 |
INTRODUCTION |
In the preceding paper (32), we show that XL
s (for
extra large s), an unusual type of G protein
subunit, is predominantly associated with the plasma membrane of
certain neuroendocrine cells. XLs consists of a novel 37-kDa XL
domain followed by a 41-kDa s domain encoded by exons 2-13 of the
Gs gene (1, 2) and, hence, contains most of the functional domains
of Gs including receptor and effector binding sites. Given the
subcellular localization of XLs and its domain structure and in
light of the observation that the C-terminal sequence of the XL domain shows a high homology to the exon 1-encoded portion of Gs (1), which
promotes binding to the 
complex, it is important to determine whether or not XLs functions, like Gs, in signal transduction.
In the present study, we addressed the following questions. First, does
XLs exchange guanine nucleotides and, if so, does this guanine
nucleotide exchange lead to a conformational change of XLs, as has
been reported for Gs (3)? Second, does XLs interact with the
complex? Third, does XLs couple to heptahelical receptors
and, if so, to which ones? And, finally, does XLs activate adenylyl
cyclase, the classical Gs effector?
| |
EXPERIMENTAL PROCEDURES |
Antibodies
The rabbit antiserum RK5 (anti-XL) against the EPAA-repeats in
the XL domain of XLs was that described in the preceding paper (32).
The rabbit antiserum against the C-terminal decapeptide of Gs and
XLs (anti-s C terminus) was the same as described previously
(1).
Plasmids
The plasmid CDM8-XLs, originally called CDM8-XL, contains a
2.6-kilobase insert starting at nucleotide 380 of the originally published sequence (1) and encodes the entire XLs protein sequence
(see correction of translational start (2)) under the control of the
cytomegalovirus promotor.
For construction of CDM8-XLs-wt, the plasmid pGEM-Gs (kindly
provided by Dr. Peter Gierschik, University of Ulm), that encodes the
entire rat Gs protein sequence, was cut with Eco47III and NsiI. The resulting 667-nucleotide fragment, corresponding
to amino acid residues 165-386 of Gs, was cloned into the
Eco47III and NsiI sites of CDM8-XLs, resulting
in a predicted XLs protein sequence containing a leucine at position
519 (corrected translational start (2)) instead of a proline as in
CDM8-XLs (amino acid residue 650 of originally published sequence
(1)). The nucleotide exchange and the ligation sites were confirmed by sequencing.
For construction of CDM8-XLs-Q548L, the plasmid pVL-1393-Gs-Q227L
(kindly provided by Dr. Peter Gierschik, University of Ulm), which
encodes the entire human Gs protein sequence, was cut with
Eco47III and NsiI. The resulting 667-nucleotide
fragment, corresponding to amino acid residues 165-386 of Gs, was
cloned into the Eco47III and NsiI sites of
CDM8-XLs, resulting in a predicted XLs protein sequence carrying
a single point mutation (Gln 
Leu) at amino acid residue 548 (corrected translational start (2); amino acid residue 679 of
originally published sequence (1). The other differences in nucleotide
sequence between human pVL-1393-Gs-Q227L and the rat CDM8-XLs do
not cause any amino acid sequence variation between the two species.)
The nucleotide exchange and the ligation sites were confirmed by
sequencing. The plasmid CDM8-Gs (kindly provided by Yanzhuang
Wang of our laboratory) encodes the entire rat Gs protein sequence
under the control of the cytomegalovirus promotor.
In Vitro Transcription
After linearization by NdeI, 10 µg of each plasmid
(CDM8-Gs and CDM8-XLs-wt) was in vitro transcribed for
4 h at 37 °C in a final volume of 100 µl containing 20 µl
of 5× transcription buffer (MBI Fermentas), 3 µl each of ATP, GTP,
CTP, and UTP (100 mM each), 2 µl of RNase Inhibitor (40 units/µl), 3 µl of T7 RNA polymerase (40 units/µl), and
nuclease-free distilled H2O. Two h after the addition of
the T7 RNA polymerase, another 3 µl of the polymerase were added. Two
µl of a 1:10 dilution of the total in vitro transcription
mixture were used directly for in vitro translation.
In Vitro Translation
Cell-free translation of in vitro transcribed RNAs
was carried out at 30 °C for 1 h using the Promega
nuclease-treated reticulocyte lysate following the manufacturer's
instructions. Briefly, a typical translation mixture contained 35 µl
of the reticulocyte lysate, 7 µl of nuclease-free distilled
H2O, 1 µl of RNase Inhibitor (40 units/µl), 1 µl of
the amino acid mixture without methionine, 4 µl of the
L-[35S]Met/Cys ProMixTM (Amersham
Pharmacia Biotech, 1000 Ci mmol
1), and 2 µl
of the 1:10 diluted total in vitro transcription mixture containing the RNA template. The non-radioactive in vitro
translation for ADP-ribosylation and the reconstitution of
S49cyc membranes was performed with 1 µl of the amino
acid mixture without methionine and 1 µl of the amino acid mixture
without cysteine.
Immunoprecipitation
All steps were performed at 4 °C. In vitro
translated Gs and XLs were mixed with two volumes of
immunoprecipitation buffer (3% Triton X-100, 1.5% sodium
deoxycholate, 0.3% SDS, 450 mM NaCl, 3 mM
EDTA, 3.75 mM phenylmethylsulfonyl fluoride
(PMSF),1 and 30 mM Tris-Cl, pH 8.0) and incubated for 30 min. Insoluble material was removed by centrifugation for 20 min at 14,000 × g, and the supernatant was incubated for 30 min with 50 µl
of a 50% slurry of protein A-Sepharose and centrifuged for 5 min at
800 × g. The supernatant was used for
immunoprecipitation using the antiserum against the common C-terminal
decapeptide of Gs and XLs or the anti-XL antiserum. The samples
were incubated with the antibody overnight followed by the addition of
protein A-Sepharose (50 µl of a 50% slurry) and further incubation
for 2 h. The Sepharose beads were pelleted and washed twice with
buffer A (0.2% Triton X-100, 150 mM NaCl, 2 mM
EDTA, 10 mM Tris-Cl, pH 8.0) and buffer B (0.2% Triton
X-100, 450 mM NaCl, 2 mM EDTA, 10 mM Tris-Cl, pH 8.0) and once with buffer C (0.1% Triton
X-100, 10 mM Tris-Cl, pH 8.0). Immunoprecipitated material
was analyzed by SDS-PAGE and phosphoimaging.
Tryptic Digestion
Tryptic digestion of Gs and XLs was performed as described
previously (4). Briefly, in vitro translated
35S-labeled proteins were incubated for 10 min at 37 °C
in TMED (25 mM Tris-Cl, pH 8.0, 10 mM
MgCl2, 1 mM EDTA, 1 mM DTT) in the absence or presence of 100 µM GTPS and then digested
for 1 h at 30 °C in the presence of various concentrations
(0-0.5 µg/µl) of trypsin (as
L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated; Sigma). Digestion was stopped by the addition of SDS sample buffer immediately followed by boiling of the samples for 5 min at
95 °C.
Cholera Toxin-catalyzed ADP-ribosylation
ADP-ribosylation of in vitro translated XLs was
performed by a modification of the procedure of Audigier (3). In
vitro translated -subunit (20 µl) was mixed with 30 µl of
20 mM HEPES-KOH, pH 7.2, 2 mM
MgCl2, and 1 mM EDTA. After 15 min of
incubation on ice, 0.5 µl of buffer (20 mM Tris-Cl, pH
8.0, 2 mM MgCl2, 1 mM EDTA, 2 mM DTT, and 11 mM CHAPS) either lacking or
containing 43 ng 12 was added to the
mixture, followed by the immediate addition of 17.5 µl of 0.5 M Na3PO4, pH 7.2, 60 mM
thymidine, 5 mM ATP, 0.5 mM GTP, 5 mM MgCl2, and 10 µCi of
32P-NAD+ (800 Ci
mmol1, PerkinElmer Life Sciences).
Cholera toxin-catalyzed ADP-ribosylation was initiated by the addition
of 10 µl of a 0.5 mg/ml solution of cholera toxin A-subunit activated
with 25 mM dithiothreitol for 30 min at 37 °C before
use. After 60 min of incubation at 30 °C, the reaction mixture was
transferred on ice, and 4 volumes of immunoprecipitation buffer (100 mM Tris-HCl, pH 7.5, 100 mM KCl, 5 mM MgCl2, 1% Triton X-100, 1% sodium
deoxycholate, 0.5% SDS, and 1 mM PMSF) were added.
Insoluble material was removed by centrifugation for 15 min at
14,000 × g at 4 °C. The supernatant was incubated
overnight with 20 µl of the rabbit antiserum against the C-terminal
decapeptide of Gs and XLs. Immune complexes were collected using
protein A-Sepharose in PBS and analyzed by SDS-PAGE and autoradiography.
Sedimentation Analysis
Sedimentation analysis using sucrose density gradient
centrifugation was performed as described previously, with minor
changes (5). Briefly, 5 µl of Gs or XLs translation medium was
incubated for 24 h at 0 °C in the absence or presence of 150 ng
of purified unlabeled 12 subunits (kindly
provided by Dr. Christiane Kleuss, Free University of Berlin) in a
buffer containing 50 mM Tris-Cl, pH 8.0, 10 mM
EDTA, 1 mM DTT, and 1 mM GDPS in a final
volume of 100 µl. Samples are loaded on top of linear 5-30% sucrose
density gradients prepared from solutions also containing 20 mM Tris-Cl, pH 8.0, 1 mM EDTA, and 10 mM -mercaptoethanol. Gradients were centrifuged for
18 h in a Beckman SW60 rotor at 55,000 rpm at 4 °C with the
deceleration setting "slow," and 20 fractions were collected per gradient.
Cell Culture and Transfection
PC12 cells were plated on 15-cm dishes and were grown to
subconfluency in Dulbecco's modified Eagle's medium supplemented with
10% horse serum and 5% fetal calf serum at 10% CO2 and
37 °C. For transient transfection, cells harvested from a
subconfluent 15-cm dish after trypsinization were subjected to
electroporation (Bio-Rad Gene Pulser; 960 µF, 300 V) in 0.8 ml of
phosphate-buffered saline containing 45 µg of circular plasmid DNA.
Transfected cells were plated on a 15-cm dish and used 2 days after
transfection, with 10 mM sodium butyrate added during the
last 16 h to increase the expression of the transgene (6).
S49cyc cells were grown in flasks to a density of 1 × 105 - 2 × 106 cells/ml in Dulbecco's
modified Eagle's medium (4.5 mg/ml glucose) supplemented with 10%
fetal calf serum at 5% CO2 and 37 °C.
Membrane Preparations
PC12 Membranes--
A post-nuclear supernatant (PNS) from PC12
cells was prepared as described (7). For the determination of adenylyl
cyclase activity, total membranes were prepared from the PNS by
centrifugation (1 h, 100,000 × g, 4 °C),
resuspended in 20 mM HEPES-KOH, pH 7.2, at 1-2 mg of
protein/ml, and snap-frozen in liquid nitrogen.
S49cyc Membranes--
S49cyc cells
(50 ml, ~106 cells/ml) were pelleted for 5 min at
800 rpm in a Heraeus Megafuge at 4 °C and washed once in ice-cold phosphate-buffered saline containing 0.5 mM PMSF. The cells
were resuspended in 10 ml of homogenization buffer (0.25 M
sucrose, 1 mM EDTA, 1 mM magnesium acetate, 1 mM DTT, 0.5 mM PMSF, and 10 mM
HEPES-KOH, pH 7.4) and pelleted for 5 min at 1600 rpm in a Heraeus
Megafuge. The cells were resuspended in 800 µl of homogenization buffer and homogenized by passage through a 22-gauge needle attached to
a 1-ml syringe followed by 10 passes through a cell cracker (EMBL,
12-µm clearance). The homogenate was centrifuged for 10 min at
900 × g at 4 °C, and the resulting PNS was
centrifuged for 1 h in a Beckman TLA45 rotor at 43,000 rpm at
4 °C. The membrane pellet was resuspended in 10 mM
HEPES-KOH, pH 7.4, 1 mM DTT to a final protein
concentration of 2 mg/ml, snap-frozen in liquid nitrogen, and stored at
80 °C.
Rat Pituitary Membranes--
Adult rats (Wistar strain) were
anesthetized with chloroform and killed by cervical dislocation. The
pituitaries were removed from the skull and transferred into ice-cold
HBS (0.3 M sucrose, 1 mM MgCl2, 1 mM EDTA, 1 mM PMSF, and 10 mM
HEPES-KOH, pH 7.4). The pituitaries were homogenized in 5 volumes of
HBS using a motor-driven glass-Teflon homogenizer at 1000 rpm. The
homogenate was centrifuged for 10 min at 1000 × g at
4 °C. The resulting PNS was centrifuged for 1 h at 100,000 × g at 4 °C. The membrane pellet was resuspended to a
protein concentration of 2 mg/ml in 10 mM Tris-Cl, pH 7.4, snap-frozen in liquid nitrogen, and stored at 80 °C.
Reconstitution of S49cyc Membranes with in
Vitro Translated Gs and XLs
Reconstitution of the S49cyc membranes was
performed as described previously (8). Briefly, 1 volume of the
in vitro translation mixture was mixed with 1 volume of
S49cyc membranes (2 mg of protein/ml) and incubated for
30 min at 30 °C, followed by centrifugation for 30 min in a Beckman
TLA45 rotor at 43,000 rpm at 4 °C. The resulting membrane pellet was
resuspended to a protein concentration of 1 mg/ml in 10 mM
Hepes, pH 7.4, 1 mM DTT. The reconstituted membranes were
kept on ice and immediately used for the adenylyl cyclase assay.
Adenylyl Cyclase Assay
The activity of adenylyl cyclase was determined by the method of
Salomon (9) with minor modifications as follows. The activity of the
S49cyc membranes (20 µg of protein) was assayed in a
final volume of 100 µl containing 100 µM ATP, 10 mM MgCl2, 500 µM cAMP, 10 mM creatine phosphate, 0.5 mg/ml creatine kinase, 1 mM DTT, 25 mM Tris-Cl, pH 8.0, 2 µCi of
[-32P]ATP, and GTP, GTPS, and isoproterenol as
indicated in the figure legend. Reactions were carried out for 30 min
at 30 °C.
The adenylyl cyclase activity in PC12 membranes was assayed in the
presence of 165 µM ATP, 5 mM
MgCl2, 10 mM creatine phosphate, 0.5 mg/ml
creatine kinase, 0.5 mM DTT, 3 µM GTP, 1 mg/ml bovine serum albumin, 3 mM HEPES-KOH, pH 7.2, and 1 µCi [-32P]ATP. Reactions were carried out for 30 min
at 30 °C.
Photoaffinity Labeling of -Subunits with
[32P]GTP-azidoanilide (AA-GTP)
Photoaffinity Labeling Using Pituitary Membranes--
AA-GTP
labeling was performed essentially as described previously (10).
Briefly, membranes (100 µg of protein) were pelleted (10 min,
14,000 × g) and resuspended in 30 µl of incubation
buffer (0.2 mM EDTA, 10 mM MgCl2,
60 mM NaCl, 100 mM HEPES-KOH, pH 7.5, 2 mM benzamidine, and 2-200 µM GDP). Samples
then received 10 µl without or with the indicated receptor agonists
(see text under "Results" and the legend to Fig. 8), were
incubated for 3 min at 30 °C, received 20 µl of
[-32P]AA-GTP (1.4 × 106 cpm/µl),
and were further incubated for 1-10 min at 30 °C. Samples were
transferred on ice and centrifuged for 5 min at 14,000 × g at 4 °C. The membranes were rapidly resuspended in 60 µl of labeling buffer (0.1 mM EDTA, 5 mM
MgCl2, 30 mM NaCl, 50 mM HEPES-KOH, pH 7.5, 1 mM benzamidine, 2 mM glutathione) and
immediately irradiated at 265 nm for 10-15 s at 4 °C. The samples
were centrifuged as above, and the membranes were solubilized in 160 µl of immunoprecipitation buffer (1% Nonidet P40, 1% sodium
deoxycholate, 0.5% SDS, 150 mM NaCl, 1 mM DTT,
1 mM EDTA, 0.2 mM PMSF, 10 µg/ml aprotinin, and 10 mM Tris-Cl, pH 7.4) for 10-15 min on ice. Insoluble
material was removed by centrifugation as above, and the supernatant
was used for immunoprecipitation using an antiserum (5 µl) against the common C-terminal decapeptide of Gs and XLs. The samples were
incubated with the antibody overnight at 4 °C followed by addition
of protein A-Sepharose (60 µl of a 10% slurry) and further incubation for 2 h at 4 °C. The Sepharose beads were pelleted and washed twice with buffer A (1% Nonidet P40, 0.5% SDS, 600 mM NaCl, 50 mM Tris-Cl, pH 7.4) and once with
buffer B (300 mM NaCl, 10 mM EDTA, 100 mM Tris-Cl, pH 7.4). Immunoprecipitated material was
analyzed by SDS-PAGE and phosphoimaging.
Photoaffinity Labeling Using PC12 Cell Membranes--
Membranes
were washed once with 50 mM Hepes-NaOH, pH 7.4, and the
membrane proteins (50 µg) were subjected to photoaffinity labeling
with [-32P]AA-GTP as described above in the presence
of GTP or ATP as indicated in the figure and with the following
modifications. (i) The incubation buffer was 1 mM
MgCl2 and 50 mM NaCl, and GDP was omitted; (ii) no receptor agonist was added; (iii) incubation was for 10 min.
SDS-PAGE
SDS-PAGE (7.5 or 10% gels) and immunoblotting was performed
according to standard procedures. Dried gels were either
autoradiographed followed by densitometric scanning or exposed to
phosphoimager plates, and the intensity of the bands was determined
using a Fuji phosphoimager and MacBAS software.
Identification of Mouse XLs--
The IMAGE cDNA clone
1499201 was identified by BLAST searching with the rat and human
XLs sequences. It was subsequently obtained from Research Genetics
and completely sequenced. The clone encodes the full-length mouse
XLs protein. The sequence was submitted to GenBankTM
(accession number AF116268).
| |
RESULTS |
Construction of an XLs cDNA for in Vitro
Transcription/Translation--
The originally published sequence of
XLs contains a point mutation in the s-part (1) that leads to a
single amino acid exchange (L P) at position 519 (corrected
translational start (2); amino acid residue 650 of originally published
sequence (1)). This point mutation was not found in XLs mRNAs
isolated from rat pituitary or PC12 cells as revealed by reverse
transcription-polymerase chain reaction analysis (data not shown),
suggesting that it reflects an aberration from the physiological XLs
sequence introduced in the course of construction of the cDNA
library from which the original XLs cDNA was isolated (1). We
therefore constructed an XLs cDNA, referred to as XLs-wt, in
which this point mutation in the s-part of XLs was reversed to
the physiological sequence. This cDNA was used in the in
vitro transcription/translation experiments performed in this study.
In Vitro Translation of XLs and Gs--
In vitro
translation in the rabbit reticulocyte lysate of a Gs mRNA,
obtained by in vitro transcription of a Gs cDNA, gave rise to a 46-kDa labeled band (Fig. 1,
lane 1). Translation of the in vitro transcribed
XLs-wt mRNA generated a major band of Mr
94,000 and additional bands of lower apparent molecular weight (Fig. 1,
lane 4). Immunoprecipitation of the translation products using an antiserum against the C-terminal decapeptide of Gs showed that XLs (Fig. 1, lane 6) contains the same C-terminal
epitope as Gs (Fig. 1, lane 3). Upon immunoprecipitation
using an antiserum against the EPAA repeats in the N-terminal region of
the XL domain (1), referred to as anti-XL antibody, only XLs
mRNA-derived translation products of Mr 
94,000 (Fig. 1, lane 5), but no Gs mRNA-derived
translation products (Fig. 1, lane 2), were obtained. This
showed that (i) the anti-XL antibody was indeed specific for the XL
domain of XLs, and (ii) the major XLs mRNA-derived translation products of Mr < 94,000 (Fig. 1,
lanes 4 and 5) were truncated in the s domain.
In vitro translated Gs and XLs, as shown in Fig. 1,
lanes 1 and 4, were used in the subsequent experiments.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 1.
Characterization of the translation products
obtained from Gs and
XLs mRNAs. Transcription products
were generated from linearized plasmids (lanes 1-3,
CDM8-Gs; lanes 4-6, CDM8-XLs-wt) and translated in
the reticulocyte lysate in the presence of
[35S]methionine/cysteine. The 35S-labeled
translation products were subjected to SDS-PAGE either before (total)
or after immunoprecipitation using either the anti-XL antiserum or the
antiserum against the common C-terminal decapeptide of Gs and XLs
(anti-s C-term.) and visualized by
phosphoimaging. Arrow, full-length XLs;
arrowhead, Gs. The results shown are representative of
three independent experiments.
|
|
Trypsin Resistance of the s Domain of XLs in the GTP-bound
State--
The activation of G subunits, i.e. the
replacement of GDP by GTP, is associated with a conformational change
that is reflected in their differential sensitivity to degradation by
trypsin (3). As shown in Fig. 2, the
presence of GTPS affected the tryptic digestion of in
vitro translated XLs in the same way as it was described
previously for Gs (3). In the absence of GTPS, XLs, like
Gs, was fully degraded by increasing concentrations of trypsin. In
contrast, for both Gs and XLs in the presence of GTPS,
37-35-kDa fragments were protected from tryptic digestion. This
strongly suggests that (i) XLs binds GTPS and (ii) upon GTPS
binding, the s domain undergoes the same conformational change as
Gs.
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 2.
Differential sensitivity of
XLs to digestion by trypsin in the absence or
presence of GTPS. Transcription products
were generated from linearized plasmids (CDM8-Gs or CDM8-XLs-wt)
and translated in the reticulocyte lysate in the presence of
[35S]methionine/cysteine. The translation products,
either 35S-labeled Gs (A) or XLs (B), were incubated
for 10 min at 37 °C in the absence (GTPS)
or presence (+GTPS) of 100 µM
GTPS and then digested for 60 min at 30 °C with the indicated
concentrations of trypsin. The 35S-labeled products were
subjected to SDS-PAGE and visualized by phosphoimaging.
Arrow, full-length XLs; arrowhead, full-length
Gs; triangles show the position of the tryptic fragments
(37-35 kDa) that are protected from further digestion in the presence
of GTPS. The results shown are representative of three independent
experiments.
|
|
Formation of an XLs- Heterotrimer--
In XLs, the
exon 1-encoded portion of Gs is replaced by the XL domain (1). The
exon 1-encoded portion of Gs is involved in the interaction of Gs
with the complex (5, 11). Interestingly, the C-terminal region
of the XL domain shows a high degree of homology to the exon 1-encoded
portion of Gs and other types of G subunits (1) (Fig.
3, A and B). In
particular, the residues that are known to directly contact the
dimer (11, 12) are conserved not only between rat, mouse, and human
XLs (Fig. 3B) but also between XLs and the various
types of G subunits (1). It is therefore possible that XLs, like
Gs, binds to the complex. To investigate this issue, we used
(i) ADP-ribosylation by cholera toxin to detect even weak or transient
XLs- interactions and (ii) sucrose density gradient
centrifugation to search for stable heterotrimer formation.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 3.
Sequence comparison of two highly conserved
regions in the XL domain of human, mouse, and rat
XLs. A, domain organization of
rat XLs (1). EPAA, region containing the EPAA repeats;
ARAA, region containing the AARA repeats; P,
proline-rich region; C, cysteine-rich region; , region
containing the putative binding site; numbers refer
to the corrected translational start of XLs (2). B and
C, comparison of the C-terminal amino acid sequence of the
XL domain of human (h, Ref. 25), mouse (m, Ref.
31 and GenBankTM accession number AF116268), and rat
(r, Refs. 1 and 2) XLs with the corresponding N-terminal
sequence of rat Gs (B) and of the proline-rich region of
XLs across the three species (C). Boxes indicate
conserved residues; asterisks in B indicate
residues that are known to directly contact the complex (11,
30).
|
|
Effect of Subunits on the ADP-ribosylation of XLs by
Cholera Toxin--
The ability of activated cholera toxin to catalyze
the ADP-ribosylation of purified (i.e. monomeric), native
Gs (13) as well as of in vitro translated Gs (3) is
very poor. However, in either case, the addition of subunits
before the addition of activated cholera toxin greatly increases the
extent of ADP-ribosylation (3, 13). ADP-ribosylation by cholera toxin
is an irreversible modification and is therefore a very sensitive
method for detecting even weak or transient interactions of in
vitro translated Gs and XLs with subunits. In the
absence of subunits, in vitro translated XLs was
indeed found to be a poor substrate for cholera toxin catalyzed
ADP-ribosylation (Fig. 4, lane
4), whereas the addition of exogenous subunits resulted in
a >2.2-fold increase in the labeling of both the in vitro
translated XLs (Fig. 4, lane 5) as well as the endogenous
Gs present in the reticulocyte lysate (Fig. 4, lanes 2 and 5). The labeling of XLs and Gs in the absence of
added subunits (Fig. 4, lanes 1 and 4) is
most likely due to the presence of some endogenous in the
reticulocyte lysate (14).
View larger version (52K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of
subunits on the ADP-ribosylation of XLs
by cholera toxin. Reticulocyte lysate either lacking (lanes
1 and 2) or containing (lanes 3-5) in
vitro translated XLs was incubated for 60 min at 30 °C with
32P-NAD+ in the absence () or presence (+) of
purified 12 subunits and cholera toxin
(Ctx) as indicated, followed by immunoprecipitation with the
antiserum against the C-terminal decapeptide of Gs and XLs.
Immune complexes were analyzed by SDS-PAGE and autoradiography. The
results shown are representative of three independent
experiments.
|
|
Effect of Subunits on the Sedimentation Behavior of
XLs--
To look for stable heterotrimer formation, we examined the
effects of purified subunits on the sedimentation behavior of in vitro translated XLs and, for comparison, Gs, using
sucrose density gradients. In the absence of subunits, XLs
(Fig. 5B, open
circles) and Gs (Fig. 5A, open circles)
sedimented at a rate corresponding to a calculated molecular mass of 78 and 46 kDa, respectively, and hence a monomeric state, as revealed by comparison to the 68-kDa hemoglobin tetramer (Fig. 5, bars).
In the presence of subunits, XLs (Fig. 5B,
filled circles) and confirming previous results (5), Gs
(Fig. 5A, filled circles), sedimented at a faster
rate. However, in the presence of subunits, XLs (Fig.
5B, filled circles) sedimented at a slower rate
than Gs (Fig. 5A, filled circles), although
the total molecular mass of an XLs- heterotrimer is greater
than that of the Gs- heterotrimer. This suggests that XLs,
upon contact with the dimer, undergoes a conformational change,
which alters its sedimentation behavior.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5.
Effect of
subunits on the sedimentation behavior of
XLs. Transcription products were
generated from linearized plasmids (CDM8-Gs or CDM8-XLs-wt) and
translated in the reticulocyte lysate in the presence of
[35S]methionine/cysteine. The translation products,
either 35S-labeled Gs (A) or XLs
(B), were incubated in the presence of GDPS and in the
absence (open circles) or presence (filled
circles) of purified unlabeled subunits, followed by
centrifugation on a linear 5-30% sucrose gradient. An aliquot of each
fraction (fraction 1 = top of gradient) was subjected
to SDS-PAGE, and the 35S-labeled Gs (A) or
XLs (B) was visualized by phosphoimaging and quantified.
The amount of 35S-labeled Gs (A) or XLs
(B) recovered in each fraction is expressed as percent of
the total (sum of the values of all fractions). The bar
indicates the position of the hemoglobin tetramer (68 kDa), derived
from the rabbit reticulocyte lysate, which served as an internal
molecular mass standard. The results shown are representative of four
independent experiments.
|
|
Activation of Adenylyl Cyclase by a GTPase-deficient Mutant of
XLs--
To determine whether XLs is capable of activating
adenylyl cyclase, we constructed a GTPase-deficient mutant of XLs,
XLsQ548L. This mutation corresponds to the Q227L mutation in Gs,
which is analogous to the Q61L mutant of p21ras. In Gs, this
mutation leads to a 100-fold reduction in the rate constant of GTP
hydrolysis (15). Hence, XLsQ548L should be constitutively activated.
Immunoblotting of PC12 cell membranes using the antibody against the
common C-terminal decapeptide of Gs and XLs (Fig.
6, bottom panel) indicated
that transient transfection with the cDNA for XLs or
XLs-Q548L substantially increased the amount of XLs above the
endogenous level. Membranes of transfected and untransfected PC12 cells
were then analyzed for adenylyl cyclase activity (Fig. 6, top
panel). Cells transfected with the activated form of XLs, XLs-Q548L, showed a massive increase in adenylyl cyclase activity as
compared with wild type or mock-transfected cells. Transfection of
XLs resulted in only a small increase in adenylyl cyclase activity,
presumably because XLs was predominantly in the GDP-bound state and
therefore inactive toward adenylyl cyclase. The addition of forskolin
to the membranes from wild type, mock-transfected, and
XLs-transfected PC12 cells increased adenylyl cyclase activity to
the level observed with membranes from XLs-Q548L-transfected cells
(data not shown), showing that adenylyl cyclase in the former membranes
could be activated. When HeLa (rather than PC12) cells were transiently
transfected with XLs-Q548L, they also showed an increase (4-fold) in
adenylyl cyclase activity as compared with mock-transfected cells (data
not shown).
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 6.
Activation of adenylyl cyclase by a
GTPase-deficient mutant of XLs
(XLs-Q548L). Wild type PC12 cells
(WT) or PC12 cells transfected with the CDM8 vector without
insert (mock), with CDM8-XLs
(XLs), or with CDM8-XLs-Q548L
(Q548L) were used. Top, adenylyl cyclase activity
in total membranes (15 µg of protein/assay). The mean of duplicate
determinations is shown. Error bars indicate the variation
of the individual values from the mean; for some conditions, these are
too small to be seen. Bottom, immunoblot of a similar
membrane preparation (30 µg of protein/lane) using the
antiserum against the common C-terminal decapeptide of Gs and
XLs. The results shown are representative of three independent
experiments.
|
|
Receptor Activation of Gs, but Not XLs, in Reconstituted
S49cyc Membranes--
S49cyc cells are
deficient in Gs (16-18) as well as XLs, as revealed by
immunoblotting (data not shown), but still express the 2-adrenergic
receptor, subunits, and the adenylyl cyclase. Given that XLs
can activate adenylyl cyclase (Fig. 6), we used S49cyc
cells to study the signal transduction properties of XLs. As reported previously (5, 8), adenylyl cyclase activity of S49cyc membranes could be stimulated upon the addition of
isoproterenol, a 2-adrenergic receptor agonist, when the membranes
had been reconstituted with in vitro translated Gs (Fig.
7A). Compared with the
addition of GTP alone, adenylyl cyclase activation upon receptor
stimulation was increased to about half of the maximal value obtained
in the presence of GTPS (Fig. 7A).
View larger version (21K):
[in this window]
[in a new window]
|
Fig. 7.
Differential effects of
Gs and XLs on the
receptor-mediated adenylyl cyclase activation in
S49cyc membranes. Transcription products
were generated from linearized plasmids (CDM8-Gs or CDM8-XLs-wt)
and translated in the reticulocyte lysate. S49cyc
membranes were reconstituted either with reticulocyte lysate alone or
with reticulocyte lysate containing in vitro translated
Gs (A) or XLs (B). Reconstituted membranes
were incubated in the presence of 100 µM GTP, 10 µM ()isoproterenol (isoprot), or 10 µM GTPS, as indicated. cAMP formation was determined
in triplicate. Note that the mean values obtained with membranes that
had been reconstituted with the reticulocyte lysate only (0.86 ± 0.28 pmol of cAMP mg1 protein
min1) were subtracted from the mean values
obtained with membranes that had been reconstituted with reticulocyte
lysate containing in vitro translated Gs or XLs. The
mean value of the GTPS condition (A, 8.8 ± 0.2 pmol
of cAMP mg1 protein
min1; B, 6.2 ± 0.3 pmol of
cAMP mg1 protein
min1) is arbitrarily set to 100, and the
other mean values are expressed relative to this. Bars
indicate the error of the final value. The results shown are
representative of three independent experiments.
|
|
Because Gs and XLs contain the same C-terminal domain and may
therefore couple to the same type of receptor (19-21), we investigated whether XLs, like Gs, was able to restore receptor stimulation of
adenylyl cyclase activation in reconstituted S49cyc
membranes. In contrast to Gs (Fig. 7A), XLs mediated
only a small, barely significant increase in adenylyl cyclase
activation upon receptor stimulation, as compared with the addition of
GTP or GTPS alone (Fig. 7B). We therefore conclude that,
in this in vitro system, XLs couples to the
2-adrenergic receptor much less efficiently than Gs, if at all.
Consistent with these in vitro findings, we observed, in
comparison with untransfected PC12 cells, an increased adenylyl cyclase
activity upon the addition of 100 µM CGS 21680 (adenosine 2A receptor agonist) in membranes of Gs-transfected, but
not XLs-transfected, PC12 cells (data not shown).
Receptor Activation of Gs, but not XLs, in Pituitary
Membranes--
Given that XLs couples only weakly to the 2
adrenergic receptor in reconstituted S49cyc membranes, we
investigated whether or not XLs is capable of coupling to a G
protein-coupled receptor in vivo. For this purpose, we used
photoaffinity labeling of G protein subunits with AA-GTP (10). When
combined with immunoprecipitation using antibodies specific for a given
G protein subunit, such photoaffinity labeling is a powerful tool
to identify the specific G protein subunit activated by a given receptor.
First, we investigated whether XLs binds AA-GTP with the same
affinity as Gs. PC12 cell membranes were incubated in the presence
of AA-GTP with increasing concentrations of either unlabeled GTP or
ATP. As shown in Fig. 8A,
binding of AA-GTP to XLs could be competed with increasing
concentrations of GTP but not ATP. The comparison of the competition
profiles obtained for XLs and Gs shows that XLs binds to
AA-GTP with virtually the same affinity as Gs.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 8.
Differential effect of PACAP receptor
stimulation on GTP binding to XLs and
Gs. A, PC12 cell membranes
were incubated for 10 min with [-32P]GTP-azidoanilide
in the presence of the indicated concentrations of unlabeled GTP or
ATP, followed by photocross-linking. XLs and Gs were
immunoprecipitated from the solubilized membranes using an antiserum
against the common C-terminal decapeptide, and immunecomplexes were
analyzed by SDS-PAGE followed by autoradiography. Top,
XLs region of the autoradiogram. Bottom, quantitation of
the 32P-labeled XLs (filled circles) and
Gs (open circles) by densitometric scanning;
a.u., arbitrary units. To facilitate comparison with XLs,
the Gs values (sum of two Gs bands) were divided by 2.5 (GTP) and
2.2 (ATP). B, rat pituitary membranes were incubated for the
indicated times with [-32P]GTP-azidoanilide in the
presence of 10 µM GDP in the absence () or presence (+)
of 1 µM PACAP, followed by photocross-linking. XLs and
Gs were immunoprecipitated from the solubilized membranes using an
antiserum against the common C-terminal decapeptide, and immune
complexes were analyzed by SDS-PAGE and phosphoimaging
(top). Bottom, quantitation of the
32P-labeled XLs and Gs. For each period of
incubation, the value obtained in the absence () of PACAP was
arbitrarily set to 100, and the value obtained in the presence (+) of
PACAP was expressed relative to this. The results shown are
representative of five independent experiments.
|
|
Second, we determined whether AA-GTP binding to XLs could be
enhanced by the activation of known Gs-coupled receptors. For this
purpose, we used rat pituitary membranes because the expression level
of XLs was highest in this tissue (32). Fig. 7B shows that the receptor for pituitary adenylyl cyclase-activating polypeptide (PACAP) was capable of activating Gs, as indicated by an increased incorporation of AA-GTP in the presence of PACAP but not XLs.
Other Gs-coupled receptors in the pituitary, i.e. that
for vasoactive intestinal polypeptide (VIP, used at 1-10
µM) and corticotropin releasing factor (CRF, used at 10 µM) as well as the adenosine 2A receptor of PC12 cells
(22, 23) (stimulated by 10 µM CGS 21680), were
also found to activate Gs but not XLs (data not shown). Changing
the experimental conditions, e.g. by the addition of various
concentrations of GDP to suppress the basal rate of guanine nucleotide
exchange, or using different labeling times also did not reveal any
significant receptor stimulation of guanine nucleotide exchange on
XLs (data not shown).
| |
DISCUSSION |
Our study shows that XLs shares many, but not all, functional
properties of Gs. XLs (i) forms a heterotrimer with
subunits, (ii) binds GTP and undergoes a conformational change upon GTP binding, and (iii) activates, when in the GTP state, adenylyl cyclase.
However, XLs does not appear to be activated by known Gs-coupled receptors.
Binding to the Dimer--
We used two methods to
demonstrate the ability of XLs to interact with subunits: (i)
ADP-ribosylation by cholera toxin, which in the case of Gs is
greatly promoted by its heterotrimeric state (3, 13, 24), and (ii)
sucrose density gradient centrifugation. The addition of
subunits to in vitro translated XLs increased its
ADP-ribosylation as well as its sedimentation rate in sucrose density
gradients, indicating that XLs forms heterotrimers with
subunits. Remarkably, however, comparison of the sedimentation behavior
of Gs and XLs in the presence of dimers revealed that
XLs sedimented more slowly than Gs, although the molecular mass
of an XLs- heterotrimer is greater than that of a Gs- heterotrimer. The observations that (i) heterotrimeric rather than
monomeric Gs (3, 13, 24) and XLs (Fig. 4) is a substrate for
cholera toxin-catalyzed ADP-ribosylation and (ii) XLs, like Gs,
undergoes ADP-ribosylation by cholera toxin in vivo (1) strongly suggest that XLs exists in the heterotrimeric state in vivo. It is therefore likely that the sedimentation of
XLs observed in the presence of subunits reflected that of
heterotrimeric XLs rather than that of a monomeric XLs that had
adopted a different conformation upon transient contact with the
dimer. This in turn suggests that the heterotrimerization of XLs
with the dimer is associated with a stable conformational change
of XLs toward a more rod-like state, resulting in a slower
sedimentation of the XLs- heterotrimer than the Gs-
heterotrimer. The proline-rich region of the XL domain of XLs (1)
(Fig. 3, A and C) could serve as a hinge for this
conformational change. It is worth noting that this region, like the
C-terminal region of the XL domain implicated in binding (1)
(Fig. 3, A and B) is more highly conserved in
XLs of various species (Fig. 3C) than other regions of
the XL domain (for a comparison of human and rat XLs, see Hayward
et al. (25)).
The ability of XLs to form a heterotrimeric complex with
subunits also has implications for the observations reported in the
preceding paper (32) that immunoreactive and ADP-ribosylatable XLs
show a distinct distribution upon subcellular fractionation, whereas
this is not the case for Gs. Specifically, the subpopulation of
XLs molecules that were poorly, if at all, ADP-ribosylated by
cholera toxin were preferentially recovered in fractions containing plasma membrane, whereas the subpopulation that was well
ADP-ribosylated was preferentially recovered in fractions containing
Golgi membranes and certain subdomains of the plasma membrane (see
preceding paper (32)). This raises the possibility of an interplay
between binding to XLs and its subcellular loca
-Subunit
XL
,
**,
TOP
ABSTRACT
INTRODUCTION
