Receptor Types

Home , Deltorphin, Dynorphin B

Proc. Natl. Acad. Sci. USA Vol. 87, pp. 3180-3184, April 1990 Pharmacology Chimeric opioid peptides: Tools for identifying opioid receptor types (dynorphin/dermorphin/deltorphin/monoclonal antibody/panning) Guo-xi XIE*t, ATSUSHI MIYAJIMA*, TAKASHI YOKOTA*, KEN-ICHI ARAI*, AND AVRAM GOLDSTEINt *Department of Molecular Biology, DNAX Research Institute of Molecular and Cellular Biology, Palo Alto, CA 94304; and tDepartment of Pharmacology, Stanford University, Stanford, CA 94305 Contributed by Avram Goldstein, January 23, 1990

ABSTRACT We synthesized several chimeric peptides in was assumed that the C-terminal amide group ofdermorphin, which the N-terminal nine residues of dynorphin-32, a peptide deltorphins, and DSLET and the alcohol group of DAGO selective for the K opioid receptor, were replaced by opioid could be removed without affecting opioid binding. By anal- peptides selective for other opioid receptor types. Each chi- ogy to dyn-32, which binds selectively to K opioid sites, meric peptide retained the high affminty and type selectivity DAGO-DYN and dermorphin-DYN should bind selectively characteristic of its N-terminal sequence. The common C- to p.; deltorphins-DYN and DSLET-DYN should bind selec- terminal two-thirds of the chimeric peptides served as an tively to 8. mAbs 17.M and 39 should act as nonblocking epitope recognized by the same monoclonal antibody. When antibodies to all these peptides. bound to receptors on a cell surface or membrane preparation, In the present study, we have demonstrated that the these peptides could still bind specifically to the monoclonal chimeric peptides do maintain the high affinities and type antibody. These chimeric peptides should be useful for isolating selectivities of their N-terminal sequences. mAbs 17.M and ,A, 8, and c opioid receptors and for identifying opioid recep- 39 bind to these peptides (as to dyn-32), even after the tors on transfected cells in expression cloning procedures. The peptides are bound to receptors on brain membranes or on general approach using chimeric peptides should be applicable intact NG108-15 neuroblastoma-glioma hybrid cells. to other peptide receptors. MATERIALS AND METHODS Several peptide ligands selective for different opioid receptor Peptides and mAbs. dyn-32 and all the chimeric peptides types have been isolated from natural sources or synthesized. were synthesized on an Applied Biosystems 430A peptide Among them [D-Ala2,N-MePhe4,Gly-ol5]enkephalin synthesizer. Their purities and sequences were confirmed by (DAGO) (1) and dermorphin (2) have an optimal combination HPLC analysis and by sequence analysis on the Applied ofhigh affinity and type selectivity for , binding sites. [D-Pen2, Biosystems 477A protein/peptide sequencer. D-Pen5]enkephalin (DPDPE) (3) and deltorphin I and II (4) mAb 17.M was purified from mouse ascites fluid, by using have similar properties with respect to 8 sites, as do dynorphin a rat anti-mouse IgG antibody affinity column (Boehringer A (dyn A) (5) and its derivatives (6) for K sites. Each of these Mannheim). mAb 39 was produced in hybridoma cell culture peptides has a subnanomolar dissociation constant at its in RPMI 1640 medium (J. R. Scientific, Woodland, CA) with preferred binding site, at least two orders ofmagnitude greater fetal calf serum reduced gradually from 10% to 0%. The affinity than at its next-preferred binding site. supernatant was collected by centrifuging (200 x g, 20C, 10 For experiments on the expression cloning of opioid re- min) and concentrated on a Diaflo ultrafilter (YM type with ceptors we wished to develop a system in which peptide 5000-Da limit, Amicon) under N2 pressure. Finally, the ligands and antibodies against the peptides could be used for antibody was purified by immunoaffinity chromatography as affinity purification or for isolation of receptor-bearing cells. above. Dynorphin-32 (dyn-32) (Fig. 1), our model peptide, is, in Peptides and mAbs were labeled with 1251 by the chloram- effect, a fusion product of two peptides-the 17-residue dyn ine-T method. 1251I-labeled peptides were purified on HPLC. A at the N terminus and the 13-residue dynorphin B (dyn B) 1251I-labeled IgG was purified on a Sephadex G-50 column and at the C terminus, connected by a "bridge sequence" Lys- then on the rat anti-mouse IgG antibody affinity column. Arg. dyn-32 itself is a K opioid agonist (7). Of several ELISA Assays. In antibody titration (dilution) assays, Mi- monoclonal antibodies (mAbs) raised against dyn-32, two crotiter plate (Dynatech) wells were coated with 100 ,ul of0.1 were useful in this study. mAb 17.M requires the bridge and M acetate buffer (pH 4.5) containing chimeric peptide at 5 C-terminal sequences; mAb 39 recognizes only the C- ,ug/ml, incubated at 4°C overnight, and then washed three terminal domain (8). Opioid peptides require their N-terminal times with Dulbecco's phosphate-buffered saline (PBS) sequences for binding. The immediately adjacent residues are (GIBCO, containing calcium and magnesium)/0.05% Tween responsible for the binding-site selectivities, and further 20. One hundred microliters of PBS/0.1% bovine serum C-terminal extensions, in general, contribute little or nothing albumin (BSA) was added to the wells and incubated over- to opioid binding (9). Accordingly, and as documented later night at 40C to block uncoated sites. The plates were washed in this paper, mAbs 17.M and 39 did not block the K-receptor again with PBS/Tween. Different concentrations of mAb binding of dyn-32. Based on the above considerations and the structure of Abbreviations: BSA, bovine serum albumin; DAGO, [D-Ala2, dyn-32, we synthesized several chimeric peptides with the N-MePhe4,Gly-ol5]enkephalin; Pen, penicillamine; DPDPE, [D- sequences of DAGO, dermorphin, deltorphins and [D- Pen2,D-Pen5]enkephalin; DSLET, [D-Ser2,Leu5]enkephalin-Thr; dyn, Ser2,Leu5]enkephalin-Thr (DSLET) at the N termini, fol- dynorphin; dyn-32, dynorphin-32; KHB, Krebs-Hepes buffer; PBS, lowed in every case by residues 10-32 of dyn-32 (Fig. 1). It Dulbecco's phosphate-buffered saline; U50,488, trans-3,4-dichloro- N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]benzeneacetamide methanesulfonate; U69,593, (5a,7a,8,8)-(+)-N-methyl-N-[7-(1-pyrrol- The publication costs of this article were defrayed in part by page charge idinyl)-1-oxaspiro-(4,5)dec-8-yl]benzeneacetamide; mAb, monoclonal payment. This article must therefore be hereby marked "advertisement" antibody. in accordance with 18 U.S.C. §1734 solely to indicate this fact. tTo whom reprint requests should be addressed. 3180 Downloaded by guest on October 1, 2021 Pharmacology: Xie et al. Proc. Natl. Acad. Sci. USA 87 (1990) 3181

Tyr-Gly-Gly-Phe-Leu-Arg-Arg-lIe-Arg-Pro-Lys-Leu-Lys-Trp-Asp-Asn-Gin-Lys-Arg- Opioid Receptor Binding Assay. Guinea pig brain membranes Tyr-Gly-Gly-Phe-Leu-Arg-Arg-GIn-Phe-Lys-Val-Val-Thr were prepared as described (10). About 50,000 cpm per tube Dyn-32 (1-ml assay) of [3H]DAGO (47 Ci/mmol, NEN), [3H]DPDPE (28 Ci/mmol, NEN) and [3H]U69,593 (42 Ci/mmol, NEN) Tyr-D-Ala-Gly-NMe-Phe-Gly-Pro-Lys-Leu-Lys-Trp-Asp-Asn-Gln-Lys-Arg-Tyr-Gly- Gly-Phe-Leu-Arg-Arg-Gln-Phe-Lys-Val-Val-Thr (U69,593 is (5a,7a,8,8)-(+)-N-methyl-N-[7-(1-pyrrolidinyl)- 1-oxaspiro-(4,5)dec-8-yl]benzeneacetamide) was used to label DAGO-DYN ,u, 8, and K sites, respectively, of which 2000-3000 cpm was bound. Different concentrations ofchimeric peptides were used Tyr-D-Ala-Phe-Gly-Tyr-Pro-Ser-Gly-Pro-Lys-Leu-Lys-Trp-Asp-Asn-GIn-Lys-Arg- Tyr-Gly-Gly-Phe-Leu-Arg-Arg-GIn-Phe-Lys-Val-Val-Thr to compete. A set of parallel tubes containing radioligands, competing peptides, and additionally 0.5 ,uM unlabeled DAGO Dermorphin-DYN or DPDPE (Peninsula Laboratories) or U50,488 (trans- Tyr-D-Ala-Phe-Asp-Val-Val-Gly-Phe-Leu-Thr-Pro-Lys-Leu-Lys-Trp-Asp-Asn-Gln- 3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)-cyclohexyl]ben- Lys-Arg-Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Gln-Phe-Lys-Val-Val-Thr zeneacetamide methanesulfonate; Upjohn) was used for sub- Deltorphin I-DYN traction of nonspecific binding (30-50o of total). The binding assays were carried out in Krebs-Hepes buffer (KHB) [118 mM Tyr-D-Ala-Phe-Glu-Val-Val-Gly-Phe-Leu-Thr-Pro-Lys-Leu-Lys-Trp-Asp-Asn-GIn- NaCl/4.8 mM KCl/2.5 mM CaCl2/1.2 mM MgC12/25 mM Lys-Arg-Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Gln-Phe-Lys-Val-Val-Thr Hepes (Hepes is N-2-hydroxyethyl piperazine-N'-2-ethane- Deltorphin Il-DYN sulfonic acid; Research Organics), pH adjusted to 7.4 with NaOH] in 96-tube Microtiter plates (1-ml capacity, Bio-Rad) in Tyr-D-Ser-Gly-Phe-Leu-Thr-Pro-Lys-Leu-Lys-Trp-Asp-Asn-GIn-Lys-Arg-Tyr-Gly- triplicate. After incubation (room temperature, 2 hr), samples Gly-Phe-Leu-Arg-Arg-GIn-Phe-Lys-Val-Val-Thr were kept on ice for 10 min and then filtered on glass filter strips DSLET-DYN (Schleicher & Schuell, grade 32) on a 12-channel LKB har- vester. After three washes with 5 ml of cold KHB, filters were FIG. 1. Sequences of dyn-32 and synthetic chimeric peptides. cut, placed in scintillation vials, and counted. N-terminal segment in boldface type represents unique sequence of For the NG108-15 cells, the binding assay was performed each peptide; remaining C-terminal segment is common to all pep- on intact cells. Cells were cultured in Dulbecco's modified tides. Eagle's medium (J. R. Scientific) containing 10% fetal calf serum (HyClone), vitamins (Sigma, biotin at 7.3 ,g/liter, 17.M or 39 in PBS/BSA were added (100 per well) and lipoate at 200 ,g/liter, B12 at 10 mg/liter) and 100 ,uM incubated at room temperature for 2 hr. In peptide compe- hypoxanthine/0.4 ,uM aminopterin/16 ,uM thymidine (HAT; tition assays, the wells were coated only with dyn-32, and a Sigma) at 37°C with 5% CO2. At half confluent density, cells fixed amount of mAb solution (1 ng per well in 100 jul of were harvested, washed with KHB, and then resuspended in PBS/BSA) was added with different concentrations of chi- KHB/0.05% BSA. The 1-ml assay contained 105 cells. After meric peptides. After incubation, the plates were washed, incubation, cells were separated and washed three times with and horseradish peroxidase-conjugated rabbit anti-serum cold KHB/BSA by centrifugation (200 x g, 2°C, 5 min). Cell against mouse IgG, IgA, and IgM (both heavy and light pellets were dissolved in scintillation solution and counted. chains) (Zymed Laboratories, diluted 1/4000 with PBS/BSA, Integrity of Chimeric Peptides in the Binding Assays. Pep- 100 Aul per well) was added, followed by incubation for 1 hr tide (100 nM, final concentration) containing =50,000 cpm of at room temperature. After washing, 100-Al substrate 2,2'- 125I-labeled peptide was incubated with a guinea pig brain azino-di-(3-ethylbenzthiazoline sulfonate) (ABTS) solution membrane preparation or NG108-15 cells at room tempera- (Boehringer Mannheim) was added, followed by incubation ture for 2 hr. After chilling on ice for 5 min and centrifuging for another 1 hr; then the plates were read at 405 nm on an (10,000 x g, 2°C, 10 min), supernatants were transferred to ELISA reader. new tubes and lyophilized; then they were redissolved in 200

mAb 17.M mAb 39 4 4'

3 3

LO 0 2 2 - 0

0 -7 -6 -5 -4 -3 -2 - 1 -7 -6 -5 -4 -3 -2 -1I 0 Log concentration of mAb, Ag/well

FIG. 2. (Top row) ELISA re- 0.5- 0.5- sults for titration of mAbs 17.M and 39 binding to chimeric pep- LO0 010.4 0.4' tides. (Bottom row) Chimeric a peptides competing for the bind- 0.3 0.3. ing ofmAb 17.M and 39 (1 ng per 0.2 well) to dyn-32. *, dyn-32; o, 0.2- DAGO-DYN; *, dermorphin- 0.11 0.11 DYN; o, deltorphin I-DYN; *, deltorphin II-DYN; A, DSLET- n.L -15 -13 x, -14 -12 -11 -10 -9 -8 -7 -6 -15 -14 -13 -12 -;I -10 -9 -8 -7 -4 DYN; dyn A; +, /3-endor- phin. Points are means of tripli- Log concentration of peptide [Ml cates from single experiments. Downloaded by guest on October 1, 2021 3182 Pharmacology: Me et al. Proc. Natl. Acad. Sci. USA 87 (1990) ,pl of methanol/0. 1 M HCl (vol/vol, 1: 1) and injected onto a (GIBCO, without calcium and magnesium)/0.5 mM EDTA/ 1LBondapak C18 reversed-phase column (30-min linear gradi- 0.02% sodium azide. After suspending once in KHB and ent of 20-50% acetonitrile in 5 mM trifluoroacetic acid, 1.5 centrifuging (200 x g, 2°C, 5 min), cells (2 x 105 per tube) ml/min). Each fraction (0.6 ml) was collected, and radioac- were resuspended in 1 ml of KHB/0.1% BSA containing 100 tivity was determined on a y counter. Counts from fractions nM dyn-32 or chimeric peptide. After 1.5-hr incubation at representing intact 1251I-labeled peptide peaks were compared room temperature and 30 min on ice, cells were washed three with controls, which were peptides incubated without guinea times with cold KHB by centrifugation as above, then pig brain membranes or with boiled membranes. resuspended in 1 ml of PBS/0.5 mM EDTA/0.1% BSA and Antibody-Peptide Sandwich Binding Assay. Different con- plated on the antibody-coated dishes. After 2 hr at room centrations of chimeric peptides were incubated with guinea temperature, the dishes were washed gently three times with pig brain membranes in 1 ml KHB at room temperature for 3 ml of PBS. Cells remaining on the dishes were observed 1.5 hr followed by 30 min on ice. In another set of tubes 0.5 under the microscope. The specificity of a given peptide for ItM DAGO, DPDPE, or U50,488 was included. After incu- panning the NG108-15 cells was tested by competition with bation, samples were washed three times with ice-cold KHB type-selective opioid ligands. by centrifugation (10,000 x g, 20C, 20 min). Pellets were RESULTS resuspended in 1 ml of cold KHB, and about 5 x 105 cpm of Antibody Recognition of Chimeric Peptides. Fig. 2 (top row) 125I-labeled mAb 17.M or 39 was added, followed by incu- shows that mAbs 17.M and 39 recognized dyn-32 and all chi- bation on ice for 1 hr. After washing three times by centrif- meric peptides but not dyn A. In the competition assays (Fig. ugation, pellets were transferred to new tubes, and radioac- 2, bottom row) all chimeric peptides were able to block the tivity was determined. Assays without guinea pig brain binding of mAbs 17.M and 39 to dyn-32; IC50 values were in membranes or without peptides served for background de- the high picomolar to low nanomolar range. dyn A and 8-en- termination. dorphin, which lack the C-terminal epitope, were ineffective. Panning of NG108-15 Cells. We used a procedure modified Opioid Receptor Binding Selectivities of Chimeric Peptides. from that described by Seed and Aruffo (11). Petri dishes DAGO-DYN and dermorphin-DYN exhibited typical high (60-mm diameter) were coated with mAb 17.M (30 ,ug per dish in 3 ml of 50 mM Tris-HCI buffer, pH 9.5) for 2 hr and then washed three times with 0.15 M NaCl. Three milliliters of 0.1% BSA/PBS was added, followed by incubation at 4°C overnight. NG108-15 cells at half confluent density were detached from cell culture dishes by incubation with PBS 0 x 100 E 80 0c 60 40 20 0 HPLC fraction no.

100 1201 b

C=' 80 mC: 60 o 40 0 CD 20 4- C 0 0 0 0 4-a) 0) 100- cu C 01) 40 80- a) IL 60 20 40 201 &o DA AF A -10 -9 -8 -7 -6 -5 Log concentration [M] Guinea. pig brain membranes NG108-15 cells FIG. 3. Competition for binding by chimeric peptides in guinea FIG. 4. HPLC analysis of peptide degradation in receptor- pig brain membrane preparation. (a) ,u ligand [3H]DAGO. (b) 8 ligand binding assays with guinea pig brain membranes or NG108-15 cells. [3H]DPDPE. (C) K ligand [3H]U69,593. Specific binding of tritiated (a) -. . , Control 125I-labeled dyn-32 incubated with boiled ligands without competition was normalized to 100%. Points are membrane preparation; -, 1251I-labeled dyn-32 incubated with mem- means of triplicates from single experiments. ---, Competition by branes. (b) Chimeric peptides; each bar represents the percentage DAGO (a), DPDPE (b), or U50,488 (c). *, dyn-32; o, DAGO-DYN; area of intact 125I-labeled peptide HPLC peak compared with boiled *, dermorphin-DYN; o, deltorphin I-DYN; *, deltorphin Il-DYN; A, membrane control (left) or cell-free control (right). DERM, dermor- DSLET-DYN. phin; DEL, deltorphin. Downloaded by guest on October 1, 2021 Pharmacology: Xie et aL Proc. Natl. Acad. Sci. USA 87 (1990) 3183 affinities and selectivities for Au sites in competing with 1U was 410,000 sites per cell for [3H]DPDPE. Ifeither the peptide ligand [3H]DAGO (Fig. 3a), 8 ligand [3H]DPDPE (Fig. 3b), ligand or antibody was omitted, very few cells were retained and K ligand [3H]U69,593 (Fig. 3c) in standard binding assays. on the dish (data not shown). Typical panning results are Deltorphin I-DYN, deltorphin I-DYN, and DSLET-DYN shown in Fig. 6. Cells adherent to a plate coated with mAb had high affinity and selectivity for 8 sites, and the deltorphin- 17.M after incubation with DAGO-DYN are shown before DYN peptides did not compete for K binding even at 10 ILM (Fig. 6a) and after (Fig. 6b) the wash step in the panning concentration. dyn-32 displayed its well-known high affinity procedure; no significant adherence of cells to the dish was and selectivity for K sites. observed. The same negative result was seen when cells were Integrity of Chimeric Peptides in Binding Assays. After 2-hr incubated with dermorphin-DYN or dyn-32 (data not shown), incubation with guinea pig brain membranes, only 23% of consistent with the absence of A and K receptors from these added dyn-32 remained intact in the supernatant; the remain- cells. In contrast, after incubation with deltorphin I-DYN, der was degraded or formed complexes with membrane most cells were retained on the antibody-coated dishes after components that were eluted in different fractions on HPLC the wash, even when a K or, ligand (e.g., DAGO, Fig. 6c) was (Fig. 4a). All the chimeric peptides were degraded to about present during the incubation. Similar results were obtained the same extent (Fig. 4b). In the binding assay with NG108-15 with deltorphin II-DYN or DSLET-DYN (data not shown). cells, 40-50% of the chimeric delta ligands were degraded However, the positive panning result could be prevented by (Fig. 4b). incubation with 0.5 ,uM DPDPE (Fig. 6d). Antibody Recognizes Chimeric Peptides Bound to Recep- tors. mAbs 17.M and 39 were able to bind not only the free DISCUSSION form of the chimeric peptides but also the receptor-bound To use peptide ligands and antibodies against the peptides for form. Fig. 5 shows that the binding of 125I-labeled mAb 17.M identifying and isolating peptide receptors, three important to the peptide-receptor complex was dependent on peptide requirements should be realized: (i) The peptide should have ligand concentration and was competitively reduced by a high high affinity and type-selectivity for a particular receptor concentration of the corresponding type of selective ligand. type. (ii) The peptide should be long enough so that when From the specific activity of radiolabeled IgG (41000 Ci/ bound to the receptor, sufficient free epitope still exists for mmol), the amount of membrane preparation, and the as- antibody recognition. (iii) Binding of the antibody to the sumption that each IgG molecule binds two molecules of peptide should not affect binding of the peptide to the peptide, we could calculate the numbers ofreceptor sites that receptor. In the case of opioid receptors, most selective were saturated by ligand and recognized by antibody. The peptide ligands are relatively short, leaving little to protrude highest numbers found [pmol/g of brain (wet weight)] were from the cell membrane after binding. And even for peptides 0.44 for ,u, 0.39 for 8, and 0.28 for K. Corresponding Bma,, with suitable length, raising a nonblocking antibody to each values from standard ligand binding assays were 0.82, 1.1, one separately would be a laborious procedure. and 2.8, respectively (10). We have simplified the approach by synthesizing chimeric To test whether the mAb was able to bind chimeric peptides peptides with high-affinity, type-selective sequences at the N that had already bound to receptors on cell surfaces, we terminus (which, in the case of opioid peptides, binds to the carried out the experiment ofpanning NG108-15 cells on mAb receptor), and we have placed at the C terminus a common 17.M-coated dishes. These cells express only 8 opioid recep- peptide sequence containing the epitopes for either of two tors (12); under our culture and binding assay conditions Bm., existing mAbs. As expected, these peptides possessed high 6 .a DAGO-DYN b Dermorphin-DYN 5. 5- foa ,,,3 4. 4 ,--"- A 3 _0 Speciic _ LJ n' _~~~' I 2 -W R - - D -+-A -D I- Specific C 0 2 4 6 8 10 2 4 6 8 10 2

0 10 7 - Deltorphin l-DYN Deltorphin Il-DYN X 8 d E 5 -- CL * ._ 4 I" ope :5 C - --. '2- ,o1,-A' Specific E3 ~~Specific0 ~~~~- _0 1I0 - -41- 0 5 10 1! 1 2 3

, 7 f Dynorphin-32 6 e DSLET-DYN - Iro',3 6 aOa.

6 W r 5A FIG. 5. Binding of mAb 4- k.0 17.M to the chimeric peptide- 3 .0I -I. 0 3 _ membrane receptor complexes. 2 -A-S- 2 13 Points are means of triplicates A from single experiments. ---, To- 1 ~~~~~~~Specific tal binding in the presence of0.5 0 5 10 15 ,uM DAGO (a, b), DPDPE (c-e) U V.,O I.U 1.:> Z.Q or U50,488 (f); -, specific bind- Concentration of chimeric peptide [nM] ing (difference plot). Downloaded by guest on October 1, 2021 3184 Pharmacology: Me et al. Proc. Natl. Acad. Sci. USA 87 (1990) guinea pig brain membranes (10 mg/ml, based on initial wet weight of brain) in a very short incubation time. Diluting the membranes could reduce the extent of degradation, but this procedure may be impractical in a binding assay. Here we observed that dyn-32 and otherchimeric peptides, even at very high concentration (100 nM), were degraded as much as 70% by membranes and 40-50% by intact cells under binding assay conditions. Boiling the membrane preparation prevented such degradation. It seems that a membrane-bound, high-capacity peptidase is responsible for the degradation. Our observed IC5o values for these peptides and the labeling of receptor- bound peptides by antibody are probably affected by degradation, and this may contribute to the substantially lower binding with the antibody method when compared with Bm. values obtained in a direct radioreceptor assay. However, as was shown directly with another C-terminally extended dynorphin (6), peptides that are already bound to both receptor and d antibody may be less susceptible to attack by peptidases. The chief reason for the low binding is probably that although the binding sites were certainly saturated by the high ligand concentrations used, the conditions ofour experiments did not necessarily favor the complexing of antibody to all bound ligand molecules, and dissociation of ligands or ligand- antibody complexes is not excluded. Despite all these poten- tial problems, specific recognition of the three types of occu- pied sites by the mAbs was satisfactory, and the complexing was sufficient for specific panning of cells expressing 8 receptors. The utility of these chimeric peptides is obvious. Immobi- lizing mAbs 17.M or 39 or both should facilitate isolating different types of opioid receptors. We have transiently trans- FIG. 6. Panning of NG108-15 cells on mAb 17.M-coated dishes. fected COS-7 cells with a guinea pig brain cDNA library, then (Top row) Cells were incubated with DAGO-DYN, and free ligand incubated them with dyn-32, and panned them on 17.M coated was removed by centrifuging; then cells were plated on antibody- dishes as described here. Plasmids were recovered from the coated dishes for 2 hr. (a) Before wash step in panning procedure. (b) panned cells and used for another transfection and panning. After wash step. (Bottom row) Cells were incubated with deltorphin I-DYN together with another ligand (DAGO (c); DPDPE (d)) and After several such cycles, we observed an apparently suc- then subjected to the wash step. cessful panning result (similar to that shown with cells bearing 8 receptors in Fig. 6), possibly due to expressed K receptors, affinities and selectivities for the several types of opioid as it could be prevented specifically by U50,488 (unpublished receptors, very similar to these properties of the original data). The chimeric peptide method described here should be ligands. mAbs 17.M and 39 recognized these chimeric pep- generally applicable to other peptide receptors. tides and dyn-32 (against which they were raised) equally well. All We thank Grace Hsu (DNAX) for synthesis and sequence analysis the chimeric peptides could compete for binding of of the peptides. Dr. Ronald W. Barrett (Affymax Research Institute) mAbs 17.M and 39 to dyn-32. kindly commented on the manuscript and also provided mAb 17.M The results from guinea pig brain membrane binding assays ascites fluid, mAb 39 hybridomas, and NG108-15 cells. DNAX and from panning of NG108-15 cells indicate that after the Research Institute is supported by Schering-Plough Corporation. chimeric peptides bound to opioid receptors through their N-terminal domains, their C-terminal extensions were long 1. Gillan, M. G. C. & Kosterlitz, H. W. (1982) Br. J. Pharmacol. 77, 461-468. enough for antibody recognition. It has been known that the 2. Montecucchi, P. C., de Castiglione, R., Piani, S., Gozzini, L. & N-terminal 13 residues of the endogenous K ligand dyn A are Erspamer, V. (1981) Int. J. Pept. Protein Res. 17, 275-283. necessary and sufficient for affinity and specificity (9). dyn 3. Mosberg, H. I., Hurst, R., Hruby, V. J., Gee, K., Yamamura, A-analogue kappa ligand (DAKLI), a dyn A-derived peptide H. I., Galligan, J. J. & Burks, T. F. (1983) Proc. Nat!. Acad. Sci. ligand designed on this principle, has high affinity and selec- USA 80, 5871-5874. 4. Erspamer, V., Melchiorri, P., Falconieri-Erspamer, G., Negri, L., tivity for K opioid receptors; and biotinylated DAKLI, even Corsi, R., Severini, C., Barra, D., Simmaco, M. & Kreil, G. (1989) after it is bound to the receptors, can bind avidin (6). In the Proc. Natl. Acad. Sci. USA 86, 5188-5192. present study, the C-terminal amide group of dermorphin, 5. Goldstein, A., Fischli, W., Lowney, L. I., Hunkapiller, M. & Hood, deltorphins, and DSLET, as well as the alcohol group of L. (1981) Proc. Natl. Acad. Sci. USA 78, 7219-7223. DAGO were removed, and a 23-residue C-terminal extension 6. Goldstein, A., Nestor, J. J., Jr., Naidu, A. & Newman, S. R. (1988) Proc. Natl. Acad. Sci. USA 85, 7375-7379. was added. The maintenance of high affinity and type- 7. Fischli, W., Goldstein, A., Hunkapiller, M. & Hood, L. (1982) Proc. selectivity by these synthetic chimeric peptides implies that Natl. Acad. Sci. USA 79, 5435-5437. their C-terminal sequences are not directly involved in re- 8. Barrett, R. W. & Goldstein, A. (1985) Neuropeptides 6, 113-120. ceptor binding. 9. Chavkin, C. & Goldstein, A. (1981) Proc. Natl. Acad. Sci. USA 78, Degradation of peptide ligands in receptor-binding assays 6543-6547. 10. Goldstein, A. & Naidu, A. (1989) Mol. Pharmacol. 36, 265-272. can be tested by HPLC or equivalent analytical method. 11. Seed, B. & Aruffo, A. (1987) Proc. Natl. Acad. Sci. USA 84, Goldstein et al. (6) demonstrated that dynorphin peptides, at 3365-3369. low concentrations (<1 nM), were completely degraded by 12. Chang, K.-J. & Cuatrecasas, P. (1979)J. Biol. Chem. 2,2610-2618. Downloaded by guest on October 1, 2021