* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Purification, Cloning, and Tissue Distribution of a 23
Amino acid synthesis wikipedia , lookup
Epitranscriptome wikipedia , lookup
Artificial gene synthesis wikipedia , lookup
Size-exclusion chromatography wikipedia , lookup
Biosynthesis wikipedia , lookup
Gene expression wikipedia , lookup
Paracrine signalling wikipedia , lookup
Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup
Expression vector wikipedia , lookup
Signal transduction wikipedia , lookup
Genetic code wikipedia , lookup
Biochemistry wikipedia , lookup
Magnesium transporter wikipedia , lookup
Point mutation wikipedia , lookup
Bimolecular fluorescence complementation wikipedia , lookup
Metalloprotein wikipedia , lookup
G protein–coupled receptor wikipedia , lookup
Ancestral sequence reconstruction wikipedia , lookup
Interactome wikipedia , lookup
Clinical neurochemistry wikipedia , lookup
Homology modeling wikipedia , lookup
Nuclear magnetic resonance spectroscopy of proteins wikipedia , lookup
Protein structure prediction wikipedia , lookup
Protein–protein interaction wikipedia , lookup
Western blot wikipedia , lookup
Purification, Cloning, and Tissue Distribution of a 23-kDa Rat Protein Isolated by Morphine Affinity Chromatography David K. Grandy*, Eric Hanneman, James Bunzow, Marjorie Shih, Curtis A. Machida, Jean M. Bidlack, and Olivier Civelli Vollum Institute for Advanced Biomedical Research (D.K.G., J.B., M.S., O.C.) and the Department of Cell Biology and Anatomy (O.C.), Oregon Health Sciences University, Portland, Oregon 97201; the Department of Zoology, University of Arizona (E.H.), Tucson, Arizonia 85721; the Oregon Regional Primate Center (C.A.M.), Beaverton, Oregon 97006; and the Department of Pharmacology, University of Rochester Medical Center (J.M.B.), Rochester, New York 14642 A 23-kDa (p23k) rat brain protein was stereospecifically eluted from a 14/3-bromoacetamidomorphine affinity column, purified to apparent homogeneity by reverse phase HPLC, and partially sequenced. Three degenerate oligodeoxynucleotide probes were synthesized based on this partial amino acid sequence. A rat brain cDNA library was screened using these probes, and a full-length cDNA was isolated. The deduced protein, 187 amino acids long, is rich in glutamic and aspartic acid residues, endowing p23k with a net negative charge at neutral pH. The protein lacks a signal sequence as well as any transmembrane domains. Based on predictions of secondary structure, p23k is a globular protein composed of 30% a-helices and 18% ^-pleated sheets. Northern blot analysis revealed p23k transcripts in rat brain, liver, and the mouse x rat neuroblastoma-glioma NG108-14 cell line. Although not an opioid receptor itself, this protein may be associated with such a receptor or be related to a protein that has been shown to be cross-linked to the opioid peptide 0endorphin. (Molecular Endocrinology 4: 1370-1376, 1990) detergent-solubilized membrane proteins has been shown to be contained within a large phospholipidglycoprotein complex (1-5) of 100-1000 kDa (6-10) with a Stokes radius of 700A (11,12). With the development of highly specific ligands and the use of derivatized morphine affinity columns, proteins with ^-opiate receptor-like binding profiles have been purified to homogeneity from bovine (13) and rat brain membranes (14). Recently, Scholfield et al. (15) cloned a bovine cDNA which encodes a 58-kDa morphine-binding protein. Surprisingly, the deduced amino acid sequence of this morphine-binding protein was found to be homologous to that of members of the immunoglobulin protein superfamily that are thought to be involved in cell adhesion. Previously, we reported that a detergent-solubilized rat brain membrane preparation is enriched in ^-opiatebinding activity after affinity chromatography over 14/3bromoacetamidomorphine (7, 16, Bidlack, J., W. E. O'Malley, D. K. Grandy, and O. Civelli, submitted). When this eluate is run in sodium dodecyl sulfate (SDS)polyacrylamide gels and stained with silver, several proteins are resolved. Two of these, 23 kDa (p23) and 36 kDa, were of interest to us because of reports that a protein of 23-25 kDa can be covalently cross-linked to human /?-endorphin (/?h-endorphin) (18-20), the possibility that the 23- and 36-kDa proteins are proteolytic cleavage products of a protein with the mol wt expected of an opioid receptor, and the fact that they are two of the more abundant proteins in the affinity column eluate. To determine whether p23k is a proteolytic fragment of a larger protein and to begin an analysis of its tissue distribution we have undertaken its molecular characterization. Here we report the purification of p23k and the cloning of its cDNA. INTRODUCTION The opiate receptors are a heterogeneous group of proteins which bind a number of medically important plant alkaloids and vertebrate brain peptides. The best studied of these receptors is n, which has a high affinity for morphine. The stereospecific binding of morphine to 0888-8809/90/1370-1376$02.00/0 Molecular Endocrinology Copyright © 1990 by The Endocrine Society 1370 Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009 1371 cDNA of a Rat Morphine Affinity Column Binding Protein RESULTS AND DISCUSSION Isolation and Purification of p23k Approximately 200 mg Triton X-100-solubilized rat brain membrane protein were applied to the 14/?-bromdacetamidomorphine (/3AM1) affinity column and eluted with either 100 MM dextrorphan (DEX) or 100 I*M levorphanol (LEV), the inactive and active stereoisomers of morphine, respectively. When DEX was used, less than 10 ng protein were eluted from the column. When half of this material was run in a SDS-polyacrylamide gel, a faint band, with a relative molecular mass of 36 kDa (Fig. 1A, inset), was stained with silver. Reverse phase HPLC of the remainder of the DEX-eluted material revealed little absorbance at 214 nm (Fig. 1A). In contrast, 100 MM LEV reproducibly eluted 100200 Mg protein from the /3AM column. SDS-polyacrylamide gel electrophoresis (SDS-PAGE) of a fraction of the LEV eluate resolved proteins of 28,36, and 43 kDa, which were heavily stained with silver, and proteins of 23, 29, 54, 58, 66, 80, and 90 kDa, which stained weakly (Fig. 1B, inset). A similar staining pattern was observed when the /3AM column was eluted with 100 2 4 5 M M [D-Ala ,MePhe ,Gly-ol ]enkephalin, 100 MM etorphine, or 100 MM morphine (16). When assayed for /3h- endorphin-binding activity, the LEV-eluted protein complex typically binds [125l]/3h-endorphin with a binding capacity of 125 pmol/mg protein and a Kd of 2.2 nM, representing a 1200-fold enrichment of opioid-binding activity (Bidlack, J., W. E. O'Malley, D. K. Grandy, and O. Civelli, submitted). A typical UV chromatogram of the LEV eluate after reverse phase HPLC is presented in Fig. 1B. Nearly identical absorbance profiles were obtained when morphine (16), etorphine, and [D-Ala2,MePhe4,Gly-ol5]enkephalin (data not shown) were used as the eluting ligands. SDS-PAGE and silver staining of individual HPLC fractions revealed that the two major peaks of absorbance, with retention times of 21 and 42 min, corresponded to proteins of 23 kDa (p23k) and 36 kDa (p36k), respectively, and that both were purified to apparent homogeneity. It is noteworthy that p23k is a predominant species in the LEV preparation, but reacts very weakly in the silver-staining reaction. The reverse phase HPLC of the LEV eluate also resulted in the purification of a second protein of 36 kDa, with a retention time of 25 min, as well as proteins of 28, 43, 58, and 80 kDa. Due to reports that proteins of 23-25 kDa are stereospecifically labeled by [125l]/3h-endorphin (18-20) and in order to determine whether the 23- and 36-kDa proteins in our preparation are proteolytic fragments of a 58- to 60-kDa protein, we proceeded with the cloning of p23k. 25 Time (min) Fig. 1. SDS-PAGE and Reverse Phase HPLC of Proteins Stereospecifically Eluted from the /3AM Affinity Column Silver-stained SDS-polyacrylamide gel of the DEX eluate revealed a very faint band of 36 kDa (inset to A), whereas the LEV eluate contained a number of proteins (inset to B) before reverse phase HPLC. After chromatography of the C4 reverse phase HPLC column, the 214 nm absorbance profiles of the DEX (A) and LEV (B) eluates were recorded. Individual HPLC fractions were analyzed by SDS-PAGE and silver staining, and the peaks of absorbance were assigned a molecular mass, as indicated, relative to those of protein mol wt standards, expressed as Mr x 10~3. Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009 Vol 4 No. 9 MOL ENDO-1990 1372 Trypsin Digestion and Partial Amino Acid Sequence of p23k Two attempts to obtain N-terminal sequence information from purified p23k were unsuccessful, probably due to a modification of the protein's N-terminus. To obtain partial amino acid sequence information, p23k was digested with trypsin, and the resulting peptides were fractionated by reverse phase HPLC on a C4 column in an acetonitrile gradient. The elution profile of the digest is shown in Fig. 2. A total of nine tryptic peptides were subjected to N-terminal gas phase microsequence analysis. Six peptides (a, b, c, d, i, and I) yielded nonoverlapping sequence information, whereas peptides m, q, and s could not be sequenced. The inability to sequence these peptides may have been due to a blocked N-terminal residue. Cloning of p23R1-1.1 The amino acid sequences determined for tryptic peptides b, d, and I were used to design three pools of degenerate oligodeoxynucleotide probes (see Materials and Methods). A Xgt1O rat brain cDNA library was screened with these probes, and 90 positive clones were identified among 600,000 recombinants. Thirty of these were isolated, and their inserts analyzed. The longest insert, a 1.1-kilobase (kb) EcoRI fragment, was subcloned into pUC18 (p23R1-1.1). A partial restriction map of p23R1-1.1 and a summary of the dideoxy sequencing strategy are diagrammed in Fig. 3A. The complete nucleotide and deduced amino acid sequence of the p23k cDNA clone is presented in Fig. 3B. The first AUG codon found in the cDNA is situated within the Kozak consensus sequence, GUCAUGG, which has been identified as an efficient site for the initiation of translation (21). Beginning with this codon, the nucleotide sequence predicts an acidic protein of 187 amino acids, with a molecular mass of 20,806 daltons. Several pieces of evidence support the conclusion that the 1.1-kb cDNA codes for the complete p23k protein; all six of the sequenced tryptic peptides are contained within this polypeptide (Fig. 3B), the deduced amino acid sequence is in agreement with the amino acid analysis of acid-hydrolyzed p23k protein (data not shown), and, as noted below, both the p23k cDNA and the mRNA prepared from rat tissues are approximately 1.1 kb in length (Fig. 4). The apparent discrepancy between the relative molecular mass estimated from SDS-PAGE analysis and the deduced amino acid sequence may be due to posttranslational modifications. One site for N-linked glycosylation is found at Asn140, and Ser51 and Ser153 are potential target sites for protein kinases-A and -C, respectively (22). Also worthy of note are two potential endopeptidase cleavage sites (23) located at residues Arg76Lys77 and Arg155Lys156l_ys1S7. The open reading frame, terminated by UAG, is followed by 417 bases of the 3' sequence that contain the consensus polyadenylation signal sequence, AAUAAA. Tissue Distribution of p23k mRNA A preliminary analysis of the tissue distribution of p23k mRNA revealed a major transcript of 1.1 kb in rat brain (minus the cerebellum), rat cerebellum, and rat liver (Fig. 4). Poly(A)+ mRNA prepared from the 5-opioid receptorbearing rat x mouse hybrid cell line NG1-8-15 was also examined and was found to express both a 1.1- h ij 8 C •8 O 0.03 Time (minutes) Fig. 2. Reverse Phase HPLC of p23k Tryptic Peptides Reverse phase HPLC-purified p23k was digested with trypsin, as described in Materials and Methods, and the resulting peptides were purified by a single pass over a C4 reverse phase HPLC column. The peaks specific to p23k are labeled a-s. Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009 1373 cDNA of a Rat Morphine Affinity Column Binding Protein ATG Hincll Sal I Pvull TAG Pstl A l-t Eco Rl (0) .(n) Eco Rl (1047) scale:l B 100 bp ' -61 T AAACGCAGGG CGATTGCATT CGGGACTCGG CGCGCGCGCG TGTCTGTTCT CTCCATCGTC +1 ATG GCC GCC GAC ATC AGC CAG TGG GCC GGG CCG CTG TCA TTA CAG GAG GTG GAT GAG CCG MET Ala Ala Asp lie Ser Gin Trp Ala Gly Pro Leu Ser Leu Gin Glu Val Asp Glu Pro +61 CCC CAG CAC GCC CTG AGG GTC GAC TAC GGC GGA GTA ACG Pro Gin His Ala Leu Arg Val Asp Tyr Gly Gly Val.Thr d * +121 CTG ACG CCC ACC CAG GTC ATG AAT AGA CCA AGT AGC ATT Leu Thr Pro Thr Gin Val MET Asn Arg Pro Ser Ser Tip +60 GTG GAC GAG CTG GGC AAA GTG +120 Val Asp Glu Leu Gly T.vs Val TCA TGG GAT GGC CTT GAT CCT +180 Ser Trp Asp Gly Leu Asp Pro +181 GGG AAG CTC TAC ACC CTG GTC CTC ACA GAC CCC GAT GCT CCC AGC AGG AAG GAC CCC AAA +240 Gly Lys Leu Tvr Thr Leu Val.Leu Thr Asp Pro Asp Ala Pro Ser Arg Lys Asp Pro Lys +241 TTC AGG GAG TGG CAC CAC TTC CTG GTG GTC AAC ATG AAG GGC AAC GAC ATT AGC AGT GGC +300 Phe Arg Glu Trp His His Phe Leu Val Val Asn MET Lys Gly Asn Asp lie Ser Ser Gly +301 ACT GTC CTC TCC GAA TAC GTG GGC TCC GGA CCT CCC AAA GAC ACA GGT CTG CAC CGC TAC +360 Thr Val Leu Ser Glu Tyr Val Gly Ser Gly Pro Pro Lys Asp Thr Gly Leu His Arg Tyr 3 +361 GTC Val +421 AAG Lys TGG Trp TCT Ser CTG Leu GGA Gly GTG Val GAC Asp TAT Tyr AAC Asn GAG Glu CGC Arg CAG Gin GGC Gly GAG Glu AAG Lys CAG Gin TTC Phe CCT Pro AAG Lys CTG Leu GTG Val AAC Asn GAG Glu TGT Cys TCC Ser GAC Asp TTC Phe GAG Glu CGC Ara CCC Pro AAG Lys ATC lie AAG Lys CTC Leu TAC Tyr AGC Ser CAC His AAC +420 Asn CTG +480 Leu +481 GGA GCC CCG GTG GCC GGC ACG TGC TTC CAG GCA GAG TGG GAT GAC TCT GTG CCC AAG CTG +540 Gly Ala Pro Val Ala Gly Thr Cys Phe Gin Ala Glu Trp Asp Asp Ser Val Pro Lys Lfiu + 541 CAC GAT CAG CTG GCT GGG AAG TAGGGGCGCT GCAGAGCCCG CAGCCCCGGG GACCCCACAG TACAG His Asp,Gin Leu Ala Glv Lvs b +607 TCAAG TCGTATAAAG CATGTGCTGT GGGGTGTCCC CCACGCCCAT CCTTCCTTCC CACCCTCTCA TAGGG +606 +676 +677 AGTTC TCAGTTGTGC TAGGTTACAG CTCTAGGATG TCTTCCACTT TGTCCAGGAC CAGGCCCAGT AACAT +746 +747 CTTTT GGGGTGGGCT TATCAATCCT CCCATCTCGG CTGAGCCCTG ACCGCCCAGG TCAGATGGCT GCATA +816 +817 GTTAT CAATATTCCT GGGCTGCTGC TCAGCAGTGC TGCTGTGTGG AGGCCAGCTG TGGAGAGAGA CCCTG +886 +887 TTAGC CCTTACATCC CAGTGGGATA AGCAAAAGTC ACCGGAGTTG CTGGGCGTGT TAAACCTCAT CAAAT +956 +957 ACAAA TAAAGGGCAT TGCATTCAAA AAAAA +986 Fig. 3. Restriction Map, Sequencing Strategy, and the Complete Nucleotide and Deduced Amino Acid Sequence of the p23R11.1 cDNA A, The p23k-coding region is outlined by the open box, with the restriction sites used in the sequencing strategy marked. The arrows indicate the direction and extent of the nucleotide sequence determined. B, The complete nucleotide and deduced amino acid sequences of p23R1-1.1 are shown. The tryptic peptides that were sequenced are underlined and identified with letters corresponding to the peaks in Fig. 2. The polyadenylation signal sequence is indicated by the dotted line. Potential target sites for posttranslation modification are marked, with the N-linked glycosylation site boxed and the sites for kinase activity indicated by an asterisk. kb and a 1.2-kb transcript. The detection of two transcripts in NG108-15 may reflect size differences between mouse and rat mRNAs, but also raises the possibility of alternative splicing of p23k pre-mRNA. Structural Analysis of p23k A hydrophobicity analysis (24) predicted that p23k does not contain a transmembrane domain or a good signal sequence (25). From these analyses it is unlikely that p23k is an integral membrane protein; however, in our preparation it comigrates with solubilized membrane proteins, suggesting that it is membrane associated. A Chou-Fassman analysis (26) of secondary structure predicts that p23k is a globular molecule consisting of 30% X-helices and 18% j8-pleated sheets. The net electrical charge of p23k is - 6 at neutral pH. However, the C-terminal third (residues 126-187) contains 43% of the molecule's charged residues, bestowing on this region a net charge of +3. Fast DBr and Fast Ar searches of nucleic acid (GenBank, European Molecular Biology Laboratory, Heidelberg, Germany) and amino acid sequences (PIR) failed to find any other published sequence with significant identity with p23k. Therefore, one can presently only guess at the biological activity Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009 MOL ENDO-1990 1374 Vol 4 No. 9 distribution of its mRNA, preclude p23k from being a G-protein-coupled opioid receptor. However, it may be an opioid receptor-associated protein or a protein with some enzymatic or structural function. The expression of p23k in liver is intriguing and suggests the additional possibility that it may be involved in the catabolism of opiates. The availability of the cDNA encoding p23k will allow us to test these hypotheses by expression in eucaryotic cells. MATERIALS AND METHODS Animals Male Sprague-Dawley rats were purchased from Charles River Breeding Laboratories (Wilmington, MA) and treated in accord with the Guidelines for Care and Use of Experimental Animals. - 1 Materials Fig. 4. Tissue Distribution of p23k-Encoding mRNA Ten micrograms of poly(A)+ mRNA from NG108-15 neuroblastoma x glioma cells (A), rat brain minus cerebellum (B), rat liver (C), and cerebellum (D) were electrophoresed, blotted, and probed with the nick-translated p23R1-1.1 insert, as described in Materials and Methods. The sizes, in kilobases, of the major transcripts are indicated relative to sizes of RNA mol wt markers. A single band of 1.1 kb is present in all tissues. A doublet is observed in the hybrid cell line. of this protein. Our chromatographic experiments suggest that the opioid-binding protein is eluted as part of a complex of proteins, although several of the detected proteins might coelute fortuitously. p23k is probably not the opioid receptor per se, but its similarities to the jS-endorphin-cross-linkable protein suggest that it could modulate opioid binding. In summary, a 23-kDa rat brain protein, which is stereospecifically eluted from a derivatized morphine affinity column, was purified by reverse phase HPLC. Based on tryptic peptides derived from p23k, degenerate oligonucleotide probes were made and used to clone its cDNA. These cloning experiments demonstrated that p23k is not a proteolytic fragment of a larger protein. Furthermore, when the amino acid and nucleotide sequences of p23k were used to query the most current data bases, no significant alignments were found. Therefore, we conclude that p23k is a membrane-associated protein of unknown function and that it is enriched by morphine affinity chromatography. Currently, our efforts are directed at establishing whether p23k can be cross-linked to /?-endorphin. Three findings, the abundance of the protein, that the protein lacks a hydrophobic domain, and the tissue Opioid peptides were obtained from Bachem (Torrance, CA) or Peninsula Laboratories (Belmont, CA). DEX, LEV, and etorphine were obtained from the NIDA. HPLC grade acetonitrile was purchased from J. T. Baker (Philipsburg, NJ), and Sequanol grade trifluoroacetic acid (TFA) was obtained from Pierce Chemical Co. (Rockford, IL). The HPLC system consisted of a pair of Waters series 510 pumps (Milford, MA), a U6k injector, automated gradient controller, and a model 411 absorbance detector. A Vydac (Hesperia, CA) 25-mm C4 reverse phase HPLC column with a particle size of 300 A was used in conjunction with an Uptight guard column packed with C-2 resin. Protein electrophoresis grade reagents were purchased from Bio-Rad (Richmond, CA). Colony/plaque screen filters, intensifying screens, [o-32P]ATP, [a-32P]ATP, and [a32 P]dCTP were acquired from NEN-DuPont (Boston, MA). Reverse transcriptase was prepared by Seikagaku, Inc. (St. Petersburg, FL). EcoRI-digested and dephosphorylated Xgt10 arms and Gigapak Plus were purchased from Stratagene (La Jolla, CA). DNA restriction and modification enzymes were purchased from New England BioLabs (Beverly, MA), Boehringer Mannheim (Indianapolis, IN), Bethesda Research Laboratories (Gaithersburg, MD), Promega (Madison, Wl), and IBI (New Haven, CT) and used according to the manufacturers' specifications. Membrane Preparation, Solubilization, AM Affinity Column Chromatography, and SDS-PAGE Membranes were prepared from rat brains as previously described (7), suspended at a final protein concentration of 10 mg/ml in 50 ITIM Tris-HCI, pH 7.5, with protease inhibitors, and stored at - 7 0 C. The inhibitors added were 1 HIM K2EDTA, 2.5 Mg/ml bacitracin, 10 /xg/ml soybean trypsin inhibitor, and 5 Mg/ml leupeptin. Triton X-100 solubilization of membrane protein was performed as previously described (7). Protein concentration was determined by the method of Bradford (27), with BSA as the standard. Approximately 200 mg soiubilized membrane protein were applied at a flow rate of 15 ml/h to 10 ml /3AM affinity resin at 4 C. The /3AM affinity matrix was prepared as previously described (7). The column was washed with at least 20 column vol 50 mw Tris-HCI, pH 7.5, until the 225 nm absorbance was zero, and then a ligand was applied to the column. The eluate was pooled and concentrated on an Amicon PM-10 membrane (Lexington, MA) at 4 C. The sample was either analyzed immediately or stored at - 7 0 C. For gel analysis, approximately 3 M9 protein were boiled for 5 min in Laemmli buffer with 5% 2-mercaptoethanol and analyzed by SDS-PAGE (28). Before electrophoresis samples prepared from reverse phase HPLC fractions were dried, suspended, and boiled in Laemmli buffer with 2-mercaptoethanol, and if Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009 cDNA of a Rat Morphine Affinity Column Binding Protein 1375 necessary, the pH was neutralized with Tris base. Gels were stained with silver by the method of Wray et al. (29). h, washed in 2 x SSC-0.1% SDS at 55 C, and autoradiographed with an intensifying screen at - 7 0 C. Reverse Phase HPLC of the /3AM Affinity Column Eluates Computer Analyses Dialyzed affinity column eluates were adjusted to 25% acetonitrile, 0.1% TFA, and injected onto the C4 column at room temperature in 25% acetonitrile-0.1 % TFA (solvent A). A linear gradient was developed to 95% acetonitrile-0.1 % TFA (solvent B) over 60 min at a flow rate of 1 ml/min. Absorbance was monitored at 214 nm, and individual peaks were collected in polypropylene tubes, capped, and stored at - 2 0 C. A Digital MiniVax computer was used to store the GenBank (release 62.0), EMBL (release 21.0), and PIR (release 21.0) databases. The complete nucleotide (including 5' and 3' untranslated regions) and deduced amino acid sequences were used to query these databases using Intelligentics FastDB and Wisconsin GCG FastA algorithms. Protein secondary structure and hydrophobicity were evaluated by Intelligenetics software based on the methods of Kyte and Doolittle (24) and Chou and Fassman (26). Trypsin Digestion and Partial p23k Amino Acid Sequence Determination Approximately 20 ^g HPLC-purified p23k were dried and brought up in 100 ^l 0.1 M NH4HCO3 (pH 8.3)-1 mM CaCI2. Trypsin in water was added at a protein to enzyme ratio of 50:1 (wt/wt), and digestion was initiated at 37 C. After 12 h of incubation, additional trypsin was added, and digestion was continued for a total of 24 h. A trypsin blank sample was prepared and injected onto the C4 column before the p23k digest was fractionated. Peptides were eluted in a linear gradient of acetonitrile (0-50%) developed over 120 min in 0.1% TFA at a flow rate of 1 ml/min. Absorbance was monitored at 214 nm, and each peak, unique to p23k, was collected as a separate fraction. Tryptic peptides were subjected to Nterminal gas phase microsequence analysis on a model 470A Applied Biosystems protein sequencer coupled to a 120A PTH analyzer and a model 7000A SICA chromatogram processor. Rat Brain cDNA Library Preparation and Screening Poly(A)+ mRNA was prepared by standard procedures (30) from rat brain after removal of the cerebellum. The cDNA was synthesized from 5 fig poly(A)+ mRNA according to the method of Gubler and Hoffman (31), followed by the addition of EcoRI adaptors; the cDNA library was constructed by ligating the cDNA to EcoRI-digested Xgt10 arms (Stratagene), followed by in vitro packaging using the Gigapak Plus packaging system (Stratagene). Approximately 600,000 phages were plated on c600 hfl" at a density of 50,000 plaques/plate, and nylon Colony/Plaque Screen filters were used to pull replicas of the library. Three oligonucleotide probes, based on partial amino acid sequence data, were synthesized on a model 388 Applied Biosystems automated DNA synthesizer. The probes were radioactively labeled by T4 polynucleotide kinase (30), and the filters were hybridized and washed according to the manufacturer's recommendations. Plaques that hybridized in duplicate with all three probes were selected and purified. p23R1-1.1 Restriction Site Mapping, Subcloning, and Sequencing Phage cDNA was prepared from liquid lysates (30), and the longest cDNA was subcloned into EcoRI-cut pUC18 to generate p23R1 - 1 . 1 . DNA sequence was obtained by the dideoxy chain termination method (32). The entire sequence of the coding region was confirmed by sequencing both strands of DNA. Northern Blot Analysis Poly(A)+ mRNA was prepared from rat brain minus the cerebellum, cerebellum, and liver and from the 5-opioid receptorbearing cell line NG108-15. Approximately 10 ^g of each were run on a 0.8% agarose-glyoxal gel (30) and transferred to Nytran (Schleicher and Scheull, Keene, NH). The blot was probed with nick-translated p23R1-1.1 insert in 5 x SSC, 50% formamide, 1 % SDS, and 5 x Denhardfs at 42 C for 72 Note Added in Proof A computer search conducted on a 1990 data bank (NBRF Release 23) has led to the identification of the 23 kDa protein as being a cytosolic phosphatidylethanolamine binding protein (Schoentgen F, et al. 1987 Eur Biochem 166:333-338. This finding can be explained by considering that: 1) the morphinan matrix might have enough structural or chemical features to attract phosphatidyl-binding molecules; or 2) there might exist interactions (possibly through lipid moieties) between the opioid binding proteins and the phosphatidyl-binding proteins which lead to the coelution of the latter. Acknowledgments We would like to thank Charles Jimenez for conducting the amino acid analysis of p23k; David Stein for synthesizing the oligonucleotides; Dr. E. Weber for his advice during the protein purification and peptide sequencing; and Dr. O. Humberto Viveros for critical discussions of the manuscript. We would also like to thank Julie Tasnady for typing the manuscript, and Nancy Kurkinen and June Shiigi for preparing the art work. Received April 6, 1990. Revision received June 14, 1990. Accepted June 18,1990. Address requests for reprints to: Dr. Olivier Civelli, Vollum Institute for Advanced Biomedical Research, L-474, Oregon Health Science University, 3181 SW Sam Jackson Park Road, Portland, Oregon 97201. This work was supported by a research grant from the NIDDK(DK-3731;toO.C). * Holder of a fellowship from the NIH. REFERENCES 1. Lin MK, Simon EJ 1978 Phospholase A inhibition of opiate receptor binding can be reversed by albumin. Nature 271:383-384 2. Cho TM, Ge BL, Yamamoto C, Smith AP, Loh HH 1983 Isolation of opiate binding components by affinity chromatography and reconstitution of binding activity. Proc Natl Acad Sci USA 80:5176-5180 3. Pasternak AW, Snyder SM 1974 Opiate receptor binding: enzymatic treatments that discriminate between agonist and antagonist interactions. Mol Pharmacol 11:478-484 4. Howells RD, Gioannini TL, Hiller JM, Simon EJ 1982 Solubilation and characterization of active opiate binding sites from mammalian brain. J Pharmacol Exp Ther 222:629-634 5. Gioannini T, Foucaud B, Hiller JM, Hatten ME, Simon EJ Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009 MOL ENDO-1990 1376 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 1982 Lectin biology of solubilized opiate receptors: evidence for their glycoprotein nature. Biochem Biophysic Res Commun 105:1128-1134 Simon EJ, Hiller JM, Edelman I 1975 Solubilization of a stereospecific opiate-macromolecular complesx from rat brain. Science 190:389-390 Bidlack JM, Abood LG, Osei-Gyimah P, Archer S 1981 Purification of the opiate receptor from rat brain. Proc Natl Acad Sci USA 78:646-639 Zukin RS, Kream RM 1979 Chemical crosslinking of a solubilized enkephalin macromolecular complex. Proc Natl Acad Sci USA 76:1593-1597 Chow T, Zukin S 1983 Solubilization and preliminary characterization of mu and kappa opiate receptor subtypes from rat brain. Mol Pharmacol 24:203-212 Lai FA, Newman EL, Peers E, Barnard EA 1984 Sizes of opioid receptor types in rat brain membranes. Eur J Pharmacol 103:349-354 Demilou-Mason CD, Barnard EA 1986 Characterization of opioid receptor subtypes in solution. J Neurochem 46:1129-1136 Hammonds Jr RG, Nicolas P, Li CH 1982 Characterization of /3h-endorphin binding protein (receptor) from rat brain membranes. Proc Natl Acad Sci USA 79:6494-6496 Gioannini TL, Howard AD, Hiller JM, Simon EJ 1985 Purification of an active opioid-binding protein from bovine striatum. J Biol Chem 260:15117-15121 Cho TM, Hasegawa J-l, Ge BL, Loh HH 1986 Purification to apparent homogeneity of a M-type opioid receptor from rat brain. Proc Natl Acad Sci USA 83:4138-4142 Schofield PR, McFarland KC, Hayflick JS, Wilcox JN, Cho TM, Roy S, Lee NM, Loh HH, Seeburg PH 1989 Molecular characterization of a new immunoglobulin superfamily protein with potential roles in opioid binding and cell contact. EMBO J 8:489-495 Civelli O, Machida CM, Bunzow J, Albert P, Hanneman E, Salon J, Bidlack J, Grandy DK 1987 The next frontier in the molecular biology of the opioid system-the opioid receptor. Mol Neurobiol 1:373-391 Deleted in proof. Howard AD, de La Baume S, Gioannini TL, Hiller JM, Simon EJ 1985 Covalent labeling of opioid receptors with Vol 4 No. 9 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. radioiodinated human 0-endorphin. J Biol Chem 260:10833-10839 Howard AD, Same Y, Gioannini TL, Hiller JM, Simon EJ 1986 Identification of distinct binding site subunits of n and 5 opioid receptors. Biochemistry 25:357-360 Helmeste DM, Hammonds Jr RG, Li CH 1986 Preparation of [125l-TYR27,Leu5]/3h-endorphine and its use for crosslinking of opioid binding sites in human striatum and NG108-15 neuroblastoma-glioma cells. Proc Natl Acad Sci USA 83:4622-4625 Kozak M 1986 Point mutations define a sequence flanking the aug initiation codon that modulates translation by eukaryotic ribosomes. Cell 44:283-292 Kishimoto A, Nishiyama K, Nakanishi H, Uratsuji Y, Nomura H, Takeyama Y, Nishizuka Y 1985 Studies on the phosphorylation of myelin basic protein by protein kinase C and adenosine 3'-5' monophosphate-dependent protein kinase. J Biol Chem 260:12492-12495 Turner AJ 1986 Processing and metabolism of neuropeptides. Essays Biochem 22:69-119 Kyte J, Doolittle RF 1982 A simple method for displaying the hydropathic character of a protein. J Mol Biochem 157:105-132 Von Heinje G 1983 Patterns of an acid near signal-sequence change sites. Eur J Biochem 133:17-21 Chou PY, Fassman GD 1978 Prediction of the 20 structure of proteins from their amino acid system. Adv Enzyme 47:45-148 Bradford MM 1976 A rapid and sensitive method for the quantitation of micogram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248254 Laemmli UK, Farve M 1973 Maturation of the head of bacteriophage T4. J Mol Biol 80:575-599 Wray W, Boulikas T, Wray VP, Hancock R 1981 Silver staining of proteins in polyacrylamide gels. Anal Biochem 118:197-203 Maniatis T, Fritsch EF, Sambrook J 1982 Molecular Cloning-A Laboratory Manuel. Cold Spring Harbor Laboratory, Cold Spring Harbor Gubler U, Hoffman BJ 1983 A simple and very efficient method for generating cDNA libraries. Gene 25:263-269 Sanger F, Nichlen S, Coulson AR 1977 DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 74:5463-5467 Downloaded from mend.endojournals.org at Cornell University Library on December 2, 2009