Download Purification, Cloning, and Tissue Distribution of a 23

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