Identification of distinct binding site subunits of mu and delta opioid receptors

Iodinated human beta-endorphin was affinity-cross-linked to opioid receptors present in membrane preparations from bovine frontal cortex, bovine striatum, guinea pig whole brain, and rat thalamus. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis followed by autoradiography revealed covalently labeled peptides of 65, 53, 41, and 38 kilodaltons (kDa). The 65- and 38-kDa peptides were present in all four tissues. The 41-kDa peptide was seen only in bovine caudate and guinea pig whole brain while the 53-kDa peptide was absent in rat thalamus. All four labeled peptides were constituents of opioid receptors since their labeling was fully suppressed by the presence of excess opiates, such as bremazocine, during binding. The distribution and levels of the labeled species in the brain tissues examined and, in earlier work, in the neuroblastoma X glioma NG 108-15 cell line suggested that the 65-kDa peptide is a binding component of mu receptors while the 53-kDa peptide is a binding subunit of delta receptors. This result was strongly supported by the finding that the labeling of the 65-kDa peptide is selectively reduced by the presence of the highly mu-selective ligand Tyr-D-Ala-Gly-(N-Me)Phe-Gly-ol (DAMGE) during binding, while while the labeling of the 53-kDa peptide is selectively reduced or eliminated by the highly mu-selective ligand [D-Pen2, D-Pen5]enkephalin (DPDPE). The labeling of the 41- and 38-kDa bands was reduced by either DAMGE or DPDPE. The relationship of these lower molecular weight opioid-binding peptides to mu and delta receptors is not understood. Several possible explanations are presented.     

Characterization of beta-125I-endorphin cross-linked proteins in NG108-15 cell membranes. A 25-kilodalton protein with properties of delta-opioid-binding site.

Cross-linking of beta-125I-endorphin to NG108-15 cell membranes labeled bands with molecular masses of 55, 35, and 25 kDa on sodium dodecyl sulfate-polyacrylamide gel electrophoresis. We applied several criteria to evaluate the relevance of these cross-linked bands to delta-opioid receptors, including selectivity, stereospecificity, affinity, G-protein coupling, down-regulation, and correlation with opioid receptor level in different well-characterized cell lines. Only the 25 kDa protein adequately fulfilled all these criteria. Thus, cross-linking to the 25-kDa band was selectively inhibited by ligands with delta-opioid affinity, but not by mu-opioid, kappa-opioid, or optically inactive opioid ligands or by non-opioid ligands. Based on inhibition of cross-linking, we calculated an affinity of [D-Ala2,D-Leu5]enkephalin binding to the 25-kDa and (Kd = 6 nM) that is similar to that reported for [D-Ala2,D-Leu5]enkephalin binding to NG108-15 membranes; this affinity decreased approximately 10-fold in the presence of Na+/guanyl-5'-yl imidodiphosphate. Chronic agonist treatment of NG108-15 cells reduced cross-linking to the 25-kDa band, but not to others, in a manner parallel to down-regulation of opioid receptors. Finally, the amount of the 25-kDa band was roughly proportional to the level of opioid receptors present in N18TG2, NS20Y, ST7-3, and ST8-4 cells. The 25-kDa band was absent in PC12h, NIH3T3, and C6BU1 cells as well as in liver, all of which had no detectable opioid binding.     

Rapid agonist-induced loss of 125I-beta-endorphin opioid receptor sites in NG108-15, but not SK-N-SH neuroblastoma cells.

We have measured mu and delta opioid receptor sites on intact SK-N-SH and NG108-15 neuroblastoma cells, respectively, in culture. Use of 125I-beta-endorphin (beta E) as a tracer, together with beta E(6-31) to block high-affinity non-opioid binding in both cell lines, permitted the measurement of cell surface mu and delta opioid receptor sites. Labeling was at delta sites in NG108-15 cells and predominantly at mu sites in SK-N-SH cells. Pretreatment with the mu and delta agonist, DADLE, caused a rapid loss of cell surface delta receptor sites in NG108-15 cells, but failed to reduce significantly mu receptor density in SK-N-SH cells.     

125I-beta-endorphin binding to neuroblastoma X glioma NG108-15 cells: distribution of delta opioid receptors.

The mouse neuroblastoma x rat glioma hybrid NG108-15 was previously shown to express delta opioid receptors. Because neuroblastoma cells display different phenotypes and cloned cell lines are heterogenous, we studied the characteristics and distribution of human 125I-beta-endorphin (125I-beta E) binding sites in cultures of NG108-15 cells with the use of micro-autoradiography and light microscopy. 125I-beta E labeled delta sites in NG108-15 in the presence of the non-opioid blocking peptide, beta-endorphin (6-31) (beta E (6-31)). Silver grains resulting from 125I-beta E binding to the opioid sites occurred in diffuse patches over several cells, with preferential location in dense cell patches. Pretreatment of NG108-15 with the delta agonist DADLE, previously shown to decrease beta E binding to delta sites on intact cells, also reduced silver grain density; however, some cells located in dense cell clusters were resistant to substantial agonist induced loss of labeling. These results suggest that delta opioid binding has a heterogenous cellular distribution in NG108.     

Target size analysis of opioid receptors. No difference between receptor types, but discrimination between two receptor states

Target size analysis of opioid receptor is complicated by the presence of multi-exponential inactivation curves. Irradiation of intact frozen tissue proved essential to eliminate such artifacts, due to indirect irradiation effects. Upon irradiation condition, opioid binding activity was inactivated in a single mono-exponential manner. Identical inactivation curves were obtained for mu, delta and kappa binding activities in brain membranes from rat, guinea-pig and frog and in NG 108-15 cells: the molecular mass obtained was 98 +/- 2 kDa. However, when opioid binding was assayed in the presence of Na+, Mg2+ and GTP, the molecular mass was found to be only 56 +/- 4.4 kDa. We suggest that the opioid recognition site comprises a unit of 56 kDa and that in the absence of Na+, Mg2+ and GTP an additional membrane component of 40-44 kDa is necessary for high-affinity opioid binding.     

A photoactivatable naltrexone derivative labels glycoproteins of different molecular weight corresponding to the mu- and kappa-opioid receptors.

The synthesis and characterization of a novel opioid receptor photoaffinity probe [3H]naltrexyl urea phenylazido derivative ([3H]NUPA) is described. In the absence of light, [3H]NUPA binds with high affinity in a reversible and saturable manner to rat brain and guinea pig cerebellum membranes. Dissociation constants and binding capacities (Scatchard plots) are 0.11 nM and 250 fmol/mg of protein for rat brain and 0.24 nM and 135 fmol/mg of protein for guinea pig cerebellum. Competition experiments indicate that this ligand interacts with high affinity at both mu- and kappa-opioid binding sites while exhibiting low affinity at delta sites (Ki = 21 nM). On irradiation, [3H]NUPA incorporates irreversibly into rat brain and guinea pig cerebellum membranes. SDS gel electrophoresis of rat brain membranes reveals specific photolabeling of a 67-kDa molecular mass band. Conversely, a major component of 58 kDa and a minor component of 36 kDa are obtained from [3H]NUPA-labeled guinea pig cerebellum membranes. Different photolabeling patterns are obtained in rat brain (mu/delta/kappa, 4/5/1) and guinea pig cerebellum (mu+delta/kappa, 1,5/8,5) membranes in the presence of selective opioid ligands indicating labeling of mu and kappa sites, respectively. Thus, [3H]NUPA behaves as an efficient photoaffinity probe of mu- and kappa-opioid receptors, which are probably represented by distinct glycoproteins of 67 and 58 kDa, respectively.     

Dimeric pentapeptide enkephalin: a novel probe of delta opiate receptors

A dimeric pentapeptide enkephalin (DPE2) consisting of two molecules of [D-Ala 2, Leu 5] enkephalin linked at C-terminal leucine with ethylenediamine, (H-Tyr-D-Ala-Gly-Phe-Leu-NH-Ch2)2 is a bivalent ligand for the delta enkephalin receptors of rat brain and neuroblastoma-glioma hybrid (NG108-15) cells. This new enkephalin analog shows dramatically increased affinity in radioligand assays using whole brain membranes when delta but not mu specific radioligands are employed. When membranes from NG108-15 cells are used, the dimer shows greatly increased activity irrespective of the mu or delta specificity of the tracer. The dimer DPE2 shows a four-fold, "sodium shift" in its IC50 for competition with [3H]naloxone, suggestive of agonist behavior. Agonist activity was confirmed by demonstrating that DPE2 inhibits cyclic AMP production in prostaglandin E1 stimulated NG108-15 cells, and by demonstrating very high potency in the mouse vas deferens bioassay. DPE2 binds to the same delta sites as the delta-selective monomer [D-Ala2, D-Leu5] enkephalin, since the two ligands show complete crossdisplacement. Radiolabeled 3H-DPE2 shows a five-fold higher affinity constant, a 2.5-fold higher association rate constant, and a two-fold lower dissociation rate than the monomer. These results are consistent with the hypothesis that the dimeric pentapeptide enkephalin can bridge two delta receptors. This enkephalin dimer provides a valuable new probe of opiate receptors and their organization in cell membranes.     

Biphasic competition between opiates and enkephalins: does it indicate the existence of a common high affinity ("mu-1") binding site?

Displacement from brain membranes of labeled opiates by low concentrations of enkephalins and of labeled enkephalins by low concentrations of opiates has been previously explained by the existence of a common high affinity site termed mu-1. An alternative interpretation of the same results is that the trough seen in the low concentration zone of the displacement curves represents cross binding of mu and delta opioid ligands to delta and mu receptors, respectively. In three sets of experiments with brain membranes, the size of the trough is shown to be dependent on the labeled ligand used: The ratio between the size of troughs seen with [3H]D-Ala, D-Leu enkephalin and with [3H]morphine varies with experimental conditions (storage of membranes at 4 degrees C for 72 h), with ratio of mu:delta receptors (e.g. in thalamus and cortex which are enriched in mu and delta sites, respectively) and with pretreatment of membranes with naloxonazine. These results can not be explained by a common high affinity site, but rather by binding of [3H]D-Ala, D-Leu enkephalin to mu and of [3H]morphine to delta opioid receptors.     

Synthesis and binding properties to DNA and to opioid receptors of enkephalin-ellipticinium conjugates

In order to specifically direct cytotoxic agents against tumor cells bearing delta opioid receptors, the DNA intercalating agents ellipticine and 9-OH-ellipticine were coupled by quaternarization of the pyridine nitrogen to an enkephalin modified pentapeptide through a short chemical linker. The ellipticine ring of these conjugates was shown to intercalate into DNA, with DNA affinity constants close to those of the non-conjugated ellipticines. Despite the addition of a polycyclic ring to the C-terminal amino acid, the D-Ala2-D-Leu5-enkephalin-ellipticine conjugates bind to the opioid receptor from rat brain and NG 108-15 cells with an affinity constant close to 10(8) M-1. Other derivatives were synthesized as a control using a tripeptide which does not bind to the opioid receptor.     

Binding characteristics of a series of dimeric tripeptide enkephalins for delta opiate receptors in rat brain and NG108-15 cells

The N-terminal tripeptide enkephalin analogue, Tyr-D-Ala-Gly, was dimerized at the C-terminus systematically with a series of alpha,omega-diaminoalkanes, NH2-(CH2)n-NH2 (n = 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, and 22). The binding affinities of dimers for delta opiate receptors in rat brain were evaluated and compared with those for delta receptors in NG108-15 cells. Although the monomeric tripeptide amide was almost inactive, dimers showed a dramatic increase in binding affinity (8-900 times). The enhancement of affinity was apparently related to the number of methylene chains in the crosslinking spacer moiety, and it was maximal at n = 14-18 in the rat brain. In NG cells the activity increased progressively from n = 2 to n = 22 without reaching any apparent peak. These results suggest that delta receptors in rat brain and NG cells may have slight structural differences.     

Different molecular weight forms of opioid receptors revealed by polyclonal antibodies

Polyclonal antibodies were raised against a purified opioid receptor from bovine brain (Cho, et. al., 1986), and shown to inhibit 3H-diprenorphine binding to this receptor in a dose-dependent fashion. These antibodies were then used to characterize opioid-binding material present in rat brain and in NG108-15 neuroblastoma-glioma hybrid cells. Western blot analysis revealed that the antibodies reacted with a single species of 58,000 molecular weight in rat brain membranes; this closely corresponds in size to the bovine opioid receptor used to raise the antibodies. In contrast, the polyclonal antibodies reacted with a 45,000 molecular weight species in NG108-15 neuroblastoma-glioma hybrid cells; moreover, this band was specifically reduced in NG108-15 cells in which opioid receptors had been down-regulated by incubation with D-ala2-D-leu5-enkephalin for 24 hours. Thus at least two distinct opioid receptor molecules have been identified, which have antigenic similarities.     

Monoclonal antiidiotypic antibodies against delta opioid receptors as an electron microscopy probe.

Four different rat monoclonal antibodies were produced against delta opioid receptor using an antiidiotypic approach in which antibodies directed against the opioid agonist DADLE were used as immunogen. In the first step, seven hybridomas were selected on the basis of their ability to inhibit the DADLE-anti-DADLE antibody interaction. After purification from ascitic fluids, these monoclonal antibodies were characterized. Four antiidiotypic antibodies, named 5, 11, 16, and 51, directed toward different epitopes, recognized the delta opioid receptor: (i) they bound directly to the NG108-15 cells, (ii) they inhibited the [3H]DADLE binding on the NG108-15 cells, (iii) they immunoprecipitated a 52,500 dalton protein present on the surface of the NG108-15 cells. The four monoclonal antiidiotypic anti-opioid receptor antibodies were used to immunocytologically detect the opioid receptors under light and electron microscopy in the rat spinal cord. The regional distribution of the immunoreactivity corresponded to layers known to be rich delta opioid receptor subtype. Moreover, at the ultrastructural level, the labeling was located mainly on plasma membranes, especially on non-synaptic zones. Our results show that monoclonal antiidiotypic antibodies constitute a valuable tool for visualizing cell surface receptors.     

Covalent labeling of opioid receptors with 3H-D-Ala2-Leu5-enkephalin chloromethyl ketone. I. Binding characteristics in rat brain membranes

The chloromethyl ketone derivative of D-Ala2-Leu5-enkephalin was synthesized in a radioactive form, and the resulting compound (3H-DALECK) was used to label opioid receptors. 3H-DALECK binds with high affinity, specificity and saturability to rat brain membranes. The number of sites labeled is 130 fmoles/mg protein. Unlabeled opioids inhibited the binding of 3H-DALECK; etorphine and DAGO being most potent. A 10-fold preference for mu sites over delta was seen in site-specific competition experiments; while DALECK displayed low affinity for kappa sites of rat brain. DALECK irreversibly blocked a certain population of sites. Approximately 40% of 3H-DALECK binding at 15 min, and 60% at 60 min association time did not dissociate in the presence of a large excess of unlabeled DALECK and was resistant to washing. Autoradiography performed after SDS-PAGE revealed specific alkylation of proteins with molecular weight of 74, 65, 56, 43 and 34 kD. These results demonstrate the applicability of using 3H-DALECK to covalently label opioid receptors.     

Isolation and partial characterization of high affinity laminin receptors in neural cells

Neural cells in culture (NG-108, PC12, chick dorsal root ganglion, chick spinal cord, and rat astrocytes) bind laminin with an apparent Kd of congruent to 10(-9) M. Laminin affinity chromatography of chick brain membranes washed with 150 mM NaCl and eluted with 0.2 M glycine buffer, pH 3.5, yields a single protein with an apparent molecular mass of 67 kDa by sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions. Isoelectric focusing and peptide mapping indicate that the 67-kDa protein is distinct from bovine serum albumin (68 kDa) but indistinguishable from high affinity laminin receptors isolated from skeletal muscle. After electroblotting onto nitrocellulose paper and probing with 125I-laminin, this putative laminin receptor binds laminin specifically (100 ng/ml). A second protein (congruent to 120-140 kDa) is also detected with 125I-laminin (100 ng/ml) in the laminin affinity-purified membrane proteins. Both 67- and congruent to 120-140-kDa proteins can be laminin affinity-purified from cultures enriched for neurons (greater than 90%) following metabolic labeling with [35S]methionine. Our data suggest that neural cells (dorsal root ganglion, central nervous system neurons, astrocytes, and several neural cell lines) have high affinity binding sites for laminin and that two membrane proteins, 67- and congruent to 120-140-kDa, are responsible at least in part for this binding.     

Insulin-like growth factor-binding protein from human plasma. Purification and characterization

Insulin-like growth factor (IGF)-binding protein (BP) has been purified from Cohn fraction IV of human plasma by acidification, ion exchange to remove endogenous ligands, and affinity chromatography on agarose-IGF-II. The pure protein appeared as a single peak by high performance reverse-phase and gel permeation chromatography (molecular mass, 45-50 kDa), but on sodium dodecyl sulfate-polyacrylamide gel electrophoresis gave a major band at 53 kDa and a minor band at 47 kDa, unreduced, or 43 and 40 kDa, respectively, reduced. The two bands stained for both protein and carbohydrate. After storage at 2 degrees C for 5 months at pH 3, two additional bands, at 26 and 22 kDa on unreduced gels, were also present. Autoradiography after affinity labeling with IGF-I or IGF-II tracer revealed a single labeled band of 61 kDa. BP, quantitated using a specific radioimmunoassay, was retained by agarose-immobilized IGF-I, IGF-II, concanavalin A, and wheat germ lectin, but not Helix pomatia lectin. Competitive binding curves using pure BP and human IGF-I and IGF-II as both labeled and unlabeled ligands indicated association constants of 2-3 X 10(10) liters/mol for both peptides, with a slightly higher affinity for IGF-II than IGF-I, and 0.9 binding sites for either peptide per 53-kDa protein. The exact relationship of this acid-stable IGF BP to the 150-kDa complex from which it is derived remains to be determined.     

Exploration of universal cysteines in the binding sites of three opioid receptor subtypes by disulfide-bonding affinity labeling with chemically activated thiol-containing dynorphin A analogs.

A ligand containing an SNpys group, i.e. 3-nitro-2-pyridinesulfenyl linked to a mercapto (or thiol) group, can bind covalently to a free mercapto group to form a disulfide bond via the thiol-disulfide exchange reaction. This SNpys chemistry has been successfully applied to the discriminative affinity labeling of mu and delta opioid receptors with SNpys-containing enkephalins [Yasunaga, T. et al. (1996) J. Biochem. 120, 459-465]. In order to explore the mercapto groups conserved at or near the ligand binding sites of three opioid receptor subtypes, we synthesized two Cys(Npys)-containing analogs of dynorphin A, namely, [D-Ala2, Cys(Npys)8]dynorphin A-(1-9) amide (1) and [D-Ala2, Cys(Npys)12]dynorphin A-(1-13) amide (2). When rat (mu and delta) or guinea pig (kappa) brain membranes were incubated with these Cys(Npys)-containing dynorphin A analogs and then assayed for inhibition of the binding of DAGO (mu), deltorphin II (delta), and U-69593 (kappa), the number of receptors decreased sharply, depending upon the concentrations of these Cys(Npys)-containing dynorphin A analogs. It was found that dynorphin A analogs 1 and 2 effectively label mu receptors (EC50 = 27-33 nM), but also label delta receptors fairly well (160-180 nM). However, for kappa receptors they showed drastically different potencies as to affinity labeling; i.e., EC50 = 210 nM for analog 1, but 10,000 nM for analog 2. Analog 2 labeled kappa receptors about 50 times more weakly than analog 1. These results suggested that dynorphin A analog 1 labels the Cys residues conserved in mu, delta, and kappa receptors, whereas analog 2 only labels the Cys residues conserved in mu and delta receptors.     

Discrete mapping of brain Mu and delta opioid receptors using selective peptides: quantitative autoradiography, species differences and comparison with kappa receptors

The opioid peptides, [3H]DAGO and [3H]DPDPE, bound to rat and guinea pig brain homogenates with a high, nanomolar affinity and to a high density of mu and delta receptors, respectively. [3H]DAGO binding to mu receptors was competitively inhibited by unlabelled opioids with the following rank order of potency: DAGO greater than morphine greater than DADLE greater than naloxone greater than etorphine much greater than U50488 much greater than DPDPE. In contrast, [3H]DPDPE binding to delta receptors was inhibited by compounds with the following rank order of potency: DPDPE greater than DADLE greater than etorphine greater than dynorphin(1-8) greater than naloxone much greater than U50488 much greater than DAGO. These profiles were consistent with specific labelling of the mu and delta opioid receptors, respectively. In vitro autoradiographic techniques coupled with computer-assisted image analyses revealed a discrete but differential anatomical localization of mu and delta receptors in the rat and guinea pig brain. In general, mu and delta receptor density in the rat exceeded that in the guinea pig brain and differed markedly from that of kappa receptors in these species. However, while mu receptors were distributed throughout the brain with "hotspots" in the fore-, mid- and hindbrain of the two rodents, the delta sites were relatively diffusely distributed, and were mainly concentrated in the forebrain with particularly high levels within the olfactory bulb (OB), n. accumbens and striatum. Notable regions of high density of mu receptors in the rat and guinea pig brain were the accessory olfactory bulb, striatal "patches" and "streaks," amygdaloid nuclei, ventral hippocampal subiculum and dentate gyrus, numerous thalamic nuclei, geniculate bodies, central grey, superior and inferior colliculi, solitary and pontine nuclei and s. nigra. Tissues of high delta receptor concentration included, OB (external plexiform layer), striatum, n. accumbens, amygdala and cortex (layers I-II and V-VI). Delta receptors in the guinea pig were, in general, similarly distributed to the rat, but in contrast to the latter, the hindbrain regions such as the thalamus, geniculate bodies, central grey and superior and inferior colliculi of the guinea pig were apparently more enriched than the rat. These patterns of mu and delta site distribution differed dramatically from that of the kappa opioid sites in these species studied with the peptide [125I]dynorphin(1-8).     

The erythropoietin receptor of rat erythroid progenitor lens. Characterization and affinity cross-linkage

Commercially available 125I-labeled erythropoietin, obtained by genetic engineering from a human gene, was used to characterize receptors for this hormone on the cell surface of rat erythroid progenitor cells. A low number of high affinity binding sites (487 +/- 32 sites/cell, Kd = 167 +/- 14 pm) were found. Nonerythroid cells and erythrocytes did not exhibit specific binding. The high affinity binding was reversible and displaced by unlabeled erythropoietin, but not by other hormones and growth factors. After incubation at 37 degrees C, nearly 35% of the specifically bound erythropoietin seemed to be internalized, as judged by resistance to acidic buffer treatment. Thus, binding showed characteristics of a hormone-receptor association. 125I-Erythropoietin-labeled cells were treated with the bifunctional reagent dissucinimidyl suberate. Analysis of the cellular extracts by polyacrylamide gel electrophoresis under denaturing and reducing conditions revealed that erythropoietin can be cross-linked to two molecules of 94 and 78 kDa, respectively. Both labeled bands disappeared when the cells were labeled in the presence of an excess of unlabeled erythropoietin. Under nonreducing conditions, a cross-linked band of 230-255 kDa was observed. The relationships between these bands are discussed.     

Affinity labeling of delta opioid receptors by an enkephalin-derivative alkylating agent, DSLET-Mal

Opioid binding properties of Tyr-D-Ser-Gly-Phe-Leu-Thr-NH-NH-Gly-Mal (DSLET-Mal), a novel enkephalin-framed affinity label, was determined in rat brain membranes. In competition studies the ligand showed high affinity for the delta opioid sites, labelled by [(3)H][Ile(5,6)]deltorphin II (K(i) = 8 nM), whereas its binding to the mu ([(3)H]DAMGO) and kappa ([(3)H]EKC) sites was weaker. Preincubation of the rat brain membranes with DSLET-Mal at micromolar concentrations resulted in a wash-resistant and dose-dependent inhibition of the [(3)H][Ile(5,6)]deltorphin II binding sites (96% blocking at 10 microM concentration). Intracerebroventricular (ICV) administration of DSLET-Mal reduced the density of delta opioid receptors and had no effect on mu and kappa receptors, as determined by saturation binding studies. [Ile(5, 6)]deltorphin II-stimulated [(35)S]GTPgammaS binding was determined in membrane preparations of different brain areas of the ICV-treated animals. In both frontal cortex and hippocampus DSLET-Mal significantly decreased G protein activation by the delta agonist, having no effect on DAMGO stimulated [(35)S]GTPgammaS binding. DSLET-Mal had qualitatively similar effects on both receptor binding and G protein activation. These characteristics of the compound studied suggest that DSLET-Mal can serve as an affinity label for further studies of the delta-opioid receptors.