- Email: [email protected]
Interactions of morphine with putative neurotransmitters in the mesencephalic reticular formation
Life Sciences, Vol. 32, pp. 759-769 Printed in the U.S.A.
INTERACTIONS
Pergamon Press
OF MORPHINE WITH PUTATIVE NEUROTRANSI41TTERS MESENCEPHALIC RETICULAR FORMATION
IN THE
Henry J. Haigler* and Timothy P. O'Neill Department of Pharmacology Emory University School of Medicine Atlanta, Georgia 30322 (Received in final form Novemher 2, 1982) Summary Acetylcholine (ACh) and norepinephrine (NE) have been identified previously as putative nociceptive neurotransmitters in the mesencephalic reticular formation (MRF) of the rat because they frequently mimic the change in neuronal firing (usually an increase) evoked by a noxious stimulus (NS). The purpose of this study was to determine if i.) morphine (M) acts to prevent the increase in firing evoked by a NS by blocking the effects of either of these two neurotransmitters and 2.) if this effect is a specific narcotic effect. Using the technique of microiontophoresis in conjunction with extracellular recording, we located single units in the MRF in which i.) neuronal firing was accelerated by a NS: 2.) M blocked this response; and 3.) either ACh or NE mimicked the effect of the NS. Neurons meeting these three criteria were studied further to determine if morphine would also block the response to either of the neurotransmitters and if this was a specific narcotic effect. We found that morphine blocked the increase in neuronal firing evoked by the NS and ACh or the NS and NE in over 50% of the cells meeting the above criteria. Some neurons were found in which both ACh and NE mimicked the NS and M blocked all three responses. This blockade of these neurotransmitters was a specific narcotic effect because it could be reversed by the systemic administration of naloxone. These data lead to the tentative hypothesis that M, acting via an opiate receptor, blocks the increase in neuronal firing evoked by a NS by blocking the postsynaptic effects of either ACh or NE. This may be one of the mechanisms by which morphine acts to produce analgesia. The mesencephalic reticular formation (MRF) of the rat is apparently an area where morphine acts to produce analgesia. When morphine is injected directly into this area it produces an analgesic effect (i). When morphine is administered microiontophoretically in the MRF, it blocks the increase in neuronal firing evoked by a nociceptive stimulus (noxious foot pinch; NS)(2). Both of these are specific narcotic effects (1,3). Therefore the MRF is an area where the mechanism of action of morphine can be studied.
*Current address: Section Head, CNS Pharmacology, Department of Biological Research, Searle Research and Development, Div. of G.D. Searle and Co., 4901 Searle Parkway, SkDkie, IL 60077. 0024-3205/83/070759-11503.00/0 Copyright (c) 1983 Pergamon Press Ltd.
760
Morphine Interactions With Neurotransmitters
Vol. 32, No. 7, 1983
Morphine may produce its analgesic effect by interfering with cholinergic or noradrenergic transmission in the brain. Both acetylcholine (ACh) and norepinephrine (NE) frequently mimic the response to the NS in the MRF and therefore have been tentatively identified as nociceptive neurotransmitters (3). More recent evidence that NE is a neurotransmitter mediating nociception in the MRF is that phentolamine, a relatively specific a-adrenoceptor antagonist, blocks both the response to the NS and the response to NE (4). A hypothesis based on the above data is that morphine produces its analgesic effect by interfering with noradrenergic and cholinergic transmission in the MRF. Morphine may act to interfere with the release of either NE or ACh from the presyneptic terminals or block their effects on the postsynaptic neuron. Obviously, there are other possibilities. To determine if morphine interacts with either of these neurotransmitters and how it might interact with them, we carried out the following experiments using the technique of microiontophoresis. The neurons of primary interest in this study were those in which both the NS and one of the putative neurotransmitters evoked an increase in firing. We report here that when morphine blocks the response to a NS, it typically blocks the response to NE, ACh or both. This effect was a specific narcotic effect because it could be reversed by naloxone administered systemically.
METHODS A total of 71 male Sprague-Dawley rats (Charles Rivers Laboratories, Wilmington, Mass) weighing between 220-330 g, were used. All animals were initially anesthetized with chloral hydrate (400 mg/kg) and kept anesthetized throughout the experiment by additional i.v. injections of chloral hydrate administration through a tail vein. Lidocaine (Xylocaine) jelly was used to coat all wDund edges and the earbars and mouthpiece of the stereotaxio apparatus. The body temperature of all animals was maintained between 36-38°C. Microiontophoresls: Five barrel glass mlcropipettes were preloaded with a few strands of fiberglass in each barrel prior to "pulling". The tips of the pipettes were broken back to 5 microns under microscopic control; solutions were then injected directly into the various barrels. The central (recording) barrel was filled with 2M NaCI solution saturated with fast green. One side barrel, the balance barrel, was filled with 4 M NaCI. The remaining three barrels were filled with one of the following drugs: acetylcholine chloride (0.2M, pH 4.2, Sigma); l-norepinephrlne HCI (0.2M pH 4.2 Aldrich); and morphine sulfate (0.05M, pH 4.2, Merck). Circuitry for the iontophoretic ejection of drugs and current balance was similar to that previously described (5), except a solid state system was used. The current balance was kept to within ± 1.0 nA. Impedances, measured at i000 Hz, were 2 to 4 megohms in the recording barrels; 9-11 megohms in the balance barrels; and 12-17 megohms in the drug barrels. The retaining current for all drugs was 5 nA. The duration of ejection for all drugs was 1-2 min. Current controls indicated that the ejection currents had no effects by themselves. Additional evidence that the ejection current had no effect on its own is that in cells in which one putative neurotransmltter was ineffective an ejection current as high as i00 nA was tested, typically without effect. Cells usually responded to ejection currents in range between 20-60 nA.
Vol.
32, No.
7, 1983
Morphine
Interactions
With N e u r o t r a n s m i t t e r s
761
N o c i c e p t i v e Stimulus. The NS used in this study was pressure d e l i v e r e d to the left hind foot by an analgesy m e t e r (Stoelting, no. 51302) using a method m o d i f i e d from (6). This protocol has been described in detail p r e v i o u s l y by H a i g l e r and Spring (3). Briefly, the foot was trapped on a pedestal w i t h a blunt, tapered stylus at 90 ibs/in 2. After I0 seconds the pressure was increased to 540 ibs/in 2 over a 12 second period. This pressure was then m a i n t a i n e d for 30 seconds before the foot was released. Protocol: When m o r p h i n e was a d m i n i s t e r e d m i c r o i o n t o p h o r e t i c a l l y , the ejection of m o r p h i n e began 20 seconds before either placing the foot in the analgesy meter or the iontophoresis of either NE or ACh. In those experiments in which naloxone was tested to d e t e r m i n e if it would reverse the effects of morphine, it was injected i.v. via a tail vein; the dose was calculated on the basis of the free base. Data Analysis. A computer was used to q u i c k l y determine if a drug or the NS had produced a significant effect on spontaneous firing in the following way. A stable baseline was present when there was no significant d i f f e r e n c e between the firing rates of two c o n s e c u t i v e 10-second intervals based on a one-way analysis of variance (ANOVA). Once a stable baseline was established, the m e a n and variance of the last I0 second interval was stored in the computer m e m o r y as the control period. Immediately after the beginning of m i c r o iontophoretic drug ejection or nociceptive stimulation, the firing rate over the next five 10-second periods (test period) was stored in the computer memory. If there were indications that the firing rate was not going to r e t u r n to control quickly, an additional 50 seconds of test data was collected; this procedure was typically n e c e s s a r y in the case of NE. In those cases in which two c o n s e c u t i v e 50 sec intervals were collected, both were compared to the control using a o n e - w a y ANOVA. The null hypothesis of no change in firing rate was rejected if there was a significant change in firing rate at the 0.01 level (F > 3.46, w i t h 5 over 54 degrees of freedom). The control mean firing rate was always determined immediately prior to each e x p e r i m e n t a l test (i.e., drug ejection or the a p p l i c a t i o n of the NS). It is possible that two i0 second intervals are not an adequate estimate in the basal neuronal firing rate. Therefore, if there were any apparent changes in the basal firing rate on the analog record, another control rate was obtained. There were m u l t i p l e control rates obtained during an experiment; each basal mean firing rate was typically ± 2 of the initial basal firing rate throughout the experiment. Furthermore, when one n e u r o t r a n s m i t t e r had no effect but another did, the "no effect" response served as a control. If a n e u r o t r a n s m i t t e r did not produce an effect a 20-60 nA, it was typically tested at i00 n A to make sure that it would not have an effect. We typically studied a n e u r o n of interest for 1.5 to 4.5 hours d u r i n g which time the basal firing rate r a r e l y changed. The response to the NS and the n e u r o t r a n s m i t t e r s remained stable throughout the experiment in the absence of any other drug. When it was n e c e s s a r y to administer additional chloral hydrate anesthesia, the baseline and the evoked r e s p o n s e to the NS and one of the n e u r o t r a n s m i t t e r s was retested. Any d e p r e s s a n t effects of the supplemental dose of chloral hydrate (30-60 m g / k g ) were transient and did n o t depress the cell for longer than 3-5 min. Criteria. The f o l l o w i n g are criteria that we established for the neurons that would be studied extensively. First, neurons had to be affected by the noxious stimulus. Second, m o r p h i n e had to block this response. Third, either NE or ACh had to mimic the r e s p o n s e to the NS (c.f. Fig. 2 and 3, ref. 3 for an illustration of h o w ACh and NE m i m i c k e d the r e s p o n s e to the NS). If the cell could be held long enough, naloxone was injected i.v. to d e t e r m i n e if it would reverse m o r p h i n e ' s blockade of the NS or the two neurotransmitters.
762
Morphine Interactions With Neurotransmltters
Vol. 32, No. 7, 1983
Histology. Recording sites were marked by passing a 25 microamp cathodal current through the barrel for i0 min. This resulted in the deposition of fast green in a discrete spot (7) with a usual diameter of 40 microns. Animals were then perfused and the histological localization for each recording site was determined as previously described (8).
RESULTS ACh mimicked the response to the NS in 27 of 47 (57%) of the neurons studied which is in good agreement with a previous study (3). In the 27 neurons in which ACh mimicked the response to the NS, morphine blocked or attenuated the response to both ACH and the NS in 15 of 27 (56%) of these neurons (Fig. i). NE mimicked the response to the NS in 28 of 43 (65%)
P
ACh
MS
60
MS
40
0 tu ¢0 ¢0 UJ O.
40
ACh
ACh
60
NE
MS
40
NE
MS 40
40
NE
MS
60
40 P
60
40
NE
MS
60
40
ACh
v
60
T
15] o
.Ji
MS NE
. . . . . . .
A
. . . . . . .
". . . . . . . . . . . . . . . . .
40
MS
60
40
......
,,,
MS
40
ACh
60
T
I
2 MIN
I
FIG. i Morphine administered microiontophoretically attenuates the response to the NS (p), acetylchollne (ACh) and norepinephrlne (NE). Mlcroiontophoretically administered acetylcholine at 60 nA (ACh 60) and norepinephrine at 60 nA (NE 60) produce a response that is similar to that produced by the NS. The NS is foot pressure, the arrows beneath the horizontal llne labeled 'p' indicate an increase in pressure from 90-540 ibs/in 2. The horizontal bar underneath ACh, NE and MS (for morphine sulfate) indicates the duration of the ejection current; the number indicates the ejection current in nA. The ordinate represents the integrated rate of firing in spikes/sec. The three arrows (third and fourth trace) mark the point at which naloxone was administered; each arrow indicates the administration of a dose of I mg/kg.
Vol. 32, No. 7, 1983
Morphine Interactions With Neurotransmitters
763
of the neurons studied• In a previous study (3) NE mimicked the response to the NS in 67% of the neurons studied. In these neurons morphine blocked or attenuated response to both NE and the NS in 19 of 28 (68%) neurons (Fig. I). In eight of the neurons that are described above, morphine blocked all three responses; the response to the NS, the response to ACh and the response to NE (Fig. I). Evidence that morphine did not produce a non-specific depressant effect on neuronal firing is that it did not depress the increase in spontaneous firing which was apparent after the administration of naloxone (Fig. i). The increase in spontaneous firing after the second and third dose of naloxone is not readily explained, but could arise from the transient blockade of an endogenous opioid that modulates the activity of this neuron. Note, however, that the firing rate was allowed to return to control before retesting morphine against ACh. A crucial question is whether or not morphine's ability to block the response to NE and Ach is a specific narcotic effect. If it is, naloxone should reverse the effects of morphine. Naloxone reversed the ability of morphine to block the increase in firing produced by ACh and NE (Fig. i and Fig. 2). In 4 of 5 cells naloxone reversed morphine's ability to block
/
20 A: N S
o
C: ACh
/ o
~
~o
,
o
1o
2o
30
40
D U R A T I O N OF T E S T ( S E C )
5o
21o 5o 7o D U R A T I O N OF T E S T
80 (SEC)
100
10
2o
3o
4o
50
D U R A T I O N OF T E S T ( S E C )
FIG. 2 A plot of the mean firing rates with respect to time from Fig. i showing that naloxone reverses morphine's ability to attenuate the response to NE and ACh in a dose-related way. A. The response to the NS (O) produces a mean firing rate of 21-22 spikes per second. B. NE (O) at 60 nA produces a similar increase with a slower onset; note that the abscissa is from 0 to i00 seconds for NE to show the complete response• This is necessary because the latency of the response to NE is longer than either that for the NS or for ACh. C. ACh (O) at 60 nA produces a peak response that is slightly less than that to the NS;, but there is no significant difference between the two peak effects (Student's t-test; p>O.l). In the presence of morphine (Q) the response to the NS, NE and ACh is reduced to approximately 50% of control. After naloxone (I mg/kg), the curve for NE begins to recover but morphine's blockade of ACh is apparently enhanced ( • ) . After a second dose of naloxone ( I mg/kg; ~:) the NE curve returns close to the response that was apparent in the absence of morphine as does the ACh curve• After a third dose of naloxone (cumulative dose, 3 mg/kg; ~ ) the ACh curve is almost indistinguishable from the control and may actually show a slight enhancement.
764
Morphine
Interactions
With Neurotransmitters
A 1610
A 350
A 1020
A 160
P 100
Vol.
32, No.
7, 1983
A 620
P 300
FIG. 3 Histological location of recording sites. The location of recording sites in which morphine blocked the response to the NS and NE are marked with filled squares ( • ) . Recording sites in which morphine blocked the response to the NS and ACh are marked by triangles ( • ); those in which morphine blocked the response to the NS, ACh and NE are marked by asterisks ( 0 ) . The recording sites at which morphine blocked the response to the NS but not to ACh nor NE are marked by a clear circle (O). Recording sites where M blocked the response to the NS but where the response is not mimicked by either ACh or NE are not shown although the sites were also in the MRF. The numbers under each section are the distances of the presented section from a frontal section taken in the zero plane and corresponds to the legend in the atlas by K~nig and Klippel (9); MRF is mesencephalic reticular formation and PAG is periaqueductal gray. All but two of the neurons in which morphine blocked both the NS and either NE or ACh are in the MRF according to the KSnig and Klippel (9) atlas. response to the NS and the response to ACh. In two cells naloxone reversed morphine's effect on the NS and NE. In two additional cells naloxone reversed morphine's blocking effect on NE and ACh but the cell was lost ~ h e n retestlng the NS. Because naloxone can reverse the ability of morphine to attenuate or block the increase in firing produced by NE or ACh as well as that produced by the NS it is apparently a specific narcotic effect. Another line of evidence that morphine did not produce a non-specific depressant effect is that there were 13 cells in which morphine did not
Vol. 32, No. 7, 1983
Morphine Interactions With Neurotransmitters
765
block the response to the NS. In these cells, NE mimicked the response to the NS in 5 (39~) but morphine blocked the response to NE in only I of 5 (20~). Similarly, there were 13 other cells in which morphine did not block the response to the NS. ACh mimicked the response to the NS in 5 (39~) and morphine blocked the response to ACh in only i of 5 (20~). The congruence of these two sets of data is coincidental. The neurons in which morphine blocked both the NS and the response to NE and ACh were in the MRF (Fig. 3). There is a cluster of recording sites at A 160 and P i00 where morphine blocks both the NS and the response to NE. There is also an apparent organization of cholinoceptive neurons because 7 of 8 neurons in which morphine blocked the NS and ACh are clustered together at A 620. There is no apparent organization of the neurons in which morphine apparently blocked the NS and both ACh and NE. In 14 of 40 (35%) neurons, morphine blocked the response to the NS but did not block the response to either NE or ACh. However, 7 of these 14 were all at A 350 (Fig. 3), indicating that there may be another nociceptive neurotransmitter, as yet unidentified, that is blocked by morphine in this particular area of the MRF. The other neurons in this category were either on the fringes or outside the MRF.
DISCUSSION In considering the mechanism of how morphine might act to block the increase in firing evoked by the noxious stimulus, there are two possible hypotheses. One is that morphine blocks the release of the neurotransmitter from the presynaptic terminal. The other hypothesis is that morphine prevents the neurotransmitter from producing its effect on the postsynaptic neuron. Based on the data presented in this paper, morphine apparently acts via the latter mechanism; specifically it acts to prevent the postsynaptic neuron from responding to either NE or ACh as well as preventing the response to the NS. This was a specific narcotic effect because it could be reversed by naloxone. It is possible that morphine acts on neurons in the MRF to produce a hyperpolarization as it does in the locus coeruleus (i0). In addition to producing a hyperpolarization, morphine apparently blocks the I-glutamate depolarization in the spinal cord by impairing the Na + influx triggered at the postsynaptlc Na + membrane (ii). Our data do not allow us to determine if the morphine's blockade of the response to the NS and NE or ACh is produced by either of these mechanisms. A second more complicated hypothesis that could also explain these data is that NE and ACh affect an interneuron in such a way that it leads to an increase in firing and morphine blocks the neurotransmitter released from this Interneuron. Our data do not allow us to support or refute this possibility. Even if this is the case, the point remains that morphine interferes with synaptic transmission in pathways that are associated with ACh and NE. Support for the hypothesis that the interruption of cholinergic transmission in the brain is related to analgesia is given below. According to one preliminary report, the injection of scopolamine, a cholinergic antagonist, into the cerebral ventricle of the monkey produces analgesia (16). In the guinea pig, atropine (2.5 mg/kg) produced an analgesic effect (measured as the response to a radiant heart stimulus) similar in magnitude to that of morphine (7.5 mg/kg) although appropriate controls to eliminate the possibility that this effect arose from a peripheral effect were not
766
Morphine Interactions With Neurotransmitters
Vol. 32, No. 7, 1983
carried out (17). Further evidence that morphine may affect the cholinergic system is that when ineffective doses of atropine (0.5 mg/kg) and morphine (3 mg/kg) were given together, they produced significant analgesia (18). Therefore, it is possible that morphine administered microiontophoretically in the MRF blocks the increase in neuronal firing evoked by the NS by blocking the effects of ACh on a postsynaptic neuron. Data supporting the hypothesis that morphine acts on the cholinergic system is supported by the following data. Morphine increases the levels of ACh in the striatum of both the mouse and the rat (13) indicating that it blocks the release of ACh from the synaptic terminals. Morphine blocks the release of ACh from nerve terminals in the guinea pig ileum (14). Furthermore, morphine blocks the release of ACh in the caudate of the cat (15). However our data indicate that morphine may act via a mechanism other than interferring with the release of ACh from synaptic terminals. The effects of morphine in the cortex and caudate may be related to effects of morphine other than those related to analgesia. The cholinergic receptor involved in the analgesic effects of ACh may be muscarinic or nicotinic. The short latency excitatory effect produced by ACh has been reported previously in the rat and is apparently mediated by both muscarinic and nicotinic receptors (19). Nicotine also apparently produced a desensitization to the excitatory response in these neurons (19). Other research (using behavioral measures in mice) indicates that morphine acts on central muscarinic sites but not nicotinic sites (20). This discrepancy may represent a species difference or a difference in technique. In mice, ACh administered intraventricularly enhanced the analgesia produced by morphine; this effect was reversed by atropine indicating that a muscarinic receptor was involved (21). Similar results have also been reported by others (22). Although these data apparently contradict the hypothesis that the analgesic effects of morphine entail in part a blockade of ACh, it is possible that ACh administered intraventricularly blocks a part of the nociceptive pathway by producing a depolarization block or producing a desensitization of the cholinergic receptor. If this assumption is correct, these data would also support the hypothesis that the blockade of the cholinergic pathway will enhance morphine analgesia. Morphine may act via a specific narcotic effect to suppress any increase in firing in neurons throughout the neuroaxis. However, previous evidence, as well as evidence in this study, indicates that there are some cells in which there is an increase in firing evoked by the NS that is not blocked by morphine (2,23,24). Morphine may act to produce a non-specific depression of neuronal firing in cells in which it blocked the response to the NS or one of the two putative neurotransmitters. Evidence against this possibility is that morphine did not always block the increase in firing produced by either NE or ACh, even when it blocked the response to the NS. Furthermore, morphine did not typically block the increase in firing produced by all three stimuli (NS, ACh, and NE). For example, morphine might block the increase in firing produced by the NS and ACh but not that produced by NE. Therefore, it is difficult to accept the hypothesis that morphine produces a non-specific effect that decreases the excitability of the neuron. If this were true, morphine should have always blocked the response to NE and ACh whenever it blocked the response to the NS. Further evidence that morphine does not block the response to NE and ACh by producing a non-selective depression of neuronal firing is as follows.
Vol. 32, No. 7, 1983
Morphine Interactions With Neurotransmitters
767
In those cases in which morphine did not block the response to the NS but NE and ACh mimicked the response to the NS, morphine rarely blocked the response to NE and ACh. These data suggest that the ability of morphine to block NE and ACh are related to a specific effect on neurons in a nociceptive pathway. One possible interpretation of these data is that morphine is acting as a functional antagonist, a concept explained elsewhere (12), of ACh and NE in some neurons that are in the nociceptive pathway. The dose of naloxone that was necessary to reverse morphine's blockade of NE and ACh was rather high (1-3 mg/kg). In a previous study (2) naloxone reversed morphine's ability to block the NS at a lower dose (0.5-1.5 mg/kg). It is important to realize that the dose (or concentration) or morphine at a particular synapse achieved by microiontophoresie may need to be higher than the concentration at this synapse achieved when an analgesic dose of morphine is administered systemically. When morphine is administered systemically it may act at several synapses in the neuroaxis to attenuate the nociceptive signal. However, if only one synapse is involved, the concentration of morphine would have to be higher in order to completely block the response to the NS. Similarly, if the cell is responding to a particular neurotransmitter administered mlcroiontophoretically (which is flooding the cell) a high concentration of morphine would be needed to block the response. Therefore, because the concentration of morphine is higher, the systemic dose of naloxone needed to reverse this effect would also have to be higher. In addition to naloxone reversal, other criteria that should be met in order for the effect of morphine to be considered a specific narcotic effect include stereospecificity. Stereospeclficity is when a compound, levorphanol, has morphine-like activity, but its stereoisomer, dextrorphan, lacks this activity. In the MRF of the rat, levorphanol produces a morphinelike effect in blocking the response to the NS in 5 of 9 cells, but in 4 other cells it produced a prolonged excitatory response (2). Therefore, it was not feasible to test stereoisomers in the MRF in this experiment. A third criterion that should be met for an effect to be considered a specific narcotic effect is that morphine agonists should mimic the effects of morphine. Both oxymorphone and methadone mimic morphine in blocking the response to the NS (2). Based on these previous data, one would predict that oxymorphone and methadone would have effects like morphine on NE and ACh. This is a hypothesis that can be tested in future experiments. It was not feasible to carry out these tests in the present experiments because the technical and pharmacological complexity of these experiments limited the number of drugs that could be tested to three. Because there is no significant interaction between ACh and NE, it is now possible to carry out experiments in which morphine and an agonlst are compared with respect to their effect on the NS and either ACh or NE. In such a paradigm, morphine and one of its agonist could be compared to each other with respect to their ability to block the response to ACh and the NS. The cluster of cells in which NE mimicked the NS is in the same area in which phentolamine blocks both the response to NE and the NE (4). This area also corresponds to the rostral portion of the nucleus reticularls pontis oral~s (25). The cluster of cells at A 620 in which morphine blocked the response to both ACh and the NS indicates that the cholinerglc system that is related to nociceptlon may represent some type of nuclear grouping. In this regard,
it is puzzling why the neurons in which morphine blocks
768
Morphine Interactions With Neurotransmitters
Vol. 32, No. 7, 1983
the response to both ACh and NE do not have a mor~ clear-cut organization. The rather random scatter of these neurons might indicate that there are numerous areas in the MEF where both cholinergic and noradrenergic neurons converge and where nociceptive information is integrated. The grouping of neurons (in the category of morphine blocking the response to the nociceptlve stimulus but not the response NE or ACh) at A 350 may indicate that morphine acts to block some other neurotransmitter in the MRF in this region. This possibility remains to be explored. Based on the above data we propose the following tentative hypotheses. The first hypothesis is that the MRF is a site where morphine acts to produce analgesia. The second hypothesis is that ACh and NE are two neurotransmitters (although not necessarily the only two) in a nociceptive pathway that projects to the MRF. The third hypothesis is that morphine acts at the neuronal level by blocking the response to the nociceptive stimulus that is mediated by these two neurotransmitters. Obviously these hypotheses need to be buttressed by additional data before they should be accepted.
ACKNOWLEDGEMENTS We would like to thank Merck and Company for supplying the morphine sulfate used in this study. We also thank Ms. S. Burson for technical assistance. This study was supported in part by NIDA Grant No. I-RO-I-DA01344-06 (H.J.H.) and NIGMS grant T32-GM-07594 (T.P.O.).
References i. 2. 3. 4. 5. 6. 7. 8. 9.
i0. ii. 12.
H.J. HAIGLER and D.D. SPRING, Life Sci., 23, 1229-1240 (1978) H.J. HAIGLER, Life Sci., 19, 841-858 (1976). H.J. HAIGLER and D.D. SPRIN'--G, Life Sci., 29, 33-43 (1981). T.P. O'NEILL and H.J. HAIGLER, J. Pharmacol. Exp. Therap., 222, 555-561 (1982). G.C. SALMOIRAGHI and F. WEIGHT, Anesthesiol. 28, 54-63 (1967). L.O. RANDALL and J. SELITTO, Arch. Int. Pharmacody., ii, 409-419 (1957). R.C. THOMAS and V.J. WILSON, Nature, 206, 211-213 (1965). H.J. HAIGLER, Eur. J. Pharmacol., 51, 361-376 (1978). J.F.R. KONIG and R.A. KLIPPEL, The R---atBrain: A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brainstem, The Robert Krieger Publishing Co., New York (1970). C.M. PEPPER and G. HENDERSON, Science 209, 394-396 (1980). W. ZIEGLGANSBERGER and H. BAYERL, Brain Res., 115, 111-128 (1976). C.K. BUKNER and R.K. SAINI, J. Pharmacol. Exp. Therap., 194, 564-574
(1975). 13. 14. 15. 16. 17. 18. 19. 20.
J.P. GREEN, S.D. GLICK, A.M. CRANE and P.I.A. SZILAGYI, Eur. J. Pharmacol., 39, 91-99 (1976). M. WEINSTOCK, in The Effects of Narcotic Drugs: Biochemical Pharmacology ed. by D.H. Clouet, Plenum Press, New York, N.Y. pp. 254-261 (1971). T.---L.YAKSH and H.I. YAMAMURA, Brain Res., 83, 520-524 (1975). A. PERT, T.L. YAKSH and J.R. BLAIR, Fed. Proc., 33, 1564 (1974). D.K. JONGH, Acta Physiol Pharmacol Neurl., [, 540-545 (1952). IBID, 3, 164-172 (1954). P.B. BRADLEY and A. DRAY, Br. J. Pharmacol., 45, 372-374 (1972). W.S. DEWEY, L.S. HARRIS, J.F. HOWES and J.A. NUITE, J. Pharmacol. Exp. Ther., 175, 435-442 (1970).
Vol. 32, No. 7, 1983
21. 22. 23.
24. 25.
Morphine Interactions With Neurotransmitters
N.W. PEDIGO, W.S. DEWEy and L.S. HARRIS, J. Pharmacol. Exp. Ther. 193, 845-852 (1975). F.C. TULUNAY, I. YANO and A.E. TAKEMORI, Eur. J. Pharmacol., 35, 285-292 (1976). D.A. HOSFORD and H.J. HAIGLER, J. Pharmacol. Exp. Ther., 219, 496-504 (1981). H.J. HAIGLER and D.D. SPRING, Eur. J. Pharmacol., 67, 64-74 (1980). F. VALVERDI, J. Comp. N e u r o l . , 119, 25-54 (1962).
769
INTERACTIONS
Pergamon Press
OF MORPHINE WITH PUTATIVE NEUROTRANSI41TTERS MESENCEPHALIC RETICULAR FORMATION
IN THE
Henry J. Haigler* and Timothy P. O'Neill Department of Pharmacology Emory University School of Medicine Atlanta, Georgia 30322 (Received in final form Novemher 2, 1982) Summary Acetylcholine (ACh) and norepinephrine (NE) have been identified previously as putative nociceptive neurotransmitters in the mesencephalic reticular formation (MRF) of the rat because they frequently mimic the change in neuronal firing (usually an increase) evoked by a noxious stimulus (NS). The purpose of this study was to determine if i.) morphine (M) acts to prevent the increase in firing evoked by a NS by blocking the effects of either of these two neurotransmitters and 2.) if this effect is a specific narcotic effect. Using the technique of microiontophoresis in conjunction with extracellular recording, we located single units in the MRF in which i.) neuronal firing was accelerated by a NS: 2.) M blocked this response; and 3.) either ACh or NE mimicked the effect of the NS. Neurons meeting these three criteria were studied further to determine if morphine would also block the response to either of the neurotransmitters and if this was a specific narcotic effect. We found that morphine blocked the increase in neuronal firing evoked by the NS and ACh or the NS and NE in over 50% of the cells meeting the above criteria. Some neurons were found in which both ACh and NE mimicked the NS and M blocked all three responses. This blockade of these neurotransmitters was a specific narcotic effect because it could be reversed by the systemic administration of naloxone. These data lead to the tentative hypothesis that M, acting via an opiate receptor, blocks the increase in neuronal firing evoked by a NS by blocking the postsynaptic effects of either ACh or NE. This may be one of the mechanisms by which morphine acts to produce analgesia. The mesencephalic reticular formation (MRF) of the rat is apparently an area where morphine acts to produce analgesia. When morphine is injected directly into this area it produces an analgesic effect (i). When morphine is administered microiontophoretically in the MRF, it blocks the increase in neuronal firing evoked by a nociceptive stimulus (noxious foot pinch; NS)(2). Both of these are specific narcotic effects (1,3). Therefore the MRF is an area where the mechanism of action of morphine can be studied.
*Current address: Section Head, CNS Pharmacology, Department of Biological Research, Searle Research and Development, Div. of G.D. Searle and Co., 4901 Searle Parkway, SkDkie, IL 60077. 0024-3205/83/070759-11503.00/0 Copyright (c) 1983 Pergamon Press Ltd.
760
Morphine Interactions With Neurotransmitters
Vol. 32, No. 7, 1983
Morphine may produce its analgesic effect by interfering with cholinergic or noradrenergic transmission in the brain. Both acetylcholine (ACh) and norepinephrine (NE) frequently mimic the response to the NS in the MRF and therefore have been tentatively identified as nociceptive neurotransmitters (3). More recent evidence that NE is a neurotransmitter mediating nociception in the MRF is that phentolamine, a relatively specific a-adrenoceptor antagonist, blocks both the response to the NS and the response to NE (4). A hypothesis based on the above data is that morphine produces its analgesic effect by interfering with noradrenergic and cholinergic transmission in the MRF. Morphine may act to interfere with the release of either NE or ACh from the presyneptic terminals or block their effects on the postsynaptic neuron. Obviously, there are other possibilities. To determine if morphine interacts with either of these neurotransmitters and how it might interact with them, we carried out the following experiments using the technique of microiontophoresis. The neurons of primary interest in this study were those in which both the NS and one of the putative neurotransmitters evoked an increase in firing. We report here that when morphine blocks the response to a NS, it typically blocks the response to NE, ACh or both. This effect was a specific narcotic effect because it could be reversed by naloxone administered systemically.
METHODS A total of 71 male Sprague-Dawley rats (Charles Rivers Laboratories, Wilmington, Mass) weighing between 220-330 g, were used. All animals were initially anesthetized with chloral hydrate (400 mg/kg) and kept anesthetized throughout the experiment by additional i.v. injections of chloral hydrate administration through a tail vein. Lidocaine (Xylocaine) jelly was used to coat all wDund edges and the earbars and mouthpiece of the stereotaxio apparatus. The body temperature of all animals was maintained between 36-38°C. Microiontophoresls: Five barrel glass mlcropipettes were preloaded with a few strands of fiberglass in each barrel prior to "pulling". The tips of the pipettes were broken back to 5 microns under microscopic control; solutions were then injected directly into the various barrels. The central (recording) barrel was filled with 2M NaCI solution saturated with fast green. One side barrel, the balance barrel, was filled with 4 M NaCI. The remaining three barrels were filled with one of the following drugs: acetylcholine chloride (0.2M, pH 4.2, Sigma); l-norepinephrlne HCI (0.2M pH 4.2 Aldrich); and morphine sulfate (0.05M, pH 4.2, Merck). Circuitry for the iontophoretic ejection of drugs and current balance was similar to that previously described (5), except a solid state system was used. The current balance was kept to within ± 1.0 nA. Impedances, measured at i000 Hz, were 2 to 4 megohms in the recording barrels; 9-11 megohms in the balance barrels; and 12-17 megohms in the drug barrels. The retaining current for all drugs was 5 nA. The duration of ejection for all drugs was 1-2 min. Current controls indicated that the ejection currents had no effects by themselves. Additional evidence that the ejection current had no effect on its own is that in cells in which one putative neurotransmltter was ineffective an ejection current as high as i00 nA was tested, typically without effect. Cells usually responded to ejection currents in range between 20-60 nA.
Vol.
32, No.
7, 1983
Morphine
Interactions
With N e u r o t r a n s m i t t e r s
761
N o c i c e p t i v e Stimulus. The NS used in this study was pressure d e l i v e r e d to the left hind foot by an analgesy m e t e r (Stoelting, no. 51302) using a method m o d i f i e d from (6). This protocol has been described in detail p r e v i o u s l y by H a i g l e r and Spring (3). Briefly, the foot was trapped on a pedestal w i t h a blunt, tapered stylus at 90 ibs/in 2. After I0 seconds the pressure was increased to 540 ibs/in 2 over a 12 second period. This pressure was then m a i n t a i n e d for 30 seconds before the foot was released. Protocol: When m o r p h i n e was a d m i n i s t e r e d m i c r o i o n t o p h o r e t i c a l l y , the ejection of m o r p h i n e began 20 seconds before either placing the foot in the analgesy meter or the iontophoresis of either NE or ACh. In those experiments in which naloxone was tested to d e t e r m i n e if it would reverse the effects of morphine, it was injected i.v. via a tail vein; the dose was calculated on the basis of the free base. Data Analysis. A computer was used to q u i c k l y determine if a drug or the NS had produced a significant effect on spontaneous firing in the following way. A stable baseline was present when there was no significant d i f f e r e n c e between the firing rates of two c o n s e c u t i v e 10-second intervals based on a one-way analysis of variance (ANOVA). Once a stable baseline was established, the m e a n and variance of the last I0 second interval was stored in the computer m e m o r y as the control period. Immediately after the beginning of m i c r o iontophoretic drug ejection or nociceptive stimulation, the firing rate over the next five 10-second periods (test period) was stored in the computer memory. If there were indications that the firing rate was not going to r e t u r n to control quickly, an additional 50 seconds of test data was collected; this procedure was typically n e c e s s a r y in the case of NE. In those cases in which two c o n s e c u t i v e 50 sec intervals were collected, both were compared to the control using a o n e - w a y ANOVA. The null hypothesis of no change in firing rate was rejected if there was a significant change in firing rate at the 0.01 level (F > 3.46, w i t h 5 over 54 degrees of freedom). The control mean firing rate was always determined immediately prior to each e x p e r i m e n t a l test (i.e., drug ejection or the a p p l i c a t i o n of the NS). It is possible that two i0 second intervals are not an adequate estimate in the basal neuronal firing rate. Therefore, if there were any apparent changes in the basal firing rate on the analog record, another control rate was obtained. There were m u l t i p l e control rates obtained during an experiment; each basal mean firing rate was typically ± 2 of the initial basal firing rate throughout the experiment. Furthermore, when one n e u r o t r a n s m i t t e r had no effect but another did, the "no effect" response served as a control. If a n e u r o t r a n s m i t t e r did not produce an effect a 20-60 nA, it was typically tested at i00 n A to make sure that it would not have an effect. We typically studied a n e u r o n of interest for 1.5 to 4.5 hours d u r i n g which time the basal firing rate r a r e l y changed. The response to the NS and the n e u r o t r a n s m i t t e r s remained stable throughout the experiment in the absence of any other drug. When it was n e c e s s a r y to administer additional chloral hydrate anesthesia, the baseline and the evoked r e s p o n s e to the NS and one of the n e u r o t r a n s m i t t e r s was retested. Any d e p r e s s a n t effects of the supplemental dose of chloral hydrate (30-60 m g / k g ) were transient and did n o t depress the cell for longer than 3-5 min. Criteria. The f o l l o w i n g are criteria that we established for the neurons that would be studied extensively. First, neurons had to be affected by the noxious stimulus. Second, m o r p h i n e had to block this response. Third, either NE or ACh had to mimic the r e s p o n s e to the NS (c.f. Fig. 2 and 3, ref. 3 for an illustration of h o w ACh and NE m i m i c k e d the r e s p o n s e to the NS). If the cell could be held long enough, naloxone was injected i.v. to d e t e r m i n e if it would reverse m o r p h i n e ' s blockade of the NS or the two neurotransmitters.
762
Morphine Interactions With Neurotransmltters
Vol. 32, No. 7, 1983
Histology. Recording sites were marked by passing a 25 microamp cathodal current through the barrel for i0 min. This resulted in the deposition of fast green in a discrete spot (7) with a usual diameter of 40 microns. Animals were then perfused and the histological localization for each recording site was determined as previously described (8).
RESULTS ACh mimicked the response to the NS in 27 of 47 (57%) of the neurons studied which is in good agreement with a previous study (3). In the 27 neurons in which ACh mimicked the response to the NS, morphine blocked or attenuated the response to both ACH and the NS in 15 of 27 (56%) of these neurons (Fig. i). NE mimicked the response to the NS in 28 of 43 (65%)
P
ACh
MS
60
MS
40
0 tu ¢0 ¢0 UJ O.
40
ACh
ACh
60
NE
MS
40
NE
MS 40
40
NE
MS
60
40 P
60
40
NE
MS
60
40
ACh
v
60
T
15] o
.Ji
MS NE
. . . . . . .
A
. . . . . . .
". . . . . . . . . . . . . . . . .
40
MS
60
40
......
,,,
MS
40
ACh
60
T
I
2 MIN
I
FIG. i Morphine administered microiontophoretically attenuates the response to the NS (p), acetylchollne (ACh) and norepinephrlne (NE). Mlcroiontophoretically administered acetylcholine at 60 nA (ACh 60) and norepinephrine at 60 nA (NE 60) produce a response that is similar to that produced by the NS. The NS is foot pressure, the arrows beneath the horizontal llne labeled 'p' indicate an increase in pressure from 90-540 ibs/in 2. The horizontal bar underneath ACh, NE and MS (for morphine sulfate) indicates the duration of the ejection current; the number indicates the ejection current in nA. The ordinate represents the integrated rate of firing in spikes/sec. The three arrows (third and fourth trace) mark the point at which naloxone was administered; each arrow indicates the administration of a dose of I mg/kg.
Vol. 32, No. 7, 1983
Morphine Interactions With Neurotransmitters
763
of the neurons studied• In a previous study (3) NE mimicked the response to the NS in 67% of the neurons studied. In these neurons morphine blocked or attenuated response to both NE and the NS in 19 of 28 (68%) neurons (Fig. I). In eight of the neurons that are described above, morphine blocked all three responses; the response to the NS, the response to ACh and the response to NE (Fig. I). Evidence that morphine did not produce a non-specific depressant effect on neuronal firing is that it did not depress the increase in spontaneous firing which was apparent after the administration of naloxone (Fig. i). The increase in spontaneous firing after the second and third dose of naloxone is not readily explained, but could arise from the transient blockade of an endogenous opioid that modulates the activity of this neuron. Note, however, that the firing rate was allowed to return to control before retesting morphine against ACh. A crucial question is whether or not morphine's ability to block the response to NE and Ach is a specific narcotic effect. If it is, naloxone should reverse the effects of morphine. Naloxone reversed the ability of morphine to block the increase in firing produced by ACh and NE (Fig. i and Fig. 2). In 4 of 5 cells naloxone reversed morphine's ability to block
/
20 A: N S
o
C: ACh
/ o
~
~o
,
o
1o
2o
30
40
D U R A T I O N OF T E S T ( S E C )
5o
21o 5o 7o D U R A T I O N OF T E S T
80 (SEC)
100
10
2o
3o
4o
50
D U R A T I O N OF T E S T ( S E C )
FIG. 2 A plot of the mean firing rates with respect to time from Fig. i showing that naloxone reverses morphine's ability to attenuate the response to NE and ACh in a dose-related way. A. The response to the NS (O) produces a mean firing rate of 21-22 spikes per second. B. NE (O) at 60 nA produces a similar increase with a slower onset; note that the abscissa is from 0 to i00 seconds for NE to show the complete response• This is necessary because the latency of the response to NE is longer than either that for the NS or for ACh. C. ACh (O) at 60 nA produces a peak response that is slightly less than that to the NS;, but there is no significant difference between the two peak effects (Student's t-test; p>O.l). In the presence of morphine (Q) the response to the NS, NE and ACh is reduced to approximately 50% of control. After naloxone (I mg/kg), the curve for NE begins to recover but morphine's blockade of ACh is apparently enhanced ( • ) . After a second dose of naloxone ( I mg/kg; ~:) the NE curve returns close to the response that was apparent in the absence of morphine as does the ACh curve• After a third dose of naloxone (cumulative dose, 3 mg/kg; ~ ) the ACh curve is almost indistinguishable from the control and may actually show a slight enhancement.
764
Morphine
Interactions
With Neurotransmitters
A 1610
A 350
A 1020
A 160
P 100
Vol.
32, No.
7, 1983
A 620
P 300
FIG. 3 Histological location of recording sites. The location of recording sites in which morphine blocked the response to the NS and NE are marked with filled squares ( • ) . Recording sites in which morphine blocked the response to the NS and ACh are marked by triangles ( • ); those in which morphine blocked the response to the NS, ACh and NE are marked by asterisks ( 0 ) . The recording sites at which morphine blocked the response to the NS but not to ACh nor NE are marked by a clear circle (O). Recording sites where M blocked the response to the NS but where the response is not mimicked by either ACh or NE are not shown although the sites were also in the MRF. The numbers under each section are the distances of the presented section from a frontal section taken in the zero plane and corresponds to the legend in the atlas by K~nig and Klippel (9); MRF is mesencephalic reticular formation and PAG is periaqueductal gray. All but two of the neurons in which morphine blocked both the NS and either NE or ACh are in the MRF according to the KSnig and Klippel (9) atlas. response to the NS and the response to ACh. In two cells naloxone reversed morphine's effect on the NS and NE. In two additional cells naloxone reversed morphine's blocking effect on NE and ACh but the cell was lost ~ h e n retestlng the NS. Because naloxone can reverse the ability of morphine to attenuate or block the increase in firing produced by NE or ACh as well as that produced by the NS it is apparently a specific narcotic effect. Another line of evidence that morphine did not produce a non-specific depressant effect is that there were 13 cells in which morphine did not
Vol. 32, No. 7, 1983
Morphine Interactions With Neurotransmitters
765
block the response to the NS. In these cells, NE mimicked the response to the NS in 5 (39~) but morphine blocked the response to NE in only I of 5 (20~). Similarly, there were 13 other cells in which morphine did not block the response to the NS. ACh mimicked the response to the NS in 5 (39~) and morphine blocked the response to ACh in only i of 5 (20~). The congruence of these two sets of data is coincidental. The neurons in which morphine blocked both the NS and the response to NE and ACh were in the MRF (Fig. 3). There is a cluster of recording sites at A 160 and P i00 where morphine blocks both the NS and the response to NE. There is also an apparent organization of cholinoceptive neurons because 7 of 8 neurons in which morphine blocked the NS and ACh are clustered together at A 620. There is no apparent organization of the neurons in which morphine apparently blocked the NS and both ACh and NE. In 14 of 40 (35%) neurons, morphine blocked the response to the NS but did not block the response to either NE or ACh. However, 7 of these 14 were all at A 350 (Fig. 3), indicating that there may be another nociceptive neurotransmitter, as yet unidentified, that is blocked by morphine in this particular area of the MRF. The other neurons in this category were either on the fringes or outside the MRF.
DISCUSSION In considering the mechanism of how morphine might act to block the increase in firing evoked by the noxious stimulus, there are two possible hypotheses. One is that morphine blocks the release of the neurotransmitter from the presynaptic terminal. The other hypothesis is that morphine prevents the neurotransmitter from producing its effect on the postsynaptic neuron. Based on the data presented in this paper, morphine apparently acts via the latter mechanism; specifically it acts to prevent the postsynaptic neuron from responding to either NE or ACh as well as preventing the response to the NS. This was a specific narcotic effect because it could be reversed by naloxone. It is possible that morphine acts on neurons in the MRF to produce a hyperpolarization as it does in the locus coeruleus (i0). In addition to producing a hyperpolarization, morphine apparently blocks the I-glutamate depolarization in the spinal cord by impairing the Na + influx triggered at the postsynaptlc Na + membrane (ii). Our data do not allow us to determine if the morphine's blockade of the response to the NS and NE or ACh is produced by either of these mechanisms. A second more complicated hypothesis that could also explain these data is that NE and ACh affect an interneuron in such a way that it leads to an increase in firing and morphine blocks the neurotransmitter released from this Interneuron. Our data do not allow us to support or refute this possibility. Even if this is the case, the point remains that morphine interferes with synaptic transmission in pathways that are associated with ACh and NE. Support for the hypothesis that the interruption of cholinergic transmission in the brain is related to analgesia is given below. According to one preliminary report, the injection of scopolamine, a cholinergic antagonist, into the cerebral ventricle of the monkey produces analgesia (16). In the guinea pig, atropine (2.5 mg/kg) produced an analgesic effect (measured as the response to a radiant heart stimulus) similar in magnitude to that of morphine (7.5 mg/kg) although appropriate controls to eliminate the possibility that this effect arose from a peripheral effect were not
766
Morphine Interactions With Neurotransmitters
Vol. 32, No. 7, 1983
carried out (17). Further evidence that morphine may affect the cholinergic system is that when ineffective doses of atropine (0.5 mg/kg) and morphine (3 mg/kg) were given together, they produced significant analgesia (18). Therefore, it is possible that morphine administered microiontophoretically in the MRF blocks the increase in neuronal firing evoked by the NS by blocking the effects of ACh on a postsynaptic neuron. Data supporting the hypothesis that morphine acts on the cholinergic system is supported by the following data. Morphine increases the levels of ACh in the striatum of both the mouse and the rat (13) indicating that it blocks the release of ACh from the synaptic terminals. Morphine blocks the release of ACh from nerve terminals in the guinea pig ileum (14). Furthermore, morphine blocks the release of ACh in the caudate of the cat (15). However our data indicate that morphine may act via a mechanism other than interferring with the release of ACh from synaptic terminals. The effects of morphine in the cortex and caudate may be related to effects of morphine other than those related to analgesia. The cholinergic receptor involved in the analgesic effects of ACh may be muscarinic or nicotinic. The short latency excitatory effect produced by ACh has been reported previously in the rat and is apparently mediated by both muscarinic and nicotinic receptors (19). Nicotine also apparently produced a desensitization to the excitatory response in these neurons (19). Other research (using behavioral measures in mice) indicates that morphine acts on central muscarinic sites but not nicotinic sites (20). This discrepancy may represent a species difference or a difference in technique. In mice, ACh administered intraventricularly enhanced the analgesia produced by morphine; this effect was reversed by atropine indicating that a muscarinic receptor was involved (21). Similar results have also been reported by others (22). Although these data apparently contradict the hypothesis that the analgesic effects of morphine entail in part a blockade of ACh, it is possible that ACh administered intraventricularly blocks a part of the nociceptive pathway by producing a depolarization block or producing a desensitization of the cholinergic receptor. If this assumption is correct, these data would also support the hypothesis that the blockade of the cholinergic pathway will enhance morphine analgesia. Morphine may act via a specific narcotic effect to suppress any increase in firing in neurons throughout the neuroaxis. However, previous evidence, as well as evidence in this study, indicates that there are some cells in which there is an increase in firing evoked by the NS that is not blocked by morphine (2,23,24). Morphine may act to produce a non-specific depression of neuronal firing in cells in which it blocked the response to the NS or one of the two putative neurotransmitters. Evidence against this possibility is that morphine did not always block the increase in firing produced by either NE or ACh, even when it blocked the response to the NS. Furthermore, morphine did not typically block the increase in firing produced by all three stimuli (NS, ACh, and NE). For example, morphine might block the increase in firing produced by the NS and ACh but not that produced by NE. Therefore, it is difficult to accept the hypothesis that morphine produces a non-specific effect that decreases the excitability of the neuron. If this were true, morphine should have always blocked the response to NE and ACh whenever it blocked the response to the NS. Further evidence that morphine does not block the response to NE and ACh by producing a non-selective depression of neuronal firing is as follows.
Vol. 32, No. 7, 1983
Morphine Interactions With Neurotransmitters
767
In those cases in which morphine did not block the response to the NS but NE and ACh mimicked the response to the NS, morphine rarely blocked the response to NE and ACh. These data suggest that the ability of morphine to block NE and ACh are related to a specific effect on neurons in a nociceptive pathway. One possible interpretation of these data is that morphine is acting as a functional antagonist, a concept explained elsewhere (12), of ACh and NE in some neurons that are in the nociceptive pathway. The dose of naloxone that was necessary to reverse morphine's blockade of NE and ACh was rather high (1-3 mg/kg). In a previous study (2) naloxone reversed morphine's ability to block the NS at a lower dose (0.5-1.5 mg/kg). It is important to realize that the dose (or concentration) or morphine at a particular synapse achieved by microiontophoresie may need to be higher than the concentration at this synapse achieved when an analgesic dose of morphine is administered systemically. When morphine is administered systemically it may act at several synapses in the neuroaxis to attenuate the nociceptive signal. However, if only one synapse is involved, the concentration of morphine would have to be higher in order to completely block the response to the NS. Similarly, if the cell is responding to a particular neurotransmitter administered mlcroiontophoretically (which is flooding the cell) a high concentration of morphine would be needed to block the response. Therefore, because the concentration of morphine is higher, the systemic dose of naloxone needed to reverse this effect would also have to be higher. In addition to naloxone reversal, other criteria that should be met in order for the effect of morphine to be considered a specific narcotic effect include stereospecificity. Stereospeclficity is when a compound, levorphanol, has morphine-like activity, but its stereoisomer, dextrorphan, lacks this activity. In the MRF of the rat, levorphanol produces a morphinelike effect in blocking the response to the NS in 5 of 9 cells, but in 4 other cells it produced a prolonged excitatory response (2). Therefore, it was not feasible to test stereoisomers in the MRF in this experiment. A third criterion that should be met for an effect to be considered a specific narcotic effect is that morphine agonists should mimic the effects of morphine. Both oxymorphone and methadone mimic morphine in blocking the response to the NS (2). Based on these previous data, one would predict that oxymorphone and methadone would have effects like morphine on NE and ACh. This is a hypothesis that can be tested in future experiments. It was not feasible to carry out these tests in the present experiments because the technical and pharmacological complexity of these experiments limited the number of drugs that could be tested to three. Because there is no significant interaction between ACh and NE, it is now possible to carry out experiments in which morphine and an agonlst are compared with respect to their effect on the NS and either ACh or NE. In such a paradigm, morphine and one of its agonist could be compared to each other with respect to their ability to block the response to ACh and the NS. The cluster of cells in which NE mimicked the NS is in the same area in which phentolamine blocks both the response to NE and the NE (4). This area also corresponds to the rostral portion of the nucleus reticularls pontis oral~s (25). The cluster of cells at A 620 in which morphine blocked the response to both ACh and the NS indicates that the cholinerglc system that is related to nociceptlon may represent some type of nuclear grouping. In this regard,
it is puzzling why the neurons in which morphine blocks
768
Morphine Interactions With Neurotransmitters
Vol. 32, No. 7, 1983
the response to both ACh and NE do not have a mor~ clear-cut organization. The rather random scatter of these neurons might indicate that there are numerous areas in the MEF where both cholinergic and noradrenergic neurons converge and where nociceptive information is integrated. The grouping of neurons (in the category of morphine blocking the response to the nociceptlve stimulus but not the response NE or ACh) at A 350 may indicate that morphine acts to block some other neurotransmitter in the MRF in this region. This possibility remains to be explored. Based on the above data we propose the following tentative hypotheses. The first hypothesis is that the MRF is a site where morphine acts to produce analgesia. The second hypothesis is that ACh and NE are two neurotransmitters (although not necessarily the only two) in a nociceptive pathway that projects to the MRF. The third hypothesis is that morphine acts at the neuronal level by blocking the response to the nociceptive stimulus that is mediated by these two neurotransmitters. Obviously these hypotheses need to be buttressed by additional data before they should be accepted.
ACKNOWLEDGEMENTS We would like to thank Merck and Company for supplying the morphine sulfate used in this study. We also thank Ms. S. Burson for technical assistance. This study was supported in part by NIDA Grant No. I-RO-I-DA01344-06 (H.J.H.) and NIGMS grant T32-GM-07594 (T.P.O.).
References i. 2. 3. 4. 5. 6. 7. 8. 9.
i0. ii. 12.
H.J. HAIGLER and D.D. SPRING, Life Sci., 23, 1229-1240 (1978) H.J. HAIGLER, Life Sci., 19, 841-858 (1976). H.J. HAIGLER and D.D. SPRIN'--G, Life Sci., 29, 33-43 (1981). T.P. O'NEILL and H.J. HAIGLER, J. Pharmacol. Exp. Therap., 222, 555-561 (1982). G.C. SALMOIRAGHI and F. WEIGHT, Anesthesiol. 28, 54-63 (1967). L.O. RANDALL and J. SELITTO, Arch. Int. Pharmacody., ii, 409-419 (1957). R.C. THOMAS and V.J. WILSON, Nature, 206, 211-213 (1965). H.J. HAIGLER, Eur. J. Pharmacol., 51, 361-376 (1978). J.F.R. KONIG and R.A. KLIPPEL, The R---atBrain: A Stereotaxic Atlas of the Forebrain and Lower Parts of the Brainstem, The Robert Krieger Publishing Co., New York (1970). C.M. PEPPER and G. HENDERSON, Science 209, 394-396 (1980). W. ZIEGLGANSBERGER and H. BAYERL, Brain Res., 115, 111-128 (1976). C.K. BUKNER and R.K. SAINI, J. Pharmacol. Exp. Therap., 194, 564-574
(1975). 13. 14. 15. 16. 17. 18. 19. 20.
J.P. GREEN, S.D. GLICK, A.M. CRANE and P.I.A. SZILAGYI, Eur. J. Pharmacol., 39, 91-99 (1976). M. WEINSTOCK, in The Effects of Narcotic Drugs: Biochemical Pharmacology ed. by D.H. Clouet, Plenum Press, New York, N.Y. pp. 254-261 (1971). T.---L.YAKSH and H.I. YAMAMURA, Brain Res., 83, 520-524 (1975). A. PERT, T.L. YAKSH and J.R. BLAIR, Fed. Proc., 33, 1564 (1974). D.K. JONGH, Acta Physiol Pharmacol Neurl., [, 540-545 (1952). IBID, 3, 164-172 (1954). P.B. BRADLEY and A. DRAY, Br. J. Pharmacol., 45, 372-374 (1972). W.S. DEWEY, L.S. HARRIS, J.F. HOWES and J.A. NUITE, J. Pharmacol. Exp. Ther., 175, 435-442 (1970).
Vol. 32, No. 7, 1983
21. 22. 23.
24. 25.
Morphine Interactions With Neurotransmitters
N.W. PEDIGO, W.S. DEWEy and L.S. HARRIS, J. Pharmacol. Exp. Ther. 193, 845-852 (1975). F.C. TULUNAY, I. YANO and A.E. TAKEMORI, Eur. J. Pharmacol., 35, 285-292 (1976). D.A. HOSFORD and H.J. HAIGLER, J. Pharmacol. Exp. Ther., 219, 496-504 (1981). H.J. HAIGLER and D.D. SPRING, Eur. J. Pharmacol., 67, 64-74 (1980). F. VALVERDI, J. Comp. N e u r o l . , 119, 25-54 (1962).
769