TRPV1 Antagonists and Chronic Pain: Beyond Thermal Perception

In the last decade, considerable evidence as accumulated to support the development of Transient Receptor Potential Vanilloid 1 (TRPV1) antagonists for the treatment of various chronic pain conditions. Whereas there is a widely accepted rationale for the development of TRPV1 antagonists for the treatment of various inflammatory pain conditions, their development for indications of chronic pain, where conditions of tactical, mechanical and spontaneous pain predominate, is less clear. Preclinical localization and expression studies provide a firm foundation for the use of molecules targeting TRPV1 for conditions of bone pain, osteoarthritis and neuropathic pain. Selective TRPV1 antagonists weakly attenuate tactile and mechanical hypersensivity and are partially effective for behavioral and electrophysiological endpoints that incorporate aspects of spontaneous pain. While initial studies with TRPV1 antagonist in normal human subjects indicate a loss of warm thermal perception, clinical studies assessing allelic variants suggests that TRPV1 may mediate other sensory modalities under certain conditions. The focus of this review is to summarize the current perspectives of TRPV1 for the treatment of conditions beyond those with a primary thermal sensitivity.


Introduction
Chronic neuropathic pain is the result of a lesion or disease that affects somatosensory processing. Despite consensus that this syndrome arises from injury to the peripheral and/or central nervous systems, the numerous mechanistic differences of the maladaptive responses to deal with pain represent a major obstacle in the development of novel and effective medications. Chronic pain disorders, including neuropathic pain, post-herpetic neuralgia, fibromyalgia, osteoarthritis, and bone cancer pain, have collectively emerged as a serious public health concern. In the United States alone, chronic pain syndromes affect 25% to 30% of the population and nearly 60% of people over the age of 65 [1]. Recent statistics from the World Health Organization (WHO) estimate that 5% of the global population, 70% of patients with cancer, and 95% of patients with spinal cord injuries suffer from some form of chronic pain. Furthermore, chronic pain disorders are often co-morbid with other diseases including cancer, metabolic disease, and psychiatric disorders including anxiety and depression [2]. Taken together, these (and other) pain disorders represent one of the most underestimated health-care burdens costing more than $200 billion in annual health-care expenses [3].
The Transient Receptor Potential Vanilloid 1 (TRPV1) channel is one of the most researched and targeted mechanisms for the development of novel analgesics for inflammatory pain owing to its distribution and function. These channels are predominately expressed in small sensory C-fibers and to a lesser extent in Aδ-fibers [4], both of which terminate in the spinal dorsal horn, where TRPV1 is localized to both pre-and post-synaptic neurons in lamina I and II, as well as to glial cells [5]. In addition to localization at the spinal level, TRPV1 has supraspinal localization and contributes to descending modulation of nociceptive stimuli [6]. TRPV1 is activated by heat (>43 °C), as well as endogenous eicosanoids and protons (pH < 5.9). With respect to pain, the expression of TRPV1 channels are upregulated in preclinical models of inflammation [7,8], as well as clinical inflammatory conditions, such as osteoarthritis and rheumatoid arthritis [9], inflammatory bowel disease [10,11], gastro-esophageal reflux disease [12,13], chronic pelvic pain [14], and chronic cough [15,16]. In addition to disease related changes in receptor expression, inflammatory mediators such as cytokines, prostaglandin, bradykinin, glutamate, serotonin, and nerve growth factor all have been shown to increase the phosphorylation state of TRPV1, thereby increasing channel activity [17][18][19]. Collectively, these data suggest that TRPV1 is a critical integrator of inflammatory pain signaling [20][21][22][23].
Initial approaches targeting TRPV1 utilized agonists for the treatment of pain. Capsaicin is the prototypical TRPV1 agonist that is found in many topical formulations [24]. Agonist treatment for pain is related to the initial excitation of sensory neurons followed by a refractory state of desensitization, where the neuron becomes unresponsive to TRPV1 agonists and other inflammatory mediators. Repeated or high dose application produces a reversible ablation of the nerve fiber that further reduces sensitivity to cutaneous stimuli such as tactile, heat, mechanical and cold [25]. However, because this later approach essentially produces a neurolytic lesion of TRPV1-associated epidermal nerve fibers, the physiology for pain alleviation is mechanistically different from antagonist approaches.
The discovery of selective TRPV1 antagonists has provided additional support for the role of TRPV1 channels in inflammatory pain conditions. Pre-clinically, TRPV1 antagonists are effective at blocking thermal hypersensivity to numerous inflammogens (e.g., carrageenan and complete Freund's adjuvant; CFA), without modulating associated inflammatory responses, as measured by edema. Initial clinical results indicate that TRPV1 antagonist decrease thermal pain perception in normal subjects and elevate core body temperature [26][27][28]. Whereas tolerance appears to develop to the hyperthermic effects of TRPV1 antagonists, there does not appear to be tolerance to thermal hypoethetic effects. Collectively, these findings substantiate the role of TRPV1 as a logical mechanism associated with thermal hypersensitivity.
Although warm/hot thermal sensitivities represent some of the symptomatology, chronic pain is also characterized by allodynia, hyperalgesia and spontaneous pain where substantial heterogeneity in responsiveness to sensory stimuli exists. Using quantitative sensory testing stimuli, neuropathic patients can be sensitive to multiple stimuli including blunt pressure, pinprick, heat, cold, brushing of the skin, paradoxical heat sensation and enhanced pinprick wind-up, a form of central sensitization [29]. Moreover, gains of nociceptive sensitivity (hyperalgesia), loss of nociceptive sensitivity (hypoalgesia) or combinations of both are often observed, with some disease populations demonstrating differing phenotypes. For example, in a study by Maier et al. [29], 25% of peripheral nerve injury patients had hyperalgesia to hot pain, whereas 16% had hypoalgesia to this same stimulus. Until clinical trials for the effectiveness of TRPV1 antagonists are conducted, the generality of TRPV1 for pain conditions that are primarily non-inflammatory will remain unknown. Herein review and summarize the role of TRPV1 in stimulus modalities beyond warm thermal and whether sensitization to one sensory modality (e.g., thermal) could contribute to altered sensitivities to other modalities.

Expression Changes in TRPV1 Channels in Chronic Pain Conditions
Numerous preclinical studies suggest that TRPV1 expression is altered under conditions of chronic pain. For example, in normal animals, TRPV1 is predominately expressed in small sensory C-fibers and to a lesser extent in Aδ-fibers [4], both of which terminate in the spinal dorsal horn, where TRPV1 is localized to both pre-and post-synaptic neurons (and glial cells) in lamina I and II [5]. Following nerve injury, TRPV1 is down-regulated in the spinal cord after rhizotomy [4] and in the somata of damaged dorsal root ganglion (DRG) nerves two weeks following nerve transection or spinal nerve ligation (SNL) [30]. Despite this loss of TRPV1 in axotomized DRGs, TRPV1 was detected proximal to the neuronal site of lesion. After partial sciatic nerve ligation (PSNL), TRPV1 protein was increased in a population of undamaged DRG neurons [30]. Similarly, after lumbar (L) 5 SNL, TRPV1 expression was decreased in the damaged L5 DRG, whereas it was increased in the non-ligated L4 DRG, with a 3-fold increase expression observed in A-fibers. These findings were corroborated by independent laboratories [31][32][33]; however, see [34,35].
Changes in TRPV1 expression were also observed in the chronic constriction injury (CCI) model of neuropathic pain. TRPV1 expression increased by 149% and 167% in the ipsilateral spinal cord seven and 14 days, respectively, after injury, whereas no changes in expression were observed at earlier time points (one or three days) or in the contralateral spinal cord [36]. At day 14, capsaicin-evoked calcitonin gene-related peptide (CGRP) release was significantly higher (170%) in spinal cord slices from CCI animals compared to sham animals, suggestive that increased expression has functional effects on spinal sensitization. Thus, it has been hypothesized that increased TRPV1 expression, and its enhanced activity due to phosphorylation by local injury and glial derived inflammatory mediators, could contribute to spontaneous neuronal activity by reducing the thermal threshold, whereby TRPV1 becomes activated at body temperature [17,18].
With respect to osteoarthritis, which initially begins with a peripheral inflammatory component, preclinical studies suggest that chronic osteoarthritis produces central sensitization phenomena similar to that observed in neuropathic pain models [37][38][39]. Only a few studies have evaluated TRPV1 expression under osteoarthritic conditions. In patients with osteoarthritis, TRPV1 is expressed on synovium, as well as synovial fibroblasts suggesting both a neuronal and a non-neuronal role of TRPV1 in this condition [9,40]. In rats, TRPV1 is expressed in DRG neurons and knee joint synoviocytes [41,42]. Additionally, in the mono-iodoacetate (MIA) model of osteoarthritis, joint afferents in the DRG, as determined by Fast Blue staining, expressed a greater amount of TRPV1 (72%) compared to normal joint afferents (54%) [40].
Lastly, preclinical studies and the clinical presentation of pain associated with chronic bone cancer suggest similarities to neuropathic pain [43][44][45]. In humans, TRPV1 is up regulated in osteoclasts from osteoporotic patients [46]. In mice, TRPV1 is expressed on sensory fibers in mineralized bone and bone marrow, DRGs and in the spinal cord [47]. In an osteosarcoma model of bone cancer, the percentage of TRPV1 positive neurons was increased by 7% in the DRG neurons of bone cancer mice compared to control mice [48]. Overall, the studies highlighted above provide compelling evidence for a pivotal role for TRPV1 channels in a variety of pain conditions.

Behavioral Changes in Chronic Pain Conditions
Numerous approaches for the behavioral evaluation of TRPV1 in pathologic pain have been used, including knockout, knockdown and antagonist strategies. In knockout studies, mice lacking TRPV1 show a reduced sensitivity to heat and reduced thermal hypersensivity in response to incision or to the administration of inflamogens, such as mustard oil, carrageenan and CFA [49][50][51][52]. TRPV1 knockout mice show normal responses to other stimulus modalities (e.g., tactile and pressure stimuli) and develop similar magnitudes of hypersensitivity to both thermal and tactile stimuli after PSNL [49]. These findings were substantiated in another study wherein a similar magnitude of mechanical hypersensitivity (thermal sensitivity was not evaluated) was observed after L5/L6 SNL in both knockout and normal mice [53]. In general, these knockout studies indicate that TRPV1 is exclusively a thermal transducer. However, in a bone cancer pain model, TRPV1 knock-out mice displayed approximately 40-50% less spontaneous pain behaviors and palpation-induced flinching compared to wild-type mice [47].
In contrast to most knockout studies, knockdown approaches have demonstrated that TRPV1 modulates stimulus modalities other than thermal. Following intrathecal administration of a TRPV1 selective small interfering RNA (siRNA), CCI-induced cold hypersensitivity (cold plate) was significantly reduced by approximately 50% for up to five days [54]. This effect was transient, as cold hypersensitivity returned to control levels after seven days. TRPV1 knockdown was demonstrated by a lack of capsaicin-induced behaviors following rectal application of capsaicin. Similarly, twice daily intrathecal administration of TRPV1 antisense (but not mismatch) oligonucleotides significantly (~50%), reversed L5/L6 SNL-induced tactile hypersensivity, as measured by an electronic von Frey apparatus [34]. Behavioral studies were also conducted with RNA interference (RNAi) against TRPV1 and directly compared with TRPV1 knockout animals. Consistent with previous studies, TRPV1 knockout mice developed tactile hypersensitivity, whereas mice treated with intrathecal RNAi did not develop tactile hypersensitivity [53]. Since knockdown strategies have greater similarity to antagonist approaches compared to knockout strategies, these contrasting results between knockout and knockdown studies suggest functional differences between gene deletion and gene silencing approaches.
Consistent with their effects in inflammatory pain models, TRPV1 antagonists also attenuate thermal hypersensitivity in neuropathic pain models. The TRPV1 selective antagonist A-425619 completely attenuated thermal hypersensitivity 10 days after SNL surgery in mice [35]. Against non-thermal stimuli, A-425619 produced a 36% reversal of both L5/L6 SNL-and CCI-induced tactile (von Frey filament) hypersensitivity following i.p. administration in rats [57]. However, in another study, A-425619 and a structurally dissimilar antagonist A-840257 did not produce any significant reversal of L5/L6 SNL-induced tactile hypersensitivity at doses that attenuated CFA-induced thermal hypersensivity [58]. Although the central nervous system (CNS) penetration of A-840257 was not disclosed, A-425619 exhibited low (~5%) CNS penetration [57], which might limit its effects under neuropathic pain conditions that are known to be mediated by spinal and supraspinal processes [review . Consistent with this notion, intrathecal (i.t.) administration of A-425619 produced a 33% reversal of CCI-induced tactile hypersensitivity [57].
To the extent that bone cancer pain has a neuropathic component, Ghilardi et al. [47] have demonstrated that daily administration of JNJ-17203212, a prototype TRPV1 antagonist, attenuated spontaneous and palpitation-induced flinching by approximately 50% compared to vehicle treated mice. Consistent with these findings, a single dose of ABT-102 produced an 18-19% reversal of ongoing pain-related behaviors, spontaneous ambulation and palpitation evoked pain-related behaviors [60] in a bone cancer pain model. The effectiveness of ABT-102 on these endpoints increased to approximately 43-45% reversal after daily administration for 12 days, which was not due to compound accumulation.
TRPV1 antagonists have been evaluated in the MIA-induced model of osteoarthritic pain. In this model, i.p. A-425619 produced a relatively small (24%) normalization of weight bearing [57]. These researchers also studied the importance of CNS exposure and the role of central TRPV1 actions in the MIA osteoarthritis model. Two compounds exhibiting similar in vitro TRPV1 potency and oral potency for reversing CFA-induced thermal hypersensitivity were evaluated after central and systemic administration [61]. Similar potencies were obtained for both compounds after i.t. administration in the MIA model, using weight bearing as the behavioral measure. However, the CNS penetrant molecule (A-784168) was more potent than the poorly CNS penetrant molecule (A-795614) after oral administration. Moreover, A-784168 reversed MIA-induced weight bearing differences by 85%, ~78% and 65% after oral, i.t., or intracerebroventricular routes of administration, respectively [61]. Consistent with these results, ABT-889425 dose dependently reversed MIA-induced impairment of grip strength with a dose of 300 μmol/kg producing a complete reversal [62]. These results highlight the potential importance of peripheral, spinal and supraspinal TRPV1 receptors in pain conditions.
Since patients with chronic pain disorders require repeated administration of medications, preclinical studies have instigated the effectiveness of TRPV1 antagonist following chronic dosing. In one study, ABT-102 was evaluated on hind limb grip strength after acute and chronic administration in the MIA osteoarthritis model [60]. Animals received ABT-102 daily at either a low, non-effective (5%) or high, partially-effective (47%) dose after a single administration. Following 12 days of dosing, the low dose produced a 62% improvement, whereas the high dose produced a 98% improvement in grip strength. These results were confirmed with another TRPV1 antagonist, A-995662, a single high dose of which produced an 84% reversal in hind limb grip strength compared to a 67% reversal with the non-steroidal anti-inflammatory drug celecoxib [38]. Following 12 days of dosing with a partially effective (22%) dose of A-995662, there was a significant restoration of grip force (91%), which was not due to compound accumulation. Overall, chronic dosing appears to alter the effectiveness of TRPV1 antagonists, with changes in behavioral responses correlated with significant reductions in capsaicin-induced glutamate (26% decrease) and CGRP (41% decrease) release in the dissociated (rat) spinal cords.
In addition to the contribution of TRPV1 to mechanical hypersensivity as measured through behavioral endpoints, other studies have provided evidence that mechanical hypersensitivity can be observed at the level of the nociceptive neuron. Increases in spontaneous firing and mechanically evoked firing of spinal wide dynamic range neurons have been observed in inflammatory pain models (i.e., CFA), as well as MIA model of OA pain. The TRPV1 antagonist A-889425 attenuated both spontaneous and mechanically evoked firing in WDR and nociceptive specific neurons when evaluated at the level of the dorsal spinal cord after von Frey stimulation of the OA knee (62). Changes in neuronal activity also have been observed in other chronic inflammatory pain models likely containing a neuropathic component such as sickle cell disease [63,64]. Mice expressing the human sickle hemoglobin (HbSS) exhibit pathology and behavioral phenotypes consistent with human sickle cell patients. In mice, behavioral hypersensivity was observed with tactile, cold thermal and warm thermal stimuli. Along with attenuating these behavioral effects, the TRPV1 antagonist A-425619 substantially attenuated mechanical neuronal discharges in sensitized C-fibers to levels similar to control [64].

Allelic Variations in TRPV1 Channels
A growing body of research has underscored gender, psychological, cultural and genetic factors contributing to pain perception. For TRPV1, a number of allelic variants have been identified and the allele frequency can vary substantially among ethnic populations. For example, the TRPV1 1911A > G (rs8065080, I585V) minor allele in Caucasian populations is a major allele in Han Chinese and Japanese populations [65]. In female subjects with this variant, higher cold withdrawal tolerance to experimental cold water hand emersion has been reported [66] (however, see [67]). Moreover, this allele variant was associated with a significantly lower risk of painful knee osteoarthritis (OA) by comparing symptomatic and asymptomatic OA patients. However, because of its expression profile, it was also hypothesized to play a role in OA-induced cartilage morphology in addition to effects on nociception [68]. In neuropathic pain patients, this allele variant was associated with diminished cold sensitivity and in a subset of patients characterized by preserved sensory function, significantly less heat hyperalgesia, pinprick hyperalgesia and mechanical hypaesthesia [69]. Importantly, while the genetic variant contributed to the diminished somatosensory hypersensitivity in neuropathic pain patients, it was not linked to the susceptibility for developing neuropathic pain. A similar finding was observed in that TRPV1 was not associated with susceptibility for developing chronic pancreatitis, though the impact on pain perception was not fully quantified [70]. Another TRPV1 allele variant 1103C > G (rs222747, M315I) was associated with cold hypaesthesia [69] suggesting that there might be additional TRPV1 genetic variability.
Whereas these genetic association studies suggest a role for TRPV1 variants in multiple clinical pain states, in vitro studies have not demonstrated robust differences in channel function among alleles. For example, the TRPV1 1911A > G allele had normal response to capsaicin, pH and temperature [71] and other alleles evaluated in vitro by Xu et al. [65] had similar EC 50 values for capsaicin compared to TRPV1 wild-type control. However, slight variations were observed with some alleles including changes in the Hill slope coefficient and maximal responsiveness to anandamide. The later effect appeared related to increased TRPV1 expression in cells caused by the allele in vitro. There is increasing recognition that gene expression caused by allelic variants, in the absence of functional changes, might contribute to human genetic variation [65]. In what is akin to a human knockout study, an individual was identified with complete insensitivity to capsaicin. Upon further evaluation, this individual had TRPV1 expression in the buccal mucosa that was only 38% of normal levels [72]. While this subject displayed capsaicin insensitivity, sensitivities to pH and water temperatures between 45 and 60 ºC were similar to normal subjects. Together with changes in distribution after neuronal Pharmaceuticals 2012, 5 121 injury, such results suggest that the level of TRPV1 expression has profound consequences on neuronal sensitivities.

Conclusions
The TRPV1 channel is thought to be a primary mediator of warm thermal sensation. Recent studies indicate that inflammatory and neuronal injury alters the localization of TRPV1 and sensitizes the channel to respond to stimulus modalities beyond the classical thermal profile. Preclinical localization and expression studies provide evidence that changes in TRPV1 occurs at multiple levels (e.g., peripheral nociceptors, DRG and dorsal spinal cord) along pain pathways after neuronal injury. In studies that have evaluated functional activity, a gain of TRPV1-mediated currents is typically observed, which is related to local changes in pro-inflammatory neurotransmitters, cytokines and secondary messengers responsible for sensitization of the TRPV1 channel. In relation to allelic variants, such changes suggest that the TRPV1 channel may play a role in the variability associated with pain sensation, perception and the development of hypersensitivities under disease conditions. This heterogeneity among patient populations may be important for the continued development of novel analgesics. While functional differences among alleles have not demonstrated robust changes in sensitivity to activators of TRPV1, published studies have not evaluated whether the TRPV1 antagonists under clinical development have differing profiles in patient populations. Potential differences in sensitivities to antagonists, including the observation of thermal hypoethsia and hyperthermia might be observed. These initial insights suggest a complex role of TRPV1 in somatosensory pain perception and emphasize the need for a greater understanding of such processes with an eye towards identifying approaches for stratifying patients for better, more personalized, evidence based therapies. human IBD In rectosigmoid biopsies from IBD patients with pain, TRPV1 expression was significantly (4 to 5-fold) higher compared with controls and IBD patients without pain. In IBD patients with pain, TRPV1 expression correlated with abdominal pain severity.

[41]
mouse normal TRPV1 is localized within periosteum of periarticular bone, the articular capsule and vasculature of joints. TRPV1 expression occurs in 43% and 39% of DRG neurons innervating joint afferents the ankle and knee, respectively.

[34]
rat SNL TRPV1 expression decreased in ligated L5 and L6 but not in the non-ligated L4 DRG and dorsal spinal cord.

[5]
rat mouse normal TRPV1 expression is seen in lamina I and II of the L4 dorsal spinal horn and in spinal glial cells but not in TRPV1 KO mice.
human OA/RA TRPV1 is expressed in synovial fibroblasts from OA and RA patients (no comparison with non-OA/RA patients).

[47]
mouse BCP TRPV1 expression was localized on sensory femur DRG neurons. No expression changes were observed in neurons of sarcoma-bearing femurs.

[16]
human chronic cough TRPV1 expression of bronchial epithelial nerves in patients with chronic cough was 4-fold higher than controls. Patients with chronic cough were 30-fold more sensitive to the tussive effects of capsaicin.

[12]
human esophageal reflux TRPV1 mRNA and protein is increased 2-to 3-fold in the esophageal mucosa of patients with non-erosive reflux disease and erosive esophagitis compared to controls.  Acute doses of A-995662 significantly reversed MIA-induced decreased grip strength. An acutely subeffective (22% reversal) dose significantly restored grip force (91% reversal) after chronic administration.
The duration of effectiveness was longer than the detection of compound in brain or plasma. [55] CFA,