Receptor Subtype Mediating the Adrenergic Sensitivity of Pain Behavior and Ectopic Discharges in Neuropathic Lewis Rats

Receptor Subtype Mediating the Adrenergic Sensitivity of Pain Behavior and Ectopic Discharges in Neuropathic Lewis Rats DOO HYUN LEE, 1 XIANZENG LIU, 1 HYUN TAEK KIM, 1 KYUNGSOON CHUNG, 1,2 AND JIN MO CHUNG 1 3 1 Marine Biomedical Institute; 2 Department of Anatomy and Neurosciences; 3 Department of Physiology and Biophysics, University of Texas Medical Branch, Galveston, Texas 77555-1069 Doo Hyun Lee, Xianzeng Liu, Hyun Taek Kim, Kyungsoon Chung, and Jin Mo Chung. Receptor subtype mediating the adrenergic sensitivity of pain behavior and ectopic discharges in neuropathic Lewis rats. J. Neurophysiol. 81: 2226 2233, 1999. We attempted to identify the subtype of -adrenergic receptor ( -AR) that is responsible for the sympathetic (adrenergic) dependency of neuropathic pain in the segmental spinal injury (SSI) model in the Lewis strain of rat. This model was chosen because our previous study showed that pain behaviors in this condition are particularly sensitive to systemic injection of phentolamine (PTL), a general -AR blocker. We examined the effects of specific 1 - and 2 -AR blockers on 1) behavioral signs of mechanical allodynia, 2) ectopic discharges recorded in the in vivo condition, and 3) ectopic discharges recorded in an in vitro setup. One week after tight ligation of the L5 and L6 spinal nerves, mechanical thresholds of the paw for foot withdrawals were drastically lowered; we interpreted this change as a sign of mechanical allodynia. Signs of mechanical allodynia were significantly relieved by a systemic injection of PTL (a mixed 1 - and 2 -AR antagonist) or terazosin (TRZ, an 1 -AR antagonist) but not by various 2 -AR antagonists (idazoxan, rauwolscine, or yohimbine), suggesting that the 1 -AR is in part the mediator of the signs of mechanical allodynia. Ongoing ectopic discharges were recorded from injured afferents in fascicles of the L5 dorsal root of the neuropathic rat with an in vivo recording setup. Ongoing discharge rate was significantly reduced after intraperitoneal injection of PTL or TRZ but not by idazoxan. In addition, by using an in vitro recording setup, spontaneous activity was recorded from teased dorsal root fibers in a segment in which the spinal nerve was previously ligated. Application of epinephrine to the perfusion bath enhanced ongoing discharges. This evoked activity was blocked by pretreatment with TRZ but not with idazoxan. This study demonstrated that both behavioral signs of mechanical allodynia and ectopic discharges of injured afferents in the Lewis neuropathic rat are in part mediated by mechanisms involving 1 -ARs. These results suggest that the sympathetic dependency of neuropathic pain in the Lewis strain of the rat is mediated by the 1 subtype of AR. because phentolamine (PTL), a mixed 1 - and 2 -AR antagonist, was used successfully as a diagnostic tool for identification of SMP patients (Arnér 1991; Raja et al. 1991). However, which subtype of AR is involved, 1 or 2, is highly controversial. It is obviously important to identify the subtype to improve the means of treatment of SMP patients. Sympathetic dependency of pain behaviors has also been shown in animal models of neuropathic pain. Surgical or chemical sympathectomy has been shown to be effective in relieving pain behaviors in various rat models (Kim et al. 1997; Lee et al. 1997; Neil et al. 1991; Shir and Seltzer 1991). However, -AR blockers, such as PTL, have not consistently proven to be effective in reducing neuropathic pain behaviors. The lack of a consistent effect of -AR blockers makes it difficult to study the subtype of -AR involved in neuropathic pain behaviors. In our recent study (Lee et al. 1997), we found a striking difference in adrenergic sensitivity of neuropathic pain behaviors among different strains of rats. Lewis rats in particular showed a powerful and consistent antiallodynic response to systemically injected PTL. Because of this robust effect of a mixed 1 - and 2 -AR antagonist on the Lewis neuropathic rat, we examined the subtypes of -AR that mediate both the adrenergic dependency of pain behaviors and the ectopic discharges of injured sensory neurons in this strain. Preliminary data were presented in abstract form (Lee and Chung 1997; Lee et al. 1998). METHODS The experimental protocols were approved by the Animal Care and Use Committee of the University of Texas Medical Branch and were performed in accordance with the NIH guidelines. INTRODUCTION It was well documented that neuropathic pain resulting from peripheral nerve injury can be relieved in some patients by blocking sympathetic outflow to the affected region (Bonica 1990; Loh and Nathan 1978; Loh et al. 1980). This type of neuropathic pain is referred to as sympathetically maintained pain (SMP) and is contrasted to sympathetically independent pain, which is not influenced by sympathetic manipulation (Roberts 1986). At least a part of this sympathetic dependency appears to be mediated by -adrenergic receptors ( -AR) The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Animals and surgery Ninety-nine Lewis strain male rats (Harlan Sprague-Dawley, Indianapolis, IN), weighing 150 250 g, were used in this study. The animals were housed in groups of three, in plastic cages, with soft bedding, under a 12/12-h reversed light dark cycle (light cycle: 9:00 P.M. to 9:00 A.M.; dark cycle: 9:00 A.M. to 9:00 P.M.). They were kept in the same room at a constant ambient temperature and had free access to food and water. The rats were kept 7 days under these conditions before experimental manipulations. The study was performed during the dark cycle, which is the rat s active period. Neuropathic surgery was done as previously described in detail (Choi et al. 1994; Kim and Chung 1992). Briefly, under gaseous anesthesia with a mixture of halothane (2% for induction and 0.8% for maintenance) and a 2:1 flow of O 2 and N 2 O, the left L5 and L6 spinal nerves were ligated tightly with 6 0 silk threads. Wounds were 2226 0022-3077/99 $5.00 Copyright 1999 The American Physiological Society

ADRENERGIC RECEPTOR SUBTYPE MEDIATING NEUROPATHIC PAIN 2227 closed and anesthesia was discontinued. Animals were kept on a warm plate until they recovered from anesthesia completely and resumed normal activities. Behavioral tests MECHANICAL ALLODYNIA. Mechanical sensitivity of the paw was measured by determining the median 50% foot withdrawal threshold with von Frey filaments with the up down method (Chaplan et al. 1994). The rats were placed under a plastic cover (8 8 18 cm) on a metal mesh floor, and von Frey filaments were applied from underneath to the plantar surface of the foot. The area tested was the proximal one-half of the third or fourth toe. The toe was stimulated with a series of eight von Frey filaments with logarithmically incremental bending forces (4.4, 7.4, 12.3, 21.6, 32.3, 53.9, 82.3, and 142.1 mn). The von Frey filament was presented perpendicular to the toe surface with sufficient force to bend it slightly and was held for 2 3 s. An abrupt withdrawal of the foot (hind limb flinching) during stimulation or immediately after the removal of stimulus was considered to be a positive response. On the basis of electrophysiological recordings, we found that thresholds for mechanoreceptors in the rat foot did not exceed 25 mn, whereas those of nociceptors were rarely lower than that level (Leem et al. 1993). However, the threshold for hind limb flinching after neuropathic surgery often became 10 mn, suggesting that a nociceptive reflex was elicited by the activation of mechanoreceptors in this situation. Therefore a significant reduction in the mechanical threshold for hind limb flinching was interpreted as a sign of mechanical allodynia. -AR ANTAGONISTS. The effects of systemic (intraperitoneal) injection of -AR antagonists on mechanical hypersensitivity were tested at the 1-wk postoperative (PO) time point. Tested -AR antagonists were PTL (a mixed 1 - and 2 -AR antagonist, from RBI), terazosin (TRZ, an 1 -AR antagonist), idazoxan HCl (IDZ, an 2 -AR antagonist, from Sigma), rauwolscine HCl (an 2 -AR antagonist, from RBI), and yohimbine HCl (an 2 -AR antagonist, from Sigma). Electrophysiological studies IN VIVO STUDY. Single-unit recordings were made from filaments of the left L5 dorsal root in neuropathic rats at a time between 7 and14 PO days. The rats were anesthetized with a mixture of halothane and a 2:1 ratio of O 2 and N 2 O. The left jugular vein was cannulated with a polyethylene tube (PE-20) for systemic drug administration. The right carotid artery was cannulated to monitor blood pressure throughout the experiment. When the diastolic blood pressure dropped to 60 mmhg for 30 min, the experiment was discontinued. Under artificial ventilation, animals were paralyzed with pancuronium bromide (Parvlon: a single bolus of 1 mg/kg iv followed by a continuous intravenous infusion, 0.4 mg kg 1 h 1 ). The ventilator was adjusted to an end-tidal CO 2 level between 4 and 5% throughout the experiment. The spinal cord was exposed by a laminectomy of the L1 L6 vertebrae. The animal was mounted on a spinal investigation frame, and a heated mineral oil pool (36 C) was made over the exposed tissue to prevent it from drying. The L5 dorsal root was cut near the spinal cord, and the distal stump was placed on a mirror plate. Fine filaments were dissected until a single spontaneous unit could be isolated on the basis of its amplitude and waveform. The unit activity was amplified with an AC-coupled amplifier (WPI, DAM-5A) and led to a window discriminator (Mentor, N-750). The output of the window discriminator was used to compile peristimulus time histograms by a data acquisition system (CED-1401, Spike 2). IN VITRO STUDY. For the in vitro study, the left L4 and L5 spinal nerves were ligated. Seven to 14 days later, animals were anesthetized with halothane, and the L4 and L5 dorsal root ganglia (DRG), along with dorsal roots and spinal nerves, were removed. The DRG were placed in an in vitro recording chamber that consisted of two separate compartments, one for the dorsal root and the other for the DRG and spinal nerve. The DRG and spinal nerve compartment was perfused with oxygenated (95% O 2-5% CO 2 ) artificial cerebrospinal fluid [composition (in mm): 130 NaCl, 3.5 KCl, 1.25 NaH 2 PO 4, 24 NaHCO 3, 10 dextrose, 1.2 MgCl 2, 1.2 CaCl 2, ph 7.3] at a rate of 4 5 ml/min. The dorsal root compartment was filled with mineral oil. The temperature was kept at 35 1 C by means of a temperaturecontrolled water bath. Ectopic discharges were recorded from the teased dorsal root fascicles, and the spinal nerve was stimulated with a suction electrode. Fiber types were classified according to their conduction velocity: 14 m/s for A, 2 14 m/s for A, and 2 m/s for C fibers (Harper and Lawson 1985; Ritter and Mendell 1992; Waddell and Lawson 1990). For analysis of the effects of -AR antagonists and agonists, the number of spikes per minute was calculated, and the numbers were compared before and after a treatment. Statistical treatments Data are displayed as box plots, and differences between groups were tested with the Kruskal-Wallis one-way analysis of variance followed by the Dunnett s post hoc multiple comparisons test. Twotailed P values 0.05 were considered to be significant. RESULTS Effects of -AR antagonists on mechanical allodynia Nine rats were used to examine the time course of neuropathic pain behaviors in Lewis strain rats after segmental spinal nerve injury (SSI). All rats showed mechanical allodynia (decreased hind limb flinching threshold) that reached the peak level 1 day after nerve injury. This high level of mechanical sensitivity was maintained for the next 8 wk. Although the mechanical threshold gradually increased beyond 8 wk, a significant level of hypersensitivity was maintained for the entire observation period of 20 wk. The effects of -AR antagonists on mechanical hypersensitivity were examined on 36 (9 in each group) neuropathic rats, and the results are shown in Fig. 1. Neuropathic surgery produced mechanical hypersensitivity at 1 wk PO, and so the hind limb flinching threshold was reduced to a very low level (BASE in Fig. 1). Intraperitoneal injection of PTL (5 mg/kg, a mixed 1 - and 2 -AR antagonist) or TRZ (5 mg/kg, an 1 -AR antagonist) produced a significant elevation of the threshold for 1 4 h. On the other hand, neither saline nor IDZ (5 mg/kg, an 2 -AR antagonist) had any effect on mechanical hypersensitivity. These data suggest that the mechanical allodynic behavior of neuropathic Lewis rats is in part maintained by an 1 -AR mediated mechanism. Because IDZ had no effect, we tested two other 2 -AR antagonists to make certain of the ineffectiveness of 2 -AR antagonists. Nine neuropathic rats (1 wk PO) were prepared, and mechanical sensitivity was measured before and after intraperitoneal injection of 2 -AR antagonists. On a given day, one of the following three substances, rauwolscine (5 mg/kg), yohimbine (5 mg/kg), or saline (0.2 ml), was administered to each of three groups of randomly selected rats. The procedure was repeated a total of three times on 3 consecutive days according to the Latin Square design so that each rat received all three substances in random order over 3 days. None of these compounds had any effect on the mechanical thresholds for foot withdrawal. Therefore it seems clear that 2 -ARs are not

2228 D. H. LEE, X. LIU, H. T. KIM, K. CHUNG, AND J. M. CHUNG FIG. 1. Box plots showing the effects of -adrenergic receptor ( -AR) blockers on mechanical sensitivity of the foot. Mechanical sensitivity was expressed as foot withdrawal thresholds determined by the up down method with graded strengths of von Frey filaments. Four groups (n 9 each) of rats were tested before neuropathic surgery (PRE), 1 week after the surgery (BASE), and at 3 time points (1, 4, and 24 h) after an intraperitoneal injection of 1 of the following 4 agents: saline (SAL, vehicle control), idazoxan (IDZ, 5 mg/kg, an 2 -AR antagonist), phentolamine (PTL, 5 mg/kg, a mixed 1 - and 2 -AR antagonist), and terazosin (TRZ, 5 mg/kg, an 1 -AR antagonist). Box plot code: horizontal bar within each box, the median value; the top and bottom borders of each box, 75 and 25th percentile values; and error bars of each box, 90 and 10th percentile values. *: values significantly different from those in the saline control group (P 0.05 by the Kruskal-Wallis ANOVA on ranks followed by the Dunnett s multiple comparisons test). involved in the mechanical allodynia that develops in neuropathic Lewis rats. Effects of -AR antagonists on ectopic discharges (in vivo study) Because ectopic discharges of injured afferents are an important underlying mechanism of neuropathic pain behaviors, we examined the effects of -AR antagonists on the ectopic discharges of 44 units (10 units for saline, 13 for PTL, 12 for TRZ, and 9 for IDZ) recorded from the L5 dorsal roots of 25 Lewis rats between 7 and 14 days after tight ligation of the L5 and L6 spinal nerves. Many afferent fibers in the L5 dorsal root of neuropathic rats showed ongoing activity without any apparent stimulation. Because the L5 spinal nerve was tightly ligated at the time of neuropathic surgery, all afferent fibers in the L5 dorsal root were disconnected from their original sensory receptors, where normal impulse generation occurs. The recorded ongoing activity therefore must originate from sites other than the original receptors; thus they are regarded as ectopic discharges. Initially, the firing rate of each ectopic discharge was recorded for 10 min, and this rate was considered the baseline ectopic discharge rate. Then a bolus of PTL (2.5 mg/kg), TRZ (2.5 mg/kg), IDZ (2.5 mg/kg), or saline (0.2 ml) was injected intraperitoneally. The recording was maintained for 1 h after the administration of each drug and extended to2hinsome cases. Examples of single-unit recordings of ectopic discharges and the effects of -AR antagonists are shown in Fig. 2. After a 10-min recording of the baseline ongoing activity, either saline or a specific -AR antagonist was injected. Both PTL and TRZ reduced the rate of ectopic discharges with a delay of 10 20 min after the injection. The rate then recovered in 1.5 2 h. Neither saline nor IDZ had any effect. Testing with intravenous injection (1 mg/kg) of PTL in two other units (recorded from two animals) produced a similar reduction in the ectopic discharges. The effects of various -AR antagonists examined in all 44 units were analyzed. As shown in Fig. 3, PTL and TRZ significantly reduced the rate of ectopic discharges, whereas neither saline nor IDZ produced any effect. Because -AR antagonists have strong cardiovascular effects, it was necessary to monitor blood pressure very closely during ectopic discharge recordings. Figure 4 shows examples of blood pressure responses to injections of various -AR antagonists. All three -AR antagonists (IDZ, PTL, and TRZ) produced a transient decrease in blood pressure lasting for 20 min. FIG. 2. Effects of -AR blockers on ectopic discharges (in vivo preparation) in the neuropathic Lewis rat. Single unit activity was recorded from teased L5 dorsal root filaments between 7 and 14 days after tight ligation of the L5 and L6 spinal nerves. Records show examples of peristimulus time histograms compiled from 4 separate units. Various -AR antagonists (or saline) were injected intraperitoneally at time 0, and the activities were followed for either 60 or 120 min. PTL and TRZ reduced the rate of ectopic discharges, but IDZ had no effect. Top, inset: segment of action potential recording made before saline injection.

ADRENERGIC RECEPTOR SUBTYPE MEDIATING NEUROPATHIC PAIN 2229 FIG. 3. Box plots showing the effects of -AR blockers on ectopic discharges recorded in an in vivo preparation. Single-unit activity was recorded from teased L5 dorsal root filaments between 7 and 14 days after tight ligation of the L5 and L6 spinal nerves. Before any drug injection, the number of ectopic discharges during a 10-min period is collected as baseline activity, and subsequent data in 10-min blocks are expressed as the percent of the change from the baseline. Box plot codes are the same as in Fig. 1. *, values significantly different from those in the saline control group (P 0.05 by the Kruskal-Wallis ANOVA on ranks followed by the Dunnett s multiple comparisons test). Effects of -AR antagonists and agonists on ectopic discharges (in vitro study) To confirm the effects of -AR antagonists in simplified conditions, we tested them on ectopic discharges recorded in an in vitro preparation. Single-unit recordings were made from teased L4 or L5 dorsal root filaments between 7 and 14 days after tight ligation of the L4 and L5 spinal nerves. Table 1 summarizes the general characteristics of the 48 recorded units (from 20 rats) and responsiveness to exogenously applied epinephrine bitartrate (EP, from RBI). The conduction velocity was measured for 45 of 48 units. Thirty-five of 45 units (77.8%) were A fibers (CV: 14 69.4 m/s), and 10 units (22.2%) were A fibers (CV: 7.8 13 m/s). We did not find any C fibers in this study. Application of EP to the perfusion bath evoked an enhancement of the discharges by 30% over the baseline in 29 units (60.4%), whereas 16 units (33.3%) did not show any response. A small number of units (4 units, 6.3%) showed a decreased (reduction of 30% of the baseline value) ectopic discharge rate in response to EP. Because the ectopic discharges of many units responded to exogenously applied EP, we investigated the subtype of -AR mediating the responses. We tested the effects of other -AR agonists on 7 units among the 29 units that showed an excitatory evoked response to EP. Among the seven units tested, three units received L-phenylephrine HCl (PEP; an 1 -AR agonist, 10 M, from Sigma) first, and after washing out UK 14,304 (UK1; an 2 -AR agonist, 10 M, from RBI) was applied. The order of application of PEP and UK1 was reversed in the remaining four units. Figure 5A shows an example of responses to -AR agonists, and the results of all the units are summarized in Fig. 5B. Infusion of EP produced an enhancement of ectopic discharges of 104.3% (median value) over the baseline rate. Application of PEP produced a similar enhancement of discharges (median value of 45.5% over the baseline). On the other hand, UK1 failed to induce an enhancement of ectopic discharges (median value of 38.2% reduction FIG. 4. Effects of -AR blockers on systemic blood pressure in the neuropathic Lewis rat. Arterial blood pressure was monitored during ectopic discharge recording sessions as shown in Fig. 2. Various -AR antagonists (or saline) were injected intraperitoneally at time 0, and the recordings were followed for 60 min after the injection. Both 1 - (TRZ) and 2 -AR antagonists (IDZ) produced a similar pattern of blood pressure changes lasting 20 min. from the baseline). The results indicate that an 1 -AR agonist but not an 2 -AR agonist can mimic the action of EP on ectopic discharges and suggest that the ectopic discharges evoked by EP in axotomized sensory neurons are mediated by 1 -AR. In fact, an 2 -AR agonist tends to inhibit ectopic discharges. We further investigated the subtype of -AR mediating the EP evoked ectopic discharges by examining the effects of pretreatment with -AR antagonists. Among the 29 units that TABLE 1. Matrix chart summarizing the responses of 48 individually recorded units to exogenous application of EP Fiber type Frequency, Hz Response to EP Application Increase No change Decrease Total A 4.29 2.03 21 12 3 35 A 4.97 3.36 7 2 1 10 Unknown 3.72 3.33 1 2 0 3 Total 4.39 2.39 29 16 4 48 Values are means SD. Increase and decrease are defined by changes in discharge rate of 30% from the baseline. EP, epinephrine.

2230 D. H. LEE, X. LIU, H. T. KIM, K. CHUNG, AND J. M. CHUNG showed an excitatory evoked response to EP, we tested the effects of -AR antagonists on 11 units. We pretreated two units with IDZ, an 2 -AR antagonist, and two units with TRZ, an 1 -AR antagonist, before application of EP. In both cases, pretreatment with IDZ did not influence the EP-evoked ectopic discharges, whereas pretreatment with TRZ produced a longlasting blockade of the EP-induced enhancement of ectopic discharges. We tested the remaining seven units for the effect of pretreatment with IDZ first and then, after washing out the drug, the effect of TRZ. Figure 6A shows an example of this experiment, and the results of all units are summarized in Fig. 6B. Infusion of EP alone on these units produced an enhancement of ectopic discharges of 95.8% (median value) over the baseline rate. After pretreatment with IDZ, the application of EP still produced a similar enhancement of discharges (median value of 53.8% increase over the baseline). On the other hand, pretreatment with TRZ completely blocked EP-evoked enhancement of ectopic discharges (median value of 44.7% reduction from the baseline). Pretreatment with TRZ produced long-lasting effects. On six units, reapplication of EP 30 60 min after pretreatment with TRZ evoked an inhibition (median value of 35.9% reduction from the baseline) rather than excitation. Thus application of a mixed 1 - and 2 -AR ligand (EP) in the absence of 1 -AR (because of blockade by TRZ) produced an inhibition, presumably through activation of 2 -ARs. FIG. 5. Effects of -AR agonists on ectopic discharges recorded in an in vitro preparation. The L4 and L5 spinal nerves were tightly ligated in Lewis rats. Seven to 14 days later, single-unit activity was recorded from teased L4 or L5 dorsal root filaments with an in vitro recording setup, and -AR agonists were applied to the perfusion bath. An example of responses to -AR agonists is shown in A. Application of epinephrine (EP) evoked an enhancement of ectopic discharges, whereas UK 14,304 (UK1) produced a mild reduction. Application of phenylephrine (PEP) mimicked the effect of EP. Top, inset: segment of action potential recording made before EP injection. In B, responses to -AR agonists on all tested units are shown as box plots. Activity was counted every minute, and data are expressed as percent changes from the baseline activity (before application of any agent). Box plot codes are the same as in Fig. 1. *, value significantly different from that in the EP group (P 0.05 by the Kruskal-Wallis ANOVA on ranks followed by the Dunnett s multiple comparisons test). FIG. 6. Effects of -AR antagonists on ectopic discharges recorded in an in vitro preparation. The L4 and L5 spinal nerves were tightly ligated in Lewis rats. Seven to 14 days later, single-unit activity was recorded from teased L4 or L5 dorsal root filaments with an in vitro recording setup and -AR antagonists (along with EP) were applied to the perfusion bath. Examples of responses to -AR antagonists are shown in A. Application of EP evoked an enhancement of ectopic discharges, and this enhancement was blocked by application of TRZ but not by IDZ. In B, responses to -AR antagonists on all tested units are shown as box plots. Activity was counted every minute, and data are expressed as percent changes from the baseline activity (before application of any agent). Box plot codes are the same as in Fig. 1. *, value significantly different from that in the EP group (P 0.05 by the Kruskal- Wallis ANOVA on ranks followed by the Dunnett s multiple comparisons test). The results indicate that the 1 -AR antagonist but not the 2 -AR antagonist blocks the action of EP on ectopic discharges and again suggest that the ectopic discharges evoked by EP in axotomized sensory neurons are mediated by 1 -AR. Furthermore, activation of 2 -AR seems to produce an inhibition of ectopic discharges. DISCUSSION The role of the sympathetic nervous system in neuropathic pain is a complex and controversial issue. In particular, although it is generally accepted that the -AR is involved in SMP patients, it is not clear which subtype is playing the important role. One line of evidence, mainly obtained from human patients, supports the importance of 1 -AR (Davis et al. 1991). In fact, Stevens et al. (1993) reported that TRZ, an 1 -AR antagonist, effectively relieved SMP and vasospasm in a human patient. On the other hand, a number of animal studies suggests that 2 -AR is more important (Chen et al. 1996; Leem et al. 1997; Sato and Perl 1991; Xie et al. 1995). In our previous study, we reported that the mechanical hypersensitivity that develops in the SSI model in the Lewis strain of rats was greatly reduced by intraperitoneal injection of PTL (a

ADRENERGIC RECEPTOR SUBTYPE MEDIATING NEUROPATHIC PAIN 2231 mixed 1 - and 2 -AR antagonist) (Lee et al. 1997). This study expands the previous work by showing that the subtype of -AR mediating the reduction of mechanical hypersensitivity is 1 -AR, not 2 -AR. Furthermore, both in vivo and in vitro physiological experiments showed that adrenergic sensitivity of ectopic discharges in an axotomized sensory neuron is mediated by 1 -AR. It is well known that sensory neurons develop abnormal, spontaneous activity after being separated from their peripheral receptors by axotomy (Devor and Jänig 1981; Devor et al. 1994; Kajander and Bennett 1992; Korenman and Devor 1981; Scadding 1981; Wall and Gutnick 1974). This abnormal, spontaneous activity is due to ectopically generated discharges and is considered to be an important contributor to central sensitization (Woolf 1995), and thus the activity plays a major role in the development and maintenance of neuropathic pain (Gracely et al. 1992; Sheen and Chung 1993; Yoon et al. 1996). Although the precise generation mechanisms and factors influencing ectopic discharges are not clear yet, sympathetic manipulations are known to influence the rate of the discharges. It was reported that sympathetic stimulation (Devor et al. 1994; McLachlan et al. 1993) or application of norepinephrine (Chen et al. 1996; Devor et al. 1994; Wall and Gutnick 1974; Xie et al. 1995) increases the rate of ectopic discharges. These sympathetically evoked ectopic discharges are usually blocked by an 2 -AR antagonist (yohimbine or idazoxan) but not by an 1 -AR antagonist (prazosin) (Chen et al. 1996; Leem et al. 1997; Xie et al. 1995). From these observations, it was suggested that 2 -AR is responsible for mediating the adrenergic dependency of ectopic discharges. The results of these studies appear to contradict those of the current study because our data show that the 1 -AR mediates neuropathic pain behaviors as well as ectopic discharges. However, it should be emphasized that the 2 -AR mediated responses reported by all of these previous studies are for sympathetically evoked discharges, not for ongoing discharges. In fact, until the current study, ongoing discharges have never been shown to be affected by any -AR antagonists. Our data suggest that ongoing ectopic discharges are in part maintained by an 1 -AR mediated mechanism. It is also possible that different subtypes of -AR are involved in mediating adrenergic sensitivity of pain behaviors in different strains of animals. The current study used the Lewis strain of rats, whereas the studies by Devor et al. (1994) and Chen et al. (1996) used the Wistar-derived Sabra strain of rats, and the one by Xie et al. (1995) used the Sprague-Dawley strain. Considering that the Lewis strain of rats is known to release a smaller quantity of norepinephrine in a stress condition than other strains (Dhabhar et al. 1993; Sternberg et al. 1992), it is possible that the Lewis strain represents a unique population of rats. The higher degree of adrenergic dependency of pain behaviors in Lewis rats compared with other strains (Lee et al. 1997) also suggests that strain difference may be an important factor in determining the degree of adrenergic dependency of pain behaviors as well as the subtype of -AR mediating the effect. Another complicating factor for comparing different studies is potential nonspecificity of various adrenergic agents used by each study. TRZ is a close structural analogue of the wellknown 1 -AR antagonist prazosin. It is more soluble in water than prazosin, and it has a longer half-life (Hoffman and Lefkowitz 1996). UK 14,304 is known to have higher specificity for the 2 -AR than clonidine (Andorn et al. 1988; Paris et al. 1989). IDZ has an imidazoline structure and interacts with both I 1 and I 2 imidazoline receptors (Langin et al. 1990), but the specificity of IDZ for 2 -AR is approximately five times higher than that of yohimbine (YHB) (Doxey et al. 1984). YHB is a nonimidazoline 2 -AR antagonist, but it is known to block sodium channels in the giant squid axon (Lipicky et al. 1978) as well as in the mouse brain (Huang et al. 1978; Zimanyi et al. 1988) in a use-dependent manner. In this study, neuropathic pain behavior was not reduced by either IDZ or YHB. For physiological study, we used IDZ exclusively because the voltage-gated sodium channels were suggested to be an important source of ectopic discharge generation, and YHB may interfere with these channels. Although IDZ may interact with imidazoline receptors, this seems to be more likely a problem in the CNS (King et al. 1995). Because -AR antagonists have strong cardiovascular effects, it is possible that the changes in the rate of ectopic discharges in in vivo experiments are secondary to the changes in blood pressure of the animal. However, monitoring systemic blood pressure while recording ectopic discharges revealed that 1) most blood pressure changes occur during the first 20 min after injection of -AR antagonists, whereas reductions of ectopic discharges do not begin until 30 min after the injection, and 2) both 1 - and 2 -AR antagonists influence blood pressure similarly, whereas a reduction of ectopic discharges occurs only after injection of 1 - but not after 2 -AR antagonist injection. These mismatches between changes in blood pressure and ectopic discharges suggest that the former is not a direct cause of the latter. In addition, the specificity of 1 -AR in modulation of ectopic discharges in in vitro experiments further suggests that the action of the 1 -AR antagonist is not a side effect. Our approach of recording ectopic discharges may appear to be inconsistent because we focused on spontaneous activity in the in vivo study on one hand and on evoked activity in the in vitro study on the other hand. Because our goal is to find out the subtype of -AR that is involved in SMP, it is essential to study sympathetically maintained ectopic discharges. Sympathetically maintained ectopic discharges are the activity that is present in the resting state, and hence these are influenced only by the release of adrenergic compounds because of basal sympathetic tone and circulating catecholamines. Therefore we examined spontaneous ectopic discharges in the in vivo experiments. The experimental conditions of the in vitro study, however, were different from that of the in vivo study. During isolation of the tissue for recording, the sympathetic supply is invariably denervated, and hence there is no longer basal sympathetic tone. Because we did not add catecholamines to the perfusion solution, there were no agents comparable with circulating catecholamines either. Therefore, when we evoke activity by adding adrenergic agonists to the perfusion solution in the in vitro study, we presumably mimic the presence of basal sympathetic tone and circulating catecholamines found in the in vivo condition. Therefore we focused on evoked responses in the in vitro study. The results of in vitro experiments showed that an 1 -AR agonist evokes an enhancement of ectopic discharges, whereas an 2 -AR agonist depresses them. An 2 -AR mediated depression of ectopic discharges could also be seen when EP was applied after pretreatment with TRZ, an 1 -AR blocker. These

2232 D. H. LEE, X. LIU, H. T. KIM, K. CHUNG, AND J. M. CHUNG excitatory and inhibitory actions of 1 - and 2 -AR agonists are in agreement with their general actions in the CNS (Millan et al. 1994; Pieribone et al. 1994) and in the sympathetic ganglia (Akasu et al. 1985; Brown and Caulfield 1979). In conclusion, this study showed that the 1 -AR in part mediates neuropathic pain behaviors in Lewis strain rats. Furthermore, both in vivo and in vitro electrophysiological experiments showed that ectopic discharges generated from injured afferents are also in part dependent on an 1 -AR mediated mechanism. These results suggest that the potential contribution of 1 -AR in generation of neuropathic pain in human patients needs to be examined more carefully. We thank Drs. J. Zhang and R. H. LaMotte for showing us their in vitro electrophysiological setup. We also thank Abbott Laboratories for a generous gift of TRZ. This study was supported by National Institute of Neurological Disorders and Stroke Grants NS-31680, NS-35057, and NS-11255. Address for reprint requests: J. M. Chung, Marine Biomedical Institute, University of Texas Medical Branch, 301 University Blvd., Galveston, TX 77555-1069. Received 8 September 1998; accepted in final form 26 January 1999. REFERENCES AKASU, T., GALLAGHER, J. P., NAKAMURA, T., SHINNICK-GALLAGHER, P., AND YOSHIMURA, M. Noradrenaline hyperpolarization and depolarization in cat vesicle parasympathetic neurones. J. Physiol. (Lond.) 361: 165 185, 1985. ANDORN, A. C., CARLSON,M.A.,AND GILKESON, R. C. Specific [ 3 H]UK14,304 binding in human cortex occurs at multiple high affinity states with alpha 2 - adrenergic selectivity and differing affinities for GTP. Life Sci. 43: 1805 1812, 1988. ARNÉR, S. Intravenous phentolamine test: diagnostic and prognostic use in reflex sympathetic dystrophy. Pain 46: 17 22, 1991. BONICA, J. J. Causalgia and other reflex sympathetic dystrophies. In: The Management of Pain, edited by J. J. Bonica. Philadelphia, PA: Lea & Febiger, 1990, p. 220 243. BROWN,D.A.AND CAULFIELD, M. P. Hyperpolarizing 2 -adrenoceptors in rat sympathetic ganglia. Br. J. Pharmacol. 65: 435 445, 1979. CHAPLAN, S. R., BACH, F. W., POGREL, J. W., CHUNG, J.M.,AND YAKSH, T.L. Quantitative assessment of tactile allodynia evoked by unilateral ligation of the fifth and sixth lumbar nerve in the rat. J. Neurosci. Methods 53: 55 63, 1994. CHEN, Y., MICHAELIS, M., JANIG, W., AND DEVOR, M. Adrenoreceptor subtype mediating sympathetic-sensory coupling in injured sensory neurons. J. Neurophysiol. 76: 3721 3730, 1996. CHOI, Y., YOON, Y. W., NA, H. S., KIM, S.H.,AND CHUNG, J. M. Behavioral signs of ongoing pain and cold allodynia in a rat model of neuropathic pain. Pain 59: 369 376, 1994. DAVIS, K. D., TREEDE, R. D., RAJA, S. N., MEYER, R.A.,AND CAMPBELL, J.N. Topical application of clonidine relieves hyperalgesia in patients with sympathetically maintained pain. Pain 47: 309 317, 1991. DEVOR, M. AND JÄNIG, W. Activation of myelinated afferents ending in a neuroma by stimulation of sympathetic supply in the rat. Neurosci. Lett. 24: 43 47, 1981. DEVOR, M., JÄNIG, W., AND MICHAELIS, M. Modulation of activity in dorsal root ganglion neurons by sympathetic activation in nerve-injured rats. J. Neurophysiol. 71: 38 47, 1994. DHABHAR, F. S., MCEWEN, B. S., AND SPENCER, R. L. Stress response, adrenal steroid receptor levels and corticosteroid-binding globulin levels a comparison between Sprague-Dawley, Fischer 344 and Lewis rats. Brain Res. 616: 89 98, 1993. DOXEY, J. C., LANE, A. C., ROACH, A.G.,AND VIRDEE, N. K. Comparison of the -adrenoceptor antagonist profiles of idazoxan (RX 781094), yohimbine, rauwolscine and corynanthine. Naunyn-Schmiedebergs Arch. Pharmacol. 325: 136 144, 1984. GRACELY, R. H., LYNCH,S.A.,AND BENNETT, G. J. Painful neuropathy: altered central processing maintained dynamically by peripheral input. Pain 51: 175 194, 1992. HARPER, A. A. AND LAWSON, S. N. Conduction velocity is related to morphological cell type in rat dorsal root ganglion neurones. J. Physiol. (Lond.) 359: 31 46, 1985. HOFFMAN, B. B. AND LEFKOWITZ, R. J. Catecholamines, sympathetominetic drugs, and adrenergic receptor antagonists. In: Goodman & Gilman s The Pharmacological Basis of Therapeutics, edited by J. G. Hardman, L. E. Limbird, P. B. Molinoff, R. W. Ruddon, and A. G. Gilman. New York: McGraw-Hill, 1996, p. 199 248. HUANG, L. M., EHRENSTEIN, G., AND CATTERALL, W. A. Interaction between batrachotoxin and yohimbine. Biophys. J. 23: 219 231, 1978. KAJANDER, K. C. AND BENNETT, G. J. Onset of a painful peripheral neuropathy in rat: a partial and differential deafferentation and spontaneous discharge in A and A primary afferent neurons. J. Neurophysiol. 68: 734 744, 1992. KIM, K. J., YOON, Y. W., AND CHUNG, J. M. Comparison of three rodent neuropathic pain models. Exp. Brain Res. 113: 200 206, 1997. KIM, S. H. AND CHUNG, J. M. An experimental model for peripheral neuropathy produced by segmental spinal nerve ligation in the rat. Pain 50: 355 363, 1992. KING, P. R., SUZUKI, S., LOUIS, W. J., AND GUNDLACH, A. L. Distribution of nonadrenergic [ 3 H]rilmenidine binding in rat brain and kidney. Ann. NY Acad. Sci. 763: 194 207, 1995. KORENMAN, E. M. AND DEVOR, M. Ectopic adrenergic sensitivity in damaged peripheral nerve axons in the rat. Exp. Neurol. 72: 63 81, 1981. LANGIN, D., PARIS, H., AND LAFONTAN, M. Binding of [ 3 H]idazoxan and of its methoxy derivative [ 3 H]RX821002 in human fat cells: [ 3 H]idazoxan but not [ 3 H]RX821002 labels additional non- 2 -adrenergic binding sites. Mol. Pharmacol. 37: 876 885, 1990. LEE, D. H. AND CHUNG, J. M. The effects of adrenergic blockers on neuropathic pain in an experimental animal model. Neurosci. Abstr. 23: 1534, 1997. LEE, D. H., CHUNG, K., AND CHUNG, J. M. Strain differences in adrenergic sensitivity of neuropathic pain behaviors in an experimental rat model. NeuroReport 8: 3453 3456, 1997. LEE, D. H., LIU, X., KIM,H.T.,AND CHUNG, J. M. Receptor subtype mediating the adrenergic sensitivity of ectopic discharges in axotomized sensory neurons. Neurosci. Abstr. 24: 2089, 1998. LEEM, J. W., GWAK, Y. S., NAM, T.S., AND PAIK, K. S. Involvement of 2 -adrenoceptors in mediating sympathetic excitation of injured dorsal root ganglion neurons in rats with spinal nerve ligation. Neurosci. Lett. 234: 39 42, 1997. LEEM, J. W., WILLIS, W.D.,AND CHUNG, J. M. Cutaneous sensory receptors in the rat foot. J. Neurophysiol. 69: 1684 1699, 1993. LIPICKY, R. J., GILBERT, D. L., AND EHRENSTEIN, G. Effects of yohimbine on squid axons. Biophys. J. 24: 405 422, 1978. LOH, L. AND NATHAN, P. W. Painful peripheral states and sympathetic blocks. J. Neurol. Neurosurg. Psychiatry 41: 664 671, 1978. LOH, L., NATHAN, P. W., SCHOTT,G.D.,AND WILSON, P. G. Effects of regional guanethidine infusion in certain painful states. J. Neurol. Neurosurg. Psychiatry 43: 446 451, 1980. MCLACHLAN, E. M., JÄNIG, W., DEVOR, M., AND MICHAELIS, M. Peripheral nerve injury triggers noradrenergic sprouting within dorsal root ganglia. Nature 363: 543 546, 1993. MILLAN, M. J., BERVOETS, K., RIVET, J.-M., WIDDOWSON, R. P., RENOUARD, A., MAROUILLE-GIRARDON, S. L., AND GOBERT, A. Multiple alpha-2 adrenergic receptor subtypes. II. Evidence for a role of rat R Alpha-2A adrenergic receptors in the control of nociception, motor behavior and hippocampal synthesis of noradrenaline. J. Pharmacol. Exp. Ther. 270: 958 972, 1994. NEIL, A., ATTAL, N., AND GUILBAUD, G. Effects of guanethidine on sensitization to natural stimuli and self-mutilating behaviour in rats with a peripheral neuropathy. Brain Res. 565: 237 246, 1991. PARIS, H., GALITZKY, J., AND SENARD, J. M. Interactions of full and partial agonists with HT29 cell 2 -adrenoceptor: comparative study of [ 3 H]UK- 14,304 and [ 3 H] clonidine binding. Mol. Pharmacol. 35: 345 354, 1989. PIERIBONE, V. A., NICHOLAS, A. P., DAGERLIND, Å., AND HÖKFELT, T. Distribution of 1, adrenoceptors in rat brain revealed by in situ hybridization experiments utilizing subtype-specific probes. J. Neurosci.14: 4252 4268, 1994. RAJA, S. N., TREEDE, R.-D., DAVIS, K. D., AND CAMPBELL, J. N. Systemic alpha-adrenergic blockade with phentolamine: a diagnostic test for sympathetically maintained pain. Anesthesiology 74: 691 698, 1991. RITTER, A. M. AND MENDELL, L. M. Somal membrane properties of physiologically identified sensory neurons in the rat: effects of nerve growth factor. J. Neurophysiol. 68: 2033 2041, 1992. ROBERTS, W. J. A hypothesis on the physiological basis for causalgia and related pains. Pain 24: 297 311, 1986.

ADRENERGIC RECEPTOR SUBTYPE MEDIATING NEUROPATHIC PAIN 2233 SATO, J. AND PERL, E. R. Adrenergic excitation of cutaneous pain receptors induced by peripheral nerve injury. Science 251: 1608 1610, 1991. SCADDING, J. W. Development of ongoing activity, mechanosensitivity, and adrenaline sensitivity in severed peripheral nerve axons. Exp. Neurol. 73: 345 364, 1981. SHEEN,K.AND CHUNG, J. M. Signs of neuropathic pain depend on signals from injured nerve fibers in a rat model. Brain Res. 610: 62 68, 1993. SHIR, Y. AND SELTZER, Z. Effects of sympathectomy in a model of causalgiform pain produced by partial sciatic nerve injury in rats. Pain 45: 309 320, 1991. STERNBERG,E.M.,GLOWA,J.R.,SMITH,M.A.,CALOGERO,A.E.,LISTWAK,S.J., AKSENTIJEVICH, S., CHROUSOS, G. P., WILDER, R. L., AND GOLD, P. W. Corticotropin releasing hormone related behavioral and neuroendocrine responses to stress in Lewis and Fischer rats. Brain Res. 570: 54 60, 1992. STEVENS, D. S., ROBINS, V. F., AND PRICE, H. M. Treatment of sympathetically maintained pain with terazosin. Regional Anesth. 18: 318 321, 1993. WADDELL, P. J. AND LAWSON, S. N. Electrophysiological properties of subpopulations of rat dorsal root ganglion neurons in vitro. Neuroscience 36: 811 822, 1990. WALL, P. D. AND GUTNICK, M. Ongoing activity in peripheral nerves: the physiology and pharmacology of impulses originating from a neuroma. Exp. Neurol. 43: 580 593, 1974. WOOLF, C. J. An overview of the mechanisms of hyperalgesia. Pulm. Pharmacol. 8: 161 167, 1995. XIE, Y., ZHANG, J., PETERSEN, M., AND LAMOTTE, R. H. Functional changes in dorsal root ganglion cells after chronic nerve constriction in the rat. J. Neurophysiol. 73: 1811 1820, 1995. YOON, Y. W., NA, H.S.,AND CHUNG, J. M. Contributions of injured and intact afferents to neuropathic pain in an experimental rat model. Pain 64: 27 36, 1996. ZIMANYI, I., LAJTHA, A., VIZI, E. S., AND REITH, M.E.A. Interaction of yohimbine with batrachotoxin binding to mouse brain sodium channels. Biochem. Pharmacol. 37: 641 645, 1988.