TRPV1 agonist-based therapies: mechanism of action and clinical prospects Keith R. Bley1 and Annika B. Malmberg2 1NeurogesX, Inc., 981F Industrial Road, San Carlos, CA 94070, USA; 2Elan Pharmaceuticals, 800 Gateway Boulevard, South San Francisco, CA 94080, USA Introduction Capsaicin and other naturally occurring pungent molecules have been used for cen- turies as topical analgesics to treat a variety of painful conditions. Moreover, since 1989, periodic instillations of high-concentration capsaicin and resiniferatoxin (RTX) solutions have been found useful for the management of persistent bladder pain and the symptoms of overactive bladder. However, only within the last 8 years has it been appreciated that the selective action of capsaicin and similar compounds on nociceptive sensory nerve fibers is mediated by agonism of a ligand-gated ion channel called the transient receptor potential vanilloid receptor 1 (TRPV1). The selective expression of TRPV1 on nociceptors – those nerve fibers specialized for the detection of stimuli associated with tissue injury – in skin, bladder, joints and other tissues, has resulted in this receptor becoming an important target for rational drug development. Two different, but non-mutually exclusive, strategies are being pur- sued: optimization of TRPV1 agonist-based therapies which can functionally inac- tivate nociceptive nerve fibers for extended periods, and identification of receptor antagonists which would prevent nociceptive fibers from being activated by on- going inflammatory stimuli. This chapter will focus on recent advances in the understanding of drugs and treatments which attempt to use naturally occurring or synthetic TRPV1 agonists to alter the function of nociceptive sensory nerves and consequently cause pain relief. Consistent with the experimental results reviewed, the terms nociceptor and C-fiber will often be used interchangeably, even though there is not always a complete over- lap of the two categories; nociceptors consist of not only C-fibers, but also some Aδ- fibers. Evidence for hypotheses regarding the basis of nociceptor hyperactivity in pain syndromes will be reviewed, and the prospects for efficacy of locally adminis- tered therapies against various indications will be evaluated. Turning up the Heat on Pain: TRPV1 Receptors in Pain and Inflammation, edited by Annika B. Malmberg and Keith R. Bley © 2005 Birkhäuser Verlag Basel/Switzerland 191 Keith R. Bley and Annika B. Malmberg TRPV1 agonist-induced nociceptor desensitization When activated by a combination of heat, acidosis or endogenous agonists, TRPV1 initiates signal transmission to the spinal cord by depolarizing sensory nerve endings and generating action potentials which may be experienced by the brain as either a warming or a burning sensation (Fig. 1). However, if TRPV1 is activated continu- ously by on-going exposure to an exogenous agonist (e.g. capsaicin), a local bio- chemical signal can also be generated in nerve fibers, which produces long-term effects on nociceptive fiber functionality [1]. The TRPV1 channel is highly calcium- permeable, allowing calcium to flow down its steep electrochemical gradient into the cell. Furthermore, as TRPV1 is also expressed on intracellular organelles, exter- nal capsaicin application can cause release of calcium from the endoplasmic reticu- lum and may even induce additional intracellular calcium release from internal stores via calcium-dependent calcium release [2]. If TRPV1 is activated in this con- tinuous fashion, high levels of intracellular calcium and associated enzymatic and osmotic changes can induce processes that impair nociceptor function for extended periods [1]. A persistent lack of responsiveness to stimuli that would normally cause noci- ceptor activation has been termed ‘desensitization’. The use of this term in TRPV1 agonist literature arises from psychophysical studies of human subjects who display reduced reactions to painful stimuli applied to areas pretreated with TRPV1 ago- nists, and is not narrowly confined to a direct desensitization of TRPV1 or its intra- cellular signaling mechanisms [1]. Nociceptor hyperactivity Desensitization of nociceptive nerve fibers may constitute an important therapeutic intervention if they are hyperactive. Following acute injury, the basis for the hyper- activity and/or hypersensitivity of nociceptive nerve endings in affected tissues is well established [3], and the protective behaviors which result from nociceptor hyperactivity are considered fundamental to tissue repair and avoidance of addi- tional damage. Still, TRPV1-agonist-based therapies are being developed for acute traumatic pain syndromes (see below) because it is widely recognized that even acute pain can be non-productive and interfere with healing processes. In chronically painful conditions, particularly neuropathic pain syndromes, clin- ical and nonclinical research show collectively that the most peripheral aspects of damaged sensory nerves often display aberrant ‘pathophysiological’ electrical hyper- activity [4]. However, direct correlations between aberrant activity of nociceptors and patient pain reports have proven difficult to demonstrate, due to the technical complexity of measuring electrical activity in small-diameter nerve fibers. A tech- nique known as microneurography – which measures action potential extracellular- 192 TRPV1 agonist-based therapies: mechanism of action and clinical prospects Figure 1 Activation of TRPV1 leads the sensation of heat or burning pain and can also result in localized nociceptor desensitization. ly – can be used, but this diagnostic procedure is somewhat invasive and may cause discomfort [5]. Moreover, microneurography has the greatest utility for the long nerves of the legs and arms, and thus those chronic pain syndromes with primary presentations in the trunk or face (e.g. lower-back pain or post-herpetic neuralgia) are difficult to analyze. Consequently, there are very few published studies that have correlated successfully nociceptive nerve fiber hyperactivity with patient reports of chronic pain, and they all involve neuropathic pain of the extremities. For instance, a systematic study of hyperactive nociceptors in patients with erythromelalgia (burning pain of the feet) showed altered conduction velocities and spontaneous activity or sensitization in some mechano-insensitive C-fibers [6]. In contrast, spontaneous activity of distal nociceptive fibers following nerve injury has been recorded extensively in nonclinical models, and correlated directly with pain behaviors. For instance, transection of the sciatic nerve in rodents is a long-standing model in which spontaneous or ‘ectopic’ electrical activity of the resulting neuroma (the injured tip of a nerve fiber) develops, and nerve fibers ter- 193 Keith R. Bley and Annika B. Malmberg minating in the neuroma become extremely sensitive to stimuli [7]. However, although highly instructive regarding basic mechanisms of neural excitability, large nerve neuromas represent only a very small fraction of clinically presenting periph- eral neuropathies. Polyneuropathies in which some innervation of the skin or other target organ remains intact – such as due to diabetes – are much more common and clinically important [8]. Accordingly, several recent nonclinical studies have focused on the excitability of intact nociceptors following mechanical injuries to surrounding nerve fibers. For instance, one day following ligation and transaction of the L5 spinal nerve in rats, about one-half of the uninjured C-fiber nociceptors in the L4 spinal nerve develop spontaneous activity [9]. Similarly, 7 days following rhizotomy of L5 ventral roots (which leads predominantly to degeneration in myelinated fibers) in rats, a marked decrease in paw withdrawal thresholds occurred concomitantly with increased low- frequency C-fiber spontaneous activity [10]. Furthermore, after partial denervation of the dorsum of the foot was induced by tight ligations of spinal nerve L6 in pri- mates [11], there is a significantly higher incidence of spontaneous activity observed in uninjured single C-fibers in the superficial peroneal nerve recorded using an in vitro skin/nerve preparation. Bases for nociceptor hyperactivity The projections of nociceptors into target organs can be visualized and quantified by immunostaining of antigens selectively expressed in neurons. Protein gene prod- uct 9.5 (PGP 9.5) is the most commonly studied marker, although substance P, cal- citonin-gene-related peptide (CGRP) or others have also been used. Antibodies to PGP 9.5 stain all nerve fibers, but because nerve fibers in the epidermis of the skin, uroepithelium of the bladder and mucosa are almost exclusively nociceptors (see below), changes in the density of these fibers may be quantified [12]. In order to understand the mechanisms underlying aberrant activity of nocicep- tive nerve fibers in chronic pain syndromes, one key factor may be changes in the density of innervation of the target organ. That is, in most pain syndromes, which would be considered neuropathic, sensory neuron axons are lost due either to cell body or nerve fiber damage following viral, metabolic, traumatic or chemical insults. In contrast, in chronically painful conditions, which do not involve direct injury to axons, there may be increased nociceptor innervation of target organs (see below). Immunohistochemical analyses have indicated that the density of epidermal nerve fibers in skin is decreased in a wide range of neuropathic pain