Molecular Biology of Sensory Transduction

SUMMARY

The perception of pain arising from a noxious stimulus starts with conversion of the energy of the stimulus into an electrical signal in the primary afferent neurons innervating the site of the stimulus. This process of energy conversion is called transduction. The three general modalities of noxious stimuli that impinge on the body are chemical, thermal, and mechanical, although each of these groups can be broken down into the specific nature of the stimulus, including the type of chemical, the temperature, or unique properties of the mechanical stimulus such as torque, sheer, or stretch. Specific proteins or groups of proteins called transducers underlie the process of transduction. There has been tremendous progress over the past decade in the identification and characterization of transducers responsive to all three modalities of stimuli. Our understanding of chemotransduction has progressed the farthest with the detailed maps that are now available for some chemotransducers of chemical binding sites and the conformational changes in protein structure with ligand binding. Putative transducers responsive to temperatures ranging from noxious cold to noxious hot have been identified, as have transducers responsive to a variety of mechanical stimuli, but it is now clear that still more of both types of transducers have yet to be identified. Although transduction of noxious stimuli was once thought to be an intrinsic property of nociceptive afferents, mounting evidence indicates that transduction also occurs in a variety of cells surrounding the afferent terminals; we are only just beginning to tease apart the impact of the interplay between these direct and indirect transduction processes. The critical interaction between transducers and the ion channels that control the excitability of afferent terminals has long been appreciated. However, the molecular identity of many of these channels has now been determined. Finally, despite evidence that there are still transducers to be identified, the contribution of many transducers to injury-induced changes in sensitivity has now been characterized. Advances on all three of these fronts have suggested novel approaches for the treatment of pain that are being actively pursued.

Introduction

Pain, a sensory and emotional experience , is in the brain. As discussed elsewhere in this textbook, there are clearly cases, such as stroke, where pain can originate from within the central nervous system. However, the vast majority of the pain that we experience, including chronic pain associated with peripheral nerve injury and inflammatory disorders, arises from activity in primary afferent neurons. Moreover, the vast majority of this activity is due to the impact of thermal, chemical, and/or mechanical stimuli. Afferent activity may arise spontaneously under pathological conditions as a result of changes in the relative balance of ionic currents in the membrane ( ; ), although even “spontaneous” activity may ultimately depend on membrane depolarization driven by a mechanical, thermal, or chemical stimulus impinging on the afferent, even though the source of the stimulus may not be readily apparent ( ). The focus of this chapter is on the mechanisms that enable thermal, mechanical, and chemical stimuli to initiate neural activity.

By definition, sensory transduction is conversion of the energy of a stimulus into an electrical signal. For the special senses (vision, audition, olfaction, taste), transduction occurs in specialized organs via cellular events specific to the stimuli associated with these senses. Sensory information arising from the body, referred to as somatosensation, may also involve specialized sense organs. For example, Golgi tendon organs and Meissner’s corpuscles are involved in the transduction of tension on tendons and low-threshold mechanical stimuli on glabrous skin, respectively. The primary afferents or sensory neurons innervating these structures tend to have rapidly conducting myelinated axons and anatomically distinct, specialized endings that often incorporate non-neuronal cells ( ). In contrast, afferents referred to as nociceptors respond to noxious or potentially tissue-damaging stimuli that are normally perceived as painful. Axons of these neurons tend to have slowly conducting unmyelinated (C fibers) or thinly myelinated (Aδ fibers) axons with peripheral terminals that are not associated with specific structures or cell types ( ). Thus, nociceptors are said to have free nerve endings. Although recent data, discussed below, have forced investigators to rethink the contribution of other cell types to sensory transduction, an important implication of the free nerve ending is that the molecular machinery necessary for transduction of noxious stimuli must be intrinsic to the nociceptive afferents. The subsequent demonstration that subpopulations of isolated sensory neurons are responsive to thermal (both hot and cold) ( ; ; ; ; ), mechanical ( ), and a variety of algogenic chemical stimuli ( ) is consistent with the idea that transduction is an intrinsic property of nociceptive afferents.

The available evidence indicates that the resting membrane potential of nociceptive afferents is negative to −40 mV, with values at the cell body ranging between −50 and −75 mV ( ). Evidence from study of the putative nociceptive afferent somata in vitro suggests that the action potential threshold is relatively high at greater than −35 mV ( ; ; ; ; ). Thus, because action potential generation is necessary for propagation of sensory information to the central nervous system, transduction of nociceptive stimuli must ultimately result in membrane depolarization. The membrane depolarization resulting from a transduction event is called a generator potential.

A generator potential can be initiated in three primary ways. The first and most direct way involves the opening of an ion channel with an ion permeability ratio such that the equilibrium potential for the net charge movement through the channel is depolarized to the action potential threshold. In this case, sufficient activation of this ion channel will drive the membrane potential above threshold and thereby result in an action potential that can be propagated toward the central nervous system. Activation of the transient receptor potential vanilloid type 1 (TRPV1) channel is an example of such a transduction mechanism. As discussed below, the TRPV1 channel is activated or opened by thermal (heat) stimuli ( ), as well as by a variety of chemical stimuli ( ). It is a non-selective cation channel that is permeable to Ca ²⁺ , Na ⁺ , and K ⁺ such that the equilibrium potential, or the potential at which there is no net flux of charge, is approximately 0 mV for this channel. Because this equilibrium potential is above the action potential threshold, activation of enough TRPV1 channels can ultimately result in action potential generation ( Fig. 2-1 ).

Figure 2-1, At least three mechanisms underlie the initiation of a generator potential.

A second mechanism underlying the generator potential involves the closing of a channel responsible for a hyperpolarizing current. K ⁺ channels are the only channels capable of contributing such a current in nociceptive afferents. This is because of the distribution of ions inside and outside nociceptive afferents. That is, interstitial fluid has a relatively high concentration of Na ⁺ , Ca ²⁺ , and Cl ⁻ and a low concentration of K ⁺ . In contrast, in nociceptive afferents, the intracellular concentration of K ⁺ is high, that of Cl ⁻ is relatively high ( ), and that of Na ⁺ and Ca ²⁺ is low. Closing of a K ⁺ current is clearly an indirect mechanism of sensory transduction since it will result in a generator potential only if a resting depolarizing current is simultaneously active with the hyperpolarizing K ⁺ current. Relatively high K ⁺ conductance in the face of relatively low Na ⁺ conductance will still enable the neuron to maintain a resting membrane potential in the expected range. If the decrease in K ⁺ channels is sufficient in such a neuron, the result will be a generator potential capable of driving the membrane potential above the action potential threshold (see Fig. 2-1 ).

The third mechanism underlying a generator potential is also indirect, but in contrast to the second mechanism, it is dependent on a relatively close association between an ion channel capable of driving membrane depolarization and a low-threshold voltage-gated ion channel capable of pushing the membrane potential above the action potential threshold. That is not to say that the localization of ion channels is not critical for the ultimate success (i.e., generation of action potentials) of transduction via all three mechanisms, but this is particularly true for Cl ⁻ because of the unique regulation of Cl ⁻ in sensory neurons. There is evidence in some neurons that the concentration of intracellular Cl ⁻ may be high enough that the Cl ⁻ equilibrium potential is above the action potential threshold. Consequently, activation of a Cl ⁻ channel in these neurons, such as bradykinin-induced activation of the Ca ²⁺ -dependent Cl ⁻ channel TMEM16, may result in a generator potential sufficient for generation of an action potential ( ). Though depolarized relative to the resting membrane potential in almost all sensory neurons, the Cl ⁻ equilibrium potential is still below the action potential threshold in many sensory neurons. However, if there are low-threshold voltage-activated channels such as the T-type Ca ²⁺ channel Ca _v 3.2 in close association with Cl ⁻ channels, activation of a Cl ⁻ channel may still be sufficient for action potential generation even if the Cl ⁻ equilibrium potential is below the action potential threshold (see Fig. 2-1 ).

Many chemical stimuli act on the G protein–coupled receptors (GPCRs) expressed on nociceptors. In these cases, as discussed below, subsequent intracellular signaling cascades are needed to modify ion channel activity and drive initiation of the generator potential.

Chemo-, Thermo-, and Mechanotransducers

Transducers are often categorized according to the stimuli to which they are responsive. This is a useful way to think about transducers, particularly in the context of a particular type of pain or altered sensitivity such as cold allodynia or heat hyperalgesia, because it is reasonable to assume that these “types” of pain are due to afferent activity evoked with a specific stimulus. However, many, if not most of the putative thermo- and mechanotransducers respond to more than one stimulus modality and are therefore said to be polymodal. Given evidence that a variety of transducers are present and functional in non-neural tissue, it is also reasonable to categorize transducers according to whether they are intrinsic or extrinsic to the primary afferent. Furthermore, given evidence that at least one putative transducer (TRPV1) may be present and functional on subcellular organelles ( ), it is at least worth considering transducer localization despite evidence that the vast majority of transducers are membrane bound. Nociceptive afferents and consequently transducers are present throughout the body. Although a pinprick and noxious stretch are both mechanical stimuli, visceral afferents such as those innervating the colon are far more sensitive to stretch (i.e., colon distention) than other forms of mechanical stimuli ( ), whereas pinprick is a highly effective stimulus for activating nociceptive afferents innervating the skin ( ). Consequently, it is important to consider the nature of the stimulus and the tissue being affected. Finally, even though a number of chemotransducers are activated by noxious chemicals in the environment, the majority, if not all, are responsive to endogenous chemicals. Therefore, it is also important to consider the source of the stimulus.

Chemotransducers

Of the three primary modalities of somatosensory stimuli, the process of chemotransduction is the most well understood. Specificity for one chemical over another is achieved through binding sites in the transducer that are unique, or at least relatively so, for a particular chemical. In the most direct form of chemotransduction, the transducer has a binding site, or receptor, for the chemical stimulus and is also an ion channel. Binding of the chemical to the receptor drives a conformational change in the transducer protein that opens the ion channel (e.g., see ). Thus, these transducers are also referred to as ionotropic receptors ( Fig. 2-2 ). This is the most rapid form of chemical transduction, with signaling possible on the microsecond time scale. There is also an indirect form of chemotransduction whereby the conformational change in the transducer driven by chemical binding results in the activation of an intracellular signaling cascade. These transducers are referred to as metabotropic receptors (see Fig. 2-2 ). This form of chemical transduction is slower and occurs on a time scale of milliseconds to minutes. Guanine nucleotide–binding proteins, or GPCR receptors, are by far the most common type of metabotropic receptor, with the type of G protein being responsible for both initiation of the cellular signaling cascade and the type of cascade initiated ( ). Additional metabotropic receptors found in sensory neurons include receptors bearing intrinsic protein tyrosine kinase domains (i.e., Trk receptors), receptors that associate with cytosolic tyrosine kinases (i.e., non–tyrosine kinase receptors such as cytokine receptors, integrins), and protein serine/threonine kinases (i.e., transforming growth factor-α [TGF-α] receptors) (see for review). This second form of signaling is very widespread and responsible for changes in the regulation of a variety of cellular processes, including ion channel properties ( ), cellular properties such as the regulation of intracellular Ca ²⁺ ( ) and neurite extension ( ), and gene expression ( ). Second-messenger signaling is complex with multiple points of convergence and interaction ( ). However, to keep this chapter tractable, I will consider only metabotropic receptor–mediated transduction events that are coupled to an ion channel that may initiate a generator potential.

Figure 2-2, Ionotropic receptors are directly coupled to ion channels.

At least three additional factors have an impact on the efficacy of chemoreceptor signaling. First, the spatial distribution of transducers relative to other ion channels in the membrane, particularly those responsible for action potential initiation, is a critical determinant of whether a generator potential will result in an action potential. To add to the complexity of metabotropic receptor signaling, there is evidence that second-messenger coupling is dynamic and changes in response to a number of different conditions, including hormonal status ( ) and history of prior stimulation ( ). Furthermore, it is dependent not only on the appropriate localization of transducers and targeted ion channel but also on the appropriate cellular signaling machinery. Consequently, metabotropic receptor activation may not always result in a generator potential. Second, allosteric modulation of chemotransducers, where a second chemical binding site is located at a site different from that of the chemical that activates the receptor, is common and can result in profound changes in chemotransducer activity. An extreme example is the N -methyl- d -aspartate (NMDA) type of ionotropic glutamate receptor: for glutamate to activate the receptor, it must also be bound to glycine ( ). Third, a number of chemotransducers may be activated by several distinct chemicals. In the case of TRPV1, a transducer activated by protons and capsaicin (the pungent component of chili peppers), the binding sites or receptors for these compounds are on distinct parts of the protein ( ).

Ionotropic Receptor Families

Acid-Sensing Ion Channels

Ionotropic receptors are generally classified according to their structure and genetic homology. The acid-sensing ion channels (ASICs) are, as their name implies, activated by extracellular protons, although this is true for only three of the four genes encoding ASIC channel subunits (ASIC1, 3, and 4) since ASIC2 does not appear to be activated by protons ( ). All four subunits are detected in sensory neurons ( ), including the two splice variants of ASIC1 (1a and 1b) and ASIC2 (2a and 2b). The ASICs are trimeric proteins with homology to the epithelial Na ⁺ channel (ENaC)/degenerin family that can form homomeric or heteromeric channels ( ). ASIC3 was originally thought to be specific to dorsal root and trigeminal ganglion neurons ( ), where it is enriched in nociceptive afferents, but it has subsequently been shown to be more widely expressed ( ). It is the most sensitive to protons—activated by a decrease of less than 0.2 pH units—and this sensitivity is dramatically enhanced with lactate ( ). Intense muscle use produces lactic acidas, a metabolic produce that is thought to contribute to exercise and ischemic muscle pain. Thererfore, ASIC3 is one of several transducers present in muscle afferents that may also be referred to as metaboreceptors ( ). ASIC3 is enriched in specific subpopulations of afferents, including those innervating the heart ( ) and dura ( ), where it has been suggested to contribute to the pain associated with coronary ischemia and migraine, respectively. Evidence from null mutant mice suggests that ASIC3 contributes to hyperalgesia of an inflamed muscle (primary hyperalgesia) whereas ASIC1 contributes to the hyperalgesia observed at sites distant from the inflamed muscle (secondary hyperalgesia) ( ), although the details underlying the basis for this distinct pattern have yet to be fully clarified. Recent data indicating that venom from the western coral snake can directly activate ASIC1 ( ) further support a role for this subunit in the response to pain-producing stimuli and suggest that these channels are responsive to a wider variety of stimuli than originally thought. The observation that the pH sensitivity of ASIC2a was dramatically increased by coral snake venom also led to the suggestion that this subunit may function as a coincidence detector for as yet to be identified compounds. Along these same lines, recent evidence suggests that ASIC3 and P2X ₅ may act as coincidence detectors since binding of adenosine 5′-triphosphate (ATP) to P2X ₅ dramatically increases the sensitivity of ASIC3 to protons ( ).

Cys-Loop Family of Ligand-Gated Ion Channels

ACh/5-HT ₃ /GABA _A

One of the larger families of ligand-gated ion channels is the Cys-loop family, so named because of the characteristic loop in the extracellular N-terminal domain of the α subunit formed by a disulfide bond between two cysteine (Cys) residues ( ). Members of three of the four major subfamilies of Cys-loop receptors are present in sensory neurons. These include nicotinic acetylcholine receptors (nAChRs), serotonin type 3 (5-HT ₃ ) receptors, and type A γ-aminobutyric acid (GABA) receptors (GABA _A ). All Cys-loop receptors contain five subunits, which may be homomeric or heteromeric, depending on the receptor subtype and subunit composition.

Both homomeric (α ₇ ; ) and heteromeric (α _3–6 , β _2–4 ; ) nAChRs composed of two α and three β subunits are present in sensory neurons, with evidence that both forms contribute to nociceptive processing ( ). Although there are several potential non-neural sources of acetylcholine (ACh) in the periphery, it is important to note that ACh levels are increased in inflamed tissue ( ). Furthermore, peripheral administration of ACh- or nAChR-selective agonists results in burning pain (although recent evidence suggests that at least some of the pain associated with nicotine may be due to activation of TRPA1 ( ). Interestingly, recent evidence also suggests that the relative contribution of nAChR subtypes to peripheral pain may depend on the target of innervation. That is, α ₇ nAChRs appear to suppress activity in colonic dorsal root ganglion (DRG) neurons ( ), whereas the burning pain associated with the application of nAChR agonists to the skin is probably mediated by heteromeric receptors ( ).

The 5-HT ₃ receptor is the only ionotropic serotonin receptor. It may exist as a homomeric receptor composed of five 5-HT _3A subunits or as a heteromeric receptor composed of a 5-HT _3A subunit with one of the other four 5-HT ₃ subunits (5-HT _3B–3E ) ( ). It was originally thought to have a relatively limited role in nociception because of its expression pattern in afferents with a medium to large cell body diameter ( ). Subsequent analysis, however, suggested that this receptor is critical for both serotonin-induced pain and, more importantly, full expression of the second phase of pain behavior in the formalin test ( ), a behavior thought to reflect spontaneous inflammatory pain. Given that the gastrointestinal (GI) tract is the largest source of serotonin in the body, it should not be surprising that this receptor plays an even more important role in visceral pain. In fact, several 5-HT ₃ receptor antagonists have been approved for use in treating the pain associated with irritable bowel and other visceral pain syndromes ( ).

Unlike the nAChRs and 5-HT ₃ receptors, which have ion channels permeable to Na ⁺ , K ⁺ , and to varying degrees, Ca ²⁺ , GABA _A receptors have an ion pore selective for anions (primarily Cl ⁻ and to a less degree HCO ₃ ⁻ ) ( ). Like nAChRs and 5-HT ₃ receptors, GABA _A receptors may exist as homomeric or heteromeric proteins. The homomeric proteins are composed of one of the three ρ subunits (originally called GABA _C receptors) and have been most studied extensively in the retina ( ). Recent evidence suggests that ρ subunit–containing receptors are present in sensory neurons. However, heteromeric GABA _A receptors are far more common throughout the nervous system, including sensory neurons. Although 19 subunits of the GABA _A receptor have been identified, the number of possible receptors is constrained by a subunit stoichiometry that consists of two α subunits, two β subunits, and one of the remaining seven non-ρ subunits. GABA _A receptor subunit composition determines the pharmacological and biophysical properties of the receptor, as well as its cellular distribution ( ).

Given that GABA _A receptors underlie the vast majority of fast inhibitory synaptic transmission in the adult central nervous system, it may be surprising to find these receptors among a list of chemotransducers. However, because, as noted above, the concentration of intracellular Cl ⁻ is maintained at levels considerably higher in nociceptive afferents than in neurons in the central nervous system ( ), GABA _A receptor activation results in membrane depolarization in nociceptive afferents ( ), which may still be inhibitory in the absence of tissue injury. However, following tissue injury, GABA _A receptor activation may result in action potential generation in nociceptive afferents ( ). This can occur at the central terminals of nociceptive afferents, a process referred to as the dorsal root reflex. More importantly in the context of the present discussion, GABA _A receptor activation in the periphery can excite nociceptive afferents ( ). It should be noted that although there do not appear to be sources of peripheral GABA necessary to achieve the concentrations needed for activation of low-affinity synaptic receptors, recent evidence suggests that high-affinity receptors, which may be activated by GABA concentrations close to the resting levels observed in extracellular fluid, are present in sensory neurons ( ). Thus, even though there is evidence that a shift in GABA _A signaling in the spinal cord contributes to both the initiation and maintenance of inflammatory hypersensitivity, high-affinity GABA _A receptors in the periphery may also contribute to ongoing afferent drive in the presence of inflammation.

Glutamate Receptors

Ionotropic glutamate receptors play a critical role in fast excitatory synaptic transmission in the central nervous system and are therefore generally studied in the context of synaptic transmission. These receptors have historically been classified according to their response to three selective agonists: α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), kainate, and NMDA ( ). Subsequent analysis has indicated that these receptors are composed of heteromeric combinations of four subunits, with AMPA receptors being composed of GluR1–4, kainate receptors being composed of GluR5–7 (GluK1–3) and/or KA1–2 (GluK4–5), and NMDA receptors being composed of a combination of NR1, NR2A–D, and/or NR3A–B. All three types of receptors are present and functional in primary afferent neurons ( ). Interestingly, even though these receptors are present on the central terminals of primary afferents, where they appear to contribute to synaptic transmission in the spinal cord, they are also present in the periphery, where they can drive action potential generation ( ; ) and pain ( ). There are a number of potential sources of peripheral glutamate, including the afferent itself, where peripherally released glutamate could serve as a form of feedback excitation to amplify injury-induced activation of nociceptive afferents.

HCN Channels

Hyperpolarization-activated, cyclic nucleotide–gated (HCN) ion channels are not typical chemotransducers since they are not directly activated by extracellular ligand binding ( ). Furthermore, because these non-selective cation channels are activated by membrane hyperpolarization and close with membrane depolarization, their biophysical properties argue against a role in the initiation of a generator potential capable of driving the membrane above the action potential threshold. Under resting conditions, these channels are activated only during the increase in membrane potential associated with the after-hyperpolarization that follows an action potential, where they provide a depolarizing drive for subsequent spike initiation ( ). There are four HCN family members (HCN1–4), which consistent with their homology to Kv family members, form homomeric tetramers ( ). mRNA for all four HCN channels is detectable in sensory neurons, with HCN1 being differentially distributed in large- and small-diameter neurons and HCN3 enriched in small-diameter neurons ( ).

These channels have been included in the list of chemotransducers for several reasons. The voltage dependence of channel activation is regulated by intracellular cyclic adenosine monophosphate (cAMP) such that an increase in cAMP drives a depolarizing shift in the voltage dependence of channel activation. This shift can be sufficiently large that HCN channels contribute to the depolarization of resting membrane potential and, more relevantly, to a depolarizing drive that facilitates action potential generation ( ). The channel is also a putative target for a number of metabotropic receptors coupled to second-messenger pathways that result in an increase in cAMP. There is an increase in HCN current density in large-diameter neurons following traumatic peripheral nerve injury, where the increase appears to be responsible for ectopic activity ( ). Recent evidence from null mutant mice suggests that HCN2 in Na _v 1.8-expressing neurons plays a particularly important role in the generation of inflammatory thermal hyperalgesia, as well as peripheral nerve injury–induced thermal and mechanical hypersensitivity ( ).

Purinergic Receptors—P2X

Ionotropic purinergic, or P2X, receptors are activated by ATP. The functional receptor is a trimer. Seven subunits have been identified, P2X _1–7 ; all but P2X ₆ can form functional homomers, and each appears to be able to form a heteromeric protein with at least one other subunit ( ). At least six, if not all seven, subunits are present in sensory neurons, with all but P2X ₄ being differentially distributed among subpopulations of DRG neurons ( ). P2X ₃ , which forms a heteromer with P2X ₂ , has been the most extensively studied member of this family with respect to nociceptor activation ( ). This subunit was originally shown to be enriched in a subpopulation of neurons with a small cell body diameter that did not express the neuropeptides substance P or calcitonin gene–related peptide (CGRP) ( ), the so-called non-peptidergic afferents. Although the distribution of P2X ₃ receptors among specific subpopulations of sensory neurons was subsequently shown to depend on the target of innervation (e.g., see ), pharmacological and molecular biological analysis confirmed that this subunit plays a dominant role in mediating the nociceptive response to the application of ATP, as well as the response to a variety of noxious stimuli applied to various tissues ( ). Interestingly, this subunit appears to mediate the activation of nociceptive afferents observed following damage to neighboring cells ( ), thus making this transducer a critical player in the initial pain observed in response to tissue injury. As noted above, more recent work has highlighted a potential role for P2X ₅ in the sensitization of ASIC currents in sensory neurons and consequently the pain associated with muscle ischemia ( ).

Transient Receptor Potential Channels

The TRP channels involved in chemotransduction come from a large family of ion channels that encompass eight subfamilies ranging from TRPA (for ankyrin) to TRPV (for vanilloid), with TRPC (for canonical), TRPM (for melastatin), TRPML (for mucolipins), TRPP (for polycystins), and TRPN (for NO-mechanopotential C) in between ( ). Unifying features of this family include a channel protein formed from four subunits, each with a structure analogous to that for voltage-gated K ⁺ ion channel subunits with a six-transmembrane segment and a pore loop between transmembrane segments 5 and 6. Specific to TRP channels are ankyrin domains on the intracellular N-terminus and proline-rich domains on the intracellular C-terminus. All TRP family members form Ca ²⁺ -permeable cation channels. Many of the family subunits have a voltage sensor in segment 4 and consequently exhibit voltage-sensitive properties. As discussed below, several of the channels are activated by different stimulus modalities, with the biophysical properties of the channel activity appearing to depend on how the channel is activated. For example, capsaicin-induced activation of TRPV1 desensitizes in the presence of extracellular Ca ²⁺ , whereas heat-evoked activation of TRPV1 does not ( ). The channels also appear to be used differentially across phyla. For example, TRPA1 functions as a cold receptor among other modalities of transduction in mammals ( ) but underlies infrared detection in snakes ( ). The relative contribution of various TRP channels to nociceptive processing is an active area of investigation.

TRPA1

TRPA1 was originally identified through a combined bioinformatic and expression strategy designed to identify additional TRP family members ( ). It is the only member of its subfamily defined by the exceptional number (14) of N-terminal ankyrin repeats. Although TRPA1 was first thought to function primarily as a cold transducer (see below), subsequent analysis has indicated that TRPA1 is responsive to a variety of noxious compounds. This list first included allyl isothiocyanates (i.e., mustard oil) and cannabinoids ( ) and was rapidly expanded to include the pungent ingredients of garlic ( ), cinnamon, wintergreen oil, clove oil, and ginger ( ), as well as a variety of environmental irritants such as acrolein, CO ₂ , and formalin ( ). TRPA1 is also activated by a number of endogenous mediators, including products of oxidative stress ( ) and cyclooxygenase-dependent fatty acid metabolites ( ). TRPA1 appears to be expressed in a subset of TRPV1-expressing neurons and, like TRPV1, appears to be a target of endogenous algogenic compounds such as bradykinin ( ). There is also evidence that TRPA1 may interact with TRPV1 at the level of a complex ( ). As a result, activation of one channel influences the response resulting from activation of the other. Interestingly, despite its role in mediating the acute painful response to a variety of chemicals, a gain-of-function mutation in TRPA1 is not associated with ongoing pain. Rather, individuals with the TRPA1 mutation experience episodic upper body pain triggered by fasting or physical stress ( ).

TRPM2

TRPM2 is widely expressed throughout the body but is particularly enriched in immune cells such as macrophages ( ). It was originally suspected that TRPM2 would have enzymatic properties in addition to its putative function as a Ca ²⁺ channel based on the presence of a Nudix box on its C-terminus, a motif common to enzymes, particularly those that degrade nucleoside diphosphates ( ). It is likely that this motif serves as a binding site for adenosine diphosphate (ADP)-ribose, which was subsequently shown to activate the channel. It was soon realized, however, that TRPM2 plays a significant role in mediating the cellular response to stress since it is activated by reactive oxygen species such as hydrogen peroxide ( ). The channel has recently been shown to be present in sensory neurons, where it enables sensitivity to hydrogen peroxide ( ). Recent data indicate that the channel plays a prominent role in facilitating the inflammatory response to infection ( ) and is likely to contribute to inflammatory hypersensitivity. However, though present in sensory neurons ( ), the pro-nociceptive role of this channel is likely to be indirect via facilitation of the release of chemokines such as CXCL2 from immune cells and microglia.

TRPM3

TRPM3 is another member of the melastatin subfamily of TRP receptors implicated in nociception. This TRP channel is expressed in a variety of different tissues, including a relatively broad population of sensory neurons ( ). The most potent known agonist for TRMP3 is pregnenolone sulfate. Although this compound acts at a number of receptors and ion channels, it has been shown to induce nociceptive behavior when administered to mice, behavior that is eliminated in TRPM3 null mutant mice ( ).

TRPM8

The discovery of TRPM8 was reported almost simultaneously by two different groups that used complementary strategies of bioinformatics ( ) and expression cloning ( ) to identify novel TRP channels. This channel was shown to be responsive to both cooling and menthol. The channel is present in a small subpopulation of non-peptidergic afferents that does not overlap with TRPV1/TRPA1-expressing neurons ( ). Because the sensation of menthol is not usually described as painful, this transduction mechanism would generally be grouped with those associated with the transduction of non-noxious stimuli. It is included here for the sake of completeness and, as noted below, because of evidence that the channel contributes to the perception of cold pain ( ). Interestingly, recent genome-wide association studies have linked TRPM8 to migraine without aura (ref PMID:21666692 and 22683712), although it remains to be determined how a polymorphism in this channel contributes to the increased risk for the presence of migraine.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here