|
|
|
|||||||
|
|||||||||
(Received for publication, October 14, 1994; and in revised form, January
17, 1995) From the
We recently characterized (Moebius, F. F., Burrows, G. G.,
Striessnig, J., and Glossmann H.(1993) Mol. Pharmacol. 43,
139-144) and purified (Moebius, F. F., Hanner, M., Knaus, H. G.,
Weber, F., Striessnig, J., and Glossmann, H.(1994) J. Biol. Chem. 269, 29314-29320) a binding protein for the phenylalkylamine
Ca The phenylalkylamine Ca Here we report the cloning and
heterologous expression of human and guinea pig cDNAs coding for
proteins with the pharmacological properties of EBP. We also present
the tissue distribution of EBP and EBP-mRNA and discuss the structural
features which EBP shares with drug transporters.
Microsomal membranes from different tissues (see legend to Fig. 2B) were prepared as described
previously(3) . Microsomal protein was solubilized in sample
buffer containing 4% (w/v) SDS and 10 mM dithiothreitol and
separated on 13% (w/v) polyacrylamide gels. After electrophoretic
transfer to a polyvinylidene difluoride membrane (Immobilon P,
Millipore), immunostaining with affinity purified
anti-EBP
Figure 2:
Tissue distribution of EBP in guinea pig. A, Northern blot hybridization with a guinea pig EBP cDNA
probe. 15 µg of total RNA/lane were separated on a 1.2% (w/v)
agarose gel and transferred to a nylon membrane (Boehringer Mannheim).
Hybridization with a DIG-labeled guinea pig antisense cDNA probe was
carried out as described under ``Experimental Procedures''
with an alkaline phosphatase-conjugated second antibody and
chemoluminescence detection. EBP was encoded in guinea pig and human
liver by a 1.3-kilobase mRNA. The human EBP hybridized to a lesser
extent due to the sequence difference to the guinea pig cDNA probe. B, Western blot analysis with affinity-purified
anti-EBP
Figure 1:
A, comparison of the deduced amino acid
sequences of guinea pig (G.P.) and human (H.S.) EBP.
cDNAs were isolated from guinea pig and human liver cDNA libraries. The
four hydrophobic putative transmembrane segments are shaded.
The consensus sequence for phosphorylation by cAMP-dependent
proteinkinase (
Nonspecific [
Figure 3:
Characterization of human and guinea pig
EBP expressed in S. cerevisiae. A,
[
Saturation analysis with
[ The
pharmacological profile of the expressed guinea pig and human EBP
revealed no major difference (see Table 1) to the EBP in guinea
pig liver. Minor decreases in affinity were observed for haloperidol,
NaCl (human and guinea pig EBP), and (+)-verapamil (human EBP).
Accordingly, [ We previously reported
that disuccinimidyl suberate, SANPAH, and glutaraldehyde led to
dimerization of the liver EBP(3) . Dimerization after
incubation with the above cross-linkers was confirmed for the expressed
guinea pig (see Fig. 3E) and human (results not shown)
EBP. EBP is a high affinity binding protein for the antiischemic
phenylalkylamine Ca
The functional expression of the
cloned EBP-cDNAs in S. cerevisiae demonstrated that they
encode proteins which are able to form the
[
The localization of EBP in the
endoplasmic reticulum membrane hampers in vivo demonstration
of drug transport due to intracellular drug compartimentalization. In
contrast to the P-glycoprotein, the EBP amino acid sequence contains no
ATP-binding cassette. Although EBP carries a sodium ion-binding site (1) as in sodium-dependent
transporters(26, 27, 28) , the driving force
of a potential drug transport by EBP is unknown making in vitro studies difficult. The molecular cloning of a high affinity
binding protein for phenylalkylamine Ca The nucleotide
sequence(s) reported in this paper has been submitted to the
GenBank(TM)/EMBL Data Bank with accession number(s) Z37985 [GenBank]and Z37986[GenBank].
Volume 270,
Number 13,
Issue of March 31, 1995 pp. 7551-7557
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
Antagonist Binding Protein
MOLECULAR CLONING, TISSUE DISTRIBUTION, AND HETEROLOGOUS EXPRESSION (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
antagonist emopamil. The emopamil-binding protein
(EBP) acts as a high affinity acceptor for several antiischemic drugs
and thus represents a potential common molecular target for
antiischemic drug action. Degenerate oligonucleotides were synthesized
according to the N-terminal amino acid sequence of purified EBP and
used to amplify a guinea pig cDNA with reverse transcriptase-polymerase
chain reaction and to clone full-length cDNAs from guinea pig and human
liver cDNA libraries. The cDNAs coded for 229 (guinea pig) and 230
(human) amino acid 27-kDa polypeptides without significant sequence
homology with any known protein. However, EBP shared structural
features with pro- and eukaryotic drug transport proteins. The amino
acid identity between human and guinea pig EBP was 73%. Hydrophobicity
plots predicted four transmembrane segments. The C terminus contained a
lysine-rich consensus sequence for the retrieval of type I integral
membrane proteins to the endoplasmic reticulum. The heterologous
expression of human and guinea pig EBP in Saccharomyces cerevisiae demonstrated that the expression of EBP alone is sufficient to
form high affinity drug- and cation-binding domains identical to the
[
H]-emopamil-binding site of guinea pig liver.
Northern and Western blot analysis revealed high abundance of EBP in
guinea pig epithelial tissues as liver, bowel, adrenal gland, testis,
ovary, and uterus and low densities in brain, cerebellum, skeletal
muscle, and heart. EBP is suggested to be the first structurally
characterized member of a family of high affinity microsomal drug
acceptor proteins carrying so called 
-binding sites.
antagonist emopamil
labels a high affinity binding protein (emopamil-binding protein, EBP) (
)for a variety of structurally different compounds in
guinea pig liver(1, 2) . Some of these drugs such as
emopamil, ifenprodil, opipramol, trifluoperazine, and chlorpromazine
exert antiischemic effects in animal models of stroke, but their
neuroprotective action is not fully understood at the molecular level.
Since the possibility exists that EBP is involved in ischemia-related
cellular events biochemical studies were undertaken to investigate the
physiological function of EBP. These studies revealed that the EBP is
an integral membrane protein of the endoplasmic reticulum that migrated
with a relative molecular mass of 22.5 kDa in SDS-PAGE(2) .
Ferguson analysis revealed a molecular mass of 27.2 kDa (3) .
Hydrodynamic studies and the extensive characterization of its
drug-binding properties (2, 4) demonstrated striking
similarities with other microsomal drug-binding proteins, so called
receptors (see (5) for review). This led to the proposal
that EBP and sigma receptors are members of a superfamily of high
affinity drug-binding proteins in the endoplasmic reticulum of
different tissues(3) .
Materials
(-)[H]Emopamil
(67 Ci/mmol), (±)[H]emopamil (49 Ci/mmol),
(-)-[H]azidopamil (87 Ci/mmol), and the
unlabeled phenylalkylamines were kindly provided by Knoll A.G.
(Ludwigshafen, Germany). Sigma ligands were a gift of Dr. Traber
(Tropon, Cologne, Germany). Other chemicals were obtained from the
following sources: opipramol, Ciba-Geigy (Vienna, Austria); Bradford
protein reagent, electrophoresis reagents and molecular weight markers,
Bio-Rad; restriction enzymes and polymerases, Promega (Vienna,
Austria); oligonucleotide synthesis reagents, Millipore (Vienna,
Austria); all other chemicals, Sigma (Vienna, Austria).PCR Amplification of EBP cDNA Fragments
The
N-terminal sequence of purified guinea pig EBP was obtained by Edman
degradation (peptide A,(3) ). Degenerate oligonucleotides
encoding amino acid residues PLHPYW
(5`-CC(CGAT)TT(CGAT)CA(TC)CC(CGAT)TA(TC)TGG-3` (sense primer)) and
DHFVPN (5`-TC(GA)TT(CGAT)GG(CGAT)AC(GA)AA(GA)TG(GA)TC-3` (antisense
primer)) were synthesized and used in the polymerase chain reaction
(PCR) (Bio-med, Thermocycler 60). First strand cDNA was synthesized
from guinea pig total RNA with an oligo(dT) primer (Pharmacia Biotech.
Inc., first strand cDNA synthesis kit). PCR was performed with 35
cycles at low stringency (1 min at 94 °C, 0.5 min at 37 °C, 1.5
min at 72 °C). The amplified products were separated on a 20% (w/v)
polyacrylamide gel. Products with a size of 50- 60 bp were eluted into
destilled water (12 h, 37 °C), cloned into pCR
II
vector (Invitrogen) and sequenced. One clone with a 56-bp insert coded
for the amino acid sequence PLHPYWPRHLRLDHFVPN (residues 7-24) of
peptide A. The cDNA for EBP was amplified (1 min at 94 °C, 0.5 min
at 56 °C, 1.5 min at 72 °C) with an oligonucleotide according
to the amino acid sequence HLRLDHF (5`-ACCTGCGGCTGGA(TC)CA(CT)TT-3`
(sense primer)) in conjunction with the oligo(dT) primer (3`-antisense
primer from the first strand cDNA synthesis kit, Pharmacia Biotech
Inc.). After cloning into the pCRII vector, one clone (C7,
940 bp) was sequenced with the dideoxy termination method (6) using standard and specific sequencing primers. This clone
coded for a protein containing all partial peptide sequences previously
determined by Edman degradation (3) .Library Construction and Screening
An
oligo(dT)-primed cDNA library was constructed in phage 
gt10 using
size-selected (>500 bp) poly(A)
RNA derived cDNA
from human and guinea pig liver. Recombinant phages were plated on Escherichia coli C600 hfl at a density of 60,000
plaque-forming units/135-mm plate. After plaque formation, plaques from
five plates were transferred to Biodyne B membrane (Pall) and
hybridized with a C7-specific antisense cDNA probe labeled with
digoxigenin 11-dUTP (DIG) following a single strand-labeling PCR
protocol(7) . From 30 to 40 hybridizing plaques/plate 10
plaques from each library were isolated. The cDNA inserts were excised
with EcoRI and subcloned into pBluescript IISK+
(Stratagene). The longest cDNAs were sequenced on both strands as
described above. DNA and protein sequence analysis were performed with
the GCG sequence analysis software package (Genetics Computer Group
Inc., Madison, WI).RNA Preparation and Northern Blot Analysis
Total
RNA from liver, kidney, adrenal gland, lung, brain, cerebellum, spleen,
heart, and skeletal muscle were prepared as described(8) . 15
µg of each RNA were separated on a 1.2% (w/v) agarose gel in 30%
(v/v) formaldehyde and blotted. RNAs were probed with a DIG-labeled
specific antisense probe (guinea pig cDNA, nucleotides 227-709).
After hybridization (42 °C, 50% (v/v), formamide) blots were washed
at low stringency (0.2 
SSC (1 SSC is 0.15 M NaCl, 0.01 M sodium citrate, pH 7.0), 0.1% (w/v) SDS at
22 °C), and bands were detected according to the DIG protocol using
chemolumiscence detection (Boehringer Mannheim).Expression of EBP in Saccharomyces
cerevisiae
Guinea pig and human EBP cDNAs were subcloned
into the yeast episomal plasmid YEp351ADC1(9) . Truncated cDNAs
carrying a 5`-HindIII restriction site and AAA triplet before
the initial ATG (10) (human and guinea pig EBP) and a
3`-NotI restriction site behind the stop codon (human EBP)
were generated with PCR. PCR products were cloned into pBluescript
IISK+ and sequenced before subloning into YEp351ADC1.
Transformation of S. cerevisiae JB811 was performed as
described(11) . Cells were harvested at an OD
of
1.2 and lysed with glass beads in 50 mM Tris-HCl (pH 7.4, 37
°C), 1 mM EDTA, 0.1 mM phenylmethylsulfonyl
fluoride. The lysate was spun for 5 min at 500 g (4
°C), and the supernatant was pelleted for 45 min at 100,000 g (4 °C). This microsomal pellet was washed with 0.5 M KCl, 0.15 M Tris HCl (pH 8, 4 °C), centrifuged at
100,000 g (4 °C), resuspended in 5% (v/v)
glycerol, 20 mM Tris HCl (pH 7.4, 37 °C) at a protein
concentration of 3-6 mg/ml, and frozen in liquid N
.Immunological Techniques
A polyclonal antiserum
was raised in New Zealand White rabbits against a synthetic peptide
corresponding to the C terminus of human EBP with an additional
N-terminal lysine residue (peptide EBP
). The
production of a sequence directed antibody against the N terminus of
guinea pig EBP (anti EBP
) was described
previously(3) . Peptide synthesis, coupling to bovine serum
albumin, immunization, enzyme-linked immunosorbent assay, and affinity
purification of antisera were performed as described(12) . and anti-EBP was
carried out as described previously (3) .
antibody. Microsomal membranes were
prepared as described(2) , and 30 µg of microsomal protein
were separated on 13% (w/v) SDS-polyacrylamide gels. Immunoblotting was
performed as described previously(3) . The arrow indicates the migration of EBP. The
(±)[H]emopamil binding activity was
measured at a ligand concentration of 1.89 nM in parallel.
Nonspecific binding was defined as the binding in the presence of 30
µM ZnCl to selectively inhibit
(±)[H]emopamil binding to EBP. The binding
activity expressed as percent binding relative to guinea pig liver
microsomes (1.95 ± 0.17 pmol of
(±)[H]emopamil bound/mg of protein (n = 2)) is given in parenthesis (means ±
standard deviations, n = 2). The following
abbreviations were used: Li, guinea pig liver (100%); Lh, human liver (not determined); Sp, spleen (5.1
± 0.4); Lu, lung (8.2 ± 0.3); Br, brain
(2.4 ± 0.8); Sm, skeletal muscle (1.2 ± 0.1); Ki, kidney (4.5 ± 0.9); Hm, heart muscle (1.9
± 0.9); Ce, cerebellum (2.1 ± 0.4); Ad,
adrenal gland (15.1 ± 0.1); Oe, oesophagus (5.3
± 1.8); St, stomach (0 ± 0); Il, ileum
(10.8 ± 2.2); Co, colon (20.0 ± 0.4); Pc, pancreas (0 ± 0); Sk, skin (3.6 ±
0.1); Eye (5.0 ± 0.3); Te, testis and epididymidis (9.7
± 1.4); Ov, ovary (8.0 ± 0.2); Ut,
uterus (4.8 ± 2.1).
Binding Assays, Photoaffinity Labeling, and
SDS-PAGE
These were performed according to standard
procedures(2, 3) . Additional experimental detail is
given in the figure legends. Protein concentrations were determined
according to Bradford(13) , using bovine serum albumin as a
standard. Binding parameters (IC
, slope factor (n
), K
, and B
)
were calculated by non-linear curve fitting to the general
dose-response equation (14) (inhibition data) or a rectangular
hyberbola (saturation data). Data are given as means ± standard
deviations.Hep G2 Cell Culture
Hep G2 cells were grown in
Hanks' modified Dulbecco's modified Eagle's medium in
4.5% (w/v) glucose. Cells were trypsinized, collected, and microsomal
membranes were prepared as described above.
Cloning of EBP cDNAs
guinea pig and human liver
gt10 libraries were screened with the 940-bp EBP cDNA obtained by
reverse transcriptase-PCR (see ``Experimental Procedures'').
From each library 10 hybridizing clones were isolated. After
characterization by restriction enzyme mapping the longest clones
(1005-bp clone GP5 for guinea pig EBP and 1073-bp clone HS3
for human EBP) were subjected to sequencing. The deduced amino acid
sequences of human and guinea pig EBP are shown in Fig. 1A.
), the polylysine cluster for retrieval into the
endoplasmic reticulum (underlined,(17, 18) ),
the sequences of the peptides (EBP and
EBP) used for antibody raising (bold),
and the amino acid sequences previously determined by Edman degradation
of tryptic peptides A-E of the purified EBP (overlined,(3) ) are indicated. The numbers of the
amino acid residues are given on the right. B, hydrophobicity
analysis of guinea pig EBP. Hydropathy plotting of the EBP amino acid
sequence was performed according to Kyte and Doolittle (19) with an averaging window size of 19 plotted at one-residue
intervals. On the ordinate hydrophobicity and hydrophilicity
are indicated by positive and negative numbers, respectively.
TM1-4 designate the putative transmembrane segments. C,
proposed topology model of guinea pig EBP. Based on the endoplasmic
reticulum retrieval sequence (17, 18) and the above
hydrophobicity analysis of the amino acid sequence the protein is
postulated to have cytoplasmic N and C termini, four transmembrane
segments, and two connecting loops in the lumen of the endoplasmic
reticulum (ER) (
, hydrophobic residues; &cjs2125;,
positively charged residues; , negatively charged residues;
, others).
Sequence Analysis
The open reading frame of guinea
pig and human cDNAs coded for 229 and 230 amino acid residues
corresponding to a molecular mass of 26.683 and 26.356 Da,
respectively. The polyadenylation signal AATAAA of the human EBP cDNA
occurred 239-244 nucleotides downstream from the stop codon of
the putative EBP and 13 nucleotides upstream from a putative poly(A)
tail (not shown). Although the 5` non-translated regions of both cDNAs
(nucleotides 1-114 of human EBP) were 64% homologous (results not
shown) the first methionine codon was flanked by a consensus sequence
for the initiation of translation (15) indicating that it was
indeed the start codon. The N-terminal amino acid sequence of the
purified EBP determined by Edman degradation only lacked the initial
methionine residue(3) . The identity and similarity (determined
as described in (16) ) of the amino acid sequences between
human and guinea pig EBP were 73 and 85%, respectively (see Fig. 1A). Sequence comparison in protein and DNA
databases (Swiss Prot, EMBL/GenBank) showed no significant homology
(>20%) with known sequences. Both EBPs contained potential
phosphorylation sites for protein kinases A (see Fig. 1A) and C (not shown). The C termini of both
proteins were heterologous. They contained a polylysine motif (KVMKSKGK
in guinea pig and KAKSKKN in human) known to mediate the retrieval of
type I integral membrane proteins into the endoplasmic
reticulum(17, 18) . Hydropathy plots according to Kyte
and Doolittle (19) computed with a window of 19 amino acid
residues predicted four transmembrane segments (TMS) (see Fig. 1B). The connecting loop between TMS3 and TMS4 was
also highly hydrophobic (see Fig. 1B). The TMS2 and
TMS3 contained two glutamate residues conserved in human and guinea pig
EBP whereas the TMS1 and TMS4 contained no charged residues (see Fig. 1A). All cysteine residues were localized in the
TMS2 and 3 (see Fig. 1A). A high content of aromatic
amino acid residues in the transmembrane segments was determined as
described in (20) and was 28 and 23% for guinea pig and human
EBP, respectively. The topology model of EBP (Fig. 1C)
predicts that most of the protein is buried in the lipid bilayer. From
the endoplasmic reticulum retrieval sequence for type I integral
membrane proteins(17, 18) , we conclude that the N and
C termini face the cytoplasm. In this model the potential
phosphorylation site for cAMP-dependent protein kinase (see Fig. 1, A and C) faces the lumen of the
endoplasmic reticulum. Protein kinase A did not phosphorylate the
detergent-purified EBP (results not shown).Northern and Western Blot Analysis of Tissue Distribution
in Comparison to [
The
tissue distribution of EBP was studied in guinea pig by comparing the
mRNA levels as well as EBP immunoreactivity with the microsomal
[H]Emopamil BindingH]emopamil binding activity. In Northern blots a
1.3-kilobase mRNA was detected in different tissues (Fig. 2A). No other band hybridized under conditions of
low stringency indicating that RNAs with high sequence homology are
either absent or present only at lower abundance. The migration of EBP
immunoreactivity in different tissues was indistinguishable from liver (Fig. 2B). Immunolabeling was specifically blocked by
0.03 µM of synthetic peptide (results not shown). H]emopamil binding was defined
with 30 µM ZnCl which is known (2) to
selectively block [H]emopamil binding to EBP. The
distribution of EBP mRNA and immunoreactivity in different tissues
correlated with the binding activity (see Fig. 2A and B and legend to Fig. 2). The highest EBP densities were
found in epithelial tissues as liver, ileum, and colon. Urogenital
tissues as kidney, adrenal gland, testis, ovary, and uterus contained
also high densities of EBP. Lower densities were present in spleen,
oesophagus, stomach, and eyes. No binding activity was detected in
microsomes from pancreas and stomach perhaps due to proteolysis during
incubation at 22 °C. Very low densities were found in excitable
tissues as brain, cerebellum, heart, and skeletal muscle. Since the
microsomes used for Western blot analysis and binding studies are
mainly derived from the endoplasmic reticulum (2) the
differential distribution of EBP rather reflects differences in EBP
density than differences in the cellular endoplasmic reticulum content. Expression of EBP in S. cerevisiae
To
clarify if EBP alone is able to form the high affinity emopamil binding
site and to investigate if the isolated homologous human cDNA also
forms a high affinity emopamil acceptor, we expressed both cDNAs.
Several mammalian cell lines (e.g. PC12, COS-7 (not shown) and
Hep G2 (see Fig. 3, A and B)) were found
unsuitable for this purpose due to the presence of endogenous
[H]emopamil binding activity and
anti-EBP immunoreactivity. In contrast, the S.
cerevisiae strain JB811 lacked both activities (Fig. 3, A and B) and was therefore used for expression with
the yeast episomal plasmid YEp351ADC1(9, 21) . The
complete cDNA (GP
and HS, see Fig. 3, A and B) did not yield detectable
expression. We therefore deleted the 5`-untranslated region (GP, HS, see Fig. 3, A and B) which resulted in high expression of
[H]emopamil binding activity (Fig. 3A) and anti-EBP immunoreactivity (Fig. 3B). Since comparable mRNA
levels were found for GP and GP as well as
HS and HS (results not shown) this was not due
to differences in transcription or mRNA stability. Instead it could
indicate that the highly homologous 5`-nontranslated region of both
cDNAs represses translation in S. cerevisiae. With both
truncated cDNAs, expression levels similar to the density in guinea pig
liver microsomes were observed as quantitated by
[H]emopamil binding activity (see Fig. 3A) and anti-EBP immunoreactivity (see Fig. 3B). The apparent
molecular mass of the expressed protein in SDS-PAGE was
indistinguishable from the EBP in guinea pig liver microsomes (see Fig. 3B). The identity of the expressed proteins was
further confirmed with a sequence-directed antibody
(anti-EBP) specific for the human EBP (see Fig. 1A). This antibody (not shown) as well as
anti-EBP (see Fig. 3B)
specifically recognized EBP in the human hepatoma cell line Hep G2 and
the expressed human EBP whereas no cross-reactivity with guinea pig EBP
was observed (results not shown).
H]emopamil binding to microsomal membranes from
guinea pig liver (Li), Hep G2 cells (He), and S.
cerevisiae transfected with nonrecombinant vector (Co) or
vector with human (HS, HS),
and guinea pig (GP, GP) EBP
cDNAs. Microsomal membranes were prepared as described under
``Experimental
Procedures.''(-)-[H]emopamil binding
was performed at a ligand concentration of 1.1-1.7 nM.
The results shown are the mean of 4-11 experiments. EBP cDNAs
were expressed in S. cerevisiae strain JB811 (obtained from
Kim Nasmyth, Vienna) using the yeast episomal plasmid YEp351 (21) with an ADC1 cassette(9) . Yeast cells were lysed
with glass beads (see ``Experimental Procedures''). HS and GP are the entire clone inserts HS3 and
GP5, respectively; in GP the 5`-noncoding region, and
in HS the 5` and the 3`-noncoding region were removed by
PCR (see ``Experimental Procedures''). B,
anti-EBP immunoblotting. 7 µg of microsomal
membrane protein (He, HepG 2, 50 µg of protein)
were separated on a 13% (w/v) SDS-polyacrylamide gel and analyzed by
immunostaining with anti-EBP. The higher molecular
mass observed for EBP in Hep G2 cells could be due to higher protein
load. C, saturation analysis of
(-)-[H]emopamil binding to human EBP
(HS) expressed in S. cerevisiae. Saturation
analysis was performed by decreasing the specific activity of
(-)-[H]emopamil by dilution with unlabeled
(-)-emopamil at a protein concentration of 19 µg/ml. A
B of 75 pmol/mg protein and a K
of 76 nM were obtained. The inset shows
the Scatchard transformation of these data. D,
[H]azidopamil photoaffinity labeling of
microsomal protein from vector-transfected yeast cells (lane
1), guinea pig liver microsomes (lane 7), and yeast cells
transfected with the human cDNA (HS).
Photoaffinity labeling was carried out in the absence (lanes
1, 2, 6, and 7) or presence of 0.15
µM of(-)-emopamil (lane 3), opipramol (lane 4), and ifenprodil (lane 5) by incubation of 21
nM [H]azidopamil with 0.5 mg microsomal
protein/ml. 50 µg of microsomal protein were separated on a 15%
(w/v) polyacryamide gel. For fluorography the gel was equilibrated in
Amplify
(Amersham), dried, and exposed to Kodak X-Omat AR5
(10-day exposure).The migration of EBP is indicated by the arrow. E, chemical cross-linking. Microsomal
membranes from yeast cells expressing guinea pig EBP (GP) were solubilized in in 1% (w/v) digitonin,
150 mM NaCl, 20 mM K
H
PO4, pH
7.8, at a protein concentration of 4 mg/ml. After dilution in 150
mM NaCl, 20 mM KHPO4, pH 7.8,
to a final digitonin concentration of 0.2% (w/v) samples were incubated
2 h at 4 °C in the absence (lane 1) or presence of
chemical cross-linkers (20 mM disuccinimidyl suberate, lane 2; 20 mM SANPAH, lane 3; 10 mM glutaraldehyde, lane 4). After photolysis of SANPAH (3
min, 3 cm distance) with a Sylvania GTE germicide lamp 10 µg of
protein were separated by SDS-PAGE and immunoblotted as described under
``Experimental Procedures.''
H]emopamil (see Fig. 3C)
revealed similar densities of EBP in microsomes from S. cerevisiae cells expressing EBP (HS, B 70 ±
4 (n = 3) pmol/mg membrane protein; GP,
B 29 ± 7 (n = 3) pmol/mg membrane
protein) and guinea pig liver (B 35 pmol/mg membrane
protein,(1) ). The dissociation constants of human (K 15 ± 1 (n = 3)
nM) and guinea pig EBP (K 10 ± 3 (n = 3) nM) expressed in S. cerevisiae were similar to the value measured in guinea pig liver (K 13 nM,(1) ).
H]azidopamil photoaffinity
labeling of HS (see Fig. 3D) reflected the
properties of [H]emopamil binding, i.e. complete block of labeling in the presence of 0.15 µM of emopamil, opipramol, and ifenprodil. No labeling of a protein
with similar apparent molecular mass was observed in cells transfected
with the expression vector without EBP cDNA.
antagonist
[H]emopamil and the photoaffinity label
[H]azidopamil. A variety of compounds with
antiischemic effects in animal models of cerebral ischemia inhibited
[H]emopamil binding to EBP with high affinity. It
could therefore represent a common molecular target of antiischemic
drug action(2) . This prompted further biochemical studies to
eventually clarify its physiological function. The close
pharmacological and biochemical relationship of EBP with so-called
receptors (5) led to the proposal that EBP and
receptors are members of a superfamily of high affinity microsomal
drug-binding proteins(3) .EBP Forms the High Affinity Emopamil-binding Domain
The
purification of EBP from guinea pig liver allowed us to determine its
N-terminal sequence by Edman degradation(3) . This enabled us
to clone its cDNA from guinea pig and human liver cDNA libraries. The
cDNAs coded for proteins with a calculated molecular mass of 26.7 and
26.4 kDa, respectively. This molecular mass of approximately 27 kDa was
in agreement with the relative molecular mass of 27.2 kDa determined by
Ferguson analysis(3) .H]emopamil-binding site. Chemical cross-linking
of the guinea pig and human EBP expressed in yeast suggested that the
protein forms a functional homodimer.EBP Carries an Endoplasmic Reticulum Retrieval
Sequence
Our previous finding of a subcellular localization of
EBP in the endoplasmic reticulum (2) was substantiated by the
presence of a lysine-rich consensus sequence in the C terminus of human
and guinea pig EBP. This motif has been demonstrated in other proteins
to be a signal sequence for the retrieval of type I integral membrane
proteins into the endoplasmic reticulum(17, 18) .
Therefore the N and C termini of EBP must face the cytoplasm (Fig. 1C).EBP Resembles Drug Transporters
Since the primary
structure of EBP lacked similarity with available protein sequences no
conclusion about its biological function could be drawn. However, EBP
shares structural features with bacterial and eukaryontic drug
transporting proteins like the bacterial drug pump smr(22) . Similar to smr EBP has four
putative transmembrane segments and contains two conserved glutamate
residues in TMS2 and 3. In smr a glutamate residue seems to be
involved in the transport of cationic amphiphilics(23) .
Another prominent feature of EBP is its high content of aromatic amino
acid residues (>23%) in its transmembrane segments. A similarly high
content (18.4%) is characteristic (20) for the P-glycoprotein,
a drug pump of the plasma membrane. These aromatic amino acid residues
have been suggested to be involved in the drug transport by the
P-glycoprotein(20) . Like P-glycoprotein EBP has a broad drug
specificity preferring drugs containing positively charged amines and
lipophilic side chains as found in compounds from different
pharmacological classes(1, 2, 24) . Although
the tissue distribution of EBP and the P-glycoprotein are not identical
both are mainly expressed in epithelial tissues (Fig. 2, A and B(25) ). antagonists
demonstrates a unique primary structure distinct from the drug binding
subunit of L-type Ca channels and
rules out a relationship with known drug metabolizing enzymes. It was
previously demonstrated that EBP shares many pharmacological and
biochemical features with so-called
receptors(2, 4) . We therefore proposed (3) the existence of a superfamily of microsomal high affinity
drug acceptor proteins comprising binding polypeptides for
[H]pentazocine,
[H]ditolylguanidine, and
[H]SKF10047 and a high affinity
[H]opipramol-binding site(29) . EBP is
the first structurally characterized member of this family. Of
considerable interest is the molecular organization of the promiscous
drug-binding domain of these proteins in comparison to other
drug-binding domains with a narrower drug specificity, e.g. the phenylalkylamine binding domain of L-type Ca channels(30) . This can be investigated now by
site-directed mutagenesis employing the efficient expression system
presented here. The comparison of human and guinea pig cDNAs will allow
the cloning of EBP and related proteins from other species to identify
evolutionary conserved motifs important for the function of these
proteins. Furthermore, our work paves the way for the cloning and
disruption of the mouse EBP gene to generate EBP-deficient animals
which could give insight into the pharmacological significance and
biological function of this new family of microsomal drug acceptor
proteins.
), maximal density of binding
sites; KHPO
,
KHPO/KHPO; PAGE,
polyacrylamide gel electrophoresis; PCR, polymerase chain reaction;
SANPAH, N-succinimidyl-6(4`-azido-2`-nitrophenylamino)hexanoate; TMS,
transmembrane segment; bp, base pair(s); DIG, digoxigenin 11-dUTP.
We thank Drs. I. Graziadei and W. Vogel for the gift
of Hep G2 cells, Dr. Peter Kaiser for the yeast expression plasmid
YEp351ADC1, M. Froschmayr for helpful advice on protein expression in S. cerevisiae, and M. Holzer for skilled technical assistance.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Gioti, J. M. Pradier, E. Fournier, P. Le Pecheur, C. Giraud, D. Debieu, J. Bach, P. Leroux, and C. Levis A Botrytis cinerea Emopamil Binding Domain Protein, Required for Full Virulence, Belongs to a Eukaryotic Superfamily Which Has Expanded in Euascomycetes Eukaryot. Cell, February 1, 2008; 7(2): 368 - 378. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Kedjouar, P. de Medina, M. Oulad-Abdelghani, B. Payre, S. Silvente-Poirot, G. Favre, J.-C. Faye, and M. Poirot Molecular Characterization of the Microsomal Tamoxifen Binding Site J. Biol. Chem., August 6, 2004; 279(32): 34048 - 34061. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Misawa, T. Horiba, N. Arimura, Y. Hirano, J. Inoue, N. Emoto, H. Shimano, M. Shimizu, and R. Sato Sterol Regulatory Element-binding Protein-2 Interacts with Hepatocyte Nuclear Factor-4 to Enhance Sterol Isomerase Gene Expression in Hepatocytes J. Biol. Chem., September 19, 2003; 278(38): 36176 - 36182. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Vidal, G. Mondesert, S. Galiegue, D. Carriere, P.-H. Dupuy, P. Carayon, T. Combes, E. Bribes, J. Simony-Lafontaine, A. Kramar, et al. Identification and Pharmacological Characterization of SRBP-2: A Novel SR31747A-binding Protein Cancer Res., August 15, 2003; 63(16): 4809 - 4818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. E. Herman Disorders of cholesterol biosynthesis: prototypic metabolic malformation syndromes Hum. Mol. Genet., April 2, 2003; 12(90001): R75 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Souter, J. Topping, M. Pullen, J. Friml, K. Palme, R. Hackett, D. Grierson, and K. Lindsey hydra Mutants of Arabidopsis Are Defective in Sterol Profiles and Auxin and Ethylene Signaling PLANT CELL, May 1, 2002; 14(5): 1017 - 1031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. F. Moebius, R. J. Reiter, K. Bermoser, H. Glossmann, S. Y. Cho, and Y.-K. Paik Pharmacological Analysis of Sterol Delta 8-Delta 7 Isomerase Proteins with [3H]Ifenprodil Mol. Pharmacol., September 1, 1998; 54(3): 591 - 598. [Abstract] [Full Text]
|
|
|
|
 |
|
 B. U. Fitzky, M. Witsch-Baumgartner, M. Erdel, J. N. Lee, Y.-K. Paik, H. Glossmann, G. Utermann, and F. F. Moebius Mutations in the Delta 7-sterol reductase gene in patients with the Smith-Lemli-Opitz syndrome PNAS, July 7, 1998; 95(14): 8181 - 8186. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Paul, S. Silve, N. De Nys, P.-H. Dupuy, C. L.-L. Bouteiller, J. Rosenfeld, P. Ferrara, G. Le Fur, P. Casellas, and G. Loison Both the Immunosuppressant SR31747 and the Antiestrogen Tamoxifen Bind to an Emopamil-Insensitive Site of Mammalian Delta 8-Delta 7 Sterol Isomerase J. Pharmacol. Exp. Ther., June 1, 1998; 285(3): 1296 - 1302. [Abstract] [Full Text]
|
|
|
|
|
|
F. F. Moebius, B. U. Fitzky, J. N. Lee, Y.-K. Paik, and H. Glossmann Molecular cloning and expression of the human Delta 7-sterol reductase PNAS, February 17, 1998; 95(4): 1899 - 1902. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Lenz and J. W. Kleineke Hormone-induced rise in cytosolic Ca2+ in axolotl hepatocytes: properties of the Ca2+ influx channel Am J Physiol Cell Physiol, November 1, 1997; 273(5): C1526 - C1532. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Silve, P. H. Dupuy, C. Labit-Lebouteiller, M. Kaghad, P. Chalon, A. Rahier, M. Taton, J. Lupker, D. Shire, and G. Loison Emopamil-binding Protein, a Mammalian Protein That Binds a Series of Structurally Diverse Neuroprotective Agents, Exhibits Delta 8-Delta 7 Sterol Isomerase Activity in Yeast J. Biol. Chem., September 13, 1996; 271(37): 22434 - 22440. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |