Platelets. Platelets differ from other formed elements of blood in expressing mechanisms for uptake, storage, and endocytotic release of 5-HT. 5-HT is not synthesized in platelets, but is taken up from the circulation and stored in secretory granules by active transport, similar to the uptake and storage of serotonin by serotonergic nerve terminals. Thus, Na+-dependent transport across the surface membrane of platelets, via the 5-HT transporter, is followed by VMAT2-mediated uptake into dense core granules creating a gradient of 5-HT as high as 1000:1 with an internal concentration of 0.6 M in the dense core storage vesicles. Measuring the rate of Na+-dependent 5-HT uptake by platelets provides a sensitive assay for 5-HT uptake inhibitors.
Main functions of platelets include adhesion, aggregation, and thrombus formation to plug holes in the endothelium; conversely, the functional integrity of the endothelium is critical for platelet action. A complex local interplay of multiple factors, including 5-HT, regulates thrombosis and hemostasis (Chapters 30 and 33). When platelets make contact with injured endothelium, they release substances that promote platelet aggregation, and secondarily, they release 5-HT (Figure 13–4). 5-HT binds to platelet 5-HT2A receptors and elicits a weak aggregation response that is markedly augmented by the presence of collagen. If the damaged blood vessel is injured to a depth where vascular smooth muscle is exposed, 5-HT exerts a direct vasoconstrictor effect, thereby contributing to hemostasis, which is enhanced by locally released autocoids (thromboxane A2, kinins, and vasoactive peptides). Conversely, 5-HT may interact with endothelial cells to stimulate production of NO and antagonize its own vasoconstrictor action, as well as the vasoconstriction produced by other locally released agents.
Cardiovascular System. The classical response of blood vessels to 5-HT is contraction, particularly in the splanchnic, renal, pulmonary, and cerebral vasculatures. 5-HT also induces a variety of responses by the heart that are the result of activation of multiple 5-HT receptor subtypes, stimulation or inhibition of autonomic nerve activity, or dominance of reflex responses to 5-HT (Kaumann and Levy, 2006). Thus, 5-HT has positive inotropic and chronotropic actions on the heart that may be blunted by simultaneous stimulation of afferent nerves from baroreceptors and chemoreceptors. Activation of 5-HT3 receptors on vagus nerve endings elicits the Bezold-Jarisch reflex, causing extreme bradycardia and hypotension. The local response of arterial blood vessels to 5-HT also may be inhibitory, the result of the stimulation of endothelial NO production and prostaglandin synthesis and blockade of norepinephrine release from sympathetic nerves. On the other hand, 5-HT amplifies the local constrictor actions of norepinephrine, angiotensin II, and histamine, which reinforce the hemostatic response to 5-HT. Presumably, these constrictor effects result, at least in part, from direct action of 5-HT on vascular smooth muscle. In response to IV injection of 5-HT, these disparate effects on blood pressure may be temporally distinct.
GI Tract. Enterochromaffin cells in the gastric mucosa are the site of the synthesis and most of the storage of 5-HT in the body and are the source of circulating 5-HT. These cells synthesize 5-HT from tryptophan, by means of TPH1, and store 5-HT and other autacoids, such as the vasodilator peptide substance P and other kinins. Basal release of enteric 5-HT is augmented by mechanical stretching, such as that caused by food, and by efferent vagal stimulation. Released 5-HT enters the portal vein and is subsequently metabolized by MAO-A in the liver. 5-HT that survives hepatic oxidation may be captured by platelets or rapidly removed by the endothelium of lung capillaries and inactivated. 5-HT released from enterochromaffin cells also acts locally to regulate GI function (Gershon and Tack, 2007). Motility of gastric and intestinal smooth muscle may be either enhanced or inhibited by at least six subtypes of 5-HT receptors (Table 13–3). The stimulatory response occurs at nerve endings on longitudinal and circular enteric muscle (5-HT4 receptors), at postsynaptic cells of the enteric ganglia (5-HT3 and 5-HT1P receptors), and by direct effects of 5-HT on the smooth-muscle cells (5-HT2A receptors in intestine, 5-HT2B receptors in stomach fundus). In esophagus, 5-HT acting at 5-HT4 receptors causes either relaxation or contraction, depending on the species. Abundant 5-HT3 receptors on vagal and other afferent neurons and on enterochromaffin cells play a pivotal role in emesis (Chapter 46). Serotonergic nerve terminals have been described in the myenteric plexus. Enteric 5-HT triggers peristaltic contraction when released in response to acetylcholine, sympathetic nerve stimulation, increases in intraluminal pressure, and lowered pH.
CNS. A multitude of brain functions are influenced by 5-HT, including sleep, cognition, sensory perception, motor activity, temperature regulation, nociception, mood, appetite, sexual behavior, and hormone secretion. All of the cloned 5-HT receptors are expressed in the brain, often in overlapping domains. Although patterns of 5-HT receptor expression in individual neurons have not been extensively defined, it is likely that multiple 5-HT receptor subtypes with similar or opposing actions are expressed in individual neurons (Bonsi et al., 2007), leading to a tremendous diversity of actions.
The principal cell bodies of 5-HT neurons are located in raphe nuclei of the brainstem and project throughout the brain and spinal cord (Chapter 14). In addition to being released at discrete synapses, release of serotonin also seems to occur at sites of axonal swelling, termed varicosities, which do not form distinct synaptic contacts. 5-HT released at nonsynaptic varicosities is thought to diffuse to outlying targets, rather than acting on discrete synaptic targets. Such non-synaptic release with ensuing widespread effects is consistent with the idea that 5-HT acts as a neuromodulator as well as a neurotransmitter (Chapter 14).
Serotonergic nerve terminals contain the proteins needed to synthesize 5-HT from L-tryptophan (Figure 13–2). Newly formed 5-HT is rapidly accumulated in synaptic vesicles (through VMAT2), where it is protected from MAO. 5-HT released by nerve-impulse flow is reaccumulated into the pre-synaptic terminal by the 5-HT transporter, SERT (SLC6A4; Chapter 5). Pre-synaptic reuptake is a highly efficient mechanism for terminating the action of 5-HT released by nerve-impulse flow. MAO localized in postsynaptic elements and surrounding cells rapidly inactivates 5-HT that escapes neuronal reuptake and storage.
Electrophysiology. The physiological consequences of 5-HT release vary with the brain area and the neuronal element involved, as well as with the population of 5-HT receptor subtype(s) expressed (Bockaert et al., 2006). 5-HT has direct excitatory and inhibitory actions (Table 13–4), which may occur in the same preparation, but with distinct temporal patterns. For example, in hippocampal neurons, 5-HT elicits hyperpolarization mediated by 5-HT1A receptors followed by a slow depolarization mediated by 5-HT4 receptors.
5-HT1A receptor–induced membrane hyperpolarization and reduction in input resistance results from an increase in K+ conductance. These ionic effects, which are blocked by pertussis toxin, are independent of cAMP, suggesting that 5-HT1A receptors couple, by means of Gβγ subunits, to receptor-operated K+ channels (Andrade et al., 1986). Somatodendritic 5-HT1A receptors on raphe cells also elicit a K+-dependent hyperpolarization. The G protein involved is pertussis toxin–sensitive, but the K+ current apparently is different from the current elicited at postsynaptic 5-HT1A receptors in the hippocampus. The precise signaling mechanism involved in inhibition of neurotransmitter release by the 5-HT1B/1D autoreceptor at synaptic terminals is not known, although inhibition of voltage-gated Ca++ channels likely contributes.
Slow depolarization induced by 5-HT2A receptor activation in areas such as the prefrontal cortex and nucleus accumbens involves a decrease in K+ conductance. A second, distinct mechanism involving Ca2+-activated membrane currents enhances neuronal excitability and potentiates the response to excitatory signals such as glutamate. The role of intracellular signaling cascades in these physiological actions of 5-HT2A receptors has not been clearly defined. In areas where 5-HT1 and 5-HT2A receptors co-exist, the effect of 5-HT may reflect a combination of the two opposing responses: a prominent 5-HT1 receptor–mediated hyperpolarization and an opposing 5-HT2A receptor–mediated depolarization. When 5-HT2A receptors are blocked, hyperpolarization is enhanced. In many cortical areas, 5-HT2A receptors are localized on both GABAergic interneurons and pyramidal cells. Activation of interneurons enhances GABA (γ-aminobutyric acid) release, which secondarily slows the firing rate of pyramidal cells. Thus, there is the potential for the 5-HT2A receptor to differentially regulate cortical pyramidal cells, depending on the specific target cells (interneurons versus pyramidal cells). Activation of 5-HT2C receptors has been shown to depress a K+ current, an effect that may contribute to the excitatory response. The 5-HT4 receptor, which is coupled to activation of adenylyl cyclase, also elicits a slow neuronal depolarization mediated by a decrease in K+ conductance. It is not clear why two distinct 5-HT receptor families linked to different signaling pathways can elicit a common neurophysiological action. Yet another receptor, the 5-HT1P receptor, elicits a slow depolarization. This receptor, which couples to activation of adenylyl cyclase, is restricted to the enteric nervous system (Gershon and Tack, 2007). The unique pharmacological profile of the 5-HT1P receptor is most consistent with the 5-HT7 receptor.
The fast depolarization elicited by 5-HT3 receptors reflects direct gating of an ion channel intrinsic to the receptor structure itself. The 5-HT3 receptor–induced inward current has the characteristics of a cation-selective, ligand-operated channel. Membrane depolarization is mediated by simultaneous increases in Na+ and K+ conductance, comparable to the nicotinic cholinergic receptor. 5-HT3 receptors have been characterized in the CNS and in sympathetic ganglia, primary afferent parasympathetic and sympathetic nerves, and enteric neurons. The pharmacological properties of 5-HT3 receptors, which are different from those of other 5-HT receptors, suggest that multiple 5-HT3 receptor subtypes may exist and may correspond to different combinations of subunits (Jensen et al., 2008).
Behavior. The behavioral alterations elicited by drugs that interact with 5-HT receptors are extremely diverse. Many animal behavioral models for initial assessment of agonist and antagonist properties of drugs depend on aberrant motor or reflex responses, such as startle reflexes, hind-limb abduction, head twitches, and other stereotypical behaviors. Operant behavioral paradigms, such as drug discrimination, provide models of specific 5-HT receptor activation and are useful for exploring the action of CNS-active drugs, including agents that interact with 5-HT. For example, investigations of the mechanism of action of hallucinogenic drugs have relied heavily on drug discrimination (Winter, 2009). The following discussion focuses on animal models that may relate to pathological conditions in humans and will not attempt to cover the voluminous literature dealing with 5-HT and behavior. See King and associates (2008), Lesch and coworkers (2003), Lucki (1998), and Swerdlow and colleagues (2000) for excellent reviews on this topic.
Sleep-Wake Cycle. Control of the sleep-wake cycle is one of the first behaviors in which a role for 5-HT was identified. Following pioneering work in cats (Jouvet, 1999), many studies showed that depletion of 5-HT with p-chlorophenylalanine, a tryptophan hydroxylase inhibitor, elicits insomnia that is reversed by the 5-HT precursor, 5-hydroxytryptophan. Conversely, treatment with L-tryptophan or with nonselective 5-HT agonists accelerates sleep onset and prolongs total sleep time. 5-HT antagonists reportedly can increase and decrease slow-wave sleep, probably reflecting interacting or opposing roles for subtypes of 5-HT receptors. One relatively consistent finding in humans and in laboratory animals is an increase in slow-wave sleep following administration of a selective 5-HT2A/2C-receptor antagonist such as ritanserin.
Aggression and Impulsivity. Studies in laboratory animals and in humans suggest that 5-HT serves a critical role in aggression and impulsivity. Many human studies reveal a correlation between low cerebrospinal fluid 5-HIAA and violent impulsivity and aggression. As with many effects of 5-HT, pharmacological studies of aggressive behavior in laboratory animals are not definitive, but suggest a role for 5-HT. Two genetic studies have reinforced and amplified this notion. Knockout mice lacking the 5-HT1B receptor exhibited extreme aggression (Saudou et al., 1994), suggesting either a role for 5-HT1B receptors in the development of neuronal pathways important in aggression or a direct role in the mediation of aggressive behavior. A human genetic study identified a point mutation in the gene encoding MAO-A, which was associated with extreme aggressiveness and mental retardation (Brunner et al., 1993), and this has been confirmed in knockout mice lacking MAO-A (Cases et al., 1995).
Anxiety and Depression. The effects of 5-HT–active drugs in anxiety and depressive disorders, like the effects of selective serotonin reuptake inhibitors (SSRIs), strongly suggest a role for 5-HT in the neurochemical mediation of these disorders. Mutant mice lacking the 5-HT transporter display anxiety and a "depressive-like" phenotype (Fox et al., 2007). 5-HT-related drugs with clinical effects in anxiety and depression have varied effects in classical animal models of these disorders, depending on the experimental paradigm, species, and strain. For example, the anxiolytic buspirone (Chapter 15), a partial agonist at 5-HT1A receptors, does not reduce anxiety in classical approach-avoidance paradigms that were instrumental in development of anxiolytic benzodiazepines. However, buspirone and other 5-HT1A receptor agonists are effective in other animal behavioral tests used to predict anxiolytic effects. Furthermore, studies in 5-HT1A receptor knockout mice suggest a role for this receptor in anxiety, and possibly depression. Agonists of certain 5-HT receptors, including 5-HT2A and 5-HT2C receptors (e.g., m-chlorophenylpiperazine), have anxiogenic properties in laboratory animals and in human studies. Similarly, these receptors have been implicated in the animal models of depression, such as learned helplessness.
An impressive finding in humans with depression is the abrupt reversal of the antidepressant effects of drugs such as SSRIs by manipulations that rapidly reduce the amount of 5-HT in the brain. These approaches include administration of p-chlorophenylalanine or a tryptophan-free drink containing large quantities of neutral amino acids (Delgado et al., 1990). Curiously, this kind of 5-HT depletion has not been shown to worsen or to induce depression in nondepressed subjects, suggesting that the continued presence of 5-HT is required to maintain the effects of these drugs. This clinical finding adds credence to somewhat less convincing neurochemical findings that suggest a role for 5-HT in the pathogenesis of depression.