Systemic Physostigmine Shows Antiallodynic Effects in Neuropathic Rats

Abstract

The aim of this study was to examine the antiallodynic and antinociceptive effects of subcutaneously administered physostigmine (50, 100, 200 [micro sign]g/kg), compared with morphine (2.5, 5, 10 mg/kg) and NaCl after spinal nerve ligation in rats. The following stimuli were used: acetone (cold allodynia), von Frey hairs (mechanical allodynia), and paw flick test (thermal nociception). Motility boxes were used to investigate the effects of the drugs on motor performance. Physostigmine attenuated both mechanical and cold allodynia dose-dependently but had no effect on the paw flick test. The effect was antagonized by atropine (muscarinic receptor antagonist) but not by mecamylamine (nicotinic receptor antagonist) or naloxone (opioid receptor antagonist). Morphine produced dose-dependent antiallodynic and antinociceptive effects. In the antiallodynic doses, morphine caused severe rigidity. Physostigmine 200 [micro sign]g/kg impaired locomotor activity, but no rigidity was observed. Implications: Physostigmine has different effects on allodynia and nociception, which suggests that different cholinergic (muscarinic) mechanisms may be involved in neuropathic and nociceptive pain.

(Anesth Analg 1999;89:428-33)

Tactile allodynia and poor response to opioids are two common features of neuropathic pain. Pain after nerve damage remains a major clinical problem because its mechanisms are still insufficiently understood. Most research has focused on the alpha_2-adrenergic and GABAergic mechanisms ^[1,2], tachykinin and glutamate ^[3] neurotransmitter systems, and Na⁺ channel mechanisms ^[4]. The potential cholinergic mechanisms of neuropathic pain have not been studied until recently when it was shown that both systemically administered ABT-954, a novel nicotinic agonist ^[5], and intrathecally administered neostigmine ^[6,7], produced a significant antiallodynic effect in a rat model of neuropathic pain.

Cholinergic drugs, such as muscarinic and nicotinic agonists and acetylcholinesterase-inhibitors (AchEInhs), produce dose-related antinociceptive activity in thermally, chemically, and mechanically evoked nociceptive tests in rodents and sheep ^[5,8,9]. AchEInhs can also potentiate the antinociceptive effects of opioids and clonidine ^[10,11]. IV physostigmine and intrathecal neostigmine also relieve nociceptive pain in humans ^[12-14]. There are no published studies about the efficacy of the AchEInhs in neuropathic pain in humans, and only a few studies have been published on their effects in animal models of neuropathic pain ^[6,7]. No studies that compare the antinociceptive and antiallodynic effects of AchEInhs are available.

We designed this study to examine the effects of systemically administered physostigmine compared with placebo and morphine in a rat model of neuropathic pain. The effects of these drugs were also studied in thermal nociception.

Methods

After approval by our institutional animal investigation committee, 40 male Sprague-Dawley rats (B&K Universal Ab, Solletuna, Sweden) weighing 210-340 g underwent spinal nerve ligation during halothane-induced anesthesia as described by Kim and Chung ^[15]. In brief, the left L5 and L6 spinal nerves were exposed by removing a small piece of the paravertebral muscle and by removing a small piece of the left spinous process of the L5 lumbar vertebra. The L5 and L6 spinal nerves were then carefully isolated and ligated tightly with 6-0 silk sutures distal to the dorsal root ganglion. Before and after the recovery from surgery, rats were housed in groups of seven in plastic cages. Lab chow and water were available ad libitum, and artificial lighting with a fixed 12-h light-dark cycle was provided. The experiments were performed according to the guidelines for animal research by local authorities and the International Association for the Study of Pain.

Two weeks after the surgery, the rats were placed on a metal mesh with a plastic dome for the assessment of mechanical and cold allodynia. They were allowed to habituate until exploratory behavior diminished. The threshold for mechanical allodynia was measured with a series of von Frey hairs ranging from 0.217 to 12.5 g (Semmes-Weinstein monofilaments, Stoelting, IL). The allodynic areas of the ventral surface of the paw were located using a 12.5-g von Frey hair. The animals that responded to a von Frey force <12.5 g were considered neuropathic and were used in the tests. After a withdrawal response of the paw to the stimulation was obtained, the next weaker hair was used until the threshold was found. In the following sessions, testing was started with the weakest hair that had produced a withdrawal response previously to minimize excessive stimulation. If the strongest hair did not elicit a response, 12.5 g was recorded as the threshold.

The temperature of the skin in the palm of the paw was measured by using an infrared thermometer. The sensor (3.0 mm) was placed directly under the paw that was in contact with the metal mesh. The distance between the skin and the sensor was approximately 2 mm and was kept constant by pressing a plastic ring surrounding the sensor against the metal mesh. After 1-2 s of stabilization, the reading was taken.

Cold allodynia was measured as the number of foot withdrawal responses after application of a drop of acetone to the heel of the rat with a syringe connected to a thin polyethylene tube. A brisk foot withdrawal was regarded as a sign of cold allodynia. The contralateral, nonneuropathic paw was tested first. Both paws were tested five times with an interval of 2 min between each test. The rat was considered to have cold allodynia if three or more withdrawals were observed.

The paw flick test was used to study the effects of the drugs on thermal nociception. A thermal (heat) light beam was set to 40 (units of the scale of the apparatus 0-90), and a cutoff time of 16 s was used to avoid tissue damage. The stimulus was begun only when the tested paw was set on the glass floor of the device. The values of two repeated paw flick measurements were averaged for both sides.

Spontaneous motility in a dark field ^[16] was assessed by using an automatic measurement system (Kungsbacka Mat & Reglerteknik AB, Kungsbacka, Sweden). The rats were placed in a sound isolated box (70 x 70 x 35 cm) equipped with two series of photocells located 2 cm and 12 cm above the floor. The cover of the box was closed to isolate it from ambient light and noises. The lower series of photocells detected movement of the animal as crossings of the photocell lines, and the upper photocells registered rearing. Twelve 5-min measurement periods were used to cover a 60-min assessment time. The rats were put into the boxes 15 min after the administration of the drugs. A decrease in spontaneous motility in a dark field could reflect, for example, motor dysfunction or sedation.

For neuropathic and nociceptive testing, doses of 50, 100, and 200 [micro sign]g/kg physostigmine and 2.5, 5, and 10 mg/kg morphine and saline were injected subcutaneously in volumes of 1.0 mL/kg using a randomized, blinded, cross-over design. Each dose group consisted of six rats. The same rat was used only after a 3-day interval. Neuropathic and nociceptive tests were performed before the drug administration and 30, 60, 90, 120, 180, and 240 min thereafter. At each time, the temperature of both paws was measured. Between these sessions, changes in the behavior of the rats, such as rigidity, were noted. In a separate trial, the rats received 200 [micro sign]g/kg physostigmine, followed by atropine (5 mg/kg), mecamylamine (2 mg/kg), naloxone (2 mg/kg), or saline in a randomized and blinded manner. Each dose group consisted of six rats. Allodynia was tested before and 30 min after these two injections.

For the motility tests, each rat was randomized to receive one of the following injections: 50 and 100 [micro sign]g/kg physostigmine or 5 mg/kg morphine or saline. Each dose group consisted of four rats. All drugs were dissolved in saline for injection. Atropine was obtained from YA-Kemia Oy, Helsinkin, Finland; physostigmine was obtained from Alcon-Couvreur, Puur, Belgium; and morphine and mecamylamine were obtained from RBI, Natick, MA.

Analysis of variance for repeated measurements, followed by Bonferroni's t-test, was used for the statistical analysis that consisted of continuous data. Pair-wise comparisons were performed by using Student's t-tests. P < 0.05 was considered to represent a significant difference. Continuous variables are presented as mean +/- SEM.

Results

Saline had no effects on the threshold levels to stimulation by von Frey hairs, which ranged from 3.3 +/- 0.4 to 4.5 +/- SEM 0.7 during the study. Compared with saline, statistically significant attenuation of mechanical allodynia was observed after the administration of physostigmine 100-200 [micro sign]g/kg and morphine 10 mg/kg (Figure 1). A significant antiallodynic effect of physostigmine 200 [micro sign]g/kg and morphine 5-10 mg/kg also was seen in acetone-induced cold allodynia (Figure 2). The antiallodynic effect of physostigmine was reversed by atropine but not by naloxone, mecamylamine, or saline (Table 1).

Figure 1: A, The effects of a subcutaneous injection of physostigmine ([black square] = 50, [square] = 100, and [black diamond] = 200 [micro sign]g/kg) and saline ([white circle]). B, The effects of morphine ([up triangle, filled] = 2.5, [up triangle, open] = 5, and [triangle down, filled] = 10 mg/kg) and saline ([white circle]) on the withdrawal threshold to stimulation with von Frey filaments (mechanical allodynia). Mean values +/- SEM of six rats are given. *P < 0.05, **P < 0.001 statistically significant differences between physostigmine or morphine and saline.

Figure 2: A, The effects of a subcutaneous injection of physostigmine ([black square] = 50, [square] = 100, and [black diamond] = 200 [micro sign]g/kg) and saline ([white circle]). B, the effects of morphine ([up triangle, filled] = 2.5, [up triangle, open] = 5, and [triangle down, filled] = 10 mg/kg) and saline ([white circle]) on the withdrawal threshold to stimulation using five consecutive drops of acetone (cold allodynia). Mean values +/- SEM of six rats are given. *P < 0.05, **P < 0.01 statistically significant differences between physostigmine or morphine and saline.

Table 1: Effects of Opioid and Cholinergic Antagonists on Physostigmine-Induced Reversal of Allodynia

In the paw flick test, physostigmine had no effect, but morphine caused a dose-dependent antinociceptive effect (Figure 3). No difference in responses was found between the neuropathic and control paws.

Figure 3: The effects of subcutaneous physostigmine ([black square] = 50, [square] = 100, and [black diamond] = 200 [micro sign]g/kg) and saline ([white circle]) on the thermal nociception in the control paw (A) and the effects of subcutaneous morphine ([up triangle, filled] = 2.5, [up triangle, open] = 5, and [triangle down, filled] = 10 mg/kg) and saline ([white circle]) on the thermal antinociception in the control paw (B). The mean paw flick latencies +/- SEM of six rats are given. *P < 0.05, **P < 0.01 statistically significant differences between morphine and saline.

During the testing, three of six rats became rigid after the administration of morphine 5 mg/kg, and six of six rats became rigid after the administration of morphine 10 mg/kg. Although rigidity was not graded systematically, its duration was longest (up to 60-120 min) after the largest dose of morphine. Neither rigidity nor salivation was observed after physostigmine administration. Compared with saline, locomotor activity decreased significantly after physostigmine 200 [micro sign]g/kg and morphine 5 mg/kg administration (Figure 4). The largest dose of morphine was not tested because it caused rigidity in all rats.

Figure 4: The effects of an injection of physostigmine ([square] = 100, and [black diamond] = 200 [micro sign]g/kg), morphine ([white circle] = 5 mg/kg) and saline ([white circle]) on the total activity of the rats in the motility box. Recordings started 15 min after the injection of the drug. Mean values +/- SEM of four rats are given. *P < 0.05, **P < 0.01 statistically significant differences between the physostigmine or morphine and saline.

The baseline temperature of the left (allodynic) paw was 30.8 +/- 0.3[degree sign]C and that of the control paw was 31.4 +/- 0.6[degree sign]C (P < 0.05). Physostigmine 200 [micro sign]g/kg decreased the temperature of both paws, with a mean value of 1.5 +/- 0.2[degree sign]C in a statistically significant (P < 0.05) manner. Other drugs had no effects on the temperature.

Discussion

Our main finding was that physostigmine produced a dose-dependent antiallodynic effect after spinal nerve ligation but had no effect on thermal nociception in the rat. The antiallodynic effects of physostigmine are in agreement with the findings by Lavand'homme et al. ^[6] and Hwang et al. ^[7], who recently reported that intrathecal neostigmine reversed tactile allodynia in the spinal nerve ligation model of neuropathic pain ^[15].

Some previous studies ^[8,10,11,17] have reported antinociceptive effects of physostigmine and neostigmine in the hot plate test. The 50% effective dose for IM physostigmine in the rat has been reported to be 550 [micro sign]g/kg in the hot plate test ^[18]. We used smaller doses, which may be one reason why we did not find antinociceptive effects after the administration of physostigmine. In a study by Beilin et al. ^[10], the maximal possible antinociceptive effect of 100 [micro sign]g/kg subcutaneous physostigmine combined with 2 [micro sign]g of intrathecal morphine was only 40%, and one study failed to demonstrate any effect of 15 [micro sign]g of intrathecal physostigmine in the hot plate test ^[19]. The duration of antinociception caused by a single systemic or intrathecal injection of an AchEInh is considerably shorter than the effect of morphine ^{[10,11,17,20]}, which may be another reason for the lack of antinociceptive effect 30 minutes after the administration of physostigmine in our study. The most likely explanation for our observations is that different cholinergic mechanisms operate in thermal nociception and allodynia.

AchEInhs are believed to produce antinociception by inhibiting the breakdown of endogenous acetylcholine, which leads to the stimulation of muscarinic and nicotinic cholinergic receptors. Cholinergic binding sites are found in many areas in the central and peripheral nervous systems associated with the transmission of pain, such as in the thalamic region ^[21], in pedunculopontine tegmental and raphe magnus nuclei ^[22], and in substantia gelatinosa of the dorsal horn ^[23]. Additionally, by binding directly to the peripheral endings of primary afferents, acetylcholine may hyperpolarize nociceptive neurons, reduce the amount of pronociceptive neurotransmitters, and activate the nitric oxide-cyclic guanosine monophosphate pathway ^[24].

The mechanisms of allodynia in neuropathic pain are different from those of nociception. Abnormal sensory changes observed after peripheral nerve injury may occur because of structural changes in the central nervous system. Sprouting of both large A beta fibers into lamina II ^[25] and noradrenergic perivascular axons into dorsal root ganglia ^[26] has been described. Transsynaptic degeneration in spinal dorsal horn neurons is found after a constriction injury of the nerve ^[27]. Additionally, spontaneous discharge in A beta and A delta fibers develops after the ligation of sciatic nerve in the rat ^[28]. The role of the cholinergic system in the modulation of these events has not been studied thoroughly. Our study and two other recent articles ^[6,7] provide evidence for an important role of cholinergic mechanisms in the development of allodynia. We did not test whether the muscarinic antagonists would worsen the neuropathic symptoms in the rats, but they have previously been shown to decrease the nociceptive mechanical (but not thermal) withdrawal thresholds in rats ^[29]. It would be interesting to determine whether such a tonic endogenous cholinergic inhibitory control mechanism exists in neuropathic pain.

As a liposoluble substance, physostigmine penetrates the blood-brain barrier and may cause analgesia after subcutaneous administration by both central and peripheral mechanisms. Based on the present findings, we cannot define the site of action because all the antagonists we used may also cause both central and peripheral effects. The fact that atropine (a muscarinic receptor antagonist), but not mecamylamine (a nicotinic receptor antagonist) or naloxone (an opioid receptor antagonist), antagonized the antiallodynic effects of physostigmine indicates that the antiallodynic effect of physostigmine was mediated via muscarinic receptors. This observation is in agreement with the recent findings of Hwang et al. ^[7]. According to previous studies, both postsynaptic muscarinic M1 and presynaptic muscarinic M2 receptors are involved in supraspinal antinociception ^[8], but spinal cholinergic antinociception is mediated predominantly via the M1 receptor ^[30].

Systemic and supraspinal, but not spinal [micro sign]-opioid receptor, agonists reduce tactile allodynia in a dose-dependent manner after spinal nerve ligation in rats ^[31]. In the present study, morphine produced an antiallodynic effect only in doses that caused severe rigidity in all rats; some of these animals raised their paws lightly but failed to produce a clear response after stimulation with the von Frey filaments. Although Lee et al. ^[31] showed that similar systemic doses of morphine (6-10 mg/kg) reversed tactile allodynia without causing catalepsy, we would argue that the unresponsiveness of rats to tactile stimuli after such doses of morphine may be attributed to the motor impairment.

We observed a slight but statistically significant decrease in the paw temperature caused by the largest dose of physostigmine, which may be because of increased sympathetic outflow in these rats ^[32]. The same dose (200 [micro sign]g/kg) of physostigmine impaired motor activity in the motility boxes, which agrees with previous findings with intrathecal physostigmine ^[23]. The motor dysfunction caused by AchEInhs has been suggested to be caused by both a direct depolarization of the motor neuron and inhibition mediated via Renshaw cells ^[23]. However, during the testing of allodynia, we did not notice motor dysfunction or rigidity in the behavior of the rats treated with physostigmine. The antiallodynic effect of physostigmine cannot be explained by motor dysfunction caused by the drug because the antiallodynic effect lasted longer than the motor impairment. In addition, during the nociceptive testing, the animals moved freely after physostigmine administration, but not after administration of larger doses of morphine.

The use of cholinomimetic drugs, either alone or in combination with other drugs, for the relief of neuropathic pain seems to be a promising approach. Synergistic antinociceptive interaction has been shown between AchEInhs and opioids and alpha_2-adrenergic agonists ^[9-11]. Synergism has also been reported for the antiallodynic effect of intrathecal neostigmine and clonidine ^[6]. However, potential adverse effects, such as nausea, motor weakness, or bradycardia, may limit the wider clinical use of AchEInhs. Whether these disadvantages can be overcome by selecting the optimal compound (M1/M2 receptor-specific muscarinic agonist) and dose and route of administration remains to be studied.

In conclusion, systemic physostigmine attenuates both mechanical and cold allodynia in a dose-dependent manner. The effect was reversed by atropine, which suggests that physostigmine acted via muscarinic receptors. Morphine caused a dose-dependent antiallodynic and antinociceptive effect and caused pronounced rigidity. At a dose of 200 [micro sign]g/kg, physostigmine impaired motor activity and increased the sympathetic outflow in the rats.