Refine
Year of publication
Document Type
- Article (16)
- Doctoral Thesis (1)
Has Fulltext
- yes (17)
Is part of the Bibliography
- no (17)
Keywords
- dorsal root ganglia (3)
- axon bifurcation (2)
- mice (2)
- neuropathic pain (2)
- pain (2)
- spinal cord (2)
- Adenosine A1 receptor (1)
- Cyclic GMP (1)
- FRET imaging (1)
- Genetic transduction (1)
Institute
Die NO/cGMP-Kaskade spielt bei der nozizeptiven Transmission im Hinterhorn des Rückenmarks eine wichtige Rolle. In der vorliegenden Arbeit wurden bekannte cGMP-Targets (PKG-1, CNG-Kanäle, PDE-2 und -3) sowie synaptische Vesikelproteine (Synapsin 2, Rabphilin) als potentielle Targets der NO/cGMP-Kaskade mit Hilfe von molekularbiologischen Methoden und nozizeptiven Verhaltensstudien hinsichtlich einer Beteiligung an der nozizeptiven Transmission im Rückenmark untersucht. Im Formalintest reduzierte der PKG-1-Inhibitor Rp-8-Br-cGMPS (0,1 - 0,5 µmol i.t.) die nozizeptive Antwort, während der PKG-1-Aktivator 8-Br-cGMP in hoher Dosis (2,5 µmol i.t.) einen gegenteiligen Effekt zeigte. Überraschenderweise wirkte 8-Br-cGMP in niedriger Dosis (0,1 - 0,25 µmol i.t.) antinozizeptiv, was durch die gleichzeitige Applikation des PKG-Inhibitors weiter verstärkt wurde. Im Gegensatz zu Rp-8-Br-cGMPS oder 8-Br-cGMP beeinflussten weder der CNG-Kanal-Inhibitor L-cis-Diltiazem (0,5 mg i.t.), noch die PDE-Inhibitoren EHNA (0,25 µmol i.t.) oder Milrinon (5 - 10 mg/kg i.p.) die nozizeptive Antwort im Formalintest. Mit Western Blot-Analysen konnte gezeigt werden, dass die Formalininjektion in eine Hinterpfote im Lumbalmark nach 48 - 96 h eine Steigerung der PKG-1-Proteinkonzentration zur Folge hat. Dies wurde durch Vorbehandlung der Versuchstiere mit Rp-8-Br-cGMPS (0,1 - 0,5 µmol i.t.) oder Morphin (2,5 - 5 mg/kg i.p.) verhindert, während 8-Br-cGMP (2,5 µmol i.t.) die Formalin-induzierte Steigerung der PKG-1-Konzentration im Lumbalmark verstärkte. Die Formalininjektion in eine Hinterpfote veränderte auch die Synapsin 2b-Konzentration im Lumbalmark: 10 min bis 8 h nach der Injektion wurde die Synapsin 2b-Proteinkonzentration gesenkt, nach 48 h war jedoch eine Zunahme zu beobachten. Diese späte Zunahme der Synapsin 2b-Proteinkonzentration wurde durch eine Steigerung der Genexpression hervorgerufen, denn mit quantitativer Realtime RT-PCR wurden erhöhte mRNA-Konzentrationen 24 - 48 h nach der Formalininjektion gemessen. Die rasche Formalin-induzierte Abnahme der Synapsin 2b-Proteinkonzentration ging jedoch weder mit Änderungen der mRNA-Konzentration, noch mit veränderten Solubilisierungseigenschaften bei der Proteinaufbereitung einher, und wurde durch Vorbehandlung der Versuchstiere mit Morphin (10 mg/kg i.p.), Diclofenac (10 mg/kg i.p.), Metamizol (1 g/kg i.p.) oder dem NOS-Inhibitor L-NAME (10 - 100 mg/kg i.p.) verhindert. Demgegenüber führte die Vorbehandlung mit dem NO-Donor NOC-5 (4 - 20 µg i.t.), 8-Br-cGMP (0,1 - 2,5 µmol i.t.), Rp-8-Br-cGMPS (0,1 - 0,25 µmol i.t.) oder der Kombination von 8-Br-cGMP und Rp-8-Br-cGMPS (0,1 + 0,1 und 0,5 + 0,5 µmol i.t.) zu einer Verstärkung der Formalin-induzierten raschen Senkung der Synapsin 2b-Konzentration. Die funktionelle Relevanz dieser Befunde wurde in mehreren nozizeptiven Tiermodellen überprüft. Durch eine kontinuierliche i.t. Infusion von Antisense-Oligonukleotiden wurde die Synapsin 2-Konzentration im Lumbalmark der Ratte gesenkt, was eine Reduktion der nozizeptiven Antwort im Formalintest zur Folge hatte. Bei Synapsin 2-Knockout-Mäusen war im Vergleich zu Wildtyp-Mäusen eine verminderte nozizeptive Antwort im Formalintest und eine Reduktion der mechanischen Hyperalgesie bei Zymosan-induzierter Pfotenentzündung zu beobachten. Im Hot-Plate-Test zeigten die Knockout-Mäuse im Vergleich zu Wildtyp-Mäusen kürzere Latenzzeiten. Im Gegensatz zu Synapsin 2b wurde die Rabphilin-Konzentration im Lumbalmark durch Formalininjektion in eine Hinterpfote nicht beeinflusst. Allerdings führte die Verabreichung von Metamizol (500 mg/kg i.p.) oder Diclofenac (5 mg/kg i.p.) nach 1 h zu einer Steigerung der Rabphilin-Proteinkonzentration, welche nicht von Änderungen der mRNA-Expression begleitet war. Eine Senkung der Rabphilin-Proteinkonzentration wurde durch Applikation von NOC-5 (4 - 20 µg i.t.), 8-Br-cGMP (0,5 - 2,5 µmol i.t.), oder Rp-8-Br-cGMPS (0,1 - 0,25 µmol i.t.) hervorgerufen. Zusammenfassend bestätigen diese Ergebnisse die Hypothese, dass im Rückenmark PKG-1 einen Effektor der NO-induzierten Hyperalgesie darstellt. Darüber hinaus konnte gezeigt werden, dass NO/cGMP über noch unbekannte Mechanismen die Verfügbarkeit bestimmter synaptischer Vesikelproteine moduliert, die vor allem bei starker oder anhaltender nozizeptiver Erregung für die Transmitterausschüttung und damit für die nozizeptive Transmission notwendig sind. Interessanterweise führt NO/cGMP über diese Mechanismen eher zur Hemmung der Nozizeption, was die bei niedrigen intrathekalen Dosen beobachteten antinozizeptiven Effekte von 8-Br-cGMP erklären kann. Die These der NO-induzierten Hyperalgesie kann aufgrund der Untersuchungen in dieser Arbeit und früherer Studien um eine insbesondere in niedriger Dosis auftretende NO/cGMP-vermittelte Antinozizeption erweitert werden.
Although the Nobel Prize for the discovery of nitric oxide (NO) dates back almost 20 years now, the knowledge about cGMP signaling is still constantly increasing. It looks even so that our understanding of the role of the soluble guanylyl cyclase (sGC) and particulate guanylyl cyclase (pGC) in health and disease is in many aspects at the beginning and far from being understood. This holds even true for the therapeutic impact of innovative drugs acting on both the NO/sGC and the pGC pathways. Since cGMP, as second messenger, is involved in the pathogenesis of numerous diseases within the cardiovascular, pulmonary, renal, and endocrine systems and also plays a role in neuronal, sensory, and tumor processes, drug applications might be quite broad. On the 8th International Conference on cGMP, held in Bamberg, Germany, world leading experts came together to discuss these topics. All aspects of cGMP research from the basic understanding of cGMP signaling to clinical applicability were discussed in depth. In addition, present and future therapeutic applications of cGMP-modulating pharmacotherapy were presented (http://www.cyclicgmp.net/index.html).
Impaired NO-cGMP signaling has been linked to several neurological disorders. NO-sensitive guanylyl cyclase (NO-GC), of which two isoforms—NO-GC1 and NO-GC2—are known, represents a promising drug target to increase cGMP in the brain. Drug-like small molecules have been discovered that work synergistically with NO to stimulate NO-GC activity. However, the effects of NO-GC stimulators in the brain are not well understood. In the present study, we used Förster/fluorescence resonance energy transfer (FRET)-based real-time imaging of cGMP in acute brain slices and primary neurons of cGMP sensor mice to comparatively assess the activity of two structurally different NO-GC stimulators, IWP-051 and BAY 41-2272, in the cerebellum, striatum and hippocampus. BAY 41-2272 potentiated an elevation of cGMP induced by the NO donor DEA/NO in all tested brain regions. Interestingly, IWP-051 potentiated DEA/NO-induced cGMP increases in the cerebellum and striatum, but not in the hippocampal CA1 area or primary hippocampal neurons. The brain-region-selective activity of IWP-051 suggested that it might act in a NO-GC isoform-selective manner. Results of mRNA in situ hybridization indicated that the cerebellum and striatum express NO-GC1 and NO-GC2, while the hippocampal CA1 area expresses mainly NO-GC2. IWP-051-potentiated DEA/NO-induced cGMP signals in the striatum of NO-GC2 knockout mice but was ineffective in the striatum of NO-GC1 knockout mice. These results indicate that IWP-051 preferentially stimulates NO-GC1 signaling in brain slices. Interestingly, no evidence for an isoform-specific effect of IWP-051 was observed when the cGMP-forming activity of whole brain homogenates was measured. This apparent discrepancy suggests that the method and conditions of cGMP measurement can influence results with NO-GC stimulators. Nevertheless, it is clear that NO-GC stimulators enhance cGMP signaling in the brain and should be further developed for the treatment of neurological diseases.
Neuropathic pain is a debilitating and commonly treatment-refractory condition requiring novel therapeutic options. Accumulating preclinical studies indicate that the potassium channel Slack (KNa1.1) contributes to the processing of neuropathic pain, and that Slack activators, when injected into mice, ameliorate pain-related hypersensitivity. However, whether Slack activation might reduce neuropathic pain in humans remains elusive. Here, we evaluated the tolerability and analgesic efficacy of loxapine, a first-generation antipsychotic drug and Slack activator, in neuropathic pain patients. We aimed to treat 12 patients with chronic chemotherapy-induced, treatment-refractory neuropathic pain (pain severity ≥ 4 units on an 11-point numerical rating scale) in a monocentric, open label, proof-of-principle study. Patients received loxapine orally as add-on analgesic in a dose-escalating manner (four treatment episodes for 14 days, daily dose: 20, 30, 40, or 60 mg loxapine) depending on tolerability and analgesic efficacy. Patient-reported outcomes of pain intensity and/or relief were recorded daily. After enrolling four patients, this study was prematurely terminated due to adverse events typically occurring with first-generation antipsychotic drugs that were reported by all patients. In two patients receiving loxapine for at least two treatment episodes, a clinically relevant analgesic effect was found at a daily dose of 20–30 mg of loxapine. Another two patients tolerated loxapine only for a few days. Together, our data further support the hypothesis that Slack activation might be a novel strategy for neuropathic pain therapy. However, loxapine is no valid treatment option for painful polyneuropathy due to profound dopamine and histamine receptor-related side effects.
Clinical Trial Registration: www.ClinicalTrials.gov, identifier NCT02820519.
More than 70 human adenoviruses with type-dependent pathogenicity have been identified but biological information about the majority of these virus types is scarce. Here we employed multiple sequence alignments and structural information to predict receptor usage for the development of an adenoviral vector with novel biological features. We report the generation of a cloned adenovirus based on human adenovirus type 17 (HAdV17) with high sequence homology to the well characterized human adenovirus type 37 (HAdV37) that causes epidemic keratoconjunctivitis (EKC). Our study revealed that human CD46 (CD46) is involved in cell entry of HAdV17. Moreover, we found that HAdV17 infects endothelial cells (EC) in vitro including primary cells at higher efficiencies compared to the commonly used human adenovirus type 5 (HAdV5). Using a human CD46 transgenic mouse model, we observed that HAdV17 displays a broad tropism in vivo after systemic injection and that it transduces ECs in this mouse model. We conclude that the HAdV17-based vector may provide a novel platform for gene therapy.
Tissue injury and inflammation may result in chronic pain, a severe debilitating disease that is associated with great impairment of quality of life. An increasing body of evidence indicates that members of the Rab family of small GTPases contribute to pain processing; however, their specific functions remain poorly understood. Here, we found using immunofluorescence staining and in situ hybridization that the small GTPase Rab27a is highly expressed in sensory neurons and in the superficial dorsal horn of the spinal cord of mice. Rab27a mutant mice, which carry a single-nucleotide missense mutation of Rab27a leading to the expression of a nonfunctional protein, show reduced mechanical hyperalgesia and spontaneous pain behavior in inflammatory pain models, while their responses to acute noxious mechanical and thermal stimuli is not affected. Our study uncovers a previously unrecognized function of Rab27a in the processing of persistent inflammatory pain in mice.
A cGMP signaling cascade composed of C-type natriuretic peptide, the guanylyl cyclase receptor Npr2 and cGMP-dependent protein kinase I (cGKI) controls the bifurcation of sensory axons upon entering the spinal cord during embryonic development. However, the impact of axon bifurcation on sensory processing in adulthood remains poorly understood. To investigate the functional consequences of impaired axon bifurcation during adult stages we generated conditional mouse mutants of Npr2 and cGKI (Npr2fl/fl;Wnt1Cre and cGKIKO/fl;Wnt1Cre) that lack sensory axon bifurcation in the absence of additional phenotypes observed in the global knockout mice. Cholera toxin labeling in digits of the hind paw demonstrated an altered shape of sensory neuron termination fields in the spinal cord of conditional Npr2 mouse mutants. Behavioral testing of both sexes indicated that noxious heat sensation and nociception induced by chemical irritants are impaired in the mutants, whereas responses to cold sensation, mechanical stimulation, and motor coordination are not affected. Recordings from C-fiber nociceptors in the hind limb skin showed that Npr2 function was not required to maintain normal heat sensitivity of peripheral nociceptors. Thus, the altered behavioral responses to noxious heat found in Npr2fl/fl;Wnt1Cre mice is not due to an impaired C-fiber function. Overall, these data point to a critical role of axonal bifurcation for the processing of pain induced by heat or chemical stimuli.
Cyclic GMP (cGMP) is a second messenger that regulates numerous physiological and pathophysiological processes. In recent years, more and more studies have uncovered multiple roles of cGMP signalling pathways in the somatosensory system. Accumulating evidence suggests that cGMP regulates different cellular processes from embryonic development through to adulthood. During embryonic development, a cGMP-dependent signalling cascade in the trunk sensory system is essential for axon bifurcation, a specific form of branching of somatosensory axons. In adulthood, various cGMP signalling pathways in distinct cell populations of sensory neurons and dorsal horn neurons in the spinal cord play an important role in the processing of pain and itch. Some of the involved enzymes might serve as a target for future therapies. In this review, we summarise the knowledge regarding cGMP-dependent signalling pathways in dorsal root ganglia and the spinal cord during embryonic development and adulthood, and the potential of targeting these pathways.
LINKED ARTICLES
This article is part of a themed issue on cGMP Signalling in Cell Growth and Survival. To view the other articles in this section visit http://onlinelibrary.wiley.com/doi/10.1111/bph.v179.11/issuetoc
Inflammation or injury to the somatosensory nervous system may result in chronic pain conditions, which affect millions of people and often cause major health problems. Emerging lines of evidence indicate that reactive oxygen species (ROS), such as superoxide anion or hydrogen peroxide, are produced in the nociceptive system during chronic inflammatory and neuropathic pain and act as specific signaling molecules in pain processing. Among potential ROS sources in the somatosensory system are NADPH oxidases, a group of electron-transporting transmembrane enzymes whose sole function seems to be the generation of ROS. Interestingly, the expression and relevant function of the Nox family members Nox1, Nox2, and Nox4 in various cells of the nociceptive system have been demonstrated. Studies using knockout mice or specific knockdown of these isoforms indicate that Nox1, Nox2, and Nox4 specifically contribute to distinct signaling pathways in chronic inflammatory and/or neuropathic pain states. As selective Nox inhibitors are currently being developed and investigated in various physiological and pathophysiological settings, targeting Nox1, Nox2, and/or Nox4 could be a novel strategy for the treatment of chronic pain. Here, we summarize the distinct roles of Nox1, Nox2, and Nox4 in inflammatory and neuropathic processing and discuss the effectiveness of currently available Nox inhibitors in the treatment of chronic pain conditions.
Accumulating evidence indicates that increased generation of reactive oxygen species (ROS) contributes to the development of exaggerated pain hypersensitivity during persistent pain. In the present study, we investigated the antinociceptive efficacy of the antioxidants vitamin C and vitamin E in mouse models of inflammatory and neuropathic pain. We show that systemic administration of a combination of vitamins C and E inhibited the early behavioral responses to formalin injection and the neuropathic pain behavior after peripheral nerve injury, but not the inflammatory pain behavior induced by Complete Freund's Adjuvant. In contrast, vitamin C or vitamin E given alone failed to affect the nociceptive behavior in all tested models. The attenuated neuropathic pain behavior induced by the vitamin C and E combination was paralleled by a reduced p38 phosphorylation in the spinal cord and in dorsal root ganglia, and was also observed after intrathecal injection of the vitamins. Moreover, the vitamin C and E combination ameliorated the allodynia induced by an intrathecally delivered ROS donor. Our results suggest that administration of vitamins C and E in combination may exert synergistic antinociceptive effects, and further indicate that ROS essentially contribute to nociceptive processing in special pain states.
Oral presentation from 4th International Conference of cGMP Generators, Effectors and Therapeutic Implications ; Regensburg, Germany. 19–21 June 2009 Background: An exaggerated pain sensitivity is the dominant feature of inflammatory and neuropathic pain both in the clinical setting and in experimental animal models. It manifests as pain in response to normally innocuous stimuli (allodynia), increased response to noxious stimuli (hyperalgesia) or spontaneous pain, and can persist long after the initial injury is resolved. Research over the last decades has revealed that several signaling pathways in the spinal cord essentially contribute to the pain sensitization. To test the contribution of cGMP produced by NO-sensitive guanylyl cyclase (NO-GC) to pain sensitization, we investigated the localization of NO-GC in the spinal cord and in dorsal root ganglia, and we characterized the nociceptive behavior of mice deficient in NO-GC (GC-KO mice). Results: We show that NO-GC (β1 subunit) is distinctly expressed in neurons of the mouse spinal cord, while its distribution in dorsal root ganglia is restricted to non-neuronal cells. GC-KO mice exhibited a considerably reduced nociceptive behavior in models of inflammatory or neuropathic pain, but their responses to acute pain were not impaired. Moreover, GC-KO mice failed to develop pain sensitization induced by spinal administration of drugs releasing NO. Surprisingly, during spinal nociceptive processing cGMP produced by NO-GC may activate signaling pathways different from cGMP-dependent protein kinase I (cGKI), while cGKI can be activated by natriuretic peptide receptor-B (NPR-B) dependent cGMP production. Conclusion: Taken together, our results provide evidence that NO-GC has a dominant role in the development of exaggerated pain sensitivity during inflammatory and neuropathic pain. Furthermore, beside the NO-mediated cGMP synthesis, cGMP produced by NPR-B contributes to pain sensitization by activation of cGKI.
First paragraph (this article has no abstract) Persistent stimulation of nociceptors results in sensitization of nociceptive sensory neurons, which is associated with hyperalgesia and allodynia. The release of NO and subsequent synthesis of cGMP in the spinal cord are involved in this process. cGMP-dependent protein kinase I (PKG-I) has been suggested to act as a downstream target of cGMP, but its exact role in nociception hadn't been characterized yet. To further evaluate the NO/cGMP/PKG-I pathway in nociception we assessed the effects of PKG-I inhibiton and activaton in the rat formalin assay and analyzed the nociceptive behavior of PKG-I-/- mice. Open access article.
The experience of pain is mediated by a specialized sensory system, the nociceptive system. There is considerable evidence that the cGMP/cGMP kinase I (cGKI) signaling pathway modulates the nociceptive processing within the spinal cord. However, downstream targets of cGKI in this context have not been identified to date. In this study we investigated whether cysteine-rich protein 2 (CRP2) is a downstream effector of cGKI in the spinal cord and is involved in nociceptive processing. Immunohistochemistry of the mouse spinal cord revealed that CRP2 is expressed in superficial laminae of the dorsal horn. CRP2 is colocalized with cGKI and with markers of primary afferent C fibers. Importantly, the majority of CRP2 mRNA-positive dorsal root ganglion (DRG) neurons express cGKI and CRP2 is phosphorylated in a cGMP-dependent manner. To elucidate the functional role of CRP2 in nociception, we investigated the nociceptive behavior of CRP2-deficient (CRP2-/-) mice. Touch perception and acute thermal nociception were unaltered in CRP2-/- mice. However, CRP2-/- mice showed an increased nociceptive behavior in models of persistent pain as compared to wild type mice. Intrathecal administration of cGKI activating cGMP analogs increased the nociceptive behavior in wild type but not in CRP2-/- mice, indicating that the presence of CRP2 was essential for cGMP/cGKI-mediated nociception. These data indicate that CRP2 is a new downstream effector of cGKI-mediated spinal nociceptive processing and point to an inhibitory role of CRP2 in the generation of inflammatory pain.
Lack of efficacy of a partial adenosine A1 receptor agonist in neuropathic pain models in mice
(2021)
Previous studies suggest that adenosine A1 receptors (A1R) modulate the processing of pain. The aim of this study was to characterize the distribution of A1R in nociceptive tissues and to evaluate whether targeting A1R with the partial agonist capadenoson may reduce neuropathic pain in mice. The cellular distribution of A1R in dorsal root ganglia (DRG) and the spinal cord was analyzed using fluorescent in situ hybridization. In behavioral experiments, neuropathic pain was induced by spared nerve injury or intraperitoneal injection of paclitaxel, and tactile hypersensitivities were determined using a dynamic plantar aesthesiometer. Whole-cell patch-clamp recordings were performed to assess electrophysiological properties of dissociated DRG neurons. We found A1R to be expressed in populations of DRG neurons and dorsal horn neurons involved in the processing of pain. However, administration of capadenoson at established in vivo doses (0.03–1.0 mg/kg) did not alter mechanical hypersensitivity in the spared nerve injury and paclitaxel models of neuropathic pain, whereas the standard analgesic pregabalin significantly inhibited the pain behavior. Moreover, capadenoson failed to affect potassium currents in DRG neurons, in contrast to a full A1R agonist. Despite expression of A1R in nociceptive neurons, our data do not support the hypothesis that pharmacological intervention with partial A1R agonists might be a valuable approach for the treatment of neuropathic pain.
The transient receptor potential (TRP) ankyrin type 1 (TRPA1) channel is highly expressed in a subset of sensory neurons where it acts as an essential detector of painful stimuli. However, the mechanisms that control the activity of sensory neurons upon TRPA1 activation remain poorly understood. Here, using in situ hybridization and immunostaining, we found TRPA1 to be extensively co-localized with the potassium channel Slack (KNa1.1, Slo2.2, or Kcnt1) in sensory neurons. Mice lacking Slack globally (Slack−/−) or conditionally in sensory neurons (SNS-Slack−/−) demonstrated increased pain behavior after intraplantar injection of the TRPA1 activator allyl isothiocyanate. By contrast, pain behavior induced by the TRP vanilloid 1 (TRPV1) activator capsaicin was normal in Slack-deficient mice. Patch-clamp recordings in sensory neurons and in a HEK cell line transfected with TRPA1 and Slack revealed that Slack-dependent potassium currents (IKS) are modulated in a TRPA1-dependent manner. Taken together, our findings highlight Slack as a modulator of TRPA1-mediated, but not TRPV1-mediated, activation of sensory neurons.
Keywords: TRPA1; slack; dorsal root ganglia; pain; mice
Functional coupling of Slack channels and P2X3 receptors contributes to neuropathic pain processing
(2021)
The sodium-activated potassium channel Slack (KNa1.1, Slo2.2, or Kcnt1) is highly expressed in populations of sensory neurons, where it mediates the sodium-activated potassium current (IKNa) and modulates neuronal activity. Previous studies suggest that Slack is involved in the processing of neuropathic pain. However, mechanisms underlying the regulation of Slack activity in this context are poorly understood. Using whole-cell patch-clamp recordings we found that Slack-mediated IKNa in sensory neurons of mice is reduced after peripheral nerve injury, thereby contributing to neuropathic pain hypersensitivity. Interestingly, Slack is closely associated with ATP-sensitive P2X3 receptors in a population of sensory neurons. In vitro experiments revealed that Slack-mediated IKNa may be bidirectionally modulated in response to P2X3 activation. Moreover, mice lacking Slack show altered nocifensive responses to P2X3 stimulation. Our study identifies P2X3/Slack signaling as a mechanism contributing to hypersensitivity after peripheral nerve injury and proposes a potential novel strategy for treatment of neuropathic pain.