| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Nociceptin/orphanin FQ (N/OFQ) is the endogenous ligand for the N/OFQ receptor (NOP), whose characteristics in the dog are unknown. We therefore compared [3H]N/OFQ binding in dog and rat brain membranes. Radioligand saturation/competition studies with these membranes and leucyl-[3H]N/OFQ(117)OH or the novel radioligand [3H]N/OFQ(113)NH2 were performed to determine receptor density and ligand affinity. The density of classic opioid receptors was determined by using [3H]diprenorphine. Leucyl-[3H]N/OFQ(117)OH binding was concentration dependent and saturable in dog (maximum binding capacity [Bmax], 28.7 ± 2.8 fmol/mg of protein; equilibrium dissociation constant as negative log [pKd], 10.27 ± 0.11) and rat (Bmax, 137.0 ± 12.9 fmol/mg of protein; pKd, 10.41 ± 0.05). In comparison, the Bmax and pKd of [3H]diprenorphine were, respectively, 77.7 ± 5.3 fmol/mg of protein and 9.74 ± 0.09 in dog and 79.1 ± 18.2 fmol/mg of protein and 9.51 ± 0.04 in rat. In dog, [3H]N/OFQ(113)NH2 binding to NOP receptors was also saturable (Bmax, 23.7 ± 2.0 fmol/mg of protein; pKd, 10.16 ± 0.12). In both species, leucyl-[3H]N/OFQ(117)OH was displaced by various NOP ligands. Dynorphin A, N/OFQ(15)NH2, and nocistatin were essentially inactive. There was a significant positive correlation (r2 = 0.95; P < 0.0001) between pKi values (an estimate of affinity) obtained in displacement studies in rat and dog. We have demonstrated a low density of NOP receptors, measured with two radioligands, in dog, and these receptors display a high degree of pharmacological similarity with those natively expressed in the rat. IMPLICATIONS: Experimentally, the dog represents a species used in various anesthetic studies, yet little is known regarding the expression of nociceptin receptors (with opioid-like activity) in this model. In comparison with rat, dog membranes express a small density of pharmacologically identical binding sites whose functional activity remains to be determined.
The nociceptin/orphanin FQ (N/OFQ) receptor is a member of the guanine nucleotide-binding regulatory protein (G protein)-coupled opioid family (1,2). In line with recent International Union of Pharmacology (IUPHAR) recommendations, this receptor will be referred as "NOP" (previously ORL-1 or OP4) (3). Activation of NOP receptors by N/OFQ results in an inhibition of adenylyl cyclase activity, thereby decreasing cellular concentrations of cyclic adenosine monophosphate, stimulating an outward K+ current, and inhibiting voltage-gated Ca2+ channels. Depending on cellular NOP localization, presynaptic inhibition of neurotransmitter release or postsynaptic inhibition of neuronal excitability will result (4). The distribution of NOP is widespread throughout the central nervous system and in peripheral tissues such as the spleen, kidney, vas deferens, intestine, and retina. In general, receptor expression correlates with the localization of the prepronociceptin gene, the gene for the precursor of nociceptin, and other bioactive peptides, such as nocistatin (5). After systemic or central administration of N/OFQ to mice, rats, and guinea pigs, a range of physiological changes can be observedfor example, modulation of pain behavior, including antinociception or pronociception, depending on the site of administration; anxiolysis; modulation of feeding; and cardiovascular depression (4,6,7). Because of the homology with the classic opioid system (8), this novel receptor-transmitter system is of major anesthetic relevance, particularly with reference to pain, anxiolytic activity, and cardiovascular control. Many studies have examined the pharmacology and distribution of NOP receptors in rodents, the group in which there is the most evidence that N/OFQ acts as an important neuromodulator and in which the NOP receptor was initially identified. However, the distribution of this receptor and peptide in the dog has yet to be determined. Dogs are an important model for studying in vivo cardiovascular variables after drug administration (9), and it is likely that N/OFQ modulates cardiovascular activity in dogs. Indeed, dogs are an important model with which to study anesthetic action (10,11), because there are many similarities to humans. To examine N/OFQ in this species and as an intermediate between small laboratory mammals and humans, a basic understanding of NOP receptor pharmacology is urgently required. This information will enable more detailed autoradiographic and functional analyses to be performed.
Therefore, in this study we have performed a series of radioligand-binding studies with two [3H]N/OFQ radioligands, full-sequence [3H]N/OFQ and truncated [3H]N/OFQ(113)NH2, to 1) determine NOP receptor density in the dog brain, 2) examine the pharmacology of a range of commonly used NOP and opioid receptor ligands, and 3) compare these data with the NOP receptor in the rat whose identity has been confirmed in both cloning and pharmacological studies. In addition, we have compared NOP receptor densities with those of the classical µ-opioid (MOP),
Permission for this study was received from the relevant local animal research committee. N/OFQ, N/OFQ(15)NH2 (N/OFQ5), N/OFQ(19)NH2 (N/OFQ9), N/OFQ(113)NH2 (N/OFQ13), [Phe1 (CH2-NH)Gly2]N/OFQ(113)NH2 ([F/G]), [Nphe1]N/OFQ(113)NH2 ([Nphe1]), nocistatin, AcRYYRWKNH2 (CTD), and J-113397 were synthesized at one of our institutions. Ro64-6198 [(1S,3a]S)-8-(2,3,3a,4,5,6-hexa-hydro-1H-phenalen-1-yl)-1-phenyl-1,3,8-triaza-spiro [4.5]decan-4-one was from F. Jenck and J. Wichmann of F. Hoffmann-La Roche AG, Switzerland; III-BTD was purchased from Neosystem, France; and naloxone benzoylhydrazone (NalBzOH) and dynorphin A were from Sigma, UK. The radioligand leucyl-[3H]N/OFQ(117)OH (149 Ci/mmol specific activity) was purchased from Amersham Life Sciences, UK, and [3H]diprenorphine ([3H]DPN; 56 Ci/mmol specific activity) was purchased from NEN.
The novel NOP radioligand [3H]N/OFQ(113)NH2 was prepared by catalytic dehalogenation of a precursor peptide, p-iodo-Phe1-N/OFQ(113)NH2, with SPPS methodology by using tritium gas and PdO/BaSO4. Briefly, Brains from male and female Beagle dogs (912 mo; 914 kg) killed by pentobarbital overdose and female Wistar rats (200250 g) killed by stunning and cervical dislocation were used. Four dog brains were used for four separate preparations. Briefly, the brains were rapidly dissected, the cerebellum and brainstem were removed, and the remaining brain (i.e., cortical and midbrain tissue) was processed further by homogenization and centrifugation (13,500 rpm for 10 min at 4°C) in assay buffer (50 mM Tris-HCl [Sigma] and 5 mM MgSO4 [Fisher Scientific UK, Ltd.], pH7.4). This process was repeated three times, and the pellet was finally resuspended in assay buffer. Rat cerebrocortical homogenates were prepared in the same manner as the dog, described previously. The total protein concentration of the membrane preparations was determined (13), and 1-mL stocks were frozen at -70°C until further use. The assay protocol was essentially as described previously (14,15). All assays were performed in a total volume of 0.5 mL of assay buffer supplemented with 0.5% bovine serum albumin (Sigma) and 10 µM peptidase inhibitors (amastatin, bestatin, captopril, and phosphoramidon; Sigma) for 60 min at room temperature (see below). Membranes (100 µg for rat or 200 µg for dog) were incubated with various concentrations of either leucyl-[3H]N/OFQ(117)OH or [3H]N/OFQ(113)NH2 for various times. Nonspecific binding (NSB) was determined in all experiments by the addition of either 1 µM unlabeled N/OFQ (for leucyl-[3H]N/OFQ(117)OH) or N/OFQ13 (for [3H]N/OFQ(113)NH2).
To determine opioid receptor density for comparison with NOP expression, a series of saturation studies were performed with 200 µg of rat or dog membranes in 50 mM Tris buffer, pH 7.4, with In all binding protocols, bound and free radioligands were separated by rapid vacuum filtration by using a Brandel 24 place cell harvester onto Whatman GF/B filters. Filters were soaked for 1 h in 0.5% polyethylenimine (Sigma) to reduce NSB and loaded onto the harvester wet. Bound radioactivity was extracted for at least 8 h in 4.5 mL of Wallac Optiphase "safe" scintillation fluid (Fisher Scientific UK, Ltd.) before filter-bound radioactivity was assessed by standard scintillation spectroscopy. The pharmacological terminology adopted in this study is in line with IUPHAR recommendations (17). Data are expressed as mean ± SEM. Saturation binding isotherms were analyzed by using a log transformation to obtain the maximum binding capacity (Bmax) and the equilibrium dissociation constant (as negative log; pKd) by computer-assisted curve fitting of individual curves with Prism Version 3.0 (GraphPad, San Diego, CA). In competition studies, the concentration of displacer producing 50% inhibition of radiolabel binding was corrected for the competing mass of radiolabel according to Cheng and Prusoff (18) to yield Ki an estimate of affinity, where the Kd used was determined in this study. Where appropriate, data were analyzed with analysis of variance and Students t-test (two tailed) or linear regression, as appropriate, with P < 0.05 considered as significant.
The binding of leucyl-[3H]N/OFQ(117)OH was time dependent such that at 0.2 nM, an apparent equilibrium was reached at approximately 15 min in both dog and rat, and this was stable for at least 1 h (Fig. 1A). When these data were fitted to a one-phase exponential association, a mean observed rate of association of 0.133 and 0.196/min was estimated for dog and rat, respectively. To ensure that equilibrium was achieved at smaller ligand concentrations, all subsequent binding studies were performed with a 60-min incubation.
The binding of leucyl-[3H]N/OFQ(117)OH was concentration dependent and saturable in both dog (Fig. 1B) and rat (Fig. 1C) membrane preparations. Bmax and pKd were calculated for four different dog preparations (Table 1), and there was some variation in the Bmax obtained (n = 35 for each preparation). There were significantly fewer ( 5-fold) N/OFQ binding sites (Bmax) in the dog (Fig. 1D) compared with the rat (P = 0.0002). pKd values were consistent. Differences in the Bmax between species could not be attributed to loss of receptor over time, because membranes stored at -70°C for approximately 5 mo resulted in an approximately 6% decrease in receptor density. This failed to reach statistical significance as compared with freshly prepared samples. There was no change in pKd (Table 1).
The binding of [3H]N/OFQ(113)NH2 in dog and rat membranes was also concentration dependent and saturable (Fig. 2); Bmax and pKd values are shown in Table 1. In agreement with studies using the full-sequence radioligand, there was a significantly smaller density of NOP binding sites in the dog compared with the rat (P < 0.0001). However, there was a small ( 3-fold) but significant (P < 0.005) difference in radioligand pKd, with a higher affinity obtained in the rat. The number of N/OFQ binding sites in the dog revealed with this radiolabel was smaller ( 23%) than with the full-sequence radioligand in paired membrane preparations.
To assess total (MOP, DOP, and KOP) opioid density in the dog and hence allow further comparison of our data with published information, we performed a series of [3H]DPN saturation binding studies. The binding of this radioligand was concentration dependent and saturable; Bmax and pKd are shown in Table 1. There were approximately threefold more classic opioid than NOP receptors in dog brain preparations. In contrast, there were half as many classical opioid as NOP receptors in the rat. The binding of leucyl-[3H]N/OFQ(117)OH was displaced in a concentration-dependent manner in both dog and rat membranes by a range of NOP and classic opioid ligands (Fig. 3) encompassing both peptide and nonpeptide structures. Calculated pKi values for these data are presented in Table 2. NalBzOH displaced with relatively low affinity, and dynorphin A, N/OFQ5, and nocistatin were inactive over the range of concentrations used. A small but significant (P = 0.03) difference for [F/G] was found between the two species. There was a strong positive correlation (r2 = 0.95; P < 0.0001) between pKi values in both species (Fig. 4). This correlation could be improved to 0.98 if [F/G] was excluded from the comparison.
In this study we have clearly demonstrated that the dog brain possesses a relatively low density of N/OFQ binding sites compared with rat and within the opioid family when compared with classic MOP, DOP, and KOP opioid receptors. These binding sites probably correspond to the nociceptin receptor identified by Reinscheid et al. (2) and Meunier et al. (1). The levels of receptor expression in dog and rat membrane preparations assessed with leucyl-[3H]N/OFQ(117)OH differ markedly, with Bmax values of 29 and 137 fmol/mg of protein, respectively. The binding affinity of this ligand was species independent. There have been several studies examining the binding of both [3H]N/OFQ(117)OH and [125I]Y14N/OFQ(117)OH in a range of species which generally demonstrated a significantly higher density of NOP receptors in these species compared with the dog. However, in a study using [3H]CTD, Thomsen et al. (19) reported a very low density (22 fmol/mg of protein) of high-affinity NOP receptors in the rat. Reliability of the data from these previous studies is uncertain because each used different assay conditions to determine NOP density, and the integrity of, particularly, [3H] radioligands is questionable. The relatively low density of NOP receptors in the dog has a number of experimental and functional consequences. Indeed, in this study, a glance at the numerical data presented in competition binding assays shows greater variability when compared with the rat. As for functional consequences, these may become important when examining the pharmacology of partial agonists in functional studies (see below). We have also examined NOP receptor density in dog and rat by using a novel [3H]N/OFQ radioligand, [3H]N/OFQ(113)NH2 (20). Truncation of N/OFQ from 17 to 13 amino acids results in an agonist with essentially the same activity (in binding and functional assays) as the full-sequence peptide (15,21). In addition, amidation may protect the molecule from enzymic degradation, although in this study we included peptidase inhibitors in all our assays. Increased functional potency of the amidated peptide has been shown in isolated vas deferens (21). Data obtained in the dog with this radioligand essentially confirm the relatively low density of NOP receptors in this species. There was no difference in binding affinity when comparing dog by using the [3H] full sequence, and, interestingly, this was also the case for the rat. In displacement studies with leucyl-[3H]N/OFQ(117)OH, there was a statistically significant correlation between pKi values obtained in both species, indicating a high degree of identity of N/OFQ binding sites. In experiments using leucyl-[3H]N/OFQ(117)OH as the primary labeling ligand, a range of molecules with activity at the NOP receptor produced a concentration-dependent displacement. The rank order of pKi obtained in both species was essentially identical and agrees with that reported by us (15) and others (7,21) for the rat and human. However, one discrepancy with our previous data is the higher affinity of [F/G], although, notably, previous data were obtained by using [125I]Y14N/OFQOH as the radioligand. Affinity values for the antagonists[Nphe1] (22), J-113397 (23), and III-BTD (24)and agonistRo 64-6198 (25)are essentially identical to values obtained with recombinant human NOP receptors and [3H]N/OFQ(117)OH as the radioligand (26). Ligands with variable activity at NOP receptors and classic opioid receptors displayed weak affinity at NOP receptors in both dog and rat. Indeed, NalBzOH, which is a mixed opioid agonist/antagonist and a partial agonist at the NOP receptor (22,27), dis- placed with relatively low affinity (similar to that of N/OFQ9), and the KOP agonist dynorphin A was inactive despite some homology between this and the N/OFQ. Similarly, nocistatin was inactive. One ligand that requires further comment is the peptide partial agonist [F/G], for which there is a discrepancy in the affinity values for both species, with the rat NOP receptor displaying a sixfold higher affinity. It is tempting to suggest that there may be differences in the structures of the receptors in both species, but on the basis of these limited data, such a conclusion cannot be supported. We have no explanation for these differences, but they remain the same for different batches of the synthetic peptides and membrane preparations. In conclusion, we have compared the binding of NOP ligands in rat and dog brain. Despite the almost identical pharmacology between dog and rat NOP receptors, dog brain contains approximately one-fifth the density of receptors as found in rat. Further detailed autoradiographic mapping (areas of the brain involved in cardiovascular and pain control) and whole-animal functional studies in this species are clearly warranted.
Supported by a Pfizer-sponsored studentship (EEJ). We would like to thank C. De Risi of the Department of Pharmacological Sciences, University of Ferrara, Italy, for providing J-113397; Dr. F. Jenck and Dr. J. Wichmann of F. Hoffmann-La Roche AG, Switzerland, for Ro64-6198; and Dr. G. Calo of the Department of Experimental and Clinical Medicine, University of Ferrara, Italy, for helpful discussions.
All experiments complied with the current ethical standards for the United Kingdom, and permission was received from the relevant local animal committee.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|