Abstract
G protein activation by the agonist-occupied nociceptin- (orphanin FQ-) receptor in rat cerebral cortex was studied by characterizing the nociceptin-stimulated binding of the radiolabeled guanylyl triphosphate (GTP) analog 35S-guanylyl-5′-O-(γ-thio)-triphosphate (GTPγS). Using 3H-Tyr14- and125I-Tyr14-nociceptin in saturation and displacement receptor binding studies, a single high-affinity (Kd 21.6–116.7 pM) and high-capacity binding site for nociceptin (orphanin FQ) in membranes and sections of rat cerebral cortex was identified. Stable GTP analogs and NaCl lowered the affinity only moderately by 2- to 3-fold, but under these conditions nociceptin stimulated the binding of35S-GTPγS to G proteins in the membranes with a potency about 100-fold lower (EC50 9.11 nM). It was estimated that this stimulation was due to a 29-fold increase in the affinity from Kd 45.8 to 1.57 nM of only about 6.5% of the basal binding sites for GTPγS, and that at least 10 G protein binding sites could be stimulated by one receptor site. The link of this nociceptin-stimulated binding of GTP to the nociceptin receptor was further evidenced by the specificity of stimulation, as seen with nociceptin, nociceptin(1–13),d-Ala7-nociceptin and nociceptin(1–9), which paralleled that of their receptor affinities. Furthermore, the distribution in rat brain regions of the binding of35S-GTPγS stimulated by nociceptin differed from that stimulated by the mu opioid agonist [d-Ala2, N-Me-Phe4, Gly5-ol)]-enkephalin. Especially, no stimulation by nociceptin was observed in caudate putamen, where also the absence of ORL1 receptors had been reported. The putative coupling of the high-affinity nociceptin receptor to the low-potency stimulation of GTPγS binding in rat cerebral cortex might be explained by the switch of a low part of occupied nociceptin binding sites to a very low-affinity state being stabilized at high peptide concentrations and catalytically stimulating the GTP binding.
Despite the sequence homology of the G protein-coupled ORL1 to themu, delta and kappa opioid receptor, none of the opioid receptor ligands exhibited activity at ORL1 expressed in COS or CHO cells (Mollereau et al., 1994;Lachowicz et al., 1995), and a novel heptadecapeptide named orphanin FQ (Reinscheid et al., 1995) or nociceptin (Meunieret al., 1995) has been identified to be an endogenous ligand for ORL1. Nociceptin was found to specifically activate ORL1 expressed in cells (Meunier et al., 1995; Reinscheid et al., 1995, 1996; Butour et al., 1997) and a nociceptin receptor in neuroblastoma x glioma NG108–15 hybrid cells (Ma et al., 1997) and in membranes from mouse brain (Mathis et al., 1997) to inhibit the adenylate cyclase. However, nociceptin has been shown to bind to any of the opioid receptors only with extremely low affinity (Shimohigashi et al., 1996).
Furthermore, the pharmacological effects of nociceptin in various species differed from those of the opioids (reviewed by Henderson and McKnight, 1997). After isolation of nociceptin, a hyperalgesic response to the peptide when injected intracerebrovascularly into mouse was found (Reinscheid et al., 1995; Meunier et al., 1995). After further studies, however, the interpretation of the actions of nociceptin has become more complicated, especially when it was observed that intrathecally administered peptide produced an analgesic, rather than a hyperalgesic, response (see review byHenderson and McKnight, 1997) and that the acute hyperalgesia induced by intracerebrovascularly injected nociceptin was followed by a delayed analgesic response (Rossi et al., 1996). Nevertheless, it seems now to be clear that nociceptin supraspinally reverses opioid-mediated antinociception (Mogil et al., 1996; Tianet al., 1997) and produces persistent hyperalgesia when the analgesic effect of the endogenous opioids is blocked by opioid antagonists (Rossi et al., 1997).
Specific binding sites for nociceptin, thought to mediate the actions of the peptide in brain, have been found in membranes obtained from rat (Dooley and Houghten, 1996; Makman et al., 1997; Ardatiet al., 1997), guinea pig (Shimohigashi et al., 1996) and mouse brain (Mathis et al., 1997) as well as from neuroblastoma x glioma NG108–15 hybrid cells (Ma et al., 1997). Remarkably, their affinities reported differed widely (Kd 0.004–5.0 nM), which was even true for those reported (Kd 0.021–1.2 nM) for ORL1 expressed in CHO (Reinscheid et al., 1995, 1996;Ardati et al., 1997; Butour et al., 1997; Fukudaet al., 1997) or HEK cells (Ardati et al., 1997;Shimohigashi et al., 1996). The presence of more than one receptor affinity state for agonist binding to the nociceptin receptor might indicate different functional interaction of the G protein-coupled receptor with the signal-transducing G proteins.
The aim of this study was, therefore, to investigate G protein activation by the activated nociceptin receptor in rat brain. In membrane preparations, such an activation process is frequently monitored by studying the agonist stimulation of high-affinity GTPase activity of the G protein α-subunit or the shift in receptor binding affinity by a stable GTP analog. Whereas GTPase activity reflects only the steady-state kinetics of the overall G protein activity cycle, the shift in affinity by GTP is rather variable and sometimes very small in size. Therefore, to get insights in the process of coupling of the nociceptin receptor to the G proteins in a more quantitative manner, we characterized the binding of the radiolabeled GTP analog35S-GTPγS to the G proteins within the rat brain membranes during stimulation by nociceptin agonists. As a result we found that in rat brain a single high-affinity nociceptin binding site is coupled to low-potency stimulation of GTP binding.
Methods and Materials
Substances and buffer.
Nociceptin [nociceptin(1–17)], its fragments nociceptin(1–13) and nociceptin(1–9), and its analogd-Ala7-nociceptin(1–17) were synthesized in our institute (Berlin).3H-Tyr14-nociceptin (45 Ci/mmol),125I-Tyr14-nociceptin (2200 Ci/mmol) and 35S-GTPγS (1208 Ci/mmol) were obtained from NEN (Boston, MA). Gpp(NH)p, GTPγS, GDP, naltrindole, U-69593, DAMGO and naloxone were from Sigma (Delsenhofen, Germany). A total of 50 mM Tris/HCl, pH 7.4, containing 3 mM MgCl2 and 0.2 mM EGTA with additions as specified was used as buffer in all experiments.
Nociceptin binding to rat cerebral cortex membranes.
The cerebral cortex of male Wistar rats (about 250 g) was homogenized with an Ultra-Turrax in buffer, and the total membrane fraction was obtained by centrifugation. The receptor binding experiments were performed as described earlier for the mu opioid receptor (Albrecht et al., 1997). Briefly, about 250 μg of membrane protein in buffer containing 0.1% BSA and the protease inhibitors aprotinin (0.0015%) and bacitracin (0.15 mM) was incubated with increasing concentrations of3H-Tyr14-nociceptin (saturation experiment) or with the constant concentration of about 50 pM 3H-Tyr14-nociceptin and 0.001 to 10,000 nM unlabeled nociceptin (displacement experiment) in a total volume of 1 ml at 30°C for 90 min. In the displacement experiments, nonspecific binding in presence of 10 to 1000 nM unlabeled nociceptin was as low as <2%. For the parameters of nociceptin receptor binding to be compared with those of receptor-stimulated GTPγS binding, in some binding experiments 100 mM NaCl, 10 μM Gpp(NH)p or GTPγS and 20 μM GDP each alone or in combination were added to the incubations. These substances were included in the experiments on nociceptin-stimulated GTPγS binding. Furthermore, in one series of three experiments, displacement curves of the binding of3H-Tyr14-nociceptin using nociceptin, nociceptin(1–13), nociceptin(1–9) andd-Ala7-nociceptin(1–17) were obtained from incubations in presence of 100 mM NaCl/20 μM GDP at 25°C for 2 hr. The samples were filtered through Whatman GF/B filters using a Brandel-Harvester, and the filters were counted for3H-activity. The receptor binding parameters,i.e., the dissociation constantKd and the binding capacity Bmax, were estimated using the program RADLIG 4.0 (BIOSOFT, Cambridge, UK). The displacement by nociceptin of the binding of the iodinated tracer peptide125I-Tyr14-nociceptin was compared once with that of the 3H-labeled tracer. Using 3H-labeled nociceptin at concentrations of as low as 40 pM in the displacement studies and in the beginning part of the tracer saturation binding curve (fig. 1) resulted in a maximum specific binding of about 50% of the added tracer amount. These unusual conditions were necessary for obtaining the parameters of the high-affinity binding, and the program RADLIG 4.0 is able to estimate the binding parameters at sufficient accuracy even under such conditions.
Nociceptin binding in coronal sections of rat cerebral cortex.
Rats were killed by decapitation. Their brains and spinal cords were rapidly removed, frozen on dry ice and stored at −70°C until use. Coronal sections (10 μm) of both tissues were cut on a cryostat (CM 3000 Leica), thaw-mounted onto gelatin-coated slides and stored at -70°C until further processing. Sections of frontal cortex were preincubated in buffer at 25°C for 15 min. After washing two times for 30 sec with buffer, the sections were incubated with 50 pM125I-Tyr14-nociceptin and different concentrations of unlabeled nociceptin (0.001 up to 5 nM; displacement experiment) at 25°C for 2 hr in buffer containing 0.0015% aprotinin, 0.15 mM bacitracin and 0.1% BSA. After this time, the slides were washed with chilled 50 mM Tris/HCl and deionized water, dried and, together with autoradiographic125I-micro-scales (Amersham, Braunschweig, Germany), exposed to UR-imaging plates (FUJI Photo Film Co., Tokyo, Japan) for 16 hr. The plates were scanned and analyzed using the bio-imaging analysis system BAS-3000 (FUJI) linked to the micro computer imaging device system from Imaging Research Inc. (St. Catherines, Canada). When the binding of 0.01 to 100 nM3H-Tyr14-nociceptin was determined (saturation experiment), TR-plates (FUJI) were used for the exposition of the slides for at least 60 hr. The binding capacity was quantified by 3H-micro-scales (Amersham), and the binding parameters were determined as described above.
Nociceptin-stimulated 35S-GTPγS binding to rat cerebral cortex membranes.
A total of 10 to 20 μg of membrane protein prepared from rat cortex as described above was incubated with 50 pM 35S-GTPγS in presence or absence of nociceptin at 30°C for the indicated times in a total volume of 1 ml of buffer supplemented with 20 μM GDP and 100 mM NaCl. The reaction was terminated by filtration through Whatman GF/B filters using a Brandel-Harvester, and the filters were counted for35S-activity. Basal, nonspecific and nociceptin-stimulated binding of 35S-GTPγS were defined as binding of the tracer in absence of nociceptin, in presence of 10 μM unlabeled GTPγS, and as difference of binding in presence and absence of nociceptin, respectively. The affinity and capacity of the basal and nociceptin-stimulated binding of GTPγS were estimated from the displacement of the binding of35S-GTPγS by unlabeled GTPγS as described for receptor binding. EC50 values for the stimulation of 35S-GTPγS binding by nociceptin were calculated from concentration-response curves by a four-parameter logistic curve fitting program. Additionally, nociceptin concentration-response curves were compared with those of the shortened sequences nociceptin(1–13) and nociceptin(1–9) and the analogd-Ala7-nociceptin at 25°C.
Nociceptin- and DAMGO-stimulated35S-GTPγS binding in coronal sections of rat brain and spinal cord.
35S-GTPγS binding to slide-mounted sections obtained from rat brain and spinal cord as described above was studied essentially under conditions as recently described in literature (Sim et al., 1995, 1996a). The sections were incubated in buffer containing 100 mM NaCl for 10 min at 25°C followed by incubation in the same buffer containing additionally 2 mM GDP for 15 min. Then they were incubated in this medium for 2 hr at 25°C with 50 pM35S-GTPγS in presence and absence of 3 μM nociceptin or 3 μM DAMGO to obtain maximal stimulation of the binding of 35S-GTPγS compared to basal binding. After rinsing in cold buffer and deionized water, the slides were dried and, together with 14C-microscales (Amersham) for quantification, exposed to UR-imaging plates for 16 hr. The plates were scanned and analyzed using the BAS-micro computer imaging device combination. We confirmed that the concentration of GDP as high as 2 mM was necessary to suppress the basal binding of35S-GTPγS to an extent that allowed maximum stimulation of binding of the nucleotide by the ligands over basal binding (Sim et al., 1995), in contrast to the binding to membranes, which was studied at much lower concentrations of GDP.
Results
Receptor binding of nociceptin.
At 30°C the binding of 0.15 and 1.5 nM3H-Tyr14-nociceptin to membranes from the rat cortex reached equilibrium after 60 min, was stable until 150 min and increased linearly with the concentration of membrane protein (200–1300 μg in 1 ml incubate). In presence of 100 mM NaCl/20 μM GDP, i.e., the medium used for studies on binding of 35S-GTPγ S, binding equilibrium was also reached within 60 min.
Tracer (3H-Tyr14-nociceptin) saturation binding (fig. 1, upper panel) experiments revealed a high-affinity and high-capacity binding site with Kd of 21.6 ± 8.2 pM and Bmax 290.8 ± 70.5 fmol/mg protein (mean ± S.D., n = 4). In agreement with these results, displacement of bound 3H-peptide by unlabeled native nociceptin (fig. 1, lower panel) was fitted to one binding site with Kd of 30.08 ± 8.34 pM and Bmax of 230.8 ± 28.8 fmol/mg protein (mean ± S.D., n = 4). Gpp(NH)p (10 μM) lowered the binding of nociceptin with a shift in theKd value by 2.55 ± 0.43-fold (mean ± S.D., n = 3). The same affinity shift in binding was produced by 100 mM NaCl (fig. 1, lower panel) or 10 μM GTPγS, but the shift in presence of both 100 mM NaCl and 10 μM GTPγS was the same as in presence of each substance alone. GDP (20 μM) had no influence at all on the binding in absence or presence of NaCl, however, prevented the shift in affinity by GTPγS (data not shown). To summarize the influence of NaCl, GTPγS and GDP, in all cases any low-affinity shift was less than 3-fold.
Compared with nociceptin (fig. 5, upper panel), nociceptin(1–13) andd-Ala7-nociceptin had 7.9- and 36.7-fold lower affinities, respectively, but identical capacities (Bmax 201.4–223.4 fmol/mg). Nociceptin(1–9) was nearly devoid of binding activity (>10,000-fold lower affinity). No part of the total binding of nociceptin was displaced by the opioid antagonist naloxone (1 μM) or any of the selective ligands for themu, delta and kappa opioid receptor (1 μM DAMGO, 1 μM naltrindole and 10 μM U-69593, respectively).
Before using125I-Tyr14-nociceptin instead of the 3H-labeled ligand in the autoradiographic studies, the displacement curves by 0.01 to 1 nM unlabeled nociceptin of the binding of the two tracer peptides at each 50 pM to cortex membranes were found to be nearly exactly superimposed, verifying that both labeled ligands gave almost identical results in displacement binding studies. From three independent experiments at 25°C (fig. 2) the quantitative autoradiographic data were fitted to one specific binding site withKd of 116.7 ± 5.08 pM and Bmax of 1.661 ± 0.96 pmol/mg wet weight (mean ± S.D.). Furthermore, for control one autoradiographic saturation binding experiment with3H-Tyr14-nociceptin was performed, resulting in a Kd of 59.0 pM.
Nociceptin-stimulated binding of GTPγS.
In absence of any peptide ligand, the basal binding of 35S-GTPγS at 50 pM concentration to rat cortex membranes was increased linearly with protein concentrations of 2 to 20 μg per assay tube. This binding was nearly totally displaced by 10 μM unlabeled GTPγS (nonspecific binding <5% of basal binding). Nociceptin dose-dependently (1–3000 nM) stimulated the binding (fig.3) with an apparent EC50 value of 9.11 ± 1.00 (S.E.) nM. At 1 μM, it maximally stimulated the binding 2.73 ± 0.17-fold (mean ± S.D.) over basal binding independent of the incubation time of 5 to 240 min (fig. 4). During this long time, neither basal binding nor nociceptin-stimulated binding reached equilibrium. However, due to their proportional increase the stimulation factor obtained with the peptide remained constant with time (fig. 4).
Compared with nociceptin, nociceptin(1–13) andd-Ala7-nociceptin reached full activity in stimulating 35S-GTPγS binding, but their potencies were 4.5- and 30-fold lower (fig.5, lower panel), which corresponded well with their lower affinities in receptor binding (see above, fig. 5, upper panel). Nociceptin(1–9) was almost inactive, as expected from its extremely low receptor binding activity.
GTPγS binding to the membranes as studied by displacement of the bound amount of 35S-GTPγS by unlabeled GTPγS (fig. 6) resulted in an apparentKd value and apparent Bmax of 1.57 ± 0.62 (S.E.) nM and 3.03 ± 0.44 (S.E.) pmol/mg protein, respectively, for the binding maximally stimulated by nociceptin, i.e., the difference between binding in presence and absence of 1 μM nociceptin. Basal binding showed much lower affinity but much higher capacity, theKd and Bmax being 45.8 ± 4.37 (S.E.) nM and 46.56 ± 5.60 (S.E.) pmol/mg protein, respectively.
Using sections, nociceptin concentrations higher than 0.1 μM were necessary to obtain maximum stimulation of the binding of35S-GTPγS in the different areas of rat brain and spinal cord (fig. 7). From the regions studied, only in caudate putamen was no stimulation observed. The highest amount of nociceptin-stimulated binding was found in the amygdala whereas cortex showed the highest factor of stimulation of35S-GTPγS binding over basal binding. At the same time, the stimulation of binding by nociceptin in most regions was found to differ from that by the μ-opioid agonist DAMGO (fig. 7).
Discussion
In this study, rat cerebral cortex was used to investigate the binding of nociceptin to its receptor and the coupling of the receptor to G proteins through nociceptin-stimulated binding of GTPγS, because with brain sections it was found that, of all central nervous system regions studied, rat cortex showed the highest factor for the stimulation of 35S-GTPγS binding by the peptide over basal binding (fig. 7). Using membranes and sections of rat cortex, two labeled nociceptin peptides,3H-Tyr14-nociceptin and125I-Tyr14-nociceptin, and two types of binding studies (saturation, displacement), one single high-affinity and high-capacity binding site for nociceptin in rat brain was found (figs. 1 and 2), the Kd value ranging from 21.6 to 116.7 pM, dependent on the preparation and kind of method used. This high affinity essentially agreed with that found for the nociceptin receptor in brain membranes from guinea pig (Kd 16 pM, Shimohigashi et al., 1996), mouse (Kd 100 pM, Mathis et al., 1997), and rat (Kd 50–100 pM,Ardati et al., 1997; Makman et al., 1997) and in CHO cells (Kd 50–190 pM, Reinscheidet al., 1995, 1996; Ardati et al., 1997; Butouret al., 1997). However, it contrasted with the much lower affinities found for NG108–15 cells (Ma et al., 1997), rat brain membranes (Dooley and Houghten, 1996), CHO cells (Fukuda et al., 1997) and HEK cells (Shimohigashi et al., 1996) with Kd values between 1 and 5 nM. Therefore, displacement of the binding of 1.5 nM3H-Tyr14-nociceptin to rat cortex membranes was performed (0.1–1000 nM unlabeled nociceptin), resulting in an apparent displacement curve with an IC50 value of about 2 nM (data not shown). However, the detailed analysis of the curve showed that it represented exactly isotopic dilution of the tracer binding to a site already fully saturated at 1.5 nM labeled peptide. Therefore we did not find any evidence for a second binding site with a Kdvalue around 1 nM in rat cortex, in contrast to the findings in mouse brain (Mathis et al., 1997). Different temperatures as used here and in various studies (30°C/25°C) cannot be responsible for the discrepancies in the affinities observed because we found the Kd values at 25°C and 30°C to differ not more than by 10% (data not shown).
Before the identification of nociceptin as endogenous ligand for the ORL1 receptor, a lot of studies had already shown that none of the selective ligands for the mu, delta andkappa opioid receptor activated the ORL1 despite the exceptionally high structural homology among these receptors (reviewed in Reinscheid et al., 1995; Meunier, et al., 1995). In agreement with this fact, naloxone and selective opioid ligands did not compete with nociceptin for the binding sites in rat cortex at all. However, as the homology profile of the receptors would indicate, the ORL1 exhibits signaling mechanisms similar to the opioid receptors. Activation of the receptor was found to inhibit adenylate cyclase, indicative of its coupling to Gi-proteins (Reinscheid et al., 1995, 1996; Meunier et al., 1995; Mathis et al., 1997; Ma et al., 1997; Butour et al., 1997). Further evidence is given by our findings that the stable GTP analogs Gpp(NH)p and GTPγS as well as NaCl decreased the nociceptin receptor affinity in rat cortex (fig. 1), typical of opioid receptors. However, the shift was only moderate (2- to 3-fold) and not additive. This contrasts with results on ORL1 expressed in CHO cells (Butouret al., 1997), where NaCl increased the capacity of a preexisting very low-affinity nociceptin binding site, and where the shift was further increased by Gpp(NH)p. Furthermore, Gpp(NH)p drastically shifted the proportion of high to low affinity sites in HEK cells, but at the same time only moderately the affinity of a single site in CHO cells (Ardati et al., 1997). These differences show that the existence of a low-affinity binding site and the exact coupling of the receptor to the G proteins may depend on the type of cell expressing the receptor and on its expression level (Ardatiet al., 1997), and that conclusions from the situation in cells expressing the receptor to that in native tissues have to be made cautiously.
The coupling of the nociceptin receptor to G proteins was directly observed by the stimulation of the binding of35S-GTPγS in the membranes of cortex (fig. 3) as well as in sections from brain and spinal cord (fig. 7). Both basal and nociceptin-stimulated binding of the labeled nucleotide to the membranes proceeded at the same very slow rate, not reaching equilibrium even after incubation for 4 hr at 30°C (fig. 4). Obviously, the kinetics of both kinds of binding as seen in figure 4 is mostly dictated by the dissociation of GDP, required in the incubations to inactivate the G proteins and thus providing low basal levels of binding, from the GDP/GTP binding site of the G protein in the GDP-GTP exchange reaction (Wieland and Jakobs, 1994). Such an explanation should only be true if the basal binding of GTPγS observed was binding to a specific site and not just nonspecific. With this in line is the fact that not only the stimulated but also the basal binding sites for GTPγS were found to be specific withKd values of 1.57 and 45.8 nM, respectively (fig. 6), but with the capacity of the basal binding sites being 15-fold higher than that of the nociceptin-stimulated binding sites.
The specific nature of the basal binding of GTPγS might be interpreted as basal coupling of unoccupied receptors to G proteins. Then, the manifold higher capacity of basal over nociceptin-stimulated GTP binding sites found here would reflect constitutive activities of a lot of receptors. However, this seems to be unlikely due to the differences in affinity between basal and receptor-stimulated GTP binding, as seen here for the nociceptin receptor, characterizing basal and stimulated GTP binding sites as different states. More likely, basal binding of GTPγS may be assumed to be binding to the nucleotide site in the Gα subunit within the heterotrimeric G protein in equilibrium with GDP, as opposed to binding to Gα dissociated from the Gβγ dimer after stimulation by receptor occupancy. Conclusively, in a native cell the nucleotide binding site in the heterotrimeric G protein may be occupied not only by GDP but also partly by GTP.
The GTPγS binding isotherms could be evaluated according to the law of mass action, and from the parameters calculated it is concluded that nociceptin stimulated the binding of GTPγS to G proteins by decreasing the Kd fromKdlow aff. 45.8 nM toKdhigh aff. 1.57 nM, i.e.,by increasing the affinity of the GTP binding site by 29-fold. This shift is higher than the about 3-fold increase in the binding affinity of GTPγS by μ-opioid receptor activation in rat striatal membranes (Sim et al., 1996b). According to the law of mass action, if a binding site changes from low to high affinity, the amount of bound ligand in equilibrium with the free ligand concentration λ is increased by the factor q = (λ+Kdlow aff.))/(λ+Kdhigh aff.). With the free 35S-GTPγS concentration being about 50 pM in the stimulation experiments, for the nociceptin receptor-coupled G proteins the amount of35S-GTPγS binding stimulated by nociceptin over basal binding is calculated to be 28-fold. However, the maximum stimulation in the bound amount observed, i.e., the quotient of 35S-GTPγS binding in presence and absence of 1 μM nociceptin, was only 2.7 (fig. 4). Taking the 15-fold excess of the total basal (Bmax 46.56 pmol/mg) over stimulated binding sites (Bmax 3.03 pmol/mg) into account, it was calculated that a measurable factor of only 2–3 could be obtained as was the case (fig. 4). Generally, a sufficient part of G proteins involved in activation, a high increase in their affinity to GTP, and a low basal binding of GTP to the total of all G proteins will enable the stimulation of binding of a GTP analog to G proteins to be observed in any receptor activation in the tissue or cell under study.
In contrast to the parameters of the binding of nociceptin to its receptor, those of GTPγS were not obtained at equilibrium (fig. 4) and are, therefore, only apparent parameters. TheKd values for the basal and nociceptin-stimulated binding of GTPγS and especially their relation, however, should be rather robust due to the factor of stimulated over basal binding being unchanged with time (fig. 4). The binding capacities estimated, however, are not correct in absolute terms and are only meaningful when compared under identical conditions. From the capacity of the nociceptin receptor (290 fmol/mg) and of the nociceptin-stimulated GTPγS binding sites after incubation for 4 hr (3 pmol/mg) it is concluded that one receptor site is able to stimulate at least 10 G protein sites for GTP.
Ligand binding by a G protein-coupled receptor is immediately coupled to the stimulation of the G proteins by stimulating the binding of GTP. Therefore, the main problem in the interpretation of the data on GTP binding lies in the low potency of nociceptin in stimulating the GTPγS binding in cortex membranes (fig. 3) as compared with its high affinity to its receptor (fig. 1). As with membranes, sections from the cortex exhibited high-affinity receptors (fig. 2), but only at 0.1 to 1 μM concentrations of nociceptin was saturation of binding of35S-GTPγS reached. Nociceptin stimulated the nucleotide binding in membranes with an EC50 of 9.11 nM, which compares well with 19.8 nM as found in another study (Sim et al., 1996a), but which was much higher than theKd of 0.03 nM for receptor binding found here. Generally, when G proteins are activated by agonist-stimulated binding of GTP, the affinity of the receptors is lowered. This was also observed in this study for the nociceptin receptor but the decrease of the Kd by <3-fold by the GTP analogs Gpp(NH)p (fig. 1) or GTPγS was rather moderate. The same was true for the influence of NaCl (fig. 1), which was included in the GTP binding assays. Furthermore, 20 μM GDP, also included in the assay for lowering basal binding of35S-GTPγS, had no influence on the nociceptin receptor affinity in absence or presence of NaCl. Remarkably, GDP prevented the receptor affinity shift by 10 μM GTPγS, although under this condition the nucleotide binding site was fully occupied by GTPγS (Kd 1.57 nM). Therefore, not competition by GDP but an allosteric influence of it on the GTPγS-induced receptor affinity shift may be responsible for this observation.
In conclusion, under identical experimental conditions the affinity of nociceptin for its receptor was about 100-fold higher than its potency in stimulating the GTP binding. This would mean that about 90% of the concentration-response curve was accomplished between 90 to 100% occupancy of the receptor (fig. 3) or that the stimulation was mediated through a low-affinity site or state not detectable in the binding studies. When the specificity of the coupling between receptor binding and 35S-GTPγS binding was studied, the order of magnitude of the receptor affinities of nociceptin, nociceptin(1–13),d-Ala7-nociceptin, and nociceptin(1–9) (Kd 1:7.9:36.7:>10,000) was nearly the same (fig. 5) as that of their potencies in stimulating the 35S-GTPγS binding (EC50 1:4.5:30:not measurable). Such was also the specificity seen in the inhibition of cAMP accumulation in CHO cells stably expressing the ORL1-receptor (Reinscheid et al., 1996). Remarkably, whereasd-Ala7-nociceptin was a partial agonist in inhibiting cAMP accumulation (Reinscheid et al., 1996) it showed in this study full maximum activity in stimulating the GTPγS binding. In agreement with an earlier report (Sim et al., 1996a), no stimulation of nucleotide binding was identified in the caudate putamen, an area in which themu opioid agonist DAMGO highly stimulated the binding (fig.7). This parallels the very low levels of ORL1 immunoreactivity in this region, in contrast to other regions (Anton et al., 1996) that also expressed stimulation of GTP binding (fig. 7). Taken together, the specificity of the nociceptin-stimulated binding of GTPγS corresponded to the properties of the high-affinity nociceptin receptor identified in rat brain.
In summary, our data show for the first time that a high-affinity nociceptin receptor in rat brain is coupled to a comparably very low-potency stimulation of GTPγS binding. The exact mechanism of this coupling remains open. Because of the specificity of the stimulation of GTPγS binding corresponding to that of the high-affinity receptors, it appears to be unlikely that an independent second low-affinity receptor site is responsible for the coupling. Rather, it seems reasonable to suggest the involvement of a low-affinity state of the receptor. Several lines of evidence have indicated the existence of multiple forms of a receptor complex (Gudermann et al., 1996), including opioid receptors (Werling et al., 1988;Wong et al., 1994), at any one time. Therefore, one explanation for our results on the nociceptin receptor would be that a small part of agonist-occupied receptors is switched to a low-affinity state, not detectable in binding studies in the presence of an excess of sites in high-affinity state. Because of the high capacity of the nociceptin receptors and of nociceptin-stimulated GTP binding sites, even a small part of the receptors in low-affinity state in equilibrium with the high-affinity state and stabilized at high agonist concentrations could provide enough receptor sites that catalytically stimulate GTP binding.
Acknowledgments
The authors thank Dr. M. Beyermann and A. Klose (Berlin) for the synthesis of the nociceptin peptides and M. Georgi for excellent technical assistance.
Footnotes
-
Send reprint requests to: Dr. Hartmut Berger, Research Institute of Molecular Pharmacology, Alfred-Kowalke-Str. 4, D-10315 Berlin, F.R.G.
-
↵1 This work was supported in part by a grant to N.N.S. from the Deutscher Akademischer Austauschdienst e.V. (Bonn, F.R.G.).
- Abbreviations:
- BSA
- bovine serum albumin
- CHO
- Chinese hamster ovary
- EGTA
- ethylene glycol bis(2-aminoethylether)-N, N,N′,N′-tetraacetic acid
- DAMGO
- [d-Ala2, N-Me-Phe4, Gly5-ol)]-enkephalin
- Gpp(NH)p
- 5′-guanylylimidodiphosphate
- GDP
- guanosine-5′-diphosphate
- GTPγS
- guanylyl-5′-O-(γ-thio)-triphosphate
- HEK
- human embryonic kidney
- ORL1 receptor
- opioid receptor-like receptor 1
- Received November 20, 1997.
- Accepted April 6, 1998.
- The American Society for Pharmacology and Experimental Therapeutics