Brain, Vol. 123, No. 6, 1238-1246,
June 2000
© 2000 Oxford University Press
ATP in human skin elicits a dose-related pain response which is potentiated under conditions of hyperalgesia
Neuroscience Research Centre, Guy's, King's College and St Thomas' School of Biomedical Sciences, London, UK
Correspondence to:
Sara Hamilton, Neuroscience Research Centre, Guy's, King's College and St Thomas' School of Biomedical Sciences, St Thomas' Campus, Lambeth Palace Road, London SE1 7EH, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Despite the considerable interest in the possibility that ATP may function as a peripheral pain mediator, there has been little quantitative study of the pain-producing effects of ATP in humans. Here we have used iontophoresis to deliver ATP to the forearm skin of volunteers who rated the magnitude of the evoked pain on a visual analogue scale. ATP consistently produced a modest burning pain, which began within 20 s of starting iontophoresis and was maintained for several minutes. Persistent iontophoresis of ATP led to desensitization within 12 min but recovery from this was almost complete 1 h later. Different doses of ATP were delivered using different iontophoretic driving currents. Iontophoresis of ATP produced a higher pain rating than saline, indicating that the pain was specifically caused by ATP. The average pain rating for ATP, but not saline, increased with increasing current. Using an 0.8 mA current, subjects reported pain averaging 27.7 ± 2.8 (maximum possible = 100). Iontophoresis of ATP caused an increase in blood flow, as assessed using a laser Doppler flow meter. The increase in blood flow was significantly greater using ATP than saline in both the iontophoresed skin (P < 0.01) and in the surrounding skin, 3 mm outside the iontophoresed area (P < 0.05). The pain produced by ATP was dependent on capsaicin-sensitive sensory neurons, since in skin treated repeatedly with topical capsaicin pain was reduced to less than 25% of that elicited on normal skin (2.1 ± 0.4 compared with 9.3 ± 1.5 on normal skin). Conversely, the pain-producing effects of ATP were greatly potentiated in several models of hyperalgesia. Thus, with acute capsaicin treatment when subjects exhibited touch-evoked hyperalgesia but no ongoing pain, there was a threefold increase in the average pain rating during ATP iontophoresis (22.7 ± 3.1) compared with pre-capsaicin treatment (7.8 ± 2.6). Moreover, ATP iontophoresed into skin 24 h after solar simulated radiation (2 x minimal erythymic dose) resulted in double the pain rating of normal skin, increasing from 15.3 ± 4.1 to 32.7 ± 4.1. The pain response to saline was not significantly altered after UV irradiation at any time-point studied. We conclude that ATP produces pain by activating capsaicin-sensitive nociceptive afferents when applied to skin. The possibility that ATP activates nociceptors indirectly via its degradation products cannot be ruled out. The effects of ATP are dose-dependent and responses desensitize only slowly. In inflammatory conditions, ATP may be a potent activator of nociceptors and an endogenous mediator of pain.
pain; adenosine triphosphate; hyperalgesia; capsaicin; UV irradiation
MED = minimal erythema dose; SSR = solar simulated radiation; UVB = ultraviolet B; VAS = visual analogue scale
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The suggestion that ATP might have important neural actions is not new. Forty years ago, Pamela Holton demonstrated the release of ATP from some peripheral nerve fibres. In the intervening years, evidence of widespread and powerful neuronal actions of ATP has accumulated (Burnstock, 1997
). All of the classical criteria required to demonstrate that a substance acts as a neurotransmitter have slowly been proven for ATP. Thus, it is present in many neurons, and on release, it produces receptor-mediated post-synaptic effects and there are degradative mechanisms to terminate its actions (Brake and Julius, 1996). This role as a neurotransmitter is now generally accepted in the sympathetic nervous system (for review, see Sneddon et al., 1996), the dorsal horn of the spinal cord (Jahr and Jessell, 1983; Fyffe and Perl, 1984; Salter and Henry, 1985) and the medial habenula of the brain (Edwards et al., 1992).
In the last decade there have been major advances in our understanding of the nature, distribution and functioning of receptors for ATP, most notably the identification of a family of ligand-gated ion channels, the P2X family. There are currently seven known members of this family, all of which have been cloned, named P2X1–7 system (for review, see Ralevic and Burnstock, 1998). Studies of the distribution of these receptors have led to a resurgence of interest in the possibility that ATP might have functions in addition to that of a neurotransmitter. In particular, in situ hybridization and Northern blotting techniques have revealed that six out of seven members of the P2X family are located on primary sensory neurons (Kidd et al., 1995; Lewis et al., 1995; Collo et al., 1996). One of these, the P2X3 receptor, is exclusively distributed on small diameter neurons in the dorsal root ganglion that are thought to be nociceptors (Chen et al., 1995; Lewis et al., 1995). An issue of considerable current interest, therefore, is whether ATP might be released into peripheral tissues following injury and activate nociceptors, i.e. act as a peripheral mediator of pain. Obviously, if true, this finding could lead to the development of novel classes of analgesic drugs, based on antagonism of ATP receptors.
Animal studies have provided considerable circumstantial evidence for this suggestion (see Discussion). Briefly, ATP produces aversive nocifensive behaviour when injected intradermally (Bland-Ward and Humphrey, 1997; Hamilton et al., 1999a); it can activate nociceptors in vivo, and it can produce strong inward currents in these neurons in vitro (Krishtal et al., 1983; Bean, 1990). This circumstantial evidence has led to the theory that, in vivo, ATP mediates nociception via a receptor encompassing the P2X3 subunit, located on sensory neurons (Chen et al., 1995; Kennedy and Leff, 1995; Burnstock, 1997).
There are, however, much fewer data relating to the effects of ATP on human pain-signalling systems; only two studies have been performed. Bleehen and Keele applied, among other things, ATP to blister bases raised in human volunteers (Bleehen and Keele, 1977). They reported that even dilute solutions (as low as 1–3 µM, 0.2 ml applied for 2 min) produced pain with a delayed onset and typically lasting for 30–100 s. The majority of the experiments were carried out on one subject. In the other published study (Coutts et al., 1981) ATP was injected intradermally. These authors also reported that ATP produced pain, but there was no information regarding the magnitude or duration of the pain experienced. These two studies report conflicting information about the threshold dose required to elicit a pain response. Coutts and colleagues report that pain was only elicited with doses of about 250 nmol or more of ATP (deduced from their data) (Coutts et al., 1991), whereas Bleehen and Keele found doses of 0.2–0.6 nmol effective (Bleehen and Keele, 1977).
In summary, early works provide conflicting information on the threshold dose of ATP required to elicit pain and very limited information about the nature of this pain. Therefore, we have undertaken a quantitative study of ATP-induced pain, paying particular attention to the duration of the response, the desensitization properties of the response and how the response varies in normal and hyperalgesic skin in healthy volunteers.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Subjects
A group of 21 healthy adults (male and female, aged 21–44 years) participated in the experiments after having given informed consent. All subjects were familiar with the psychophysiological techniques employed in this study. All protocols had been approved by the ethical committee of St Thomas' Hospital.
Iontophoresis of ATP or saline
The experimental area of skin (always on the volar forearm) was cleaned with an alcohol swab before attaching a 9 mm diameter Perspex iontophoresis chamber (Moor Instruments, Axminster, UK) to the ventral surface of the subject's forearm using an adhesive disc. The chamber was then filled with the test solution, which was connected to the cathode of the iontophoretic circuit. The indifferent electrode was held in the subject's palm. The iontophoresor (Phoresor II PM 700, IOMED, Salt Lake City, UT, USA) delivered constant currents (0.1–0.8 mA) for specific times, and ATP and saline were iontophoresed into different experimental sites on the ventral forearm of the subject. Each site was used for only one iontophoretic trial (except in the experiments on desensitization and recovery, see below). The subject was blind as to which solution was being applied. In the first set of experiments the iontophoresis was carried out over a period of 4 min.
Psychophysiological magnitude estimations of induced pain
During iontophoresis, the subject was required to assess the degree of pain they were experiencing using an electronic visual analogue scale (VAS), generated on a personal computer. The subject was required to rate pain intensity by moving a computer mouse every 20 s on a scale of 0–100, where 0 corresponded to no pain and 100 to the worst pain imaginable for that area of skin. The average pain rating over a 4-min iontophoresis session was calculated for each subject.
Blood flow responses to iontophoresis of ATP and saline
The blood flow response to iontophoresis of ATP and saline at 0.2 mA for 4 min was assessed using a laser Doppler blood flow monitor (MBF3D, Moor Instruments). The blood flow directly over the iontophoresed area was monitored using a cylindrical probe fixed over the ATP or saline solution. Blood flow outside of the area of iontophoresis was monitored by a needle probe fitted into a hole in the Perspex, 3 mm from the outside edge of the iontophoresis chamber. Neither of the probes touched the skin but were fixed just above it. The output of the blood flow monitor was recorded on a Macintosh Performa 460 (Apple Computer Inc., Cupentino, Calif., USA) using a MacLab/4e and Chart software (v3.5, AD Instruments, Castle Hill, Australia).
Time-course of desensitization and subsequent recovery
In another set of experiments, ATP was iontophoresed continuously for 12 min at one site at 0.3 mA current. Having established that the response to ATP diminishes within that time-period, a second set of experiments was carried out whereby ATP was iontophoresed for an initial period of 8 min and then again for 4 min either 5, 35 or 60 min after the end of the initial iontophoresis period (i.e. each subject had two trials at the same site; an initial 8 min session at t = 0 followed by a second 4-min iontophoresis session starting at t = 13, 43 or 68 min).
Desensitization following repeated topical capsaicin treatment
The effect of iontophoresing ATP on to skin desensitized by repeated topical application of capsaicin was assessed (Carpenter and Lynn, 1981). Nine subjects participated in this study. An area of skin 6 x 3 cm was delineated on the flexor/volar surface of the forearm. Capsaicin solution (1% w/v) was applied using a cotton bud and the area then covered with an occlusion dressing. This procedure was repeated at 2 h intervals five times on day 1 and a further three times on day 2. Eight applications were usually sufficient to abolish the flare and pain response to capsaicin; if not, further applications were made until the response was abolished. Two hours after the final application, ATP and saline were iontophoresed on to the experimental area (using different sites) and an equivalent area on the contralateral arm. A current of 0.3 mA was used and the subject remained blind as to which solution was being tested. The pain response was monitored as before.
Hyperalgesia after topical capsaicin treatment
The response to ATP was compared between normal skin and skin rendered hyperalgesic after acute topical application of 1% capsaicin solution. Nine subjects participated in the study. A small cylinder (1.5 cm in diameter, 1 cm in depth) was glued on to the volar surface of the subject's forearm using Cyanoacrylate-gel (RS Components, Northampton, UK). Capsaicin solution (0.05 ml) was applied to the area of skin within the tube. After 45 min, the capsaicin solution and rubber tubing were removed and the experimental area cleaned with alcohol. At this time-point it was established that the spontaneous pain had subsided but touch-induced pain (provoked with a cotton swab brushed lightly across the skin) was still present. ATP and saline were iontophoresed on to the experimental area and, subsequently, to an area of untreated skin on the contralateral forearm. The pain response was monitored using the VAS. The current used was 0.3 mA. The subject was blind as to which solution was being iontophoresed at any one time.
UV irradiation
The response to ATP was compared between normal skin and skin which had received a standard dose of UV irradiation. Solar simulated radiation (SSR) from a 1000 W xenon arc solar stimulator (Oriel, Leatherhead, UK) was used to administer a specific dose of radiation to 1 cm2 sites on the subjects' forearms. About 90% of the spectrum of SSR falls within the UVB range (280–320 nm). Initial experiments were carried out to establish the minimal erythema dose (MED) for each subject. For this purpose, six 1 cm2 sites were exposed to increments of UV radiation (dose separation factor = 
2). The MED is the dose that results in erythema that is just perceptible after 24 h, as measured using an electronic erythema meter (Dia-Stron, Andover, UK). Having established this for each subject, two sites were then irradiated with a dose of 2 x MED. At 24, 48 and 72 h following irradiation, ATP and saline were iontophoresed for 4 min on to normal and irradiated patches of skin (different skin areas for each trial). The current used in all cases was 0.3 mA. The pain induced by ATP or saline was monitored using the VAS, as above. For each trial the subject was not aware of which experimental site was being iontophoresed or the solution administered.
Solutions
ATP (disodium salt) was dissolved in distilled water to make 1 or 10 mM solutions which were stored at -70°C and thawed on the day of use. Pilot experiments confirmed that, for a given current, 1 or 10 mM solutions elicit the same magnitude of pain. This finding was expected since the current at which a solution is iontophoresed determines the dose applied. Therefore, ATP solutions of either concentration were used interchangeably. Once thawed, the ATP was stored on ice until use. Commercially prepared (Steripak Ltd, Cheshire, UK) sterile saline (0.9% w/v) ampoules were used as the control solution.
The 1% capsaicin solution was prepared by dissolving 0.05g of 8-methyl-N-vanillyl-6-nonenamide (capsaicin powder; Sigma Chemicals Ltd, Poole, UK) in 0.5 ml of ethyl alcohol. Once dissolved 0.5 ml of Tween 80 was added and the whole solution diluted in 4 ml of sterile saline (0.9%) to make a 1% solution in 10% Tween. Once made up, the solution was kept in a light-protective container at 4°C.
Data analysis and statistical analysis
Data were processed on a PC computer. Parametric tests [t-test and ANOVA (analysis of variance)] were employed, as appropriate. All values are given as mean ± standard error of the mean. A level of 5% was taken as evidence of statistical significance.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
ATP-induced pain in normal skin
In all subjects (n = 12) the iontophoresis of ATP (1 or 10 mM) caused a sensation of tingling or burning pain. The pain typically began 20 s after the onset of iontophoresis and increased in magnitude over the succeeding 120 s (Fig. 1
). A plateau level of pain was maintained over the next few minutes. When the iontophoresis was terminated, the pain subsided very quickly, in a matter of seconds. The average magnitude of pain produced by ATP correlated positively with the iontophoretic current (P < 0.001, one-way ANOVA for current versus magnitude of pain elicited by ATP) (Fig. 2). The threshold current which consistently produced pain was 0.3 mA (P < 0.05, Tukey test; n = 21). With the highest current used (0.8 mA), the VAS rating averaged 27.8 ± 0.8 over the 4-min application period.
|
|
Iontophoresis of saline caused almost no pain sensation when low currents were used. Higher currents were associated with a tingling sensation or very low levels of pain. Unlike ATP, there was no correlation between the pain produced by saline and current (P = 0.2, one-way ANOVA for current versus magnitude of pain elicited by saline).
Changes in blood flow
Following iontophoresis of ATP (1 or 10 mM), a flare occurred in the surrounding area. Iontophoresis of ATP (0.2 mA) caused a rise in blood flow in the treated skin which started after about 30 s and had usually reached a plateau by the end of the 4-min period of iontophoresis (Fig. 3). Blood flow measured 3 mm outside the iontophoresed area also showed a progressive, but smaller, rise. Iontophoresis of saline caused a much smaller increase in blood flow in the skin being treated, with little or no change in blood flow in the surrounding skin.
|
The areas under the blood flow curves are significantly greater for ATP than saline for both the directly iontophoresed skin (ATP 14 956 ± 3547, saline 1315 ± 1722 arbitrary units/min; P < 0.01) and the skin 3 mm outside the iontophoresed area (ATP 5668 ± 4884, saline -76 ± 360 arbitrary units; P < 0.05).
Rate of desensitization and subsequent recovery
Figure 4 shows the time-course of the pain elicited during 12 min of iontophoresis of ATP (1 or 10 mM). On average, the pain caused by ATP reached a peak at about 3 min and then began to diminish. After 5 min, the pain caused by ATP had subsided to less than half its peak rating. By 10 min, the pain was one-fifth of that experienced at the peak. Figure 5 shows the average degree of desensitization by pooling VAS ratings into three 4-min bins (0–4, 4–8 and 8–12 min), the last of which is statistically different from the first (P < 0.05, paired t-test). In different trials, the rate of recovery from this state of desensitization was studied. An initial desensitizing stimulus lasting 8 min was applied. A test stimulus lasting 4 min was subsequently applied to the same site after varying intervals. If the skin remained desensitized, we would have expected ratings similar to those seen after 8–12 min of iontophoresis, as above. However, if there was a complete recovery from the desensitization, we would have expected ratings similar to the initial 4 min of ATP application. As shown in Fig. 5, average pain ratings were restored by ATP (1 or 10 mM) about 1 h after the desensitizing stimulus. At 13 and 43 min, but not 68 min, the response to ATP was still significantly less than that experienced in the initial 4 min (P < 0.05, paired t-test).
|
|
Effects of capsaicin-desensitization on ATP-induced pain
Patches of forearm skin were repeatedly treated with 1% topical capsaicin solution until this treatment was painless (see Methods). The capsaicin-desensitized skin was then tested with ATP (1 or 10 mM) or saline. As shown in Fig. 6
, ATP produced a much attenuated sensation under these circumstances. The average pain rating elicited on skin treated with capsaicin was reduced to less than 25% of that elicited on normal skin (2.1 ± 0.4 compared with 9.3 ± 1.5 on normal skin; P < 0.01, paired t-test). The effect of iontophoresing saline produced a very limited sensation of pain in normal skin (average 1.5 ± 0.4), which is attributable to the effect of the current. This too is reduced in capsaicin-desensitized skin (0.4 ± 0.2; P < 0.05, paired t-test). However, the reduction in the pain caused by ATP is clearly over and above that which can be explained by reduction in current-induced pain. Thus, the pain induced by iontophoresis of ATP is mediated by capsaicin-sensitive primary afferent neurons.
|
Hyperalgesia after acute topical capsaicin treatment
In contrast to the consequences of repeated capsaicin treatment, a single topical treatment produced a strong burning pain, which waned after 40 min, and hyperalgesia to touch, which persisted for 1–2 h. We applied ATP (1 or 10 mM) to skin treated with capsaicin at a time when the pain had disappeared but the hyperalgesia remained. When this was done, a threefold increase in the average pain rating during ATP iontophoresis (22.7 ± 3.1) was observed, compared with pre-capsaicin treatment (7.8 ± 2.6) (Fig. 7
). The potentiation of ATP-induced pain by capsaicin is highly statistically significant (P < 0.001, paired t-test). The responses to saline iontophoresis increased but this increase does not reach statistical significance (P > 0.05, paired t-test). Even if there was some sensitization to current it does not account for all of the potentiation seen in response to ATP in capsaicin-treated skin.
|
UV irradiation
UV irradiation also caused hyperalgesia, but unlike that induced by capsaicin, it evolved over many hours rather than minutes. The pain produced by iontophoresis of ATP (1 or 10 mM) or saline on normal skin was compared with skin treated with a dose of 2 x MED of UVB irradiation at 24, 48 and 72 h after UV irradiation (Fig. 8
). At 24 h, the average pain rating elicited by ATP on UV-treated skin was double that elicited on normal skin, increasing from 15.3 ± 4.1 to 32.7 ± 4.1 (P < 0.05, paired t-test). At 48 h, the mean VAS response to ATP was still greater than on untreated skin, but the difference does not reach statistical significance. At 72 h, the mean VAS response to ATP on UV irradiated skin was further reduced and does not differ significantly from the response on untreated skin. The very mild pain induced by saline iontophoresis (Fig. 8, lower panel) was clearly not affected by UV irradiation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
There is much circumstantial evidence to suggest that ATP may function as a peripheral mediator of pain. However, no previous study has quantified the magnitude and duration of pain elicited by ATP or correlated the magnitude of pain elicited with dose in human volunteers. In this study, we have characterized the pain-producing effects of iontophoretically applied ATP. This means of delivery overcomes the potential problem of rapid breakdown of injected ATP and allows prolonged effects to be studied. Iontophoresis delivers the drug over a defined area and is quantitative, i.e. the amount delivered is proportional to the iontophoretic driving current. Iontophoresis is also atraumatic. One limitation, however, is that the actual amount of drug delivered is not known.
Features of ATP-induced pain
In nearly all subjects, the pain induced by ATP increased during the initial 0–20 s of delivery, presumably as the concentration of ATP in the skin increased. The plateau level of pain produced in normal skin was modest. With the highest driving current used (0.8 mA) the maximum induced pain averaged 27.8 ± 0.8 on the VAS. The pain produced by ATP was essentially abolished in skin desensitized by repeated capsaicin application, demonstrating that it is mediated by capsaicin-sensitive nociceptors. Which functional subgroup of nociceptors, if any, is activated selectively by ATP was not revealed by these experiments. However, in rodent nociceptors studied electrophysiologically, ATP preferentially activated C-mechanoheat nociceptors (Hamilton et al., 1999b). In normal and desensitized human skin, no sensations other than pain were reported with ATP iontophoresis. The notion that ATP might function as a mediator of tactile stimuli following its release from Merkel cells (Malinovsky and Pac, 1985) is not supported by our observations here (nor indeed by our electrophysiological experiments in which we did not see excitation of large myelinated mechanoreceptive afferents with ATP).
The pain-producing effects of ATP were greatly potentiated in hyperalgesic skin. In both UV inflammation and in capsaicin-induced hyperalgesia, ATP elicited more than twice the expected pain. These findings are consistent with the observed increase in nociceptive effects of ATP analogues in animal models of hyperalgesia (Hamilton et al., 1999a). Our observations in human skin may help reconcile important differences that have been reported in the only other two studies of ATP-induced pain in humans. Coutts and colleagues report that pain was only elicited with a minimum dose of 250 nmol of ATP in 50 µl (Coutts et al., 1981), whereas Bleehen and Keele found doses of 0.2–0.6 nmol in 200 µl to be effective (Bleehen and Keele, 1977). The former study was undertaken on normal skin, while in the latter, compounds were applied to human blister bases, a condition that may have sensitized nociceptors to the agonists. It is not clear what mechanism underlies the increased sensitivity to ATP. In both the models studied here (capsaicin hyperalgesia and UV inflammation), there is insufficient time for altered gene expression (for instance, upregulation of ATP receptors) to contribute to the altered sensitivity. The responsiveness of some P2X receptors is known to be enhanced by low pH (Wildman et al., 1997), but this too is unlikely to explain the altered sensitivity of capsaicin-treated skin. Phosphorylation of receptor proteins appears to be a common event in a variety of models of hyperalgesia (Dray, 1995) and might occur in the case of P2X receptors. Whatever the mechanism, our observations suggest that a modest release of endogenous ATP would be capable of contributing to the activation of nociceptors in some persistent pain states.
Desensitization phenonomen
The pain produced by ATP diminished after several minutes, despite the provision of ATP by continuing iontophoresis. The time-course of this desensitization is similar to that seen when ATP is injected intradermally in rats and nocifensive behaviour is studied (Hamilton et al., 1999a). In the present study, we have shown that recovery of the response to ATP takes over an hour in human skin. Electrophysiological studies have characterized the members of the purinoceptor family according to their desensitization properties. The P2X3 channel, studied in vitro, desensitizes within milliseconds. The native dorsal root ganglion ion channel, studied in vitro, appears to desensitize in a manner which is replicated by the recombinantly expressed heterodimer formed from P2X3 and P2X2. However, even in this `slowly' desensitizing heterodimer the time-course for desensitization is within seconds, which contrasts markedly to the situation in vivo.
Mechanism of ATP-induced pain
One limitation of this study is that we cannot use potentially toxic analogues of ATP that are selective for the different P2X receptors. Therefore, we can only speculate as to the nature of the receptor mediating activation of nociceptors by ATP. In animal studies, we and others have been able to use the more selective ATP analogue ß-methylene ATP. The nocifensive response elicited by ß-methylene ATP in rats was similar to that induced by ATP in humans, with respect to latency and duration, potentiation in hyperalgesic skin, and dependency on capsaicin-sensitive nociceptors. ß-methylene ATP is specific for P2X1 and P2X3 receptors. However, P2X1 mRNA is present in very limited amounts in the dorsal root ganglion. Therefore, in rats it appears that the activation of nociceptors by ATP analogues is mediated, at least in part, by a receptor encompassing the P2X3 protein. Since the features of ATP-induced pain in humans parallel those seen with the selective P2X3 analogue in rat, it seems reasonable to suggest that the same conclusion might be drawn for human skin. A number of studies have been carried out to elucidate the population of nociceptors which express the P2X3 receptor. Nociceptors can be subdivided into two subsets according to their dependency on neurotrophins and the presence of certain immunohistochemical markers. The P2X3 receptor is present in about 35% of dorsal root ganglion neurons (Bradbury et al., 1998). The majority (85%) of P2X3-postive cells fall into the population of small diameter unmyelinated nociceptors which are dependent on GDNF (glial cell line derived neurotrophic factor); these nociceptors do not express neuropeptides and they bind the plant lectin IB4 (Bradbury et al., 1998). They terminate with a specific distribution within the inner lamina II in the dorsal horn of the spinal cord.
There is a possibility that ATP activates nociceptors indirectly via its degradation products. Extracellular endonucleotidases are responsible for cleaving ATP to produce ADP, which in turn is rapidly broken down into adenosine. Coutts and colleagues reported that subjects experienced pain with higher doses of ADP (Coutts et al., 1981). Bleehen and Keele also reported that ADP, AMP and adenosine produced algogenic actions on the human blister base (Bleehen and Keele, 1977). However, in the animal study carried out by Bland-Ward and colleagues adenosine did not elicit the same nocifensive paw-lifting response seen with ATP (Bland-Ward et al., 1997).
Changes in blood flow
In humans, ATP consistently elicits a modest flare response (also reported by Coutts et al., 1981). It is likely that part of the increased blood flow occurring directly under the ATP-treated skin was due to a direct effect on vascular smooth muscle. Human blood vessels contain several different types of purinoceptors, some mediating vasoconstrictor and others vasodilator responses. In human subcutaneous resistance vessels studied in vitro, low concentrations of ATP caused relaxation while high concentrations caused contraction (Martin et al., 1991). Intra-arterial infusion of ATP in healthy humans, however, produced only vasodilatation. Alternatively, the flare may be mediated by histamine released from mast cells by ATP (Hagermark et al., 1974; Coutts et al., 1981).
Clinical significance
All our data are consistent with the idea that ATP may be a peripheral mediator of pain. An important unresolved issue is under what context ATP might function as such a mediator. There are a variety of pathological situations where ATP is known to be present at high levels either extracellularly or intracellularly; for example, during reactive hyperaemia ATP is released from vascular endothelial cells. Burnstock has postulated that ATP may circulate and act on P2X receptors in adventitia (Bodin and Burnstock, 1995). Platelets release ATP during aggregation, particularly during migraine (Burnstock and Wood, 1996). ATP has also been implicated in the pain associated with malignancy on the basis that tumour cells contain exceptionally high levels of ATP (Maehara et al., 1987). Ferguson and colleagues demonstrated the release of ATP from urothelial cells in the bladder as a result of electrical stimulation or stress stimuli (Ferguson et al., 1997). Finally, ATP levels are shown to be raised in damaged and inflamed tissue (Gordon, 1986); for instance, ATP levels were high in articular fluid removed from arthritic knee joints (Ryan, 1991). Intracellular stores of ATP can be released by cell lysis, for example, as shown to be the case for erythrocytes (Bleehen and Keele, 1977), or by calcium-dependent mechanisms. Thus, there are many clinical situations where the release of endogenous ATP might plausibly contribute to ongoing pain. If this proves to be the case, selective antagonists of P2X receptors might prove to be effective analgesic agents.
|
Acknowledgments |
|---|
We wish to thank Elton Woo and Sue Walker for help with UV irradiation, Vivian Cheah for excellent technical help and Martin Koltzenburg for his helpful comments about the manuscript. We are grateful to the Special Trustees of St Thomas' Hospital who have provided support for Sara Hamilton.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Bean BP. ATP-activated channels in rat and bullfrog sensory neurons: concentration dependence and kinetics. J Neurosci 1990; 10: 1–10.[Abstract]
Bland-Ward PA, Humphrey PP. Acute nociception mediated by hindpaw P2X receptor activation in the rat. Br J Pharmacol 1997; 122: 365–71.[Web of Science][Medline]
Bleehen T, Keele CA. Observations on the algogenic actions of adenosine compounds on the human blister base preparation. Pain 1977; 3: 367–77.[Web of Science][Medline]
Bodin P, Burnstock G. Synergistic effect of acute hypoxia on flow-induced release of ATP from cultured endothelial cells. Experientia 1995; 51: 256–9.[Web of Science][Medline]
Bradbury EJ, Burnstock G, McMahon SB. The expression of P2X3 purinoreceptors in sensory neurons: effects of axotomy and glial-derived neurotrophic factor. Mol Cell Neurosci 1998; 12: 256–68.[Web of Science][Medline]
Brake AJ, Julius D. Signaling by extracellular nucleotides. [Review]. Annu Rev Cell Dev Biol 1996; 12: 519–41.[Web of Science][Medline]
Burnstock G. The past, present and future of purine nucleotides as signalling molecules. [Review]. Neuropharmacology 1997; 36: 1127–39.[Web of Science][Medline]
Burnstock G, Wood JN. Purinergic receptors: their role in nociception and primary afferent neurotransmission. [Review]. Curr Opin Neurobiol 1996; 6: 526–32.[Web of Science][Medline]
Carpenter SE, Lynn B. Vascular and sensory responses of human skin to mild injury after topical treatment with capsaicin. Br J Pharmacol 1981; 73: 755–8.[Web of Science][Medline]
Chen CC, Akopian AN, Sivilotti L, Colquhoun D, Burnstock G, Wood JN. A P2X purinoceptor expressed by a subset of sensory neurons. Nature 1995; 377: 428–31.[Medline]
Collo G, North RA, Kawashima E, Merlo-Pich E, Neidhart S, Suprenant A, et al. Cloning of P2X5 and P2X6 receptors and the distribution and properties of an extended family of ATP-gated ion channels. J Neurosci 1996; 16: 2495–507.
Coutts AA, Jorizzo JL, Eady RAJ, Greaves MW, Burnstock G. Adenosine triphosphate-evoked vascular changes in human skin: mechanism of action. Eur J Pharmacol 1981; 76: 391–401.[Web of Science][Medline]
Dray A. Inflammatory mediators of pain. [Review]. Br J Anaesth 1995; 75: 125–31.
Edwards FA, Gibb AJ, Colquhoun D. ATP receptor-mediated synaptic currents in the central nervous system. Nature 1992; 359: 144–7.[Medline]
Ferguson DR, Kennedy I, Burton TJ. ATP is released from rabbit urinary bladder epithelial cells by hydrostatic pressure changes – a possible sensory mechanism? J Physiol (Lond) 1997; 505: 503–11.
Fyffe RE, Perl ER. Is ATP a central synaptic mediator for certain primary afferent fibers from mammalian skin? Proc Natl Acad Sci USA 1984; 81: 6890–3.
Gordon JL. Extracellular ATP: effects, sources and fate. [Review]. Biochem J 1986; 233: 309–19.[Web of Science][Medline]
Hagermark O, Diamant B, Dahlquist R. Release of histamine from human skin induced by intracutaneous injection of adenosine-5'-triphosphate. Int Arch Allergy Appl Immunol 1974; 47: 167–74.[Web of Science][Medline]
Hamilton SG, Wade A, McMahon SB. The effects of inflammation and inflammatory mediators on nociceptive behaviour induced by ATP analogues in the rat. Br J Pharmacol 1999a; 126: 326–32.[Web of Science][Medline]
Hamilton SG, McMahon SB, Lewin GR. Selective activation of nociceptors by ATP and the P2X agonist ,ß-methylene ATP in the adult rat [abstract]. Soc Neurosci Abstr 1999b; 25: 683.
Holton P. The liberation of adenosine triphosphate on antidromic stimulation of sensory nerves. J Physiol (Lond) 1959; 145: 494–504.
Jahr CE, Jessell TM. ATP excites a subpopulation of rat dorsal horn neurones. Nature 1983; 304: 730–3.[Medline]
Kennedy C, Leff P. Painful connection for ATP [news]. Nature 1995; 377: 385–6.[Medline]
Kidd EJ, Grahames CB, Simon J, Michel AD, Barnard EA, Humphrey PP. Localization of P2x purinoceptor transcripts in the rat nervous system. Mol Pharmacol 1995; 48: 569–73.[Abstract]
Krishtal OA, Marchenko SM, Pidoplichko VI. Receptor for ATP in the membrane of mammalian sensory neurones. Neurosci Lett 1983; 35: 41–5.[Web of Science][Medline]
Lewis C, Neidhart S, Holy C, North RA, Buell G, Suprenant A. Coexpression of P2X2 and P2X3 receptor subunits can account for ATP-gated currents in sensory neurons. Nature 1995; 377: 432–5.[Medline]
Maehara Y, Kusumoto H, Anai H, Kusumoto T, Sugimachi K. Human tumor tissues have higher ATP contents than normal tissues. Clin Chim Acta 1987; 169: 341–3.[Web of Science][Medline]
Malinovsky L, Pac L. Is the Merkel cell a secondary sensory cell? [Review]. Z Mikrosk Anat Forsch 1985; 99: 119–28.[Web of Science][Medline]
Martin GN, Thom SA, Sever PS. The effects of ATP and related purines on human subcutaneous and omental resistance arteries. Brit J Pharm 1991; 102: 645–50.[Web of Science][Medline]
Ralevic V, Burnstock G. Receptors for purines and pyrimidines. [Review]. Pharmacol Rev 1998; 50: 413–92.
Ryan LM, Rachow JW, McCarty DJ. Synovial fluid ATP: a potential substrate for the production of inorganic pyrophosphate. J Rheumatol 1991; 18: 716–20.[Web of Science][Medline]
Salter MW, Henry JL. Effects of adenosine 5'-monophosphate and adenosine 5'-triphosphate on functionally identified units in the cat spinal dorsal horn. Evidence for a differential effect of adenosine 5'-triphosphate on nociceptive vs non-nociceptive units. Neuroscience 1985; 15: 815–25.[Web of Science][Medline]
Sneddon P, McLaren GJ, Kennedy C. Purinergic cotransmission: sympathetic nerves. Sem Neurosci 1996; 8: 201–5.
Wildman SS, King BF, Burnstock G. Potentiation of ATP-responses at a recombinant P2X2 receptor by neurotransmitters and related substances. Br J Pharmacol 1997; 120: 221–4.[Web of Science][Medline]
Received October 17, 1999. Revised December 14, 1999. Accepted December 20, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. G. Lechner and G. R. Lewin Peripheral sensitisation of nociceptors via G-proteindependent potentiation of mechanotransduction currents J. Physiol., July 15, 2009; 587(14): 3493 - 3503. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Ambalavanar and D. Dessem Emerging Peripheral Receptor Targets for Deep-tissue Craniofacial Pain Therapies Journal of Dental Research, March 1, 2009; 88(3): 201 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Whitsel, O. V. Favorov, Y. Li, M. Quibrera, and M. Tommerdahl Area 3a Neuron Response to Skin Nociceptor Afferent Drive Cereb Cortex, February 1, 2009; 19(2): 349 - 366. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Yu and W. C. de Groat Sensitization of pelvic afferent nerves in the in vitro rat urinary bladder-pelvic nerve preparation by purinergic agonists and cyclophosphamide pretreatment Am J Physiol Renal Physiol, May 1, 2008; 294(5): F1146 - F1156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Donnelly-Roberts, S. McGaraughty, C.-C. Shieh, P. Honore, and M. F. Jarvis Painful Purinergic Receptors J. Pharmacol. Exp. Ther., February 1, 2008; 324(2): 409 - 415. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Tashiro, K. Okamoto, S. B. Milam, and D. A. Bereiter Differential Effects of Estradiol on Encoding Properties of TMJ Units in Laminae I and V at the Spinomedullary Junction in Female Rats J Neurophysiol, December 1, 2007; 98(6): 3242 - 3253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Ren, X. Zou, L. Fang, and Q. Lin Involvement of Peripheral Purinoceptors in Sympathetic Modulation of Capsaicin-Induced Sensitization of Primary Afferent Fibers J Neurophysiol, November 1, 2006; 96(5): 2207 - 2216. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. K. Ford, M. Matchett, J. E. Krause, and W. Yu The P2X3 Antagonist P1, P5-Di[inosine-5'] Pentaphosphate Binds to the Desensitized State of the Receptor in Rat Dorsal Root Ganglion Neurons J. Pharmacol. Exp. Ther., October 1, 2005; 315(1): 405 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G. Jones, R. Slater, H. Cadiou, P. McNaughton, and S. B. McMahon Acid-Induced Pain and Its Modulation in Humans J. Neurosci., December 1, 2004; 24(48): 10974 - 10979. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Wynn, B. Ma, H. Z. Ruan, and G. Burnstock Purinergic component of mechanosensory transduction is increased in a rat model of colitis Am J Physiol Gastrointest Liver Physiol, September 1, 2004; 287(3): G647 - G657. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. North P2X3 receptors and peripheral pain mechanisms J. Physiol., January 15, 2004; 554(2): 301 - 308. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C Kennedy, T S Assis, A J Currie, and E G Rowan Crossing the pain barrier: P2 receptors as targets for novel analgesics J. Physiol., December 15, 2003; 553(3): 683 - 694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sawynok Topical and Peripherally Acting Analgesics Pharmacol. Rev., March 1, 2003; 55(1): 1 - 20. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Jarvis, E. C. Burgard, S. McGaraughty, P. Honore, K. Lynch, T. J. Brennan, A. Subieta, T. van Biesen, J. Cartmell, B. Bianchi, et al. A-317491, a novel potent and selective non-nucleotide antagonist of P2X3 and P2X2/3 receptors, reduces chronic inflammatory and neuropathic pain in the rat PNAS, December 24, 2002; 99(26): 17179 - 17184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. North Molecular Physiology of P2X Receptors Physiol Rev, October 1, 2002; 82(4): 1013 - 1067. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G.-Y. Xu and L.-Y. M. Huang Peripheral Inflammation Sensitizes P2X Receptor-Mediated Responses in Rat Dorsal Root Ganglion Neurons J. Neurosci., January 1, 2002; 22(1): 93 - 102. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. A. Chizh and P. Illes P2X Receptors and Nociception Pharmacol. Rev., December 1, 2001; 53(4): 553 - 568. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Pare, R. Elde, J. E. Mazurkiewicz, A. M. Smith, and F. L. Rice The Meissner Corpuscle Revised: A Multiafferented Mechanoreceptor with Nociceptor Immunochemical Properties J. Neurosci., September 15, 2001; 21(18): 7236 - 7246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Vlaskovska, L. Kasakov, W. Rong, P. Bodin, M. Bardini, D. A. Cockayne, A. P. D. W. Ford, and G. Burnstock P2X3 Knock-Out Mice Reveal a Major Sensory Role for Urothelially Released ATP J. Neurosci., August 1, 2001; 21(15): 5670 - 5677. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. G Hamilton, S. B McMahon, and G. R Lewin Selective activation of nociceptors by P2X receptor agonists in normal and inflamed rat skin J. Physiol., July 15, 2001; 534(2): 437 - 445. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||