|
|
|
|||||||
|
|||||||||
(Received for publication, November 16, 1994; and in revised form, May 15, 1995 ) From the
The isoquinoline carboxamide photoaffinity probe PK14105, a
ligand with selectivity for mitochondrial benzodiazepine receptors, has
been established to photolabel an 18-kDa protein. When this radioactive
probe is used to photolabel rat mitochondrial preparations, a protein
of 10 kDa, in addition to the 18-kDa protein, is identified following
electrophoretic separation and extended autoradiography. These proteins
are referred to herein as pk10 and pk18, respectively. Both proteins
exhibited the same specificity to a series of ligands used in
competition photolabeling studies and are mutually present at
apparently similar ratios across multiple tissues. Subcellular
fractionation of rat adrenals indicated that pk10 and pk18 comigrated
with the mitochondrial marker enzyme cytochrome c oxidase. In
numerous paradigms examining specificity, photolabeling of pk18
invariably coincided with photolabeling of pk10. In
detergent-solubilized extracts of rat adrenal mitochondria, pk18 and
pk10 coimmunoprecipitated when using antisera raised against pk18.
Furthermore, purification of the photolabeled proteins using
nondenaturing conditions demonstrated that pk18 and pk10 copurify
substantiating their intimate association. A set of three antisera,
specific to different regions of pk18, did not recognize pk10 on
Western blots. Likewise, partial amino acid sequence of peptide
fragments indicate that pk10 is not derived from proteolytic cleavage
of pk18. These data suggest that pk10 represents another component of
mitochondrial benzodiazepine receptors whose identity is not apparent
with any known protein.
The principal pharmacological actions of benzodiazepines, a
class of widely prescribed anxiolytic drugs, is believed to be mediated
primarily through binding to a specific allosteric modulatory site
located on ionotropic Despite the
prevalent therapeutic use of benzodiazepines, the potential secondary
drug effects mediated through MBR are not understood, but recent
progress has been made to understand the function of this protein
complex(5) . In steroidogenic tissues MBR were found to be
coupled to the regulation of steroid
biosynthesis(6, 7, 8, 9) . This
activity appears to involve a specific interaction with a cytosolic
protein known as diazepam binding inhibitor (10, 11, 12, 13) to facilitate
intramitochondrial cholesterol transport, the rate-limiting step of
steroidogenesis(14) . What remains to be elucidated, however,
is the precise molecular mechanism of MBR in this process and what role
MBR plays in cells which do not synthesize steroids. It seems probable
that MBR might play a fundamental role in the activity of the discreet
mitochondrial populations that contain these receptors. With relevance
to understanding the relationship of MBR to mitochondrial function
several proteins which comprise this complex have tentatively been
identified(15) . In attempts to elucidate MBR function a
number of other pharmacological agents have been discovered such as
aryl indoleacetamides(16) , imidazopyridines(17) ,
quinoline propanamides(18) , and isoquinoline
carboxamides(19) , along with several other more recently
reported classes of compounds(20) . Isoquinoline carboxamides
include the highly selective fluoro-nitro photoaffinity probe PK14105 (21) which has been widely reported to photolabel an 18-kDa
protein almost exclusively. This mitochondrial protein has been
purified (22, 23) and the corresponding cDNA
cloned(24, 25, 26, 27) .
Transfection experiments with the cDNA demonstrate the expression of
binding sites for benzodiazepines and isoquinoline carboxamides with
characteristic MBR specificity(24, 25, 28) . To gain further insight into other proteins associated with MBR, we
have made further use of PK14105 as a selective photoaffinity probe.
The studies reported here present data showing the existence of a
second protein specifically photolabeled by PK14105 which is associated
with the 18-kDa protein described above.
In some experiments, rat adrenals
were subfractionated into four subcellular compartments. A
nuclear-enriched pellet was obtained from the centrifugation at 750
Protein of all subcellular
fractions was quantified by the method of Bradford (29) using
bovine
For immunizations, the peptide conjugates were
emulsified with an equal volume of complete Freund's adjuvant and
800 µl, corresponding to about 800 µg of antigen, was injected
intradermally into New Zealand White rabbits. For boostings, the
conjugates were emulsified in Freund's incomplete adjuvant and
injected subcutaneously 2 months and 8 months following the initial
injections. The quantities of antigen used for these subsequent
injections were 250 and 100 µg, respectively. One week following
boosting injections, rabbits were bled from ear veins and antisera were
prepared.
For immunoblot analysis the nitrocellulose
membranes were blocked with Dulbecco's phosphate-buffered saline
containing 0.1% Tween 20 and 3% bovine serum albumin for 2 h, incubated
with antisera (1:1000) in blocking buffer for 2 h, then washed three
times (10 min each) with phosphate buffer. The membranes were incubated
for 60 min with a 1:1000 dilution of secondary antibody (alkaline
phosphatase-linked goat anti-rabbit IgG) in blocking buffer, washed
three times for 10 min with phosphate buffer and developed in 0.1 M Tris-HCl (pH 9.5), 0.1 M NaCl, 0.05 M MgCl
Partially purified receptor fractions were
next applied to Q-Sepharose columns previously equilibrated with 0.1%
digitonin in Buffer A at a ratio of 4 ml of column bed volume per 50 mg
of initial mitochondrial protein used in each preparation. All the
columns were packed to a diameter:length ratio of 1:50 and a flow rate
of 0.2 ml/min was used where 3-ml fractions were collected. Under such
conditions the photolabeled protein did not bind. The flow-through
fractions were thus pooled, supplemented with additional digitonin to
1%, and applied to another Q-Sepharose column equilibrated in 0.1 M glycine (pH 9.5), 0.1% digitonin. Again, with this second
buffer system, the photolabeled protein was found predominantly in the
flow-through fractions. The pooled MBR-containing fractions from the
second column were concentrated to about 2 ml with an Amicon
Centriprep-30 unit and applied to a 20-40% sucrose gradient
prepared with 0.025% digitonin in Buffer A. The gradient was
centrifuged overnight at 40,000 rpm in a Sorvall TH-641 rotor.
Fractions of 1 ml were collected where the radioactive peak was found
near the middle of the gradient. These fractions were pooled, diluted
to 9 ml in Buffer A (without digitonin) and placed in the centrifuge
tube underlaid with 25% sucrose in Buffer A. After another overnight
centrifugation at 40,000 rpm the photolabeled protein was concentrated
at the bottom. The upper portion of the discontinuous gradient was
carefully removed. The sucrose fraction was collected, diluted with an
equal volume of ethanol, and kept on ice. This caused the protein to
precipitate within 10 min where centrifugation at 15,000
The autoradiograms in Fig. 1show photolabeling of mitochondrial fractions from 10
different rat organs. Adrenals exhibit specific photolabeling of three
proteins having apparent molecular masses of about 36, 15, and 10 kDa.
In order to simplify referring to these specifically photolabeled
proteins in the context of this report they are signified as pk36,
pk18, and pk10, respectively. Because a protein of 10 kDa was
identified, all studies described here were performed using a
Tris/Tricine/SDS-PAGE system (30) which affords excellent
resolution of polypeptides in the range of 100 to 1 kDa. With this
buffer system pk18 migrates with an apparent mass of 15 kDa despite the
fact that the corresponding cDNA has been cloned and sequenced, from
which a mass of about 18 kDa is predicted for this
protein(24) . Furthermore, earlier studies using the SDS-PAGE
protocol of Laemmli (31) give a relative migration more
consistent with 18 kDa (21, 22) .
Figure 1:
Specific
[
All of the rat
mitochondrial preparations shown in Fig. 1are known to have
MBR, but their levels vary by 2 orders of magnitude. For example,
mitochondrial fractions from rat adrenals contain densities of >100
pmol/mg of protein, whereas those from the skeletal muscle or neuronal
tissue have 1-3 pmol/mg of protein(2) . In all of these
mitochondrial preparations pk18 and pk10 were detected as proteins
specifically photolabeled by [ In
contrast to the similar pattern observed for these two proteins, pk36
was detected only in adrenals. Although MBR was shown to be
functionally coupled to steroid biosynthesis (6, 7, 8, 9) the presence of pk36
does not seem to be a feature common to all steroidogenic tissues
because this protein was not identified in mitochondrial fractions from
testes or ovaries (Fig. 1). In several instances, other
photolabeled proteins were observed in different tissues. Many of these
are nonspecifically labeled as discriminated using 10 µM PK11195 as a competitor, but in certain tissues such as kidney or
skeletal muscle other specifically labeled proteins are faintly
detected. The remaining experiments were performed with adrenals
because this tissue contains the highest MBR levels.
Figure 2:
Specificity of
[
The competitor ligands tested are derived
from five classes of aromatic compounds, each having a well
characterized specificity for MBR. In understanding the results
obtained with rat adrenal mitochondria it is best to consider each
class of compounds individually. 1) The isoquinoline carboxamide
PK11195, a high-affinity congener of PK14105 (19) , effectively
competed against photolabeling of all three proteins. 2) A pair of
quinoline propanamide stereoisomers,(-)-PK14067 and
(+)-PK14068, display stereoselective specificity of MBR. The
(-)-isomer exhibits a low nanomolar affinity being over 2 orders
of magnitude more potent than the (+)-isomer(18) . This
stereoselective specificity was observed in the competition
photolabeling of all three proteins. 3) Alpidem and zolpidem are
imidazopyridines showing subnanomolar and micromolar affinities for
MBR, respectively(17) , a pattern which is again reflected in
the photolabeling paradigm. 4) The benzodiazepines Ro5-4864 and
flumazenil show reciprocal selectivity for MBR and type A
An equivalent
scenario can be drawn with the photolabeling of bovine adrenocortical
mitochondria (Fig. 2, lower panel). There are several
important differences, however, with respect to the results from rat
adrenals. Several other proteins with masses of 30-60 kDa are
found to be photolabeled by PK14105, but these represent
nonspecifically labeled adducts as neither PK11195 nor any of the other
compounds competed against their photolabeling. Therefore, a bovine
counterpart of pk36 is not observed, but it may be masked by the
nonspecifically labeled proteins evident here. The photolabeling of
pk18 and pk10 is easily detected in bovine samples and these show the
same sensitivity to the different ligands as was found in the rat
except for one major difference; Ro5-4864 did not appreciably compete
against their photolabeling. This is an important point because it is
well documented that Ro5-4864, while having nanomolar affinity for
rodent MBR, exhibits micromolar affinity in bovine
tissues(25) . As this result was anticipated with pk18, a
conspicuous relationship is evident in that photolabeling of pk10 not
only shows the chemical and stereoselective specificities of MBR, but
it also displays this characteristic species-specific difference. These
results alone provide strong evidence that pk10 may be another
component of MBR in close association with pk18.
Figure 3:
Subcellular distribution of photolabeled
proteins. Rat adrenal homogenates were subfractionated to yield pellets
at 750, 8,000, and 100,000
Figure 4:
Comparison of photolabeling properties for
pk18 and pk10. All photolabeling experiments included rat adrenal
mitochondrial fractions at a concentration of 100 µg of protein/ml.
In Panel A the concentration of
[
Another
criterion that was exploited in this comparison was the fact that MBR
affinity for Ro5-4864 at 4 °C is about 1 order of magnitude higher
than it is at 37 °C(32) . This property seems particular to
benzodiazepines as binding of isoquinoline carboxamides is not
appreciably temperature dependent. Given these parameters it would be
expected that Ro5-4864 should be a potent inhibitor of photolabeling at
4 °C and not as potent at 37 °C, whereas competition by PK11195
should be independent of temperature. Competition photolabeling
experiments with these two ligands at 100 nM demonstrate this
expected behavior for both pk18 and pk10 (Fig. 5). This
criterion again reveals the cognate features that pk18 and pk10 have in
their photolabeling properties. It is worthy to note that photolabeling
of pk36 does not show a similar temperature dependence in photolabeling (Fig. 5, upper panel). This finding places further
skepticism on the possibility that pk36 is associated with MBR.
Figure 5:
Temperature dependence of photolabeling
competition by Ro5-4864. Rat adrenal mitochondrial fractions at 100
µg of protein/ml were incubated and photolabeled with 10 nM [
A
photolabeled preparation of rat adrenal mitochondria was determined to
have 17-fold higher level of radioactivity incorporated into pk18
relative to pk10. This preparation was subjected to immunoblot analysis
wherein a 17-fold difference in the amounts of protein were loaded into
pairs of neighboring lanes during SDS-PAGE. With this experimental
design, if the antisera detect pk18 in the lanes having 17-fold less
protein, then the neighboring lane would have an equivalent level of
pk10 (based on radioactivity) which should be detected if it is a
fragment of pk18. All three antisera vividly revealed pk18 in the lanes
loaded with less protein, however, pk10 was not recognized by any of
the three antisera (Fig. 6). The largest peptide segment of pk18
spanning a region not recognized by any of the three antisera is about
6.5 kDa, hence, it is unlikely that a 10-kDa fragment of pk18 could
escape detection by the battery of antisera. This immunochemical
evidence suggests that pk10 is a protein immunologically distinct from
pk18. Additional studies to rule against proteolytic cleavage of pk18
are that extended incubations (1-2 h) at 37 °C either before
and during (Fig. 5) or after photolabeling (data not shown) do
not change the proportion of photolabeled pk10 relative to pk18.
Furthermore, the presence of a diversity of protease inhibitors (see
``Experimental Procedures'') did not affect the photolabeling
pattern when compared with experiments omitting these inhibitors.
Figure 6:
Immunoblot analysis with antisera against
different regions of pk18. A preparation of photolabeled rat adrenal
mitochondria containing 17-fold higher radioactivity associated with
pk18 compared to pk10 was electrophoresed in four sets of double lanes,
one lane (17
Figure 7:
Immunoprecipitation with antiserum against
pk18. Rat adrenal mitochondria were photolabeled at a concentration of
500 µg of protein/ml in the presence of 10 nM [
Further support for this possibility was
obtained by developing a purification scheme to isolate the
photolabeled proteins under nondenaturing conditions. The protocol
devised for this was adapted from our previous studies (22, 33) using a detergent different than those used
for immunoprecipitation. In all purification methods we tested, pk18
and pk10 were always found to cofractionate as exemplified by the
purified preparation shown in Fig. 8. These purified fractions
typically show several protein species of which pk18 and pk10 are major
components. Several proteins with a mass of 30-34 kDa were
verified to be the voltage-dependent anion channel and ADP/ATP carrier
using specific antibodies. These proteins were reported earlier to
copurify with pk18 in a similar purification scheme with rat kidney
mitochondria(15) . The relative abundance of pk18 is much
greater, however, using the methods described here, probably because
adrenal mitochondria were used which contain much higher MBR levels.
The presence of additional proteins at 80, 38, and 5 kDa is under
investigation concerning their possible relationship with MBR. Of
particular concern with these findings is that the voltage-dependent
anion channel and ADP/ATP carrier are known to be among the most
abundant mitochondrial proteins (apparently with the 5-kDa protein as
well, as seen in Fig. 8) and their cofractionation with pk18 may
simply be an unavoidable consequence of their contamination in
detergent micelles where pk18 was solubilized. Nevertheless, the
photolabeling experiments provide strong support that pk10 is closely
tied with MBR and therefore its copurification with pk18 most likely
reflects this intrinsic association. The staining intensities seen in Fig. 8suggest that the molar ratios of pk18 and pk10 are more
comparable than the relative intensities by which both proteins are
photolabeled.
Figure 8:
Purification of pk18 and pk10. Rat adrenal
mitochondria photolabeled with [
From preparations as shown in Fig. 8, pk10 is
easily resolved from other proteins by SDS-PAGE. This has enabled us to
purify pk10 for generation of various peptide fragments. Among these, a
tryptic digest of pk10 gave a peptide bearing the sequence XLADK
demonstrating that pk10 is not a fragment of pk18. Additional sequence
not reported here indicates that pk10 is not currently represented in
all sequence data bases of the National Center for Biotechnology
Information. Further studies are thus necessary to clone a cDNA for
pk10 and reveal the chemical nature of this newly discovered MBR
protein. The development of PK14105 has proven to be a crucial step in
unveiling proteins associated with MBR. The protein referred to as pk18
in this report has been unequivocally shown to be an essential
component for expressing the specific drug recognition sites of this
complex(24, 25, 26, 27, 28) .
Other proteins have been reported to be photolabeled specifically by
PK14105 (15, 23, 25, 35) but a
salient association of these proteins with MBR has not been
demonstrated. Of concern in some of these reports is that very long
periods of photolabeling are necessary to observe the additional
proteins, in which case a broader photolabeling selectivity and damage
through extensive ultraviolet irradiation may be occurring. By
comparison, all photolabeling studies described here involved only 15
min of irradiation and were scrutinized for specificity with a series
of competitor ligands. The identification of pk36 and pk10 is a
subject which had gone unnoticed in many previous studies. The primary
reason for this is that the level of photolabeling of these two
proteins is much lower than that of pk18. Much longer times of
autoradiography are required to detect these less apparent adducts and
thus are normally undetected in the time it takes to clearly visualize
pk18. An earlier study reported observing pk36(35) , where it
was also indicated that in mitochondrial fractions from several rat
tissues, this adduct was only found in adrenals as is confirmed here.
The disclosure of pk10 is a novel finding and, for the case of MBR
structure, is the most relevant discovery in these studies. This
photolabeled species had been observed by us previously (see (2) and (13) ), however, it was inconsistently detected
and initially believed to result from a breakdown of pk18. SDS-PAGE
performed by the procedure of Laemmli (31) gives poor
resolution and reproducibility in this molecular weight range
accounting primarily for the inconsistency we initially observed in
detecting pk10. Changing to a different SDS-PAGE system which permitted
superb resolution as low as 1 kDa proved to greatly facilitate the
analysis of pk10. As a result of this improvement in methodology, pk10
is invariably detected when pk18 is present. Because several tests
showed discrepancies in the photolabeling of pk36 compared to pk18, it
is difficult to validate an association of pk36 with MBR. In the
attempt to correlate other proteins with MBR, the revelation of pk10
fulfills many stringent criteria. 1) pk10 and pk18 are mutually present
in all tissues examined containing MBR. The apparent ratio of PK14105
photoincorporation into these two proteins also seems invariable
despite the fact that MBR levels vary by 2 orders of magnitude among
different tissues. 2) Photolabeling of pk10 and pk18 exhibit identical
responses when testing chemical specificity, stereoselectivity, and
species-dependent variation in benzodiazepine recognition. 3)
Equivalent patterns in photolabeling of both proteins are observed at
PK14105 concentrations between 1 and 100 nM and the same
IC The most plausible explanation to
account for this long list of coincident photolabeling patterns between
pk18 and pk10 is that these two proteins are associated together in a
heteromeric complex. A number of laboratories have verified that pk18
is essential for expression of MBR drug binding
sites(15, 24, 25, 26, 27, 28) .
The fact that PK14105 intensely labels pk18 upholds the proposal that
this protein constitutes the binding domain for isoquinoline
carboxamides. It is feasible that pk10 lies in close proximity to this
domain such that in only 5% of instances pk10 is covalently modified by
PK14105, whereas pk18 is labeled in the remaining 95% of successful
photolabeling events. This ratio can account for the 15-20-fold
difference in photolabeling efficiency consistently observed between
these two proteins. Such a possibility can account for the observation
that the molar ratios of pk10 and pk18 are not as disparate as judged
by silver staining following SDS-PAGE of subfractionated receptor
preparations. pk10 may actually comprise a small facet of the binding
domain, or be labeled by virtue of being a near neighbor, where
photoactivated ligand may dissociate from its binding site and diffuse
a short distance in this excited state to encounter pk10. In studies
not shown here, the use of a scavenger such as
diethylamine(21) , does not inhibit the photolabeling of pk18
or pk10, implying that PK14105 may not need to leave its binding domain
in order to label pk10. The partial amino acid sequence derived from
purified pk10 establishes that this protein is not a fragment of pk18,
nor is it the 10-kDa cytosolic protein diazepam binding inhibitor which
is suspected of interacting with
MBR(10, 11, 12, 13) . An identical
protein containing this and other sequence obtained is not represented
in any of the current sequence data bases. Because pk10 appears to show
a distinct association with pk18 it seems credible that this protein
has not been identified in another context outside of MBR. Evidence
is mounting that MBR exists as a hetero-oligomeric complex (15, 33, 36, 37) and identification
of all proteins in this entity are essential to understand its
biochemical role. Strong data have been reported for the function of
MBR in steroidogenesis (5) and mitochondrial respiratory
control (38) but the molecular mechanisms underlying these
findings remains a mystery. Current research goals include revealing
the identity of pk10 to hopefully obtain novel insight into MBR
function and the pharmacological consequences of drugs which bind to
these sites.
Volume 270,
Number 35,
Issue of September 01, pp. 20285-20291, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
-aminobutyric acid receptors. In addition to
binding to this class of inhibitory neurotransmitter receptors, there
is another class of high affinity binding sites for benzodiazepines and
other drugs found in virtually all tissues, primarily localized on
mitochondria(1, 2, 3) , hence they are
referred to as peripheral-type or mitochondrial benzodiazepine
receptors (MBR)(
)(4, 5) .
Subcellular Fractionation
Tissues from adult
Sprague-Dawley rats or bovine adrenal cortices were homogenized in 20
volumes of ice-cold 10 mM HEPES (pH 7.4), 0.32 M sucrose using 12 strokes in a tight fitting Potter-Elvehjem glass
tissue grinder. The HEPES/sucrose buffer ordinarily contained a
protease inhibitor mixture consisting of 0.1 mM 4-(2-aminoethyl)-benzenesulfonylfluoride, 0.1 µM Pepstatin A, 1 µg/ml leupeptin, and 1 mM EDTA,
although experimental results were found to be identical when performed
in the absence of these inhibitors. The homogenates were then
centrifuged for 5 min at 750 g. The resulting
supernatants were recovered and centrifuged a second time at 750
g. The supernatants were removed again, then
centrifuged at 5,200
g for 10 min yielding membrane
pellets that were resuspended in 5 volumes of HEPES/sucrose buffer and
centrifuged for 10 min at 9,000
g resulting in
mitochondrial-enriched pellets.
g which was resuspended in HEPES/sucrose buffer and
centrifuged again at 750
g yielding the nuclear
pellet. The mitochondrial-enriched fractions in these experiments were
obtained by centrifugation of the 750
g supernatants
for 10 min at 8,000
g, wherein the pellets were washed
by resuspension and centrifugation. A final centrifugation of the 8,000
g supernatant at 100,000
g for 30 min
produced a pellet consisting of microsomal membranes and a supernatant
representing the cytosolic fraction.
-globulin as a standard. All fractions were stored in
HEPES/sucrose buffer at -70 °C until ready for use.
Generation of Antibodies
The peptides
YGSYIIWKELGGFTE and LNYYVWRDNSGRRGGSRL, corresponding to an internal
sequence and the carboxyl-terminal region, respectively, of the 18-kDa
MBR protein (24) were synthesized. Each synthetic peptide was
coupled to keyhole limpet hemocyanin at a concentration of 2 mg/ml of
peptide and protein, each dissolved in Dulbecco's
phosphate-buffered saline. To these mixtures
1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide was added at 2 mg/ml and
allowed to react for 1 h at room temperature followed by an overnight
incubation at 4 °C. The peptide-protein conjugates were dialyzed
twice against 1 liter of Dulbecco's phosphate-buffered saline and
stored frozen.Photoaffinity Labeling
Photolabeling of
mitochondrial or subcellular fractions were performed as described in
earlier reports (8, 22) with several modifications.
Membranes were suspended in 10 mM HEPES (pH 7.4), 0.32 M sucrose at concentrations from 50 to 1000 µg of protein
per ml in the presence of 10 nM [
H]PK14105, unless indicated otherwise.
After 90 min at 4 °C, when binding of the radioligand had reached
equilibrium, the 1-ml samples were placed in separate wells of 24-well
tissue culture dishes and irradiated for 15 min from a distance of 2 cm
with a 15 watt ultraviolet light having maximum emission at 366 nm. The
samples were then centrifuged at 15,000 g and the
pellets were solubilized in sample loading buffer for electrophoresis.
SDS-PAGE and Transblot Analysis
Electrophoresis of
proteins was performed using the Tris/Tricine/SDS-PAGE system described
by Schägger and von Jagow(30) . As
according to this reference, the polyacrylamide gels of 1.5-mm
thickness were 12% T/3% C (interpreted as a 12% total monomer
concentration of which 3% of the original monomer concentration was
bisacrylamide) with a stacking gel of 4% polyacrylamide.
Electrophoresis was performed at 10 V/cm in a Bio-Rad Mini Protean II
electrophoresis unit. Protein was transferred from gels to
nitrocellulose membranes using a Bio-Rad Mini Protean II transblot
apparatus at 75 V for 60 min at 4 °C in 25 mM Tris, 200
mM glycine, 20% methanol. Autoradiography of photolabeled
proteins transferred to nitrocellulose was performed by spraying the
membranes with EnHance (Du Pont) and exposing them to x-ray
film at -70 °C.
with nitro blue tetrazolium
chloride/5-bromo-4-chloro-3-indolylphosphate.Radioactive Quantitation of Photolabeled Protein
Species
To quantify the extent of photolabeling in distinct
protein species each sample was loaded onto polyacrylamide gels with a
small amount of Rainbow Protein Molecular Weight markers from Amersham.
After transfer to nitrocellulose the migration of the molecular weight
markers in each lane was clearly visible and served as a reference to
facilitate excising the small patches of membrane containing isolated
photolabeled protein bands. These pieces of nitrocellulose membrane
were placed in scintillation vials and incubated overnight with liquid
scintillation fluid prior to radioactive determination. Quantitative
variation using this method was <10% comparing identical
photolabeled samples electrophoresed in several lanes.Immunoprecipitation
Photolabeled rat adrenal
mitochondrial fractions were solubilized at 4 °C in 50 mM
Tris-HCl (pH 8.0), 150 mM NaCl, 1% Triton X-100, 0.5% sodium
deoxycholate at a final protein concentration of 1 mg/ml. After 30 min,
the mixtures were centrifuged at 20,000 g for 10 min
and the supernatants were collected. To 250-µl aliquots of
supernatant, 10 µl of antiserum were added and the samples were
incubated overnight at 4 °C. Immunoprecipitates were collected by
adding 25 µl of a 1:1 slurry of Protein
A-Sepharose/Dulbecco's phosphate-buffered saline, gently mixed
for 2 h at 4 °C, and centrifuged at 12,000
g for
30 s. The pelleted beads were washed 3 times with 0.5 ml of
phosphate-buffered saline, once with 10 mM Tris-HCl (pH 7.4),
and bound protein was dissociated by heating to 95 °C in 100 ml of
10 mM Tris-HCl (pH 7.4), 0.5% SDS for 5 min. The supernatants
were recovered and combined with a subsequent wash of the beads
accomplished with 50 µl of 10 mM Tris-HCl (pH 7.4).
Purification of MBR Proteins
All buffers used in
these procedures included the protease inhibitor mixture described
under ``Subcellular Fractionation.'' Pellets of photolabeled
rat adrenal mitochondria were resuspended at a protein concentration of
5 mg of protein/ml in 4 M urea, Buffer A (50 mM NaCl,
1 mM EDTA, 50 mM Tris-HCl, pH 8.0) and incubated on
ice for 60 min. The samples were then diluted 2-fold with ice-cold
Buffer A and centrifuged at 27,000 g for 45 min. The
supernatant was discarded and the pellets were resuspended to a final
concentration of 10 mg of protein/ml with 2% digitonin (Lot PTP 9176,
Wako Chemicals, Richmond, VA) in Buffer A and kept on ice for 60 min
with constant stirring. The samples were then centrifuged for 20 min at
12,000
g after which the digitonin-solubilized
supernatant was transferred to 12-ml Sorvall polyallomer tubes (14.7
88.9 mm). To the bottom of each tube a 1-ml layer of 30%
sucrose in Buffer A with 2% digitonin was underlaid. Overnight
centrifugation at 150,000
g resulted in concentration
of photolabeled protein into the sucrose layer with a small residual
insoluble pellet at the bottom. The upper layer from each tube was
removed carefully and the sucrose cushion was diluted 3-fold with 2%
digitonin in Buffer A.
g for 10 min produced a protein pellet relatively free of digitonin,
suitable for electrophoretic separation after dissolution in SDS-PAGE
loading buffer.
Identification of Multiple Proteins Specifically
Photolabeled by PK14105
Previous studies by this and other
laboratories had shown that the isoquinoline carboxamide PK14105
specifically photolabels an 18-kDa mitochondrial
protein(21, 22, 23, 25) . This
protein has been clearly demonstrated to be associated with high
affinity recognition sites for isoquinoline carboxamides and
benzodiazepines(24, 25, 28) . When
[H]PK14105 photolabeled mitochondrial
preparations from nearly any mammalian tissue source are subjected to
electrophoresis, autoradiography of the corresponding gels for
relatively short times typically reveals only one specifically labeled
protein. We have noticed, however, that when autoradiography is
extended for much longer times (5-10-fold longer) other
specifically labeled proteins may be detected.
H]PK14105 photolabeling in mitochondrial
fractions from various rat tissues. Mitochondrial-enriched membrane
fractions from different rat organs (Ad, adrenal; Ov,
ovary; Te, testis; He, heart; Ki, kidney; Lu, lung; Sp, spleen; SM, diaphragm skeletal
muscle; Pa, pancreas; SC, spinal cord) at
concentrations of 1 mg of protein/ml were photolabeled in the presence
of 10 nM [H]PK14105 as described under
``Experimental Procedures.'' Nonspecific photolabeling (N), performed with 10 µM PK11195 included in the
incubation buffer, is compared with total photolabeling (T)
where no competitor was introduced. After solubilization in
electrophoresis sample buffer the following amounts of protein were
electrophoresed in their respective lanes: A-SM, 100 µg; P-SC, 150 µg. Following transfer to nitrocellulose, the
membranes were autoradiographed for either 7 days (A-Sp) or 21 days
(SM-SC). Migration of molecular weight markers is indicated and the
three specifically photolabeled protein bands of adrenals are
denoted.
H]PK14105. Note
also that the relative ratio of intensity between these two
photolabeled adducts remained fairly constant despite the fact that MBR
density is highly variable among the different tissue sources.Ligand-specific Competition of Photolabeling
To
examine whether the nature of these proteins to be photoaffinity
labeled by PK14105 exhibits specificity of MBR several discriminatory
ligands were used as competitive displacers at concentrations of 1
µM in the photolabeling experiments. Rat adrenal and
bovine adrenocortical mitochondrial preparations were compared for this
purpose (Fig. 2).
H]PK14105 photolabeling in rat and bovine
adrenal mitochondrial fractions. Either rat adrenal (100 µg of
protein/ml) or bovine adrenocortical (300 µg of protein/ml)
mitochondrial fractions were photolabeled with 10 nM [H]PK14105 in the presence of 1 µM of the following competing ligands: Co, control (no
competitor); PK, PK11195; Q-,(-)-PK14067; Q+, (+)-PK14068; A, alpidem; Z,
zolpidem; Ro, Ro5-4864; FM, flumazenil; FG,
FGIN-2. The quantities of photolabeled mitochondrial protein
electrophoresed on each lane were 20 µg of protein for rat samples
and 60 µg of protein for bovine samples. Autoradiograms of the
corresponding transblots were exposed for 14 days. Beneath each
autoradiogram are listed values in µM of the IC each competing ligand displayed in competing for specific
[
H]PK14105 binding. These binding experiments
were performed as described previously (33) in an incubation
volume of 0.2 ml containing mitochondrial protein and
[H]PK14105 at concentrations equal to those used
for photolabeling in this experiment. Five concentrations of each
competing ligand were tested to displace specific
[H]PK14105 binding for estimation of IC values by nonlinear regression
analysis.
-aminobutyric acid receptors(20) . Consistent with the
specificity of mitochondrial benzodiazepine recognition sites, Ro5-4864
competed against the photolabeling of all three proteins whereas
flumazenil had no apparent effect. 5) FGIN-2, an aryl indoleacetamide
displaying high affinity for MBR(16) , also proved to be an
effective competitor. In summary, with rat adrenal mitochondria,
photolabeling of pk36 and pk10 exhibited the same sensitivity to all
competitors as the established pk18 MBR protein.
Subcellular Fractionation Studies
To corroborate
whether pk36 and pk10 are directly associated with MBR, subcellular
fractionation of rat adrenals was performed to examine whether these
proteins are found predominantly with mitochondrial fractions. Four
subfractions were obtained corresponding to preparations enriched in
nuclear, mitochondrial, microsomal, and cytosolic
compartments(2) . Photolabeling of all four subfractions was
performed in addition to measuring in each the specific activity of the
mitochondrial marker enzyme cytochrome c oxidase (Fig. 3). The levels of photolabeled pk18 and pk10 showed
distributions parallel with that of the mitochondrial marker. In
contrast to this profile, pk36 cofractionated partly with mitochondria
and also with the cytosolic fraction. Subsequent experiments have
therefore focussed on the mutual relationship that pk10 with pk18
consistently manifest because the relevance of pk36 to MBR is dubious
based on several observations discussed to this point.
g, indicated as P-750,
P-8K, and P-100K, respectively. These membrane pellets
and the supernatant remaining following the 100,000
g centrifugation (Supe) were photolabeled at a
concentration of 50 µg of protein/ml. Membranes were pelleted
whereas the supernatant samples were concentrated to 0.1 ml on
Centricon-30 units and 10 µg of protein was processed for Western
transfer and autoradiography. The lower panel of the figure
shows quantitation of radioactivity migrating with each protein
species. The open bars represent specific activity of
cytochrome c oxidase measured as described in earlier
studies(2) . Quantitation of enzyme activity and radioactivity
associated with each of the three protein species are normalized
relative to the values determined in P-8K which were: cytochrome c oxidase (open bars), 381 nmol/min/mg of protein; pk36 (hatched bars), 5.2
10
dpm; pk18 (solid bars), 2.0 10
dpm; pk10 (cross-hatched bars), 1.2 10
dpm. Means
± S.D. of at least three photolabeling experiments are
shown.
Comparison of pk18 and pk10 Photolabeling
Properties
The dissociation constant of MBR for PK14105 is
typically about 5 nM. Photolabeling of pk18 and pk10 at
[H]PK14105 concentrations in the range of
1-100 nM show equivalent patterns which approach
saturation at 100 nM (Fig. 4A). Photolabeling
competition studies with nonradioactive PK14105 and(-)-PK14067 at
concentrations from 10 nM to 10 µM were also
performed. The IC values for each ligand to compete
against photolabeling by 10 nM
[
H]PK14105 was about 20 nM for both pk18
and pk10 (Fig. 4B), consistent with their reported
affinities for MBR. These findings continue to highlight the close
correlations between photolabeling of pk18 and pk10.
H]PK14105 was varied and radioactivity migrating
with pk18 (
) and pk10 (
) was quantified following
electrophoresis and transfer to nitrocellulose. Means ± S.E. of
duplicate experiments are shown. Note that both axes are plotted on log
scales. In Panel B 10 nM [H]PK14105 was used in the presence of
different concentrations of either nonradioactive PK14105
or(-)-PK14067 and radioactivity was determined for both proteins. Symbols indicate mean values of duplicate determinations; the
S.E. was <5% in competition photolabeling experiments with
(-)-PK14067. Data using nonradioactive PK14105 are virtually
superimposable with the data plotted for(-)-PK14067. The
competition curve was fitted by nonlinear regression analysis (assuming
a Hill coefficient of 1) from the data for pk18 using nonradioactive
PK14105 as a displacer.
H]PK14105 at either 4 or 37 °C in the
absence (Co) or presence of 100 nM PK11195 (PK) or Ro5-4864 (Ro). Autoradiography (upper
panel) and quantitation of radioactivity (lower panel) in
the pk18 (solid bars) and pk10 (hatched bars) bands
are shown.
pk10 Is Not a Proteolytic Fragment of pk18
To rule
out the possibility that pk10 is a proteolytic fragment derived from
pk18 several antisera specific for different regions of pk18 were used.
The first antiserum (aNI) was generated against synthetic
peptides corresponding to the sequences from Ser to
Met and Gly
to Leu
(26) giving specificity for internal and amino-terminal
segments of pk18. Antiserum aI was specific to another
internal sequence from Tyr
to Glu
and the
third antiserum aC was directed to the carboxyl-terminal
region of pk18 from Leu
to Leu
.
) having 17 times more protein than its
neighboring lane (1
). The amount of protein signified
as 1
corresponds to 9 µg except for the lanes under aC where this level is 20 µg. After transfer to nitrocellulose
one pair of lanes was autoradiographed ([
H]PK)
whereas the other lanes where processed for immunoblot analysis using
three antisera generated to different segments of pk18. Antiserum from
a rabbit immunized with keyhole limpet hemocyanin alone (NS)
is used as an indicator of nonspecificity arising from immunizations
using this carrier protein conjugated with peptides corresponding to
amino-terminal/internal (aNI), internal (aI), or
carboxyl-terminal (aC) sequences of pk18 as described in the
text.
Molecular Association and Purification of pk18 and
pk10
By all methods used up to this point, photolabeling of pk10
displays precisely the same ligand specificity and physicochemical
sensitivity as that of pk18 suggesting that these two proteins are
closely associated with each other. To investigate this possibility
more directly immunoprecipitation experiments were performed with
antiserum against pk18. From detergent-solubilized preparations of
photolabeled rat adrenal mitochondria, antiserum aI coprecipitated pk18 and pk10, whereas, preimmune serum or an
appropriate control antiserum to hemocyanin did not precipitate either
protein (Fig. 7). Note also that pk36 did not precipitate with
this complex. Identical results were obtained with antiserum aNI (data not shown). These data, therefore, provide more conclusive
evidence that pk18 and pk10 are likely to be tightly associated in a
heteromeric complex.
H]PK14105 and processed through an
immunoprecipitation scheme as described under ``Experimental
Procedures.'' The protein recovered from this procedure was
lyophilized, dissolved in 20 µl of water, and prepared for
SDS-PAGE, transfer to nitrocellulose and autoradiography. Shown are
immunoprecipitates obtained using antiserum aI (see Fig. 6)
against pk18, rabbit antiserum against keyhole limpet hemocyanin as a
nonspecific control (Co) for antiserum aI, and
preimmune serum (Pre).
H]PK14105 were
processed through the purification scheme detailed under
``Experimental Procedures.'' The left panel shows
the electrophoretic pattern of proteins stained with silver according
to the procedure of Morrissey(34) . From a preparation starting
with 50 mg of mitochondrial protein the lanes represent specific
proportions of different fractions as follows: A, 0.5% of
total mitochondrial protein; B, 0.5% of the urea-washed
membrane pellet; C, 3% of the final purified preparation. The right panel shows an autoradiogram of fraction
C.
values are obtained in competition photolabeling
studies with nonradioactive PK14105 or(-)-PK14067. 4)
Photolabeling of pk10 displays the characteristic MBR behavior that
affinity for Ro5-4864 is markedly temperature dependent, whereas the
binding of PK11195 is insensitive to temperature. 5) pk10
cofractionates with mitochondrial markers plus, with two different
detergent solubilization schemes, it immunoprecipitates and copurifies
with pk18 suggesting a tight physical association between the two
proteins. Furthermore, partial amino sequence and the use of three
separate antisera recognizing defined regions of pk18 argues against
the possibility that pk10 is a proteolytic fragment of, or is
antigenically related to pk18.
)
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
ABSTRACT
This article has been cited by other articles:
|
|
 |
|
  M. Bose, R. M. Whittal, W. L. Miller, and H. S. Bose Steroidogenic Activity of StAR Requires Contact with Mitochondrial VDAC1 and Phosphate Carrier Protein J. Biol. Chem., April 4, 2008; 283(14): 8837 - 8845. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Giatzakis and V. Papadopoulos Differential Utilization of the Promoter of Peripheral-Type Benzodiazepine Receptor by Steroidogenic Versus Nonsteroidogenic Cell Lines and the Role of Sp1 and Sp3 in the Regulation of Basal Activity Endocrinology, March 1, 2004; 145(3): 1113 - 1123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Yeliseev, K. E. Krueger, and S. Kaplan A mammalian mitochondrial drug receptor functions as a bacterial "oxygen" sensor PNAS, May 13, 1997; 94(10): 5101 - 5106. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Segade, B.én Hurlé, E.ía Claudio, S.ía Ramos, and P. S. Lazo Identification of an Additional Member of the Cytochrome c Oxidase Subunit VIIa Family of Proteins J. Biol. Chem., May 24, 1996; 271(21): 12343 - 12349. [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 |