Journal of Neurochemislry Raven Press, Ltd., New York 0 1992 International Society for Neurochemistry
Opioid Control of the In Vitro Release of CholecystokininLike Material from the Rat Substantia Nigra *tJ. J. Benoliel, *?A. Mauborgne, *S. Bourgoin, TJ. C. Legrand, *M. Hamon, and *TF. Cesselin * I N S E W U288, Neurobiologie Cellulaire et Fonctionnelle, and ?Service de Biochimie Midicale, Faculte‘ de Mkdecine Pitik-Salp&riPre,Paris, France
Abstract: Possible interactions between Met-enkephalin and cholecystokinin (CCK)-containing neurons in the rat substantia nigra were investigated by looking for the effects of various opioid receptor ligands and inhibitors of enkephalindegrading enzymes on the K+-evoked overflow of CCK-like material (CCKLM) from substantia nigra slices. The 6-opioid agonists ~-Pen~,~-Pen~-enkephalin (50 p M ) and Tyr-D-ThrGly-Phe-Leu-Thr (DTLET; 3 p M ) enhanced, whereas the popioid agonists Tyr-D-Ala-Gly-MePhe-Gly-ol (DAGO; 10 p M ) and MePhe3, D-Pro4-morphiceptin (PL 017; 10 p M ) decreased, the K+-evoked release of CCKLM. By contrast, the K-opioid agonist U-50488 H (5 p M ) was inactive. The stimulatory effect of DTLET could be prevented by the d antagonist ICI-154129 (50 p M ) , but not by the p antagonist naloxone ( 1 p M ) . Conversely, the latter drug, but not ICI154129, prevented the inhibitory effect of DAGO and PL 017. A significant increase in CCKLM overflow was observed upon tissue superfusion with the peptidase inhibitors kela-
torphan or bestatin plus thiorphan. This effect probably resulted from the stimulation of d-opioid receptors by endogenous enkephalins protected from degradation, because it could be prevented by ICI- 154 129 (50 p M ) . Furthermore, the peptidase inhibitors did not enhance CCKLM release further when 6-opioid receptors were stimulated directly by DTLET (3 p M ) . These data indicate that opioids acting on 6 and p receptors may exert an opposite influence, i.e., excitatory and inhibitory, respectively, on CCK-containing neurons in the rat substantia nigra. Because CCK has antiopioid properties, the 6-opioid control of CCKLM release might participate in the central mechanisms of opiate tolerance. Key Words: Cholecystokinin-K+-evoked releaseSubstantia nigra-p- and d-opioid receptors-Peptidase inhibitors. Benoliel J. J. et al. Opioid control of the in vitro release of cholecystokinin-like material from the rat substantia nigra. J. Neurochem. 58, 916-922 (1992).
Numerous data in the literature support the view that cholecystokinin (CCK) interacts with opioids in pain mechanisms. Thus, large doses of CCK have been shown to induce a naloxone-reversibleanalgesia (Jurna and Zetler, 198l), whereas small doses of this peptide can prevent the antinociceptive action of opioids (Fans et al., 1983). As expected from CCK acting as a natural (but indirect) opioid antagonist, the blockade of CCK receptors by proglumide (Watkins et al., 1985), devazepide (Dourish et al., 1988), or L-365,260 (Dourish et al., 1990) results in a marked enhancement of morphine-induced analgesia. Furthermore, tolerance to the analgesic effect of morphine is delayed when CCK receptors are blocked (Watkins et al., 1985; Dourish et
al., 1988, 1990),leading to the suggestion that chronic stimulation of opioid receptors by morphine may tngger a progressive compensatory increase in the activity of CCK-containing neurons in the CNS. In the present work, possible opioid-CCK interactions were investigated at the level of the substantia nigra, where CCK is located in a subpopulation of dopaminergic neurons (Seroogy et al., 1989). That the substantia nigra may be one of the few brain areas critically involved in opioid-CCK interactions is supported by several observations. Thus, opposite modulations of dopaminergic systems by CCK (inhibitory; see Wang et al., 1984) and opioids (excitatory; see Koob and Bloom, 1983) have been shown to occur within
Received January 3, 1991; revised manuscript received June 22, 199 1;accepted July 30, I99 I . Address correspondence and reprint requests to Dr. J. J. Benoliel at INSERM U288, Facult6 de MCdecine PitiC-SalpCtritre, 91, Boulevard de I’HGpital, 75634 Paris cedex 13, France. Abbreviations used; ACSF, artificial cerebrospinal fluid; CCK,
cholecystokinin; CCK-8S, CCK octapeptide; CCKLM, CCK-like DPDPE, DPen2,Dmaterial; DAGO, Tyr-D-Ala-Gly-MePhe-Gly-ol; Pen5-enkephalin;DTLET, Tyr-D-Thr-Gly-Phe-Leu-Thr; GABA, yaminobutyricacid; ME, Met-enkephalin; MELM, ME-like material; PL 0 I 7, MePhe3,~Pro4-rnorphiceptin;RIA, radioimmunoassay.
916
OPiOID-CCK INTERA CTiONS IN THE SUBSTANTIA NiGRA
the rat mesencephalon, including the substantia nigra. Furthermore, the latter area is involved in both the behavioral tolerance to or dependence on morphine (Stinus et al., 1989) and the processing of pain-related messages (Baumeister et al., 1988),for which clear evidence of opioid-CCK interactions has been reported (see above). Investigations reported herein on the possible modulation of CCK-containing neurons by opioids consisted of looking for changes in the K+-evoked, Ca2+dependent, release of CCK-like material (CCKLM) from substantia nigra slices exposed to selective ligands of the p-, 6-, or K-Opioid receptors. In addition, the release of CCIUM was also measured during tissue superfusion with various peptidase inhibitors known to protect Met-enkephalin (ME) from degradation, therefore leading to the stimulation of opioid receptors by endogenous opioids (Bourgoin et al., 1986). MATERIALS AND METHODS Chemicals Thiorphan, Tyr-D-Thr-Gly-Phe-Leu-Thr (DTLET), DPen',~-Pen'-enkephalin (DPDPE), Tyr-D-Ala-Gly-MePheGly-01 (DAGO), and MePhe3,D-Pro4-morphiceptin(PL 0 17) were from Bachem (Bubendorf, Switzerland). Other compounds included the following: ICI- 154129 (Imperial Chemical Industries, plc, Macclesfield, U.K.); naloxone (Endo Laboratories, Garden City, NY,U.S.A.); U-50488 H (Upjohn Co., Kalamazoo, MI, U.S.A.); EGTA, phosphoramidon, bestatin, and leupeptin (Sigma, St. Louis, MO, U.S.A.); and captopril (Squibb, Princeton, NJ, U.S.A.). Kelatorphan was generously given by Prof. B. P. Roques (INSERM U266, Paris, France). The tracers for the radioimmunoassays (RIAs) were "'Ihuman gastrin (2,000 Ci/mmol; CEA, Saclay, France) and I2'I-ME (2,000 Ci/mmol; New England Nuclear, Boston, MA, U.S.A.).
Superfusion of substantia nigra slices Adult male Sprague-Dawley rats (Centre d'Elevage R. Janvier, Le Genest, France) weighing 250-300 g were killed by decapitation, and the substantiae nigrae were immediately dissected at 4°C according to Torrens et al. (1981). For each experiment, tissues from 15 rats were collected and suspended in an artificial cerebrospinal fluid (ACSF; containing in mM: NaCI, 136; NaHC03, 16.2; KC1, 5.4; NaH2P04, 1.2; CaClZ, 2.2; MgCl2, 1.2; glucose, 5) adjusted to pH 7.3 by bubbling with an O2/CO2 mixture (955). Tissues were then sliced (thickness: 0.3 mm) using a McIlwain tissue chopper, resuspended in ACSF, and then dispersed into 12 thermostated (37°C) chambers for their superfusion at a flow rate of 1 ml/ 4 min with the same medium (for details, see Bourgoin et al., 1986). After a washing period of 20 min, necessary to obtain a steady "spontaneous" outflow of CCKLM and ME-like material (MELM), 1-ml fractions were collected at 0°C and divided immediately into 0.2-ml and 0.5-ml aliquots which were kept at -30°C until the measurement of their CCKLM and MELM contents. Fifteen fractions were collected for each experiment. Tissue depolarization was achieved by increasing KC1 concentration from 5.4 mMto 30 mMin the superfbsing fluid during collection of fractions 3 and 4 and fractions 12
91 7
and 13. When [KCI] was raised to 30 mM, [NaCI] was reduced to I 11.4 mM in order to maintain the isotonicity of the superfusing medium. Compounds to be tested were added from the beginning of the eighth fraction up to the end of the experiment (see Fig. I). Because the ratio of K+-induced overflow of CCKLM or MELM during the second depolarization (K2) to that during the first (K,) pulse was constant in the absence of drugs (see Results), any change in this ratio in the presence of a given substance could be ascribed to the effect of this particular substance on the K+-evoked, Ca2'dependent release of CCKLM or MELM (see Bourgoin et al., 1986).
Measurement of CCKLM The RIA of CCK previously described by Zouaoui et al. (1 990) was used with slight modifications. The buffer for di-
luting the specific CCK antiserum (see Zouaoui et al., 1990) and preparing the IZ5I-humangastrin solutions and the charcoal suspension was 0.05 M barbital-HC1, pH 8.5, containing 1 g/L sodium azide and 0.01 M MgC12. For the measurement of CCKLM in tissues, the two substantiae nigrae from one rat were homogenized in 10 volumes (vollwt) of 0.1 M HCI and heated for 15 min at 95°C. After centrifugation (38,000 g, 10 min, 4"C), the supernatant was adjusted to pH 7.0 with 1 M Tris base. The resulting precipitate was spun down at 6,000 g for 10 min at 4"C, and the clear extract was assayed at three appropriate dilutions. Fifty microliters of each dilution were mixed with 50 pi of the CCK antiserum (l/1,500,000 final dilution), 50 p1 ofthe '*'Ihuman gastrin solution (corresponding to 2,000-2,500 cpm), and 150 pl of 0.05 M barbital-HCI, pH 8.5, supplemented with sodium a i d e and MgCIz (see above). After 48 h at 4"C, the assay was stopped by adsorbing the free tracer onto active dextran T70-coated charcoal (4 and 40 g/L, respectively, in the barbital-HC1 buffer containing 10%horse serum; 1 ml of suspensionftube). The tubes were centrifuged immediately at 6,000 g for 10 min at 4"C, and the radioactivity in the supernatants was estimated by y spectrometry (Beckman 5500 counter). Standard curves were drawn from RIAs of 0.25-50 pg of authentic sulfated CCK octapeptide (CCK-8s) per tube (Bachem, Bubendorf, Switzerland). For the measurement of CCKLM released from slices, 200 pI of each collected fraction were incubated with 50 pl of the antiserum (final dilution: I / 1,500,000) and 50 pl of the barbital-HC1 buffer containing 7.5 g/L bovine serum albumin. After 48 h at 4"C, 50 p1 of the 1z51-humangastrin solution were added, and the incubation proceeded for a further 2024 h. The assay was stopped as described for the tissue extracts. Standard curves were drawn from RIAs of 0.125-25 pg of authentic CCK-8S in 50 ~1 of the barbital-HC1 buffer supplemented with bovine serum albumin. These aliquots were mixed with 50 ~l of the antiserum dilution and 200 p1 of ACSF, and RIAs then proceeded as described above. Each time a compound was added to the superfusing ACSF, a complete standard curve was drawn in the presence of this compound at the same concentration as that used for the superfusion experiments. This allowed accurate determination of CCKLM released in fractions containing an excess of K+, and/or opioid receptor ligands or peptidase inhibitors. Under these conditions, as little as 0.25 pg of CCK3S per tube could be estimated quantitatively. In every case, CCKLM content was expressed as CCKJS equivalents, i.e., in picograms of CCK-8s producing the same displacement of bound 12'I-human gastrin under standard RIA conditions.
J. Neurochem., Vol. 58, No. 3, 1992
J. J. BENOLIEL ET AL.
918 Measurement of MELM
MELM in superfusate fractions (500-pl aliquots) was radioimmunoassayed as previously described (see Bourgoin et al., 1986). MELM content was expressed as ME equivalents, i.e., in picograms of authentic ME producing the same displacement of bound Iz5I-MEunder standard RIA conditions. Statistical analyses were made according to Snedecor and Cochran (1 967). When thep value (Student's t test) was higher than 0.05, a difference was considered to be nonsignificant.
RESULTS Characteristics of CCKLM and MELM outflow from substantia nigra slices The spontaneous CCKLM outflow was stable throughout the experiment with a mean level of 2.6 f 0.1 pg of CCK-8S equivalents/ml (mean 2 SEM, n = 2 1 independent experiments) during the 60-min superfusion period with normal ACSF. Because the superfused tissues contained 2.21 2 0.08 ng/chamber (mean f SEM, n = 12), the spontaneous outflow corresponded to 0.03% (fractional rate constant) of CCKLM tissue contents being released per minute. K+induced depolarization produced a marked enhancement of CCKLM release inasmuch as a fivefold overflow was observed for the collection of fractions 3-5 (K,, Fig. 1). Then CCKLM outflow returned to base-
K+30mM K2
2ol I
line levels 4-8 min after switching the K+-enriched medium to the normal ACSF. The second exposure to 30 mM K+ also induced a significant, but less pronounced, increase in CCKLM release during the collection of fractions 12-14 (K2, Fig. 1). Under these conditions, i.e., in the absence of drugs, the ratio K2/ KI,corresponding to the CCKLM contents in fractions 12-14 over those in fractions 3-5, was remarkably constant: 0.53 +- 0.02 (mean k SEM, n = 21). When Ca2+was omitted from and 0.1 mM EGTA was added to the superfusing fluid, the Kf-evoked overflow was prevented completely, but the spontaneous outflow of CCKLM was not significantly different from that observed upon tissue superfusion with normal ACSF (data not shown). In the same experiments, the spontaneous MELM outflow also remained essentially stable during the 60min superfusion period, with a mean rate of 9.00 f 0.94 pg of ME equivalents/ml (mean f SEM, n = 44 separate experiments), corresponding to 2.25 pg of MELM being released per minute. K+-induced depolarization (K,) produced a marked enhancement of MELM release, because the mean levels of the peptide in fractions 3-5 were about five times those found in the corresponding fractions under resting conditions. Recovery of the baseline (spontaneous) level occurred 4-8 min after switching the K+-enriched medium to the normal ACSF. Then a second exposure (K2) to 30 mMK+ also induced a significant, but lower, enhancement of MELM release so that the ratio K2/KI for MELM was also less than 1 .O (0.5 1 2 0.03, mean -t SEM, n = 10).
r-----
*
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
FRACTIONS
FIG. 1. Effects of K+-induced depolarization on CCKLM outflow from slices of the rat substantia nigra. Tissues were superfused with normal ACSF (flow rate: 1 m1/4 min) except for the collection of fractions 3-4 and 12-1 3, where the concentration of KCI was raised from 5.4 mM to 30 mM (hatched bars). The CCKLM content of each fraction (1 ml) is expressed in picogramsof CCK8S equivalents. Each bar is the mean SEM of data obtained in 21 independent experiments. Kl and K2correspond to the CCKLM overflow due to the first and second K+-induced depolarizations, respectively. Each time a drug was tested on the K+evokedCCKLM overflow, it was added to the superfusing ACSF from the beginning of the eighth fraction up to the end of the experiment. ' p < 0.001, when compared to the mean CCKLM content in fractions collected before and after K1 and K2.
+
J. Neurochem., Vol. 58, No. 3, 1992
Effects of various opioid receptor ligands on CCKLM release from substantia nigra slices None of the tested drugs induced any modification of the basal release of CCKLM. Only the K+-induced CCKLM overflow was affected by opioid agonists and/ or antagonists. The two selective 6-opioid agonists DTLET (3 p M , Zajac et al., 1983) and DPDPE (50 p M , Mosberg et al., 1983) enhanced the K+-evoked CCKLM overflow as shown by significant increases in the K2/K1 ratio upon tissue superfusion with these drugs (+49% and +44%, respectively; Fig. 2). The effects of both agents could be prevented by the selective 6 antagonist ICI154129 (50 pM, Shaw et al., 1982), but 1 phfnaloxone did not block that due to DTLET. However, tissue superfusion with ICI- 1 54 129 or naloxone alone was unable to modify K'-evoked CCKLM overflow (Fig. 2). By contrast, the two selective p-opioid agonists DAGO (10 p M , Gillan and Kosterlitz, 1982) and PL 017 (10 pM, Chang et al., 1983) decreased the K+induced CCKLM overflow (-27% and -257'0, respectively), and this effect could be antagonized by naloxone ( 1 p M ) , but not by ICI-154129 (Fig. 3). Under the same conditions as those showing opposite modulations of K+-evoked CCKLM overflow by 6- and p-opioid agonists, the x-opioid selective ag-
OPIOID-CCK INTERACTIONS IN THE SUBSTANTIA NIGRA l C l b l 5 4 l a 50 pM
3Control
*
DTLET 3 pM
3DPDPE 50 ph
FIG. 2. Effects of ICI-154129 and naloxone on the increased
CCKLM overflow from substantia nigra slices superfused with the 6-opioid agonists DTLET and DPDPE. Experiments were as described in the legend to Fig. 1 with the addition of 3 pM DTLET or 50 pM DPDPE alone or together with 50 pM ICI-154129 or 1 pM naloxone, in the ACSF superfusing the slices from the eighth to the 15th (last) collected fractions. K2/K1 is the ratio of CCKLM contents in fractions 12-14 over those in fractions 3-5 (see Fig. 1). Each bar is the mean t SEM of at least eight separate determinations. The dotted line indicates the K2/K1value in the absence of drugs (NONE, Control). *p < 0.001, when compared to the respective control value (open bars). The effects of DPDPE plus naloxone have not been investigated.
919
onist U-50488 H (5 pM, Von Voigtlander et al., 1983) was inactive [K2/KI = 0.56 f 0.03 (mean k SEM, n = 8), not significantlydifferent from the ratio calculated from experiments without drugs (0.53 k 0.02, see above)]. Effects of various peptidase inhibitors on CCKLM and MELM release from substantia nigra slices Data in Table 1 indicate that neither thiorphan (0.55 p M ) nor phosphoramidon (0.1-10 p M ) , two "enkephalinase" inhibitors (see Schwartz et al., 1985), affected K+-induced CCKLM overflow. Similarly, the K2/KIratio for CCKLM release remained unchanged upon tissue superfusion with the aminopeptidase inhibitor bestatin (20 pM, Umezawa et al., 1976), captopril (1-100 pM), a potent blocker of angiotensinconverting enzyme (see Ondetti et al., 1977), and the mixed thiolserine proteinase inhibitor leupeptin (0.550 pM, see Loh et al., 1984) (Table 1). By contrast, the mixed inhibitor kelatorphan (0.5-5 pM, Bourgoin et al., 1986) and the combination of thiorphan (5 p M ) plus bestatin (20 p M ) enhanced the Kf-evoked overflow of CCKLM (Table 1). Interestingly, the two latter treatments, but not the tissue superfusion with thiorphan (5 p M ) alone, phosphoramidon (10 p M ) , or capTABLE 1. Effects of various peptiduse inhibitors on K+induced CCKLM and MELM overflowfrom substantia nigra slices K2K1 (90)
NALOXONE 1 pM
u
CONTROL DAGOlOfl PL 017 10 pM
Peptidase inhibitors Kelatorphan 0.5 pM 5PM Thiorphan t bestatin 5 pM, 20 p M Thiorphan 0.5 pM 5 PM Phosphoramidon 0.1 p M 1 PM 10 &M Bestatin 20 p M Captopril 144
10 p M FIG. 3. Effects of naloxone and ICI-154129 on the reduction in CCKLM overflow due to superfusion of substantia nigra slices with the p-opioid agonists DAGO and PL 017. Tissues were depolarized by 30 mM KCI in the absence (K,) and the presence (K2) of 10 f l DAGO or 10 pM PL 017 alone or together with 1 pM naloxone or 50 pM ICI-154129, following the protocol described in the legend to Fig. 1. The ratio KZ/K1(see the legend to Fig. 2) is the mean of at least eight independent determinations for each condition. The dotted line indicates the K2/K1value in the absence of drugs (NONE, Control). "p < 0.001, when compared to the respective control value (open bars). The effects of PL 017 plus ICI-154129 have not been investigated.
100 p M Leupeptin 0.5 p M 5 FM 10 culrl 50 pM
CCKLM
MELM
121 k 32 147" 2 11
298'5 71
136' k 8
265 t 32
111 f 8 l o o t 11
114k6
92 t 16 104 f 8 106 f 9
1 1 6 r 10
89 f 6
146a+ 10
108 2 9 1 0 4 f 15 98 f 9
116 F 12
111 t 8 106 t 9 89 f 8 106 f 1 1
-
Each inhibitor was added to the superfusing ACSF from the beginning of the eighth fraction up to the end of the experiment. The ratio K2/KI for each condition is expressed as a percentage of the respective control value, calculated from experiments without drugs. Data are the means k SEM of at least six independent determinations. ' p < 0.02; ' p < 0.001, when compared to K2/KI in the absence of drugs (100%).
J. Neurochem., Vol. 58, No. 3, 1992
920
J. J. BENOLIEL ET AL.
topril (100 p M ) , produced a marked elevation of the K+-evoked MELM overflow from substantia nigra slices (Table 1). Bestatin (20 p M ) alone also enhanced MELM overflow, but to a much lower extent (+46%) than that observed with kelatorphan (+ 198%)or thiorphan plus bestatin (+ 165%)(Table 1).
Effects of DTLET and ICI-154129 on the increased CCKLM overflow induced by thiorphan plus bestatin In order to examine whether the increased CCKLM overflow in the presence of thiorphan plus bestatin could result from the protection of this material from degrading enzymes or from some stimulatory effect of the released MELM through the activation of 6-opioid receptors, further experiments were carried out with selective ligands of these receptors. As shown in Fig. 4, the stimulatory effect of thiorphan plus bestatin was no longer observed when K+-evokedCCKLM overflow was raised by 3 p M DTLET. Furthermore, the 6-opioid antagonist ICI- 154 129 (50 p M ) completely suppressed the increase in the peptide overflow normally induced by the combination of thiorphan plus bestatin (Fig. 4). DISCUSSION The present data showed that the selective p-opioid agonists DAGO and PL 0 17 decreased the K+-induced, Caz+-dependentrelease of CCKLM from slices of the rat substantia nigra. Although this effect was observed using a high concentration ( 10 pM, to ensure maximal response) of these agonists, it can be concluded that it resulted from the selective stimulation of p-opioid reNONE
DTLET 3 uM
0 CON7 THlOl
0.8
.
\TIN 20 w h 4
x
06
E Y N
04
0.2
0
FIG. 4. Effects of thiorphan plus bestatin on K+-inducedCCKLM overflow from substantia nigra slices superfused with or without 3 pM DTLET or 50 pM lCl-154129. Experimentswere as described in the legend to Fig. 1 with or without the addition of 5 pM thiorphan plus 20 jdd bestatin, and with or without 3 pM DTLET or 50 pA.4 ICI-154129, to the ACSF superfusing the slices from the eighth to the 15th collected fractions. Each bar is the mean k SEM of at least eight independentdeterminations of K2/K, for each condition. The dotted line indicates the KdK, value in the absence of drugs (NONE, Control). ' p < 0.001,when compared to the control value.
J . Neurochem., Vol. 58. No. 3, 1992
ceptors because of the following: (a) The partially selective p antagonist naloxone (Magnan et al., 1982), but not the &selective antagonist ICI-154129 (Shaw et al., 1982), prevented the inhibitory effect of DAGO on CCKLM overflow. (b) In contrast to p agonists, the 6 agonists DTLET and DPDPE enhanced CCKLM overflow, and the K agonist U-50488 H exerted no effect on the peptide release. Therefore, the decreased CCKLM overflow upon tissue superfusion with DAGO and PL 017 could not result from some nonspecific stimulation of 6- and K-opioid receptors by these p agonists. It may appear paradoxical that opioid ligands having nanomolar affinity for their receptor binding sites in brain homogenates were still selectiveof their respective targets when used at micromolar concentrations in superfusion experiments. However, it has to be pointed out that affinity measured in binding studies with brain homogenates in the absence of guanine nucleotides, ions, etc., is optimal and may not reflect the actual situation in intact tissues, such as brain slices superfused with ACSF. Furthermore, opioid ligands in the superfusion medium have to diffuse into the slices to reach the receptors, and the actual concentration of these drugs within tissues is probably less than that in the medium. Previous studies aimed at examining possible opioid modulations of the in vitro release of ME (Bourgoin et al., 199I), substance P (Mauborgne et al., 1987), and calcitonin gene-related peptide (Pohl et al., 1989) from the rat spinal cord have demonstrated clearly that p- and 6-opioid ligands used at the same concentrations as those selected in the present study act selectively on their respective receptors. In contrast to the p agonists, 6-opioid agonists enhanced the in vitro release of CCKLM from the rat substantia nigra. Thus, 3 pM DTLET, as well as 50 pM DPDPE, significantly increased the ratio Kz/K,. This effect probably resulted from the stimulation of 6 receptors, because it could be reversed by a 6 antagonist (ICI-154129), but not by a p antagonist (1 pM naloxone; see above). The stimulatory effect of 6-opioid agonists on CCKLM release is compatible with the involvement of 6-, but not p- or K - , opioid receptors in the possible activation of CCK-containing neurons by opiates (see Baber et al., 1989). However, it must be emphasized that our in vitro experiments may be relevant to the acute effects of opioids, whereas a compensatory increase in the activity of central CCKergic systems has been proposed in the case of chronic treatment with these drugs (see introductory section). Whether chronic morphine administration also produced a &mediated increase in CCKLM release within the rat substantia nigra is currently under investigation in our laboratory. The second part of the present study consisted of exploring whether the effects of exogenous opioid agonists on CCKLM release from substantia nigra slices could be mimicked by endogenous opioids and are, therefore, of some relevance for the physiological sit-
OPIOID-CCK INTERACTIONS IN THE SUBSTANTIA NIGRA uation. For this purpose, we used various peptidase inhibitors for possibly enhancing the concentrations of endogenous opioids, notably ME, in the vicinity of opioid receptors. In agreement with previous data obtained with spinal cord (Bourgoin et al., 1986) and striatal (Waksman et al., 1985) slices, we found that kelatorphan, a mixed inhibitor of “enkephalinase” (neutral endopeptidase, EC 3.4.24.1 l), aminopeptidase(s), and dipeptidylaminopeptidase(s),and the association of the “enkephalinase” inhibitor thiorphan plus the aminopeptidase inhibitor bestatin markedly increased the outflow of MELM from slices of the rat substantia nigra. Interestingly, kelatorphan alone and the combination of thiorphan plus bestatin were also able to enhance the CCKLM overflow from substantia nigra slices. The latter effect of thiorphan plus bestatin was not additive with that of DTLET (3 p M ) , suggesting that both treatments acted through a common mechanism. Moreover, the selective 6-opioid antagonist ICI- 154129 completely abolished the increase in CCKLM overflow elicited by the association of the two peptidase inhibitors thiorphan and bestatin. These results strongly suggest that the enhancement of CCKLM overflow observed in the presence of thiorphan plus bestatin (or kelatorphan alone) was due to the protective effect of these substances on endogenous enkephalins, which exert an excitatory influence upon CCKcontaining neurons through the stimulation of 6-opioid receptors. In addition to kelatorphan or thiorphan plus bestatin, the latter peptidase inhibitor alone also produced some increase in the outflow of MELM from substantia nigra slices. However, bestatin alone did not affect the CCKLM overflow probably because the resulting stimulation of 6-opioid receptors by MELM (and possibly other endogenous opioids) was too small under this condition. Indeed, the MELM overflow due to bestatin (20 p M ) alone was only -25% of that found with kelatorphan or thiorphan plus bestatin (see Table 1). An excitatory effect of 6-opioid receptor stimulation is an unusual observation, because opioid agonists generally exert an inhibitory influence on the activity of various neuronal types in the CNS (Miller, 1984). Nevertheless, electrophysiological studies already provided another example of opioid-induced excitation of neuronal activity, i.e., that found on CAI pyramidal cells in the hippocampus (Lynch et al., 1981). In this particular case, it is well established that the opioid effect results from a primary inhibition of inhibitory y-aminobutyric acid (GABA) interneurons impinging onto pyramidal cells (Swearengen and Chavkin, 1989). A similar indirect mechanism might well account for the excitatory influence of 6-opioid receptor stimulation on CCK-containing neurons within the rat substantia nigra. Indeed, Lacey et al. (1989) have shown that nigral dopaminergic neurons, including very probably those where dopamine and CCK are co-localized (Seroogy et al., 1989), are insensitive to 6- (and p) opioid re-
921
ceptor agonists. Therefore, d-opioid agonists might well increase CCKLM release through the blockade of a tonic inhibitory (GABAergic ?) control of CCK-containing neurons in the rat substantia nigra. Further experiments (using GABA antagonists to prevent possibly the effects of 6-opioid agonists) are in progress to test this hypothesis. Except for kelatorphan and the association of thiorphan plus bestatin, none among the various peptidase inhibitors which were tested in the present study was able to increase CCKLM overflow, as expected from the blockade of the catabolism of this endogenous material (see Bourgoin et al., 1986). These results suggest that neither angiotensin-converting enzyme (tested with its selective inhibitor captopril), bestatin-sensitive aminopeptidase(s), nor “enkephalinase” (tested with thiorphan and phosphoramidon) is involved in the degradation of endogenous CCKLM within the rat substantia nigra. Other peptidases have been shown to inactivate exogenous CCK in vitro: a metallopeptidase, distinct from “enkephalinase” (Steardo et al., 1985), a thiol endopeptidase (McDermott et al., 1983), and a serine endopeptidase (Rose et al., 1988). However, their participation in the degradation of the endogenous peptide has yet to be established, especially because the present data suggested that these enzymes are insensitive to the mixed thiol- and serine-proteinase inhibitor leupeptin, at least in the substantia nigra. In conclusion, the present data indicate that opioid agonists acting on 6 and p receptors exerted opposite (i.e., stimulatory and inhibitory, respectively) effects on the in vitro release of CCKLM from slices of the rat substantia nigra. At least the stimulatory effect of 6-opioid agonists might be physiologically relevant, because it could be induced by endogenous opioids (including MELM). Because CCK exerts a central antiopioid action (Faris et al., 1983), these data provide a possible explanation for the enhancement of the behavioral effects of opiates and the prevention of opiate tolerance by CCK antagonists (Fans et al., 1983; Dourish et al., 1988, 1990). Acknowledgment: This research has been supported by grants from INSERM, Universitt!Pans VI, and La Fondation pour la Recherche Medicale. We are grateful to Prof. B. P. Roques for his generous gift of kelatorphan.
REFERENCES Baber N. S., Dourish C. T., and Hill D. R. (1989) The role of CCK, caerulein, and CCK antagonists in nociception. Pain 39, 307328. Baumeister A. A., Anticich T. G., Hawkins M. F., Liter J. C., Thibodeaux H. F., and Guillory E. C. (1988) Evidence that the substantia nigra is a component of the endogenous pain suppression system in the rat. Brain Rex 447, 116- I2 1. Bourgoin S., Le Bars D., Artaud F., Clot A. M., Bouboutou R., FournibZaIuski M. C., Roques B. P., Hamon M., and Cesselin F. (1986) Effects of kelatorphan and other peptidase inhibitors on the in vitro and in vivo release of methionine-enkephalin-like material from the rat spinal cord. J. Pharmacol. Exp. Ther. 238, 360-366.
J. Neurochem., Vol. 58, No. 3, 1992
922
J. J. BENOLIEL ET AL.
Bourgoin S., Collin E., Benoliel J. J., Chantrel D., Mauborgne A,. Pohl M., Hamon M. and Cesselin F. (199 1) Opioid control of the release of met-enkephalin-like material from the rat spinal cord. Brain Res. 551, 178-184. Chang K. J., Wei E. T., Killian A,, and Chang J. K. (1983) Potent morphiceptin analogs: structure activity relationships and morphine-like activities. J. Pharmacol. Exp. Ther. 227, 403-4 10. Dourish C. T., Hawley D., and lversen S. D. (1988) Enhancement of morphine analgesia and prevention of morphine tolerance in the rat by the cholecystokinin antagonist L-364,718. Eur. J. Pharmacol. 147,469-472. Dourish C. T., ONeill M. F., Coughlan J., Kitchener S . J., Hawley D., and Iversen S. D. (1990) The selective CCK-B receptor antagonist L-365,260 enhances morphine analgesia and prevents morphine tolerance in the rat. Eur. J. Pharmacol. 176, 35-44. Fans P. L., Komisaruk B. R., Watkins L. R., and Mayer D. J. (1983) Evidence for the neuropeptide cholecystokinin as an antagonist of opiate analgesia. Science 219, 310-312. Gillan M. G. C. and Kosterlitz H. W. (1982) Spectrum of the p, 6and K-binding sites in homogenates of rat brain. Br. J. Pharmacol. 77,46 1-469. Jurna I. and Zetler G. ( 1 98 1) Antinociceptive effect of centrally administered caerulein and cholecystokinin octapeptide (CCK-8). Eur. J. Pharmacol. 73, 323-33 1. Koob G. F. and Bloom F. E. (1983) Behavioural effects of opioid peptides. Br. Med. Bull. 39, 89-94. Lacey M. G., Mercuri N. B., and North R. A. (1989) Two cell types in rat substantia nigra zona compacta distinguished by membrane properties and the actions of dopamine and opioids. J. Neurosci. 9, 1233-1241. Loh Y. P., Brownstein M. J., and Gainer H. (1984) Proteolysis in neuropeptide processing and other neural functions. Annic. Reti. Neurosci. 7, 189-222. Lynch G. S., Jensen R. A,, McGauch J. L., Davila K., and Oliver M. W. (198 I ) Effects of enkephalin, morphine, and naloxone on the electrical activity of the in vitro hippocampal slice preparation. Exp. Neurol. 71, 527-540. Magnan J., Paterson S. J., Tavani A,, and Kosterlitz H. W. (1982) The binding spectrum of narcotic analgesic drugs with different agonist and antagonist properties. Naunyn-Schmiedebergs Arch Pharmacol. 319, 197-205. Mauborgne A., Lutz 0..Legrand J. C., Hamon M., and Cesselin F. (1987) Opposite effects of d and p opioid receptor agonists on the in vitro release of substance P-like material from the rat spinal cord. J. Neurochem. 48, 529-531. McDermott J. R., Dcdd P. R., Edwardson J. A,, Hardy J. A,, and Smith A. J. (1983) Pathway of inactivation of cholecystokinin octapeptide (CCK-8) by synaptosomal fractions. Neurochem Int. 5, 641-647. Miller R. ( 1984) How do opiates act? Trends Neurosci. 7, 184-1 85. Mosberg H. I., Hurst R., Hruby V. J., Gee K., Yamamura H. I., Galligan J. J., and Burks T. F. (1983) Bis-penicillamine enkephalins possess highly improved specificity toward 6-opioid receptors. Proc. Natl. Acad. Sci. USA 80, 5871-5874. Ondetti M. A,, Rubin B., and Cushman D. W. (1977) Design of specific inhibitors of angiotensin converting enzyme: new class of orally active antihypertensive agents. Science 196, 44 1-444. Pohl M., Lombard M. C., Bourgoin S., Carayon A,, Benoliel J. J., Mauborgne A., Besson J. M., Hamon M., andcesselin F. (1989) Opioid control of the in vitro release of calcitonin gene-related
J. Neurochem., Vol. 58, No. 3, 1992
peptide from primary afferent fibres projecting in the rat cervical cord. Neuropeptides 14, 15 1-159. Rose C., Camus A., and Schwartz J. C. (1988) A serine peptidase responsible for the inactivation of endogenous cholecystokinin in brain. Proc. Natl. Acad. Sci. USA 85, 8326-8330. Schwartz J. C.,Costentin J., and Lecomte J. M. (1985) Pharmacology of enkephalinase inhibitors. Trends Pharmacol. Sci. 6,472-476. Seroogy K., Schalling M., BrenC S., Dagerlind A., Chai S. Y., Hokfelt T., Persson H., Brownstein M., Huan R., Dixon J., Filer D., Schlessinger D., and Goldstein M. (1989) Cholecystokinin and tyrosine hydroxylase messenger RNAs in neurones of rat mesencephalon: peptide/monoamine coexistence studies using in situ hybridization combined with immunocytochemistry. Exp. Brain Res. 74, 149-162. Shaw J. S., Miller L., Turnbull M. J., Gormley J. J., and Morley J. S. (1982) Selective antagonists at the opiate delta-receptor. Life Sci. 31, 1259-1262. Snedecor G. W. and Cochran W. G. (1967) Statistical Methods, 6th ed. Iowa State College Press, Ames, Iowa. Steardo L., Knight M., Tamminga C. A., Barone P., Kask A. M., and Chase T. N. (1985) CCK26-33degrading activity in brain and nonneural tissue: a metalloendopeptidase. J. Neurochem. 45,784-790. Stinus L., Cesselin F., Deminiire J.-M., Bourgoin S.,Hamon M., and Le Moal M. (1989) Effets toxicophiliques des morphinomimktiques. Interactions dopamine-enkkphaline dans I’aire tegmentale ventrale et dans le noyau accumbens. Enckphale 15, 95-104. Swearengen E. and Chavkin C. ( 1 989) Comparison of opioid and GABA receptor control of excitability and membrane conductance in hippocampal CA 1 pyramidal cells in rat. Neuropharmacology 28,689-697. Torrens Y., Beaujouan J. C., Besson M. J., Michelot R., and Glowinski J. (1981) Inhibitory effects of GABA, L-glutamic acid and nicotine on the potassium-evoked release of substance P in substantia nigra slices of the rat. Eur. J. Pharmacol. 71, 383-392. Umezawa H., Aoyagi T., Suda H., Hamada M., and Takeuchi T. (1976) Bestatin, an inhibitor of aminopeptidase B, produced by actinomycetes. J. Antihiof. (Tokyo) 29, 97-99. Von Voigtlander P. F., Lahti R. A., and Lundens J. H. (1983) A selective and structurally novel non-mu (kappa) opioid agonist. J. Pharmacol. Exp. Ther. 224, 7- 12. Waksman G., Bouboutou R., Devin J., Bourgoin S., Cesselin F., Hamon M., Fournit-Zaluski M. C., and Roques B. P. (1985) In vitro and in vivo effects of kelatorphan on enkephalin metabolism in rodent brain. Eur. J. Pharrnacol. 117,233-243. Wang R. Y., White F. J.. and Voigt M. M. (1984) Cholecystokinin, dopamine and schizophrenia. Trends Pharmacol. Sci. 5, 436438. Watkins L. R., Kinscheck 1. B., and Mayer D. J. (1985) Potentiation of morphine analgesia by the cholecystokinin antagonist proglumide. Brain Res. 327, 169-180. Zajac J . M., Gacel G., Petit F., Dodey P., Rossignol P., and Roques B. P. ( 1983) Deltakephalin, Tyr-DThr-Gly-Phe-Leu-Thr, a new highly potent and fully specific agonist for opiate &receptors. Biochem. Biophys. Res. Commun. 111, 390-397. Zouaoui D., Benoliel J. J., Conrath M., and Cesselin F. (1990) Cholecystokinin-like immunoreactivity in the dorsal horn of the rat spinal cord: an attempt to analyse contradictory results between immunocytochemistry and radioimmunoassay. Neuropeptides 17, 177-185.
Opioid Control of the In Vitro Release of CholecystokininLike Material from the Rat Substantia Nigra *tJ. J. Benoliel, *?A. Mauborgne, *S. Bourgoin, TJ. C. Legrand, *M. Hamon, and *TF. Cesselin * I N S E W U288, Neurobiologie Cellulaire et Fonctionnelle, and ?Service de Biochimie Midicale, Faculte‘ de Mkdecine Pitik-Salp&riPre,Paris, France
Abstract: Possible interactions between Met-enkephalin and cholecystokinin (CCK)-containing neurons in the rat substantia nigra were investigated by looking for the effects of various opioid receptor ligands and inhibitors of enkephalindegrading enzymes on the K+-evoked overflow of CCK-like material (CCKLM) from substantia nigra slices. The 6-opioid agonists ~-Pen~,~-Pen~-enkephalin (50 p M ) and Tyr-D-ThrGly-Phe-Leu-Thr (DTLET; 3 p M ) enhanced, whereas the popioid agonists Tyr-D-Ala-Gly-MePhe-Gly-ol (DAGO; 10 p M ) and MePhe3, D-Pro4-morphiceptin (PL 017; 10 p M ) decreased, the K+-evoked release of CCKLM. By contrast, the K-opioid agonist U-50488 H (5 p M ) was inactive. The stimulatory effect of DTLET could be prevented by the d antagonist ICI-154129 (50 p M ) , but not by the p antagonist naloxone ( 1 p M ) . Conversely, the latter drug, but not ICI154129, prevented the inhibitory effect of DAGO and PL 017. A significant increase in CCKLM overflow was observed upon tissue superfusion with the peptidase inhibitors kela-
torphan or bestatin plus thiorphan. This effect probably resulted from the stimulation of d-opioid receptors by endogenous enkephalins protected from degradation, because it could be prevented by ICI- 154 129 (50 p M ) . Furthermore, the peptidase inhibitors did not enhance CCKLM release further when 6-opioid receptors were stimulated directly by DTLET (3 p M ) . These data indicate that opioids acting on 6 and p receptors may exert an opposite influence, i.e., excitatory and inhibitory, respectively, on CCK-containing neurons in the rat substantia nigra. Because CCK has antiopioid properties, the 6-opioid control of CCKLM release might participate in the central mechanisms of opiate tolerance. Key Words: Cholecystokinin-K+-evoked releaseSubstantia nigra-p- and d-opioid receptors-Peptidase inhibitors. Benoliel J. J. et al. Opioid control of the in vitro release of cholecystokinin-like material from the rat substantia nigra. J. Neurochem. 58, 916-922 (1992).
Numerous data in the literature support the view that cholecystokinin (CCK) interacts with opioids in pain mechanisms. Thus, large doses of CCK have been shown to induce a naloxone-reversibleanalgesia (Jurna and Zetler, 198l), whereas small doses of this peptide can prevent the antinociceptive action of opioids (Fans et al., 1983). As expected from CCK acting as a natural (but indirect) opioid antagonist, the blockade of CCK receptors by proglumide (Watkins et al., 1985), devazepide (Dourish et al., 1988), or L-365,260 (Dourish et al., 1990) results in a marked enhancement of morphine-induced analgesia. Furthermore, tolerance to the analgesic effect of morphine is delayed when CCK receptors are blocked (Watkins et al., 1985; Dourish et
al., 1988, 1990),leading to the suggestion that chronic stimulation of opioid receptors by morphine may tngger a progressive compensatory increase in the activity of CCK-containing neurons in the CNS. In the present work, possible opioid-CCK interactions were investigated at the level of the substantia nigra, where CCK is located in a subpopulation of dopaminergic neurons (Seroogy et al., 1989). That the substantia nigra may be one of the few brain areas critically involved in opioid-CCK interactions is supported by several observations. Thus, opposite modulations of dopaminergic systems by CCK (inhibitory; see Wang et al., 1984) and opioids (excitatory; see Koob and Bloom, 1983) have been shown to occur within
Received January 3, 1991; revised manuscript received June 22, 199 1;accepted July 30, I99 I . Address correspondence and reprint requests to Dr. J. J. Benoliel at INSERM U288, Facult6 de MCdecine PitiC-SalpCtritre, 91, Boulevard de I’HGpital, 75634 Paris cedex 13, France. Abbreviations used; ACSF, artificial cerebrospinal fluid; CCK,
cholecystokinin; CCK-8S, CCK octapeptide; CCKLM, CCK-like DPDPE, DPen2,Dmaterial; DAGO, Tyr-D-Ala-Gly-MePhe-Gly-ol; Pen5-enkephalin;DTLET, Tyr-D-Thr-Gly-Phe-Leu-Thr; GABA, yaminobutyricacid; ME, Met-enkephalin; MELM, ME-like material; PL 0 I 7, MePhe3,~Pro4-rnorphiceptin;RIA, radioimmunoassay.
916
OPiOID-CCK INTERA CTiONS IN THE SUBSTANTIA NiGRA
the rat mesencephalon, including the substantia nigra. Furthermore, the latter area is involved in both the behavioral tolerance to or dependence on morphine (Stinus et al., 1989) and the processing of pain-related messages (Baumeister et al., 1988),for which clear evidence of opioid-CCK interactions has been reported (see above). Investigations reported herein on the possible modulation of CCK-containing neurons by opioids consisted of looking for changes in the K+-evoked, Ca2+dependent, release of CCK-like material (CCKLM) from substantia nigra slices exposed to selective ligands of the p-, 6-, or K-Opioid receptors. In addition, the release of CCIUM was also measured during tissue superfusion with various peptidase inhibitors known to protect Met-enkephalin (ME) from degradation, therefore leading to the stimulation of opioid receptors by endogenous opioids (Bourgoin et al., 1986). MATERIALS AND METHODS Chemicals Thiorphan, Tyr-D-Thr-Gly-Phe-Leu-Thr (DTLET), DPen',~-Pen'-enkephalin (DPDPE), Tyr-D-Ala-Gly-MePheGly-01 (DAGO), and MePhe3,D-Pro4-morphiceptin(PL 0 17) were from Bachem (Bubendorf, Switzerland). Other compounds included the following: ICI- 154129 (Imperial Chemical Industries, plc, Macclesfield, U.K.); naloxone (Endo Laboratories, Garden City, NY,U.S.A.); U-50488 H (Upjohn Co., Kalamazoo, MI, U.S.A.); EGTA, phosphoramidon, bestatin, and leupeptin (Sigma, St. Louis, MO, U.S.A.); and captopril (Squibb, Princeton, NJ, U.S.A.). Kelatorphan was generously given by Prof. B. P. Roques (INSERM U266, Paris, France). The tracers for the radioimmunoassays (RIAs) were "'Ihuman gastrin (2,000 Ci/mmol; CEA, Saclay, France) and I2'I-ME (2,000 Ci/mmol; New England Nuclear, Boston, MA, U.S.A.).
Superfusion of substantia nigra slices Adult male Sprague-Dawley rats (Centre d'Elevage R. Janvier, Le Genest, France) weighing 250-300 g were killed by decapitation, and the substantiae nigrae were immediately dissected at 4°C according to Torrens et al. (1981). For each experiment, tissues from 15 rats were collected and suspended in an artificial cerebrospinal fluid (ACSF; containing in mM: NaCI, 136; NaHC03, 16.2; KC1, 5.4; NaH2P04, 1.2; CaClZ, 2.2; MgCl2, 1.2; glucose, 5) adjusted to pH 7.3 by bubbling with an O2/CO2 mixture (955). Tissues were then sliced (thickness: 0.3 mm) using a McIlwain tissue chopper, resuspended in ACSF, and then dispersed into 12 thermostated (37°C) chambers for their superfusion at a flow rate of 1 ml/ 4 min with the same medium (for details, see Bourgoin et al., 1986). After a washing period of 20 min, necessary to obtain a steady "spontaneous" outflow of CCKLM and ME-like material (MELM), 1-ml fractions were collected at 0°C and divided immediately into 0.2-ml and 0.5-ml aliquots which were kept at -30°C until the measurement of their CCKLM and MELM contents. Fifteen fractions were collected for each experiment. Tissue depolarization was achieved by increasing KC1 concentration from 5.4 mMto 30 mMin the superfbsing fluid during collection of fractions 3 and 4 and fractions 12
91 7
and 13. When [KCI] was raised to 30 mM, [NaCI] was reduced to I 11.4 mM in order to maintain the isotonicity of the superfusing medium. Compounds to be tested were added from the beginning of the eighth fraction up to the end of the experiment (see Fig. I). Because the ratio of K+-induced overflow of CCKLM or MELM during the second depolarization (K2) to that during the first (K,) pulse was constant in the absence of drugs (see Results), any change in this ratio in the presence of a given substance could be ascribed to the effect of this particular substance on the K+-evoked, Ca2'dependent release of CCKLM or MELM (see Bourgoin et al., 1986).
Measurement of CCKLM The RIA of CCK previously described by Zouaoui et al. (1 990) was used with slight modifications. The buffer for di-
luting the specific CCK antiserum (see Zouaoui et al., 1990) and preparing the IZ5I-humangastrin solutions and the charcoal suspension was 0.05 M barbital-HC1, pH 8.5, containing 1 g/L sodium azide and 0.01 M MgC12. For the measurement of CCKLM in tissues, the two substantiae nigrae from one rat were homogenized in 10 volumes (vollwt) of 0.1 M HCI and heated for 15 min at 95°C. After centrifugation (38,000 g, 10 min, 4"C), the supernatant was adjusted to pH 7.0 with 1 M Tris base. The resulting precipitate was spun down at 6,000 g for 10 min at 4"C, and the clear extract was assayed at three appropriate dilutions. Fifty microliters of each dilution were mixed with 50 pi of the CCK antiserum (l/1,500,000 final dilution), 50 p1 ofthe '*'Ihuman gastrin solution (corresponding to 2,000-2,500 cpm), and 150 pl of 0.05 M barbital-HCI, pH 8.5, supplemented with sodium a i d e and MgCIz (see above). After 48 h at 4"C, the assay was stopped by adsorbing the free tracer onto active dextran T70-coated charcoal (4 and 40 g/L, respectively, in the barbital-HC1 buffer containing 10%horse serum; 1 ml of suspensionftube). The tubes were centrifuged immediately at 6,000 g for 10 min at 4"C, and the radioactivity in the supernatants was estimated by y spectrometry (Beckman 5500 counter). Standard curves were drawn from RIAs of 0.25-50 pg of authentic sulfated CCK octapeptide (CCK-8s) per tube (Bachem, Bubendorf, Switzerland). For the measurement of CCKLM released from slices, 200 pI of each collected fraction were incubated with 50 pl of the antiserum (final dilution: I / 1,500,000) and 50 pl of the barbital-HC1 buffer containing 7.5 g/L bovine serum albumin. After 48 h at 4"C, 50 p1 of the 1z51-humangastrin solution were added, and the incubation proceeded for a further 2024 h. The assay was stopped as described for the tissue extracts. Standard curves were drawn from RIAs of 0.125-25 pg of authentic CCK-8S in 50 ~1 of the barbital-HC1 buffer supplemented with bovine serum albumin. These aliquots were mixed with 50 ~l of the antiserum dilution and 200 p1 of ACSF, and RIAs then proceeded as described above. Each time a compound was added to the superfusing ACSF, a complete standard curve was drawn in the presence of this compound at the same concentration as that used for the superfusion experiments. This allowed accurate determination of CCKLM released in fractions containing an excess of K+, and/or opioid receptor ligands or peptidase inhibitors. Under these conditions, as little as 0.25 pg of CCK3S per tube could be estimated quantitatively. In every case, CCKLM content was expressed as CCKJS equivalents, i.e., in picograms of CCK-8s producing the same displacement of bound 12'I-human gastrin under standard RIA conditions.
J. Neurochem., Vol. 58, No. 3, 1992
J. J. BENOLIEL ET AL.
918 Measurement of MELM
MELM in superfusate fractions (500-pl aliquots) was radioimmunoassayed as previously described (see Bourgoin et al., 1986). MELM content was expressed as ME equivalents, i.e., in picograms of authentic ME producing the same displacement of bound Iz5I-MEunder standard RIA conditions. Statistical analyses were made according to Snedecor and Cochran (1 967). When thep value (Student's t test) was higher than 0.05, a difference was considered to be nonsignificant.
RESULTS Characteristics of CCKLM and MELM outflow from substantia nigra slices The spontaneous CCKLM outflow was stable throughout the experiment with a mean level of 2.6 f 0.1 pg of CCK-8S equivalents/ml (mean 2 SEM, n = 2 1 independent experiments) during the 60-min superfusion period with normal ACSF. Because the superfused tissues contained 2.21 2 0.08 ng/chamber (mean f SEM, n = 12), the spontaneous outflow corresponded to 0.03% (fractional rate constant) of CCKLM tissue contents being released per minute. K+induced depolarization produced a marked enhancement of CCKLM release inasmuch as a fivefold overflow was observed for the collection of fractions 3-5 (K,, Fig. 1). Then CCKLM outflow returned to base-
K+30mM K2
2ol I
line levels 4-8 min after switching the K+-enriched medium to the normal ACSF. The second exposure to 30 mM K+ also induced a significant, but less pronounced, increase in CCKLM release during the collection of fractions 12-14 (K2, Fig. 1). Under these conditions, i.e., in the absence of drugs, the ratio K2/ KI,corresponding to the CCKLM contents in fractions 12-14 over those in fractions 3-5, was remarkably constant: 0.53 +- 0.02 (mean k SEM, n = 21). When Ca2+was omitted from and 0.1 mM EGTA was added to the superfusing fluid, the Kf-evoked overflow was prevented completely, but the spontaneous outflow of CCKLM was not significantly different from that observed upon tissue superfusion with normal ACSF (data not shown). In the same experiments, the spontaneous MELM outflow also remained essentially stable during the 60min superfusion period, with a mean rate of 9.00 f 0.94 pg of ME equivalents/ml (mean f SEM, n = 44 separate experiments), corresponding to 2.25 pg of MELM being released per minute. K+-induced depolarization (K,) produced a marked enhancement of MELM release, because the mean levels of the peptide in fractions 3-5 were about five times those found in the corresponding fractions under resting conditions. Recovery of the baseline (spontaneous) level occurred 4-8 min after switching the K+-enriched medium to the normal ACSF. Then a second exposure (K2) to 30 mMK+ also induced a significant, but lower, enhancement of MELM release so that the ratio K2/KI for MELM was also less than 1 .O (0.5 1 2 0.03, mean -t SEM, n = 10).
r-----
*
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
FRACTIONS
FIG. 1. Effects of K+-induced depolarization on CCKLM outflow from slices of the rat substantia nigra. Tissues were superfused with normal ACSF (flow rate: 1 m1/4 min) except for the collection of fractions 3-4 and 12-1 3, where the concentration of KCI was raised from 5.4 mM to 30 mM (hatched bars). The CCKLM content of each fraction (1 ml) is expressed in picogramsof CCK8S equivalents. Each bar is the mean SEM of data obtained in 21 independent experiments. Kl and K2correspond to the CCKLM overflow due to the first and second K+-induced depolarizations, respectively. Each time a drug was tested on the K+evokedCCKLM overflow, it was added to the superfusing ACSF from the beginning of the eighth fraction up to the end of the experiment. ' p < 0.001, when compared to the mean CCKLM content in fractions collected before and after K1 and K2.
+
J. Neurochem., Vol. 58, No. 3, 1992
Effects of various opioid receptor ligands on CCKLM release from substantia nigra slices None of the tested drugs induced any modification of the basal release of CCKLM. Only the K+-induced CCKLM overflow was affected by opioid agonists and/ or antagonists. The two selective 6-opioid agonists DTLET (3 p M , Zajac et al., 1983) and DPDPE (50 p M , Mosberg et al., 1983) enhanced the K+-evoked CCKLM overflow as shown by significant increases in the K2/K1 ratio upon tissue superfusion with these drugs (+49% and +44%, respectively; Fig. 2). The effects of both agents could be prevented by the selective 6 antagonist ICI154129 (50 pM, Shaw et al., 1982), but 1 phfnaloxone did not block that due to DTLET. However, tissue superfusion with ICI- 1 54 129 or naloxone alone was unable to modify K'-evoked CCKLM overflow (Fig. 2). By contrast, the two selective p-opioid agonists DAGO (10 p M , Gillan and Kosterlitz, 1982) and PL 017 (10 pM, Chang et al., 1983) decreased the K+induced CCKLM overflow (-27% and -257'0, respectively), and this effect could be antagonized by naloxone ( 1 p M ) , but not by ICI-154129 (Fig. 3). Under the same conditions as those showing opposite modulations of K+-evoked CCKLM overflow by 6- and p-opioid agonists, the x-opioid selective ag-
OPIOID-CCK INTERACTIONS IN THE SUBSTANTIA NIGRA l C l b l 5 4 l a 50 pM
3Control
*
DTLET 3 pM
3DPDPE 50 ph
FIG. 2. Effects of ICI-154129 and naloxone on the increased
CCKLM overflow from substantia nigra slices superfused with the 6-opioid agonists DTLET and DPDPE. Experiments were as described in the legend to Fig. 1 with the addition of 3 pM DTLET or 50 pM DPDPE alone or together with 50 pM ICI-154129 or 1 pM naloxone, in the ACSF superfusing the slices from the eighth to the 15th (last) collected fractions. K2/K1 is the ratio of CCKLM contents in fractions 12-14 over those in fractions 3-5 (see Fig. 1). Each bar is the mean t SEM of at least eight separate determinations. The dotted line indicates the K2/K1value in the absence of drugs (NONE, Control). *p < 0.001, when compared to the respective control value (open bars). The effects of DPDPE plus naloxone have not been investigated.
919
onist U-50488 H (5 pM, Von Voigtlander et al., 1983) was inactive [K2/KI = 0.56 f 0.03 (mean k SEM, n = 8), not significantlydifferent from the ratio calculated from experiments without drugs (0.53 k 0.02, see above)]. Effects of various peptidase inhibitors on CCKLM and MELM release from substantia nigra slices Data in Table 1 indicate that neither thiorphan (0.55 p M ) nor phosphoramidon (0.1-10 p M ) , two "enkephalinase" inhibitors (see Schwartz et al., 1985), affected K+-induced CCKLM overflow. Similarly, the K2/KIratio for CCKLM release remained unchanged upon tissue superfusion with the aminopeptidase inhibitor bestatin (20 pM, Umezawa et al., 1976), captopril (1-100 pM), a potent blocker of angiotensinconverting enzyme (see Ondetti et al., 1977), and the mixed thiolserine proteinase inhibitor leupeptin (0.550 pM, see Loh et al., 1984) (Table 1). By contrast, the mixed inhibitor kelatorphan (0.5-5 pM, Bourgoin et al., 1986) and the combination of thiorphan (5 p M ) plus bestatin (20 p M ) enhanced the Kf-evoked overflow of CCKLM (Table 1). Interestingly, the two latter treatments, but not the tissue superfusion with thiorphan (5 p M ) alone, phosphoramidon (10 p M ) , or capTABLE 1. Effects of various peptiduse inhibitors on K+induced CCKLM and MELM overflowfrom substantia nigra slices K2K1 (90)
NALOXONE 1 pM
u
CONTROL DAGOlOfl PL 017 10 pM
Peptidase inhibitors Kelatorphan 0.5 pM 5PM Thiorphan t bestatin 5 pM, 20 p M Thiorphan 0.5 pM 5 PM Phosphoramidon 0.1 p M 1 PM 10 &M Bestatin 20 p M Captopril 144
10 p M FIG. 3. Effects of naloxone and ICI-154129 on the reduction in CCKLM overflow due to superfusion of substantia nigra slices with the p-opioid agonists DAGO and PL 017. Tissues were depolarized by 30 mM KCI in the absence (K,) and the presence (K2) of 10 f l DAGO or 10 pM PL 017 alone or together with 1 pM naloxone or 50 pM ICI-154129, following the protocol described in the legend to Fig. 1. The ratio KZ/K1(see the legend to Fig. 2) is the mean of at least eight independent determinations for each condition. The dotted line indicates the K2/K1value in the absence of drugs (NONE, Control). "p < 0.001, when compared to the respective control value (open bars). The effects of PL 017 plus ICI-154129 have not been investigated.
100 p M Leupeptin 0.5 p M 5 FM 10 culrl 50 pM
CCKLM
MELM
121 k 32 147" 2 11
298'5 71
136' k 8
265 t 32
111 f 8 l o o t 11
114k6
92 t 16 104 f 8 106 f 9
1 1 6 r 10
89 f 6
146a+ 10
108 2 9 1 0 4 f 15 98 f 9
116 F 12
111 t 8 106 t 9 89 f 8 106 f 1 1
-
Each inhibitor was added to the superfusing ACSF from the beginning of the eighth fraction up to the end of the experiment. The ratio K2/KI for each condition is expressed as a percentage of the respective control value, calculated from experiments without drugs. Data are the means k SEM of at least six independent determinations. ' p < 0.02; ' p < 0.001, when compared to K2/KI in the absence of drugs (100%).
J. Neurochem., Vol. 58, No. 3, 1992
920
J. J. BENOLIEL ET AL.
topril (100 p M ) , produced a marked elevation of the K+-evoked MELM overflow from substantia nigra slices (Table 1). Bestatin (20 p M ) alone also enhanced MELM overflow, but to a much lower extent (+46%) than that observed with kelatorphan (+ 198%)or thiorphan plus bestatin (+ 165%)(Table 1).
Effects of DTLET and ICI-154129 on the increased CCKLM overflow induced by thiorphan plus bestatin In order to examine whether the increased CCKLM overflow in the presence of thiorphan plus bestatin could result from the protection of this material from degrading enzymes or from some stimulatory effect of the released MELM through the activation of 6-opioid receptors, further experiments were carried out with selective ligands of these receptors. As shown in Fig. 4, the stimulatory effect of thiorphan plus bestatin was no longer observed when K+-evokedCCKLM overflow was raised by 3 p M DTLET. Furthermore, the 6-opioid antagonist ICI- 154 129 (50 p M ) completely suppressed the increase in the peptide overflow normally induced by the combination of thiorphan plus bestatin (Fig. 4). DISCUSSION The present data showed that the selective p-opioid agonists DAGO and PL 0 17 decreased the K+-induced, Caz+-dependentrelease of CCKLM from slices of the rat substantia nigra. Although this effect was observed using a high concentration ( 10 pM, to ensure maximal response) of these agonists, it can be concluded that it resulted from the selective stimulation of p-opioid reNONE
DTLET 3 uM
0 CON7 THlOl
0.8
.
\TIN 20 w h 4
x
06
E Y N
04
0.2
0
FIG. 4. Effects of thiorphan plus bestatin on K+-inducedCCKLM overflow from substantia nigra slices superfused with or without 3 pM DTLET or 50 pM lCl-154129. Experimentswere as described in the legend to Fig. 1 with or without the addition of 5 pM thiorphan plus 20 jdd bestatin, and with or without 3 pM DTLET or 50 pA.4 ICI-154129, to the ACSF superfusing the slices from the eighth to the 15th collected fractions. Each bar is the mean k SEM of at least eight independentdeterminations of K2/K, for each condition. The dotted line indicates the KdK, value in the absence of drugs (NONE, Control). ' p < 0.001,when compared to the control value.
J . Neurochem., Vol. 58. No. 3, 1992
ceptors because of the following: (a) The partially selective p antagonist naloxone (Magnan et al., 1982), but not the &selective antagonist ICI-154129 (Shaw et al., 1982), prevented the inhibitory effect of DAGO on CCKLM overflow. (b) In contrast to p agonists, the 6 agonists DTLET and DPDPE enhanced CCKLM overflow, and the K agonist U-50488 H exerted no effect on the peptide release. Therefore, the decreased CCKLM overflow upon tissue superfusion with DAGO and PL 017 could not result from some nonspecific stimulation of 6- and K-opioid receptors by these p agonists. It may appear paradoxical that opioid ligands having nanomolar affinity for their receptor binding sites in brain homogenates were still selectiveof their respective targets when used at micromolar concentrations in superfusion experiments. However, it has to be pointed out that affinity measured in binding studies with brain homogenates in the absence of guanine nucleotides, ions, etc., is optimal and may not reflect the actual situation in intact tissues, such as brain slices superfused with ACSF. Furthermore, opioid ligands in the superfusion medium have to diffuse into the slices to reach the receptors, and the actual concentration of these drugs within tissues is probably less than that in the medium. Previous studies aimed at examining possible opioid modulations of the in vitro release of ME (Bourgoin et al., 199I), substance P (Mauborgne et al., 1987), and calcitonin gene-related peptide (Pohl et al., 1989) from the rat spinal cord have demonstrated clearly that p- and 6-opioid ligands used at the same concentrations as those selected in the present study act selectively on their respective receptors. In contrast to the p agonists, 6-opioid agonists enhanced the in vitro release of CCKLM from the rat substantia nigra. Thus, 3 pM DTLET, as well as 50 pM DPDPE, significantly increased the ratio Kz/K,. This effect probably resulted from the stimulation of 6 receptors, because it could be reversed by a 6 antagonist (ICI-154129), but not by a p antagonist (1 pM naloxone; see above). The stimulatory effect of 6-opioid agonists on CCKLM release is compatible with the involvement of 6-, but not p- or K - , opioid receptors in the possible activation of CCK-containing neurons by opiates (see Baber et al., 1989). However, it must be emphasized that our in vitro experiments may be relevant to the acute effects of opioids, whereas a compensatory increase in the activity of central CCKergic systems has been proposed in the case of chronic treatment with these drugs (see introductory section). Whether chronic morphine administration also produced a &mediated increase in CCKLM release within the rat substantia nigra is currently under investigation in our laboratory. The second part of the present study consisted of exploring whether the effects of exogenous opioid agonists on CCKLM release from substantia nigra slices could be mimicked by endogenous opioids and are, therefore, of some relevance for the physiological sit-
OPIOID-CCK INTERACTIONS IN THE SUBSTANTIA NIGRA uation. For this purpose, we used various peptidase inhibitors for possibly enhancing the concentrations of endogenous opioids, notably ME, in the vicinity of opioid receptors. In agreement with previous data obtained with spinal cord (Bourgoin et al., 1986) and striatal (Waksman et al., 1985) slices, we found that kelatorphan, a mixed inhibitor of “enkephalinase” (neutral endopeptidase, EC 3.4.24.1 l), aminopeptidase(s), and dipeptidylaminopeptidase(s),and the association of the “enkephalinase” inhibitor thiorphan plus the aminopeptidase inhibitor bestatin markedly increased the outflow of MELM from slices of the rat substantia nigra. Interestingly, kelatorphan alone and the combination of thiorphan plus bestatin were also able to enhance the CCKLM overflow from substantia nigra slices. The latter effect of thiorphan plus bestatin was not additive with that of DTLET (3 p M ) , suggesting that both treatments acted through a common mechanism. Moreover, the selective 6-opioid antagonist ICI- 154129 completely abolished the increase in CCKLM overflow elicited by the association of the two peptidase inhibitors thiorphan and bestatin. These results strongly suggest that the enhancement of CCKLM overflow observed in the presence of thiorphan plus bestatin (or kelatorphan alone) was due to the protective effect of these substances on endogenous enkephalins, which exert an excitatory influence upon CCKcontaining neurons through the stimulation of 6-opioid receptors. In addition to kelatorphan or thiorphan plus bestatin, the latter peptidase inhibitor alone also produced some increase in the outflow of MELM from substantia nigra slices. However, bestatin alone did not affect the CCKLM overflow probably because the resulting stimulation of 6-opioid receptors by MELM (and possibly other endogenous opioids) was too small under this condition. Indeed, the MELM overflow due to bestatin (20 p M ) alone was only -25% of that found with kelatorphan or thiorphan plus bestatin (see Table 1). An excitatory effect of 6-opioid receptor stimulation is an unusual observation, because opioid agonists generally exert an inhibitory influence on the activity of various neuronal types in the CNS (Miller, 1984). Nevertheless, electrophysiological studies already provided another example of opioid-induced excitation of neuronal activity, i.e., that found on CAI pyramidal cells in the hippocampus (Lynch et al., 1981). In this particular case, it is well established that the opioid effect results from a primary inhibition of inhibitory y-aminobutyric acid (GABA) interneurons impinging onto pyramidal cells (Swearengen and Chavkin, 1989). A similar indirect mechanism might well account for the excitatory influence of 6-opioid receptor stimulation on CCK-containing neurons within the rat substantia nigra. Indeed, Lacey et al. (1989) have shown that nigral dopaminergic neurons, including very probably those where dopamine and CCK are co-localized (Seroogy et al., 1989), are insensitive to 6- (and p) opioid re-
921
ceptor agonists. Therefore, d-opioid agonists might well increase CCKLM release through the blockade of a tonic inhibitory (GABAergic ?) control of CCK-containing neurons in the rat substantia nigra. Further experiments (using GABA antagonists to prevent possibly the effects of 6-opioid agonists) are in progress to test this hypothesis. Except for kelatorphan and the association of thiorphan plus bestatin, none among the various peptidase inhibitors which were tested in the present study was able to increase CCKLM overflow, as expected from the blockade of the catabolism of this endogenous material (see Bourgoin et al., 1986). These results suggest that neither angiotensin-converting enzyme (tested with its selective inhibitor captopril), bestatin-sensitive aminopeptidase(s), nor “enkephalinase” (tested with thiorphan and phosphoramidon) is involved in the degradation of endogenous CCKLM within the rat substantia nigra. Other peptidases have been shown to inactivate exogenous CCK in vitro: a metallopeptidase, distinct from “enkephalinase” (Steardo et al., 1985), a thiol endopeptidase (McDermott et al., 1983), and a serine endopeptidase (Rose et al., 1988). However, their participation in the degradation of the endogenous peptide has yet to be established, especially because the present data suggested that these enzymes are insensitive to the mixed thiol- and serine-proteinase inhibitor leupeptin, at least in the substantia nigra. In conclusion, the present data indicate that opioid agonists acting on 6 and p receptors exerted opposite (i.e., stimulatory and inhibitory, respectively) effects on the in vitro release of CCKLM from slices of the rat substantia nigra. At least the stimulatory effect of 6-opioid agonists might be physiologically relevant, because it could be induced by endogenous opioids (including MELM). Because CCK exerts a central antiopioid action (Faris et al., 1983), these data provide a possible explanation for the enhancement of the behavioral effects of opiates and the prevention of opiate tolerance by CCK antagonists (Fans et al., 1983; Dourish et al., 1988, 1990). Acknowledgment: This research has been supported by grants from INSERM, Universitt!Pans VI, and La Fondation pour la Recherche Medicale. We are grateful to Prof. B. P. Roques for his generous gift of kelatorphan.
REFERENCES Baber N. S., Dourish C. T., and Hill D. R. (1989) The role of CCK, caerulein, and CCK antagonists in nociception. Pain 39, 307328. Baumeister A. A., Anticich T. G., Hawkins M. F., Liter J. C., Thibodeaux H. F., and Guillory E. C. (1988) Evidence that the substantia nigra is a component of the endogenous pain suppression system in the rat. Brain Rex 447, 116- I2 1. Bourgoin S., Le Bars D., Artaud F., Clot A. M., Bouboutou R., FournibZaIuski M. C., Roques B. P., Hamon M., and Cesselin F. (1986) Effects of kelatorphan and other peptidase inhibitors on the in vitro and in vivo release of methionine-enkephalin-like material from the rat spinal cord. J. Pharmacol. Exp. Ther. 238, 360-366.
J. Neurochem., Vol. 58, No. 3, 1992
922
J. J. BENOLIEL ET AL.
Bourgoin S., Collin E., Benoliel J. J., Chantrel D., Mauborgne A,. Pohl M., Hamon M. and Cesselin F. (199 1) Opioid control of the release of met-enkephalin-like material from the rat spinal cord. Brain Res. 551, 178-184. Chang K. J., Wei E. T., Killian A,, and Chang J. K. (1983) Potent morphiceptin analogs: structure activity relationships and morphine-like activities. J. Pharmacol. Exp. Ther. 227, 403-4 10. Dourish C. T., Hawley D., and lversen S. D. (1988) Enhancement of morphine analgesia and prevention of morphine tolerance in the rat by the cholecystokinin antagonist L-364,718. Eur. J. Pharmacol. 147,469-472. Dourish C. T., ONeill M. F., Coughlan J., Kitchener S . J., Hawley D., and Iversen S. D. (1990) The selective CCK-B receptor antagonist L-365,260 enhances morphine analgesia and prevents morphine tolerance in the rat. Eur. J. Pharmacol. 176, 35-44. Fans P. L., Komisaruk B. R., Watkins L. R., and Mayer D. J. (1983) Evidence for the neuropeptide cholecystokinin as an antagonist of opiate analgesia. Science 219, 310-312. Gillan M. G. C. and Kosterlitz H. W. (1982) Spectrum of the p, 6and K-binding sites in homogenates of rat brain. Br. J. Pharmacol. 77,46 1-469. Jurna I. and Zetler G. ( 1 98 1) Antinociceptive effect of centrally administered caerulein and cholecystokinin octapeptide (CCK-8). Eur. J. Pharmacol. 73, 323-33 1. Koob G. F. and Bloom F. E. (1983) Behavioural effects of opioid peptides. Br. Med. Bull. 39, 89-94. Lacey M. G., Mercuri N. B., and North R. A. (1989) Two cell types in rat substantia nigra zona compacta distinguished by membrane properties and the actions of dopamine and opioids. J. Neurosci. 9, 1233-1241. Loh Y. P., Brownstein M. J., and Gainer H. (1984) Proteolysis in neuropeptide processing and other neural functions. Annic. Reti. Neurosci. 7, 189-222. Lynch G. S., Jensen R. A,, McGauch J. L., Davila K., and Oliver M. W. (198 I ) Effects of enkephalin, morphine, and naloxone on the electrical activity of the in vitro hippocampal slice preparation. Exp. Neurol. 71, 527-540. Magnan J., Paterson S. J., Tavani A,, and Kosterlitz H. W. (1982) The binding spectrum of narcotic analgesic drugs with different agonist and antagonist properties. Naunyn-Schmiedebergs Arch Pharmacol. 319, 197-205. Mauborgne A., Lutz 0..Legrand J. C., Hamon M., and Cesselin F. (1987) Opposite effects of d and p opioid receptor agonists on the in vitro release of substance P-like material from the rat spinal cord. J. Neurochem. 48, 529-531. McDermott J. R., Dcdd P. R., Edwardson J. A,, Hardy J. A,, and Smith A. J. (1983) Pathway of inactivation of cholecystokinin octapeptide (CCK-8) by synaptosomal fractions. Neurochem Int. 5, 641-647. Miller R. ( 1984) How do opiates act? Trends Neurosci. 7, 184-1 85. Mosberg H. I., Hurst R., Hruby V. J., Gee K., Yamamura H. I., Galligan J. J., and Burks T. F. (1983) Bis-penicillamine enkephalins possess highly improved specificity toward 6-opioid receptors. Proc. Natl. Acad. Sci. USA 80, 5871-5874. Ondetti M. A,, Rubin B., and Cushman D. W. (1977) Design of specific inhibitors of angiotensin converting enzyme: new class of orally active antihypertensive agents. Science 196, 44 1-444. Pohl M., Lombard M. C., Bourgoin S., Carayon A,, Benoliel J. J., Mauborgne A., Besson J. M., Hamon M., andcesselin F. (1989) Opioid control of the in vitro release of calcitonin gene-related
J. Neurochem., Vol. 58, No. 3, 1992
peptide from primary afferent fibres projecting in the rat cervical cord. Neuropeptides 14, 15 1-159. Rose C., Camus A., and Schwartz J. C. (1988) A serine peptidase responsible for the inactivation of endogenous cholecystokinin in brain. Proc. Natl. Acad. Sci. USA 85, 8326-8330. Schwartz J. C.,Costentin J., and Lecomte J. M. (1985) Pharmacology of enkephalinase inhibitors. Trends Pharmacol. Sci. 6,472-476. Seroogy K., Schalling M., BrenC S., Dagerlind A., Chai S. Y., Hokfelt T., Persson H., Brownstein M., Huan R., Dixon J., Filer D., Schlessinger D., and Goldstein M. (1989) Cholecystokinin and tyrosine hydroxylase messenger RNAs in neurones of rat mesencephalon: peptide/monoamine coexistence studies using in situ hybridization combined with immunocytochemistry. Exp. Brain Res. 74, 149-162. Shaw J. S., Miller L., Turnbull M. J., Gormley J. J., and Morley J. S. (1982) Selective antagonists at the opiate delta-receptor. Life Sci. 31, 1259-1262. Snedecor G. W. and Cochran W. G. (1967) Statistical Methods, 6th ed. Iowa State College Press, Ames, Iowa. Steardo L., Knight M., Tamminga C. A., Barone P., Kask A. M., and Chase T. N. (1985) CCK26-33degrading activity in brain and nonneural tissue: a metalloendopeptidase. J. Neurochem. 45,784-790. Stinus L., Cesselin F., Deminiire J.-M., Bourgoin S.,Hamon M., and Le Moal M. (1989) Effets toxicophiliques des morphinomimktiques. Interactions dopamine-enkkphaline dans I’aire tegmentale ventrale et dans le noyau accumbens. Enckphale 15, 95-104. Swearengen E. and Chavkin C. ( 1 989) Comparison of opioid and GABA receptor control of excitability and membrane conductance in hippocampal CA 1 pyramidal cells in rat. Neuropharmacology 28,689-697. Torrens Y., Beaujouan J. C., Besson M. J., Michelot R., and Glowinski J. (1981) Inhibitory effects of GABA, L-glutamic acid and nicotine on the potassium-evoked release of substance P in substantia nigra slices of the rat. Eur. J. Pharmacol. 71, 383-392. Umezawa H., Aoyagi T., Suda H., Hamada M., and Takeuchi T. (1976) Bestatin, an inhibitor of aminopeptidase B, produced by actinomycetes. J. Antihiof. (Tokyo) 29, 97-99. Von Voigtlander P. F., Lahti R. A., and Lundens J. H. (1983) A selective and structurally novel non-mu (kappa) opioid agonist. J. Pharmacol. Exp. Ther. 224, 7- 12. Waksman G., Bouboutou R., Devin J., Bourgoin S., Cesselin F., Hamon M., Fournit-Zaluski M. C., and Roques B. P. (1985) In vitro and in vivo effects of kelatorphan on enkephalin metabolism in rodent brain. Eur. J. Pharrnacol. 117,233-243. Wang R. Y., White F. J.. and Voigt M. M. (1984) Cholecystokinin, dopamine and schizophrenia. Trends Pharmacol. Sci. 5, 436438. Watkins L. R., Kinscheck 1. B., and Mayer D. J. (1985) Potentiation of morphine analgesia by the cholecystokinin antagonist proglumide. Brain Res. 327, 169-180. Zajac J . M., Gacel G., Petit F., Dodey P., Rossignol P., and Roques B. P. ( 1983) Deltakephalin, Tyr-DThr-Gly-Phe-Leu-Thr, a new highly potent and fully specific agonist for opiate &receptors. Biochem. Biophys. Res. Commun. 111, 390-397. Zouaoui D., Benoliel J. J., Conrath M., and Cesselin F. (1990) Cholecystokinin-like immunoreactivity in the dorsal horn of the rat spinal cord: an attempt to analyse contradictory results between immunocytochemistry and radioimmunoassay. Neuropeptides 17, 177-185.
