and guinea pig (but not rat or mouse) hippocampal slices,¹⁴³ and glutamate in cultured hippocampal cells.¹⁴⁴ Cholinergic and noradrenergic neurons project into the hippocampus, but circuits within the hippocampus are glutamatergic.^f Thus, cannabinoids could block transmission both into and within the hippocampus by blocking presynaptic neurotransmitter release.

Pain

After nausea and vomiting, chronic pain was the condition cited most often to the IOM study team as a medical use for marijuana. Recent research presented below has shown intriguing parallels with anecdotal reports of the modulating effects of cannabinoids on pain - both the effects of cannabinoids acting alone and the effects of their interaction with opioids.

Behavioral Studies

Cannabinoids reduce reactivity to acute painful stimuli in laboratory animals. In rodents, cannabinoids reduced the responsiveness to pain induced through various stimuli, including thermal, mechanical, and chemical stimuli.^{12, 19, 46, 72, 96, 154, 174} Cannabinoids were comparable with opiates in potency and efficacy in these expeniments. ^{12, 72}

Cannabinoids are also effective in rodent models of chronic pain. Herzberg and coworkers found that cannabinoids can block allodynia and hyperalgesia

associated with neuropathic pain in rats.¹¹⁷ (Allodynia refers to pain elicited by stimuli that are normally innocuous; hyperalgesia refers to abnormally increased reactivity to pain.) This is an important advance, because chronic pain frequently results in a series of neural changes that increase suffering due to allodynia, hyperalgesia, and spontaneous pain; furthermore' some chronic pain syndromes are not amenable to therapy, even with the most powerful narcotic analgesics.¹⁰

Pain perception is controlled mainly by neurotransmitter systems within the central nervous system, and cannabinoids clearly play a role in the control of pain in those systems.⁴⁵ However, pain-relieving and pain-preventing mechanisms also occur in peripheral tissues, and endogenous cannabinoids appear to play a role in peripheral tissues. Thus, the different cannabinoid receptor subtypes might act synergistically. Experiments in which pain is induced by injecting dilute formalin into a mouse's paw have shown that anandamide and palmitylethanolamide (PEA) can block peripheral pain.^{22, 73 22} Anandamide acts primarily at the CB₁ receptor, whereas PEA has been proposed as a possible CB₂ agonist, in short, there might be a biochemical basis for their independent effects. When injected together, the analgesic effect is stronger than that of either alone. That suggests an important strategy for the development of a new class of analgesic drug: a mixture of CB₁ and CB₂ agonists. Because there are few, if any, CB₂ receptors in the brain, it might be possible to develop drugs that enhance the peripheral analgesic effect while minimizing the psychological effects.

Neural sites of altered responsiveness to painful stimuli

The brain and spinal cord mediate cannabinoid analgesia. A number of brain areas participate in cannabinoid analgesia and support the role of descending pathways (neural pathways that project from the brain to the spinal cord).^{103, 105} Although more work is needed to produce a comprehensive map of the sites of cannabinoid analgesia, it is clear that the effects are limited to particular areas, most of which have an established role in pain.

Specific sites where cannabinoids act to affect pain processing include the periaqueductal gray,¹⁰⁴ the rostral ventral medulla, ^{105, 110} and the thalamic nucleus submedius,¹⁰² the thalamic ventroposterolateral nucleus,¹⁰² dorsal horn of the spinal cord,^{64, 65} and peripheral sensory nerves.^{64, 65, 66, 131} Those nuclei also participate in opiate analgesia. Although similar to opiate analgesia, cannabinoid analgesia is not mediated by opioid receptors; morphine and cannabinoids sometimes act synergistically, and opioid antagonists generally have no effect on cannabinoid induced analgesia.¹⁷¹ However, a kappa-receptor antagonist has been shown to attenuate spinal, but not supraspinal, cannabinoid analgesia.^{153, 170, 171} (Kappa opioid receptors constitute one of the three major types of opioid receptors; the other two types are mu and delta receptors.)

Neurophysiology and neurochemistry of cannabinoid analgesia

Because of the marked effects of cannabinoids on motor function, behavioral studies in animals alone cannot provide sufficient grounds for the conclusion that cannabinoids depress pain perception. Motor behavior is typically used to measure responses to pain, but this behavior is itself affected by cannabinoids. Thus, experimental results include an unmeasured combination of cannabinoid effects on motor and pain systems. The effects on specific neural systems, however, can be measured at the neurophysiological and neurochemical levels. Cannabinoids decrease the response of immediate-early genes (genes that are activated in the early or immediate stage of response to a broad range of cellular stimuli) to noxious stimuli in spinal cord, decrease response of pain neurons in the spinal cord, and decrease the responsiveness of pain neurons in the ventral posterolateral nucleus of the thalamus.^{67, 102} Those changes are mediated by cannabinoid receptors, selective for pain neurons, and unrelated to changes in skin temperature or depth of anesthesia, and they follow the time course of the changes in behavioral responses to painful stimuli, but not the time course of motor changes.⁶⁷ Cannabinoids also modulate the responses of on-cells and off-cells in the rostral ventral medulla in a manner that is very similar to that of morphine.^{55, 110} These cells control pain transmission at the level of the spinal cord.

Endogenous cannabinoids modulate pain

Endogenous cannabinoids can modulate pain sensitivity, through both central and peripheral mechanisms. For example, animal studies have shown that pain sensitivity can be increased when endogenous cannabinoids are blocked from acting at CB₁ receptors ^{22, 62, 110, 130, 158} Administration of cannabinoid antagonists in either the spinal cord ¹³⁰ or paw ²² increase the sensitivity of animals to pain. In addition, there is evidence that cannabinoids also act at the site of injury to reduce peripheral inflammation.¹³¹

Current data suggest the endogenous cannabinoid analgesic system might offer protection against the long-lasting central hyperalgesia and allodynia that sometimes follow skin or nerve injuries ^{130, 158} These results raise the possibility that therapeutic interventions that alter the levels of endogenous cannabinoids might be useful for managing pain in humans.

Chronic Effects of THC

Most substances of abuse produce tolerance, physical dependence, and withdrawal symptoms. Tolerance is the most common response to repetitive use of the same drug (not necessarily a drug of abuse) and is the condition in which, after repeated exposure to a drug, increasing doses are needed to achieve the same effect. Physical dependence develops as a result of tolerance (adaptation) produced by a resetting of homeostatic mechanisms in response to repeated drug use. It is important to reiterate that the phenomena of tolerance, dependence, and withdrawal are not associated uniquely with drugs of abuse. Many medications that are not addicting can produce these types of effects; examples of such medications include clonidine, propranolol, and tricyclic antidepressants. The following sections discuss what is known about the biological mechanisms that underlie on tolerance, reward, and dependence; clinical studies about those topics are discussed in chapter 3.

Tolerance

Chronic administration of cannabinoids to animals results in tolerance to many of the acute effects of THC, including memory disruption,³⁴ decreased locomotion,^{2, 119} hypothermia,^{42, 125} neuroendocrine effects,¹³⁴ and analgesia.⁴ Tolerance also develops to the cardiovascular and psychological effects of THC and marijuana in humans (see also discussion in chapter 3).^{55, 56, 76}

Tolerance to cannabinoids appears to result from both pharmacokinetic (how the drug is absorbed, distributed, metabolized, and excreted) and pharmacodynamic (how the drug interacts with target cells) changes. Chronic treatment with the cannabinoid agonist, CP 55,940, increases the activity of the microsomal cytochrome P450 oxidative system.³¹ Because this is the system through which drugs are metabolized in the liver, this suggests pharrnacokinetic tolerance. Chronic cannabinoid treatments also produce changes in brain cannabinoid receptors and cannabinoid receptor mRNA levels, indicating that pharmacodynamic effects are important, as well.

Most studies have found that brain cannabinoid receptor levels usually decrease after prolonged exposure to agonists,^{42, 119, 136, 138} although some studies have reported increases ¹³⁷ or no changes² in receptor binding in brain. Differences among studies may be due to the particular agonist tested, the assay used, brain region examined, or treatment time. For example, the THC analogue, levonantradol, produces a greater desensitization of adenylyl cyclase inhibition than THC in cultured neuroblastoma cells,⁴⁰ which may be explained by the efficacy differences between these two agonists ^{18, 147} Furthermore, a time course study revealed differences in the rates and magnitudes of receptor down-regulation across brain

regions.¹⁶ These findings suggest that tolerance to different effects of cannabinoids develops at different rates

Chronic treatment with THC also produces variable effects on cannabinoid-mediated signal transduction systems. Chronic THC treatment produces significant desensitization of cannabinoid-activated G-proteins in a number of rat brain regions.¹⁴⁷ Moreover, the time course of this desensitization varies across brain regions.¹⁶

It is difficult to extend the findings of these short-term animal studies to human marijuana use In order to simulate long-term use, the doses used in animal studies are higher than normally achieved by smoking marijuana. For example, the average human will feel "high" after a 0.06 mg/kg injection of THC,¹¹⁸ compared to 10-20 mg/kg/day used in many chronic studies in rats. At the same time, doses of marijuana needed to observe behavioral changes in rats (usually changes in locomotor behavior) are substantially higher than doses at which people feel "high." In addition, pharmacokinetics of THC distribution in the body are dramatically different between rats and humans, as well as being highly dependent on the THC delivery system - that is, whether it is inhaled, injected, or swallowed. Nevertheless, it is likely that some of the same biochemical adaptations to chronic cannabinoid administration occur in both laboratory animals and humans, but the magnitude of the effects in humans may be smaller in proportion to the respective doses used.

Reward and dependence

Experimental animals that are given the opportunity to self-administer cannabinoids generally do not choose to do so, which has led to the conclusion that they are not reinforcing and rewarding.³⁸ However, behaviora^l95 and brain stimulation⁹⁴ studies have shown that THC can be rewarding to animals. The behavioral study used a "place-preference" test, in which an animal is given repeated doses of a drug in one place, and is then given a choice between a place where it did not receive the drug and one where it did; the animals chose the place where they received the THC. These rewarding effects are highly dose-dependent. In all models studied, cannabinoids are only rewarding at mid-range; doses that are too low are not rewarding, doses that are too high can be aversive. Mice will self-administer the cannabinoid agonist, WIN 55,212, but only at low doses.¹⁰⁶ This effect is specifically mediated by CB₁ receptors, and indicates that stimulation of those receptors is rewarding to the mice. Antagonism of cannabinoid receptors is also rewarding in rats; in conditioned place-preference tests, animals show a preference for the place they receive the cannabinoid antagonist, SR141716A, at both low and high doses.¹⁴⁰ Cannabinoids increase dopamine levels in the mesolimbic dopamine system of rats, a pathway associated with reinforcement.^{25, 39, 161} However, the

mechanism by which THC increases dopamine levels appears to be different from that of other abused drugs ^{51 g} (see chapter 3 for further discussion of reinforcement).

Physical dependence on cannabinoids has only been observed under experimental conditions of "precipitated withdrawal", in which animals are first treated chronically with cannabinoids and then given the CB₁ antagonist, SR141716A.^{3, 166} The addition of the antagonist accentuates any withdrawal effect by competing with the agonist at receptor sites; that is, the antagonist helps to clear agonists off and keep them off receptor sites. This suggests that, under normal cannabis use, the long half-life and slow elimination from the body of THC, and the residual bioactivity of its metabolite, 11-OH -THC, may prevent significant abstinence symptoms. The precipitated withdrawal effects produced by SR141716A have some of the characteristics of opiate withdrawal, but are not affected by opioid antagonists and affect motor systems differently. An earlier study with monkeys also suggested that abrupt cessation of chronic THC is associated with withdrawal symptoms,⁸ Monkeys in that study were trained to work for food after which they were given THC on a daily basis; when the investigators stopped administering THC, the animals stopped working for food.

A study in rats indicated that the behavioral cannabinoid withdrawal syndrome correlates with stimulation of central amygdaloid corticotropin-releasing hormone release, consistent with the consequences of withdrawal from other abused drugs.¹³⁵ However, the withdrawal syndrome for cannabinoids and the corresponding increase in corticotropin-releasing hormone are only observed following administration of the CB₁ antagonist, SR 141716A, to cannabinoidtolerant animals;^{3, 166} The implications of data based on precipitated withdrawal in animals for human cannabinoid abuse have not been established.¹⁶⁶ Furthermore, acute administration of THC also produces increases in corticotropin-releasing hormone and adrenocorticotropin release, both of which are stress-related hormones.⁷¹ This set of withdrawal studies may explain the generally aversive effects of cannabinoids in animals, and may indicate that the increase in corticotropin-releasing hormone is merely a rebound effect. Thus, while cannabinoids appear to be conforming to some of the neurobiological effects of other drugs abused by humans, the underlying mechanisms of these actions and their significance in determining the reinforcement and dependence liability of cannabinoids in humans remain undetermined.

^g These increases in dopamine are due to increases in the firing rate of dopamine cells in the ventral tegmental area by ⁹-THC⁴⁷. However, these increases in firing rate in the ventral tegmental area could not be explained by increases in the firing of the A10 dopamine cell group, where other abused drugs have been shown to act⁵¹.

Cannabinoids and the Immune System

The human body protects itself from invaders such as bacteria and viruses through the elaborate and dynamic network of organs and cells referred to as the immune system (see box on Cells of the Immune System).

Cannabinoids, especially THC, can modulate the function of immune cells in various ways - in some cases enhancing, and in others diminishing the immune response ⁸⁵ (summarized in table 2.7). However, the natural function of the cannabinoids in the immune system is not known. Immune cells respond to cannabinoids in a variety of ways, depending upon experimental factors such as drug concentration, timing of drug delivery to leukocytes in relation to antigen stimulation, and the type of cell function analyzed. Although the chronic effects of cannabinoids on the immune system have not been studied, based on acute exposure studies in experimental animals it appears that the concentrations of THC which modulate immunological responses are higher than those required for psychoactivity.

Table 2.7 Effects of Cannabinoids on the Immune System

Drug Tested	Cell Types Tested or Type Drug of Animal Experiment	Drug Concen- tration a	Result	Reference
THC 2-AG 11-OH-THC CBN	Lymphocytes and Splenocytes in vitro	0.1-30 µM	Higher doses suppress T cell proliferation	Luo, 1992; Pross,1992* Klein, 1985%; Specter,1990& Lee, 1995* Herring, 1998
THC 2-AG Anandamide	Lymphocytes and Splenocytes	0.1-25 µM	Lower doses increase T cell proliferation in vitro	Luo,1992; Lee,1995* Pross,1992*
	Splenocytes in vitro	1-25 µM	Little or no effect on T cell proliferation	Lee,1995* Devane,1992
THC, 11-OH-THC AG-2	Splenocytes in vitro	3-30 µM	Decrease B cell proliferation	Klein,1985% Lee,1995*
THC CP 55,940 WIN 55,212-2	Lymphocytes in vitro	0.1-100nM [0.0001-0.1 µM]	Increase B cell proliferation	Derocq, 1995
THC	Mice were injected with drug	>5mg/kg	Antibody production was suppressed	Baczynsky, 1983 Schatz,1993
HU-210		>0.05 mg/kg		Titishov,1989
THC 11-OH-THC CBD CP55,940 CBN	Splenocytes in vitro	1-30µM	Antibody production was suppressed	Klein,1990 Baczynasky,1983 Kaminski,1994 Kaminski,1992 Herring,1998
THC	Rodents were injected with drug	3mg/kg/day for 25days 40mg/kg/day for 2 days	Repeated low doses or a high dose of THC suppress the activity of natural killer cells	Patel,1985 Klein,1987
THC 1l-OH-THC	Natural killer cells in vitro	0.1-32 µM	Doses of >=10 µM suppress natural killer cell cytolytic activity, doses <10 µM produced no effect	Klein,1987 Luo,1989
THC	Peritoneal macrophages and monocytes	3-30 µM	Variable doses of THC suppress macophage functions in vitro	Lopez-Cepero, 1986 Specter,1991 Tang,1992

* cell density dependent;
* mitogen dependent;
% % serum dependent;
& dependent on timing of drug exposure relative to mitogen exposure.

^a Drug concentrations are given in the standard format of molarity (M). A one molar solution is the molecular weight of the compound (in grams) dissolved in 1 liter of water or other solvent. The molecular weight of THC is 314 so a 1 molar solution would be 314 grams of THC dissolved in 1 liter of solution, a 10 µM solution would be 3.14 mg THC/liter.

A 1-10 µM concentration will generally elicit a physiologically relevant response in immune cell cultures. Higher doses are often suspected of not being biologically meaningful, because they are a much larger dose than would ever be achieved in the body. The doses listed in this table are, for the most part, very high. See text for further discussion.

Continue