24: Pain

• Noxious Insults Activate Nociceptors

• Signals from Nociceptors Are Conveyed to Neurons in the Dorsal Horn of the Spinal Cord

• Hyperalgesia Has Both Peripheral and Central Origins

• Nociceptive Information Is Transmitted from the Spinal Cord to the Thalamus

  • Five Major Ascending Pathways Convey Nociceptive Information

  • Several Thalamic Nuclei Relay Nociceptive Information to the Cerebral Cortex

• Pain Is Controlled by Cortical Mechanisms

  • Cingulate and Insular Areas Are Active During the Perception of Pain

  • Pain Perception Is Regulated by a Balance of Activity in Nociceptive and Non-Nociceptive Afferent Fibers

  • Electrical Stimulation of the Brain Produces Analgesia

• Opioid Peptides Contribute to Endogenous Pain Control

  • Endogenous Opioid Peptides and Their Receptors Are Distributed in Pain-Modulatory Systems

  • Morphine Controls Pain by Activating Opioid Receptors

  • Tolerance and Addiction to Opioids Are Distinct Phenomena

• An Overall View

Pain describes the unpleasant sensory and emotional experiences associated with actual or potential tissue damage. Pricking, burning, aching, stinging, and soreness are among the most distinctive of all the sensory modalities. As with the other somatic sensory modalities—touch, pressure, and position sense—pain serves an important protective function, alerting us to injuries that require evasion or treatment. In children born with insensitivity to pain, severe injuries often go unnoticed and can lead to permanent tissue damage. Yet pain is unlike other somatic sensory modalities, or vision, hearing, and smell in that it has an urgent and primitive quality, possessing both affective and emotional components.

The perception of pain is subjective and is influenced by many factors. An identical sensory stimulus can elicit quite distinct responses in the same individual under different conditions. Many wounded soldiers, for example, do not feel pain until they have been removed from the battlefield; injured athletes are often not aware of pain until a game is over. Simply put, there are no purely "painful" stimuli, sensory stimuli that invariably elicit the perception of pain in all individuals. The variability of the perception of pain is yet another example of a principle that we have encountered in earlier chapters: Pain is not the direct expression of a sensory event but rather the product of elaborate processing by the brain of a variety of neural signals.

When pain is experienced it can be acute, persistent, or in extreme cases chronic. Persistent pain characterizes many clinical conditions and is usually the reason that patients seek medical attention. In contrast, chronic pain appears to have no useful purpose; it only makes patients miserable. Pain's highly individual and subjective nature is one of the factors that make it so difficult to define objectively and to treat clinically.

In this chapter we discuss the neural processes that underlie the perception of pain in normal individuals and explain the origins of some of the abnormal pain states that are encountered clinically.

Many organs in the periphery, including skin and subcutaneous structures such as joints and muscles, possess specialized sensory receptors that are activated by noxious insults. Unlike the specialized somatosensory receptors for light touch and pressure, most of these nociceptors are simply the free nerve endings of primary sensory neurons. There are three main classes of nociceptors—thermal, mechanical, and polymodal—as well as a more enigmatic fourth class, termed silent nociceptors.

Thermal nociceptors are activated by extremes in temperature, typically greater than 45°C (115°F) or less than 5°C (41°F). They are the peripheral endings of small-diameter, thinly myelinated Aδ axons that conduct action potentials at speeds of 5 to 30 m/s (Figure 24–1A). Mechanical nociceptors are activated optimally by intense pressure applied to the skin; they too are the endings of thinly myelinated Aδ axons. Polymodal nociceptors can be activated by high-intensity mechanical, chemical, or thermal (both hot and cold) stimuli. This class of nociceptors is found at the ends of small-diameter, unmyelinated C axons that conduct more slowly, at speeds less than 1.0 m/s (Figure 24–1A).

These three classes of nociceptors are widely distributed in skin and deep tissues and are often coactivated. When a hammer hits your thumb, you initially feel a sharp pain ("first pain") followed by a more prolonged aching and sometimes burning pain ("second pain") (Figure 24–1B). The fast sharp pain is transmitted by Aδ fibers that carry information from damaged thermal and mechanical nociceptors. The slow dull pain is transmitted by C fibers that convey signals from polymodal nociceptors.

Silent nociceptors are found in the viscera. This class of receptors is not normally activated by noxious stimulation; instead, inflammation and various chemical agents dramatically reduce their firing threshold. Their activation is thought to contribute to the emergence of secondary hyperalgesia and central sensitization, two prominent pain syndromes.

Noxious stimuli depolarize the bare nerve endings of afferent axons and generate action potentials that are propagated centrally. How is this achieved? The membrane of the nociceptor contains receptors that convert the thermal, mechanical, or chemical energy of noxious stimuli into a depolarizing electrical potential. One such protein is a member of a large family of so-called transient receptor potential (TRP) ion channels. This receptor-channel, TRPV1, is expressed selectively by nociceptive neurons and mediates the pain-producing actions of capsaicin, the active ingredient of hot peppers, and many other pungent chemicals. The TRPV1 channel is also activated by noxious thermal stimuli, which suggests that it normally transduces the sensation of painful heat. In addition, TRPV1-mediated membrane currents are enhanced by a reduction in pH, a characteristic of the chemical milieu of inflammation.

Additional members of the TRP channel family are expressed by nociceptive neurons, and the variety of TRP channels in nociceptors is thought to underlie the perception of a wide range of temperatures from extreme cold to intense heat. The TRPV2 channel is expressed predominantly in Aδ fiber terminals and is activated by very high temperatures, whereas the TRPM8 channel is activated by low temperatures and by chemicals such as menthol (Figure 24–2).

In addition to this constellation of TRP channels, other receptors and ion channels that participate in the transduction of peripheral stimuli are expressed in nociceptive sensory endings. Nociceptors selectively express tetrodotoxin-resistant Na⁺ channels. One Na⁺ channel (SCN9A, also called Na_V 1.7) plays a key role in the perception of pain in humans, as revealed by the rare pain-insensitive individuals who possess mutations in the corresponding gene. One class of mutations inactivates the SCN9A channel and results in a complete inability to sense pain. But in all other respects these individuals are healthy and exhibit normal sensory responses to touch, mild temperature, proprioception, tickle, and pressure. A second class of mutations in the SCN9A gene changes the inactivation kinetics of this channel; individuals with these mutations exhibit an inherited condition called paroxysmal extreme pain disorder, characterized by rectal, ocular, and submandibular pain.

Nociceptors also express an ionotropic purinergic receptor, PTX3, that is activated by adenosine triphosphate (ATP) released from peripheral cells after tissue damage. In addition, they express members of the Mas-related G protein-coupled receptor (Mrg) family, which are activated by peptide ligands and serve to sensitize nociceptors to other chemicals released in their local environment (see Figure 24–7). These receptors and channels provide attractive targets for the development of drugs with actions selective for nociceptive sensory neurons.

The uncontrolled activation of nociceptors is associated with several pathological conditions. Allodynia and hyperalgesia are two common pain states that reflect changes in nociceptor activity. Patients with allodynia feel pain in response to stimuli that are normally innocuous: by a light stroking of sunburned skin, by the movement of joints in patients with rheumatoid arthritis, and even by the act of getting out of bed in the morning after a vigorous workout. Nevertheless, patients with allodynia do not feel pain constantly; in the absence of a peripheral stimulus there is no pain. In contrast, patients with hyperalgesia —an exaggerated response to noxious stimuli—typically report persistent pain in the absence of sensory stimulation.

Persistent pain can be subdivided into two broad classes, nociceptive and neuropathic. Nociceptive pain results from the activation of nociceptors in the skin or soft tissue in response to tissue injury, and it usually arises from an accompanying inflammation. Sprains and strains produce mild forms of nociceptive pain, whereas arthritis or a tumor that invades soft tissue produces a much more severe nociceptive pain. Neuropathic pain results from direct injury to nerves in the peripheral or central nervous system and is often accompanied by a burning or electric sensation. Neuropathic pains include the syndromes of reflex sympathetic dystrophy, also called complex regional pain syndrome, and post-herpetic neuralgia, the severe pain experienced by patients after a bout of shingles. Other neuropathic pains include phantom limb pain, the pain that occurs after limb amputation, which we discuss below. In some instances pain can even occur without a peripheral stimulus, a phenomenon termed anesthesia dolorosa. This syndrome can be triggered following attempts to block chronic pain, for example after therapeutic transection of sensory afferent fibers in the dorsal roots.

The perception of noxious stimuli arises from signals in the peripheral axonal branches of nociceptive sensory neurons whose cell bodies are located in dorsal root ganglia or the trigeminal ganglia. The central branches of these neurons terminate in the spinal cord in a highly orderly manner. Most terminate in the dorsal horn. Primary afferent neurons that convey distinct sensory modalities terminate in different laminae (Figure 24–3B). Thus there is a tight link between the anatomical organization of dorsal horn neurons, their receptive properties, and their function in sensory processing.

Many neurons in the most superficial lamina of the dorsal horn, termed lamina I or the marginal layer, respond to noxious stimuli conveyed by Aδ and C fibers. Because they respond selectively to noxious stimulation they have been called nociception-specific neurons. This set of neurons projects to higher brain centers, notably the thalamus. A second class of lamina I neurons receives input from C fibers that are activated selectively by intense cold. Other classes of lamina I neurons respond in a graded fashion to both innocuous and noxious mechanical stimulation and thus are termed wide-dynamic-range neurons.

Lamina II, the substantia gelatinosa, is a densely packed layer that contains many different classes of local interneurons, some excitatory and others inhibitory. Some of these interneurons respond selectively to nociceptive inputs, whereas others also respond to innocuous stimuli. Laminae III and IV contain a mixture of local interneurons and supraspinal projection neurons. Many of these neurons receive input from Aβ afferent fibers that respond to innocuous cutaneous stimuli, such as the deflection of hairs and light pressure. Lamina V contains neurons that respond to a wide variety of noxious stimuli and project to the brain stem and thalamus. These neurons receive direct inputs from Aβ and Aδ fibers and, because their dendrites extend into lamina II, are also innervated by C fiber nociceptors (Figure 24–3B).

Neurons in lamina V also receive input from nociceptors in visceral tissues. The convergence of somatic and visceral nociceptive inputs onto individual lamina V neurons provides one explanation for a phenomenon called "referred pain," a condition in which pain from injury to a visceral tissue is perceived as originating from a region of the body surface. Patients with myocardial infarction, for example, frequently report pain from the left arm as well as the chest (Figure 24–4). This phenomenon occurs because a single lamina V neuron receives sensory input from both regions, and thus a signal from this neuron does not inform higher brain centers about the source of the input. As a consequence, the brain often incorrectly attributes the pain to the skin, possibly because cutaneous inputs predominate. Another anatomical explanation for instances of referred pain is that the axons of nociceptive sensory neurons branch in the periphery, innervating both skin and visceral targets.

Neurons in lamina VI receive inputs from large-diameter fibers that innervate muscles and joints. These neurons are activated by innocuous joint movement and do not contribute to the transmission of nociceptive information. In contrast, many neurons located in laminae VII and VIII, the intermediate and ventral regions of the spinal cord, do respond to noxious stimuli. These neurons typically have complex response properties because the inputs from nociceptors to these neurons are conveyed through many intervening synapses. Neurons in lamina VII often respond to stimulation of either side of the body, whereas most dorsal horn neurons receive unilateral input. The activation of lamina VII neurons is therefore thought to contribute to the diffuse quality of many pain conditions.

Nociceptive sensory neurons that activate neurons in the dorsal horn of the spinal cord release two major classes of neurotransmitters. Glutamate is the primary neurotransmitter of all primary sensory neurons, regardless of sensory modality. Neuropeptides are released as cotransmitters by many nociceptors with unmyelinated axons. These peptides include substance P, calcitonin gene–related peptide (CGRP), somatostatin, and galanin (Figure 24–5). Glutamate is stored in small, electron-lucent vesicles, whereas peptides are sequestered in large, dense-core vesicles at the central terminals of nociceptive sensory neurons (Figure 24–6). Separate storage sites permit these two classes of neurotransmitters to be released under different physiological conditions.

Of the neuropeptides expressed by nociceptive sensory neurons, the actions of substance P, a member of the neurokinin peptide family, have been studied in most detail. Substance P is released from the central terminals of nociceptive afferents in response to tissue injury or after intense stimulation of peripheral nerves. Its interaction with neurokinin receptors on dorsal horn neurons elicits slow excitatory postsynaptic potentials that prolong the depolarization elicited by glutamate. Although the physiological actions of glutamate and neuropeptides on dorsal horn neurons are different, these transmitters act coordinately to regulate the firing properties of dorsal horn neurons.

There is no efficient means for peptide reuptake into nerve terminals, so neuropeptides released from sensory terminals diffuse over a greater distance than glutamate and thus have excitatory influences on many dorsal horn neurons in the immediate vicinity of the release site. This diffuse signaling system may contribute to the poorly localized character of many pain conditions. In addition, the levels of neuropeptide expression in primary nociceptive neurons are elevated in some pathological conditions. Such reactive changes in peptide expression may contribute to the enhanced excitability of dorsal horn neurons that accompanies some chronic pain states.

Details of the interaction of neuropeptides with their receptors on dorsal horn neurons have suggested strategies for pain regulation. Infusion of substance P coupled to a neurotoxin into the dorsal horn of experimental animals results in selective destruction of neurons that express neurokinin receptors. Animals treated in this way fail to develop the central sensitization that is normally associated with peripheral injury. This method of neuronal ablation is more selective than traditional surgical interventions such as partial spinal cord transection (anterolateral cordotomy) and is being explored as a treatment for patients suffering from chronic pain syndrome.

Up to this point we have considered the mechanisms that convey noxious signals in the normal physiological state. But the normal process of sensory signaling can be dramatically altered when peripheral tissue is damaged, resulting in an increase in pain sensitivity or hyperalgesia. This condition can be elicited by sensitizing peripheral nociceptors through repetitive exposure to noxious mechanical stimuli (Figures 24–7).

The sensitization is triggered by a complex mix of chemicals released from damaged cells that accumulate at the site of tissue injury. This cocktail contains peptides and proteins such as bradykinin, substance P, and nerve growth factor, as well as molecules such as ATP, histamine, serotonin, prostaglandins, leukotrienes, and acetylcholine. Many of these chemical mediators are released from distinct cell types, but together they act to decrease the threshold of nociceptor activation.

Where do these chemicals come from and what exactly do they do? Histamine is released from mast cells after tissue injury and activates polymodal nociceptors. The lipid anandamide, an endogenous cannabinoid agonist, is released under conditions of inflammation, activates the TRPV1 channel, and may trigger pain associated with inflammation. ATP, acetylcholine, and serotonin are released from damaged endothelial cells and platelets; they act indirectly to sensitize nociceptors by triggering the release of chemical agents such as prostaglandins and bradykinin from peripheral cells. Bradykinin is one of the most active pain-producing agents. Its potency stems in part from the fact that it directly activates Aδ and C nociceptors as well as increasing the synthesis and release of prostaglandins from nearby cells.

Damaged cells also release prostaglandins, metabolites of arachidonic acid that are generated through the activity of cyclooxygenase (COX) enzymes that cleave arachidonic acid. The COX2 enzyme is preferentially induced under conditions of peripheral inflammation, contributing to enhanced pain sensitivity. The enzymatic pathways of prostaglandin synthesis are targets of commonly used analgesic drugs. Aspirin and other nonsteroidal anti-inflammatory analgesics are effective in controlling pain because they block the activity of most COX enzymes, reducing prostaglandin synthesis. The drug acetaminophen appears to exert its analgesic effects through selective inhibition of COX3.

Tissue inflammation also results from peripheral injury. The cardinal signs of inflammation are heat (calor), redness (rubor), and swelling (tumor). Heat and redness result from the dilation of peripheral blood vessels, whereas swelling results from plasma extravasation, a process in which proteins, cells, and fluids are able to penetrate post-capillary venules. Release of the neuropeptides substance P and CGRP from the peripheral terminals of C fibers can elicit each of these pathological responses. Because this form of inflammation depends on neural activity, it has been termed neurogenic inflammation (Figure 24–8).

The release of substance P and CGRP from the peripheral terminals of sensory neurons is also responsible for the axon reflex, a physiological process characterized by vasodilation in the vicinity of a cutaneous injury. Pharmacological antagonists of substance P are able to block neurogenic inflammation and vasodilation in humans; this discovery illustrates how knowledge of nociceptive mechanisms has relevance for improved clinical therapies for pain.

In addition to these small molecules and peptides, neurotrophins are causative agents in pain. Nerve growth factor (NGF) is particularly active in inflammatory pain states; the synthesis of NGF is upregulated in many inflamed peripheral tissues (Figure 24–9). NGF-neutralizing molecules are effective analgesic agents in animal models of persistent pain. Indeed, inhibition of NGF function and signaling blocks pain sensation as effectively as COX inhibitors and opiates. The expression of brain derived neurotrophic factor (BDNF) is also elevated in nociceptive neurons under conditions of inflammatory and neuropathic pain. BDNF is transported to the central terminals of the primary afferents in the spinal dorsal horn, where its release enhances the response of dorsal horn neurons to painful stimuli (Figure 24–9). Drugs that block BDNF expression in nociceptive neurons may therefore be useful as analgesics.

What accounts for the enhanced sensitivity of dorsal horn neurons to nociceptor signals? Under conditions of persistent injury C fibers fire repetitively and the response of dorsal horn neurons increases progressively (Figure 24–10A). The gradual enhancement in the excitability of dorsal horn neurons has been termed "wind-up" and is thought to involve N -methyl-D-aspartate (NMDA)-type glutamate receptors (Figure 24–10B).

Repeated exposure to noxious stimuli therefore results in long-term changes in the response of dorsal horn neurons through mechanisms that are similar to those underlying the long-term potentiation of synaptic responses in many circuits in the brain. In essence these prolonged changes in the excitability of dorsal horn neurons constitute a "memory" of the state of C fiber input. This phenomenon has been termed central sensitization to distinguish it from sensitization at the peripheral terminals of the dorsal horn neurons, a process that involves activation of the enzymatic pathways of prostaglandin synthesis.

The sensitization of dorsal horn neurons also involves recruitment of second-messenger pathways and activation of protein kinases that have been implicated in memory storage in other regions of the central nervous system. One consequence of this enzymatic cascade is the expression of immediate-early genes that encode transcription factors such as c-fos, which are thought to activate effector proteins that sensitize dorsal horn neurons to sensory inputs.

In certain clinical conditions alterations in the excitability of dorsal horn neurons can decrease pain thresholds and lead to spontaneous pain. One dramatic illustration of this phenomenon is phantom limb pain, the persistent sensations of pain that appear to originate from the region of an amputated limb (Figure 24–11). Until recently limb amputation was performed solely with a general anesthetic on the assumption that this alone should be sufficient to eliminate memory of the traumatic surgical procedure. Surgeons found, however, that even under general anesthesia the spinal cord "experiences" the insult of the surgical procedure, presumably because of ongoing central sensitization. So to reduce the risk of phantom limb pain, amputation now includes interventions that block the central sensitization of dorsal horn neurons. General anesthesia is often supplemented with direct spinal administration of an analgesic agent or local administration of anesthetic at the injury site.

Five Major Ascending Pathways Convey Nociceptive Information

Five major ascending pathways—the spinothalamic, spinoreticular, spinomesencephalic, cervicothalamic, and spinohypothalamic tracts—contribute to the central processing of nociceptive information.

The spinothalamic tract is the most prominent ascending nociceptive pathway in the spinal cord. It includes the axons of nociception-specific, thermosensitive, and wide-dynamic-range neurons in laminae I and V through VII of the dorsal horn. These axons cross the midline of the spinal cord at their segment of origin and ascend in the anterolateral white matter before terminating in thalamic nuclei (Figure 24–12). The spinothalamic tract has a crucial role in the transmission of nociceptive information. Electrical stimulation of the tract is sufficient to elicit the sensation of pain; conversely, lesioning this tract (anterolateral cordotomy) can result in a marked reduction in pain sensation on the side of the body contralateral to that of the lesion.

The spinoreticular tract contains the axons of projection neurons in laminae VII and VIII. This tract ascends in the anterolateral quadrant of the spinal cord and terminates in both the reticular formation and the thalamus (Figure 24–12). The axons of spinoreticular tract neurons do not cross the midline.

The spinomesencephalic (or spinoparabrachial) tract contains the axons of projection neurons in laminae I and V. Information transmitted along this tract is thought to contribute to the affective component of pain. This tract projects in the anterolateral quadrant of the spinal cord to the mesencephalic reticular formation and periaqueductal gray matter (Figure 24–12). Axons in this tract also project to the parabrachial nucleus. Neurons of the parabrachial nucleus project to the amygdala, a key nucleus of the limbic system that regulates emotional states. Many of the axons of this pathway course through the dorsal part of the lateral funiculus rather than in the anterolateral quadrant. In surgical procedures designed to relieve pain, such as anterolateral cordotomy, the sparing of these fibers may explain the persistence or recurrence of pain after surgery.

The cervicothalamic tract runs in the lateral white matter of the upper two cervical segments of the spinal cord and contains the axons of neurons of the lateral cervical nucleus, which receives input from neurons in laminae III and IV of the dorsal horn. Most axons in the cervicothalamic tract cross the midline and ascend in the medial lemniscus of the brain stem, terminating in midbrain nuclei and in the ventroposterior lateral and posteromedial nuclei of the thalamus. Other neurons in laminae III and IV send their axons directly into the dorsal columns and terminate in the cuneate and gracile nuclei of the medulla.

The spinohypothalamic tract contains the axons of neurons found in laminae I, V, and VIII of the dorsal horn in the spinal cord. These axons project to hypothalamic nuclei that serve as autonomic control centers involved in the regulation of the neuroendocrine and cardiovascular responses that accompany pain syndromes.

Several Thalamic Nuclei Relay Nociceptive Information to the Cerebral Cortex

The thalamus contains several relay nuclei that participate in the central processing of nociceptive information. Two of the most important regions of the thalamus are the lateral and medial nuclear groups.

The lateral nuclear group comprises the ventroposterior medial nucleus, the ventroposterior lateral nucleus, and the posterior nucleus. These three nuclei receive inputs through the spinothalamic tract from nociception-specific and wide-dynamic-range neurons in laminae I and V of the dorsal horn. The lateral thalamus is thought to be concerned with the processing of information about the precise location of an injury, information usually conveyed to consciousness as acute pain. Consistent with this view, neurons in the lateral thalamic nuclei have small receptive fields, matching those of the presynaptic spinal neurons.

Injury to the spinothalamic tract and its thalamic targets causes a severe form of pain termed central pain. An infarct in a small region of the ventroposterolateral thalamus produces the perception of a spontaneous burning pain, together with other abnormal sensations (called dysesthesias) that are perceived as originating from diverse regions of the body. This constellation of abnormal percepts has been termed the Dejerine-Roussy syndrome. Electrical stimulation of the thalamus can also result in intense pain. In one dramatic clinical case, electrical stimulation of the thalamus rekindled sensations of angina pectoris that were so realistic that the anesthesiologist thought the patient was experiencing a heart attack. This and other clinical observations suggest that in chronic pain conditions there is a fundamental change in thalamic and cortical circuitry.

The medial nuclear group of the thalamus comprises the central lateral nucleus of the thalamus and the intralaminar complex. Its major input is from neurons in laminae VII and VIII of the dorsal horn. The pathway to the medial thalamus was the first spinothalamic projection evident in the evolution of mammals and is therefore known as the paleospinothalamic tract. It is also sometimes referred to as the spinoreticulothalamic tract because it includes indirect connections through the reticular formation of the brain stem. The projection from the lateral thalamus to the ventroposterior lateral and medial nuclei is most developed in primates, and is termed the neospinothalamic tract. Many neurons in the medial thalamus respond optimally to noxious stimuli and project widely to the basal ganglia and different cortical areas.

Cingulate and Insular Areas Are Active During the Perception of Pain

Until recently research on the central processing of pain concentrated on the thalamus. But pain is a complex perception that involves many cortical areas whose activity is influenced critically by the context in which the noxious stimulus is presented, as well as by an individual's prior experience.

Neurons in several areas of the cerebral cortex respond to nociceptive input. In the somatosensory cortex neurons typically have small receptive fields and may not contribute greatly to the diffuse perception of aches and pains that characterize most clinical syndromes. The cingulate gyrus and insular cortex contain neurons that are activated strongly and selectively by nociceptive somatosensory stimuli (Box 24–1). The cingulate gyrus forms part of the limbic system and is thought to be involved in processing emotional states associated with pain. The insular cortex receives direct projections from the thalamus, specifically the medial nuclei and the ventroposterior medial nucleus. Neurons in the insular cortex process information about the internal state of the body and contribute to the autonomic component of pain responses.

Patients with lesions of the insular cortex present the striking syndrome of asymbolia for pain: They perceive noxious stimuli as painful and can distinguish sharp from dull pain but fail to display appropriate emotional responses. These observations implicate the insular cortex as an area in which the sensory, affective, and cognitive components of pain are integrated.

Pain Perception Is Regulated by a Balance of Activity in Nociceptive and Non-Nociceptive Afferent Fibers

Many projection neurons in the dorsal horn of the spinal cord respond selectively to noxious inputs, but others receive convergent inputs from both nociceptive and non-nociceptive afferents. The concept that the convergence of sensory inputs onto spinal projection neurons regulates pain processing first emerged in the 1960s.

Ronald Melzak and Patrick Wall proposed that the relative balance of activity in nociceptive and non-nociceptive afferents might influence the transmission and perception of pain. In particular, they proposed that activation of non-nociceptive sensory neurons closes a "gate" for central transmission of nociceptive signals that can be opened by the activation of nociceptive sensory neurons. In the original and simplest form of this gate-control theory the interaction between large and small fibers occurred at the first possible site of convergence on projection neurons in the dorsal horn of the spinal cord (Figure 24–14). We now know that such interactions can also occur at many supraspinal relay centers and the strict anatomical predictions of the gate-control theory remain to be established.

Nevertheless, the core concept of convergence of different sensory modalities has provided an important basis for the design of new pain therapies. Viewed in its broadest sense, the convergence of high- and low-threshold inputs at spinal or supraspinal sites provided a plausible explanation for several empirical observations about the perception of pain. The shaking of the hand that follows a hammer blow or burn is a reflexive behavior and may alleviate pain by activating large-diameter afferents that suppress the central transmission of noxious stimuli.

The idea of convergence also helped to promote the use of transcutaneous electrical nerve stimulation (TENS) and dorsal column stimulation for the relief of pain. With TENS, stimulating electrodes placed at peripheral locations activate large-diameter afferent fibers that overlap the area of injury and pain. The region of the body in which pain is reduced maps to those segments of the spinal cord in which nociceptive and non-nociceptive afferents from that body region terminate. This makes intuitive sense: You do not shake your left leg to relieve pain in your right arm.

Electrical Stimulation of the Brain Produces Analgesia

Several sites of endogenous pain regulation are located in the brain. One effective means of suppressing nociception involves stimulation of the periaqueductal gray region, the area of the midbrain that surrounds the third ventricle and the cerebral aqueduct. In experimental animals stimulation of this region elicits a profound and selective analgesia. This stimulation-produced analgesia is remarkably specific; animals still respond to touch, pressure, and temperature within the body area that exhibits analgesia, but they feel less pain. Indeed, stimulation-evoked analgesia has proved to be an effective way of relieving pain in a limited number of human pain conditions.

Stimulation of the periaqueductal gray matter is able to block spinally mediated withdrawal reflexes that are normally evoked by noxious stimulation. Few of the neurons in the periaqueductal gray matter project directly to the dorsal horn of the spinal cord. Most make excitatory connections with neurons of the rostroventral medulla, including serotonergic neurons in a midline region called the nucleus raphe magnus, or neurons of the nuclei of the raphe complex. The axons of these serotonergic neurons project through the dorsal region of the lateral funiculus to the spinal cord, where they form inhibitory connections with neurons in laminae I, II, and V of the dorsal horn (Figure 24–15). Stimulation of the rostroventral medulla thus inhibits the firing of many classes of dorsal horn neurons, including projection neurons of the spinothalamic tract that convey afferent nociceptive signals.

A second major monoaminergic descending system can also suppress the activity of nociceptive neurons in the dorsal horn. This noradrenergic system originates in the locus ceruleus and other nuclei of the medulla and pons. These projections inhibit neurons in laminae I and V of the dorsal horn through direct and indirect synaptic actions.

Since discovery by the Sumerians in 3300 BC, the opium poppy and its active ingredients, opiates such as morphine and codeine, have been recognized as powerful analgesic agents. Over the past two decades we have begun to understand many of the molecular mechanisms and neural circuits through which opiates exert their analgesic actions. In addition, we have come to realize that the neural networks involved in stimulation-produced and opiate-induced analgesia are intimately related.

Two key discoveries led to these advances. The first was the recognition that morphine and other opiates interact with specific receptors on neurons in the spinal cord and brain. The second was the isolation of endogenous neuropeptides with opiate-like activities at these receptors. The observation that the opiate antagonist naloxone blocks stimulation-produced analgesia provided the first clue that the brain contains endogenous opioids.

Endogenous Opioid Peptides and Their Receptors Are Distributed in Pain-Modulatory Systems

Opioid receptors fall into four major classes: mu (μ), delta (δ), kappa (κ), and orphanin FQ. The genes encoding each of these receptor types constitute a subfamily of G protein-coupled receptors. The μ receptors are particularly diverse; numerous μ receptor isoforms have been identified, many with different patterns of expression. This finding has prompted a search for analgesic drugs that target specific isoforms.

The opioid receptors were originally defined on the basis of the binding affinity of different agonist compounds. Morphine and other opioid alkaloids are potent agonists at μ receptors, and there is a tight correlation between the potency of an analgesic and its affinity of binding to μ receptors. Mice in which the gene for the μ receptor has been inactivated are insensitive to morphine and other opiate agonists. Many opiate antagonist drugs, such as naloxone, also bind to the μ receptor and compete with morphine for receptor occupancy without activating receptor signaling.

The μ receptors are highly concentrated in the superficial dorsal horn of the spinal cord, the ventral medulla, and the periaqueductal gray matter, important anatomical sites for the regulation of pain. Nevertheless, like other classes of opioid receptors, they are also found at many other sites in the central and peripheral nervous systems. Their widespread distribution explains why systemically administered morphine influences many physiological processes in addition to the perception of pain.

The discovery of opioid receptors and their expression by neurons in the central and peripheral nervous systems led to the definition of four major classes of endogenous opioid peptides, each interacting with a specific class of opioid receptors (Table 24–1).

Three classes—the enkephalins, β-endorphins, and dynorphins—are the best characterized. These peptides are formed from large polypeptide precursors by enzymatic cleavage (Figure 24–16) and encoded by distinct genes. Despite differences in amino acid sequence, each contains the sequence Tyr-Gly-Gly-Phe. There are two enkephalins, leucine and methionine enkephalin, which are closely related small peptides. β-endorphin is a cleavage product of a precursor that also generates the active peptide adrenocorticotropic hormone (ACTH). Both β-endorphin and ACTH are synthesized by cells in the pituitary and are released into the bloodstream in response to stress. Dynorphins are derived from the polyprotein product of the dynorphin gene. Enkephalins are active at both μ and δ receptors, whereas dynorphin is a relatively selective agonist of the κ receptor. The fourth endogenous opioid peptide is orphanin FQ or nociceptin (OFQ/N1–17). This 17-amino acid peptide is related in sequence to dynorphin and binds to the OFQ/N receptor.

Members of the four classes of opioid peptides are distributed widely in the central nervous system and individual peptides are located at sites associated with the processing or modulation of nociceptive information. Neuronal cell bodies and axon terminals containing enkephalin and dynorphin are found in the dorsal horn of the spinal cord, particularly in laminae I and II, as well as in the rostral ventral medulla and the periaqueductal gray matter. Neurons that synthesize β-endorphin are confined primarily to the hypothalamus; their axons terminate in the periaqueductal gray region and on noradrenergic neurons in the brain stem. Orphanin FQ appears to participate in a broad range of other physiological functions.

Morphine Controls Pain by Activating Opioid Receptors

Microinjection of low doses of morphine or other opiates directly into specific regions of the rat brain produces a powerful analgesia. The periaqueductal gray region is among the most sensitive sites, but local administration of morphine into other regions, including the spinal cord, also elicits a powerful analgesia.

Morphine-induced analgesia can be blocked by injection of the opiate antagonist naloxone into the periaqueductal gray region or the nucleus raphe magnus (Figure 24–15). In addition, bilateral transection of the dorsal lateral funiculus in the spinal cord blocks analgesia induced by central administration of morphine. Thus the central analgesic actions of morphine involve the activation of descending pathways to the spinal cord.

In the spinal cord, as elsewhere, morphine acts by mimicking the actions of endogenous opioid peptides. The superficial dorsal horn of the spinal cord contains interneurons that express enkephalin and dynorphin, and the terminals of these neurons lie close to synapses formed by nociceptive sensory neurons and spinal projection neurons (Figure 24–17A). Moreover, the μ, δ, and κ receptors are located on the terminals of the nociceptive sensory neurons as well as on the dendrites of dorsal horn neurons that receive afferent nociceptive input, thus placing endogenous opioid peptides in a strategic position to regulate sensory afferent input. The number of μ receptors on Aδ nociceptors, which mediate fast and acute pain or "first pain", exceeds that on C fiber nociceptors, which mediate slow persistent pain or "second pain" (Figure 24–1). This may help to explain why morphine is more effective in the treatment of persistent rather than acute pains.

Opiates regulate nociceptive transmission at synapses in the dorsal horn through two main mechanisms. First, they increase membrane K⁺ conductances in the dorsal horn neurons, hyperpolarizing the neurons and thus increasing the threshold for activation. Second, by binding to receptors on presynaptic sensory terminals opiates inhibit the release of neurotransmitter and thus decrease the extent of activation of the postsynaptic dorsal horn neurons. The decrease in neurotransmitter release appears to result from a decrease in Ca²⁺ conductance and consequent reduction of Ca²⁺ entry into the sensory nerve terminal (Figure 24–17B).

Many of the side effects of opiates are caused by the activation of opiate receptors within the brain and periphery. Opiate receptors are expressed by muscles of the bowel and anal sphincter; their activation contributes to constipation. The activation of opiate receptors in the nucleus of the solitary tract is responsible for respiratory depression and cardiovascular side effects. Confining drug administration to the spinal cord or to the periphery can minimize the side effects of systemic opiates. The release of endogenous opioid peptides from chromaffin cells of the adrenal medulla or from immune cells that migrate into injury sites may normally be involved in regulating the activation of nociceptors. The presence of peripheral opioid receptors has potential clinical relevance. Prolonged relief of pain after arthroscopic surgery can be achieved by injection of morphine into joints at doses that are ineffective when administered systemically. This peripheral route of opiate administration significantly reduces the side effects of opiate drugs.

Because the dorsal horn of the spinal cord has a high density of opioid receptors, morphine injected into the cerebrospinal fluid of the spinal cord subarachnoid space interacts with these receptors to elicit a profound and prolonged analgesia. Local administration of morphine is now commonly used in the treatment of postoperative pain, notably the pain associated with Caesarean section during childbirth. In addition to producing prolonged analgesia, intrathecal injection of morphine has fewer side effects because the drug does not diffuse far from its site of injection. Continuous local infusion of morphine to the spinal cord has also been used for the treatment of certain cancer pains.

Tolerance and Addiction to Opioids Are Distinct Phenomena

The chronic use of morphine invites major problems, most notably tolerance and addiction. The repeated use of morphine for pain relief can cause patients to develop resistance to the analgesic effects of the drug, with the consequence that progressively higher drug doses are required to achieve the same therapeutic effect.

What are the mechanisms underlying opiate tolerance? One theory holds that tolerance results from uncoupling of the opioid receptor from its G protein transducer. Nevertheless, the binding of naloxone to μ opiate receptors can precipitate withdrawal symptoms in tolerant subjects, suggesting that the opioid receptor is still active in the tolerant state. Tolerance may therefore also reflect a cellular response to the activation of opioid receptors, a response that counteracts the effects of the opiate and resets the system. Then, when the opiate is abruptly removed or naloxone is administered, this compensatory response is unmasked and withdrawal results.

Such physiological tolerance differs from addiction, a psychological craving for the drug. Psychological addiction almost never occurs when morphine is used to treat chronic pain.

Pain is a complex sensory state that reflects the integration of many sensory signals. More than most sensory modalities, its perception is influenced by emotional state and environmental contingency. Because pain is dependent on experience and varies so markedly from person to person, it remains notoriously difficult to treat.

Our understanding of the organization of central pain circuits, under both normal and pathological conditions, remains sadly incomplete. Nevertheless, over the past decade remarkable insights into the molecular mechanisms of peripheral pain transduction have developed, opening the way for new and more effective pain therapies.

First, human genetics and molecular biology have revealed that specific Na⁺ channels are expressed selectively by nociceptive sensory neurons, and that mutation in the genes encoding one of these channels results in congenital insensitivity to pain while preserving other somatosensory modalities. These findings have prompted the search for small-molecule chemicals that can block nociceptor-specific Na⁺ channels and thereby serve as selective peripheral analgesics. Second, the discovery that TRP channels encode sensory responses to a range of temperatures—from burning cold to scalding hot—and the knowledge that capsaicin and other small molecules activate these channels in a selective manner, has led to the development of many small-molecule TRP channel antagonists, some of which may prove to be effective in clinical pain syndromes.

How noxious mechanical stimuli are transduced remains a mystery, and defining the nature of the ion channels involved in the mechanotransduction of noxious stimuli remains an important challenge.

The classical finding that the balance of activity in small- and large-diameter sensory fibers modulates the perception of pain prompted the use of transcutaneous and dorsal-column electrical stimulation in the control of certain types of peripheral pain. This along with the observation that stimulation of specific sites in the brain stem produces profound analgesia has fostered efforts to control pain by activating endogenous modulatory systems. For example, the knowledge that opiates applied directly to the spinal cord elicit a potent analgesia led to the localized administration of opiates in certain clinical conditions by intrathecal and epidural routes.

Nevertheless, the uncertainties about the basic anatomy and functional organization of pain pathways mean that for most central pain syndromes there are still no effective pain therapies. Future progress in pain therapy will therefore depend on defining the logic of brain circuits that transmit nociceptive signals under normal and pathological conditions.