Intrathecal Morphine Reduces Allodynia after Peripheral Nerve Injury in Rats via  Activation of a Spinal A1 Adenosine Receptor

After approval by the Animal Care and Use Committee (Wake Forest University School of Medicine, Winston-Salem, North Carolina), male Sprague-Dawley rats weighing 200–250 g (Harlan, Indianapolis, IN) underwent spinal nerve ligation as previously reported.14Briefly, during halothane anesthesia, the L5 and L6 spinal nerves were exposed and tightly ligated using 6-0 silk sutures distal to the dorsal root ganglion. Muscle layers and skin were closed using 4-0 silk sutures. After surgery, animals were housed individually in plastic cages in a climate-controlled room under a 12 h–12 h light–dark cycle, with free access to food and water. One week after nerve ligation, an intrathecal catheter was inserted as previously described.15During halothane anesthesia, a 32-gauge polyethylene catheter (ReCathCo, Allison Park, PA) connected to a piece of Tygon external tubing (Saint-Gobain Performance Plastics, Akron, OH) was inserted 7.5 cm through the cisterna magnum until the tip lay near the lumbar enlargement. Rats showing neurologic deficits were immediately euthanized by an overdose of pentobarbital. The remaining rats were allowed to recover 1 week before behavioral testing. After all of the behavioral experiments were completed, 10 μl lidocaine, 2%, was injected in some animals to confirm correct catheter position, as indicated by transient bilateral hind limb paralysis.

Withdrawal threshold of the left paw was measured in response to application of von Frey filaments, using an up–down method.16Hypersensitivity to tactile stimulation was confirmed 2 weeks after surgery, and animals with a withdrawal threshold greater than 4 g were excluded (2 of 25 rats). All behavioral testing was performed with the experimenter blinded to drugs and doses being administered.

On the day of the experiment, rats were placed in clear plastic boxes above a wire mesh floor. After acclimation for at least 20 min, calibrated von Frey filaments in log increments of force (0.69, 1.20, 1.48, 2.04, 3.63, 5.50, 8.51, 11.75, 15.14, and 28.84 g) were applied for a duration of 5 s to the midplantar aspect of the left paw, beginning with the 1.48-g filament. In addition to withdrawal threshold testing, animals were observed for overall behavior with special attention to any reduction in overall or hind limb movements.

All drugs for intrathecal injection were made freshly immediately before the experiments by an individual not performing the behavioral assessments. Morphine was diluted with saline. DPCPX was dissolved in dimethyl sulfoxide to a final concentration of 20 μg/μl as a stock solution and then diluted in saline just before injection. Dimethyl sulfoxide, 10%, was used as vehicle. Under gentle restraint, intrathecal drug injections were performed using a 100-μl Hamilton syringe. The drug or vehicle injection (10 μl) was followed by a 10-μl saline flush. Each group consisted of six to eight rats, and each rat received a maximum of four drug injections, with experiments separated by 1 week. Rats were randomly divided into at least two groups in each series to test the effects of vehicle, DPCPX (10 μg), morphine (0.25, 0.5, 1, 2.5, and 5 μg), and combinations (2.5 or 5 μg morphine with 3, 5.5, and 10 μg DPCPX). In addition, some animals received 5 μg intrathecal naloxone 15 min before 5 μg morphine injection. Finally, to test whether higher doses of morphine alone produced excitation rather than inhibition, 16 rats were randomly assigned to receive intrathecal saline (n = 2), 10 μg morphine (n = 6), and 30 μg morphine (n = 8). Withdrawal threshold was determined before and every 30 min for 4 h after injection. Dose–response curves were constructed at the time of peak effect and were converted to percentage maximal possible effect (%MPE) according to the formula %MPE = (post drug threshold − baseline threshold)/(presurgery threshold − baseline threshold) × 100.

Morphine sulfate was purchased from Astra Pharmaceutical Inc. (Westborough, MA). DPCPX was purchased from RBI Inc. (Natick, MA). Dimethyl sulfoxide and the remaining chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).

Withdrawal thresholds and %MPE were normally distributed and are presented as mean ± SE. Sigmoidal nonlinear regression curve fitting for dose–response data was performed, and the dose of morphine producing a 50% maximum efficacy (ED⁵⁰) or that of DPCPX producing a 50% maximum inhibition of morphine (ID⁵⁰) and their 95% confidence limits were calculated. Time-course and dose–response effects of morphine and DPCPX were analyzed using two-way repeated-measures analysis of variance with the Student–Newman–Keuls post hoc  test. P < 0.05 was considered significant.

Before spinal nerve ligation, the withdrawal threshold was 20.4 ± 0.4 g (n = 23). Two weeks after nerve ligation, the withdrawal threshold decreased significantly to 1.9 ± 0.2 g and remained unchanged over the maximum 4 weeks of study.

Morphine dose-dependently reversed hypersensitivity to mechanical simulation induced by spinal nerve ligation, with a peak effect 30 min after injection, whereas vehicle had no effect (fig. 1, left). Two-way repeated-measures analysis of variance showed that all doses except 0.25 μg morphine increased withdrawal threshold compared with vehicle 30 min after injection. The dose response for morphine was sigmoidal in the log rhythmic domain, with an ED⁵⁰of 0.79 μg (95% confidence interval, 0.75–0.88 μg; fig. 1, right). The just maximum effective dose of morphine was 2.5 μg (fig. 1, right). In separate studies with higher doses of morphine, saline was ineffective, whereas 10 μg morphine produced a 95 ± 3% MPE effect and 30 μg morphine produced a 100% MPE effect.

When preceded by vehicle injection, 5 μg intrathecal morphine increased withdrawal threshold from 2.1 ± 0.9 to 22.3 ± 0.8 g, whereas when preceded by 5 μg naloxone, this same dose of morphine did not alter withdrawal threshold (1.9 ± 0.9 g before and 2.9 ± 1.9 g after morphine). Thus, naloxone completely prevented the antihypersensitivity effect of a dose of morphine (5 μg) twice the just maximum effective dose.

DPCPX dose responses were performed against a just maximum effective dose of morphine (2.5 μg) and a dose twice this large. Against the 2.5-μg dose, all doses of DPCPX reduced the efficacy of morphine (fig. 2, left), whereas all but the lowest dose of DPCPX (3 μg) reduced the effect of 5 μg morphine (fig. 2, right). Maximum inhibition achieved was 70–80% (fig. 3), and the ID⁵⁰for DPCPX was 5.6 μg (95% confidence interval, 4.8–6.6 μg) against the 2.5-μg dose of morphine and 7.4 μg (95% confidence interval, 6.5–8.9 μg) against the 5-μg dose of morphine. These ID⁵⁰doses differ by t  test (P < 0.05). None of the treatments altered gross or hind limb movements, and animals ambulated normally when removed from the testing chamber.

The potency and efficacy of morphine in the current study surprised us because we observed only minor efficacy (< 30% MPE) after a large intrathecal morphine dose (30 μg) in a previous study.17Although it is theoretically possible that this large dose, also used in other laboratory studies in which morphine did not alter withdrawal threshold after spinal nerve ligation,3,4could have induced excitatory phenomena or acute hyperalgesia, hiding the efficacy of morphine, this was not observed in the current study. Rather, the current results are similar to recent reports of intrathecal morphine efficacy after nerve injury. In rats younger than those studied in the current protocol (140–160 g), the ED⁵⁰for morphine was 0.65 μg,5and the ED⁵⁰for morphine in the spared nerve injury model in adult rats was similar, at 0.52 μg.6Although there are methodologic differences among these studies, the ED⁵⁰values for morphine to relieve hypersensitivity to tactile mechanical stimuli are remarkably consistent, and the values obtained in these previous studies are within the confidence limits of that observed in the current study. Many factors, including rat strain, diet,18and housing,19can affect development of neuropathic behavior after nerve injury and could be responsible for the discrepancy between early and current reports, in addition to testing under blinded compared with unblinded conditions.7Of these, only blinded conditions were obviously different between our historic results and the current report.

We conclude, based on this and other recent studies, that intrathecal morphine is capable of reversing hypersensitivity to mechanical stimuli after peripheral nerve injury in animal models of neuropathic pain and that it does so at doses similar to those required to reduce hypersensitivity after acute injury, such as paw incision.20Whether morphine would maintain long-term efficacy in humans cannot be accurately assessed in rats because dramatic tolerance to intrathecal morphine occurs within 4–5 days in this species, and whether long-term intrathecal morphine should be administered for the treatment of neuropathic pain is not addressed in this laboratory study. However, unlike in previous reports,3,4we do not conclude that intrathecal morphine lacks short-term efficacy.

We previously noted that inhibition of adenosine metabolism in the spinal cord increased the efficacy of intrathecal morphine in rats with spinal nerve ligation,17consistent with the hypothesis that morphine’s mechanism of action to reduce hypersensitivity in these animals involves adenosine synthesis or release in the spinal cord. This follows logically from observations that morphine increases adenosine release from spinal cord synaptosomes from normal animals,8primarily from capsaicin-sensitive primary afferent terminals,21and that antinociception from intrathecal morphine is blocked by adenosine receptor antagonists.22In addition, spinal adenosine seems to play a key role in hypersensitivity states. Therefore, in normal rats and mice, intrathecal injection of adenosine receptor antagonists induces hypersensitivity,22,23and intrathecal injection of adenosine reduces hypersensitivity, including mechanical allodynia with neuropathic pain, in rats and humans.11,12,24–26 

The current observation, that the effect of intrathecal morphine in animals with spinal nerve ligation is essentially abolished by intrathecal DPCPX, adds significantly to these observations. We chose to probe A1 adenosine receptors using DPCPX because this subtype is involved in spinal antinociception in normal and nerve-injured animals.27,28DPCPX produced a dose-dependent inhibition of the effect of morphine in the current study, with an ID⁵⁰that increased with increasing dose of morphine, consistent with physiologic antagonism. This profound reliance of the action of morphine on spinal activation of adenosine receptors is unexpected, given our previous results in spinal cord synaptosomes of decreased potency of morphine to release adenosine after spinal nerve ligation. Several explanations are possible. Because the morphine concentration response to release adenosine in synaptosomes was shifted to the right after spinal nerve ligation without an effect on slope,13it is conceivable that this effect is overcome by the very high local tissue concentrations of morphine after intrathecal injection. Alternatively, intrathecal morphine could result in adenosine receptor activation via  an indirect pathway or circuit not measured in vitro  in synaptosomes.

Finally, we did not observe increased hypersensitivity with intrathecal injection of DPCPX alone, unlike observations in normal animals.22,23These previous observations suggested a tonic inhibitory function of spinal adenosine in the normal condition, and one might expect this inhibitory function is further enhanced after spinal nerve ligation because in vitro  studies of G-protein signaling indicate an increased basal inhibitory G-protein coupling due to tonic A1 adenosine receptor activation after spinal nerve ligation.29It is conceivable that the small dynamic range of von Frey filaments used at the very small forces in the hypersensitive animals after spinal nerve ligation reduced the sensitivity to observe further enhanced hypersensitivity from DPCPX to test this hypothesis.

In summary, this study confirms recent observations that intrathecal morphine reduces hypersensitivity to mechanical stimuli in rats after spinal nerve ligation with a potency similar to that in models of inflammation and postoperative pain. Morphine works via  opioid receptor activation and relies essentially on direct or indirect release of adenosine, which acts on an A1 adenosine receptor. These data suggest that intrathecal opioids may be effective short-term treatments for neuropathic pain and that drugs that reduce adenosine uptake or metabolism may further increase opioid potency in such patients.