Nerve growth factor-induced hyperexcitability of rat sensory neuron in ...

Biomedical Research 26 (3) 123-130, 2005

Nerve growth factor-induced hyperexcitability of rat sensory neuron in culture Naoki KITAMURA1, Akihiro KONNO2, Takeshi KUWAHARA1 and You KOMAGIRI1 1

Laboratory of Physiology and 2 Laboratory of Anatomy, Department of Biomedical Sciences, Graduate School of Veterinary Medicine, Hokkaido University, North 18th West 9th, Sapporo 0600818, Japan (Received 25 April 2005; and accepted 6 May 2005)

ABSTRACT Abnormal spontaneous firing of primary sensory neurons is considered to be a cause of neuropathic pain. However, pathogenic mechanisms of hyperexcitable sensory neurons in neuropathic model animals are unclear. We examined effects of chronic treatment of nerve growth factor (NGF), one of candidate mediators for the pathogenesis, on excitability of sensory neurons by voltage-clamped recording in a cell-attached configuration. From rat dorsal root ganglion (DRG) neurons cultured without NGF, only stable holding currents without spontaneous firing activity were recorded. On the other hand, more than 20% neurons cultured in the presence of NGF for more than 3 days showed spontaneous current spikes at frequencies between 0.1 and 5 Hz. Each spikes had an initial inward phase followed by the outward phase, resulted from spontaneous transient depolarization followed by transient hyperpolarization. These spontaneous spikes were abolished by tetrodotoxin, lidocaine and reduction of extracellular concentration of Na+ from 154 mM to 100 mM, in all-or-none fashion, suggesting that spontaneous current spikes reflected spontaneous action potentials. From these results, it became evident that DRG neurons of adult rats had a nature to respond to NGF and obtained the abnormal hyperexcitability to fire spontaneously.

Peripheral nerve injury frequently results in severe chronic pain, neuropathic pain that is poorly responsive to general analgesic therapy, and remains as an unsolved problem in health. Experimental injury of the peripheral nerve in experimental animals induces hyperalgesia and allodynia (2, 5, 13, 22, 31, 32). Although primary sensory neurons of healthy animals are “silent” in electrophysiology without excitatory inputs from other neurons or sensational inputs, abnormal spontaneous firing was observed in sensory neurons of animals with nerve injury (6, 11, Address correspondence to: Naoki Kitamura, DVM, PhD, Laboratory of Physiology, Department of Biomedical Sciences, Graduate School of Veterinary Medicine, Hokkaido University, North 18th West 9 th, Sapporo 0600818, Japan Tel: +81-11-706-5201, Fax: +81-11-706-5202 E-mail: [email protected]

12, 18). Such abnormal firing is considered to be a cause of neuropathic pain. Such spontaneous firing was observed at the local injured site and also at the soma of the dorsal root ganglion (DRG) neurons (30). Subthreshold membrane potential oscillation and the spontaneously generated action potentials were recorded in DRG neurons freshly isolated from nerveinjured rats by the intracellular microelectrode method (1, 17, 26) and the whole-cell patch clamp method in the current clamp mode (20, 27). Nevertheless, the pathogenic mechanisms of the abnormal hyperexcitability of the primary sensory neurons are unclear. Concentration of nerve growth factor (NGF) in DRG was reported to increase after the artificial nerve injury (9), and injection of exogenous NGF into healthy rats induced hyperalgesia (16). These reports suggest that NGF correlates with the pathogenesis of neuropathic pain. Small DRG neurons, which respond to noxious sensation, express TrkA

124

(19, 28, 29) and p75NTR (21) as receptors for NGF, and can be potential targets of the action of NGF. Therefore, cultured DRG neurons isolated from healthy adult rats were chronically treated with NGF by adding it into culture medium, and chronic effects of NGF on electrophysiological activity of neurons were examined. MATERIALS AND METHODS Cell isolation and culture. Rat DRG neurons were isolated from adult male Sprague-Dawley rats (7–12 weeks old), using procedures that have been reported previously (14). Rats were killed by cervical dislocation after inhalation anesthesia by diethyl ether. Ganglia were dissected from the full length of the vertebral column. Axons extending from ganglia were removed as far as possible under the stereoscopic microscope. The ganglia were incubated at 37°C first in calcium-magnesium free phosphatebuffered saline (CMF-PBS) containing collagenase type IV (250 U/ml, Worthington Biochemichals, Lakewood, NJ, USA), DNase I (0.12 μg/ml, Sigma, St Louis, MO, USA) and bovine serum albumin (BSA, 1 mg/ml, Sigma) for 2 h and then rinsed with CFM-PBS to remove collagenase. Next, ganglia were incubated in CMF-PBS containing trypsin (0.25% w/v, Invitrogen, Carlsbad, CA, USA) and BSA (1 mg/ml, Sigma) for 15 min. After the enzymatic digestion, cells were gently agitated with a silicon-coated Pasteur pipette and centrifuged to remove the enzymes. The isolated cells were suspended in Dulbecco’s modified Eagle’s Medium (DMEM, Invitrogen) containing 4.5 g/l of glucose and cultured on coverslips coated with poly-D-lysine (Sigma). The cells were kept at 37°C in a humidified atmosphere of 95% air and 5% CO2 and cultured until use in each experiment. DMEM was supplemented with 10% fetal bovine serum (MP Biochemicals, Irvine, CA, USA), 100 U/ml of penicillin (Sigma), 100 ng/ml of streptomycin (Sigma), and 10-μM cytosine arabinoside (Sigma). The culture medium was changed every 2 days. NGF-7S (100 ng/ml, Sigma) was added to the medium 3–5 days after the isolation of neurons. Neurons were used in the electrophysiological and immunocytochemical experiments after 4–14 days of culture. Immunohistochemistry. Bouin-fixed and paraffin-embedded tissue sections prepared from the DRG of L3-L5 nerves and 4% paraformaldehyde-fixed 4– 10-day-cultured DRG neurons on the coverslips were subjected to immunoperoxidase staining. Tis-

N. Kitamura et al.

sue sections of 3 μm thickness were dipped in 3% H2O2/methanol for 1 h to inactivate endogenous peroxidase activity, blocked with 10% normal goat serum for 1 h and then incubated with rabbit antiTrkA antibody (Upstate Biotechnology, Lake Placid, NY, USA, Cat#: 12-301, Lot#: 22778) at 10 μg/ml and anti-p75 NTR antibody (Sigma, Cat#: N3908, Lot#: 049H4843) at 2.5 μg/ml for 2 h. In negative control sections, non-immune rabbit IgG (Molecular Probes, Eugene, OR, USA) at 20 μg/ml was applied. Sections were subsequently incubated with biotinylated goat anti-rabbit IgG followed by streptavidineperoxidase complex (Histofine Kit, Nichirei, Tokyo, Japan). Finally sections were developed with diaminobenzidine and counterstained with hematoxylin. Cultured neuron specimens were permeabilized with 0.1% Triton X-100 for 5 min, treated in 3% H2O2/ methanol, blocked with 10% normal goat serum, and then incubated with anti-TrkA and anti-p75NTR antibodies at 2 μg/ml and 0.5 μg/ml, respectively. Subsequently culture specimens were incubated with peroxidase-conjugated goat anti-rabbit IgG (Molecular Probes) at 1 μg/ml, developed with diaminobenzidine and counterstained with hematoxylin. Electrophysiology. DRG neurons with small diameter (10–30 μm) were used to record current responses. In order to record electrophysiological activity of intact neurons whose cytosol was not dialyzed with an artificial physiological solution, a conventional voltage-clamped recording in a cell-attached configuration was made at room temperature (22–25°C). Heat-polished glass electrodes with a 2- to 3-MΩ tip resistance were used. The pipette solution contained (mM): 151.6 K-methansulfonic acid; 3.4 KCl; 5 Na-methansulfonic acid; 2 MgCl2; 1.3 CaCl2; 10 ethylene glycol-bis (β-aminoethylether)-N, N, N’, N’-tetraacetic acid (EGTA, Sigma); 10 N- [2-hydeoxyethyl] piperazine-N’- [2-ethanesulfonic acid] (HEPES, pH = 7.3 adjusted with methansulfonic acid). The calculated concentration of free Ca2+ in this pipette solution was 10−8 M. The normal bath solution contained (mM): 144 NaCl; 10 NaOH; 6 KCl; 1.2 MgCl 2 ; 2.5 CaCl 2 ; 10 D-glucose; 10 HEPES (Sigma, pH = 7.4 adjusted with HCl). Na+-reduced solutions were made by isotonic replacement of Na + with N-methyl-D-glucamine + (NMDG+, Merck, Darmstadt, Germany). The liquid junction potential between the pipette solution and the bath solution (approximately −10 mV) was corrected. When the concentrations of extracellular Na+ or K+ were elevated, the normal bath and pipette solutions containing additional NMDG-Cl at 40 mM

NGF-induced hyperexcitablity

were used as control solutions, and the concentrations of Na+ or K+ were elevated by the isotonic replacement of NMDG+ in the bath solutions. Neurons were continuously superfused at a flow rate of 1 ml/ min throughout the experiments. Currents were measured with a patch-clamp amplifier (CEZ-2400, Nihon Koden, Tokyo, Japan) and data acquisition was performed at sampling frequency of 40 kHz by a personal computer (Macintosh, Apple, Cupertino, CA, USA) in conjunction with an analog/digital converter (Power Lab, AD Instruments, Castle Hill, NSW, Australia). Neither an external filter nor a built-in filter was used because of the fast gating property of focused currents. Data were analyzed with Chart software (AD Instruments), Igor Pro (Wavemetrics, Lake Oswego, OR, USA) and Excel (Microsoft, Redmond, WA, USA). Data are presented as means ± standard errors of the mean (s.e.m.). The significance of differences between the means of two experiments was assessed with Student’s t-test. The level of significance chosen was 0.05. Drugs. A concentrated stock solution of NGF-7S at 10 μg/ml was made by dissolving in DMEM and stored at −20°C until use. Na-methansulfonic acid and K-methansulfonic acid were made from NaOH and KOH by mixing methansulfonic acid at 1 : 1, respectively. All other chemicals used were of analytical grade. RESULTS AND DISCUSSION Cultured adult rat DRG neurons express TrkA and p75NTR DRG neurons had been reported to express TrkA acting as a high affinity receptor for NGF (19, 28, 29), and a low affinity NGF receptor, p75NTR (21). At first, we examined whether the expression of these NGF receptors was maintained in cultured sensory neurons isolated from adult rats. Immunohistochemistry showed immunoreactivity for TrkA and p75NTR in sections of DRG collected from rats immediately after killing (Fig. 1A, B, C) and in dissociated neurons after 7 days culture (Fig. 1D, E, F). Immunoreactivity for TrkA (Fig. 1A) and p75NTR (Fig. 1B) was observed in the part of neurons in sections, consistently with the previous reports. Under the same staining condition, non-specific immunoreactivity tested by utilizing non-immune rabbit IgG was not observed (Fig. 1C). Similar to the immunohistology of DRG sections, immunoreactivity for both NGF receptors was observed in the part of cultured neurons isolated from rat DRG. DRG neu-

125

rons were generally classified into three groups in accordance with cell sizes. C-neurons were found in neurons with smaller diameter and A-neurons with large diameter (7, 8). It is well known that neurons with small diameter possess axons with small conduction velocity (C-fiber) and are responsible to noxious stimuli to the skin, while A-neurons possessing myelinated A-fibers are mainly responsible to innoxious stimuli. Since, during the staining process, many neurons were removed from coverslips, it was difficult to compare size distribution of neurons having positive immunoreactivity between native and cultured neurons. However, these results suggested that the expression of TrkA and p75NTR was maintained in rat DRG neurons cultured under the condition in this study, and NGF added to the culture medium was possible to act directly on DRG neurons. NGF-treated DRG neurons show spontaneous discharges In the below experiments, rat DRG neurons cultured in the absence and presence of NGF (NGF-7S, 100 ng/ml) were used for the patch clamp recording. The size of used neurons was 10–30 μm in diameter that were generally classified into C-type neurons (7, 8). As shown in Fig. 1, in part of these neurons, the expression of TrkA and p75NTR was suggested. After a giga ohm seal was established, a cell-attached recording was made. The electrode potential was clamped at 0 mV. Figs. 2A and B show typical traces of current responses recorded from a neuron cultured in the absence of NGF (Fig. 2A), and from three independent neurons cultured in the presence of NGF for 2 days (Fig. 2B), respectively. In almost all control neurons untreated with NGF (81 out of 84 neurons), only stable holding current with the stable noise of about 20 pA was observed in the absence of the artificial electrical stimulations under the on-cell recording condition, regardless of the culturing period after the isolation of neurons. In contrast, in NGF-treated neurons, spontaneously occurring spike-like currents were observed in the cell-attached configuration. Such spontaneous currents were recorded in 165 of 822 neurons treated with NGF for more than 1 day, and in only 3 of 84 untreated neurons. An expanded current trace of typical spontaneous events is shown in Fig. 2C. In these experiments using the K+-rich pipette solution, the amplitudes of the currents observed in the cellattached configuration were directly proportional to the membrane potential and the differential of change in the membrane potential of the target neu-

126

N. Kitamura et al.

Fig. 1 Immunohistochemistry of DRG sections (A-C) and cultured DRG neurons (D-F) of rats. Immunoreactivities for TrkA (A, D) and p75NTR (B, E) are observed in the part of neurons. Arrows and arrow heads indicate the neurons with positive and negative immunoreactivities, respectively. C and F show images of the negative control preparations confirmed by utilizing non-immune primary antibody. Non-specific immunoreactivities are not observed. Scale bars = 100 µm (A-C) and 50 µm (D-F).

Fig. 2 Cell-attached current traces recorded from NGF-treated and -untreated DRG neurons. A. A typical current trace in control neurons cultured in the absence of NGF. B. Typical traces in 3 independent neurons cultured in the presence of NGF for more than 1 day. C. An expanded current trace of spontaneous spikes observed in NGF-treated neurons.


rons. Inward currents indicated the depolarization of neurons and outward currents, the hyperpolarization. The initial inward current followed by the outward current overcoming the original baseline level was seen in every current spikes. The mean amplitudes of the inward and outward currents in randomly selected 68 neurons were −54.1 ± 3.5 pA and 22.4 ± 1.9 pA, respectively. These transient current spikes recorded repetitively and continuously in the cell-attached configuration indicated that spontaneous depolarizing events occurred in intact NGF-treated neurons without intracellular dialysis by the artificial physiological salt solution. Because of the fast gating property of the spontaneous current spikes, a high-cut filter could not be applied. In the present experiments, the noise level of the holding current was approximately 20 pA, and the lower limit of the amplitudes of detectable fast gating current spikes was approximately 10 pA. The amplitudes of inward and outward currents of the spontaneous events were variable from neuron to neuron. However, they were relatively stable in individual neurons. In contrast to the amplitudes, the durations were relatively stable among neurons, and the mean duration, which was defined as a time between the rising point of the inward phase and the ending point of the outward phase, was 11.6 ± 1.3 ms (n = 23). Spontaneous discharges reflect action potentials The spontaneous depolarizing events under the cellattached configuration had such a rapid kinetics, suggesting that they were resulted from action potentials. It is well known that the initiation of the action potentials is highly sensitive to change in the driving force of Na+ across cytoplasmic membranes. Thus, we examined effects of decrease in the extracellular Na+ concentration on spontaneous events observed in NGF-treated neurons (Fig. 3A). When extracellular Na+ concentrations were gradually decreased, in the neuron shown in Fig. 3B, spontaneous spikes suddenly disappeared in the presence of 100 mM Na+ in accordance with the all-or-none fashion. Totally, in 2 neurons out of 12 neurons tested, spontaneous spikes were abolished at 120 mM Na+ and in remaining 10 neurons the reduction of Na+ concentration to 100 mM abolished them. In all neurons tested, reintroduction of 154 mM Na+ recovered the generation of spontaneous spikes. Similar results were obtained in the experiment, in which Na+ concentrations were reduced by an isotonic replacement of Na+ by Tris+ or choline+ instead of NMDG+ indicating that the abolition of spontaneous events was not caused by an effect of NMDG+, but

127

caused by the decrease in the extracellular Na+ concentrations. Assuming that the physiological intracellular concentration of Na + was 5 mM, the calculated reversal potential of Na+ in the presence of 154, 120 and, 100 mM extracellular Na+ were +88, +81 and, +77 mV, respectively. Since the resting membrane potentials in C-type DRG neuron of adult rats have been reported to be approximately −60 mV (4, 10), when Na+ concentration were decreased from 154 mM to 100 mM, the inward driving force of Na+ decreased by only 7.4%. Such small decrease in the driving force of Na+ was effective to block the spontaneous spikes suggesting that these spikes were resulted from spontaneously generated action potentials. Moreover, effects of increasing concentrations of extracellular Na+ on the spontaneous spikes were examined (Fig. 3B). Since, the frequency of spontaneous spikes was varied from 0 (no events) to more than 5 Hz, quantitative analysis of the frequency was difficult. However, elevation of Na+ tended to increase the frequency of the spontaneous spikes without any effect on their amplitudes in all 10 neurons tested. The depolarizing phase of the action potentials was resulted from increase in Na+ conductance caused by the activation of voltage-gated Na+ channels. Since voltagegated Na + channels were inactivated under the depolarizing condition, it was expected that the spontaneous current spikes were inhibited at the depolarized resting membrane potential if they had been resulted from the action potentials. Thus, we examined effects of an elevated concentration of K+, which induced the positive shift of the resting membrane potentials, on spontaneous spikes. As shown in Fig. 3C, when the extracellular K+ concentration was raised from 6 mM to 40 mM, the frequency of spontaneous spikes was gradually increased and their amplitudes were gradually decreased, and finally these spikes disappeared, reversibly. These results were reproducible in all 6 neurons tested, and suggested that the transient depolarizing spikes observed repetitively were resulted from the spontaneous action potentials. Since the activation of voltage-gated Na+ channels has a key role in the generation of the action potentials, the effects of voltage-gated Na+ channel blocking agents — tetrodotoxin (TTX) and a local anesthetic, lidocaine — on the spontaneous depolarizing spikes in NGFtreated DRG neurons were examined. Increasing concentrations of TTX from 0.1 nM to 100 nM and lidocaine from 1 μM to 1 mM were applied extracellularly. Each concentration of Na+ channel blocking agents was applied for 30 s. Typical traces were

128

N. Kitamura et al.

Fig. 3 Effects of changes in Na+ and K+ concentrations on spontaneous current spikes in NGF-treated DRG neurons. A. Na+ concentration was decreased from 154 mM to 100 mM as indicated in the upper panel. B. Na+ concentration was elevated from 154 mM to 200 mM. Solutions at each Na+ concentration were superfused for 1 min (A, B). C. K+ concentration was raised from 6 mM to 40 mM for 1 min.

shown in Fig. 4. In all neurons tested, TTX at 100 nM reversibly abolished spontaneous discharges in accordance with the all-or-none fashion without effects on the amplitudes at lower concentrations (Fig. 4A, n = 5). Similar to the effects of TTX, the spontaneous spikes were blocked by lidocaine at 100 μM in 1 out of 5 tested neuron and at 1 mM in remaining 4 neurons in accordance with the all-ornone fashion (Fig. 4B). This blocking effect of lidocaine was also reversible; the spontaneous current spikes were recovered after washout of lidocaine. These all-or-none blocking actions of TTX and lidocaine also supported the hypothesis that spontaneously generated action potentials resulted in the spontaneous depolarizing current spikes recorded under the cell-attached configuration. Chronic NGF treatment induces hyperexcitability in DRG neurons Fig. 5 shows the time-course of appearance of spontaneously firing neurons under the cell-attached configuration. The proportion of the number of spontaneously firing neurons to the total number of recorded neurons was plotted against the culturing period in the presence of 100 ng/ml NGF. Since the frequency of spontaneous firing was variable among

Fig. 4 Effect of tetrodotoxin (TTX) and lidocaine (Lid) on spontaneous current spikes in NGF-treated DRG neurons. A. Increasing concentrations of TTX from 10 to 100 nM was applied to the bath solution. B. Increasing concentrations of Lid from 10 to 100 µM was applied to the bath solution. Each concentration of TTX and Lid were applied for 30 s.

neurons (about 0.1–5 Hz), neurons that showed no firing within 30 s after the start of recording were considered to be the quiescent neurons. Almost of all neurons cultured in the absence of NGF showed


Fig. 5 Appearance time course of hyperexcitable DRG neurons during culture with NGF. The proportion of spontaneously firing neurons is shown by columns against the period after the beginning of the culture with NGF (100 ng/ml). Numbers on each column indicate the numbers of hyperexcitable neurons/the numbers of all neurons tested on each day.

no firing activity. The spontaneous firing activity began to appear a day after the start of NGF treatment, and more than 20% neurons cultured with NGF for more than 3 days showed spontaneous firing. These results clearly indicated that chronic NGF treatment caused abnormal hyperexcitability in DRG neurons of adult rats. It has been reported that NGF concentrations in DRG of neuropathic model rat increased from 100– 300 pg/ganglion to more than 3 ng/ganglion by the operation of nerve injury (9). And an injection of NGF at 100 ng/ml was reported to cause neuropathic pain-related behavior (15, 16). These reports suggested that NGF was a candidate of the pathogenic mediators of neuropathic pain. Since the diameter of adult rat DRG was 1–3 mm, the concentration of NGF (100 ng/ml) used in this study was pathophysiologically possible concentration. The chronic treatment of adult rat DRG neurons with NGF may be a cellular model of neuropathic pain pathogenesis. It has been reported that NGF had an acute effect on DRG neurons to increase sensitivity of VR1 receptors for capsaicin (23, 24, 25). NGF was also reported to potentiate excitability by inhibiting K+ channels via the activation of p75NTR (33, 34). These acute effects of NGF on sensory neurons may explain the spontaneous firing of DRG neurons in vivo

129

and the hyperalgesia. However, this is not the case in the hyperexcitable neurons dissociated from neuropathic model animals and neurons in this study, because electrophysiological recordings were performed in the absence of NGF. DRG neurons have been also reported to express various excitatory ionotropic receptors, such as P2X receptors (3). Since we usually performed the patch clamp recordings in the neurons superfused continuously, it is hard to expect the condition, in which excitatory mediators, e.g. ATP, spontaneously released from surrounding cells stimulate the focused neurons to fire. In addition, during the cell-attached recording, when the superfusion was stopped in order to stimulate the accumulation of mediators released from surrounding cells in the bath solutions, the frequency and amplitudes of the spontaneous discharges were not changed (n = 5, data not shown). Thus, chronic treatment with NGF may provide an intrinsic mechanism causing the abnormal hyperexcitablity of DRG neurons. Electrophysiological results in this study were obtained from sensory neurons cultured in the artificial condition, and the difference in situations between cultured neurons and DRG neurons in living animals is unclear. However, at least, it became evident that DRG neurons isolated from adult rats have a nature to respond to NGF and obtained the abnormal hyperexcitability to fire spontaneously. Acknowledgements This study was supported by Grants-in-Aid for Scientific Research from Japanese Society for the Promotion of Science and grants from the Akiyama Foundation. REFERENCES 1. Amir R, Michaelis M and Devor M (1999) Membrane potential oscillations in dorsal root ganglion neurons: role in normal electrogenesis and neuropathic pain. J Neurosci 19, 8589–8596. 2. Bennett GJ and Xie YK (1988) A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 33, 87–107. 3. Bradbury EJ, Burnstock G and McMahon SB (1998) The expression of P2X3 purinoreceptors in sensory neurons: effects of axotomy and glial-derived neurotrophic factor. Mol Cell Neurosci 12, 256–268. 4. Caffrey JM, Eng DL, Black JA, Waxman SG and Kocsis JD (1992) Three types of sodium channels in adult rat dorsal root ganglion neurons. Brain Res 592, 283–297. 5. Decosterd I and Woolf CJ (2000) Spared nerve injury: an animal model of persistent peripheral neuropathic pain. Pain 87, 149–158.

130 6. Devor M, Janig W and Michaelis M (1994) Modulation of activity in dorsal root ganglion neurons by sympathetic activation in nerve-injured rats. J Neurophysiol 71, 38–47. 7. Harper AA and Lawson SN (1985) Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones. J Physiol 359, 31–46. 8. Harper AA and Lawson SN (1985) Electrical properties of rat dorsal root ganglion neurones with different peripheral nerve conduction velocities. J Physiol 359, 47–63. 9. Herzberg U, Eliav E, Dorsey JM, Gracely RH and Kopin IJ (1997) NGF involvement in pain induced by chronic constriction injury of the rat sciatic nerve. Neuroreport 8, 1613– 1618. 10. Jeftinija S (1994) The role of tetrodotoxin-resistant sodium channels of small primary afferent fibers. Brain Res 639, 125–134. 11. Kajander KC and Bennett GJ (1992) Onset of a painful peripheral neuropathy in rat: a partial and differential deafferentation and spontaneous discharge in A beta and A delta primary afferent neurons. J Neurophysiol 68, 734–744. 12. Kajander KC, Wakisaka S and Bennett GJ (1992) Spontaneous discharge originates in the dorsal root ganglion at the onset of a painful peripheral neuropathy in the rat. Neurosci Lett 138, 225–228. 13. Kim SH and Chung JM (1992) An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 50, 355–363. 14. Komagiri Y and Kitamura N (2003) Effect of intracellular dialysis of ATP on the hyperpolarization-activated cation current in rat dorsal root ganglion neurons. J Neurophysiol 90, 2115–2122. 15. Lewin GR and Mendell LM (1993) Nerve growth factor and nociception. Trends Neurosci 16, 353–359. 16. Lewin GR, Ritter AM and Mendell LM (1993) Nerve growth factor-induced hyperalgesia in the neonatal and adult rat. J Neurosci 13, 2136–2148. 17. Liu CN, Michaelis M, Amir R and Devor M (2000) Spinal nerve injury enhances subthreshold membrane potential oscillations in DRG neurons: relation to neuropathic pain. J Neurophysiol 84, 205–215. 18. Matzner O and Devor M (1994) Hyperexcitability at sites of nerve injury depends on voltage-sensitive Na+ channels. J Neurophysiol 72, 349–359. 19. Mu X, Silos-Santiago I, Carroll SL and Snider WD (1993) Neurotrophin receptor genes are expressed in distinct patterns in developing dorsal root ganglia. J Neurosci 13, 4029–4041. 20. Petersen M, Zhang J, Zhang JM and LaMotte RH (1996) Abnormal spontaneous activity and responses to norepinephrine in dissociated dorsal root ganglion cells after chronic nerve constriction. Pain 67, 391–397.

N. Kitamura et al. 21. Sebert ME and Shooter EM (1993) Expression of mRNA for neurotrophic factors and their receptors in the rat dorsal root ganglion and sciatic nerve following nerve injury. J Neurosci Res 36, 357–367. 22. Seltzer Z, Dubner R and Shir Y (1990) A novel behavioral model of neuropathic pain disorders produced in rats by partial sciatic nerve injury. Pain 43, 205–218. 23. Shu X and Mendell LM (2001) Acute sensitization by NGF of the response of small-diameter sensory neurons to capsaicin. J Neurophysiol 86, 2931–2938. 24. Shu X and Mendell LM (1999) Nerve growth factor acutely sensitizes the response of adult rat sensory neurons to capsaicin. Neurosci Lett 274, 159–162. 25. Shu XQ and Mendell LM (1999) Neurotrophins and hyperalgesia. Proc Natl Acad Sci USA 96, 7693–7696. 26. Song XJ, Zhang JM, Hu SJ and LaMotte RH (2003) Somata of nerve-injured sensory neurons exhibit enhanced responses to inflammatory mediators. Pain 104, 701–709. 27. Study RE and Kral MG (1996) Spontaneous action potential activity in isolated dorsal root ganglion neurons from rats with a painful neuropathy. Pain 65, 235–242. 28. Verge VM, Richardson PM, Benoit R and Riopelle RJ (1989) Histochemical characterization of sensory neurons with highaffinity receptors for nerve growth factor. J Neurocytol 18, 583–591. 29. Verge VM, Richardson PM, Wiesenfeld-Hallin Z and Hökfelt T (1995) Differential influence of nerve growth factor on neuropeptide expression in vivo: a novel role in peptide suppression in adult sensory neurons. J Neurosci 15, 2081–2096. 30. Wall PD and Devor M (1983) Sensory afferent impulses originate from dorsal root ganglia as well as from the periphery in normal and nerve injured rats. Pain 17, 321–339. 31. Wall PD, Devor M, Inbal R, Scadding JW, Schonfeld D, Seltzer Z and Tomkiewicz MM (1979) Autotomy following peripheral nerve lesions: experimental anaesthesia dolorosa. Pain 7, 103–111. 32. Winkelstein BA, Rutkowski MD, Weinstein JN and DeLeo JA (2001) Quantification of neural tissue injury in a rat radiculopathy model: comparison of local deformation, behavioral outcomes, and spinal cytokine mRNA for two surgeons. J Neurosci Methods 111, 49–57. 33. Zhang YH and Nicol GD (2004) NGF-mediated sensitization of the excitability of rat sensory neurons is prevented by a blocking antibody to the p75 neurotrophin receptor. Neurosci Lett 366, 187–192. 34. Zhang YH, Vasko MR and Nicol GD (2002) Ceramide, a putative second messenger for nerve growth factor, modulates the TTX-resistant Na+ current and delayed rectifier K+ current in rat sensory neurons. J Physiol 544, 385–402.