- Email: [email protected]
Spinal vs supraspinal actions of morphine on cat spinal cord multireceptive neurons
Brain Research, 273 (1983) 1-7
1
Elsevier
Research Reports
Spinal vs Supraspinal Actions of Morphine on Cat Spinal Cord Multireceptive Neurons PETER J. SOJA and JOHN G. SINCLAIR
Division of Pharmacologyand Toxicology, Faculty of Pharmaceutical Sciences, Universityof British Columbia, Vancouver, BC V6T 114:5(Canada) (Accepted December 28th, 1982)
Key words: morphine - spinal cord - inhibition - naloxone - analgesia
To examine whether morphine elicits a supraspinal mediated spinal inhibition of nociceptive transmission, several investigators have compared the effects of morphine on nociceptive transmission in animals with the spinal cord intact vs transected or cold-blocked. The results have been conflicting, possibly due to different methods of analysis. For example, some investigators have found i.v. administered morphine produces a greater percentage decrease in nociceptive transmission when the spinal cord is intact compared to the transected state. Therefore, they concluded that morphine elicits a supraspinal-mediated inhibition. Conversely, others have reported that the increase in noxious stimulus-evoked responses of dorsal horn neurons upon cold blocking the spinal cord was reduced by i.v. morphine. They therefore concluded that morphine decreases descending inhibition. We tested the effects of i.v. morphine on spinal cord multireceptive neurons in the presence and absence of descending inhibition. Using the above methods of analysis, our results were found to be consistent with their findings which indicate that the method of analysis used is critical to the interpretation reached. To determine how these calculations would be affected by a depressant effect on the spinal cord neurons only, we performed similar experiments iontophoresing "t-aminobutyric acid (GABA) onto these dorsal horn neurons. The similarity between the morphine and GABA data suggests that the effects of systemically administered morphine on multireceptive dorsal horn neurons can be adequately explained by a spinal cord site of action. INTRODUCTION
Although there is good evidence that morphine elicits antinociception through a spinal cord site of a c t i o n 17,19.34, there is also evidence that morphine attenuates spinal cord nociceptive transmission through a supraspinal site of action (see reviews by Fields and Basbaum ~°, Yaksh and Rudy 35, and Gebhartg). The latter proposal was first made by Irwin et al) 4 who found the rat tail-flick test to be less sensitive to morphine in animals with a transected spinal cord as compared to those with an intact spinal cord. Other supporting evidence in rats includes the finding that intracerebral microinjections of morphine prolong the tailflick latency 15 and inhibit spinal cord neurons receiving noxious and inocuous input (multireceptive neurons) 2. In ad0006-8993/83/$03.00 © 1983 Elsevier Science Publishers B.V.
dition, antinociception produced by intracerebral microinjection of morphine is antagonized by dorsolateral funiculi lesions t or intrathecal injections of the narcotic antagonist, naloxone 20 (however, see ref. 33). To examine the question of whether morphine elicits a supraspinal-mediated spinal depression of nociceptive transmission, several investigators have used an approach similar to that first used by Irwin et al) 4. That is, they compared the effect of morphine on spinal cord nociceptive transmission with the spinal cord intact vs transected or cold-blocked. The results have been conflicting. Takagi et al.3 ~and Hanaoka et al) 2 have claimed that morphine increases bulbospinal inhibition impinging on spinal cord multireceptive neurons in the rabbit and cat, respectively. In contrast, Jurna and Grossman ~6
and Duggan et al. s concluded that morphine decreases supraspinal inhibition of cat spinal cord ventrolateral tract axons and dorsal horn neurons. respectively. Moreover, Le Bars et al. ~' reported that dose response curves of morphine's suppression of spinal cord multireceptive neurons from intact and spinal cord sectioned rats are superimposable, suggesting no reinforcement of descending controls. Hence, a controversy exists as to whether morphine depresses spinal cord nociceptive transmission via a supraspinal site of action. Transecting or cold-blocking the spinal cord releases multireceptive dorsal horn neurons of the cat ~ >.3°.-~2and rat 23 from a powerful tonic inhibitory impingement of supraspinal origin. This is reflected by a marked increase in the response of the cell to a noxious stimulus applied to the receptive field following release from the inhibition. This increase in excitability of these dorsal horn neurons complicates the interpretation of experiments such as those performed by Takagi et al. 3~, Hanaoka, et al?:, Jurna and Grossman ~'~and Duggan et al. s because one is testing morphine on different baseline responses. In addition, the above studies indicate that morphine does have a spinal cord site of action which may also interfere with the interpretation of these experiments. That is, one cannot determine whether the spinal cord effects of morphine are responsible for all or only part of the effects noted when the spinal cord is intact. Thus, the possibility exists that conflicting conclusions were reached due to differences in the methods of analyzing the data. We therefore have re-examined the effects of systemically administered morphine on multireceptive dorsal horn neurons in the presence and absence of tonic descending inhibition. The data were analyzed using the methods of Hanaoka et al. ~2and Duggan et al. s, respectively. To determine how these calculations would be affected if the drug had a spinal cord site of action only, we performed similar experiments testing the effects of iontophoretically applied 7-aminobutyric acid (GABA) at the recording site. GABA, rather than iontophoretic morphine, was employed in these experiments because it
has a reliable and prompt effect of inhibiting the neurons under study, an effect which is consistently observed for intravenously administered morphine (1.0 mg/kg.) On the other hand, conflicting reports exist on the effect of morphine iontophoretically applied to these neurons. Morphine has been reported to elicit predominantly depression 3, excitation 24,2~,no effect or a bursting discharge pattern 4. In addition, Duggan et al? and Sastry and Goh > have provided evidence that morphine depresses spinal cord multireceptive neurons through an action in the substantia gelatinosa. Thus, in the present study, morphine and GABA may have different sites of action within the spinal cord, but they exhibit the common effect of inhibiting spinal cord multireceptive neurons. Therefore, GABA was used as a depressant agent whose action was restricted to the spinal cord. MATERIALS AND METHODS
Surgical preparation Adult cats of either sex (2.3-4.5 kg) were initially anesthetized with a 4% halothane/oxygen mixture and subsequently maintained with chloralose (60 mg/kg i.v.). Adequacy of anesthesia was determined by lack of surges in blood pressure or changes in pupillary diameter during noxious stimulation. Additional doses of chloralose were supplemented as required. Details of the surgical preparation were reported elsewhere 3°. Briefly, the lumbosacral region of the spinal cord L I S1 was exposed by laminectomy. A Peltier thermoelectric cooling device, fitted with a silver block which was shaped to make close contact with the dorsal half of the spinal cord, was carefully positioned at L I and maintained at 37"C until the cold block technique was employed. Cooling the silver block so that the temperature at the spinal cord-silver block interface was 2.5 3.0°C resulted in a blockade of spinal cord axonal conduction within 4 min ~°. End-tidal CO, levels, arterial blood pressure as well as body and spinal cord oil pool temperatures were continuously monitored and maintained within physiological limits. The animals were paralyzed with Flaxedil and artificially re-
spired. A bilateral pneumothorax was routinely performed to minimize respiration-induced spinal cord movements.
Location, stimulation and testing of spinal cord multireceptive neurons One barrel of a 3-barrel micropipette (3-4 ~m tip diameter) was filled with 4 M KC1 and used to record single unit extracellular activity in the L7.segment of the spinal cord. The other barrels contained GABA (0.5 M, pH 3.6) and NaC1 (2.0 M). The latter barrel was used for automatic current control. A Dagan 6400 current generator was used to control the release of GABA and a retaining current of 10 nA was used to prevent its leakage. Stimulation of an intact L7 dorsal root (3.0 V, 0.1 ms, 1 Hz) was used as a search stimulus for dorsal horn neurons. Cells encountered, located mainly in lamina V with some overlap into laminae IV and VI, were generally of two types: those which responded only to non-noxious forms of stimuli such as touch or hair movement and those which responded to non-noxious as well as noxious stimuli such as pinch or radiant heat above 45 °C. Only the latter group of cells, which are termed multireceptive neurons and which
A
B
180.
180,
I'b
COICII:IOI.L ~
i
238
0,
........................................
.......
2oj ,
°'-.k 28S
,n,nlo,.o.,.o.n,hdllUd~ll~uOm
................................. I ~ .......... "'W"
:T o -
correspond to class 213, lamina V-type t9, wide dynamic range:, convergent ~s and nociceptordriven neurons 29.3°were used in this study. To consistently activate and quantify the responses of these cells, a microcomputer (Rockwell AIM 65) was programmed to cycle pulses of noxious radiant heat (47 56°C, 10-15 s, 2 min intervals, feedback control) onto the receptive field while counting the number of single spikes discharged 25 s before (spontaneous activity) and after the onset of each heat pulse. The noxious heat-evoked responses were determined by subtracting the spontaneous activity from the total spikes counted over 25 s beginning with the onset of the noxious radiant heat pulse. The temperature and duration of noxious radiant heat were carefully adjusted to produce reproducible responses. For each neuron the degree of tonic descending inhibition was determined by comparing the mean noxious heat response of at least 3 trials obtained in the normal and cold block states of the spinal cord. The cord was then rewarmed to restore tonic descending inhibition, evident by the return of heat-evoked responses of the same magnitude as those obtained before the block. In each experiment GABA was first tested on
i
~BA
3.2 r~
107
o,,m
l,,,l.,.h,llmh.ld.lkln~dulnuo.Ii.i m
~.----. NOmO~..--,--~~----eot.oatOeK--] Fig. 1. The inhibitory effect of morphine (1.0 mg/kg i.v.) and iontophoretically applied GABA on two dorsal horn neurons responding to noxious radiant heat (H). The histograms were constructed using a computer and represent the average frequency of cellular discharge over 55 s, around a noxious radiant heat pulse. Each histogram was compiled from 3 noxious radiant heat responses during the normal and cold blocked states of the spinal cord, before and after morphine (A) and iontophoretically applied GABA (B). The number over each histogram corresponds to the mean noxious heat-evoked discharge.
the noxious heat-evoked responses of a few neurons in both the normal and cold block states of the cord. The current strength used to iontophorese GABA was adjusted to suppress the heatevoked responses in the normal state of the cord by approximately 50%. The epoch time for GABA release was !15 s: 90 s before and 25 s after the onset of noxious radiant heat. Following the iontophoretic study, and using a single barrel electrode, one cell was selected for testing intravenously administered morphine. In each experiment the response of the cell to noxious radiant heat was determined in the presence and absence of tonic descending inhibition both before and after the infusion of morphine. Morphine sulphate (1.0 mg/kg) was administered slowly either as a single infusion over 20 min or in two divided doses of 0.5 m g / k g approximately 30 min apart. The narcotic antagonist, naloxone HCI ( 0 . 1 2.0 mg/kg) was administered slowly about 35 min following the single infusion of morphine (1.0 mg/kg).
A SPONTANEOUS
m
co.~_
m
I~
~ SO t 1.O rng/kg n= 7
~/'/~ IONTOPHORETIC QABA n+12
80O.
co.~
co
eo~
~o
I° 20
t,K~I~MAL
COLD BLOCK
2i_ NORMAL
COLD BLOCK
B NOXIOUS HEAT-EVOKED m
CONTROL 1.0 mg/kg n - 7
m
o~,¢rRot
tv,,~,~mow,o~rne GABA n . 1 2
to
)oi
RESULTS NORMAL
A total of 24 multireceptive neurons were examined in this study. Cold blocking the spinal cord resulted in an increase in the spontaneous activity in most of these cells and a marked enhancement of the noxious heat-evoked responses in all neurons examined (Figs. 1 and 2). An example of the suppressive effects of morphine (1.0 m g / k g i.v.) and iontophoretically applied GABA on individual multireceptive neurons is illustrated in Fig. 1. Fig. 2 summarizes the effects of these agents on both spontaneous and heat-evoked activity in the presence and absence of tonic descending inhibition. Systemically administered morphine in a single infusion of 1.0 m g / k g failed to alter the spontaneous activity of these cells in either state of the spinal cord (Fig. 2A). However, if data from the animals receiving morphine (1.0 mg/kg) in divided doses and as a single infusion were grouped together, morphine did significantly decrease the spontaneous activity of these cells, but only when the spinal cord was cold
COlD BLOCK
tq~MAL
COLDBLOCK
Fig. 2. A comparison of systemically administered morphine and iontophoretically applied GABA on dorsal horn neurons with the spinal cord in the normal and cold-blocked states. Stars indicate a significant difference (P ~ 0.05) from corresponding controls. (See text for full description.)
blocked (P ~ 0.05, paired Student's t-test, n = 12). The iontophoretic release of GABA (2.7 12.5 nA, mean 6.3 __. 1.1 S.E.M., n = 12) at the recording site resulted in a significant decrease in the spontaneous activity in either state of the spinal cord (Fig. 2A). Morphine significantly reduced the noxious heat-evoked responses by 70% and 50% with the spinal cord in the normal and cold blocked state, respectively. Similarly, iontophoretically applied GABA reduced these responses by 50% and 40% in the presence and absence of tonic supraspinal inhibition, respectively (Fig. 2B). Morphine, unlike GABA, produced a mild hypotension in these experiments. The mean blood pressure was 109.5 ___ 4.9 m m Hg before and 91.9 ± 6.4 m m Hg 15-20 min after the infu-
sion (P ~ 0.05, n = 7). The temperature and the duration of noxious radiant heat pulses were the same in the two groups (49.8 _ 0.5 °C, 11.6 _ 0.7 s, n = 7 for morphine and 51.9 _ 0.9 °C, 11.5 _ 0.5 s, n = 12 for GABA). After examining the effects of a single infusion of morphine (1.0 mg/kg), naloxone was infused in a dose ranging from 0.1 to 2.0 mg/kg i.v. Of the 5 cells that were 'held' sufficiently long to examine the effects of this narcotic antagonist, a reversal of morphine's effect was seen in 4 cases while the other was not affected. The mean noxious heat-evoked response of the 3 cells tested with the spinal cord in the normal state was 212 _ 31 and 62 _ 14 before and after morphine, respectively, and 210 _ 42 after naloxone. The noxious heat-evoked response for one cell tested with the spinal cord cold blocked before and after morphine was 855 and 685, respectively, and 837 after naloxone. Morphine, when administered in a dose of 0.5 mg/kg, failed to reduce the noxious heat-evoked response whether the spinal cord was in the normal state (296 _ 93 to 242.3 ± 113.6, n = 5) or cold-blocked (780.5 ± 145.0 to 774 ± 173). However, the addition of a second 0.5 mg/kg dose depressed the noxious heat-evoked response to the same level as that produced by a single infusion of I mg/kg morphine (92 ± 46 and 68 ± 30 with the spinal cord in the normal state; 477 _ 109 and 392 _ 98 during the cold blocked state of the spinal cord). DISCUSSION
Noxious radiant heat ( > 4 5 °C) applied to the receptive field was used in this study to activate the dorsal horn neurons. We had previously found this stimulus to be effective and specific for activating dorsal horn neurons receiving input from nociceptors 28. The use of noxious radiant heat has been criticized on the basis that drug-induced circulatory changes in the receptive field would alter the dissipation of the heat and therefore alter the extent of nociceptor activation 5~8. For example, a vasodepressor agent would decrease the transfer of heat and increase the nociceptor response. To circumvent this
problem, some investigators activated dorsal horn neurons by stimulating the ipsilateral tibial nerve at an intensity sufficient to activate C-fibersS.8. The disadvantage of the use of noxious radiant heat can be largely negated in the present study since we used feedback from a thermocoupie positioned on the receptive field to control the surface temperature. In addition, the morphine was administered slowly to minimize blood pressure changes, and finally, morphine produced a hypotension but decreased the response of dorsal horn neurons. We should also point out that the use of electrical stimulation of peripheral nerves to activate dorsal horn neurons is not without disadvantages. An intense stimulation of a nerve trunk will produce an inhomogeneous barrage of afferent impulses which may result in a variety of interactions within the spinal cord, including inhibition of the neuron under study. For example, Light et al. 22 found that certain dorsal horn neurons were excited by natural cutaneous stimuli but could not be activated by intense stimulation of the dorsal roots. If the drug under study alters such an inhibition, misinterpretation of the data may result. Morphine (1.0 mg/kg i.v.) has been shown to depress the response of multireceptive neurons in the presence and absence of tonic descending inhibition. The fact that morphine depressed the responses with the spinal cord Cold blocked clearly indicates that morphine has a spinal cord site of action. It is not immediately clear whether this spinal cord effect contributes to all or only part of the suppressive effect of morphine when conduction in the spinal cord is normal. The method of analysis employed in this type of experiment may lead to contrasting interpretations. Hanaoka et al. ~2, analyzing their data on a percentage change basis, found that morphine produced a greater percentage decrease during normal spinal cord conduction than during the absence of tonic descending inhibition. They therefore concluded that morphine, through a supraspinal action, attenuates the activity of spinal cord dorsal horn neurons. Our results are consistent with these findings in that morphine decreased the heat-evoked responses by 70%
with normal spinal cord conduction and 50(,~ with the cord cold-blocked. However, GABA also suppressed the noxious heat-evoked responses of all neurons tested with the spinal cord in the normal and cold blocked states by 50~ and 40%, respectively. In this case there is no possibility of GABA acting via a supraspinal site since it was applied by iontophoresis in very small quantities. Thus, calculating the changes on a 'percentage of control' basis would not seem to be valid when the control values are markedly different. Another method of analysis was used by Duggan et al?. They considered the extent of tonic descending inhibition to be reflected by the increase in the number of noxious stimulus-evoked spikes produced when the spinal cord was cold blocked. Since they found systemically administered morphine to decrease this value they concluded that morphine decreased the tonic descending inhibition. Similarly, in the present study, we found the difference in the noxious heatevoked discharges obtained with the spinal cord in the normal vs cold-blocked state to be decreased by morphine from a mean of 558 324. However, GABA, iontophoretically applied to the spinal cord multireceptive neurons, also reduced this absolute difference from a mean of 658 411. Thus, the difference in the evoked discharge with the spinal cord in the normal vs cold-blocked state was decreased by systemically administered morphine and iontophoretically applied GABA. It seems more likely to us that this decrease was due to a reduction in the initial baseline levels than a decrease in tonic descending inhibition. Furthermore, if morphine decreases tonic descending inhibition, one would expect morphine to enhance the noxious heatevoked responses during normal spinal cord conduction as is observed when tonic inhibition is removed by cold-blocking the spinal cord. Such a response would, of course, be offset to some extent by the spinal cord depressant effects of morphine. REFERENCES 1 Basbaum. A. I.. Marlev. N. ,1. E., O'Keefe. J. and Clan-
in summary, we have shown that i.v. administered morphine depresses multireceptive dorsal horn neurons in the presence and absence of tonic descending inhibition. We have pointed out how different methods of analysis have led to conflicting interpretations of the data with regard to whether morphine elicits a supraspinal mediated descending inhibition. In addition, the similarity in effect between i.v. administered morphine and iontophoretically applied GABA to spinal cord multireceptive neurons suggests that morphine's actions can be adequately accounted lbr by a spinal cord site of action. Although the design of these experiments does not allow one to rule out the possibility of a supraspinal influence of morphine, in our view such an effect would appear to be minor. We do not imply, however, that these multireceptive neurons are not under supraspinal influences. Many studies have shown that electrical stimulation of brainstem sites such as the periaqueductal grey, the nucleus raphe magnus and the nucleus gigantocellularis produce inhibition of these neurons ~t.2L-~5. Furthermore, as shown in this and other studies, these neurons are under a powerful supraspinal tonic inhibition although the neurotransmitters mediating this inhibition have not been identified 727>3°. Finally. although morphine has been shown in this study to have a spinal cord site of action, it does not preclude the possibility of the drug also having a supraspinal site of action not involving descending inhibition. For example, morphine may inhibit ascending nociceptive transmission at the supraspinal level. ACKNOWLEDGEMENTS
The authors wish to thank David Harris for writing the computer programs used in this study. Naloxone was kindly provided by Endo Laboratories. This work was supported by a grant from the Medical Research Council of Canada to J.G.S. ton. C. H.. Reversal of morphine and stimulation-produced analgesia by subtotal spinal cord lesions, Pain, 3 (1977)43 56.
2 Bennett, G. J. and Mayer, D. J.,Inhibition of spinal cord interneurones by narcotic microinjection and focal electrical stimulation in the periaqueductal gray matter, Brain Research. 172 (1979) 243-257. 3 Calvillo, O., Henry, J. L. and Neuman, R. S., Effects of morphine and naloxone on dorsal horn neurones in the cat, Canad. J. Physiol. Pharmacoil, 52 (1974) 1207-1211. 4 Duggan, A. W., Hall, J. G. and Headley, P. M., Suppres sion of transmission of nociceptive impulses by morphine administered in the region of the substantia gelatinosa, Brit. J. Pharmacol., 61 (1977) 65-76. 5 Duggan, A. W. and Griersmith, B. T., Methyl xanthines, adenosine 3',5'-cyclic monophosphate and the spinal transmission ofnociceptive information, Brit. J. PharmaCOIl, 67 (1979) 51- 57. 6 Duggan, A. W., Griersmith, B. T., Headley, P. M. and Maher, J. B., The need to control skin temperature when using radiant heat in tests of analgesia, Exp. NeuroIl, 61 (1978)471 478. 7 Duggan, A. W., Griersmith, B. T. and Johnson, S. M., Supraspinal inhibition of the excitation of dorsal horn neurones by impulses in unmyelinated primary afterents: lack of effect by strychnine and bicuculline, Brain Research, 210 (1981) 231-241. 8 Duggan, A. W., Griersmith, B. T. and North R. A., Morphine and supra-spinal inhibition of spinal neurones: evidence that morphine decreases tonic descending inhibition in the anaesthetized cat, Brit. J. PharmacoIl, 69 (1980)461 466. 9 Gebhart, G. F., Opiate and opioid peptide effects on brain stem neurons: relevance to nociception and antinociceptive mechanisms, Pain, 12 (1982) 93 140. 10 Fields, H.L. and Basbaum, A. I., Brainstem control of spinal pain-transmission neurons, Ann. Rev. Physiol., 40 (1978) 217-248. I 1 Haber, L. H., Martin, R. F., Chatt, A. B. and Willis, W. D., Effects of stimulation in nucleus reticularis gigantocellularis on the activity of spinothalamic tract neurons in the monkey, Brain Research, 153 (1978) 163 168. 12 Hanaoka, K., Ohtani, M. Toyooka, H., Dohi, S., Ghazisaidi, K., Taub, A. and Kitahata, L. M., The relative contribution of direct and supraspinal descending effects upon spinal mechanisms of morphine analgesia, J. Pharmacoil exp. Ther., 207 (1978) 476-484. 13 Handwerker, H. O., Iggo, A. and Zimmermann, M., Segmental and supraspinal actions on dorsal horn neurones responding to noxious and non-noxious skin stimuli, Pain, 1 (1975) 147 165. 14 Irwin, S., Houde, R. W., Bennett, D. R., Hendershot, L. C. and Seevers, M. H., The effects of morphine, methadone and meperidine on some reflex responses of spinal animals to nociceptive stimulation, J. Pharmacol. exp. Ther., 101 (1951) 132 143. 15 Jacquet, Y. F. and Lajtha, A., Morphine action at central nervous system sites in rat: analgesia or hyperalgesia depending on site and dose, Science, 182 (1973) 490-491. 16 Jurna, I. and Grossmann, W., The effect of morphine on activity evoked in ventrolateral tract axons of the cat spinal cord, Exp. Brain Res., 24 (1976) 473-484. 17 Kitahata, L. M., Kosaka, Y., Taub, A., Konikos, K. and Hoffert, M., Lamina specific suppression of dorsal horn unit activity by morphine sulfate, Anesthesiology, 41 (1974)39 48. 18 Le Bars, D., Guilbaud, G., Chitour, D. and Besson, J.M., Does systemic morphine increase descending inhibitory controls of dorsal horn neurones involved in noci-
ception? Brain Research, 202 (1980) 467-481. 19 Le Bars, D., Menetrey, D. and Besson, J.-M., Effects of morphine upon the lamina V type cells activities in the dorsal horn of the decerebrate cat, Brain Research, 113 (1976) 293-310. 20 Levine, J. D., Lane, S. R., Gordon, N. C. and Fields, H. L., A spinal opioid synapse mediates the interaction of spinal and brain stem sites in morphine analgesia, Brain Research, 236 (1982) 85-91. 21 Liebeskind, J. C., Guilbaud, G., Besson~ J.-M. and Oliveras, J.-L., Analgesia from electrical stimulation of the periaqueductal gray matter in the cat: behavioral observations and inhibitory effects on spinal cord interneurons, Brain Research, 50 (1973) 441-446. 22 Light, A. R., Trevino, D. L. and Perl, E. R., Morphological features of functionally defined neurones in the marginal zone and substantia gelatinosa of the spinal dorsal horn, J. comp. Neuroil, 186 (1979) 151-172. 23 Necker, R. and Hellon, R. F., Noxious thermal input from the rat tail: modulation by descending inhibitory influences, Pain, 4 (1978) 231 242. 24 Piercey, M. F., Einspahr, F. J., Dobry, P. J. K., Schroeder, L. A. and Hollister, R. P., Morphine does not antagonize the substance P mediated excitation of dorsal horn neurons, Brain Research, 186 (1980) 421 434. 25 Oliveras, J. L., Redjemi, F., Guilband, G. and Besson, J.M., Analgesia produced by electrical stimulation of the inferior centralis of the raphe in the cat, Pain, 1 (1975) 139-145. 26 Sastry, B. R. and Goh, J. W., Actions of morphine and met-enkephalinamide on nociceptor-driven neurones in the substantia gelatinosa and deeper dorsal horn, Neuropharmacol., 22 (1983) 119-122. 27 Sinclair, J. G., Fox, R. E., Mokha, S. S. and Iggo, A., The effect of naloxone on the inhibition of nociceptor-driven neurones in the cat spinal cord, Quart. J. exp. Physioil, 65 (1980) 181-188. 28 Soja, P. J. and Sinclair, J. G., The response of dorsal horn neurones of the cat to intra-arterial bradykinin and noxious radiant heat, Neurosci. Lett., 20 (1980) 183-188. 29 Soja, P. J. and Sinclair, J. G., Evidence against a serotonin involvement in the tonic descending inhibition ofnociceptor-driven neurones in the cat spinal cord, Brain Research, 199 (1980) 225-230. 30 Soja, P. J. and Sinclair, J. G., Evidence that noradrenaline reduces tonic descending inhibition of cat spinal cord nociceptor-driven neurones, Pain, 15 (1983) 71-81. 31 Takagi, H., Satoh, M., Doi, T., Kawasaki, K. and Akaike, A., Indirect and direct depressive effects of morphine on activation of lamina V cell of the spinal dorsal horn i n duced by intra-arterial injection of bradykinin, Arch. Int. Pharmacodvn. Ther., 221 (1976) 96-103. 32 Wall, P. D., The laminar organization of the dorsal horn and effects on descending impulses, J. Physiol. (Lond.), 188 (1967) 403-423. 33 Yaksh, T. L., Direct evidence that spinal serotonin and noradrenaline terminals mediate the spinal antinociceptive effects of morphine in the periaqueductal gray, Brain Research, 160 ( i 979) 180-185. 34 Yaksh, T. L. and Rudy, T, A,, Studies on the direct spinal action of the narcotics in the production of anagesia in the rat, J. Pharmacol. exp. Ther., 202 (1977) 411-428. 35 Yaksh, T. L. and Rudy, T. A., Narcotic analgetics: CNS sites and mechanisms of action as revealed by intracerebral injection techniques, Pain, 4 (1978) 299 359.
1
Elsevier
Research Reports
Spinal vs Supraspinal Actions of Morphine on Cat Spinal Cord Multireceptive Neurons PETER J. SOJA and JOHN G. SINCLAIR
Division of Pharmacologyand Toxicology, Faculty of Pharmaceutical Sciences, Universityof British Columbia, Vancouver, BC V6T 114:5(Canada) (Accepted December 28th, 1982)
Key words: morphine - spinal cord - inhibition - naloxone - analgesia
To examine whether morphine elicits a supraspinal mediated spinal inhibition of nociceptive transmission, several investigators have compared the effects of morphine on nociceptive transmission in animals with the spinal cord intact vs transected or cold-blocked. The results have been conflicting, possibly due to different methods of analysis. For example, some investigators have found i.v. administered morphine produces a greater percentage decrease in nociceptive transmission when the spinal cord is intact compared to the transected state. Therefore, they concluded that morphine elicits a supraspinal-mediated inhibition. Conversely, others have reported that the increase in noxious stimulus-evoked responses of dorsal horn neurons upon cold blocking the spinal cord was reduced by i.v. morphine. They therefore concluded that morphine decreases descending inhibition. We tested the effects of i.v. morphine on spinal cord multireceptive neurons in the presence and absence of descending inhibition. Using the above methods of analysis, our results were found to be consistent with their findings which indicate that the method of analysis used is critical to the interpretation reached. To determine how these calculations would be affected by a depressant effect on the spinal cord neurons only, we performed similar experiments iontophoresing "t-aminobutyric acid (GABA) onto these dorsal horn neurons. The similarity between the morphine and GABA data suggests that the effects of systemically administered morphine on multireceptive dorsal horn neurons can be adequately explained by a spinal cord site of action. INTRODUCTION
Although there is good evidence that morphine elicits antinociception through a spinal cord site of a c t i o n 17,19.34, there is also evidence that morphine attenuates spinal cord nociceptive transmission through a supraspinal site of action (see reviews by Fields and Basbaum ~°, Yaksh and Rudy 35, and Gebhartg). The latter proposal was first made by Irwin et al) 4 who found the rat tail-flick test to be less sensitive to morphine in animals with a transected spinal cord as compared to those with an intact spinal cord. Other supporting evidence in rats includes the finding that intracerebral microinjections of morphine prolong the tailflick latency 15 and inhibit spinal cord neurons receiving noxious and inocuous input (multireceptive neurons) 2. In ad0006-8993/83/$03.00 © 1983 Elsevier Science Publishers B.V.
dition, antinociception produced by intracerebral microinjection of morphine is antagonized by dorsolateral funiculi lesions t or intrathecal injections of the narcotic antagonist, naloxone 20 (however, see ref. 33). To examine the question of whether morphine elicits a supraspinal-mediated spinal depression of nociceptive transmission, several investigators have used an approach similar to that first used by Irwin et al) 4. That is, they compared the effect of morphine on spinal cord nociceptive transmission with the spinal cord intact vs transected or cold-blocked. The results have been conflicting. Takagi et al.3 ~and Hanaoka et al) 2 have claimed that morphine increases bulbospinal inhibition impinging on spinal cord multireceptive neurons in the rabbit and cat, respectively. In contrast, Jurna and Grossman ~6
and Duggan et al. s concluded that morphine decreases supraspinal inhibition of cat spinal cord ventrolateral tract axons and dorsal horn neurons. respectively. Moreover, Le Bars et al. ~' reported that dose response curves of morphine's suppression of spinal cord multireceptive neurons from intact and spinal cord sectioned rats are superimposable, suggesting no reinforcement of descending controls. Hence, a controversy exists as to whether morphine depresses spinal cord nociceptive transmission via a supraspinal site of action. Transecting or cold-blocking the spinal cord releases multireceptive dorsal horn neurons of the cat ~ >.3°.-~2and rat 23 from a powerful tonic inhibitory impingement of supraspinal origin. This is reflected by a marked increase in the response of the cell to a noxious stimulus applied to the receptive field following release from the inhibition. This increase in excitability of these dorsal horn neurons complicates the interpretation of experiments such as those performed by Takagi et al. 3~, Hanaoka, et al?:, Jurna and Grossman ~'~and Duggan et al. s because one is testing morphine on different baseline responses. In addition, the above studies indicate that morphine does have a spinal cord site of action which may also interfere with the interpretation of these experiments. That is, one cannot determine whether the spinal cord effects of morphine are responsible for all or only part of the effects noted when the spinal cord is intact. Thus, the possibility exists that conflicting conclusions were reached due to differences in the methods of analyzing the data. We therefore have re-examined the effects of systemically administered morphine on multireceptive dorsal horn neurons in the presence and absence of tonic descending inhibition. The data were analyzed using the methods of Hanaoka et al. ~2and Duggan et al. s, respectively. To determine how these calculations would be affected if the drug had a spinal cord site of action only, we performed similar experiments testing the effects of iontophoretically applied 7-aminobutyric acid (GABA) at the recording site. GABA, rather than iontophoretic morphine, was employed in these experiments because it
has a reliable and prompt effect of inhibiting the neurons under study, an effect which is consistently observed for intravenously administered morphine (1.0 mg/kg.) On the other hand, conflicting reports exist on the effect of morphine iontophoretically applied to these neurons. Morphine has been reported to elicit predominantly depression 3, excitation 24,2~,no effect or a bursting discharge pattern 4. In addition, Duggan et al? and Sastry and Goh > have provided evidence that morphine depresses spinal cord multireceptive neurons through an action in the substantia gelatinosa. Thus, in the present study, morphine and GABA may have different sites of action within the spinal cord, but they exhibit the common effect of inhibiting spinal cord multireceptive neurons. Therefore, GABA was used as a depressant agent whose action was restricted to the spinal cord. MATERIALS AND METHODS
Surgical preparation Adult cats of either sex (2.3-4.5 kg) were initially anesthetized with a 4% halothane/oxygen mixture and subsequently maintained with chloralose (60 mg/kg i.v.). Adequacy of anesthesia was determined by lack of surges in blood pressure or changes in pupillary diameter during noxious stimulation. Additional doses of chloralose were supplemented as required. Details of the surgical preparation were reported elsewhere 3°. Briefly, the lumbosacral region of the spinal cord L I S1 was exposed by laminectomy. A Peltier thermoelectric cooling device, fitted with a silver block which was shaped to make close contact with the dorsal half of the spinal cord, was carefully positioned at L I and maintained at 37"C until the cold block technique was employed. Cooling the silver block so that the temperature at the spinal cord-silver block interface was 2.5 3.0°C resulted in a blockade of spinal cord axonal conduction within 4 min ~°. End-tidal CO, levels, arterial blood pressure as well as body and spinal cord oil pool temperatures were continuously monitored and maintained within physiological limits. The animals were paralyzed with Flaxedil and artificially re-
spired. A bilateral pneumothorax was routinely performed to minimize respiration-induced spinal cord movements.
Location, stimulation and testing of spinal cord multireceptive neurons One barrel of a 3-barrel micropipette (3-4 ~m tip diameter) was filled with 4 M KC1 and used to record single unit extracellular activity in the L7.segment of the spinal cord. The other barrels contained GABA (0.5 M, pH 3.6) and NaC1 (2.0 M). The latter barrel was used for automatic current control. A Dagan 6400 current generator was used to control the release of GABA and a retaining current of 10 nA was used to prevent its leakage. Stimulation of an intact L7 dorsal root (3.0 V, 0.1 ms, 1 Hz) was used as a search stimulus for dorsal horn neurons. Cells encountered, located mainly in lamina V with some overlap into laminae IV and VI, were generally of two types: those which responded only to non-noxious forms of stimuli such as touch or hair movement and those which responded to non-noxious as well as noxious stimuli such as pinch or radiant heat above 45 °C. Only the latter group of cells, which are termed multireceptive neurons and which
A
B
180.
180,
I'b
COICII:IOI.L ~
i
238
0,
........................................
.......
2oj ,
°'-.k 28S
,n,nlo,.o.,.o.n,hdllUd~ll~uOm
................................. I ~ .......... "'W"
:T o -
correspond to class 213, lamina V-type t9, wide dynamic range:, convergent ~s and nociceptordriven neurons 29.3°were used in this study. To consistently activate and quantify the responses of these cells, a microcomputer (Rockwell AIM 65) was programmed to cycle pulses of noxious radiant heat (47 56°C, 10-15 s, 2 min intervals, feedback control) onto the receptive field while counting the number of single spikes discharged 25 s before (spontaneous activity) and after the onset of each heat pulse. The noxious heat-evoked responses were determined by subtracting the spontaneous activity from the total spikes counted over 25 s beginning with the onset of the noxious radiant heat pulse. The temperature and duration of noxious radiant heat were carefully adjusted to produce reproducible responses. For each neuron the degree of tonic descending inhibition was determined by comparing the mean noxious heat response of at least 3 trials obtained in the normal and cold block states of the spinal cord. The cord was then rewarmed to restore tonic descending inhibition, evident by the return of heat-evoked responses of the same magnitude as those obtained before the block. In each experiment GABA was first tested on
i
~BA
3.2 r~
107
o,,m
l,,,l.,.h,llmh.ld.lkln~dulnuo.Ii.i m
~.----. NOmO~..--,--~~----eot.oatOeK--] Fig. 1. The inhibitory effect of morphine (1.0 mg/kg i.v.) and iontophoretically applied GABA on two dorsal horn neurons responding to noxious radiant heat (H). The histograms were constructed using a computer and represent the average frequency of cellular discharge over 55 s, around a noxious radiant heat pulse. Each histogram was compiled from 3 noxious radiant heat responses during the normal and cold blocked states of the spinal cord, before and after morphine (A) and iontophoretically applied GABA (B). The number over each histogram corresponds to the mean noxious heat-evoked discharge.
the noxious heat-evoked responses of a few neurons in both the normal and cold block states of the cord. The current strength used to iontophorese GABA was adjusted to suppress the heatevoked responses in the normal state of the cord by approximately 50%. The epoch time for GABA release was !15 s: 90 s before and 25 s after the onset of noxious radiant heat. Following the iontophoretic study, and using a single barrel electrode, one cell was selected for testing intravenously administered morphine. In each experiment the response of the cell to noxious radiant heat was determined in the presence and absence of tonic descending inhibition both before and after the infusion of morphine. Morphine sulphate (1.0 mg/kg) was administered slowly either as a single infusion over 20 min or in two divided doses of 0.5 m g / k g approximately 30 min apart. The narcotic antagonist, naloxone HCI ( 0 . 1 2.0 mg/kg) was administered slowly about 35 min following the single infusion of morphine (1.0 mg/kg).
A SPONTANEOUS
m
co.~_
m
I~
~ SO t 1.O rng/kg n= 7
~/'/~ IONTOPHORETIC QABA n+12
80O.
co.~
co
eo~
~o
I° 20
t,K~I~MAL
COLD BLOCK
2i_ NORMAL
COLD BLOCK
B NOXIOUS HEAT-EVOKED m
CONTROL 1.0 mg/kg n - 7
m
o~,¢rRot
tv,,~,~mow,o~rne GABA n . 1 2
to
)oi
RESULTS NORMAL
A total of 24 multireceptive neurons were examined in this study. Cold blocking the spinal cord resulted in an increase in the spontaneous activity in most of these cells and a marked enhancement of the noxious heat-evoked responses in all neurons examined (Figs. 1 and 2). An example of the suppressive effects of morphine (1.0 m g / k g i.v.) and iontophoretically applied GABA on individual multireceptive neurons is illustrated in Fig. 1. Fig. 2 summarizes the effects of these agents on both spontaneous and heat-evoked activity in the presence and absence of tonic descending inhibition. Systemically administered morphine in a single infusion of 1.0 m g / k g failed to alter the spontaneous activity of these cells in either state of the spinal cord (Fig. 2A). However, if data from the animals receiving morphine (1.0 mg/kg) in divided doses and as a single infusion were grouped together, morphine did significantly decrease the spontaneous activity of these cells, but only when the spinal cord was cold
COlD BLOCK
tq~MAL
COLDBLOCK
Fig. 2. A comparison of systemically administered morphine and iontophoretically applied GABA on dorsal horn neurons with the spinal cord in the normal and cold-blocked states. Stars indicate a significant difference (P ~ 0.05) from corresponding controls. (See text for full description.)
blocked (P ~ 0.05, paired Student's t-test, n = 12). The iontophoretic release of GABA (2.7 12.5 nA, mean 6.3 __. 1.1 S.E.M., n = 12) at the recording site resulted in a significant decrease in the spontaneous activity in either state of the spinal cord (Fig. 2A). Morphine significantly reduced the noxious heat-evoked responses by 70% and 50% with the spinal cord in the normal and cold blocked state, respectively. Similarly, iontophoretically applied GABA reduced these responses by 50% and 40% in the presence and absence of tonic supraspinal inhibition, respectively (Fig. 2B). Morphine, unlike GABA, produced a mild hypotension in these experiments. The mean blood pressure was 109.5 ___ 4.9 m m Hg before and 91.9 ± 6.4 m m Hg 15-20 min after the infu-
sion (P ~ 0.05, n = 7). The temperature and the duration of noxious radiant heat pulses were the same in the two groups (49.8 _ 0.5 °C, 11.6 _ 0.7 s, n = 7 for morphine and 51.9 _ 0.9 °C, 11.5 _ 0.5 s, n = 12 for GABA). After examining the effects of a single infusion of morphine (1.0 mg/kg), naloxone was infused in a dose ranging from 0.1 to 2.0 mg/kg i.v. Of the 5 cells that were 'held' sufficiently long to examine the effects of this narcotic antagonist, a reversal of morphine's effect was seen in 4 cases while the other was not affected. The mean noxious heat-evoked response of the 3 cells tested with the spinal cord in the normal state was 212 _ 31 and 62 _ 14 before and after morphine, respectively, and 210 _ 42 after naloxone. The noxious heat-evoked response for one cell tested with the spinal cord cold blocked before and after morphine was 855 and 685, respectively, and 837 after naloxone. Morphine, when administered in a dose of 0.5 mg/kg, failed to reduce the noxious heat-evoked response whether the spinal cord was in the normal state (296 _ 93 to 242.3 ± 113.6, n = 5) or cold-blocked (780.5 ± 145.0 to 774 ± 173). However, the addition of a second 0.5 mg/kg dose depressed the noxious heat-evoked response to the same level as that produced by a single infusion of I mg/kg morphine (92 ± 46 and 68 ± 30 with the spinal cord in the normal state; 477 _ 109 and 392 _ 98 during the cold blocked state of the spinal cord). DISCUSSION
Noxious radiant heat ( > 4 5 °C) applied to the receptive field was used in this study to activate the dorsal horn neurons. We had previously found this stimulus to be effective and specific for activating dorsal horn neurons receiving input from nociceptors 28. The use of noxious radiant heat has been criticized on the basis that drug-induced circulatory changes in the receptive field would alter the dissipation of the heat and therefore alter the extent of nociceptor activation 5~8. For example, a vasodepressor agent would decrease the transfer of heat and increase the nociceptor response. To circumvent this
problem, some investigators activated dorsal horn neurons by stimulating the ipsilateral tibial nerve at an intensity sufficient to activate C-fibersS.8. The disadvantage of the use of noxious radiant heat can be largely negated in the present study since we used feedback from a thermocoupie positioned on the receptive field to control the surface temperature. In addition, the morphine was administered slowly to minimize blood pressure changes, and finally, morphine produced a hypotension but decreased the response of dorsal horn neurons. We should also point out that the use of electrical stimulation of peripheral nerves to activate dorsal horn neurons is not without disadvantages. An intense stimulation of a nerve trunk will produce an inhomogeneous barrage of afferent impulses which may result in a variety of interactions within the spinal cord, including inhibition of the neuron under study. For example, Light et al. 22 found that certain dorsal horn neurons were excited by natural cutaneous stimuli but could not be activated by intense stimulation of the dorsal roots. If the drug under study alters such an inhibition, misinterpretation of the data may result. Morphine (1.0 mg/kg i.v.) has been shown to depress the response of multireceptive neurons in the presence and absence of tonic descending inhibition. The fact that morphine depressed the responses with the spinal cord Cold blocked clearly indicates that morphine has a spinal cord site of action. It is not immediately clear whether this spinal cord effect contributes to all or only part of the suppressive effect of morphine when conduction in the spinal cord is normal. The method of analysis employed in this type of experiment may lead to contrasting interpretations. Hanaoka et al. ~2, analyzing their data on a percentage change basis, found that morphine produced a greater percentage decrease during normal spinal cord conduction than during the absence of tonic descending inhibition. They therefore concluded that morphine, through a supraspinal action, attenuates the activity of spinal cord dorsal horn neurons. Our results are consistent with these findings in that morphine decreased the heat-evoked responses by 70%
with normal spinal cord conduction and 50(,~ with the cord cold-blocked. However, GABA also suppressed the noxious heat-evoked responses of all neurons tested with the spinal cord in the normal and cold blocked states by 50~ and 40%, respectively. In this case there is no possibility of GABA acting via a supraspinal site since it was applied by iontophoresis in very small quantities. Thus, calculating the changes on a 'percentage of control' basis would not seem to be valid when the control values are markedly different. Another method of analysis was used by Duggan et al?. They considered the extent of tonic descending inhibition to be reflected by the increase in the number of noxious stimulus-evoked spikes produced when the spinal cord was cold blocked. Since they found systemically administered morphine to decrease this value they concluded that morphine decreased the tonic descending inhibition. Similarly, in the present study, we found the difference in the noxious heatevoked discharges obtained with the spinal cord in the normal vs cold-blocked state to be decreased by morphine from a mean of 558 324. However, GABA, iontophoretically applied to the spinal cord multireceptive neurons, also reduced this absolute difference from a mean of 658 411. Thus, the difference in the evoked discharge with the spinal cord in the normal vs cold-blocked state was decreased by systemically administered morphine and iontophoretically applied GABA. It seems more likely to us that this decrease was due to a reduction in the initial baseline levels than a decrease in tonic descending inhibition. Furthermore, if morphine decreases tonic descending inhibition, one would expect morphine to enhance the noxious heatevoked responses during normal spinal cord conduction as is observed when tonic inhibition is removed by cold-blocking the spinal cord. Such a response would, of course, be offset to some extent by the spinal cord depressant effects of morphine. REFERENCES 1 Basbaum. A. I.. Marlev. N. ,1. E., O'Keefe. J. and Clan-
in summary, we have shown that i.v. administered morphine depresses multireceptive dorsal horn neurons in the presence and absence of tonic descending inhibition. We have pointed out how different methods of analysis have led to conflicting interpretations of the data with regard to whether morphine elicits a supraspinal mediated descending inhibition. In addition, the similarity in effect between i.v. administered morphine and iontophoretically applied GABA to spinal cord multireceptive neurons suggests that morphine's actions can be adequately accounted lbr by a spinal cord site of action. Although the design of these experiments does not allow one to rule out the possibility of a supraspinal influence of morphine, in our view such an effect would appear to be minor. We do not imply, however, that these multireceptive neurons are not under supraspinal influences. Many studies have shown that electrical stimulation of brainstem sites such as the periaqueductal grey, the nucleus raphe magnus and the nucleus gigantocellularis produce inhibition of these neurons ~t.2L-~5. Furthermore, as shown in this and other studies, these neurons are under a powerful supraspinal tonic inhibition although the neurotransmitters mediating this inhibition have not been identified 727>3°. Finally. although morphine has been shown in this study to have a spinal cord site of action, it does not preclude the possibility of the drug also having a supraspinal site of action not involving descending inhibition. For example, morphine may inhibit ascending nociceptive transmission at the supraspinal level. ACKNOWLEDGEMENTS
The authors wish to thank David Harris for writing the computer programs used in this study. Naloxone was kindly provided by Endo Laboratories. This work was supported by a grant from the Medical Research Council of Canada to J.G.S. ton. C. H.. Reversal of morphine and stimulation-produced analgesia by subtotal spinal cord lesions, Pain, 3 (1977)43 56.
2 Bennett, G. J. and Mayer, D. J.,Inhibition of spinal cord interneurones by narcotic microinjection and focal electrical stimulation in the periaqueductal gray matter, Brain Research. 172 (1979) 243-257. 3 Calvillo, O., Henry, J. L. and Neuman, R. S., Effects of morphine and naloxone on dorsal horn neurones in the cat, Canad. J. Physiol. Pharmacoil, 52 (1974) 1207-1211. 4 Duggan, A. W., Hall, J. G. and Headley, P. M., Suppres sion of transmission of nociceptive impulses by morphine administered in the region of the substantia gelatinosa, Brit. J. Pharmacol., 61 (1977) 65-76. 5 Duggan, A. W. and Griersmith, B. T., Methyl xanthines, adenosine 3',5'-cyclic monophosphate and the spinal transmission ofnociceptive information, Brit. J. PharmaCOIl, 67 (1979) 51- 57. 6 Duggan, A. W., Griersmith, B. T., Headley, P. M. and Maher, J. B., The need to control skin temperature when using radiant heat in tests of analgesia, Exp. NeuroIl, 61 (1978)471 478. 7 Duggan, A. W., Griersmith, B. T. and Johnson, S. M., Supraspinal inhibition of the excitation of dorsal horn neurones by impulses in unmyelinated primary afterents: lack of effect by strychnine and bicuculline, Brain Research, 210 (1981) 231-241. 8 Duggan, A. W., Griersmith, B. T. and North R. A., Morphine and supra-spinal inhibition of spinal neurones: evidence that morphine decreases tonic descending inhibition in the anaesthetized cat, Brit. J. PharmacoIl, 69 (1980)461 466. 9 Gebhart, G. F., Opiate and opioid peptide effects on brain stem neurons: relevance to nociception and antinociceptive mechanisms, Pain, 12 (1982) 93 140. 10 Fields, H.L. and Basbaum, A. I., Brainstem control of spinal pain-transmission neurons, Ann. Rev. Physiol., 40 (1978) 217-248. I 1 Haber, L. H., Martin, R. F., Chatt, A. B. and Willis, W. D., Effects of stimulation in nucleus reticularis gigantocellularis on the activity of spinothalamic tract neurons in the monkey, Brain Research, 153 (1978) 163 168. 12 Hanaoka, K., Ohtani, M. Toyooka, H., Dohi, S., Ghazisaidi, K., Taub, A. and Kitahata, L. M., The relative contribution of direct and supraspinal descending effects upon spinal mechanisms of morphine analgesia, J. Pharmacoil exp. Ther., 207 (1978) 476-484. 13 Handwerker, H. O., Iggo, A. and Zimmermann, M., Segmental and supraspinal actions on dorsal horn neurones responding to noxious and non-noxious skin stimuli, Pain, 1 (1975) 147 165. 14 Irwin, S., Houde, R. W., Bennett, D. R., Hendershot, L. C. and Seevers, M. H., The effects of morphine, methadone and meperidine on some reflex responses of spinal animals to nociceptive stimulation, J. Pharmacol. exp. Ther., 101 (1951) 132 143. 15 Jacquet, Y. F. and Lajtha, A., Morphine action at central nervous system sites in rat: analgesia or hyperalgesia depending on site and dose, Science, 182 (1973) 490-491. 16 Jurna, I. and Grossmann, W., The effect of morphine on activity evoked in ventrolateral tract axons of the cat spinal cord, Exp. Brain Res., 24 (1976) 473-484. 17 Kitahata, L. M., Kosaka, Y., Taub, A., Konikos, K. and Hoffert, M., Lamina specific suppression of dorsal horn unit activity by morphine sulfate, Anesthesiology, 41 (1974)39 48. 18 Le Bars, D., Guilbaud, G., Chitour, D. and Besson, J.M., Does systemic morphine increase descending inhibitory controls of dorsal horn neurones involved in noci-
ception? Brain Research, 202 (1980) 467-481. 19 Le Bars, D., Menetrey, D. and Besson, J.-M., Effects of morphine upon the lamina V type cells activities in the dorsal horn of the decerebrate cat, Brain Research, 113 (1976) 293-310. 20 Levine, J. D., Lane, S. R., Gordon, N. C. and Fields, H. L., A spinal opioid synapse mediates the interaction of spinal and brain stem sites in morphine analgesia, Brain Research, 236 (1982) 85-91. 21 Liebeskind, J. C., Guilbaud, G., Besson~ J.-M. and Oliveras, J.-L., Analgesia from electrical stimulation of the periaqueductal gray matter in the cat: behavioral observations and inhibitory effects on spinal cord interneurons, Brain Research, 50 (1973) 441-446. 22 Light, A. R., Trevino, D. L. and Perl, E. R., Morphological features of functionally defined neurones in the marginal zone and substantia gelatinosa of the spinal dorsal horn, J. comp. Neuroil, 186 (1979) 151-172. 23 Necker, R. and Hellon, R. F., Noxious thermal input from the rat tail: modulation by descending inhibitory influences, Pain, 4 (1978) 231 242. 24 Piercey, M. F., Einspahr, F. J., Dobry, P. J. K., Schroeder, L. A. and Hollister, R. P., Morphine does not antagonize the substance P mediated excitation of dorsal horn neurons, Brain Research, 186 (1980) 421 434. 25 Oliveras, J. L., Redjemi, F., Guilband, G. and Besson, J.M., Analgesia produced by electrical stimulation of the inferior centralis of the raphe in the cat, Pain, 1 (1975) 139-145. 26 Sastry, B. R. and Goh, J. W., Actions of morphine and met-enkephalinamide on nociceptor-driven neurones in the substantia gelatinosa and deeper dorsal horn, Neuropharmacol., 22 (1983) 119-122. 27 Sinclair, J. G., Fox, R. E., Mokha, S. S. and Iggo, A., The effect of naloxone on the inhibition of nociceptor-driven neurones in the cat spinal cord, Quart. J. exp. Physioil, 65 (1980) 181-188. 28 Soja, P. J. and Sinclair, J. G., The response of dorsal horn neurones of the cat to intra-arterial bradykinin and noxious radiant heat, Neurosci. Lett., 20 (1980) 183-188. 29 Soja, P. J. and Sinclair, J. G., Evidence against a serotonin involvement in the tonic descending inhibition ofnociceptor-driven neurones in the cat spinal cord, Brain Research, 199 (1980) 225-230. 30 Soja, P. J. and Sinclair, J. G., Evidence that noradrenaline reduces tonic descending inhibition of cat spinal cord nociceptor-driven neurones, Pain, 15 (1983) 71-81. 31 Takagi, H., Satoh, M., Doi, T., Kawasaki, K. and Akaike, A., Indirect and direct depressive effects of morphine on activation of lamina V cell of the spinal dorsal horn i n duced by intra-arterial injection of bradykinin, Arch. Int. Pharmacodvn. Ther., 221 (1976) 96-103. 32 Wall, P. D., The laminar organization of the dorsal horn and effects on descending impulses, J. Physiol. (Lond.), 188 (1967) 403-423. 33 Yaksh, T. L., Direct evidence that spinal serotonin and noradrenaline terminals mediate the spinal antinociceptive effects of morphine in the periaqueductal gray, Brain Research, 160 ( i 979) 180-185. 34 Yaksh, T. L. and Rudy, T, A,, Studies on the direct spinal action of the narcotics in the production of anagesia in the rat, J. Pharmacol. exp. Ther., 202 (1977) 411-428. 35 Yaksh, T. L. and Rudy, T. A., Narcotic analgetics: CNS sites and mechanisms of action as revealed by intracerebral injection techniques, Pain, 4 (1978) 299 359.