Interactions of antibiotics of the iturin group with human erythrocytes.

The peptidolipid antibiotics, iturin A and bacillomycin L have a disrupting effect on erythrocyte membrane leading to a simultaneous release of K+ ions and hemoglobin. The formation of ghosts is accompanied by a partial solubilisation of lipid components. Binding experiments with radioactive antibiotics show that about 7 X 10(7) molecules of iturin A and 4 X 10(7) molecules of bacillomycin L are bound to one erythrocyte at the concentration giving 100% hemolysis. This concentration is reduced by about 20% after treatment of erythrocytes by phospholipase A2. Scatchard plots show that the affinity for erythrocyte membrane is higher with bacillomycin L than with iturin A. This difference is probably correlated to the differences in their peptide moieties. The substitution of tyrosyl residue leads to a ten-fold increase of the concentrations giving 100% hemolysis, probably due to a low distribution coefficient of derivatives between the membrane and the solvent. This result and the humped Scatchard curves obtained with both antibiotics may be related to the self-association of the compounds in aqueous solutions.

Specific binding sites on human phagocytic blood cells for Gly-Leu-Phe and Val-Glu-Pro-Ile-Pro-Tyr, immunostimulating peptides from human milk proteins.

Two immunostimulating peptides were isolated from human milk proteins by enzymatic digestion, the tripeptide GLF and the hexapeptide VEPIPY. These peptides increased the phagocytosis of human and murine macrophages and protected mice against Klebsiella pneumoniae infection. The present study showed that this activity may be correlated to the presence of specific binding sites on human blood phagocytic cells. The receptor molecules implicated were different for the two peptides. [3H]GLF specifically bound to PMNL and monocytes, whereas [3H]VEPIPY only bound to monocytes. The leukemic promyelocytic cell line HL-60 differentiated into granulocytes or into macrophages (depending on inducer used) coroborated these results. Specific binding of [3H]GLF on plasma membrane preparations of human PMNL (20 degrees C) was saturable and Scatchard analysis indicated two classes of binding sites: high-affinity sites of Kd 2.3 +/- 1.0 nM and Bm 60 +/- 9 fmol/mg protein and low-affinity sites of Kd 26.0 +/- 3.5 nM and Bm 208 +/- 45 fmol/mg protein. [3H]GLF binding was inhibited in a concentration-dependent manner by various analogous peptides, such as LLF, GLY, LLY and RGDGLF, but not by RGD, RGDS, VEPIPY and the chemotactic peptide f-Met-Leu-Phe (f-MLF). Only at high concentrations the direct analog MLF competed with labeled GLF. An important inhibitory effect was also observed with C1q component of the complement whereas C3 and BSA were uneffective. Specific binding of [3H]VEPIPY on monocyte membranes (20 degrees C) was saturable and Scatchard analysis was consistent with one class of binding sites of Kd 3.7 +/- 0.3 nM and Bm 150 +/- 6 fmol/mg protein.

High affinity binding sites on plasma membrane obtained from the lymphoblastoid cultured 1301 cell line for highly radioactive serum thymic factor.

The interaction of the synthetic serum thymic factor (FTS, facteur thymique sérique) with a plasma membrane preparation of human T lymphocytes from the lymphoblastoid T cell line 1301 was studied using 3H-labelled FTS (specific activity 120 Ci/mmol). The binding is temperature dependent and function of the concentration of both 3H-labelled FTS and membrane proteins. At 37 degrees C, using 1 nM of 3H-labelled FTS as steady state is observed within 80 min. The binding is reversible, specific and saturable. Scatchard analysis reveals the existence of at least two binding sites with respective Kd of the order of 0.516 +/- 0.2 nM and 110 +/- 27.8 nM with concentration of 0.186 +/- 0.045 pmol and 2.026 +/- 0.367 pmol per mg of membrane protein.

Involvement of thymosin beta 4 and endoproteinase Asp-N in the biosynthesis of the tetrapeptide AcSerAspLysPro a regulator of the hematopoietic system.

It is shown that AcSDKP a new regulator of the hematopoietic system can be generated from thymosin beta 4 by a one-step enzymatic cleavage in vitro and in vivo. AcSDKP and T beta 4 were both detected in bone marrow cells (BMC). Incubation of [3H]T beta 4 with either intact or lysed BMC led to the formation of [3H]AcSDKP whereas the labelled tetrapeptide was not degraded under these conditions. Model enzymatic degradation of T beta 4 carried out with bacterial enzymes suggests that a mammalian endoproteinase Asp-N might be involved in the formation of AcSDKP through the specific cleavage of the 4Pro-5 Asp peptide bond of T beta 4.

C-terminal amidation of neuropeptides. Gly-Lys-Arg extension an efficient precursor of C-terminal amide.

Biosynthesis of the C-terminal carboxamide group of peptide hormones was studied using comparatively pGlu-His-Pro-Gly and Glu-His-Pro-Gly-Lys-Arg as putative precursors of the tripeptide, thyroliberin (TRH). Rat hypothalamus granules were found to contain an amide group forming activity which converts both peptide substrates into TRH. Comparison of the rate of conversion of the two substrates indicated that the C-terminal dibasic extension favored a 10-fold increase in the production of amidated peptide. It is suggested that this type of structure may be present in the putative biosynthetic precursor of TRH and that it may provide a better substrate for the enzyme(s) involved in C-terminal amidation.

Peptide enzymatic microsynthesis, using carboxypeptidase Y as the catalyst: application to stepwise synthesis of Leuenkephalin.

In order to develop the use of carboxypeptidase Y (CPD-Y, EC 3.4.12.1) as a catalyst for radioactive hormonal synthesis, the stepwise synthesis of a pentapeptide, Leu-enkephalin, was studied on a microscale. Each peptide bond was formed by enzymatic catalysis, using microquantities of the precursors (amino acid or peptide esters as acyl-components and amino acid ester or amide as nucleophiles). The high condensation yields obtained suggests that CPD-Y can be a useful tool for preparation of radioactive hormonal peptides.

[Binding of a tritiated enkephalinergic analog (FK 33-824) to a mitochondrial fraction of rat brain]. / Liaison d'un analogue enképhalinergique [FK 33-824] tritié à une fraction mitochondriale de cerveau de rat.

FK-33-824 (Try-D-Ala-Gly-MePhe-Met(O)ol) is a potent enkephalin analog which has been tritium labelled with a high specific radioactivity (41 Ci/mmole). The labelled drug exhibits specific and saturable binding to rat brain crude mitochondrial fraction. Specific binding is inhibited by low concentrations of morphine, levallorphan and beta-endorphin, suggesting that FK 33-824 [3H] binds preferentially to mu opiate sites. Binding studies at equilibrium and kinetics of formation and dissociation of the labelled ligand-receptor complex indicate that FK 33-824 [3H] binds to two classes of specific sites. Their affinities are distinguishable at 0 degree (KD = 1.3 and 5.8 nM) and very close to each other at 37 degree (KD = 1.9 nM).

[Radioactive analogs of neurotensin. II. Preparation of tritiated neurotensin from a synthetic multi-unsaturated derivative]. / Analogues radioactifs de la neurotensine. II. Préparation de neurotensine tritiée à partir d'un dérivé multi-insaturé de synthèse.

In this second paper on the synthesis of neurotensin analogues as precursors for radiolabelling, solid phase synthesis of two polyunsaturated peptides, [Dah6, delta Pro7,10]-neurotensin and acetyl-[delta Pro10]-neurotensin-(8-13), are described. The first one contains one triple bond and two double bonds susceptible to tritiation in the same molecule, the second one contains one double bond in the shortest sequence having neurotensin activity. The C-terminal residue, Boc-Leu, was esterified on the chloromethyl-resin by its cesium salt. For the other amino acids a double coupling was carried out, the first one with dicyclohexylcarbodimide and the second one with the amino acid hydroxybenzotriazole ester. Acylation of the second amino acid, on the resin, presented some difficulties to achieve completeness and several acetylations and benzoylations had to be performed in order to block the last 4 per cent of free amines. It seems that these difficulties are related to some batches of chloromethyl-resin. Incorporation of both acetylenic lysine, N alpha-Boc-N epsilon-Z-L-2,6-diamino-4-hexynoic acid, whose synthesis is described, and N alpha-Boc-L-3,4-dehydroproline was without problems in this synthesis. After cleavage by hydrofluoric acid the crude peptides were purified by gel filtration on Bio-Gel P2 and ion exchange chromatography on carboxymethylcellulose (CM 52). [Dah6, delta Pro7,10]-neurotensin so obtained (51 per cent compared to starting Boc-Leu-resin) was in homogeneous form as characterized by amino acid analysis, thin layer chromatography in different systems and high performance liquid chromatography. The hydrogenation or tritiation product was identical with native neurotensin. Unsaturated derivative and neurotensin obtained after catalytic hydrogenation were as active as native neurotensin in inhibition of 125I-[Trp11]-neurotensin binding to rat brain synaptic membranes and in guinea pig ileum contractility test. Substitution of proline and lysine by their dehydro-derivatives did not affect the biological properties of neurotensin. The tritiated neurotensin (160-180 Ci/mmol) should be a good agent for biological characterization of neurotensin receptors and for investigation of the peptide metabolism.

Release of proenkephalin-derived opioid peptides from rat striatum in vitro and their rapid degradation.

In a previous paper we demonstrated that the heptapeptide [Met]enkephalyl-Arg6-Phe7 was released from rat striatal slices by high K+ concentration and rapidly degraded by peptidases, even in the presence of the neutral endopeptidase 24.11 ("enkephalinase")-inhibitor, thiorphan (0.1 microM), the angiotensin-converting enzyme inhibitor, captopril (1 microM), and the aminopeptidase inhibitor, bestatin (20 microM). In this study the pattern of degradation of exogenous [3H]heptapeptide by rat striatal slices has been studied. The angiotensin-converting enzyme and aminopeptidase(s) were partly responsible for this degradation. In addition an enzymatic activity that cleaved the Phe4-Met5 bond was involved in the degradation of the heptapeptide by striatal slices. This activity was inhibited by the dipeptide Leu-Arg (1 mM) and the tripeptide Leu-Arg-Leu (1 mM). The simultaneous presence of thiorphan (0.1 microM), captopril (1 microM), bestatin (20 microM) and Leu-Arg (1 mM) almost completely inhibited the degradation of [3H]heptapeptide by striatal slices. In the presence of these peptidase inhibitors a concomitant release of [Met]enkephalin, the heptapeptide [Met]enkephalyl-Arg6-Phe7 and the octapeptide [Met]enkephalyl-Arg6-Gly7-Leu8 was evoked by KCl or veratridine. The K+-evoked release was by a Ca2+-dependent mechanism and the release evoked by veratridine was blocked by tetrodotoxin. In both cases the ratio of [Met]enkephalin to heptapeptide amounts released was close to that found in their common precursor, proenkephalin. Thus the enkephalinergic neuron appears to be capable of synthesizing, from a unique precursor, four different putative opioid neurotransmitters, namely [Met]enkephalin, [Leu]enkephalin, the heptapeptide [Met]enkephalyl-Arg6-Phe7 and the octapeptide [Met]enkephalyl-Arg6-Gly7-Leu8, to store these peptides and to release them upon depolarization.

Biological half-life and organ distribution of [3H]1-deamino-8-D-arginine-vasopressin in the rat.

The biological half-life of synthetic, radiochemically pure, biologically active [3H]1-deamino-8-D-arginine-vasopressin (dDAVP) in rats was found to be 5.33 +/- 0.28 (S.E.M.) min in the initial, transitional, fast phase and 56.28 +/- 3.27 min in the second, slow phase. The substance accumulated to the greatest extent in the kidney and small intestine and only slightly in the adenohypophysis. The results have suggested that the extended biological half-life may play a role in the more marked and longer antidiuretic effect of dDAVP. The explanation of the poor accumulation in the adenohypophysis may be that dDAVP does not possess an effect similar to that of corticotrophin releasing factor.

Biological half-life and organ distribution of [3H]8-arginine-vasopressin in the rat.

The biological half-life of synthetic, radiochemically pure, biologically active [3H]8-arginine-vasopressin ([3H]AVP), the distribution of radioactivity among the organs and the in-vivo metabolism of the hormone were studied in the rat. The half-life calculated from the [3H]AVP radioactivities isolated from the blood was found to be 1.74 +/- 0.22 (S.D.) min in the fast phase, and 16.98 +/- 1.01 min in the slow phase. The half-lives of total radioactivity were longer in both phases. The radioactivity accumulated to the greatest extents in the adenohypophysis and small intestine. The radioactive substance was accumulated more by the kidney than by the liver, but the hormone underwent inactivation more quickly in the liver.

Biological and immunological properties of tritiated salmon calcitonin.

Tritiated salmon calcitonin was prepared by methylation of the free amino groups using tritiated sodium borohydride as precursor. Specific radioactivity was measured in competitive inhibition studies with specific anticalcitonin antibodies or tubular membranes as binding sites for calcitonin. The value observed, approx. 4 Ci/mmol, corresponded to methylation of one third of the available N-H bonds. Tritiated calcitonin prepared in this way retained full biological activity as assessed in vitro by stimulation of adenylate cyclase and in vivo by rat bioassay. Tritiated calcitonin specifically bound to isolated renal cells and nonspecific binding did not exceed 10% of total binding. Equilibrium was obtained after 15 min incubation. The hormone-receptor complex could be dissociated in the presence of an excess of unlabelled calcitonin. This data shows that tritiated calcitonin can be used in metabolic and receptor studies.

The discovery of novel vasopressin V1b receptor ligands for pharmacological, functional and structural investigations.

Until recently, pharmacological studies dealing with vasopressin receptor isoforms were severely hampered by the lack of selective agonists or antagonists that recognize the pituitary V(1b) vasopressin receptor. By contrast, many selective vasopressin-related compounds are available for characterization of the vasopressor (V(1a)) or antidiuretic (V(2)) vasopressin receptor subtypes. Recently, SSR149415, a selective nonpeptide molecule, was discovered with nanomolar affinity for mammalian V(1b) receptors and good selectivity for the other vasopressin and oxytocin receptor isoforms. This molecule exhibits potent antagonist properties both in vitro and in vivo. We also designed synthetic peptides derived from [deaminocysteine(1),arginine(8)]vasopressin (dAVP), modified in position 4 by various amino acid residues. Some of these, d[cyclohexylalanine(4)]AVP or d[lysine(4)]AVP, have a high affinity and an excellent selectivity for the human V(1b) receptor subtype. However, they exhibit a mixed V(1b)/V(2) pharmacological profile for the rat vasopressin receptor isoforms. Whatever the species considered, these peptides behave as agonists both in bioassays performed in vitro and in vivo. The d[cyclohexylalanine(4)]AVP was tritiated and represents the first selective radiolabelled ligand available for studying the human V(1b) receptors. The discovery of these new selective V(1b) agonists and V(1b) antagonist allows an accurate pharmacological characterization of all the vasopressin receptor isoforms. As emphasized in this review, attention to the vasopressin and oxytocin receptor species differences is of critical importance in studies with all vasopressin and oxytocin ligands.

Characterization of [3H] CCK4 binding sites in mouse and rat brain.

We have investigated the possible occurrence of distinct CCK8 and CCK4 binding sites in the brain by comparing the binding characteristics of [3H] CCK4 to those of the CCK8 analogue, [3H] Boc (Nle28,31]CCK27-33 (BDNL-CCK7). [3H] CCK4 and [3H] BNDL-CCK7 were shown to interact with mouse brain membranes with very similar maximal binding capacities 31.7 +/- 2.1 fmol/mg prot (KD = 3.78 +/- 0.47 nM) and 38.9 +/- 2.2 fmol/mg prot (KD = 0.26 +/- 0.02 nM) respectively. The apparent affinities of five CCK analogues for the sites labelled by both probes were almost identical. Autoradiographic studies revealed that the distribution of [3H] CCK4 binding sites in rat forebrain was the same as that of [3H] BDNL-CCK7, with high densities of receptors in the cortex, nucleus accumbens, olfactory bulb and the medial striatum, moderate densities in the amygdala, the hippocampus, several nuclei of the thalamus and hypothalamus. However in the interpenduncular nucleus where there was moderate binding of [3H]BDNL-CCK7, no [3H]CCK4 labelling was observed. These studies demonstrated the occurrence of one class of high affinity binding sites for [3H] CCK4 in mouse and rat brain, with characteristics similar to those already reported with CCK33, CCK8 and pentagastrin probes. Nevertheless the presence of a small amount of very high affinity binding sites for [3H]CCK4 cannot be excluded.

Binding and metabolism studies with [3H](D-Ala2,Leu5)enkephalinamide and [3H](D-Ala2,Pro5)enkephalinamide: evidence for a selective interaction of (D-Ala2,Pro5)enkephalinamide with mu-receptors.

We have compared the binding characteristics of [3H](D-Ala2,Leu5)enkephalinamide and [3H](D-Ala2,Pro5)enkephalinamide on a washed homogenate from mouse brain. The maximum number of binding sites for [3H](D-Ala2,Leu5)enkephalinamide was twice that obtained with [3H](D-Ala2,Pro5)enkephalinamide. Hill slopes of the displacement curves for opiates and (D-Ala2,Pro5)enkephalinamide obtained against [3H](D-Ala2,Leu5)enkephalinamide were considerably lower than that of (D-Ala2,Met5)enkephalinamide and (D-Ala2,Leu5)enkephalinamide. In contrast, Hill slopes for morphine, ketocyclazocine and (D-Ala2,Leu5)enkephalinamide were similar for the displacement of [3H](D-Ala2,Pro5)enkephalinamide. These results are discussed and interpreted in terms of a selective interaction for (D-Ala2,Pro5)enkephalinamide with mu-receptors and a similar affinity of (D-Ala2,Leu5)enkephalinamide for mu- and delta-receptors. Since metabolism studies did not show any difference in the degradation rate of these two enkephalin analogs, the discrepancy between in vitro and in vivo activity cannot result from the catabolism processes. Another possible explanation is proposed.

Histamine H3 receptor binding sites in rat brain membranes: modulations by guanine nucleotides and divalent cations.

The binding of [3H](R)alpha-methylhistamine, a potent and specific agonist at histamine H3 receptors, was investigated with membranes of rat cerebral cortex. In phosphate buffer the specific binding defined with thioperamide, an H3 receptor antagonist, displayed characters of reversibility and saturability with a Bmax of approximately 30 fmol/mg protein. The KD, derived from either dissociation/association rates or saturation kinetics at equilibrium, was approximately 0.5 nM at 25 degrees C. Competition studies indicated that the binding occurred with a stereoselectivity and pharmacological specificity similar to that of functional H3 autoreceptors regulating histamine release in brain slices. However, whereas the potency of antagonists was closely similar in the two assay systems, that of agonists was approximately 10-fold higher in the binding assay. Among antagonists burimamide was the only one to compete with a pseudo-Hill coefficient significantly different from unity (nH = 0.48 +/- 0.03), indicating a possible heterogeneity among binding sites. The presence of 2.6 mM Ca2+, in a modified Krebs-Ringer medium, promoted the conversion of a larger fraction of sites into a low-affinity component with a KD of 16 nM. The presence of guanylnucleotides in the Krebs-Ringer medium with Ca2+ abolished the binding to this low-affinity component whereas in a phosphate buffer only the KD was slightly increased. It is concluded that the H3 receptor, like many other amine receptors, is coupled to its still unidentified effector system via a G-protein and regulated by Ca2+. However, unlike the latter, the H3 receptor is down-regulated by the divalent cation.

Autoradiographic distribution of mu and delta opiate receptors in rat brain using highly selective ligands.

We have characterized the autoradiographic distribution of mu and delta opiate receptors in rat brain using [3H]-Tyr-D-Ala-Gly-NMe-Phe-Gly-ol (DAGO) and [3H]D-Thr2, Thr6leu-enkephalin (DTLET), two highly selective ligands for mu and delta opiate receptors, respectively. Both ligands label an apparent single class of sites. Ligand selectivity shows that [3H]DAGO and [3H]DTLET are more selective ligands for their respective receptors than [3H]dihydromorphine (DHM) and [3H][D-Ala2,D-Leu5]enkephalin (DADLE). Autoradiographically, mu opiate receptors are highly discretely distributed in certain areas. High densities are found in "patches" in caudate-putamen, nucleus accumbens, cingulate cortex, habenula, various thalamic nuclei, amygdala, superior colliculus, and interpeduncular nucleus. Delta opiate receptors are mostly concentrated in the external plexiform layers of the olfactory bulb, caudate-putamen, olfactory tubercule, claustrum, and deep layers (layers V and VI) of the cortex. Since [3H]DAGO and [3H]DTLET are more selective ligands for their respective opiate receptor subtypes, they should be better ligands to characterize the fine distribution of mu and delta opiate receptors.

Release of the heptapeptide Met-enkephalin-Arg6-Phe7 and of the octapeptide Met-enkephalin-Arg6-Gly7-Leu8 from rat striatum in vitro and their rapid inactivation.

The heptapeptide Met-enkephalin-Arg6-Phe7 (MERF) and the octapeptide Met-enkephalin-Arg6-Gly7-Leu8 (MERGL) are potent opioid peptides present in the sequence of proenkephalin, the common precursor of Met- and Leu-enkephalin (ME and LE). We demonstrate that MERF and MERGL are released concomitantly with ME and LE from rat striatal slices following a depolarisation by K+. This release is a Ca2+-dependent process. While the ratios of ME to LE, MERF and MERGL found in the tissue (ME/LE = 2.6; ME/MERF = 3.1; ME/MERGL = 4.5) are in good agreement with the ratios found in the proenkephalin molecule (ME:LE:MERF:MERGL = 4:1:1:1), the amounts of MERF and MERGL recovered from the medium are low compared to those of ME and LE, suggesting a rapid degradation of released MERF and MERGL. In fact, when incubated with striatal slices, (3H-Tyr)-MERF is rapidly degraded by four classes of peptidases: the "enkephalinase", the angiotensin-converting enzyme (ACE), aminopeptidase(s) and an endopeptidase releasing the tetrapeptide Tyr-Gly-Gly-Phe (YGGF). Whereas the activities of the three former peptidases are reduced or abolished in the presence of thiorphan (0.1 microM), captopril (1 microM) and bestatin (20 microM), the amount of YGGF formed by the endopeptidase is not reduced in these conditions but actually increased.

[3H] Boc [Nle28, 31]CCK27-33, a new highly labelled ligand for CCK receptors: binding on brain and on pancreas.

Preliminary results on the binding of [3H]Boc[Nle28, 31]CCK27-33, designated [3H]Boc[diNle]CCK7, on mouse brain and rat pancreas membranes are presented. This new ligand for CCK receptors possesses a high specific activity (144 Ci/mmole), and binds in a saturable manner to mouse brain (Kd = 0.49 nM, Bmax = 49 fmoles/mg protein) and rat pancreas (Kd = 4.4 nM, Bmax = 696 fmoles/mg protein). Unlabelled Boc[diNle]CCK7 displaces [125I]CCK8 from its binding sites on mouse brain membranes with a high affinity, slightly superior to that of CCK8. The order of potencies to displace [3H]Boc[diNle]CCK7 from its binding sites was the same in mouse brain and rat pancreas: [3H]Boc[diNle]CCK7 greater than CCK8, Boc-CCK7 greater than non-sulfated CCK8, the pancreas binding sites being more discriminative than the brain binding sites. Thus, [3H]Boc[diNle]CCK7 is a very promising new probe for the characterization of CCK receptors and their interaction with different CCK fragments.

Brain passage of BUBU, a highly selective and potent agonist for delta opioid receptors: in vivo binding and mu versus delta receptors occupancy.

The peptidase-resistance and bioavailability of BUBU [H-Tyr-D.Ser(OtBu)-Gly-Phe-Leu-Thr(OtBu)-OH], a highly selective and potent agonist of the delta opioid receptor, have been investigated in vitro and in vivo. In vitro at 37 degrees C, the peptide was fully resistant to degradation by rat serum and strongly resistant to degradation by rat brain membranes. In vivo 0.065% of the dose of [3H]BUBU injected intravenously to the mouse was present 15 min later in the brain. The percentage determined for [3H]DAGO [H-Tyr-D.Ala-Gly-(NMe)Phe-Gly-ol], a selective ligand for mu sites, was 0.038%. Specific binding to mouse brain membranes, determined after intracerebroventricular injection of [3H]BUBU, was saturable and a high affinity (KDapp = 25 pmol) was evaluated for the delta-agonist. Competition experiments showed that BUBU is a selective ligand for delta receptors in vivo. Comparison of the analgesic potency (hot plate test) of ICV or IV administered increasing doses of BUBU and DAGO with their in vivo binding properties supports the preferential involvement of mu receptors in supraspinal analgesia. BUBU also induced an increase in spontaneous locomotion after IV administration at a dose lower than that which produced analgesia. The quantitative results obtained in the present study demonstrate that BUBU and DAGO could be used to characterize the pharmacological responses induced by selective stimulation of delta and mu receptors after systemic administration.