New York Medical College, Department of Pharmacology, New York, New York, 10019.

 

INTRODUCTION

 

The nature of opioid receptors has intrigued many investigators because a stereoisomeric requirement is coupled with an ability of structurally-related antagonists to inhibit virtually all the pharmacological actions of the opioids. Recent observations (1-3) using fragments of nerve endings or synaptosomes show low-saturability, high-affinity and stereo-specific binding of radiolabelled narcotics or their antagonists. Demonstration of opiate binding material is considered evidence for the existence of an opiate receptor since none of a wide variety of non-opiate drugs, in low concentrations, were bound by receptor substances.

 

The highest binding capacity is associated with the limbic system of the brain. The same regional distribution of radiolabelled opioid was found in human as in monkey brains. Despite the remarkable relationship between binding affinities and opioid potency the binding propekties of the receptor substance remain unchanged by induction of tolerance (4). However, Pert et al. (5) found that acute administration of opioids or antagonists slightly enhanced stereospecific binding capacity of brain extracts but this effect remained fixed despite morphine pellet implantation. They concluded that the enhanced receptor binding was unrelated to either tolerance or physical dependency and have suggested that neither tolerance or physical dependency bear any relationship to the narcotic receptor and that any change induced by chronic treatment is remote from the opioid receptor itself.

 

The fact that a morphine receptor exists prompted a search for an endogenous analog. Terenius and Wahlstrom (6) and Hughes (7), reported that low molecular weight peptides interact with the morphine receptor and this binding can be inhibited by narcotic antagonists. These findings have stimulated much new research.

 

Because some functions of the partially isolated narcotic receptor material appear disparate from pharmacological function, it is essential to determine how opiate receptors behave in the intact animal and whether responses vary from one another. If such variations exist then it may be possible to distinguish among opiate receptors and the mechanisms by which these receptors may be modified. Quantitative pharmacological measurements can estimate the pAx of the opioid receptor (8). This procedure expresses the activity of an antagonist in relation to agonistic action. When the subscript '2' is applied, the pA2 represents the negative log of the molar concentration of the antagonist needed to shift the dose response curve of the agonist two-fold to the right, implying that the antagonist now occupies half the receptor sites. Determination of the precise pA2 is always made in vitro where concentrations are known.

 

In order to obtain analagous data in the intact animal it is essential to obtain measurements during peak activity of the administered drugs. It is assumed there that the concentrations of the drugs are related to the dosage injected. Cox and Weinstock (9) utilized this procedure and reported on quantitative studies of the antagonism by nalorphine of various opioids. Their study showed that the apparent pA2 for transient cataract formation in the mouse eye was the same as the pA2 for analgesia. These investigators stated that on the basis of their findings, the receptor for cataractogenic effect and for analgesia must be the same.

 

Takemori et al. (10) showed that nalorphine or naloxone antagonized morphine analgesia and inhibition of the gastrointestinal tract, but through different receptors. The in vivo apparent pA2 for analgesia in mice was 7.01 and for intestinal inhibition, 6.6. That these receptors ought to be different is supported by the well-known observation that tolerance can develop to the analgesic effect of morphine but less easily to the constipating action of morphine. Other reasons for the observed difference include the probability that morphine has many actions on the gut including blockade of acetylcholine release and stimulation of 5-hydroxytryptamine. These investigators suggest that receptor types may be ranked using a series of agonists in several systems. Evidence for the existence of alpha and beta adrenergic receptors was developed by the same method. Miller and Anderson (11) did find differences in the relative potencies of the nor-derivatives of morphine, codeine and meperidine in the mouse, which suggested a difference in receptor types.

 

Takemori et al. (12) made the interesting and important observation that a single dose of morphine given only two hours before an antinociceptive test could significantly increase the apparent pA2. The receptor was thus sensitized to the antagonistic action of naloxone an effect which persisted for as long as four days after the injection of the morphine. Two methods for analgesic assay were undertaken. One involved the writhing test as modified by Hayashi and Takemori (13) and the other was the tail-flick procedure described by D'Amour and Smith (14). Both tests revealed that a previous injection of morphine produced a pA2 increase about three times greater than control within two hours after the injection of morphine. There was no increase in the ED50 for morphine itself indicating lack of tolerance development. Unlike morphine the mixed agonist-antagonist pentazocine did not induce a rise in pA2 suggesting that it had no effect on the conformation of the narcotic receptor. These investigators suggested that only narcotic analgesics cause a structural change in receptors with increased affinity of the receptors for antagonists. They concluded that their findings may help explain why narcotic antagonists are so effective against severe narcotic depression but not as effective in mild depression.

 

Our interest in the possible heterogeneity of narcotic receptors was prompted by our observation (15) that a single dose of levorphanol induced marked, long-term tolerance to the cataractogenic effects of the opioid whereas the respiratory response to levorphanol remained unchanged after a similar large dosage of the drug. Furthermore, we discovered that the administration of reserpine did not alter the response to the respiratory depressive effect of levorphanol, although a marked decrease was obtained in its antinociceptive action. These observations made it seem unlikely that the receptor mechanisms could be identical. In addition, cataractogenic tolerance occurred quite rapidly and persisted for long-periods of time unlike analgesic or respiratory tolerances. We therefore undertook a systematic study of each of these three characteristic opioid responses in mice: (a) antinociception; (b) respiratory depression; and (c) cataractogenic activity. We attempted to discover whether their apparent pA2 values could be differentially altered by pretreatment with an opioid with respect to dosage required and duration of effect. Significant differences were indeed found suggesting individuality of the receptors.

 

METHODS AND MATERIALS

 

Antinociception was measured using the hot plate method of Eddy and Leimbach (16). The hot plate was maintained at 55°C in a water bath. Respiratory depression was estimated indirectly by the rise in capillary blood pCO2 obtained from an incision in the ventral surface of the mouse tail near the proximal third. This capillary blood had a high p02 of 80-100 millimeters, suggesting much arterial blood with little venous mixture. In order to reduce spontaneous clotting after making an incision, the mouse was administered 100 units of heparin five minutes before the incision was made. Blood was collected without exposure to air using heparinized capillary tubes. The pCO2 and pH measurements were made with the aid of a Radiometer BMS-3. A 50% increase in respiratory depression was defined as three standard deviations above the mean pCO2 of capillary blood from untreated mice. This value is 33 mm, 7 mm above the mean pCO2 level in controls.

 

The AD50 (analgesic dose-50) was determined graphically by the method of Miller and Tainter (17). This quantal method required the use of at least thirty mice for each AD50 value. The ED50 for respiratory depression was also estimated graphically from at least four dose response curves using a minimum of fifteen mice for each curve. The ED50 for lenticular effect was measured according to the method of Weinstock

' (18) and the value obtained by the method of Litchfield and Wilcoxon (19).

 

Levorphanol tartrate was kindly provided by Hoffman La-Roche; naloxone hydrochloride was a gift from the Endo Company; and the methodone hydrochloride was given to us by Malinckrodt & Co. The dosages listed in the text are those of salt. Solutions were prepared in distilled water.

 

RESULTS

 

EFFECT OF RESERPINE ON ANALGESIC AND RESPIRATORY RESPONSES

 

When reserpine, 2 mg/kg, was injected twenty-four hours prior to hot plate testing using levorphanol, a large shift of the dose response curve was observed. The AD50 value for levorphanol rose from 7.5 mg/kg to more than 15 mg/kg whereas no change in the ED50 for respiratory depression was found as measured by the rise in the capillary blood pCO2. These observations clearly indicate differences at least in modulating mechanisms.

 

THE pA2 FOR ANTINOCICEPTION AND RESPIRATORY DEPRESSION

 

Levorphanol and naloxone were injected subcutaneously thirty minutes prior to testing. The naloxone was injected on the contralateral side. The apparent antinociceptive pA2 was found to be 7.0 whereas the respiratory pA2 was 6.99. These values for naloxone-opioid are virtually identical and indicate apparently similar affinity constants for the two response mechanisms. Studies of the cataractogenic effect, using methadone as the opioid and naloxone as the antagonist, revealed a pA2 of 7.03 identical with that of respiratory and analgesic results. Thus it would appear that each of the three test modalities reveal receptor similarity.

 

While all of the systems which have been tested with morphine-like agonists show identical pA2, Takemori et al. (20) have recently reported a higher pA2 (7.4) for respiratory depression. These investigators suggest that the respiratory opiate receptor differs qualitatively from the analgesic receptor. This difference in findings has not as yet been reconciled.

 

EFFECTS OF SINGLE TREATMENT WITH LEVORPHANOL ON THE ED50 OF THE THREE MODALITIES

 

The preceding findings indicate no differences in receptor affinities. Studies were therefore undertaken to discover what change might occur in affinity, or in tolerance development, after opioid treatment. A single dose of levorphanol, 2 mg/kg s.c., given two days prior to testing, produced a significant (p < 0.05) increase in the antinociceptive ED50, from 0.34 mg/kg to 0.58 mg/kg. These results indicate that this small dose of levorphanol had induced tolerance lasting at least two days.

 

When mice were pretreated with levorphanol in a dosage of 45 mg/kg three days prior to testing, the respiratory ED50 remained unchanged (7.0 mg/kg vs. 6.98 mg/kg). These ED50 valuds are considerably greater than the ED50 for antinociception. However, the injection of levorphanol in a dosage of 15 mg/kg s.c. given four days before testing, increased the cartaractogenic ED50 from 14 mg/kg to 21 mg/kg. Lesser dosages did not produce measureable tolerance while larger single dosages induced more profound tolerance.

 

CHANGE IN pA2 WITH PRETREATMENT

 

As shown earlier, the apparent pA2 for antinociceptive, respiratory depression or lenticular responses are similar. However, when the pA2 was determined two days after a single injection of levorphanol, 2 mg/kg s.c., a four-fold increase was obtained. The pA2 rose to 7.56. A similar increase was noted in the pA2 for respiratory depression but a larger dosage was required. Pretreatment, three days earlier in dosages of 10 to 40 mg/kg, increased the pA2 to 7.8 which is considerably greater than the mean value in controls of 6.99. Despite development of tolerance, the lenticular pA2 failed to rise when levorphanol was given four days earlier in a dosage of 15 mg/kg. The pA2 value was 6.9. This finding is of considerable interest because it indicates that no change occurred at the receptor site. Evidently tolerance to the lenticular effect develops at a site remote from the narcotic receptor initiating the event.

 

PRETREATMENT DOSAGE AND RELATION TO CHANGE IN pA2

 

Levorphanol, in a dosage of 2 mg/kg, increased antinociceptive pA2 from 7.0 to 7.6. Doubling the levorphanol dosage to 4 mg/kg, produced no further change, whereas decreasing the dosage to 1 mg/kg, modestly elevated the pA2 to 7.7. Halving this pretreatment dosage caused a substantial drop in pA2 to 7.4. These findings indicate that the change in receptor conformation occurs over a very narrow range of opioid dosage and is not increased further by raising the dosage even as much as four-fold. Very similar findings were seen with respect to respiratory depression: a dosage of 45 mg/kg injected three days earlier, raised the pA2 from 7 to 7.8 and a similar value was obtained using the dosage of only 10 mg/kg. Below this dosage a sharp falloff in pA2 to 7.4 was seen, and at the dosage of 4 mg/kg, the pA2 returned to about 7, indicating a very narrow dose relationship. The lenticular parameter was not further studied because no pA2 shift was observed in animals treated with 15 mg/kg, a dosage which produced a significant elevation in the ED50•

 

DIFFERENCES AMONG THE THREE NARCOTIC RECEPTORS

 

There is evidently a 10:1 difference in the dosage required to increase the pA2, or initiate a pharmacological response of the respiratory receptor, as compared to actions on the antinociceptive receptor. An additional difference between the respiratory and analgesic receptor mechanisms concerns the duration of the pA elevation for the respiratory response, as opposed to the antinociceptive center. The former remains elevated for up to nine days. A similar study by Takemori et a/. (12) indicates that the pA2 for the analgesic time course of morphine persists for only four days. Whether this is related to the initial dosage is not yet known.

 

DISCUSSION

 

More than two decades ago, Schneider (21) showed that treatment with reserpine attenuated the antinociceptive response to moprhine in mice. Others (22-25) have recently shown that stimulation of dopaminergeric receptors, or inhibition of serotonin synthesis leads to inhibition of morphine analgesia in mice. Cholinergic systems also are proposed as positive modulators of morphine analgesia. In our previous study we found that reserpine had no effect on the respiratory response to levorphanol, indicating that this response differed significantly from the antinociceptive response. Present studies revealed a further difference: tolerance develops to the antinociceptive action, but not to the respiratory depressive effect of levorphanol after a single dose, despite the fact that the antinociceptive and respiratory responses show the same pA2. The indifference to biogenic amine modulation shown by the respiratory receptor mechanism, and the failure to develop tolerance, strongly suggests that respiratory and analgesic responses are mediated by different receptors.

 

In an earlier study, Takemori et al. (13) found that pretreatment of mice with morphine increased the apparent pA2. This finding, using levorphanol, applies not only to the analgesic response, but to the respiratory response as well, although the latter requires nearly ten times as large a dose to initiate receptor change. This requirement corresponds to the dosage ratio for respiratory depression as compared with antinociceptive action. This differential could result from reduced perfusion of the respiratory region, so that the concentration of levorphanol might be very different from the concentration found in the region mediating antinociception. Alternatively there may be lesser numbers of receptors for the antinociceptive response, thus fewer receptors may require a lower drug concentration. Neither of these explanations would affect the pA2 findings because these represent ratios of agonist to antagonist, and do not involve, ultimately, absolute concentrations.

 

Tolerance to the lenticular or cataractogenic effect induced by levorphanol has been known for more than a decade. This phenomenon is clearly related to protein synthesis, since inhibitors such as puromycin or actinomycin-D block development of lenticular tolerance (15). Studies (22) showed that increased sensitivity to naloxone could be prevented if mice were pretreated with cycloheximide, an inhibitor of protein synthesis, indicating that the rise in pA2 results from de novo production of protein. In mice tolerant to levorphanol's cataractogenic effect, it was surprising to note that the pA2 value obtained in these mice for naloxone was no different than for naive animals. Evidently, the explanation lies in the fact that, unlike respiratory or antinociceptive receptors, cataractogenic tolerance is unassociated with conformational change in the receptor that initiates transient lenticular cataracts.

 

Cataracts develop from profound sympathetic stimulation and can be attenuated by chemical sypathectomy with reserpine, or by treatment with adrenergic antagonists (23). It would appear that sympathetic stimulation, a hallmark of opioids but not their antagonists, differs remarkably from the analgesic and respiratory depressive responses, despite the fact that the initiating receptor possesses similar constituents as judged by similarities of pA2.

 

The increase in pA2 which accompanies tolerance to the antinociceptive action, is narrowly related to the dosage of opioid used. A rise in pA2 is seen at dosages of less than 1 mg/kg but does not increase further with doses above 4 mg/kg. The fact that a rise in respiratory pA2 exhibits a similar narrow range, suggests that mechanisms for inducing conformational changes in the receptors are very similar, with the notable exception that a much larger dosage is required to produce respiratory receptor alteration. This differential in dosage is proportional roughly to the dosage required for antinociception vs. respiratory depression. Because no change took place at the site initiating the lenticular phenomenon, we conclude that this receptor site differs from the others. In short, the evidence is substantial that each of these three major effects produced by opioids, is initiated by receptors that have significant pharmacological differences. As yet, there is no evidence that they lie in different regions, although earlier studies (24) showed that the pA2 for analgesia was much larger when the levorphanol was injected i.p., whereas the pA2 for respiratory depression was nearly twenty-fold lower under the same circumstances. Naloxone was injected s.c. in each case. It may well be that the blood supply differs from one region to the other.

 

It has been suggested (6) that tolerance to the antinociceptive or other effects of opioids arises because of increases in endogenous morphine-like ligands. These ligands are variously known as endorphin or enkephalin. Increases in endorphin would decrease the available sites for opioid binding, thus requiring larger dosages of opioid to produce a given effect: hence tolerance. While this hypothesis is quite attractive, no tolerance developed to the respiratory depressant action of levorphanol, although there was a significant increase in pA2, whereas tolerance did develop to the antinociceptive action along with a rise in pA2. It seems unlikely that increases in endorphin should occur with one response and not the other. Furthermore, the pA2 was not altered for the lenticular opacity effect despite tolerance development. In conclusion, evidence has been presented which seems to distinguish three major receptor mechanisms. The evidence consists of the ability of two of the three responses to show tolerance to a single dose. But in one of these, the lenticular response, the pA2 does not increase, whereas, in a third, the increase in pA2 for the respiratory effect was not accompanied by tolerance development. Additionally there are substantial differences in sensitivity to changes in receptor conformation induced by pretreatment. An increase in the respiratory pA2 requires about ten times the initiating dosage as does a comparable rise in the antinociceptive pA2. That there may be a larger number of respiratory receptors is one explanation, but perhaps a better one is reduced access to this region as compared with the antinociceptive receptors. Our findings (24) that the pA2 for naloxone-levorphanol fell to 6.4 for respiration vs. 7.7 for antinociception when levorphanol was injected i.p. (and the naloxone s.c.), fits the hypothesis of different sites and different access. It should be mentioned that peak pharmacological effects were observed at the same test interval: thirty minutes after injections.

 

The notion that receptor sites may differ substantially is further supported by the studies of Akera et al. (25) concerning differential effects of sodium on two types of opiate binding sites. In this experiment Nat stimulated the saturable naloxone binding in thalamus-hypothalamus regions, but inhibited it in cerebellum. The authors (25) conclude that their findings strongly support the hypothesis that two typescf naloxone binding sites exist in brain tissues.

 

REFERENCES

 

1. Simon, E.J., Hiller, J.M. and Edelman, I.: Stereospecific binding of the potent narcotic analgesic 3H-etorphine to rat-brain homogenate. Proc. Nat. Acad. Sci. U.S.A. 70: 1947-1949 (1973).

2. Pert, C.B. and Snyder, S.H.: Opiate receptor: Demonstration in nervous tissue. Science 179:1011-1014 (1973).

3. Terenius, L.: Stereospecific interaction between narcotic analgesics and a synaptic plasma membrane fraction of rat cerebral cortex. J. Acta Pharmacol. Toxicol. 32:317320 (1973).

4. Snyder, S.H. and Matthysse, S.: Opiate receptor mechanisms. Neurosci. Res. Prog. Bull. 13:1-166 (1975).

5. Pert, C.B., Pasternak, G. and Snyder, S.H.: Opiate agonists and antagonists discriminated by receptor binding in brain. Science 182:1359-1361 (1973).

6. Terenius, L. and Wahlstrom, A.: Search for an endogenous ligand for the opiate receptor. Acta Physiol. Scand. 94: 74-81 (1975).

7. Hughes, J.: Isolation of an endogenous compound from the brain with pharmacological properties similar to morphine. Brain Res. 88:295-308 (1975).

8. Ariens, E.J. and Van Rossum, J.M.: pDx, pAx and pD'x values in the analysis of pharmacodynamics. Arch. Int. Pharmacodyn. 110:275-299 (1957).

9. Cox, B.M. and Weinstock, M.: Quantitative studies of the antagonism by nalorphine of some of the actions of morphine-like analgesic drugs. Brit. J. Pharmacol. 22: 289-300 (1964).

10. Takemori, A.E., Kupferberg, H.J. and Miller, J.W.: Quantitative studies of the antagonism of morphine by nalorphine and naloxone. J. Pharm. exp. Ther. 169:3945 (1969).

11. Miller, J.M. and Anderson, H.H.: The effect of N-demethylation on certain pharmacologic actions of morphine, codeine and meperidene in the mouse. J. Pharm. exp. Ther. 112:191-196 (1954).

12. Takemori, A.E., Oka, T. and Nishiyama, N.: Alteration of analgesic receptor-antagonist interaction induced by morphine. J. Pharm. exp. Ther. 186:261-265 (1973).

13. Hayashi, G. and Takemori, A.E.: The type of analgesic-receptor interaction involved in analgesic assays. Eur. J. Pharmacol. 16:63-66 (1971).

14. D'Amour, F.E. and Smith, D.L.: A method for determining loss of pain sensation. J. Pharm. exp. Ther. 72:74-79 (1941).

15. Smith, A.A., Karmin, M. and Garett, J.: Tolerance to the lenticular effects of opiates. J. Pharm. exp. Ther. 156:85-91 (1967).

16. Eddy, N.B. and Leimbach, D.: Synthetic analgesics. II. Dithienylbutenyl and dithienylbutylamines. J. Pharm. exp. Ther. 107:385-393 (1953).

17. Miller, L.C. and Tainter, M.L.: Estimation of the ED50 and its error by means of logarithmic-probit graph paper. Proc. Soc. Exp. Biol. Med. 57:261-264 (1944).

18. Weinstock, M.: Similarity between receptors responsible for the production of analgesia and lenticular opacity. Brit. J.lharmacol. 17:433-441 (1961).

19. Litchfield, J.T. and Wilcoxon, F.J.: A simplified method of evaluating dose-effect experiments. J. Pharmacol. exp. Ther. 96:99-114 (1949).

20. McGilliard, K.L. and Takemori, A.E.: Effect of acute and chronic morphine pretreatment on morphine induced respiratory depression and analgesia. Pharmacologist 18: 342 (1976).

21. Schneider, J.A.: Reserpine antagonism of morphine analgesia in mice. Proc. Soc. Exp. Biol. Med. 87:614615 (1954).

22. Tulunay, F.C. and Takemori, A.E.: Further studies on the alteration of analgesia receptor-antagonist interaction induced by morphine. J. Pharm. exp. Ther. 190: 401-407 (1974).

23. Smith, A.A.: V. Adrenergic regulation of the lenticular response to opioids in mice. Res. Publ. Ass. Res. Nerv. Ment. Dis. 46:74-88 (1968).

24. Smith, A.A. and Crofford, M.: Discrimination between analgesic and respiratory responses to opioids. Fed. Proc. (1975).

25. Akera, T., Lee, C. and Brody, T.M.: Differential effects of sodium on two types of opiate binding sites. Life Sci . 16:1801-1802 (1975).