Abstract

Findings

The purinergic P2X receptors, of which seven subtypes (P2X₁–P2X₇) have been cloned, are a family of ligand-gated cation channels activated by extracellular ATP. They have important roles in regulating neuronal and glial functions in the nervous system under physiological and pathological conditions [1-3]. There has been much recent attention paid to their roles in generating and modulating pain signaling [1-3]. Various attempts to block P2XRs pharmacologically or to suppress their expression molecularly and genetically have demonstrated that these receptors make a major contribution to pain responses evoked by tissue damage, chronic peripheral inflammation and nerve injury [4-6]. While the role of P2X₃R (and P2X₂₊₃R) has been studied extensively, recent evidence from our and other studies indicates that the P2X₄R subtype critically contributes to neuropathic pain, a highly debilitating pain condition that commonly occurs after nerve damage [7-13]. However, the precise role of P2X₄R in neuropathic pain remains to be fully determined because, in stark contrast to P2X₃R [14-16] and P2X₇R [17,18], there is a lack of sufficiently powerful pharmacological and genetic tools to selectively block P2X₄R signaling in vivo. Recently, several groups have independently developed lines of mice in which the p2rx4 gene is disrupted [19-21]. It is of particular importance to study the phenotypes of these mice to determine the in vivo functions of P2X₄R in pain signaling [22]. In the present study, we sought to characterize behavioral phenotypes in a wide range of assays of acute and chronic pain, including neuropathic pain, using a line of P2X₄R-deficient mice [19].

To examine acute physiological pain responses, the withdrawal responses from a noxious range of heat and mechanical stimuli applied by von Frey filaments were measured in both wild-type and p2rx4^-/- mice. In tail- and paw-flick tests, the latencies for animals to flick their tails and hindpaws away from radiant heat at either 30 V or 50 V were not different between wild-type and p2rx4^-/- mice (Fig. 1A, B). In a test of mechanical pain, p2rx4^-/- mice were indistinguishable from wild-type mice in terms of their paw withdrawal thresholds (Fig. ​(Fig.1C).1C). Abdominal writhing behavior in response to intraperitoneal injection with acetic acid, a model of chemical-induced visceral pain, was also comparable between the two lines of mice (Fig. ​(Fig.1D).1D). To determine the role of P2X₄R in tissue injury-induced acute and persistent pain, we assessed the pain response following the injection of formalin into the hindpaw. In wild-type mice, injection of formalin elicited biphasic biting and licking behaviors: the first phase started immediately after the injection and lasted for 5 min, and the second phase lasted for 60 min (Fig. ​(Fig.1E).1E). Neither the pattern nor the magnitude of biphasic behaviors was altered in p2rx4^-/- mice (Fig. ​(Fig.1E).1E). The amount of swelling of the formalin-injected hindpaw, as indicated by an increase in the weight of the hindpaw 60 min after the injection, was not significantly different between the two lines of mice (contralateral hindpaw: wild-type 0.137 ± 0.008 g, p2rx4^-/- 0.130 ± 0.004 g; ipsilateral: wild-type 0.173 ± 0.005 g, p2rx4^-/- 0.181 ± 0.005 g). These results indicate that pain signaling elicited by acute noxious stimuli and tissue damage is intact in p2rx4^-/- mice.

Open in a separate window

Figure 1

Acute thermal and mechanical pain and chemical-induced pain in p2rx4^-/- mice. (A) Tail-flick and (B) paw-flick tests. Values represent the latency (sec) for animals to flick their tail or paw away from the heat source (wild-type, n = 5; p2rx4^-/-, n = 4). (C) Mechanical pain test. Values indicate the threshold (g) to elicit paw withdrawal behavior in response to mechanical stimuli (wild-type, n = 5; p2rx4^-/-, n = 4). (D) Visceral pain in response to acetic acid (0.8%). Values represent the numbers of abdominal stretches (writhes) (wild-type, n = 6; p2rx4^-/-, n = 5). (E) Formalin test. Mice were injected intraplantarly with formalin (5%, 20 μL). Values represent the duration (sec) of licking and biting responses for each 5-min interval (E), from 0 to 5 min (1st phase) and for 10–60 min (2nd phase) (inset) (wild-type, n = 8; p2rx4^-/-, n = 6). All data are presented as means ± SEM.

Next, to investigate the role of P2X₄R in pain hypersensitivity under chronic pain conditions, we employed two distinct models of chronic pain, namely inflammatory and neuropathic pain. Wild-type mice with peripheral inflammation induced by intraplantar injection of complete Freund's adjuvant (CFA), used as a model of inflammatory pain, displayed decreased paw withdrawal thresholds in their ipsilateral hindpaws (Fig. ​(Fig.2A).2A). By contrast, the peripheral inflammation-induced decrease in paw withdrawal threshold was significantly smaller in p2rx4^-/- mice than wild-type mice. The paw withdrawal thresholds for the contralateral hindpaws in both genotypes of mice did not change (Fig. ​(Fig.2A).2A). In wild-type mice, a marked swelling of the ipsilateral hindpaw was observed, and the weight of the hindpaw was increased on day 14 (Fig. ​(Fig.2B).2B). In p2rx4^-/- mice, however, the CFA-induced increase in the weight of the ipsilateral hindpaw was significantly suppressed (p < 0.05, Fig. ​Fig.2B2B).

Open in a separate window

Figure 2

Reduction in the amount of peripheral inflammation-induced pain in p2rx4^-/- mice. (A) Paw withdrawal thresholds (left panel, ipsilateral hindpaw; right panel, contralateral hindpaw) of wild-type (n = 8) and p2rx4^-/- mice (n = 8) before (0) and 1, 3, 7, 10 and 14 days after intraplantar injection of CFA (0.01 mg/20 μL). *p < 0.05, **p < 0.01, ***p < 0.001 vs. wild-type mice. (B) Change in the weight (g) of the ipsilateral and contralateral hindpa0w 14 days after the injection of CFA (wild-type, n = 6; p2rx4^-/-, n = 6). *p < 0.05 vs. wild-type mice. All data are presented as means ± SEM.

To determine the role of P2X₄R in tactile allodynia under neuropathic pain conditions, we injured the forth lumbar spinal nerves of wild-type and p2rx4^-/- mice; this approach has been used to generate an animal model of neuropathic pain. While wild-type mice showed a decrease in paw withdrawal threshold after nerve injury (Fig. ​(Fig.3A),3A), this effect was markedly blunted in p2rx4^-/- mice. The attenuated tactile allodynia was observed until the last time point tested (day 1, p < 0.05; day 3, 7, 10 and 14, p < 0.001; Fig. ​Fig.3A).3A). The loss of P2X₄R did not change the paw withdrawal threshold of the contralateral hindpaw after nerve injury (Fig. ​(Fig.3A)3A) as seen in wild-type mice (except on day 1, p < 0.05 vs. wild-type, Fig. ​Fig.3A).3A). A reduction in pain behaviors is occasionally misinterpreted as a result of non-specific motor dysfunction, but the rotarod performance test demonstrated no significant difference in the time on the rotarod between the two genotypes (data not shown). We further tested whether P2X₄R deficiency also suppresses nerve injury-induced hypersensitivity to cold stimulation, a phenomenon known as cold allodynia. In striking contrast to the lack of nerve injury-induced tactile allodynia in p2rx4^-/- mice (Fig. ​(Fig.3A),3A), both wild-type and p2rx4^-/- mice showed markedly enhanced responsiveness to a cold stimulus evoked by applying acetone to the ipsilateral hindpaw after nerve injury (Fig. ​(Fig.3B).3B). There was no difference in the behavioral response to acetone in the contralateral hindpaw between the two lines of mice (wild-type, 3.2 ± 0.3; p2rx4^-/-, 3.2 ± 0.4). These results indicate that P2X₄R deficiency leads to a striking reduction in the tactile allodynia caused by nerve injury and, to the lesser extent, peripheral inflammation, without any defects in motor function.

Open in a separate window

Figure 3

Nerve injury-induced tactile allodynia is markedly blunted in p2rx4^-/- mice. Paw withdrawal thresholds (left panel, ipsilateral hindpaw; right panel, contralateral hindpaw) of wild-type (n = 5) and p2rx4^-/- mice (n = 6) before (0) and 1, 3, 7, 10 and 14 days after spinal nerve injury. (B) Cold allodynia evoked by applying acetone to the plantar surface of the ipsilateral hindpaw 7 days after nerve injury (wild-type, n = 5; p2rx4^-/-, n = 5). *p < 0.05, ***p < 0.001 vs. wild-type mice. All data are presented as means ± SEM.

By using P2X₄R-deficient mice, we revealed that P2X₄R is not required for all forms of pain responses, but rather is crucial for specialized pain states, namely chronic inflammatory and, in particular, neuropathic pain. Our present results showing the striking reduction of tactile allodynia after an injury to spinal nerves in p2rx4^-/- mice together with our previous findings in rats [7] and recent findings in a line of p2rx4^-/- mice after an injury to the sciatic nerve [22] provide compelling evidence for an essential role of P2X₄R in nerve injury-induced tactile allodynia. To elucidate the mechanisms underlying the blunted neuropathic allodynia in p2rx4^-/- mice requires further investigations, but it may involve a lack of microglial P2X₄Rs in the spinal cord, expression of which is markedly upregulated after nerve injury [7,22]. Stimulation of P2X₄R in microglia induces release of brain-derived neurotrophic factor [9,22,23], a factor that is crucial for producing aberrant excitability of dorsal horn neurons [9,24]. Therefore, the attenuated neuropathic allodynia in p2rx4^-/- mice may be associated with a reduction in the amount of pathologically altered neurotransmission in dorsal horn neurons caused by microglia-derived BDNF [22].

An interesting finding in the present study was that nerve injury-induced cold allodynia was retained in p2rx4^-/- mice, indicating that P2X₄R may have distinct roles in mechanical and cold hypersensitivities after nerve injury. In a recent study it was shown that inhibition of Src-family kinases (SFKs), whose activity is enhanced in spinal microglia after nerve injury, results in suppression of tactile allodynia without affecting cold allodynia [25]. Our previous study has implicated Lyn (microglial SFK) as a critical kinase causing P2X₄R upregulation [13]. It is thus conceivable that P2X₄R-dependent microglial signaling in the dorsal horn may participate predominantly in allodynia evoked by mechanical rather than cold stimulation after peripheral nerve injury.

Our present study showed a reduction in the level of CFA-induced inflammatory pain in p2rx4^-/- mice. In contrast to the neuropathic pain model, neither the upregulation of P2X₄R expression [7] nor activation of Lyn tyrosine kinase [13] has been demonstrated in spinal microglia following peripheral inflammation caused by CFA. The mechanisms underlying the reduction in the amount of inflammatory pain remain unknown, but the phenotype may be related to a reduction in hindpaw inflammation in p2rx4^-/- mice. This notion is supported by evidence of expression of functional P2X₄Rs in peripheral inflammatory cells, including macrophages [21,26].

In conclusion, by employing a series of acute and chronic pain tests using mice lacking P2X₄R, we demonstrated that the loss of P2X₄R leads to a marked reduction in the degree of tactile allodynia caused by nerve injury and, to a lesser extent, in peripheral inflammation, without any defects in motor function. By contrasting the roles of P2X₄R in distinct types of nerve injury-induced pain hypersensitivities, namely tactile and cold allodynia, the present study demonstrates a predominant contribution of P2X₄R to tactile allodynia rather than cold allodynia. The fact that acute physiological pain and tissue damage-induced pain were normal in p2rx4^-/- mice highlights the possibility of a therapeutic benefit of blocking P2X₄R in the treatment of chronic pain, especially tactile allodynia after nerve injury.

Methods

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

MT designed, performed and supervised the experiments, analyzed the data, and wrote the manuscript; KK and TI performed the experiments; KN, HST. analyzed the data; KI coordinated the project, helped to interpret the data, and edited the manuscript. All authors have read and approved the final manuscript.

Acknowledgements

References