Prostaglandins Inhibit Endogenous Pain Control Mechanisms Blocking
Total Page:16
File Type:pdf, Size:1020Kb
The Journal of Neuroscience, April 1988, 8(4): 1348-i 349 Prostaglandins Inhibit Endogenous Pain Control Mechanisms bY Blocking Transmission at Spinal Noradrenergic Synapses Yetunde 0. Taiwo and Jon D. Levine Departments of Medicine, Oral Surgery and Anatomy, University of California at San Francisco, San Francisco, California 94143 Spinal intrathecal injections of the nonsteroidal antiinflam- confirmed that there was a peripheral site of action (Ferreira et matory analgesics (NSAIAs) indomethacin and acetylsali- al., 1978). Subsequentreports, however, indicate that prosta- cylic acid, which inhibit prostaglandin synthesis, cause dose- glandins may also contribute to nociception by an action in the dependent hypoalgesia in the rat. lntrathecal injections of central nervous system. For example, noxious stimuli elicit the prostaglandin-E, (PGE,) produce dose-dependent hyperal- releaseof prostaglandinsfrom the spinal cord (Ramwell et al., gesia. To determine whether this action of prostaglandins 1966); low dosesof prostaglandinsapplied via a spinal intrathe- on the central nervous system is mediated through pain- cal (IT) cannula lower nociceptive thresholds (Ferreira et al., generating or analgesia pathways, we studied the effect of 1978; Yaksh, 1982; Ferreira, 1983); IT administration of low intrathecal PGE, on endogenous opioid-induced analgesia. dosesof NSAIAs antagonizesperipherally induced nociceptive lntrathecal PGE, antagonized the analgesia produced by both effects (Ferreira et al., 1978; Ferreira, 1983); and simultaneous brain stimulation and intracerebroventricular morphine. In administration of NSAIAs into a peripheral inflammatory lesion contrast, the NSAlAs synergized with brain stimulation and and into the intracerebroventricular spaceproduces analgesic morphine-induced analgesia. The alpha-adrenergic antag- effects greater than those produced at either site alone (Ferreira onist phentolamine and the catecholaminergic selective et al., 1978). neurotoxin 6-hydroxydopamine, used to block tonic cate- The site at which prostaglandinsexert their central effects on cholamine activity in endogenous opioid-mediated analge- nociception is unknown. It hasgenerally beenassumed that they sia systems, prevented the hyperalgesia induced by in- act centrally, as they do peripherally, on neurons that transmit trathecal PGE,. Phentolamine did not, however, block the the nociceptive message(Yaksh, 1982). Prostaglandinscould, hyperalgesia caused by intradermal PGE,. These findings however, act at several spinal sites of nociceptive control, in- suggest that prostaglandins can block endogenous opioid- cluding the recently discoveredneural circuitry mediatingopioid- mediated analgesia systems by inhibiting the bulbospinal induced analgesia.This circuitry includes local spinal enkeph- noradrenergic component of this analgesia pathway. alin and dynorphin neurons and serotonin and noradrenaline terminals of descendingbrain-stem projection neurons (Bas- Nonsteroidal anti-inflammatory agents(NSAIAs), including as- baum and Fields, 1984). Since E-type prostaglandinsinhibit pirin (i.e., acetylsalicylic acid) and indomethacin, are effective noradrenergic neurotransmission (Bergstrom et al., 1973; in a variety of pain syndromes,but are most often usedto treat Hedqvist, 1973), it is possiblethat the CNS system effects of the hyperalgesiaor tendernessassociated with inflammatory E-type prostaglandins is produced by their effects on neuro- lesions. This hyperalgesia is presumed to originate from the transmitters at the spinal terminals of brain stem noradrenergic liberation of various inflammatory mediators, mainly prosta- neurons. glandins, that can sensitize peripheral terminals of primary af- In this report, we provide evidence to support the hypothesis ferent nociceptors (Ferreira, 1972; Ferreira et al., 1973; Willis that the central effects of prostaglandinson nociception are me- and Cornelsen, 1973). NSAIAs inhibit cyclooxygenation of ar- diated by their interaction with an endogenous,opioid-mediated achidonic acid by inhibiting prostaglandin synthetase,thus pre- analgesiasystem, and that this action of prostaglandinsis on venting the production of hyperalgesia-inducingprostaglandins the bulbospinal noradrenergic component of the system. (Ferreira and Vane, 1974; Moncada et al., 1975). Lim and colleaguesused the crossed-perfuseddog spleenprep- Materials and Methods aration to demonstrate that NSAIAs produce analgesiaperiph- The experimentswere performed on 250-350gm male Sprague-Dawley erally, not centrally (Lim et al., 1964). Studies using local in- rats (Bantin and Kingman,Fremont, CA), which wereimplanted with jection of small amounts of NSAIAs into inflammatory lesions a spinalcannula in the lumbarintrathecal (IT) spaceand stereotaxically either with a 22-gauge stainless steel guide in the third ventricle or with a bipolar stimulating electrode (Plastic Products, Roanoke, VA) placed Received May 20, 1987; revised Sept. 14, 1987; accepted Sept. 14, 1987. in the nucleusreticularis paragigantocellularis (NRPG) (Satohet al., We thank Professors Allan Basbaum and Roger Nicoll for many helpful dis- 1980).The location of cannulae and electrodes was histologically con- cussions of this manuscript, Dr. Franklin Perry and Professor Neal Benowitz for firmed. statistical consultation, and Dr. Michael Roizen for measurements of catechol- The substances used were prostaglandin-E, (PGE,), the prostaglandin- amine levels. This work was supported in part by NIH Grants AI 19784, NS2 1647, AM32634, and DE05369 and the Louis B. Mayer Foundation. synthesis inhibitors indomethacin and acetylsalicyclic acid (Sigma, St. Correspondence should be addressed to Jon D. Levine, Division of Rheuma- Louis, MO), morphine sulfate (Elkins Sinn, Cherry Hill, NJ), the alpha- tology and Clinical Immunology, U-426 (Box 0724), University of California, San adrenergic receptor antagonist phentolamine (Ciba-Geigy, Summit, NJ), Francisco, CA 94143. and the noradrenergic-specificneurotoxin 6-hydroxydopamine(6- Copyright 0 1988 Society for Neuroscience 0270-6474/88/041346-04$02.00/O OHDA, Calbiochem-Behring, La Jolla, CA). PGE, was dissolved in a The Journal of Neuroscience, April 1988, 8(4) 1347 lndomethacin (i.t.) Acetylsalicylic Acid (i.t.) r 2oo Paw Withdrawal &--4 160-I ,001 .oi .I 1 Morphine (i.c.v.) 4 sfh 200-l 2E 160- $ DoseW 120- Figure 1. Dose-dependence relationships of the effects on paw-with- 80- drawal (0) and tail-flick (0) thresholds of intrathecal (IT) NSAIAs (in- domethacin, n = 8 for both; acetylsalicylic acid, n = 18 and n = 9, .- respectively), prostaglandin E, (PGE,) (both, n = 14) and intracere- j broventricular (ICV) morphine (n = 32 and n = 16, respectively). The 0 response is calculated as a percentage change from baseline nociceptive ,005 .d5 .5 i threshold (2 f 1 SEM). The symbols for the responses in the tail-flick test (0) are displaced slightly to the right for clarity. Dose 01g) Figure 2. Dose-dependence relationships of the effects of ICV mor- phine (0) (n = 32 and n = 16, respectively) and the combinations of ICV morphine with IT indomethacin (10 rg) (0) (n = 12 and n = 9, vehicle of 10% ethanol in saline. Indomethacin was dissolved in 2% respectively) or with acetylsalicylic acid (1 ng) Q (n = 10 and n = 9, sodium bicarbonate and then titrated to pH 7.2 with monosodium respectively) on nociceptive threshold. phosphate and 6-OHDA in saline containing 1 mg/ml ascorbic acid. The remaining drugs were dissolved in saline. Spinal IT injections were in a volume of 10 ~1, followed by a 10 ~1 flush (i.e., the volume of the cannula) with vehicle, and intracerebroventricular (ICV) injections were Results in a volume of 1 ~1. NRPG stimulation-produced analgesia was elicited with a variable-intensity 50 Hz train of 500 Fsec monophasic pulses. The independent effectsof NSAIAs (i.e., indomethacin and ace- The stimulating electrode consisted of a pair of 25 pm insulated stainless tylsalicylic acid), IT prostaglandins, and ICV morphine on no- steel wires (Plastic Products). The stimulus was turned on 10 set prior ciceptive thresholds were first tested. IT administration of to testing nociceptive thresholds. The thres_hold for stimulation-pro- indomethacinand acetylsalicylicacid produceda significantdose- duced analgesia was 134 + 9 @A (n = 39) (X +- 1 SEM) in the control condition. After completion of the experiments, an electrolytic lesion dependentanalgesia (all p’s < 0.0 1; seeFig. 1). Equivalent doses was made to allow histological confirmation of the site of stimulation. of indomethacin, given systemically (i.e., 0.5 mg/kg and 4.0 mg/ The destruction of bulbospinal noradrenergic projection neurons by ICV kg, i.p.), did not significantly affect paw-withdrawal thresholds 6-OHDA was confirmed by radioenzymatic assay of noradrenaline (7.7 f 4.3%; n = 6 and 5.7 f 3.5%; n = 6, respectively). IT (DaPrada and Zurcher, 1976). ICV 6-OHDA produced a decrease of PGE, produced a significant dose-dependenthyperalgesia in 77% in noradrenaline at the lumbosacral level of the spinal cord. Two tests of nociception-the thermal tail-flick test @‘Amour and both tail-flick and paw-withdrawal tests (both p’s < 0.01; see Smith, 194 1) and the mechanical Randall-Selitto paw-withdrawal test Fig. 1). Activation of pain control circuits by ICV injection of (Randall and Selitto, 1957)-were employed in awake, restrained rats. morphine (5 Kg) produced a much greater analgesiathan did To prevent thermally induced injury to the tail in analgesic rats, a cutoff the NSAIAs in both tail-flick