93
Pain, 42 (1990) 93-101 Elsevier
PAIN 01597
Interactions between substance P, calcitonin gene-related peptide, taurine and excitatory amino acids in the spinal cord David
H. Smullin,
Department (Received
of Veterinay
Stephen
R. Skilling
and Alice
Biology, Unioersity of Minnesota,
26 July 1989, revision received
8 November
A. Larson
St. Paul, MN 55108 (U.S.A.)
1989, accepted
10 January
1990)
Using in viva microdialysis in the dorsal spinal cord of the rat, we have previously observed increases in glutamate SUmmarY and aspartate during exposure to a noxious stimulus. The present investigation was designed to determine whether these increases may be mediated by substance P. Infusion of 1 mM of substance P in the dialysis fluid increased the concentrations of glutamate and aspartate, similar to the response seen during noxious stimulation. In addition, substance P also increased the concentrations of the inhibitory amino acids glycine and taurine. Calcitonin gene-related peptide, previously shown to enhance substance P-induced biting and scratching behavior, produced no effect on amino acid release by itself but potentiated the apparent release of taurine by substance P. To assess the importance of substance P-induced amino acid release in sensory processing, we examined the influence of taurine and of excitatory amino acid antagonists on the biting and scratching behavior produced by excitatory amino acids and substance P. Taurine selectively inhibited only substance P-induced biting and scratching while excitatory ammo acid antagonists inhibited only excitatory amino acid-induced behavior. To further explore the ability of taurine to inhibit the substance P-induced behavior, 3 tests of nociception were then used. Pretreatment with taurine inhibited the nociceptive-related writhing behavior produced by an intraperitoneal injection of acetic acid in mice but failed to alter the latency of response in the hot plate or tail flick assay. These results indicate that substance P release during nociception may mediate the release of glutamate and aspartate but that these excitatory amino acids do not appear to be involved in substance P-induced biting and scratching behavior. The selective antinociceptive effect of taurine on chemical, but not thermal, tests of nociception suggests that its release in response to substance P and enhanced release following calcitonin gene-related peptide may be important in an endogenously evoked analgesia.
Key words:
Nociception;
Substance
P; Excitatory
ammo
acids; Taurine;
Introduction The excitatory amino acids glutamate (Glu) and aspartate (Asp) and the peptide substance P (SP) have been proposed as primary afferent neurotransmitters in the spinal cord involved in nociception [3,17,21,22]. We have recently demonstrated increased concentrations of Asp and Glu
Correspondence Veterinary Biology, Building, University MN 55iO8, U.S.A.-
0304-3959/90/$03.50
to: Dr. D.H. Smullin, Department of 295 Animal Science/Veterinary Medicine of Minnesota, 1988 Fitch Ave.. St. Paul.
0 1990 Elsevier Science Publishers
Calcitonin
gene-related
peptide;
Microdialysis
in the extracellular fluid of the rat dorsal lumbar spinal cord after intradermal injection of formalin [24]. While the increase in Glu and Asp may reflect primary afferent release it is also possible that this increase reflects release from descending or spinal interneurons in response to a primary release of SP, calcitonin gene-related peptide (CGRP) or other neurotransmitter. In support of this latter conclusion it has been shown that the application of SP to the hemisected frog and rat spinal cord caused the release of Glu into the superfusate [11,12]. In addition, the intrathecal injection of excitatory amino acid (EAA) agonists,
B.V. (Biomedical
Division)
such as ~-methyl-~-aspartate (NMDA), kainic acid (KA) and quisqualate (Quis) elicit behavioral responses similar to that produced by SP [l,lO]. In addition to SP, a variety of other neuropeptides have been found in primary afferent fibers. CGRP, for example, has been associated with pain transmission and is uniquely distributed in rat dorsal horn sensory fibers of laminae I and II [26] where it has been immunocytochemically co-localized with SP [27]. While the intrathec~ administration of CGRP alone produces no behavioral effects, the injection of SP plus CGRP both potentiates and prolongs the caudally directed biting and scratching behavior produced by SP alone [27]. The purpose of the present investigation was to determine if the release of either Glu or Asp that was observed during noxious stimulation is from primary afferent fibers or if this release is mediated through an initial release of SP from primary afferents. Apparent release was determined by changes in the concentrations of putative amino acid transmitters in the extracellular fluid of rat dorsal spinal cords using in vivo microdialysis. The importance of SP-induced changes in amino acid concentrations was then examined using nocifensive responses to acetic acid injected into the peritoneal cavity and motor responses that involve caudally directed biting and scratching in response to intrathecally injected excitatory compounds.
Methods and Materials Microdia Eysis
Male Sprague-Dawley rats (275-300 g) were anesthetized with sodium pentobartial(50 mg/kg, i.p., Veterinary Laboratories, Inc.) and implanted with a microdialysis fiber (200 pm diam., 50,000 MW cut-off, Amicon Vitafiber II) through the dorsal spinal cord at the level of L5/L6 as described previously [24]. Fo~owing a 24 h recovery period, the animals were evaluated for any signs of limb paralysis or impaired movement. No impairment was observed in any of the animals in this study, and all animals appeared to behave normally. The dialysis system
was attached to a peristaltic pump (Gilson Minipuls 3) and perfused with Ringer’s solution (147 mM NaCl; 4.0 mM KCl; 2.2 mM CaCI,) at 5-6 pl/min for 90 min to establish a diffusion equilibrium. Samples were collected at 10 min intervals in polypropylene tubes and maintained at 5°C until analyzed for amino acids, within 12 h, by HPLC as described by Skilling et al. [24]. Animals were maintained at 22* C with 12 h light/dark cycles and were provided with food and water ad libitum. Following each experiment, the animals were killed and the cannulae injected with methylene blue dye. The spinal cord was then removed and fixed for gross histological confirmation of cannula ptacement. Only animals with cannulas located below lamina I and above the central canal were included in this study. In addition, animals were not included in this study if evidence of dye leakage into the cerebral spinal fluid was present. Three control samples were collected for determination of basal concentrations of amino acids. Ad~nistration of SP (Bachem), CGRP (Cambridge Research Biochemicals) or SP plus CGRP was made by infusion of 0.002-1.0 mM of SP, lo-’ 7-10p5 M of CGRP or 1.0 mM of SP plus IO--’ M CGRP in Ringer’s solution for 40-60 min. Samples were collected throughout the course of the experiments described above. The data were transformed into percent of basal concentration of extracellular fluid for each animal based on in vitro calibration experiments. The in vitro dialysis efficiency was between 1 and 3% for all of the amino acids measured. in vitro measurements. using known concentrations of peptide standards at 37 ’ C, have shown a linear relationship between SP concentration in the standards and recovery of SP in the dialysis solution as measured by HPLC. Assuming that the diffusion of SP across the dialysis membrane is approximately equal in both directions, 1.0 mM SP in the perfusate would result in an estimated maximum diffusion of only 1.0 nmole of SP into the extracellular fluid of the spinal cord during each 10 min period. The mean percent increase of the first 3 samples following treatment was transformed into log vaIues and then analyzed across treatments using analysis of
95
variance (ANOVA) and ScheffC’s F test for comparison of individual means. Intrathecal injections Male Swiss-Webster mice weighing 15-20 g (Biolab, White Bear Lake, MN) were used in all experiments in which intrathecal (i.t.) injections were made. All i.t. injections were made in unanesthetized mice at approximately the L5/L6 intervertebral space using a 30-gauge 0.5 inch needle on a 50 ~1 luer tip Hamilton syringe after the method of Hylden and Wilcox [9]. Substance P (Peninsula; 3.7 X lop3 nmol), NMDA (Sigma; 0.2 nmol), kainic acid (KA, Sigma; 0.075 nmol), Quis (Sigma; 1.0 nmol), taurine (Tau, Sigma; 0.15-12.0 nmol), Gly (Sigma; 12.0 nmol), 6-aminoethyl-3-methy1-4H-1,2,4-benzothiadiazine-l,l-dioxide (TAG, Merck Frosst, D-2-amino-5-phosphonoCanada; 4.4 nmol), valerate (APV, Sigma; 0.25-5.0 nmol) and yglutamylaminomethylsulfonic acid (GAMS, Sigma; 50.0 nmol) were administered in a volume of 5 ~1 of 0.85% saline containing 0.01 N acetic acid. Mice were placed in a plexiglass container immediately after the injection and the number of bites and scratches were counted over a 60 set period. Data were analyzed using Student’s 2-tailed t test for unpaired data or by ANOVA with Scheffe’s F test for comparison of individual means. Writhing assay Abdominal constrictions were produced in mice by the intraperitoneal (i.p.) injection of 0.3 ml of a 1% solution of acetic acid as described previously by Hayashi and Takemori [8]. The number of behaviors was counted over a 5 min interval beginning 5 min after the injection of acetic acid. Animals were killed immediately after the termination of each recording period. Data were analyzed using ANOVA with Scheffss F test for comparison of individual means. Hot plate assay In the hot plate test, mice were placed on a plate heated to 56°C and the time required for the animal to lick its front paw was recorded, at which time the animals were immediately removed
from the hot plate. If an animal showed no reaction in 30 set it was removed from the hot plate. Data were analyzed using Student’s 2-tailed t test for unpaired data. Tail flick assay The time required for a mouse to move its tail when exposed to a radiant heat source was recorded. If an animal showed no reaction in 10 set the experiment was terminated. Data were analyzed using Student’s 2-tailed t-test for unpaired data.
Results Microdialysis experiments in rats The means of 3 samples collected before treatment were used for determination of amino acid concentrations in the dorsal spinal cord extracellular fluid (ECF). Basal concentrations k S.E.M. (PM) were: Asp 3.53 & 0.25, Glu 7.45 f 2.02, Asn 7.65 + 1.46, Gly 47.87 f 11.24 and Tau 10.67 + 1.71. These values represent dialysate concentrations corrected for cannula efficiency, based on in vitro calibration. The extracellular fluid concentrations of Asp, Glu, Gly and Tau increased to 150, 150, 125 and 170% (P < 0.05) of basal concentrations respectively (Fig. 1) during a 60 min perfusion with 1.0 mM of SP. The maximal increase in concentrations of these amino acids occurred within 20 min of SP infusion. In experiments using a 40 min perfusion with 1.0 mM of SP, the concentrations of amino acids increased followed by a return to basal levels within 10 min of removal of SP from the perfusate (Fig. 2). Doses of SP lower than 1.0 mM had no significant effect on the concentrations of any of the amino acids measured. The concentration of the amino acid asparagine (Asn) did not change in response to any concentration of SP (Fig. 1). Infusion of 10-7-10-5 M CGRP by itself had no significant effect on the extracellular fluid concentrations of any of these amino acids (Fig. 1). Perfusion with 1.0 mM of SP plus 10m5 M CGRP increased Tau concentrations in the ECF 3-fold (P < 0.05), significantly greater than the increase produced by SP alone (P < 0.05) but did not alter
96 400
peptides in the spinal cord and lasted less than the first 10 min of the perfusion period. Although not quantified, no obvious differences were observed in either the intensity or duration between the behaviors produced by SP alone or SP with CGRP.
.-6 E E z
300
IJ
Ringer’s
q n
3 SP SP
El
6 0
t CGRP
Intruthecal experiments in mice Intrathecal injection of mice with 3.7 x 10 ' nmol of SP, 0.2 nmol of NMDA, 0.075 nmol of KA or 1.0 nmol of Quis resulted in caudally
z E
100
z & p. 0 Asn
GIU
TaU
GUY
Amino Acid Fig. 1. Effects of CGRP (1O-‘5 M). SP (1.0 mM) and SP+ CGRP on the concentration of extracellular amino acids in viva from the dorsal spinal cord (n = 4-5 each). Basal concentrations were determined as the mean values from 3 consecutive fractions before treatment with peptides and subsequent samples were expressed as the percent of basal concentration For statistical analysis the data were further transformed to log values. The rats were perfused with peptide for 60 mm and the results represent the mean*S.E.M. (bars) of the first 3 samples following treatment. Ten minute fractions were collected at a flow rate of 5-6 &mm. A single asterisk indicates a difference of P -C0.05 by ANOVA with Scheffe’s F test for comparison of individual means between treatment and control (Ringer’s). Two asterisks indicates a difference of P -Z0.05 by ANOVA with Scheffe’s F test for comparison of individual means between SP and SP + CGRP.
A
SP
NMDA
the response to SP for the other amino acids measured (Fig. 1). Intermittent caudally directed biting and licking behavior was observed in most animals after receiving either SP alone or SP with CGRP. This behavior, when present, began at a time that corresponded with the arrival of the
m : co z
Asp
Avg
---a---
Avg Gly Avg Tau
300 200
-
100 -
E s z a
0 -60
~30
0
30
60
90
120
I 150
Time (min)
Fig. 2. Time course of the effect of SP (1.0 mM) on the extracellular concentrations of ammo acids in vivo from the rat dorsal spinal cord (n = 3). Treatment with SP was begun at 0 min and lasted for 40 min. Details of experiment as in Fig. 1.
7
5 SP
-
+ 1liil KA
NMDA
KA
6
6
OUIS
Fig. 3. A: the effects of co-administration of Tau (12.0 mmol; striped bars) on the caudally directed biting and scratching behavior elicited by intrathecal injections of SP (3.7X 10e3 nmol), NMDA (0.2 nmol), KA (0.075 nmol) and Quis (1.0 nmol) (solid white bars) in mice. B: the effects of co-administration of GAMS (50.0 nmol; striped bars) on the caudally directed biting and scratching behavior elicited by intrathecal injections of SP (3.7x10-s nmol), NMDA (0.2 nmol), KA (0.075 nmol) and Quis (1.0 nmol) (solid white bars) in mice. Drugs were administered in 5 pl of 0.85% saline containing 0.01 N acetic acid and the number of behaviors were counted over a 60 set period. Asterisk indicates a difference of P < 0.01 between injection of peptide and of peptide plus Tau by unpaired t test. The number of mice in each group is indicated by the number at the base of each bar.
97
directed biting and scratching behavior (Fig. 3) as previously reported [l]. The co-administration of 12.0 nmol of Tau with either NMDA, KA or Quis had no effect on the number of behaviors elicited by these EAA agonists, however, at this dose Tau totally inhibited SP-induced biting and scratching (Fig. 3A). This total blockade of SP-induced behaviors by Tau was found at concentrations as low as 1.5 nmol. In contrast, co-administration of 12.0 nmol of Gly elicited only 50% inhibition of the SP-induced behaviors. The i~bito~ effect of 12.0 nmol of Tau on SP-induced behavior was blocked by co-administration of TAG, a Tau antagonist, at a dose that had no effect when injected alone (4.4 nmol) or in combination with SP (Fig. 4). Intrathecal injection of 50.0 nmol of GAMS, reportedly a KA and Quis ~tago~st with less ability to inhibit NMDA-induced excitation [6], did not affect the SP- or Quis-induced behaviors, but did inhibit the behavior caused by NMDA and KA (Fig. 3B). Similarly, co-administration of 5.0 nmol of APV, an NMDA specific antagonist, with SP had no effect on the SP-induced behavior, while 1.0 nmol was sufficient to totally inhibit NMDA-induced behavior of similar intensity (data not shown). Neither Tau nor Gly produced behavioral effects when injected alone at 12.0 nmol
I_
6
6 SP+Tau
SPtTau+TAG
SP+TAG
Treatment
Fig. 4. The effects of co-administration of Tau (12.0 nmol), TAG (4.4 rnnol) and of Tau (12.0 nmol) plus TAG (4.4 nmol) on the caudally directed biting and scratching behavior elicited by intrathecal injections of SP (3.7~ 10-s nmol) in mice. Details of experiments as in Fig. 3. Asterisk indicates a difference of Pi 0.05by ANOVA with Scheffe’s F test for comparison of individual means between SP+Tau and SP, SP + Tau + TAG and SP + TAG.
I_ t
cl_ 6
Tall
6
Tau+TAG
7 TAG
Treatment
Fig. 5. The effects of pretreatment with saline, Tau (12.0 nmol), TAG (4.4 nmol) and of Tau (12.0 nmol) plus TAG (4.4 nmol) on the writhing behavior induced by an i.p. injection of acetic acid. Writhing was produced in mice by the i.p. injection of 0.3 ml of a 1% solution of acetic acid as described previously by Hayashi and Takemori [8]. The number of writhes were counted over a 5 mm interval beginning 5 min after the injection of acetic acid. Drugs were administered intrathecally in 5 ~1 of 0.85% saline containing 0.01 N acetic acid 2 min prior to the i.p. injection of acetic acid. Asterisk indicates a difference of P < 0.05by ANOVA with Scheff& F test for comparison of individual means between Tau and Control, Tau+TAG and TAG. The number of mice in each group is indicated by the number at the base of each bar.
during the 60 set recording period but doses higher than 120.0 nmol and 200.0 nmol, respectively, elicited biting and scratching in addition to dyskinetic writhing-like movements as reported previously [14]. In order to ascertain whether Tau has an effect on behavioral tests of nociception used historically to determine whether a compound is analgesic, mice were pretreated with an it. injection of 12.0 nmol of Tau 2 min prior to an i.p. injection of acetic acid, or 1 and 5 min prior to either a hot plate or tail flick test. Taurine had no effect on either the hot plate or tail flick assays (data not shown) while the writhing behavior induced by acetic acid was inhibited by Tau (Fig. 5). This inhibition was reversed by TAG (4.4 nmol) which had no effect by itself (Fig. 5).
The present study, demonstrating a dramatic increase in the concentrations of Asp and Glu in
the ECF in response to infusion of SP (Fig. l), shows changes in EAAs that are qualitatively similar to those changes seen after formalin injection in the rat hind paw [24]. These increases are believed to reflect an increased release of these compounds from neurotransmitter pools into the ECF. The ability of infused SP to elicit this response suggests that the apparent increase in Asp and Glu release during chronic pain may be mediated, at least in part, by an initial release of SP from primary afferent fibres where SP is believed to act as a nociceptive transmitter [17,21]. The ability of SP alone to mimic the noxious stimulusinduced increases in Glu and Asp may be due to a second-order release of EAAs from interneurons in response to the release of SP from primary afferent fibers. The release of Asp and Glu during pain and SP infusion may also be due to presynaptic facilitation of the release of these EAAs by SP. The latter possibility is supported by recent findings that Glu and SP coexist in primary afferent terminals in the rat spinal cord [7]. Although the formalinand SP-induced increases in Asp and Glu could be induced by activation of nociceptive pathways, it is also possible that activation of antinociceptive pathways is responsible for these increases. The present study also demonstrates that SP infusion increases the ECF concentrations of Gly and Tau (Fig. l), a phenomenon not seen after formalin injection. The release of these inhibitory amino acids during infusion of SP may reflect a more extensive activation of SP receptor populations than would occur following endogenous SP release from primary afferents due to formalin injection. SP is found in intrinsic neurons of the spinal cord, descending fibers, as well as primary afferent fibers where it is co-localized with CGRP [26]. The addition of CGRP to the SP infusion doubled the apparent release of Tau while not affecting any of the other amino acids measured (Fig. 1). These results demonstrate a potentiative interaction between SP and CGRP, a neuroactive peptide that has been demonstrated to be exclusively localized in primary afferent fibers in the dorsal horn of the spinal cord. Based on the discrete localization of CGRP in this area of the spinal cord, this unique and selective increase in
Tau may reflect an SP-mediated response directly associated with primary afferent activity. The ECF concentration of Asn, which is not believed to be a neurotransmitter and which we have previously shown to be insensitive to veratridine or noxious stimulation induced by formalin injection in the hind paw in the rat spinal cord [24], was unaffected during infusion of SP (Fig. 1). This confirms the selectivity of the action of SP on amino acid release and supports the interpretation that changes in microdialysate concentrations reflect release of neurotransmitters in this model. In addition to an apparent release of select amino acids, the infusion of SP in the dorsal spinal cord evokes caudally directed biting and licking, a behavior similar in appearance to the biting and scratching previously reported to be caused by the intrathecal injection of SP or of excitatory amino acid agonists in rats [23] and mice [lo]. In the present study the behavioral response to infusion of SP lasted for less than 10 min while the increased ECF concentrations of Glu and Asp were maintained throughout the infusion period (Fig. 2). In a previous study, animals injected with formalin also exhibited only a transient behavioral response to this injection [24]. The discrepancy between the time-course of the behavioral response and the elevations of Glu and Asp after SP or formalin suggests receptor desensitization or release of a transmitter that inhibits this behavior. We have recently demonstrated a decreased behavioral response to repeated intrathecal injections of SP in mice [13]. The mechanism of this apparent desensitization might involve the accumulation of N-terminal metabolic fragments of SP as they have been reported to cause analgesia in the mouse [25]. Alternatively, the present data showing that both Tau and Gly are elevated in response to infusion of SP, are consistent with the possibility that release of these putative inhibitory transmitters may play a role in the modulation of SP-induced behaviors. To test the possibility that increased Gly and Tau are able to inhibit SP- or EAA-induced activity we used the biting and scratching assay in mice. The complete inhibition of SP-induced be-
99
haviors by it. injection of relatively low doses of Tau and the blockade of this inhibition by the specific Tau antagonist TAG (Fig. 4) indicates that Tau may be involved in the modulation of the activity of SP in the spinal cord. At 8 times the lowest dose of Tau that completely blocked SP activity, Gly inhibited SP-induced behaviors by less than 50%. Further examination of Tau revealed a high degree of specificity for inhibition of SP as Tau failed to inhibit EAA-induced behaviors (Fig. 3A), which are qualitatively similar to those produced by SP. These results further support the hypothesis that the release of Tau, and to a lesser extent Gly, is involved in mediating the inhibition of SP-induced behaviors. The inhibition of acetic acid-induced writhing (Fig. 5) and the lack of effect on thermal tests of nociception by Tau suggest a possible role for Tau in antinociception that is modality specific, but clarification of this role requires further investigation. In contrast to our current results showing inhibitory effects of a very low dose (12 nmol) of Tau on SP-induced behaviors, we have previously demonstrated that an i.t. injection of 10 times (120 nmol) this dose of Tau by itself elicits biting, scratching and writhing-like behavior [14]. Thus the potentiative interaction between SP and CGRP on the release of Tau (Fig. 1) may play a role in either the inhibition or potentiation of SP-induced behavior. Wiesenfeld-Hallin et al. [27] have reported an enhancement of the behavioral response to intrathecal SP by CGRP. As CGRP is a potent inhibitor of an endopeptidase involved in SP degradation [15], the potentiation of SP-induced behaviors by CGRP may be caused by either an inhibition of SP metabolism or a large CGRPmediated increase in the release of Tau. It has been hypothesized that Tau is a neuromodulator, released postsynaptically, involved in regulating postsynaptic calcium fluxes by inhibiting neuronal uptake of this cation [16]. The action of SP has recently been shown to involve an increase in intracellular calcium in isolated rat dorsal horn neurons [28], while CGRP has been shown to enhance the inward calcium current in dorsal root ganglion cells [19] and to increase the duration of calcium-dependent action potentials in dorsal horn neurons [20]. Thus the apparent
increase in Tau release in response to the infusion of SP (Figs. 1 and 2) and the specific potentiation of this release by CGRP (Fig. 1) may result from these effects of SP and CGRP on calcium flux. The inhibition of SP-induced behavior by Tau may then be postulated to result from Tau’s role in calcium homeostasis. Intrathecal injection of calcium has recently been found to be antinociceptive in mice [18], further indicating the potential importance of an interaction between Tau and this cation in regulating nociception. While APV inhibits NMDA-induced behavior in mice [2], our results indicate that it has no effect on SP-induced behavior. APV appears to have a similar spectrum of activity on SP- and EAA-induced behaviors in rats [4] as those observed in mice in the present study. Similarly, GAMS inhibits behavior induced by NMDA and KA but has no effect on SP-induced behavior (Fig. 3B). Interestingly, the lack of effect of GAMS on Quis-induced behaviors is in contrast to its previously shown order of antagonism of iontophoretically applied EAA agonists in the cat spinal cord [6]. These results, together with the differential inhibition of SP- and EAA-induced behaviors by Tau, suggest that, although similar in appearance, the behaviors produced by SP and EAAs are not mediated by a common NMDA or KA receptor. While it is unclear whether biting and scratching behavior correlates with pain perception, this assay is useful for quantifying transmitter activity. Given the evidence supporting a role for SP in pain transmission, the failure of APV and GAMS to inhibit SP-induced behaviors is nonetheless consistent with the poor or non-existent analgesic activity of the EAA antagonists [2,5]. These results also imply that the enhanced concentrations of EAAs in the ECF observed after SP infusion and after formalin injection [24] do not appear to be involved in SP-mediated nociception. This does not rule out the involvement of EAAs in an endogenous analgesic system triggered by SP. In conclusion, these results show that SP causes a significant increase in the ECF concentrations of Asp and Glu, as seen during formalin-induced nociception, as well as in Gly and Tau in the rat dorsal spinal cord, and that the addition of CGRP
100
further potentiates this effect for Tau. The apparent increase in Tau release in response to SP infusion together with the selective inhibitory effect of Tau on the behavioral responses to i.t. SP and i.p. acetic acid in mice suggests a novel role for Tau in SP activity and nociception in the spinal cord. The use and care of animals in this study were performed in accordance with the guidelines of the Minnesota Animal Care and Use Committee and those prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council (DHEW Publication No. (NIH) 78-23. revised 1978).
Acknowledgements The authors acknowledge the competent assistance of MS Cassandra Schamber in the performance of these experiments and would like to thank Dr. J.G. Atkinson from Merck Frosst Laboratories, Canada for generously supplying TAG. This research was supported by United States Public Health Service Grants DA04090, DA04190 and DA00124 to A.A.L., CA01342 to S.R.S. and DA07234 to D.H.S.
References Aanonsen, L.M. and Wilcox, G.L., Phencyclidine selectively blocks a spinal action of N-methyl-o-aspartate in mice, Neurosci. Lett., 67 (1986) 191-197. Aanonsen, L.M. and Wilcox, G.L., Nociceptive action of excitatory amino acids in the mouse: effects of spinally administered opioids, phencyclidine and sigma agonists, J. Pharmacol. Exp. Ther., 243 (1987) 9-19. Besson, J. and Chaouch, A., Peripheral and spinal mechanisms of nociception, Physiol. Rev., 67 (1987) 67-180. Bossut, D., Fret& H. and Mayer, D.J., Is substance P a primary afferent neurotransmitter for nociceptive input? IV. 2-Amino-5-phosphonovalerate (APV) and [D-PrO’,DTrp7,9]-substance P exert different effects on behaviors induced by intrathecal substance P, strychnine~%nd kainic acid, Brain Res., 455 (1988) 247-253. Cahusac, P.M.B., Evans, R.H., Hill, R.G., Rodriquez, R.E. and Smith, D.A.S., The behavioral effects of an N-methyl-
9
10
11
12
13
14
15
16
17
1x
19
20
aspartate receptor antagonist following application to the lumbar spinal cord of conscious rats, Neuropharmacology, 23 (1984) 7199724. Davies, J. and Watkins. J.C., Depressant actions of y-1). glutamylaminomethyl sulfonate (GAMS) on amino acid-induced and synaptic excitation in the cat spinal cord. Brain Res., 327 (1985) 113-120. De Biasi, S. and Rustioni, A., Glutamate and substance P coexist in primary afferent terminals in the superficial laminae of spinal cord. Proc. Nat. Acad. Sci. (U.S.A.). 85 (1988) 7820-7824. Hayashi, G. and Takemori, A.E., The type of analgesia-receptor interaction involved in certain analgesic assays. Eur. J. Pharmacol., 16 (1971) 63-66. Hylden, J.L.K. and Wilcox, G.L., Intrathecal morphine in mice: a new technique, Eur. J. Pharmacol., 67 (1980) 313. 316. Hylden, J.L.K. and Wilcox, G.L., Intrathecal substance P elicits a caudally directed biting and scratching behavior in mice, Brain Res., 217 (1981) 212-215. Kawagoe, R.. Onodera. K. and Takeuchi, A., Effects of black widow spider venom, substance P and bradykinin on the release of endogenous glutamate from the frog spinal cord, Biomed. Res., 6 (1985) 2477252. Kawagoe. R., Onodera. K. and Takeuchi, A.. The release of endogenous glutamate from the newborn rat spinal cord induced by dorsal root stimulation and substance P. Biomed. Res., 7 (1986) 253-259. Larson, A.A., Desensitization to intrathecal substance P in mice: possible involvement of opioids, Pain, 32 (1988) 367-374. Larson, A.A., Intrathecal GABA, glycine, taurine or betaalanine elicits dyskinetic movements in mice, Pharmacol. Biochem. Behav., 32 (1989) 505509. Le Greves, P.. Nyberg, F.. Terenius, L. and Hokfelt, T., Calcitonin gene-related peptide is a potent inhibitor of substance P degradation. Eur. J. Pharmacol.. 115 (1985) 3099311. Lehmann, A.. Hagberg, H. and Hamberg, A., A role for taurine in the maintenance of homeostasis in the central nervous system during hyperexcitation, Neurosci. Lett.. 52 (1984) 347-350. Lembeck, F., Zur Frage der zentralen Ubertragung afferenter Impulse. III. Mitteilung das Vorkommen und die Bedeutung der Substanz P in den dorsalen Wurzeln des Riickenmarks, Naunyn-Schmiedeberg’s Arch. Exp. Pathol. Pharmakol., 219 (1953) 197-213. Lux, F.. Welch, S.P., Brase. D.A. and Dewey, W.L.. Interaction of morphine with intrathecally administered calcium and calcium antagonists: evidence for supraspinal endogenous opioid mediation of intrathecal calcium-induced antinociception in mice, J. Pharmacol. Exp. Ther.. 246 (1988) 500-507. Ryu, P.D., Murase, K., Gerber, G. and Randic, M.. Actions of calcitonin gene-related peptide on rat sensory ganglion neurons, Physiol. Bohemoslov., 37 (1988) 259-265. Ryu, P.D., Gerber, G., Murase. K. and Randic, M., Actions
101
21
22
23
24
25
of calcitonin gene-related peptide on rat dorsal horn neurons, Brain Res., 441 (1988) 357-361. Salt, T.E. and Hill, R.G., Neurotransmitter candidates of somatosensory primary afferent fibers, Neuroscience, 10 (1983) 1083-1103. Schneider, S.P. and Perl, E.R., Selective excitation of neurons in the mammalian spinal dorsal horn by aspartate and glutamate in vitro: correlation with location and excitatory input, Brain Res., 360 (1985) 339-343. Seybold, V.S., Hylden, J.L.K. and Wilcox, G.L., lntrathecal substance P and somatostatin in rats: behavior indicative of sensation, Peptides, 3 (1982) 49-54. Skilling, S.R., Smullin, D.H., Beitz, A.J. and Larson, A.A., Extracellular amino acid concentrations in the dorsal spinal cord of freely moving rats following veratridine and nociceptive stimulation, J. Neurochem., 51 (1988) 127-132. Stewart, J.M., Hall, M.E., Harkins, J., Frederickson, CA.,
Terenius, L., Hiikfelt, T. and Krivoy, W.A., A fragment of substance P with specific central activity: SP (l-7), Peptides, 3 (1982) 851-857. 26 Tuchscherer, M.M. and Seybold, V.S., A quantitative study of the coexistence of peptides and varicosities within the superficial laminae of the dorsal horn of the rat spinal cord, J. Neurosci., 9 (1989) 195-205. 27 Wiesenfeld-Hallin, Z., Hokfelt, T., Lundberg, J.M., Forssman, W.G., Reinecke, M., Tschopp, F.A. and Fischer, J.A., Immunoreactive calcitonin gene-related peptide and substance P coexist in sensory neurons to the spinal cord and interact in spinal behavioral responses of the rat, Neurosci. Lett., 52 (1984) 199-204. 28 Womack, M.D., MacDermott, A.B. and Jessell, T.M., Sensory transmitters regulate intracellular calcium in dorsal horn neurons, Nature, 334 (1988) 351-353.
Pain, 42 (1990) 93-101 Elsevier
PAIN 01597
Interactions between substance P, calcitonin gene-related peptide, taurine and excitatory amino acids in the spinal cord David
H. Smullin,
Department (Received
of Veterinay
Stephen
R. Skilling
and Alice
Biology, Unioersity of Minnesota,
26 July 1989, revision received
8 November
A. Larson
St. Paul, MN 55108 (U.S.A.)
1989, accepted
10 January
1990)
Using in viva microdialysis in the dorsal spinal cord of the rat, we have previously observed increases in glutamate SUmmarY and aspartate during exposure to a noxious stimulus. The present investigation was designed to determine whether these increases may be mediated by substance P. Infusion of 1 mM of substance P in the dialysis fluid increased the concentrations of glutamate and aspartate, similar to the response seen during noxious stimulation. In addition, substance P also increased the concentrations of the inhibitory amino acids glycine and taurine. Calcitonin gene-related peptide, previously shown to enhance substance P-induced biting and scratching behavior, produced no effect on amino acid release by itself but potentiated the apparent release of taurine by substance P. To assess the importance of substance P-induced amino acid release in sensory processing, we examined the influence of taurine and of excitatory amino acid antagonists on the biting and scratching behavior produced by excitatory amino acids and substance P. Taurine selectively inhibited only substance P-induced biting and scratching while excitatory ammo acid antagonists inhibited only excitatory amino acid-induced behavior. To further explore the ability of taurine to inhibit the substance P-induced behavior, 3 tests of nociception were then used. Pretreatment with taurine inhibited the nociceptive-related writhing behavior produced by an intraperitoneal injection of acetic acid in mice but failed to alter the latency of response in the hot plate or tail flick assay. These results indicate that substance P release during nociception may mediate the release of glutamate and aspartate but that these excitatory amino acids do not appear to be involved in substance P-induced biting and scratching behavior. The selective antinociceptive effect of taurine on chemical, but not thermal, tests of nociception suggests that its release in response to substance P and enhanced release following calcitonin gene-related peptide may be important in an endogenously evoked analgesia.
Key words:
Nociception;
Substance
P; Excitatory
ammo
acids; Taurine;
Introduction The excitatory amino acids glutamate (Glu) and aspartate (Asp) and the peptide substance P (SP) have been proposed as primary afferent neurotransmitters in the spinal cord involved in nociception [3,17,21,22]. We have recently demonstrated increased concentrations of Asp and Glu
Correspondence Veterinary Biology, Building, University MN 55iO8, U.S.A.-
0304-3959/90/$03.50
to: Dr. D.H. Smullin, Department of 295 Animal Science/Veterinary Medicine of Minnesota, 1988 Fitch Ave.. St. Paul.
0 1990 Elsevier Science Publishers
Calcitonin
gene-related
peptide;
Microdialysis
in the extracellular fluid of the rat dorsal lumbar spinal cord after intradermal injection of formalin [24]. While the increase in Glu and Asp may reflect primary afferent release it is also possible that this increase reflects release from descending or spinal interneurons in response to a primary release of SP, calcitonin gene-related peptide (CGRP) or other neurotransmitter. In support of this latter conclusion it has been shown that the application of SP to the hemisected frog and rat spinal cord caused the release of Glu into the superfusate [11,12]. In addition, the intrathecal injection of excitatory amino acid (EAA) agonists,
B.V. (Biomedical
Division)
such as ~-methyl-~-aspartate (NMDA), kainic acid (KA) and quisqualate (Quis) elicit behavioral responses similar to that produced by SP [l,lO]. In addition to SP, a variety of other neuropeptides have been found in primary afferent fibers. CGRP, for example, has been associated with pain transmission and is uniquely distributed in rat dorsal horn sensory fibers of laminae I and II [26] where it has been immunocytochemically co-localized with SP [27]. While the intrathec~ administration of CGRP alone produces no behavioral effects, the injection of SP plus CGRP both potentiates and prolongs the caudally directed biting and scratching behavior produced by SP alone [27]. The purpose of the present investigation was to determine if the release of either Glu or Asp that was observed during noxious stimulation is from primary afferent fibers or if this release is mediated through an initial release of SP from primary afferents. Apparent release was determined by changes in the concentrations of putative amino acid transmitters in the extracellular fluid of rat dorsal spinal cords using in vivo microdialysis. The importance of SP-induced changes in amino acid concentrations was then examined using nocifensive responses to acetic acid injected into the peritoneal cavity and motor responses that involve caudally directed biting and scratching in response to intrathecally injected excitatory compounds.
Methods and Materials Microdia Eysis
Male Sprague-Dawley rats (275-300 g) were anesthetized with sodium pentobartial(50 mg/kg, i.p., Veterinary Laboratories, Inc.) and implanted with a microdialysis fiber (200 pm diam., 50,000 MW cut-off, Amicon Vitafiber II) through the dorsal spinal cord at the level of L5/L6 as described previously [24]. Fo~owing a 24 h recovery period, the animals were evaluated for any signs of limb paralysis or impaired movement. No impairment was observed in any of the animals in this study, and all animals appeared to behave normally. The dialysis system
was attached to a peristaltic pump (Gilson Minipuls 3) and perfused with Ringer’s solution (147 mM NaCl; 4.0 mM KCl; 2.2 mM CaCI,) at 5-6 pl/min for 90 min to establish a diffusion equilibrium. Samples were collected at 10 min intervals in polypropylene tubes and maintained at 5°C until analyzed for amino acids, within 12 h, by HPLC as described by Skilling et al. [24]. Animals were maintained at 22* C with 12 h light/dark cycles and were provided with food and water ad libitum. Following each experiment, the animals were killed and the cannulae injected with methylene blue dye. The spinal cord was then removed and fixed for gross histological confirmation of cannula ptacement. Only animals with cannulas located below lamina I and above the central canal were included in this study. In addition, animals were not included in this study if evidence of dye leakage into the cerebral spinal fluid was present. Three control samples were collected for determination of basal concentrations of amino acids. Ad~nistration of SP (Bachem), CGRP (Cambridge Research Biochemicals) or SP plus CGRP was made by infusion of 0.002-1.0 mM of SP, lo-’ 7-10p5 M of CGRP or 1.0 mM of SP plus IO--’ M CGRP in Ringer’s solution for 40-60 min. Samples were collected throughout the course of the experiments described above. The data were transformed into percent of basal concentration of extracellular fluid for each animal based on in vitro calibration experiments. The in vitro dialysis efficiency was between 1 and 3% for all of the amino acids measured. in vitro measurements. using known concentrations of peptide standards at 37 ’ C, have shown a linear relationship between SP concentration in the standards and recovery of SP in the dialysis solution as measured by HPLC. Assuming that the diffusion of SP across the dialysis membrane is approximately equal in both directions, 1.0 mM SP in the perfusate would result in an estimated maximum diffusion of only 1.0 nmole of SP into the extracellular fluid of the spinal cord during each 10 min period. The mean percent increase of the first 3 samples following treatment was transformed into log vaIues and then analyzed across treatments using analysis of
95
variance (ANOVA) and ScheffC’s F test for comparison of individual means. Intrathecal injections Male Swiss-Webster mice weighing 15-20 g (Biolab, White Bear Lake, MN) were used in all experiments in which intrathecal (i.t.) injections were made. All i.t. injections were made in unanesthetized mice at approximately the L5/L6 intervertebral space using a 30-gauge 0.5 inch needle on a 50 ~1 luer tip Hamilton syringe after the method of Hylden and Wilcox [9]. Substance P (Peninsula; 3.7 X lop3 nmol), NMDA (Sigma; 0.2 nmol), kainic acid (KA, Sigma; 0.075 nmol), Quis (Sigma; 1.0 nmol), taurine (Tau, Sigma; 0.15-12.0 nmol), Gly (Sigma; 12.0 nmol), 6-aminoethyl-3-methy1-4H-1,2,4-benzothiadiazine-l,l-dioxide (TAG, Merck Frosst, D-2-amino-5-phosphonoCanada; 4.4 nmol), valerate (APV, Sigma; 0.25-5.0 nmol) and yglutamylaminomethylsulfonic acid (GAMS, Sigma; 50.0 nmol) were administered in a volume of 5 ~1 of 0.85% saline containing 0.01 N acetic acid. Mice were placed in a plexiglass container immediately after the injection and the number of bites and scratches were counted over a 60 set period. Data were analyzed using Student’s 2-tailed t test for unpaired data or by ANOVA with Scheffe’s F test for comparison of individual means. Writhing assay Abdominal constrictions were produced in mice by the intraperitoneal (i.p.) injection of 0.3 ml of a 1% solution of acetic acid as described previously by Hayashi and Takemori [8]. The number of behaviors was counted over a 5 min interval beginning 5 min after the injection of acetic acid. Animals were killed immediately after the termination of each recording period. Data were analyzed using ANOVA with Scheffss F test for comparison of individual means. Hot plate assay In the hot plate test, mice were placed on a plate heated to 56°C and the time required for the animal to lick its front paw was recorded, at which time the animals were immediately removed
from the hot plate. If an animal showed no reaction in 30 set it was removed from the hot plate. Data were analyzed using Student’s 2-tailed t test for unpaired data. Tail flick assay The time required for a mouse to move its tail when exposed to a radiant heat source was recorded. If an animal showed no reaction in 10 set the experiment was terminated. Data were analyzed using Student’s 2-tailed t-test for unpaired data.
Results Microdialysis experiments in rats The means of 3 samples collected before treatment were used for determination of amino acid concentrations in the dorsal spinal cord extracellular fluid (ECF). Basal concentrations k S.E.M. (PM) were: Asp 3.53 & 0.25, Glu 7.45 f 2.02, Asn 7.65 + 1.46, Gly 47.87 f 11.24 and Tau 10.67 + 1.71. These values represent dialysate concentrations corrected for cannula efficiency, based on in vitro calibration. The extracellular fluid concentrations of Asp, Glu, Gly and Tau increased to 150, 150, 125 and 170% (P < 0.05) of basal concentrations respectively (Fig. 1) during a 60 min perfusion with 1.0 mM of SP. The maximal increase in concentrations of these amino acids occurred within 20 min of SP infusion. In experiments using a 40 min perfusion with 1.0 mM of SP, the concentrations of amino acids increased followed by a return to basal levels within 10 min of removal of SP from the perfusate (Fig. 2). Doses of SP lower than 1.0 mM had no significant effect on the concentrations of any of the amino acids measured. The concentration of the amino acid asparagine (Asn) did not change in response to any concentration of SP (Fig. 1). Infusion of 10-7-10-5 M CGRP by itself had no significant effect on the extracellular fluid concentrations of any of these amino acids (Fig. 1). Perfusion with 1.0 mM of SP plus 10m5 M CGRP increased Tau concentrations in the ECF 3-fold (P < 0.05), significantly greater than the increase produced by SP alone (P < 0.05) but did not alter
96 400
peptides in the spinal cord and lasted less than the first 10 min of the perfusion period. Although not quantified, no obvious differences were observed in either the intensity or duration between the behaviors produced by SP alone or SP with CGRP.
.-6 E E z
300
IJ
Ringer’s
q n
3 SP SP
El
6 0
t CGRP
Intruthecal experiments in mice Intrathecal injection of mice with 3.7 x 10 ' nmol of SP, 0.2 nmol of NMDA, 0.075 nmol of KA or 1.0 nmol of Quis resulted in caudally
z E
100
z & p. 0 Asn
GIU
TaU
GUY
Amino Acid Fig. 1. Effects of CGRP (1O-‘5 M). SP (1.0 mM) and SP+ CGRP on the concentration of extracellular amino acids in viva from the dorsal spinal cord (n = 4-5 each). Basal concentrations were determined as the mean values from 3 consecutive fractions before treatment with peptides and subsequent samples were expressed as the percent of basal concentration For statistical analysis the data were further transformed to log values. The rats were perfused with peptide for 60 mm and the results represent the mean*S.E.M. (bars) of the first 3 samples following treatment. Ten minute fractions were collected at a flow rate of 5-6 &mm. A single asterisk indicates a difference of P -C0.05 by ANOVA with Scheffe’s F test for comparison of individual means between treatment and control (Ringer’s). Two asterisks indicates a difference of P -Z0.05 by ANOVA with Scheffe’s F test for comparison of individual means between SP and SP + CGRP.
A
SP
NMDA
the response to SP for the other amino acids measured (Fig. 1). Intermittent caudally directed biting and licking behavior was observed in most animals after receiving either SP alone or SP with CGRP. This behavior, when present, began at a time that corresponded with the arrival of the
m : co z
Asp
Avg
---a---
Avg Gly Avg Tau
300 200
-
100 -
E s z a
0 -60
~30
0
30
60
90
120
I 150
Time (min)
Fig. 2. Time course of the effect of SP (1.0 mM) on the extracellular concentrations of ammo acids in vivo from the rat dorsal spinal cord (n = 3). Treatment with SP was begun at 0 min and lasted for 40 min. Details of experiment as in Fig. 1.
7
5 SP
-
+ 1liil KA
NMDA
KA
6
6
OUIS
Fig. 3. A: the effects of co-administration of Tau (12.0 mmol; striped bars) on the caudally directed biting and scratching behavior elicited by intrathecal injections of SP (3.7X 10e3 nmol), NMDA (0.2 nmol), KA (0.075 nmol) and Quis (1.0 nmol) (solid white bars) in mice. B: the effects of co-administration of GAMS (50.0 nmol; striped bars) on the caudally directed biting and scratching behavior elicited by intrathecal injections of SP (3.7x10-s nmol), NMDA (0.2 nmol), KA (0.075 nmol) and Quis (1.0 nmol) (solid white bars) in mice. Drugs were administered in 5 pl of 0.85% saline containing 0.01 N acetic acid and the number of behaviors were counted over a 60 set period. Asterisk indicates a difference of P < 0.01 between injection of peptide and of peptide plus Tau by unpaired t test. The number of mice in each group is indicated by the number at the base of each bar.
97
directed biting and scratching behavior (Fig. 3) as previously reported [l]. The co-administration of 12.0 nmol of Tau with either NMDA, KA or Quis had no effect on the number of behaviors elicited by these EAA agonists, however, at this dose Tau totally inhibited SP-induced biting and scratching (Fig. 3A). This total blockade of SP-induced behaviors by Tau was found at concentrations as low as 1.5 nmol. In contrast, co-administration of 12.0 nmol of Gly elicited only 50% inhibition of the SP-induced behaviors. The i~bito~ effect of 12.0 nmol of Tau on SP-induced behavior was blocked by co-administration of TAG, a Tau antagonist, at a dose that had no effect when injected alone (4.4 nmol) or in combination with SP (Fig. 4). Intrathecal injection of 50.0 nmol of GAMS, reportedly a KA and Quis ~tago~st with less ability to inhibit NMDA-induced excitation [6], did not affect the SP- or Quis-induced behaviors, but did inhibit the behavior caused by NMDA and KA (Fig. 3B). Similarly, co-administration of 5.0 nmol of APV, an NMDA specific antagonist, with SP had no effect on the SP-induced behavior, while 1.0 nmol was sufficient to totally inhibit NMDA-induced behavior of similar intensity (data not shown). Neither Tau nor Gly produced behavioral effects when injected alone at 12.0 nmol
I_
6
6 SP+Tau
SPtTau+TAG
SP+TAG
Treatment
Fig. 4. The effects of co-administration of Tau (12.0 nmol), TAG (4.4 rnnol) and of Tau (12.0 nmol) plus TAG (4.4 nmol) on the caudally directed biting and scratching behavior elicited by intrathecal injections of SP (3.7~ 10-s nmol) in mice. Details of experiments as in Fig. 3. Asterisk indicates a difference of Pi 0.05by ANOVA with Scheffe’s F test for comparison of individual means between SP+Tau and SP, SP + Tau + TAG and SP + TAG.
I_ t
cl_ 6
Tall
6
Tau+TAG
7 TAG
Treatment
Fig. 5. The effects of pretreatment with saline, Tau (12.0 nmol), TAG (4.4 nmol) and of Tau (12.0 nmol) plus TAG (4.4 nmol) on the writhing behavior induced by an i.p. injection of acetic acid. Writhing was produced in mice by the i.p. injection of 0.3 ml of a 1% solution of acetic acid as described previously by Hayashi and Takemori [8]. The number of writhes were counted over a 5 mm interval beginning 5 min after the injection of acetic acid. Drugs were administered intrathecally in 5 ~1 of 0.85% saline containing 0.01 N acetic acid 2 min prior to the i.p. injection of acetic acid. Asterisk indicates a difference of P < 0.05by ANOVA with Scheff& F test for comparison of individual means between Tau and Control, Tau+TAG and TAG. The number of mice in each group is indicated by the number at the base of each bar.
during the 60 set recording period but doses higher than 120.0 nmol and 200.0 nmol, respectively, elicited biting and scratching in addition to dyskinetic writhing-like movements as reported previously [14]. In order to ascertain whether Tau has an effect on behavioral tests of nociception used historically to determine whether a compound is analgesic, mice were pretreated with an it. injection of 12.0 nmol of Tau 2 min prior to an i.p. injection of acetic acid, or 1 and 5 min prior to either a hot plate or tail flick test. Taurine had no effect on either the hot plate or tail flick assays (data not shown) while the writhing behavior induced by acetic acid was inhibited by Tau (Fig. 5). This inhibition was reversed by TAG (4.4 nmol) which had no effect by itself (Fig. 5).
The present study, demonstrating a dramatic increase in the concentrations of Asp and Glu in
the ECF in response to infusion of SP (Fig. l), shows changes in EAAs that are qualitatively similar to those changes seen after formalin injection in the rat hind paw [24]. These increases are believed to reflect an increased release of these compounds from neurotransmitter pools into the ECF. The ability of infused SP to elicit this response suggests that the apparent increase in Asp and Glu release during chronic pain may be mediated, at least in part, by an initial release of SP from primary afferent fibres where SP is believed to act as a nociceptive transmitter [17,21]. The ability of SP alone to mimic the noxious stimulusinduced increases in Glu and Asp may be due to a second-order release of EAAs from interneurons in response to the release of SP from primary afferent fibers. The release of Asp and Glu during pain and SP infusion may also be due to presynaptic facilitation of the release of these EAAs by SP. The latter possibility is supported by recent findings that Glu and SP coexist in primary afferent terminals in the rat spinal cord [7]. Although the formalinand SP-induced increases in Asp and Glu could be induced by activation of nociceptive pathways, it is also possible that activation of antinociceptive pathways is responsible for these increases. The present study also demonstrates that SP infusion increases the ECF concentrations of Gly and Tau (Fig. l), a phenomenon not seen after formalin injection. The release of these inhibitory amino acids during infusion of SP may reflect a more extensive activation of SP receptor populations than would occur following endogenous SP release from primary afferents due to formalin injection. SP is found in intrinsic neurons of the spinal cord, descending fibers, as well as primary afferent fibers where it is co-localized with CGRP [26]. The addition of CGRP to the SP infusion doubled the apparent release of Tau while not affecting any of the other amino acids measured (Fig. 1). These results demonstrate a potentiative interaction between SP and CGRP, a neuroactive peptide that has been demonstrated to be exclusively localized in primary afferent fibers in the dorsal horn of the spinal cord. Based on the discrete localization of CGRP in this area of the spinal cord, this unique and selective increase in
Tau may reflect an SP-mediated response directly associated with primary afferent activity. The ECF concentration of Asn, which is not believed to be a neurotransmitter and which we have previously shown to be insensitive to veratridine or noxious stimulation induced by formalin injection in the hind paw in the rat spinal cord [24], was unaffected during infusion of SP (Fig. 1). This confirms the selectivity of the action of SP on amino acid release and supports the interpretation that changes in microdialysate concentrations reflect release of neurotransmitters in this model. In addition to an apparent release of select amino acids, the infusion of SP in the dorsal spinal cord evokes caudally directed biting and licking, a behavior similar in appearance to the biting and scratching previously reported to be caused by the intrathecal injection of SP or of excitatory amino acid agonists in rats [23] and mice [lo]. In the present study the behavioral response to infusion of SP lasted for less than 10 min while the increased ECF concentrations of Glu and Asp were maintained throughout the infusion period (Fig. 2). In a previous study, animals injected with formalin also exhibited only a transient behavioral response to this injection [24]. The discrepancy between the time-course of the behavioral response and the elevations of Glu and Asp after SP or formalin suggests receptor desensitization or release of a transmitter that inhibits this behavior. We have recently demonstrated a decreased behavioral response to repeated intrathecal injections of SP in mice [13]. The mechanism of this apparent desensitization might involve the accumulation of N-terminal metabolic fragments of SP as they have been reported to cause analgesia in the mouse [25]. Alternatively, the present data showing that both Tau and Gly are elevated in response to infusion of SP, are consistent with the possibility that release of these putative inhibitory transmitters may play a role in the modulation of SP-induced behaviors. To test the possibility that increased Gly and Tau are able to inhibit SP- or EAA-induced activity we used the biting and scratching assay in mice. The complete inhibition of SP-induced be-
99
haviors by it. injection of relatively low doses of Tau and the blockade of this inhibition by the specific Tau antagonist TAG (Fig. 4) indicates that Tau may be involved in the modulation of the activity of SP in the spinal cord. At 8 times the lowest dose of Tau that completely blocked SP activity, Gly inhibited SP-induced behaviors by less than 50%. Further examination of Tau revealed a high degree of specificity for inhibition of SP as Tau failed to inhibit EAA-induced behaviors (Fig. 3A), which are qualitatively similar to those produced by SP. These results further support the hypothesis that the release of Tau, and to a lesser extent Gly, is involved in mediating the inhibition of SP-induced behaviors. The inhibition of acetic acid-induced writhing (Fig. 5) and the lack of effect on thermal tests of nociception by Tau suggest a possible role for Tau in antinociception that is modality specific, but clarification of this role requires further investigation. In contrast to our current results showing inhibitory effects of a very low dose (12 nmol) of Tau on SP-induced behaviors, we have previously demonstrated that an i.t. injection of 10 times (120 nmol) this dose of Tau by itself elicits biting, scratching and writhing-like behavior [14]. Thus the potentiative interaction between SP and CGRP on the release of Tau (Fig. 1) may play a role in either the inhibition or potentiation of SP-induced behavior. Wiesenfeld-Hallin et al. [27] have reported an enhancement of the behavioral response to intrathecal SP by CGRP. As CGRP is a potent inhibitor of an endopeptidase involved in SP degradation [15], the potentiation of SP-induced behaviors by CGRP may be caused by either an inhibition of SP metabolism or a large CGRPmediated increase in the release of Tau. It has been hypothesized that Tau is a neuromodulator, released postsynaptically, involved in regulating postsynaptic calcium fluxes by inhibiting neuronal uptake of this cation [16]. The action of SP has recently been shown to involve an increase in intracellular calcium in isolated rat dorsal horn neurons [28], while CGRP has been shown to enhance the inward calcium current in dorsal root ganglion cells [19] and to increase the duration of calcium-dependent action potentials in dorsal horn neurons [20]. Thus the apparent
increase in Tau release in response to the infusion of SP (Figs. 1 and 2) and the specific potentiation of this release by CGRP (Fig. 1) may result from these effects of SP and CGRP on calcium flux. The inhibition of SP-induced behavior by Tau may then be postulated to result from Tau’s role in calcium homeostasis. Intrathecal injection of calcium has recently been found to be antinociceptive in mice [18], further indicating the potential importance of an interaction between Tau and this cation in regulating nociception. While APV inhibits NMDA-induced behavior in mice [2], our results indicate that it has no effect on SP-induced behavior. APV appears to have a similar spectrum of activity on SP- and EAA-induced behaviors in rats [4] as those observed in mice in the present study. Similarly, GAMS inhibits behavior induced by NMDA and KA but has no effect on SP-induced behavior (Fig. 3B). Interestingly, the lack of effect of GAMS on Quis-induced behaviors is in contrast to its previously shown order of antagonism of iontophoretically applied EAA agonists in the cat spinal cord [6]. These results, together with the differential inhibition of SP- and EAA-induced behaviors by Tau, suggest that, although similar in appearance, the behaviors produced by SP and EAAs are not mediated by a common NMDA or KA receptor. While it is unclear whether biting and scratching behavior correlates with pain perception, this assay is useful for quantifying transmitter activity. Given the evidence supporting a role for SP in pain transmission, the failure of APV and GAMS to inhibit SP-induced behaviors is nonetheless consistent with the poor or non-existent analgesic activity of the EAA antagonists [2,5]. These results also imply that the enhanced concentrations of EAAs in the ECF observed after SP infusion and after formalin injection [24] do not appear to be involved in SP-mediated nociception. This does not rule out the involvement of EAAs in an endogenous analgesic system triggered by SP. In conclusion, these results show that SP causes a significant increase in the ECF concentrations of Asp and Glu, as seen during formalin-induced nociception, as well as in Gly and Tau in the rat dorsal spinal cord, and that the addition of CGRP
100
further potentiates this effect for Tau. The apparent increase in Tau release in response to SP infusion together with the selective inhibitory effect of Tau on the behavioral responses to i.t. SP and i.p. acetic acid in mice suggests a novel role for Tau in SP activity and nociception in the spinal cord. The use and care of animals in this study were performed in accordance with the guidelines of the Minnesota Animal Care and Use Committee and those prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council (DHEW Publication No. (NIH) 78-23. revised 1978).
Acknowledgements The authors acknowledge the competent assistance of MS Cassandra Schamber in the performance of these experiments and would like to thank Dr. J.G. Atkinson from Merck Frosst Laboratories, Canada for generously supplying TAG. This research was supported by United States Public Health Service Grants DA04090, DA04190 and DA00124 to A.A.L., CA01342 to S.R.S. and DA07234 to D.H.S.
References Aanonsen, L.M. and Wilcox, G.L., Phencyclidine selectively blocks a spinal action of N-methyl-o-aspartate in mice, Neurosci. Lett., 67 (1986) 191-197. Aanonsen, L.M. and Wilcox, G.L., Nociceptive action of excitatory amino acids in the mouse: effects of spinally administered opioids, phencyclidine and sigma agonists, J. Pharmacol. Exp. Ther., 243 (1987) 9-19. Besson, J. and Chaouch, A., Peripheral and spinal mechanisms of nociception, Physiol. Rev., 67 (1987) 67-180. Bossut, D., Fret& H. and Mayer, D.J., Is substance P a primary afferent neurotransmitter for nociceptive input? IV. 2-Amino-5-phosphonovalerate (APV) and [D-PrO’,DTrp7,9]-substance P exert different effects on behaviors induced by intrathecal substance P, strychnine~%nd kainic acid, Brain Res., 455 (1988) 247-253. Cahusac, P.M.B., Evans, R.H., Hill, R.G., Rodriquez, R.E. and Smith, D.A.S., The behavioral effects of an N-methyl-
9
10
11
12
13
14
15
16
17
1x
19
20
aspartate receptor antagonist following application to the lumbar spinal cord of conscious rats, Neuropharmacology, 23 (1984) 7199724. Davies, J. and Watkins. J.C., Depressant actions of y-1). glutamylaminomethyl sulfonate (GAMS) on amino acid-induced and synaptic excitation in the cat spinal cord. Brain Res., 327 (1985) 113-120. De Biasi, S. and Rustioni, A., Glutamate and substance P coexist in primary afferent terminals in the superficial laminae of spinal cord. Proc. Nat. Acad. Sci. (U.S.A.). 85 (1988) 7820-7824. Hayashi, G. and Takemori, A.E., The type of analgesia-receptor interaction involved in certain analgesic assays. Eur. J. Pharmacol., 16 (1971) 63-66. Hylden, J.L.K. and Wilcox, G.L., Intrathecal morphine in mice: a new technique, Eur. J. Pharmacol., 67 (1980) 313. 316. Hylden, J.L.K. and Wilcox, G.L., Intrathecal substance P elicits a caudally directed biting and scratching behavior in mice, Brain Res., 217 (1981) 212-215. Kawagoe, R.. Onodera. K. and Takeuchi, A., Effects of black widow spider venom, substance P and bradykinin on the release of endogenous glutamate from the frog spinal cord, Biomed. Res., 6 (1985) 2477252. Kawagoe. R., Onodera. K. and Takeuchi, A.. The release of endogenous glutamate from the newborn rat spinal cord induced by dorsal root stimulation and substance P. Biomed. Res., 7 (1986) 253-259. Larson, A.A., Desensitization to intrathecal substance P in mice: possible involvement of opioids, Pain, 32 (1988) 367-374. Larson, A.A., Intrathecal GABA, glycine, taurine or betaalanine elicits dyskinetic movements in mice, Pharmacol. Biochem. Behav., 32 (1989) 505509. Le Greves, P.. Nyberg, F.. Terenius, L. and Hokfelt, T., Calcitonin gene-related peptide is a potent inhibitor of substance P degradation. Eur. J. Pharmacol.. 115 (1985) 3099311. Lehmann, A.. Hagberg, H. and Hamberg, A., A role for taurine in the maintenance of homeostasis in the central nervous system during hyperexcitation, Neurosci. Lett.. 52 (1984) 347-350. Lembeck, F., Zur Frage der zentralen Ubertragung afferenter Impulse. III. Mitteilung das Vorkommen und die Bedeutung der Substanz P in den dorsalen Wurzeln des Riickenmarks, Naunyn-Schmiedeberg’s Arch. Exp. Pathol. Pharmakol., 219 (1953) 197-213. Lux, F.. Welch, S.P., Brase. D.A. and Dewey, W.L.. Interaction of morphine with intrathecally administered calcium and calcium antagonists: evidence for supraspinal endogenous opioid mediation of intrathecal calcium-induced antinociception in mice, J. Pharmacol. Exp. Ther.. 246 (1988) 500-507. Ryu, P.D., Murase, K., Gerber, G. and Randic, M.. Actions of calcitonin gene-related peptide on rat sensory ganglion neurons, Physiol. Bohemoslov., 37 (1988) 259-265. Ryu, P.D., Gerber, G., Murase. K. and Randic, M., Actions
101
21
22
23
24
25
of calcitonin gene-related peptide on rat dorsal horn neurons, Brain Res., 441 (1988) 357-361. Salt, T.E. and Hill, R.G., Neurotransmitter candidates of somatosensory primary afferent fibers, Neuroscience, 10 (1983) 1083-1103. Schneider, S.P. and Perl, E.R., Selective excitation of neurons in the mammalian spinal dorsal horn by aspartate and glutamate in vitro: correlation with location and excitatory input, Brain Res., 360 (1985) 339-343. Seybold, V.S., Hylden, J.L.K. and Wilcox, G.L., lntrathecal substance P and somatostatin in rats: behavior indicative of sensation, Peptides, 3 (1982) 49-54. Skilling, S.R., Smullin, D.H., Beitz, A.J. and Larson, A.A., Extracellular amino acid concentrations in the dorsal spinal cord of freely moving rats following veratridine and nociceptive stimulation, J. Neurochem., 51 (1988) 127-132. Stewart, J.M., Hall, M.E., Harkins, J., Frederickson, CA.,
Terenius, L., Hiikfelt, T. and Krivoy, W.A., A fragment of substance P with specific central activity: SP (l-7), Peptides, 3 (1982) 851-857. 26 Tuchscherer, M.M. and Seybold, V.S., A quantitative study of the coexistence of peptides and varicosities within the superficial laminae of the dorsal horn of the rat spinal cord, J. Neurosci., 9 (1989) 195-205. 27 Wiesenfeld-Hallin, Z., Hokfelt, T., Lundberg, J.M., Forssman, W.G., Reinecke, M., Tschopp, F.A. and Fischer, J.A., Immunoreactive calcitonin gene-related peptide and substance P coexist in sensory neurons to the spinal cord and interact in spinal behavioral responses of the rat, Neurosci. Lett., 52 (1984) 199-204. 28 Womack, M.D., MacDermott, A.B. and Jessell, T.M., Sensory transmitters regulate intracellular calcium in dorsal horn neurons, Nature, 334 (1988) 351-353.
