Differential excitatory and inhibitory effects of opiates on non-nociceptive and nociceptive neurones in the spinal cord of the cat

Brain Research, 145 (1978) 303-314

303

~ Elsevier/North-Holland BiomedicalPress

D I F F E R E N T I A L EXCITATORY AND INHIBITORY EFFECTS OF OPIATES ON NON-NOCICEPTIVE AND NOCICEPTIVE NEURONES IN THE SPINAL CORD OF THE CAT

G. BELCHER and R. W. RYALL

Department of Pharmacology, University of Cambridge, Cambridge CB2 2QD (Great Britain) (Accepted August 8th, 1977)

SUMMARY

Morphine, levorphanol, dextrorphan and naloxone were applied microelectrophoretically to cells identified as either having nociceptive inputs or non-nociceptive inputs in the dorsal horn of the cat. Morphine excited non-nociceptive cells and depressed nociceptive cells. Naloxone reversed morphine excitations on non-nociceptive cells, but only reversed about one-third of morphine depressions on nociceptive cells. Levorphanol depressed nociceptive ceils, whilst dextrorphan ejected with similar currents caused less depression or had no effect. It is concluded that excitation of nonnociceptive cells may constitute a spinal action relevant to the analgesic action of opiates, acting synergistically with a depressant effect on nociceptive neurones.

INTRODUCTION

It is probable that morphine causes analgesia by actions at supraspinal levels and by modifying transmission in ascending spinal pathways. It has been proposed that the apparent spinal action of morphine may be due to excitation of a descending pathway which inhibits ascending nociceptive transmission42,43, 45 but microinjection47 and intravenous studies with morphine4,9,'~4,27,2s,31,32 have provided behavioural and electrophysiological evidence for a depression of nociception by a direct spinal action. The present study was undertaken in an attempt to define the direct spinal actions of iontophoretically administered morphine on carefully characterized neurones. During the course of this investigation, several laboratories have published many, apparently conflicting, results obtained with the iontophoretic technique. The differences may be due to the use of different anaesthetics, to transaction of the spinal cord, to species differences or to differences in sampling and characterising neurones. Both excitatory and inhibitory effects of morphine have been reported. Morphine

304 excites Renshaw ceils e.lv.ts-:3.~.41 and is antagonized by naloxone ~:~.1~or by dihydro-?erythroidine 2,~8,41, although it is not certain that this is not due to the presynaptic release of acetylcholine2. TM Other interneurones in the rat 1:~ but rarely m the cat ~s were found to be excited by either acetylcholine or morphine. However. 20"',~ of interneurones were excited by morphine m another study in c a r e 6. Depressant actions of morphine have been noted, particularly on nociceptive neurones 8. A depression of amino acid-induced excitation Js reported in several st udie.~'~ 10,15,16,1S,3~),48, but a consistent antagonism of the depression by naloxone was found in only one of these 4s. There is similarly conflicting evidence regarding the relative potencies of the depressant acuon of levor phanol and its analgesically inactive isomer. dextrorphan. Dextrorphan was more potent than levorphanol in one study ~s. eqmpotent in another ~Gor less potent in yet a third4'L Effects of morphine and naloxone on either spontaneous or evoked excitation are also inconsistent. Dostrovsky and Pomeranz ~(~ found that morphine depressed activity of neurones in laminae I. V VI and VII but never in lamina IV. whereas Zeiglgansberger and Bayerl ~s found that lamina IV neurones were also depressed. although larger electrophoretic currents were required. In contrast. Duggan e t a ] . z° could find no direct effect on neurones in laminae IV and V and claim that the primary action of morphine is on cells in substantia gelatinosa. Natoxone rarely antagonized the inhibitory effects of morphine in two studies s,16. but it did in two others "-'°.~s. The reasons for these many discrepancies are not entirely clear but the present study shows significant differences between different, carefully characterized populations of nociceptive and non-nociceptive neurones in their responses to morphine and the reversal of these effects in anaesthetized cats with intact spinal cords, and may provide a partial explanation of some of the discrepancies. The major difference between this and previous studies is that all neurones were carefully characterized by their response to intra-arterially injected bradykinin, which is a more specific noxmus stimulus "a than those used in other iontophoretic studies, in addition, effects have been carefully compared on spontaneous activity, stimulus-evoked and amino acid-induced excitation, because this was fotmd to influence the interpretation of the data. MATERIALS AND METHODS Thirty-seven cats of either sex weighing 2.0-3.8 kg were used in this study. The experiments were performed under chloralose anaesthesia (50-60mg/kg, i.v., sometimes supplemented by an additional 10 mg/kg ) after induction of anaesthesia by halothane. Alter setting up in the animal frame, some animals were paralyzed with intravenous gallamine triethiodide, and mechanically ventilated. Blood pressure was monitored routinely, as was rectal temperature, which was maintained between 37 and 38 ~C, and end-tidal carbon dioxide, which was maintained at approximately 4)~o. A taminec~ tomy from L1 to L7 segments exposed the spinal cord. The S 1 and L7 ipsilateral ventral roots were cut and mounted for stimulation for identification of Renshaw cells and motoneurones. The spinal cord was left intact.

305 Unit recordings from the dorsal horn of L6 and L7 segments were made using 5-barrelled microelectrodes. The centre barrel was filled with 4 M NaC1 for recording cell activity and the outer barrels filled with solutions of morphine hydrochloride (0.07 M), naloxone hydrochloride (0.1 M), DL-homocysteic acid (DLH) (0.2 M, pH 8.0), acetylcholine bromide (0.5 M), levorphanol tartrate (0.1 M or 0.025 M in 0.1 M NaCl), dextrorphan tartrate (0.1 M or 0.025 M in 0.1 M NaCI) or NaCI (4 M). Retaining voltages of 0.5 V were routinely applied to the drug barrels. The effects of electrophoretic ejection of drugs were tested on spontaneous firing of identified cells, amino acid-induced excitation and excitation by natural stimulation in the periphery. Cells were identified as either nociceptive, i.e. fired by the local injection of bradykinin (2.5-15/~g) via a side branch of the femoral artery, or non-nociceptive. Locally applied heat, which was painful when applied to the experimenter's hand, was also employed as a noxious stimulus, but was not used for classification. The skin was locally heated for controlled periods at fixed intervals by a heating coil placed on its surface. The effect of strong mechanical stimulation (pinching) was also examined on some cells. Non-nociceptive cells were driven by stimuli such as hair movement produced by puffs of air, light touch produced by a light vibrating stylus, limb extension, limb flexion or joint movement. Solutions injected into the median vein were morphine hydrochloride (10 mg/ml) and naloxone hydrochloride (2 mg/ml). Cell firing rates were counted with the aid of an amplitude discriminator and rate meter. On-going rates and rates for 5 sec epochs were displayed continuously on a pen-recorder. RESULTS

Neuronal classification of cells studied The criterion for classification of the neurones as nociceptive or non-nociceptive was based mainly upon the presence or absence of excitation by intra-arterially injected bradykinin. It has been suggested 89 that bradykinin might have a direct excitatory action on dorsal horn neurones when administered systemically. This was unlikely to have occurred with the arterial injections of 4-15/~g employed in this study, because in 4 cats the cells, 6 in number, were only excited after intra-arterial injection of 4-15/~g but not after intravenous injection. In many of the experiments the effects of noxious heat or intense mechanical stimulation were also examined, but the response to these types of stimulation were not made the basis of classification because of the possibility of stimulating temperature or mechanoreceptors respectively. All of the cells studied had receptive fields in the lower hind limb. The depth below the cord dorsum was noted, but the sites were not verified histologically. Table I shows at least three distinct populations of neurones with different inputs at different depths. The deepest cells responded only to joint or muscle movement, with no convergence from nociceptors. At an intermediate depth, cells were located which responded only to the noxious stimulus, only to light touch or with convergence from both types of

306 TABLE 1 The depth of dorsal horn interneurones from the surfiwe of the spinal cord and the peripheral stimuli that best fires them

Table shows a and b significantly (P ~- 0.05) more superficial than c, d. e; f significantly (P - 0.05~ deeper than c, d and e. Type of peripheral stimulation that best fires the cell

Number of cells

Averagedepth below the surfi~ce ol the cord z standard error ~mm

a. b. c. d. e. f.

13 58 24 31 24 17

L.620 z: 0.073 1.765 = 0.066 2.167 :a 0.099 2.200 i 0.127 2.312., 0.135 2.651 7 0.071

Hair movement and noxious Hair movement Light touch Noxious Light touch and noxious Joint or muscle movement

stimulation. Neurones responding to hair m o v e m e n t or to hair m o v e m e n t and noxious stimulation were located most superficially. This depth profile probably reflects the lamination of the spinal cord 4°. and is similar to that reported in more detailed studiesa,29,3o,46 Excitatory and inhibitory effects o f morphine on responses to stimulation o f peripheral receptors

The effects o f iontophoretically administered morphine (5-200 nA) were variable and dependent upon the type o f neurone on which it was tested and the means used to activate the cell. Thus, there were differences between nociceptive and non-nociceptlve cells in the effect of morphine and in the ability of naloxone to antagonize the effects. There were also differences depending on whether the neurones were activated by natural stimulation o f their peripheral receptors or by the local iontophoretic administration o f an excitatory amino acid ( D L H ) , or when tests were performed on the background, spontaneous activity. Some o f the differences are summarized in Fig. 1. In a population of 80 non-noclceptive cells (Fig. 1A-C), 50 were excited by morphine (Figs. I A and 2) but only 5 (Fig. 1D) of 43 nociceptive cells (Fig. I D - F ) were similarly excited. A Chi-squared test showed that these population differences were significant (Z2 .... 68. P <: 0.005). In contrast, only 3 non-nociceptive cells were depressed by morphine (Fig. I B) but a larger proportion o f the nociceptive cells (26 cells) were depressed (Figs. IE and 4). Again these population differences were significant (X2 -- 53, P <: 0.005~. Cells responding to hair m o v e m e n t were the most consistently excited (42 of 58 cells), whilst all cells (5 cells) responding only to muscle or joint m o v e m e n t were unaffected. The depressant effects were most frequently encountered on those cells responding only to noxious stimulation or on those responding to noxious stimulation and light touch (Fig. 1E). The excitation often t o o k the f o r m o f an increased tendency of the cells to fire in bursts of high frequency action potentials, even on cells in which such activity was

307 , NON- NOCICEPTIVE

NOCICEPTIVE

Facilitation No of cells

0

No of cells

[]

--

0

~==3 ~

~

~

~)epression

No Effect

o

[] T

~

o

~r-TF-q[--~

14 J Total T=Touch. H = Hair.

H+N T*N N Total J=Joint, N= Noxious,

Fig. I. Histograms of number of cells facilitated (A, D) depressed (B, E) or unaffected (C, F) by the iontophoretic administration of morphine on stimulus-evoked activity of non-nociceptive (A, B, C) and nociceptive (D, E, F) neurones. The type of stimulus which best fired each cellis shown at the bottom of the figure: T, light touch: H, hair movement; J, joint movement; N, noxious stimulation by bradykinin or local heating of the receptive field, or combinations of these.

ol

'°o,,oo'

~

,

• u ~

~

o--~--497 47

7> ~-- 45-

i

ill

"ltl

rain

lev 20

~o

,lev 20,

4'0 p

dex 30

i

~4341, 39

io

mill

io

Fig. 2. Upper record: excitation of a non-nociceptive interneurone by morphine (MOR) and its antagonism by naloxone (HAL). Lower record: levorphanol (LEV)-excited interneurone which was not excited by dextrorphan (DEX). The test responses were evoked by hair movement. Frequency of firing (spikes/sec) is shown on the ordinate scale. Electrophoretic currents are in nA.

308 TABLE II Effects of iontophoretically applied levorphanol and dextrorphan on the firing o]'cetls in the cut dorsal hort~ Type of cell

Drug and current

Non-nociceptive levorphanol (fired by hair 2(~100 nA movement or touch) dextrorphan 10-50 nA Nociceptive (fired by levorphanol heat or bradykinin) 20-50 nA dextrorphan 20-50 nA

No effect

No. of cells excited

No. o! cells depressed

Total

4

2

4

l0

7

0

3

tO

3

0

16

J9

~

0

II

]9

not observed prior to the administration of morphine. The depression caused by morphine was variable in extent but was usually within the range of 30-80 ~ reduction in the control rate of firing, and effects were evident in less than 2 min from the commencement of administration. In 14 of 26 cells, the spontaneous rate was unaffected by morphine, even though evoked responses were reduced or, on 5 other cells, increased. It has been reported la that morphine-induced excitation of uncharacterized dorsal horn interneurones in the rat is correlated with an excitatory action of acetylcholine. However, in tests on 18 non-nociceptive neurones excited by morphine in the cat in the present study, none were excited by acetylcholine, as found in other studies in catslZ,21. Results obtained with intravenous administration of 2.5 mg/kg in 8 cats were similar to those obtained with microelectrophoretic administration in that the response to a noxious input was reduced in 4 of 5 cells but there was no effect of 3 non-nociceptive cells. The effects o f levorphanol and dextrorphan

Excitatory effects of levorphanot (30 nA) were seen on only two of 10 non-nociceptive neurones on which it was tested (Table II), but on neither of these did dextrorphan, ejected with the same electrophoretic current, produce any effect. No nocicepuve neurones were excited by either levorphanol or dextrorphan. Levorphanol and dextrorphan (20-50 nA) were compared on 19 nociceptive neurones. On three, neither drug had any effect. On 5 cells, levorphanol depressed the response to noxious stimulation whereas dextrorphan did not (Fig. 3) and on two cells the depressant effect of levorphanol was greater than that of dextrorphan. On 9 cells the drugs were equieffective. Since dextrorphan is inactive as an analgesic 22, it is clear that some of the depressant effects of the opiates must be due to an action which is not explicable in terms of a specific interaction with an opiate receptor, a conclusion which was supported by the experiments with the antagonist, naloxone, and has been confirmed by other investigators16, is. Nevertheless, it is of interest that a larger proportion (60 ~o) of noci-

309

4o-

I

dex 30

I

~_ 20-

!

lev 30

f

3~ rain

Fig. 3. The effect of dextrorphan (30 nA) and levorphanol (30 nA) on the maximum firing rate (spikes/ sec) of a nociceptive dorsal horn interneurone excited by 2.5 ttg of bradykinin injected into the blood supply of its peripheral receptive field.

ceptive cells than of non-nociceptive cells (30 ~) appeared depressed by the two isomers, although this difference was not statistically significant (P > 0.05). It was also noted that levorphanol and dextrorphan often produced a broadening of the action potentials and a decrease in the amplitude, an effect which was rarely seen with morphine. Such cells were excluded from the analysis in Table II. The effects o f morphine on D L H - e v o k e d activity

The action of morphine, administered iontophoretically with currents of 10-95 nA, was examined on 34 non-nociceptive and nociceptive neurones in which excitation was evoked by the iontophoretic application of the amino acid, D L H . The results are summarized in Table III. The chief difference between these results and those obtained in which the test response was activity evoked by stimulation of peripheral receptors (Fig. 1) is that the incidence of excitation by morphine of non-nociceptive units was much less and the incidence of morphine-evoked depression of non-nociceptive cells was TABLE lII

Effects of iontophoretically applied morphine on firing of cells to DLH in the feline dorsal horn Type of cell

Current of morphine

No. of cells deNo. of cells expressed (average cited (average depression* ± S.E.) excitations* ± S.E.)

No. of cells unaffected

Totals

Non-nociceptive

10-60 nA (average 40 nA) 10-95 nA (average 29 nA)

14 (29 -k 6~)

4 (23 ± 8~)

1

19

15 (40 ± 4~)

0

0

15

Nociceptive

* Average depression or excitation is the change in firing rate expressed as a percentage of the control rate.

310 much greater. Nevertheless, the average depression of non-nociceptive unit.~ was less (P < 0.025, 't' :-: 2.15) than that of' nociceptive units (Table II1), as observed when the peripheral receptors were activated.

The action of naloxone The excitatory effects of morphine were more consistently antagonized by the iontophoretic administration of naloxone (10-100 nA) than were depressant effects. A reversal o f the excitatory effect on responses to activation o f peripheral receptors was obtained in tests on 20 of 29 non-nociceptive neurones: an example ix illustrated in Fig. 2. In contrast, it was only possible to antagonize the depressant action of morphine (Fig. 4) on 5 of 13, and of levorphanol on 1 o f 4. nociceptive neurones examined, and only once was the reversal complete. Complete reversal was observed on two cells after intravenous injection o f 0.5 mg/kg of naloxone. In the absence o f an opiate agonist, iontophoretic naloxone usually had no effect (7 non-nociceptive and 3 noclceptive cells). Three non-nociceptive cells were excited and on two (one non-nociceptive and the other noclceptive) there was depression. In experiments in which D L H was used as an excitant, morphine-evoked de-

I0 sec

t

t

sal

t

brady

pinch

i

mot i

nsI 70nA

minutes Fig. 4. The effect of microelectrophoretically applied morphine (MOR) and naloxone (NAL) on the activity (spikes/sec) of a dorsal horn interneurone fired by injection of bradykinin (BRADY, 8 t¢g) locally into the blood supply of the peripheral receptive field of the cell (bottom record). The ordinate shows the maximum activity in spikes/sec obtained after injections of bradykinin. The top record shows a ratemeter recording (spikes/sec) of the activity of the celt and the effects of local injection of saline (SAL, 0.1 ml) or bradykinin (8/tg in 0. l ml saline) and pinch by serrated forceps in the receptive field of the cell.

311 pression was antagonized by naloxone on only two nociceptive units. On 5 others, the depressant action of morphine was potentiated, an effect never observed when cells were excited by peripheral stimulation. Naloxone had no effect on one cell. Depression of amino acid-evoked excitation by morphine was also potentiated on three non-nociceptive cells and potentiation of excitatory effects of morphine occurred on two cells. On only one non-nociceptive neurone did naloxone have no effect. Thus, it was rarely possible to antagonize the depression ofneurones by morphine, especially when DLH was used as an excitant, but antagonism of excitatory responses was consistently obtained when natural stimulation, but not when DLH-evoked excitation, was employed as a test response. DISCUSSION The conflicting data concerning the effects of iontophoretically administered morphine (see Introduction) may be due to a number of differences in experimental technique. In this study in anaesthetized cats with intact spinal cords it is clear that there are different responses in different populations of neurones, classified by the convergence of peripheral inputs. The noxious stimulus used to classify the cells as nociceptive may have considerable relevance to the outcome of the study, as may the type of activity on which the drugs are tested. Bradykinin was used as the noxious stimulus for classification of cells in this investigation because it is known to be a potent analgesic agent 1,3, 33,z4 and probably acts more selectively on nociceptors 26 than do other stimuli such as heat, which may be expected to activate thermoreceptors, strong pressure or pinching which would also excite many mechanoreceptors. In other studies in spinal animals, the convergence from peripheral receptors onto spinal neurones may well have been different from that in the present investigation. The most important new finding to emerge from this study is that dorsal horn neurones which do not receive a nociceptive input are generally excited by morphine. Although other investigators have occasionally noted excitation 14,15,1s this has not been correlated with an absence of nociceptive convergence. Nociceptive neurones were generally inhibited by morphine as reported previously 8,25,37,48. The excitatory effect on non-nociceptive neurones was of particular interest because it was consistently antagonized by naloxone, in marked contrast to the inconsistency of antagonism of inhibitory actions on nociceptive cells, as found by others s,16. This antagonism of the excitatory effect of morphine is rarely found in other parts of the central nervous systemZ,7, 44, although naloxone apparently reverses morphine excitation of midbrain neurones '~3 and of Renshaw cells in cats13,18, and of cholinoceptive dorsal horn neurones in rats ~3. In agreement with other investigators 8,19intravenous naloxone was more effective in antagonizing depressant effects of morphine than was iontophoretic administration. The reason for this is not entirely clear 8, but may be related to the unpredictable release of naloxone by iontophoresis or to differences in distribution to different parts of the cell when administered by the two methods. The failure to antagonize all of the effects of morphine by naloxone is not neces-

312 sarily an indication that the effect is unrelated to analgesia. It has been suggested :38 that the opiate receptor may exist in either of two configurations, one of which has a high affinity for agonists and the other which has a high affinity for antagonists, the relative proportions of which may vary. However. this seems an unlikely explanation because dextrorphan, which is inactive as an analgesic, was equieffective with levorphanol as a depressant on about half of the neurones tested, although it was less effective on the remainder. The fact thal the depressant action was relatively specific for nociceptive cells indicates that the mechanism was unrelated to a non-specific depression of postsynaptic excitability, as with local anaesthetics ~z. but does not exclude a possible presynaptic action if the presynaptic, nociceptive input is carried by fibres which are smaller in size than those conveying non-nociceptwe impulses. The distinction between nociceptive and non-nociceptive neurones in their responses to morphine was clearly evident when the drug was tested upon responses to 'natural' stimulation of receptors, but was less evident upon spontaneous activity or activity evoked by the iontophoresis of D L H On amino acid-evoked excitation the differentiation was partly obscured by non-specific depressant effects, as evidenced by the inability of naloxone to cause reversal and by the fact that naloxone even potentiated the depressant action of morphine: this could be due to a partial agomst effect of naloxone 6. Nevertheless, even though the incidence of excitation of non-nociceptive neurones by morphine was comparatively low. the depression of amino acid-evoked firing was greater on nociceptive than on non-nocice ptive cells. In general, it is concluded that little useful information regarding the selectivity of action of opiate drugs could be derived from the experiments in which an amino acid was used to excite the cells. In summary, it is concluded that the most consistent action of morphine in the spinal cord which can be attributed to an interaction with specific opiate receptors is an excitatory effect on non-nociceptive neurones. The greater incidence of depression of nociceptive cells may also be relevant to analgesia but is more difficult to interpret. The conclusion that analgesic action should be correlated with an effect on nonnociceptive cells may at first seem a little paradoxical. It is not so problematical, however, if considered in relation to the "gate theory' of Melzack and Wall 36. in which increased activity in non-nociceptive pathways decreases activity in pain pathways. The reciprocal excitatory and depressant actions may then be seen as synergistic in the reduction of pain. ACKNOWLEDGEMENTS This study was supported by grants from the Medical Research Council and the Wellcome Trust. G. Belcher is an M R C Scholar. Bradykinin was a gift o f Sandoz (Basel), naloxone was kindly supplied by Endo Laboratories (New York), and Dextrorphan tartrate and levorphanol tartrate by Roche Products (U,K.).

313 REFERENCES 1 Armstrong, D., Dry, R. M. L., Keele, C. A. and Markham, J. W., Observations on chemical excitants of cutaneous pain in man, J. Physiol. (Lond.), 120 (1953) 326-351. 2 Belcher, G. and Ryall, R. W., Substance P and Renshaw cells: a new concept of inhibitory synaptic interactions, J. Physiol. (Lond.) (1977) in Press. 3 Besson, J. M., Conseiller, C., Hamann, K. F. and Maillard, M. C., Modifications of dorsal horn cell activities in the spinal cord after intra-arterial injection of bradykinin, J. Physiol. (Lond.), 221 (1972) 189-205. 4 Besson, J. M., Wyon-Maillard, M. C., Benoist, J. M., Conseiller, C. and Hamann, K. F., Effects of phenoperidine on lamina V cells in the cat dorsal horn, J. Pharmacol. exp. Ther., 187 (1973) 239-245. 5 Bioulac, B., Lund, J. P. and Pull, E., Morphine excitation in the cerebral cortex, Canad. J. Physiol. Pharmacol., 53 (1975) 683-687. 6 Blumberg, H. and Drayton, H. B., Naloxone and related compounds. In H. W. Kosterlitz, H. O. J. Collier and J. E. Villarreal, (Eds.), Agonist and Antagonist Actions of Narcotic Analgesic Drugs, Macmillan, London, 1972, pp. 110-119. 7 Bramwell, G. J. and Bradley, P. B., Actions and interactions of narcotic agonists and antagonists on brain stem neurones, Brain Research, 73 (1974) 167 170. 8 Calvillo, O., Henry, J. L. and Neuman, R. S., Effects of morphine and naloxone on dorsal horn neurones in the cat, Canad. J. Physiol. Pharmacol., 52 (1974) 1207 1211. 9 Conseiller, C., Menetrey, C., Le Bars, D. et Besson, J. M., Effet de la morphine sur los activit~s des interneurones de la couche V de Rexed de la corne dorsale chez le chat spinal, J. Physiol. (Paris), 65 (1972) 220-221A. 10 Curtis, D. R. and Duggan, A. W., The depression of spinal inhibition by morphine, Agents Actions, 1 (1969) 14-19. 11 Curtis, D. R. and Phillis, J. W., The action of procaine and atropine on spinal neurones, J. Physiol. (Lond.), 153 (1960)17-34. 12 Curtis, D. R., Ryall, R. W. and Watkins, J. C., The action of cholinomimetics on spinal interneurones, Exp. Brain Res., 2 (1966) 97-106. 13 Davies, J., Effects of morphine and naloxone on Renshaw cells and spinal interneurones in morphine dependent and non-dependent rats, Brain Research, 113 (1976) 31 I 326. 14 Davies, J. and Duggan, A. W., Opiate agonist-antagonist effects on Renshaw cells and spinal interneurones, Nature (Lond.), 250 (1974) 70-71. 15 Dostrovsky, J. and Pomeranz, B., Morphine blockade of amino acid putative transmitters on cat spinal cord sensory interneurones, Nature New Biol., 246 (1973) 222-224. 16 Dostrovsky, J. O. and Pomeranz, B., Interaction of iontophoretically applied morphine with responses of interneurones in cat spinal cord, Exp. Neurol., 52 0976) 325-338. 17 Duggan, A. W. and Curtis, D. R., Morphine and the synaptic activation of Renshaw cells, Neuropharmaeology, I 1 (1972) 189-196. 18 Duggan, A. W., Davies, J. and Hall, J. G., Effects of opiate agonists and antagonists on central neurones of the cat, J. Pharmacol. exp. Ther., 196 (1976) 107-120. 19 Duggan, A. W. and Hall, J. G., Morphine, naloxone and the responses of medial thalamic neurones of the cat, Brain Research, 122 (1977) 49-57. 20 Duggan, A. W., Hall, J. G. and Headley, P. M., Morphine, enkephalin and the substantia gelatinosa, Nature (Land.), 264 (1976) 456458. 21 Engberg, I. and Ryall, R. W., The inhibitory action of noradrenaline and other monoamines on spinal neurones, J. Physiol. (Lond.), 185 (1966) 298-322. 22 Fr•mherz• K.• Pharmac•••gica• effects •f 3-hydr•xy-N-methy•m•rphinan• Arch. int. Pharmac•dyn.• 83 (1951) 387 398. 23 Gent, J. P. and Wolstencroft, J. H., Actions of morphine, enkephaIin and endorphin on single neurones in the brain stem, including the raphe and periaqueductal gray of the cat. In H. W. Kosterlitz (Ed.), Opiates and Endogenous Opioid Peptides, Elsevier/North-Holland Biomedical Press, Amsterdam, 1976, pp. 21%224. 24 Grossman, W. and Jurna, I., Depression by morphine of activity in the ventrolateral tract evoked from cutaneous A-fibres, Europ. J. Pharmacol., 29 (1974) 171-174. 25 Henry, J. L. and Neuman, R. S., Morphine depression of dorsal horn neurones in the cat, Proc. Canad. Fed. biol. Soe., 17 (1974) 158.

314 26 lggo, A., Pain Receptors. In J. J. Bonica. P. Procacci and ( . A. Pagnl (Eds.7, Receipt Aduances o, Pain, Charles C. Thomas, Illinois. 1976. pp. 3 35. 27 lwata, N. and Sakai, Y., Effects of fentanyl upon the spinal interneurones activated by A~ afi'ere~t fibres of the cutaneous nerves of the cat. Jap. J. Pharmaeol.. 21 (1971) 413-416. 28 Kitahata, L. M.. Kosaka. Y.. Taub. A.. Bonikos. K. and Hoffert, M., Lamina specific suppressJo~ of dorsal horn activity by morphine sulfate. Anesthesiology, 41 (1974) 39-48. 29 Kitahata, L. M., Taub, A. and Kosaka. Y.. Lamina specific suppression of dorsal horn unit activity by ketamine hydrochloride, Anesthesiology, 38 (1973) 4- 11. 30 Kitahata, L. M., Taub, A and Sato. I.. Lamina specific suppression of dorsal horn unit actiwty by nitrous oxide and by hyperventilation. J. Pharmacol. exp. Ther.. 176 (1971) 101 108. 31 Le Bars, D., Guilbaud, G., Jurna, !. and Besson, J. M.. Differential effects of morphine on responses of dorsal horn lamina V type cells elicited by A and C fibre stimulation in the spinal cat. Brain Research, 115 (I 976) 518- 524 32 Le Bars, D., Menetrey, C.. Conseiller. C. and Besson. J. M.. Depressive effects of morphine upon lamina V cells' activities in the dorsal horn of the spinal cat, Brain Research. 98 (1975) 261-277. 33 Lira, R. K. S., Neuropharmacology of Pain and Analgesia. tn R. K. S. Lim, D. Armstrong and E. G. Pardo (Eds.), Proc. Third Int. Pharmacol. Meeting, Vol. 9 Pharmacology o/Pain, Pergamon Press, Oxford, 1968, pp. 169 219. 34 Lira, R. K. S., Krauthamer. G.. Guzman, F. and Ftup, R. R.. Central nervous system activity associated with pain evoked by bradykinin and its alteration by morphine and aspirin. Proe. nat. Acad. Sci. (Wash.), 63 (1969) 705-712. 35 Lodge, D., Headley, P. M.. Duggan, A. W. and Biscoe. T. J.. The effects of morph the, etorphine and sinomenine on the chemical sensitivity and synaptic responses of Renshaw cells and other spinal neurones in the rat, Europ. J. Pharmacol.. 26 (1974) 277-284. 36 Melzack, R. and Wall, P. D.. Pain mechanisms: a new theory, Science. 150 (19657 97t--975 37 Neuman, R. S., Calvilto. O. and Henry, J. L.. Blockade by morphine of nocicepti~e neurones in the dorsal horn of the cat. The Pharmacologist, 16 (1974) 203. 38 Pert, C. B. and Snyder, S. H., Opiate receptor binding of agonists and antagonists affected differentially by sodium, Molec. Pharmacol.. 10 (1974) 868-879. 39 Randic, M. and Yu, H. H., Effects of 5-hydroxytryptamine and bradykinin in cat dorsal horn neurones activated by noxious stimuli. Brain Research. 1 I 1 (1976) 19"/203. 40 Rexed, B., The cytoarchitectonic organisation of the spinal cord of the cat, J. comp. NeuroL. 96 (1952) 415 495. 41 Ryatl, R. W. and Belcher, G.. Substance P selectively blocks nicotinic receptors on Renshaw cells: a possible synaptic inhibitory mechanism. Brain Research. 137 (1977) 376-380. 42 Satoh, M., Nakamura, N. and Takagi, H.. Effects of morphine on bradykinin induced unitary discharges in the spinal cord of the rabbit, Europ. J. Pharmacol., 16 (19717 245-247. 43 Satoh, M. and Takagi, H.. Effects of morphine on the pre- and post-synaptic inhibitions in the spinal cord, Europ. J. PharmacoL. 14 (1970) 150 154. 44 Satoh, M., Zieglgansberger, W.. Fries. W. and Herz. A.. Opiate agonist-antagonist interactions at cortical neurones of naive and tolerant/dependent rats. Brain Research. 82 ~1974) 378-382. 45 Takagi, H., Matsunara, M.. Yanai. A. and Ogiu, K.. The effect of analgesics on the spinal cord reflex activity of the eat, Jap. J. PharmacoL. 4 (1955) 176-187. 46 Wall, P. D., The laminar organisation of the dorsal horn and the effects of descending impulses, J. Physiol. (Lond.), 188 (1967) 403-423. 47 Yaksh, T. L. and Rudy, T. A., Analgesia mediated by a direct spinal action of narcotics, Science, 192 (1976) 1357 1358. 48 Zeiglgansberger, W. and Bayerl, H.. The mechanism of inhibition of neuronal activity by opiates in the spinal cord of cat, Brain Research, 115 (1976) 111 128.