|Home | About | Journals | Submit | Contact Us | Français|
Acute pain detection is vital to navigate and survive in one’s environment. Protection and preservation occur because primary afferent nociceptors transduce adverse environmental stimuli into electrical impulses that are transmitted to and interpreted within high levels of the central nervous system. Therefore, it is critical that the molecular mechanisms that convert noxious information into neural signals be identified and their specific functional roles delineated in both acute and chronic pain settings. The Transient Receptor Potential (TRP) channel family member TRP ankyrin 1 (TRPA1) is an excellent candidate molecule to explore and intricately understand how single channel properties can tailor behavioral nociceptive responses. TRPA1 appears to dynamically respond to an amazingly wide range of diverse stimuli that include apparently unrelated modalities such as mechanical, chemical and thermal stimuli that activate somatosensory neurons. How such dissimilar stimuli activate TRPA1, yet result in modality-specific signals to the CNS is unclear. Furthermore, TRPA1 is also involved in persistent to chronic painful states such as inflammation, neuropathic pain, diabetes, fibromyalgia, bronchitis and emphysema. Yet how TRPA1’s role changes from an acute sensor of physical stimuli to its contribution to these diseases that are concomitant with implacable, chronic pain is unknown. TRPA1’s involvement in the nociceptive machinery that relays the adverse stimuli during painful disease states is of considerable interest for drug delivery and design by many pharmaceutical entities. In this review, we will assess the current knowledge base of TRPA1 in acute nociception and persistent inflammatory pain states, and explore its potential as a therapeutic pharmacological target in chronic pervasive conditions such neuropathic pain, persistent inflammation and diabetes.
Pain is a percept that results from somatosensory stimuli that trigger a cascade of adaptive responses in the body. An individual’s ability to distinguish and interpret painful signals is a unique culmination of neural coding that requires many levels of processing. In primary sensory neurons, there is a complex interaction of numerous molecular sensors within a given neuron, and there are intricate patterns of activated and silent afferent fibers converging on spinal cord interneurons and projection neurons. Together, these processes convey encoded nociceptive information to pathways of ascending hierarchical processing in the spinal cord and brain. Moreover, descending pathways from the brainstem, cortex or other CNS regions affect pain processing by facilitating or inhibiting the information ascending in nociceptive pathways. Here we will focus on the molecular sensors in primary afferent neurons where nociception is initiated and driven for several reasons: 1) peripheral receptors and ion channels are relatively accessible to pharmacological manipulation, 2) therapeutic compounds can be designed to be peripherally-restricted, thereby reducing side effects of compounds acting at CNS sites, and 3) a focus on the primary afferent input should reduce much of the inter-individual variability that is contributed by many complex higher levels of processing. The focus on peripheral transduction should make it more feasible to identify and define targets for novel pharmacological therapies that have broader treatment potential across individual patients and across multiple diseases.
Molecular sensors in the plasma membrane of primary sensory neurons initially detect noxious stimuli. These sensors respond to minute, local changes in concentrations of chemicals, pH, pressure, or temperature. Here we focus on one class of molecular sensors, the Transient Receptor Potential (TRP) channels that are intimately involved in the initial transduction of sensory stimuli into generator potentials that depolarize the sensory nerve membrane. All TRP channels are six transmembrane-spanning proteins with varying sizes of cytosolic N- and C-termini. Based on the structure of TRP Vanilloid 1 (TRPV1), revealed by electron cryomicroscopy 1, TRP channels are thought to exist as tetramers where four subunits are arranged around a pore region that is generally permeable to cations including Ca2+ and Na+. Structural analysis of TRPV1 indicates that a large cytoplasmic region containing the N- and C-termini hangs like a basket underneath the smaller, more compact transmembrane domain of the channel 1. At least 33 different TRP channels have been identified in mammals 2. Several of these have been localized to primary afferent sensory neurons and shown to be essential for temperature and pain sensation, including TRP Ankyrin 1 (TRPA1), TRP Melastatin 8 (TRPM8) and TRPV1.
Transient Receptor Potential Ankyrin 1 (TRPA1) is the only member of the ankyrin subfamily found in mammals. Originally called ANKTM1, TRPA1 was identified by a homology search for ankyrin domains and six transmembrane domains and is approximately 20% homologous at the amino acid level to TRPV1 3, 4. Like other TRP family members, the TRPA1 gene encodes a protein that contains 6 transmembrane domains. However, TRPA1 is unique in that it has a particularly long (14–18) ankyrin repeat region in the N-terminus 4, 5, and contains calcium-sensitive regions in the N-terminal EF-hand motif 6–8 and in the S4 transmembrane segment 9. The flexible ankyrin domains may provide scaffolds for protein-protein interactions 10. It is generally assumed that TRPA1 subunits are assembled into homotetrameric complexes on the cell membrane. However, TRPA1 subunits also co-localize with TRPV1 in heterologous cells 11 and may assemble into heterotetrameric complexes to tailor the single channel biophysical properties in native sensory neurons 12.
Similar to its pioneer relative TRPV1, TRPA1 appears to respond to a wide variety of diverse exogenous and endogenous chemical stimuli, as well as physical stimuli that include cold temperatures and mechanical stretch or force. How these dissimilar stimuli are encoded by a sensory neuron to nonetheless convey specific modalities to the CNS is puzzling and of significant interest, given the intense focus by many pharmaceutical companies on TRPA1’s potential as a target for novel pain therapies for broad indications. In this review, we will explore recent advancements in the understanding of TRPA1 channel function and its roles in tissue injury, nerve injury and disease.
TRPA1 was initially identified in human fibroblasts 3. Subsequently, TRPA1 was shown to be predominantly expressed in small- to medium-diameter nociceptive sensory neurons of the dorsal root ganglia (DRG) or trigeminal ganglion neurons that also express the capsaicin receptor TRPV1, suggesting it has an important role in nociception 4, 13, 14. As studies broadened and detection techniques became more sensitive, low expression levels of TRPA1 were found in other tissues, including inner hair cells 15, keratinocytes 16–20, cerebellar neurons 21, smooth muscle 22, gastrointestinal sensory neurons 23, 24, and somatosensory DRG neurons of all diameters 19. The fact that TRPA1 is so broadly expressed over a wide range of tissues suggests that this channel plays diverse functional roles. Thus, it is important to understand how TRPA1 expression is regulated in a tissue-specific manner, and the molecular basis of how TRPA1 protein is regulated in different tissues to mediate specific sensory modalities.
The TRPA1 channel is a multimodal receptor whereby diverse chemical stimuli activate nociceptive neurons via a single protein. Pungent or irritating chemicals such as mustard oil (allyl isothiocyanate; AITC), cinnamaldehyde, gas exhaust (acrolein), raw garlic and onions (allicin), formalin (formaldehyde) and heavy metals (zinc, cadmium) all directly activate TRPA1, and each of these elicit a painful burning or prickling sensation 4, 13, 14, 25–29. These electrophilic compounds directly activate or sensitize TRPA1 by covalently modifying cysteines and lysines in the N-terminus and transmembrane domains 30–33. Recent data also indicate that TRPA1 is mildly sensitive to the stimulant molecules caffeine (that many of us essentially rely on) 34 and nicotine 35, which are proposed to act through Ca2+-dependent pathways.
More physiologically, TRPA1 is now known to be activated by an increasing number of identified endogenous chemicals, including hydrogen peroxide (H2O2), hydrogen sulfide, 4-hydroxy-2-nonenal and the conventionally anti-inflammatory agent 15-deoxy-Δ12,14 prostaglandin J2, which are α,β unsaturated aldehydes generated via peroxidation or dehydration of lipid second messengers 33, 36–39. Other endogenous chemical compounds appear to drive metabotropic receptor-coupled molecular signaling pathways to sensitize downstream targets that include TRPA1. Namely, the inflammatory mediator bradykinin indirectly activates TRPA1 in nociceptive afferents via phospholipace C (PLC) and protein kinase A (PKA) signaling pathways as well as increasing intracellular Ca2+ 14, 25, 40. Another pro-inflammatory pain molecule, endothelin-1 (ET-1), sensitizes TRPA1 via a protein kinase C (PKC) signaling pathway 41.
The fact that TRPA1 itself is a Ca2+ permeable channel and that TRPA1 is directly activated by Ca2+ via an intracellular Ca2+-binding domain in the N-terminus 7, 8 and an extracellular Ca2+-binding domain in S4 region 9, open up the possibility that very broad classes of compounds and intracellular mediators may target TRPA1 in a “non-specific” manner. Therefore, many exogenous and endogenous chemical compounds may ultimately activate or sensitize TRPA1 downstream and thereby, contribute to hypersensitivity during tissue damage from external insults or the internal inflammatory responses that follow tissue injury.
A current spotlight in the somatosensory field is focused on exposing the mechanisms underlying our skin’s sensation of cool and cold stimuli and how severe cold elicits a frankly painful response. Small and medium-diameter sensory neurons, which include unmyelinated (C fiber) and thinly-myelinated (Aδ fiber) peripheral afferents, are considered essential for conveying signals about cool and cold external stimuli to the spinal cord and higher CNS 42, 43. The TRPA1 relative, menthol-sensitive channel TRPM8, is expressed on C and Aδ fiber primary afferents and was proposed to be the primary cold transducer at ambient temperatures below 26°C 44. Indeed, three lines of TRPM8-deficient animals confirm that TRPM8 is a principal transducer for temperatures over a broad range from mild cooling to intense cold (30–10 °C) 45–47. However, TRPM8-deficient mice still exhibit behavioral responses to noxious cold and have some nociceptors that respond to intense cold 45, indicating that other cold-responsive channels must exist.
TRPA1 has been vigorously debated as a noxious cold sensor since it was first characterized as cold-responsive in Chinese hamster ovary (CHO) cells by Story and colleagues in 2003 4. Some studies have described TRPA1 to be directly activated by cold in the noxious range (~17 °C) when expressed in heterologous systems including CHO or human embryonic kidney (HEK) cells 4, 14, 48. However, other laboratories found that exogenously-expressed TRPA1 is not directly activated by noxious cold 13, 25, 49. Studies that have examined the cold-sensitive properties of TRPA1 in its native milieu in somatosensory neurons have suggested that the cellular context surrounding TRPA1 is important. Neither teased C fiber or Aδ fiber afferents, nor isolated DRG or trigeminal small neurons showed any deficits in response to short-duration cold stimuli 19, 25. Behavioral studies revealed similarly disparate results as one study found TRPA1-deficient mice to have cold deficits in female mice 50, while others found no cold or cooling deficits in male mice 25. A recent study found that TRPA1-deficient mice exhibited no difference in shivering or acute paw withdrawal latencies in male or female mice, but importantly, found that sustained exposure of mouse paws to intense cold surfaces revealed defects in jumping responses in TRPA1-deficient mice 51. One plausible explanation is that the extended cold exposure induces damage and tissue injury, activating bradykinin signaling and other inflammatory mediators that ultimately increase intracellular Ca2+ that activates TRPA1. TRPA1 is involved in inflammation-induced cold hypersensitivity (below) 52, 53. Thus, TRPA1 may indeed be essential for cold-induced sensitization to cold stimuli. Together, these studies highlight an essential point that should be generally borne in mind when interpreting physiological results derived for any receptor or ion channel: the context surrounding TRPA1 critically important, both on a membrane milieu level and a complex tissue setting level. The possibility that gender-related factors could alter TRPA1 signaling on either of these levels should also be considered.
Mechanoreceptors provide essential information about internal organ distension, muscle contraction and tactile sensory stimuli skin. In skin, changes in the nature, intensity, duration and location of mechanical stimuli are sensed by peripheral receptors that are uniquely tuned to very different qualities of mechanical force: soft brush, massage stroke, pinch, needle stick, intense pressure, vibration and hair follicle movement from a light breeze or mosquito landing. We know that in general, light touch, vibration and joint position are detected by primary afferents with myelinated Aβ axons, whereas pinch and prick are quickly detected by Aδ mechanonociceptors, and intense painful pressure, particularly after tissue injury, is sensed by unmyelinated C fiber nociceptors. But exactly how these different mechanoreceptors detect and transduce specific mechanical stimuli into receptor potentials is not known. While this is amazing considering the frequency with which we utilize these senses (consciously and unconsciously) on a daily basis for a wide range of activities, it is understandable when we consider several important factors: 1) the fine mechanosensitive afferent terminals are rather sparsely distributed throughout the skin, 2) the ion channels and receptors on these mechanoreceptive terminals are furthermore expressed in low density in the sensory neuron plasma membrane, and 3) the cell bodies that synthesize these mechanoreceptive proteins are located up to a meter away from the skin in the dorsal root ganglia for the body or the trigeminal ganglion for the face.
TRPA1 is a candidate somatosensory mechanotransduction molecule. The long ankyrin repeats in the N-terminus of TRPA1 have been hypothesized to tether the channel to cytoskeletal proteins and potentially serve as a “spring” that pulls the channel open in response to extracellular force 5. This mechanotransduction role for TRPA1 is further supported by its expression patterns. The fact that it is highly expressed in C fiber and some Aδ- primary sensory neurons 4, 49, 54, 55 suggests it is in the correct location for cutaneous mechanotransduction. Recent studies indicate that its expression may be more broad, including many medium and large-diameter neurons which have Aδ and Aβ fiber axons 19. Within these fiber types, TRPA1 is located on the peripheral terminal endings of both unmyelinated C and myelinated Aβ fiber afferents in skin 19.
TRPA1 family channels have been proposed to contribute to mechanotransduction in several diverse species. Drosophila larvae deficient in a TRPA homologue called painless have decreased behavioral responses to intense mechanical stimuli75. C. elegans with mutations in a Trpa1 gene fail to show head withdrawal or stop feeding in response to nose touch93. In pig, mast cell-induced hypersensitivity to mechanical force in esophageal C-fibers is reduced following TRPA1 inhibition using the TRPA1 antagonist HC-03003156. Mice with a deletion of the pore domain of TRPA1 exhibit decreased behavioral responses to intense mechanical force in the noxious range 50, although behavioral mechanical deficits were not observed in another TRPA1 mutant mouse 25. In non-injured skin, TRPA1 is required for normal mechanical responses in several types of cutaneous afferents. Teased fiber recordings from TRPA1-deficient mice have reduced firing in C fiber nociceptors (all ranges of intensity), Aδ mechanonociceptors (only high intensity stimuli) and slowly adapting Aβ fibers (all ranges of intensity) 19. Acute pharmacological inhibition of TRPA1 in skin using HC-030031 similarly reduces cutaneous C fiber firing at all intensities 57. In visceral sensory neurons, TRPA1 is enriched in vagal, colonic and pelvic sensory neurons and their terminals, and afferents innervating each of these visceral regions show reduced mechanical firing in TRPA1-deficient mice 24. In the spinal cord, pharmacological blockade of TRPA1 using A-967079 disrupted synaptic transmission of high-intensity mechanical firing to nociceptive-specific and wide dynamic range spinal neurons in normal animals 58. Together, these data indicate that the TRPA1 channel is essential for normal mechanical firing of primary afferent neurons in cutaneous and visceral targets.
Is TRPA1 a mechanotransducer or part of a mechanotransduction complex? The answer is unclear. However, several pieces of evidence suggest that TRPA1 may function as an amplifier to modulate the mechanically-evoked sensory neuron response downstream of another mechanotransducer or mechanotransduction complex. First, when electrical search stimuli were used to locate neurons in skin-nerve preparations from TRPA1-null or wild type mice, normal proportions of mechanically-sensitive fibers were found among C, Aδ and Aβ fiber classes 19. Thus, TRPA1 is not generally essential for the functional presence of mechanically-sensitive fibers. This suggests that either TRPA1 does not likely form an essential part of the complex required for mechanotransduction to occur in cutaneous sensory neurons, or that multiple complexes exist and compensate in TRPA1-null mice. Another consideration is that in skin-nerve preparations, sensory terminals are embedded in a milieu of other neighboring cell types, including keratinocytes, melanocytes and dendritic cells, which also express TRPA1 17, 20, 59. Consequently, the mechanosensory role of TRPA1 in sensory nerve terminals versus that in non-neuronal epidermal cells needs to be determined by mechanotransduction experiments on these cell types independently. Second, either genetic ablation or pharmacological inhibition of TRPA1 markedly reduces the mechanical firing of C fiber nociceptors at high intensities in the presumably noxious range 19, 24, 57. This data suggests that TRPA1 may be an amplifier of mechanically-gated action potentials following transduction by another mechanically-gated channel(s). Amplification could conceivably occur via initial mechanotransduction by an upstream channel/protein(s) that allows extracellular Ca2+ entry or release from intracellular stores, and elevated Ca2+ levels in a microdomain near TRPA1 that subsequently activates TRPA1 via its Ca2+-sensitive EF hand domain 7. If TRPA1’s role in mechanosensation in sensory afferents is indeed that of an amplifier or potentiating modulator, the prospects of TRPA1 as a pharmaceutical target for pain conditions that involve mechanical hypersensitivity are quite exciting. Select TRPA1 compounds could conceivably be designed that would “take the edge off” mechanical allodynia and hyperalgesia but leave intact the normal tissue-protective properties of nociceptors as well as normal cutaneous sensitivity to light touch, vibration and limb position.
Rapidly growing evidence indicates that the TRPA1 channel contributes substantially to the hypersensitivity and pain behavior that occurs with many inflammatory and neuropathic diseases. This evidence spans studies focused on multiple diseases (inflammation, osteoarthritis, nerve injury and diabetes), utilizing different pharmacological inhibitors of TRPA1 or transgenic knockouts, and from targeting different sites (skin, primary afferent neurons and spinal cord neurons).
The possibility that TRPA1 has a role in inflammatory hypersensitivity was initially suggested by the cellular findings that the inflammatory mediator bradykinin targets TRPA1 via a PLC/Ca2+ signaling pathway to elicit its excitatory effects 14, 25. Pharmacological inhibition of TRPA1 (small molecule AP-18 injection into the hindpaw) reverses the mechanical hypersensitivity induced by peripheral inflammation via CFA (complete Freund’s adjuvant) 53. Signaling molecules released during peripheral inflammation upregulate expression and function of TRPA1 channels on peripheral sensory neurons 52, 60, 61. Recent cellular evidence indicates that inflammatory mediators act via protein kinase A (PKA) and PLC signaling pathways to drive the translocation and cell surface expression of TRPA1 62. In composite, studies from multiple groups using various TRPA1 inhibitors and different routes of administration (oral, systemic, intraplantar and intracerebroventricular) have documented that acute pharmacological inhibition of TRPA1 blocks both the mechanical and cold hypersensitivity that accompanies persistent inflammation in the paw and osteoarthritis in the knee joint 52, 58, 63.
TRPA1 has also been implicated in neuropathic pain by several lines of evidence. A very recent, exciting study revealed that a point mutation in the TRPA1 S4 transmembrane domain results in the rare autosomal dominant Familial Episodic Pain Syndrome, a disease characterized by episodes of severe hypersensitivity in the upper body triggered by cold, fatigue and fasting 9. This is the first example of a TRP-mutant channelopathy that underlies a specific familial pain syndrome and indicates that variations in the TRPA1 gene can underlie pain perception in humans 9. In rodent model systems, TRPA1 mRNA levels are upregulated following spinal nerve ligation 61, 64, 65. Painful neuropathy induced by the chemotherapy drug cisplatin, used to treat solid tumors, induces TRPA1 mRNA upregulation in mouse trigeminal ganglion in vivo 66. Antisense knockdown of TRPA1 alleviates cold hypersensitivity, and treatment with HC-030031 reduced mechanical hypersensitivity following spinal nerve ligation 63, 67. However, not all nerve injury studies agree. Chronic constriction injury (CCI) resulted in decreased TRPA1 mRNA, TRPA1 function and cold responsiveness in DRGs 68, 69, suggesting that the role of TRPA1 in cold-hypersensitivity after nerve injury may be model-dependent 67. Although speculative at this point, the differences between nerve-injury models suggests that TRPA1 channel expression and function may greatly depend on the context of the tissue setting. Further conjecture is that TRPA1 expressed in different neuronal populations may be considerably heterogeneous, and heterogenity may arise from different expression patterns of alternatively-spliced TRPA1 variants or point mutations9, or co-assembly of TRPA1 with other TRP channels to form heterotetrameric complexes.
TRPA1 has also been implicated in diabetic neuropathy as acute inhibition of TRPA1 (CHEM; intraperitoneal administration) attenuated mechanical withdrawal thresholds in a model of streptozotocin-induced diabetes 70. A subsequent study showed that spinal administration of the TRPA1 inhibitor CHEM markedly attenuated diabetic hypersensitivity whereas intraplantar administration produced weak inhibition of hypersensitivity 71. These findings are quite interesting because they highlight the fact that TRPA1 channels are expressed on multiple sites as well as cell types, including skin keratinocytes, peripheral terminals of sensory neurons, dorsal root ganglia, and central terminals of primary afferent fibers in the spinal cord 4, 16, 54, 72. Indeed, a very recent study indicates that the central terminals of primary afferents that express TRPA1 make excitatory synaptic connections to specific subtypes of spinal neurons in the substantia gelatinosa (lamina II) 72. These data suggest that first, TRPA1-expressing primary afferent fibers may have specific second-order ascending pathways in the spinal cord that they feed input to, and second, that TRPA1 channels on central terminals of primary afferent fibers may be key in controlling nociceptive afferent synaptic input by modulating presynaptic release of neurotransmitters to spinal cord pain pathways 72–74. Together this data provides all the more evidence that TRPA1 is likely to be an exciting pharmacological target for multiple types of soft tissue injury, nerve injury and debilitating, chronic diseases associated with mechanical or cold hypersensitivity.
One of the most challenging aspects of TRPA1 expression and function analysis is to determine how to interpret interspecies differences, and the ultimate effect these differences will have on translating TRPA1 properties to humans for useful drug design. TRPA1 has diverged from a monophyletic common ancestor (TRPA1 clade) and is distinct from other TRPAs (basal TRPAs). Vertebrates express only channels derived from the TRPA1 clade, whereas invertebrates can express TRPA channels from both or may be limited to basal TRPAs. For instance, Drosophila expresses multiple TRPA channels that include TRPA1 (clade) and Painless (basal), and C. elegans expresses TRPA1 (basal)94. TRPA1 channels belonging to the TRPA1 clade are expressed in species as divergent as Drosophila 75, mice 4, dogs 21, snakes 76 and humans 3. The TRPA1 clade class of channels conserve at least five of the six cysteine and lysine residues that are critical for electrophile sensing properties, which is strikingly different from basal TRPA1 channels that conserve only one and are not electrophile-sensitive94. These electrophilic properties are a key component to human chemical nociception via TRPA1. Thus, neither C. elegans TRPA1 nor Painless provide a valuable model for TRPA1 studies involving electrophilic chemical agonists that can be translated to humans. More importantly, human and rodent TRPA1 are ~80% identical at the amino acid level 21. Most academic labs and preclinical pharmaceutical labs rely on rats and mice for modeling pain conditions. Given the variation between rodent and human TRPA1 amino acid sequences, physiological function may differ considerably. Thus, the data from TRPA1 studies in rodents and its implications for human pain conditions should be interpreted with some caution. For instance, caffeine directly activates mouse TRPA1, whereas it suppresses human TRPA1 when expressed in heterologous cells 34. In contrast to rodents, the chimpanzee TRPA1 orthologue is 99.7% homologous to humans, suggesting they have similar biophysical properties.
While chimpanzee TRPA1 homology is very close to humans, the significant cost and scarce availability of using chimpanzees or other primates for preclinical studies makes many current physiological studies unfeasible. Human tissue banks provide another option for expression and labeling studies, but performing critical physiological investigations to determine TRPA1 function in fresh human tissue is nearly impossible. One of the most important challenges to move the field forward is to define the interpretability of how results from model systems apply to humans and determine in which contexts discoveries can be translated. It is true that non-human model systems provide an unparalleled insight into physiological mechanisms of TRPA1 activation and targeting. Yet the results of these studies must be approached with cautious optimism. Unraveling this will likely require increased research methodologies performed in vivo, further identification of specific channel residues involved in ligand-binding, and a greater understanding of how channel expression and function differs in species at molecular, cellular, tissue and behavioral levels.
There is growing interest in and evidence for the involvement of non-neural cells in acute and chronic pain. Clear evidence indicates that microglia, astrocytes and satellite cells in the CNS are key players in numerous chronic pain conditions 77–79. In the periphery, non-neuronal cells surrounding sensory terminals have come into focus. In skin, epidermal keratinocytes serve as the initial barrier for the internal body to environmental elements such as algogens, toxins, allergens, ultraviolet radiation, temperature fluctuations and physical impact from objects. Recent evidence indicates that keratinocytes do not just form a protective barrier to the external world, but that they express ion channels and receptors, and release signaling neurotransmitters, neuropeptides, cytokines and growth factors 80, 81. They may dynamically communicate with sensory afferent terminals and thereby, contribute to transduction of external stimuli by cutaneous afferents. Indeed, strong evidence shows that the TRPV3 and TRPV4 heat-sensitive channels are highly expressed on keratinocytes and these channels convey thermal information to primary afferent terminals via prostaglandin E2 (PGE2) receptor and P2X channel activation 82, 83. Recent studies in several species from rodents to humans document that TRPA1 channels are expressed in epidermal keratinocytes 16–20, 59. The TRPA1 channels expressed on keratinocytes are functional in that they respond to mustard oil and cinnamaldehyde 18, chemically induced cold (icilin) 17, bradykinin 18 and cold (decreased ambient temperature) 18. Thus, keratinocytes may conceivably contribute to sensory transduction by either directly activating the terminal, or by modulating the response of the terminal to cold, inflammatory or mechanical stimuli.
In response to these stimuli, keratinocytes release cytokine interleukin-1β (IL-β), nerve growth factor (NGF), adenosine triphosphate (ATP) and PGE2. Other signaling molecules may also modulate the excitability of the sensory afferents that are embedded in and under the keratinocytes 80–84. These molecules can act on sensory afferents to sensitize the nerve terminal membrane to thermal and mechanical stimuli 85–87. Additionally, physical displacement of keratinocytes may modulate the activation of mechanically-gated channels expressed on sensory neurons; particularly those of unmyelinated C fibers which intercalate most extensively through keratinocyte layers. This close apposition may allow for a relatively fast time course of signaling that could modulate the mechanical response during or after transduction. Therefore, keratinocytes may act to mediate pain transduction by either sensitizing sensory afferents or participating in activating them. Since several studies have now documented TRPA1 expression on keratinocytes 17–19, it is conceivable that TRPA1 expressed on keratinocytes may contribute to or modulate the response properties of sensory nerve terminals.
One of the most enigmatic challenges of pharmacological treatment is the delivery of compounds to specific cells types or to distinct body regions, as well as increasing drug specificity for their targets. TRPA1 expression levels are high in nociceptors, generating great interest in drugs that may specifically target these sensory afferents while minimally affecting other body tissues. Though its expression in soft tissues may turn out to be important, the fact that TRPA1 is highly expressed in nociceptors makes it an excellent candidate for analgesic drug design. Techniques such as electron microscopy promise to aid in structure-based drug design by providing images of TRPA1 channel structure and configuration, similar to the 3 dimensional imaging analysis undergone withTRPV1 1. Like its relative TRPV1, structural adaptations within the TRPA1 protein may allow for unique pharmacological targeting. It was recently reported that pore dilation occurs in TRPA1 and TRPV1 in response to prolonged exposure to their respective agonists 88–90. While the dilated pore can alter cation permeability and current, it can also allow large molecules to enter TRPA1- and TRPV1-expressing sensory afferents. The dilated TRPA1 and TRPV1 pore has been shown to allow the entry of membrane impermeable anesthetics, such as the large organic cations Yo-Pro and N-methyl-D-glucamine (NMDG+) 88. Therefore, by using TRPA1- and TRPV1-selective agonists to activate and dilate channels, these larger molecules can enter. Based on their expression in Aδ and C-fibers, using channel specific activators may allow the precise delivery of drugs that are potentially useful in anesthesia. One potential candidate that has been explored in sciatic nerve and trigeminal ganglion nociceptors is the lidocaine derivative QX-314, which can enter through large-pore channels to produce long-lasting anesthesia by blocking voltage-gated sodium channels. Co-application of QX-314 with capsaicin selectively opens TRPV1 channels, allowing QX-314 entry 89, 91, 92, which selectively inhibits TRPV1-expressing nociceptor excitability. TRPA1 may be another channel that could be utilized in the same manner following pretreatment of selective agonists. A more important development will be the selective delivery of these compounds to tissue-specific subsets of TRPA1-expressing neurons, so drugs can selectively decrease the sensitization and hyperexcitability of bronchopulmonary, cutaneous, orofacial and visceral sensory afferents.
We believe that TRPA1 will continue to be a clinically-relevant pharmacological target of great therapeutic potential. The mechanical and cold hypersensitivity that plagues patients with chronic neuropathic pain, inflammation and tissue injury continues to be a devastating clinical problem, with few effective therapies available. Substantial evidence indicates that TRPA1 contributes to these processes in sensory neurons at multiple levels, from peripheral terminals and their surrounding non-neuronal cells, to DRG somata, to central terminals in the spinal cord. One of the greatest challenges is to delineate TRPA1 activation by multiple, diverse stimuli and how these processes contribute to acute and chronic mechanical hypersensitivity and cold allodynia, and the protective response to electrophilic compounds after injury. Our greatest steps in developing therapeutic treatments to address TRPA1-mediated pain are ahead. We anticipate that better TRPA1-specific drugs will be designed and therapeutic treatment protocols developed at an increasingly rapid pace as cutting-edge research continues to be conducted on this unique channel.
The authors would like to thank Dr. Daniel Vilceanu, Richard Lennertz, Marie-Elizabeth Barabas and Elena Kossyreva for their helpful discussion and comments on the manuscript.