Sensitivity of Opioid Receptor-like Receptor ORL1 for Chemical Modification on Nociceptin, a Naturally Occurring Nociceptive Peptide*

Nociceptin or orphanin FQ is a novel neuropeptide that activates an opioid-like G protein-coupled receptor ORL1. This heptadecapeptide FGGFTGARKSARKLANQ resembles (cid:107) -opioid peptide dynorphin A but exhibits an opposite effect to make animals hyperreactive to nociceptive stimulations (Meunier, In the present study, it was found that guinea pig brain contains receptors to which nociceptin binds much more strongly than to ORL1 receptors expressed in out at least in triplicate. The results were analyzed by the computer program ALLFIT (15). In Vivo Nociceptive Activity— The hyperalgesic effect of peptides was examined in the tail-flick test. Responses were recorded on an analge-simeter (Muromachi Kikai, Tokyo) that focuses a noxious beam of light on the tail and automatically records the latency to removal. The peptides in a volume of 10 (cid:109) l were administered intracerebroventricu- larly (i.c.v.) to ICR mouse (25–35 g). The training sessions were per-formed until mice could flick their tails within 10 s. The latency time was recorded twice (20 and min) after i.c.v. administration. Statis- tical significance was determined by one-way analysis of variance test.

The recent finding of a naturally occurring nociceptive peptide named nociceptin (1) or orphanin FQ (2) has highlighted a possibility to design a novel analgesic. Nociceptin was isolated as a ligand for ORL1 receptor whose primary structure is most closely related to those of opioid receptors (3)(4)(5). It produced hyperalgesia in rat and mouse in both hot-plate and tail-flick assays and was found to inhibit cAMP accumulation in the Chinese hamster ovary cell-expressing ORL1 receptor.
Nociceptin resembles dynorphin A, a -receptor selective opioid peptide (6, 7) (see Fig. 1). Both nociceptin and dynorphin A are 17-amino acid peptides and contain 4 -5 basic amino acids at the C-terminal portions. It is a question of great interest whether structural similarities between nociceptin and dynorphin A not only in peptide sequences but also in their receptor structures are reflected in the structure-activity relationships of nociceptin. The N-terminal sequences of nociceptin and dynorphin A are very similar, having the same sequence of Gly-Gly-Phe at positions 2-4. Gly 3 and Phe 4 are essential for activities of opioid peptides. As to Gly at position 2, structural requirements appear to be diverse among opioid peptides and receptors. Chavkin and Goldstein (8) reported that Gly 2 having no side chain is important to the potency of dynorphins for receptors. This is in contrast to the fact that small side chain substituents for Gly 2 of enkephalin analogs lead to a favorable steric interaction at ␦ and receptors (9). The major difference between nociceptin and opioid peptides exists, however, in the N-terminal amino acid: Phe for nociceptin and Tyr for dynorphin A and other opioid peptides. It has been well known that the replacement of Tyr in opioid peptides by Phe, which means the removal of the para-hydroxyl group from ␤-phenyl, results in inactivity (10).
The importance of basic amino acid residues (Lys 13 , Lys 11 , and Arg 7 ) to the potency of dynorphin A was demonstrated by successive removal of C-terminal amino acids (8). In particular, Arg 7 was found to be an essential structural element of dynorphins for specific recognition of opioid receptors. When this Arg 7 was retained, the removal of the C-terminal octapeptide containing two of the other four basic amino acids sustained a full activity of dynorphin A (11)(12)(13). Dynorphin A has a basic amino acid pair of Arg-Arg at position 6 -7. If dynorphin A was proteolytically excised here, the resulting peptide fragment is no longer dynorphin, producing a leucine-enkephalin molecule. On the other hand, nociceptin contains a couple of such basic amino acid pairs, namely two Arg-Lys sequences at positions 8 -9 and 12-13 (see Fig. 1). It is interesting to know whether or not these pairs split nociceptin into inactive peptides.
In the present study, in order to elucidate fundamental structural essentials of nociceptin, we have synthesized Tyr 1nociceptin, nociceptin-(1-7), and nociceptin-(1-11) and evaluated their in vivo nociceptive activity and receptor binding affinity for ORL1 and opioid receptors.

MATERIALS AND METHODS
Peptide Syntheses-All peptides, nociceptin, Tyr 1 -nociceptin, Tyr 14nociceptin, nociceptin-(1-7), and nociceptin-(1-11), were synthesized by the automated peptide synthesizer ABI 470A with the Fmoc (N-(9fluorenyl)methoxycarbonyl) synthetic strategy. Peptides were first purified by the Sephadex G-15 (Pharmacia, Uppsala, Sweden) column (2.0 ϫ 138 cm) eluted with 30% AcOH after removal of resin and deprotection by reagent K. Further purification was carried out by * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. reversed-phase HPLC 1 using a prepacked column (2.5 ϫ 25 cm, Cica-Merck LiChrospher RP-18(e), 5 m) with a linear gradient of 0.1% trifluoroacetic acid and 80% acetonitrile. The fractions containing pure peptide were pooled and lyophilized. The purity was verified by analytical reversed-phase HPLC (0.4 ϫ 25 cm, LiChrospher 100 RP-18(e), 5 m) and amino acid analysis.
Expression Plasmid-The cDNA clone of human ORL1 receptor was obtained by reverse transcription and polymerase chain reaction. The first strand cDNA was synthesized with 5 g of human total brain RNA (Clontech, Palo Alto, CA) using the First strand synthesis kit (Pharmacia) according to the manufacturer's instruction. The oligonucleotide primers corresponding to the 5Ј-end (5Ј-CGGAATTCATGGAGC-CCCTCTTCCCCG-CGCCGTTCTGG-3Ј) and to the opposite strand of the 3Ј-end of the coding region (5Ј-GCTCTAGATCATGCGGGCCGCG-GTACCGTCTCAGAGGT-3Ј) were synthesized based on the published sequence (3). Polymerase chain reaction was carried out with the first strand cDNA reaction mixture and these primers for 25 cycles. The amplified DNA fragment was isolated by electrophoresis on a 1.2% agarose gel, digested with EcoRI and XbaI, and cloned into the EcoRI/ XbaI site of pcDNA3 (Invitrogen, San Diego, CA). Several cDNA clones were sequenced, and a cDNA clone without unwanted mutations was selected.
Cell Culture and Transfection-Human embryonic kidney 293 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. To generate stable transfectants, plasmid DNA (5 g) was transfected into 293 cells using the Lipofectamine method (Life Technologies, Inc.) in a 3-cm culture dish. After 24 h, cells were seeded into 10-cm dishes, and geneticin (G418) (600 g/ml, Life Technologies, Inc.) was added to select transfectant clones. Two weeks later, G418-resistant cell clones were isolated, propagated, and screened by ligand binding. The stable cell clone expressing receptors most highly was cultured up to the 25th passages, and the intact cells were used for binding assays.
Receptor Binding Assays-Radioligand receptor binding assays were carried out essentially as described previously (14).  14 ]nociceptin, human 293 cells expressing ORL1 receptor or guinea pig brain were utilized, whereas rat and guinea pig brains were utilized for assays with opioid ligands. The membranes of these brains were prepared as previously reported (14). Incubations were carried out at 25°C for 30 min in Tris-HCl buffer (pH 7.4) containing 0.1% bovine serum albumin. Bacitracin (100 g/ml) was added as an enzyme inhibitor. Dose-response curves were constructed utilizing 7-10 doses, and the assays were carried out at least in triplicate. The results were analyzed by the computer program ALLFIT (15).
In Vivo Nociceptive Activity-The hyperalgesic effect of peptides was examined in the tail-flick test. Responses were recorded on an analgesimeter (Muromachi Kikai, Tokyo) that focuses a noxious beam of light on the tail and automatically records the latency to removal. The peptides in a volume of 10 l were administered intracerebroventricularly (i.c.v.) to ICR mouse (25-35 g). The training sessions were performed until mice could flick their tails within 10 s. The latency time was recorded twice (20 and 60 min) after i.c.v. administration. Statistical significance was determined by one-way analysis of variance test.

RESULTS AND DISCUSSION
Receptor Binding Activity of Nociceptin and Its Analogs to ORL1-expressing Human 293 Cells-Tyr 14 -substituted nociceptin and its monoiodinated derivative were reported to be FIG. 1. Amino acid sequences of nociceptin, dynorphin A, and  their analogs.   FIG. 2. Hyperalgesic effect of nociceptin (A) and Tyr 1 -nociceptin (B) in the mouse tail-flick test. After i.c.v. administration, the latency time was recorded twice after 20 min (solid column) and 60 min (open column). All data are means Ϯ S.E. determined from the assays using 7ϳ8 animals in triplicate. Statistical significance was determined by an analysis of variance followed by the Bonferroni multiple comparison test; asterisks denote significant differences between the results for mice treated with peptides (three asterisks, p Ͻ 0.005; four asterisks, p Ͻ 0.001).  (Table I). Interestingly, Tyr 1 -nociceptin showed considerably high affinity for ORL1 receptors (IC 50 ϭ 3.0 nM) and was only slightly less potent than nociceptin. This indicates that the Phe 1 3 Tyr substitution, namely the replacement of the parahydrogen atom by the hydroxyl group on the Phe 1 -phenyl group, does not affect much the receptor binding characteristics of nociceptin. This is quite different from the fact that the Tyr 1 3 Phe replacement eliminates completely the ability of opioid peptides to bind to opioid receptors. ORL1 receptor is clearly insensitive to the Phe 1 3 Tyr substitution at the first position of nociceptin.
In sharp contrast, nociceptin-(1-11) was found to be inactive, lacking completely an ability to bind to ORL1 receptor (Table  I). Nociceptin-(1-11) has a sequence cleaved at the amino side of a basic amino acid pair Arg-Lys at positions 12-13 (Fig. 1). It is clear that a structural element essential for receptor binding exists in the C-terminal sequence of Arg-Lys-Leu-Ala-Asn-Gln. Nociceptin-(1-7), which has a sequence cleaved at the amino side of another Arg-Lys at positions 8 -9, was also completely inactive. Although dynorphin A is also a 17-amino acid peptide, its shortened analogs lacking C-terminal tetra-to nonapeptides have been reported to be almost as potent as parent dynorphin A (11)(12)(13). All these data indicate a distinct divergence in structural frames between nociceptin and dynorphin A in interacting with their specific receptors.
Receptor Binding Activity of Nociceptin and Its Analogs in Guinea Pig Brain-When the binding assays were carried out using guinea pig brain membrane preparations, [ 125 I-Tyr 14 ]nociceptin bound in a dose-dependent manner and showed a very high binding affinity. K d was 0.014 Ϯ 0.002 nM, and B max was 2.8 fmol/mg protein. Using this [ 125 I-Tyr 14 ]nociceptin, nociceptin and Tyr 14 -nociceptin were assayed in guinea pig brain. They exhibited a strong inhibition of the binding of [ 125 I-Tyr 14 ]nociceptin in a dose-dependent manner, and the IC 50 values were 0.0016 and 0.0055 nM, respectively (Table I). Clearly, guinea pig brain contains the receptors to which nociceptin and its analogs can bind. Tyr 1 -nociceptin was here again as potent as nociceptin, while shortened analogs were extremely weak or almost inactive (Table I).
It should be noted that all these nociceptins are much more potent (about 20 -50-fold) in guinea pig brain than in human 293 cells. Meunier et al. (1) have reported that the sequence of nociceptin appears to be conserved across mammalian species, showing that central amino acids of the peptide ARKSAR were identical in the human, rat, and mouse species. The present results may imply that there is a slight difference in sequence between rodent and human nociceptins. Alternatively, the receptors expressed in guinea pig brain might be different from ORL1. It is also possible that receptors of the intact cells has a ligand sensitivity different from that of receptors in membranes prepared from guinea pig brain.
In Vivo Hyperalgesic Effect of Nociceptin and Tyr 1 -nociceptin-Nociceptin and Tyr 1 -nociceptin were evaluated for the in vivo activity in mice after intracerebroventricular administration. Nociceptin elicited hyperalgesia with 1 ng to 1 g i.c.v. per mouse). At a dose of 1 ng, the reaction time was reduced by about 50%, and reaction time was further reduced with as doses increased ( Fig. 2A). At the highest dose (1 g), nociceptin induced a decrease in activity. When Tyr 1 -nociceptin was tested, a similar activity profile was obtained (Fig. 2B). Tyr 1nociceptin also made mice hyperreactive to nociceptive stimulation. This is certainly the reflection of its high binding affinity for ORL1 receptor.
Zhang and Yu (16) have recently reported that dynorphin A activates the potassium channel by stimulating the ORL1 receptor expressed in Xenopus oocytes. This appears to be quite rational for explaining a high activity of Tyr 1 -nociceptin. The sequence identity between dynorphin A and Tyr 1 -nociceptin is more than 40%, sharing the N-terminal tetrapeptide YGGF.
Binding Affinity of Nociceptin and Tyr 1 -nociceptin for Opioid Receptors-Nociceptin and Tyr 1 -nociceptin were assayed for opioid receptors in guinea pig brain, using specific radioligands each for ␦, , and subtypes. Nociceptin bound to ␦ and receptors extremely weakly (IC 50 ϭ 0.8 -7.4 M) (Table II). Among the subtypes, receptors were most favorable to accept nociceptin. Nociceptin-(1-11) was completely inactive for all subtypes of opioid receptors.
When Tyr 1 -nociceptin was assayed, it exhibited considerably increased (8 -37-fold) affinity to all these receptors. In particular, it bound to receptors in guinea pig and rat brains considerably strongly (IC 50 ϭ 21.7 nM). As mentioned above, Tyr 1 -nociceptin possesses exactly the same N-terminal sequence of opioid peptides, namely Tyr-Gly-Gly-Phe, and became much more similar to dynorphin A, a receptor-selective opioid ligand (Fig. 1). In spite of such sequence similarity, Tyr 1 -nociceptin bound to receptors about 10 times more preferentially than to receptors. These results emphasize the difference in binding abilities of nociceptin and dynorphin A against nociceptive and opioid receptors. Apparently, the C- Structural Requirements of Nociceptive Peptide 23644