| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
q/11 Protein to the Membrane and Induces Its Specific Internalization Independently of Receptor- G Protein Coupling in HEK-293 Cells
Institut national de la santé et de la recherche médicale U36 Collège de France 75005 Paris, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
q/11 protein,
identified by indirect immunofluorescence, and the development of a
specific software for digital image analysis together allow examination
and quantification of the cellular distribution of these
proteins before and after the binding of different agonist or
antagonist ligands. These observations result in several conclusions:
1) Expression of increasing amounts of the AT1A
receptor at the cell surface is associated with a progressive
recruitment of the cytosolic Gq/11 protein at the membrane; 2)
Internalization of the EGFP-tagged AT1A induced
by peptide ligands but not nonpeptide ligands is accompanied by a
Gq/11 protein intracellular translocation, which presents a similar
kinetic pattern but occurs predominantly in a different compartment;
and 3) This Gq/11 protein cellular translocation is dependent on
receptor internalization process, but not G protein coupling and signal
transduction mechanisms, as assessed by pharmacological data using
agonists and antagonists and the characterization of
AT1A receptor mutants
(D74N and 
329) for which the coupling and
internalization functions are modified. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The transduction machinery for GPCR is described classically as a
three-pieces system comprising 1) a receptor, which is an integral
transmembrane protein with seven hydrophobic helices, 2) a
transducer, corresponding to a heterotrimeric GTP-binding protein and
associating a ß
-complex, which is anchored in membranes via a
prenylated cystein and an -subunit that may be associated to the
membrane via myristate or palmitate moities (2), and 3) an effector,
which can be a membrane-bound enzyme or an ion channel.
Several cellular events follow the binding of Ang II to the
AT1A receptor, such as activation of the
signaling pathways, receptor-ligand internalization, and receptor
phosphorylation and desensitization. These processes involved multiple
intracellular receptor-protein interactions. On one hand, the Ang II
signals inside the cell via receptor, Gq/11protein, phospholipase C
(ß-PLC) interactions, which activate inositol phosphate/calcium
signaling (for review see Ref. 1), but perhaps also via a Jak-STAT
pathway that includes AT1/Jak2 interactions (3).
On the other hand and like most other GPCRs, the
AT1 receptor is internalized after ligand
binding. The kinetics, pharmacology, and molecular determinants (third
intracellular loop and C-terminal part of the receptor) of this
internalization process have been extensively investigated by several
groups and our laboratory (4, 5, 6, 7). However, the molecular mechanism of
this internalization process is unclear and may be not only a coated
pit-dependent or a lipid raft/caveolae-dependent mechanism (8).
A clear picture of the parallel subcellular distribution and of simultaneous trafficking for the Ang II AT1 receptor and its interacting proteins, e.g. G proteins, is not yet available due to technical limitations in the immunohistochemical, optical tools, and image analysis softwares.
In the present study, a functional Ang II AT1A
receptor tagged with the enhanced green fluorescent protein (EGFP) was
stably expressed in HEK-293 cells. Simultaneous morphological analysis
of the fluorescent receptor and its coupled endogenous Gq/11 protein
identified with specific antibodies, in the absence or presence of
specific ligands using confocal microscopy, enabled us to make two new
and interesting observations:
1. The cell surface localization of the endogenous Gq/11 protein
increases with the cell surface expression of the
AT1A receptor.
2. After Ang II binding, both the receptor and its coupled G protein are internalized, but in different intracellular compartments. This phenomenon is dependent on receptor internalization but not on receptor activation and signal transduction.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). Stimulation of the tagged receptor
with Ang II induces a 10-fold increase in total inositol phosphate (IP)
production (EC50 = 2.34 ± 0.54
nM) as reported for the wild-type receptor expressed in
Chinese hamster ovary (CHO) cells (EC50 =
0.75 ± 0.09 nM) (9). In addition, a more integrated
assay of the signaling pathway activation was performed, involving the
activation of a reporter gene under the control of a promoter regulated
by the PLC-protein kinase C (PKC) pathway (see Materials and
Methods). The wild-type AT1A and
AT1A-EGFP receptors, transiently transfected in
HEK-293 cells with the reporter construct, stimulated the reporter gene
expression to similar level and sensitivity (EC50
= 0.97 nM for the AT1A
receptor; EC50 = 0.50 nM
for the AT1A-EGFP receptor) (Fig. 1).
|
|
). The glycosylated form of
AT1A receptor corresponds to a diffuse band from
70 to 90 kDa (Fig. 2, lane II, zone 2), a value that is more or less in
agreement with the expected molecular mass of this glycoprotein
(41 kDa for AT1A + 27 kDa for EGFP + 3 N-linked
glycosylated chains) and with the few observations of its molecular
mass by SDS-PAGE in the literature (60–65 kDa for Ref. 10 and 50–90
kDa for Ref. 11). This broad band is specific since no signal was
observed in the HEK-AT1A cell extracts. Moreover,
and as observed by others in the literature, another band with a higher
molecular mass (120 kDa and more) was visible (Fig. 2, lane II, zone 1)
and may correspond to either a homodimer or a heterocomplex with
another protein. After N-deglycosylation with peptide N-glycosidase F
(PNGase F), the receptor presents a reduced molecular mass to
60–80 kDa (Fig. 2, lane III, zone 3), which confirms its
N-glycosylation. The similar heterogenous pattern of the receptor after
deglycosylation probably reflects the high sensitivity of the receptor
to cellular proteases.
|
) or for a typical
HEK-AT1A-EGFP cell (Fig. 3B). No fluorescence was
detected in nontransfected HEK-293 cells (data not shown).
Internalization was studied by adding the agonist Ang II (100
nM) at 4 C and incubating the cells at 37 C for 20 min. In
these conditions, fluorescence at the plasma membrane completely
disappeared, and the cytoplasm contained a multitude of fluorescent
vesicles as shown by a large confocal view (Fig. 3C) or a typical
AT1A-EGFP-internalized cell (Fig. 3D). Optical
sectioning of the cells by 1 µm thick serial confocal slices
confirmed the presence of fluorescent internalization vesicles since
their average size was less than 1 µm and they were not connected to
the plasma membrane (data not shown). This Ang IIinduced
internalization is specific because it was abolished by acid wash of
Ang II before incubation at 37 C.
|
).
Within the cell population and in basal conditions, there is an
extensive heterogeneity in the AT1A-EGFP
fluorescence intensity at the cell surface from one cell to another
(see Fig. 3A), which results in large variations of the S'/C' ratio
from 0.39 (low level) to 20.75 (high level). Further investigations
suggested that these variations were related to the cell cycle and that
this heterogenity is due to an absence of growth synchronization in the
cell preparation. Thus, to analyze receptor internalization more
accurately during this study, we analyzed only cells (representing the
large majority of the cellular population, see Fig. 3A) with similar
levels of receptor expression corresponding to a total mean
fluorescence of 6 to 12 in the absence of agonist (total mean
fluorescence being a fluorescence index derived from the addition of S,
C, and N values (for details see Materials and Methods and
Table 2).
|
and Table 2), without any change in total cellular
fluorescence. In addition, this method was used to quantify both the
kinetics and dose dependence of internalization. Ang II clearly induced
a dose-dependent internalization of the receptor (Fig. 3E). The
ED50 for Ang II-induced internalization was
0.31 ± 0.03 nM, an Ang II concentration similar to
the EC50 for IP activation and the dissociation
constant (Kd) of the AT1A
receptor for Ang II. Furthermore, the kinetic profiles of
AT1A-EGFP receptor internalization induced by Ang
II were similar when measured either by the S'/C' ratio using the
morphological quantification method or by the acid wash biochemical
technique (data not shown). In addition, no difference in the
internalization kinetics was observed between the wild-type
AT1A receptor and the
AT1A-EGFP receptor measured by the acid wash
biochemical method (data not shown). Altogether, these data indicate that the AT1A-EGFP receptor is internalized like the wild-type receptor after Ang II binding. Moreover, this internalization can be clearly and precisely monitored by morphological analysis on individual cells or a cellular population.
Receptor Internalization Is Dependent on the Peptide Structure but
Not on the Agonist Properties of the Ligand
The ability of large variety of AT1A
receptor ligands to induce the internalization was analyzed next. These
molecules are either natural metabolites of Ang II (Ang I, Ang III, Ang
IV), or peptide analogs of Ang II with either agonist
([Sar1] Ang II) or antagonist properties
([Sar1,Ile8] Ang II,
[Sar1,Ala8] Ang II,
[Sar1,Thr8] Ang II).
Others ligands are pseudopeptide (CGP42112A) or nonpeptide compounds
with either agonist (L162, 313) or antagonist properties (Losartan).
All these ligands were used at concentrations approximately 100-fold
greater than their reported Kd, and
internalization was analyzed as described in the previous section.
Based on the Ang II dose-dependent internalization curve (Fig. 3E),
when the S'/C' ratio is higher than 3 there is no internalization of
the receptor; when the S'/C' ratio is below 1.5 there is
internalization of the receptor. Thus, all the peptide or pseudopeptide
(CGP42112A) ligands induce internalization of the
AT1A-EGFP receptor independently of their agonist
or antagonist status (Table 3), and
nonpeptide ligands do not induce this process, again independently of
their agonist or an- tagonist status. These observations lead us
to conclude that occupancy of the peptide binding site of the
AT1A receptor, but not of the nonpeptide binding
site which is known to be different (12, 13, 14), is the key event that
triggers internalization.
|
q/11 Protein Is Colocalized with the
AT1A-EGFP Receptor at the Cell Surfaceq/11 protein, it was interesting to analyze the distribution of
the endogenous G protein parallel to the recombinant
AT1A-EGFP receptor in
HEK-AT1A-EGFP cells. The Gq/11 protein was
detected in these cells in Western blots with commercially available
antibodies as a double band around 47 kDa (data not shown) or by
immunofluorescence. In HEK-AT1A-EGFP cells,
intense labeling indicates the presence of endogenous Gq/11 protein
both in the cytoplasm and at the cell surface (Fig. 4B, panel b). This immunolabeling is
specific since both omission of the primary antibody and the addition
of an excess of immunoreactive peptide abolished fluorescence (data not
shown).
|
q/11 protein (S'/C'
0.63 ± 0.05) whereas this ratio was almost twice higher (S'/C'
0.89 ± 0.07) in a comparable population of cells transfected with
the AT1A-EGFP receptor. This ratio increases
again (S'/C' 1.31 ± 0.16) when the population of
HEK-AT1A-EGFP cells is selected for a total EGFP
fluorescence ranking between 6 and 12. Altogether, these results
suggest that the membrane expression of the AT1A
receptor may be a determinant for the cell surface localization of the
Gq/11 protein. This hypothesis was investigated further using the
heterogeneous expression of the AT1A-EGFP
receptor in HEK cells and by measuring the S'/C' ratios for both
AT1A-EGFP and Gq/11 fluorescence in about 200
individual cells expressing different levels of the
AT1A-EGFP receptor (S'/C' ratio from 0.39 to
20.75; see above). The positive correlation between the level of cell
surface expression of the AT1A receptor and the
fraction of membrane-bound Gq/11 protein (Fig. 4A)
(r = 0.394, P < 0.0001) is in favor of
our hypothesis. This correlation is also illustrated by two
representative examples of overall confocal views (Fig. 4B, panels
a–d) or single cell images (Fig. 4B, panels e-h) showing a majority of
cells (but not all) with parallel levels of
AT1A-EGFP receptor and Gq/11 protein at the
cell surface. Some cells present high surface levels of both
AT1A-EGFP (Fig. 4B, panels a1, c2, and e) and
membrane-bound Gq/11 protein (Fig. 4B, panels b1, d2, and f) or a
low level of the receptor (Fig. 4B, panels a3, c4, and g) and
membrane-bound Gq/11 protein (Fig. 4B, panels b3, d4, and h). In
addition, the S'/C' ratio for Gq/11-associated fluorescence was also
higher on a whole population of cells stably expressing either the
wild-type AT1A receptor or another
Gq/11-coupled receptor, the V1b vasopressin receptor (Fig. 4C).
All these data suggest that the membrane recruitment of the Gq/11 is
at least partly dependent on the level of membrane expression of its
cognate receptor and not on artifacts such as intercellular variations
of protein synthesis. Indeed, there was no correlation between the cell
surface fluorescence of Gq/11 and the total expression of
AT1A-EGFP or Gq/11 in either HEK-293 or
HEK-AT1A-EGFP cells (data not shown). In
addition, in HEK-AT1A-EGFP cells, a constant
total cellular expression of the Gq/11 protein was measured despite
a heterogeneous expression of the AT1A-EGFP
receptor, indicating that the two levels of expression are
independent.
Cytoplasmic Translocation of the Gq/11 Protein and the
AT1A-EGFP Receptor Is Synchronized but Occurs
in Different Compartments
The AT1A receptor (Fig. 5A, left upper panel) and the
Gq/11 protein (Fig. 5A, middle upper panel) are
colocalized at the cell surface in basal conditions as indicated by the
yellow color of the plasma membrane in the overlay of the
two fluorescences (Fig. 5A, right upper panel). Therefore,
it was interesting to investigate the cellular distribution of Gq/11
protein after Ang II addition and receptor internalization. After 20
min at 37 C in the presence of 100 nM Ang II,
both proteins are translocated from the plasma membrane to
intracellular locations with major changes in the S’/C’ ratios of
both proteins (from 6.87 to 0.70 for EGFP-associated fluorescence and
from 1.31 to 0.34 for Gq/11-associated fluorescence, n = 13 for
each) (Table 2 and Fig. 5A, lower panels). The loss of
plasma membrane fluorescence was not associated with a loss of total
cellular fluorescence intensity (Table 2).
|
q/11 proteins were
analyzed next by sequential observation of the cells for 20 min. Figure 5B shows a representative example of this observation on a plasma
membrane segment and its adjacent cytoplasm using a high resolution
imaging on confocal microscope. During the first 2 or 3 min after Ang
II application, the AT1A-EGFP receptor
(green fluorescence on lane I) and the Gq/11 protein
(red fluorescence on lane II) remain colocalized at the
plasma membrane, as indicated by the yellow labeling on the
overlay (lane III). Starting at 3 min after Ang II
application and maximum at 5 min, both the
AT1A-EGFP receptor and the Gq/11 are
intracellular. However, these localizations appear predominantly
different since the AT1A-EGFP receptor is
essentially located in large internalization vesicles, and the
Gq/11-associated fluorescence is more diffuse in the cytoplasm and
corresponds to fine granules, without any superimposition with the
receptor, as indicated by the absence of yellow labeling in the overlay
(Fig. 5B, lane III, lower panels). Confocal images were
quantified and the kinetics of intracellular translocation for both the
AT1A-EGFP receptor and the Gq/11 protein were
found to be identical (Fig. 5C). Internalization of 50% of the
proteins was obtained after similar periods of time for the two
proteins (t1/2 = 2.7 min for
AT1A-EGFP; t1/2 = 3.2 min
for Gq/11 protein; n = 4 for each point). Therefore, the
AT1A-EGFP receptor and the Gq/11 protein are
both translocated after agonist application, and, even more
interestingly, their intracellular localizations seem to be
predominantly different.
In addition, the receptor-associated Gq/11 protein is translocated
inside the cell in a specific manner after the
AT1A-EGFP activation for two reasons: 1) no
Gq/11 translocation was observed after Ang II application in
untransfected HEK-293 cells, which do not express the receptor (data
not shown), and 2) Ang II-activated AT1A-EGFP
receptor is unable to translocate a Gs/olf protein, an unrelated G
protein (data not shown).
Ang II-Induced Gq/11 Protein Translocation Is Dependent on
Receptor Internalization but Not on Receptor Coupling and Signal
Transduction
The association of this Gq/11 translocation, which depends on
Ang II binding and on the expression of the
AT1A-EGFP receptor at the cell surface, with
either receptor activation, coupling and intracellular signaling, or
receptor internalization was investigated next, using several tools: 1)
the previously described pharmacological tools for dissociating
activation (agonists vs. antagonists) and internalization
(peptides vs. nonpeptides) and 2) two mutants of the
AT1A receptor presenting major dissociations
between signaling and internalization functions: the
D74N mutant, which presents a major defect in
inositol phosphate-calcium signaling (90–100% reduction according to
Refs. 15, 16) but is still internalized after ligand binding (same
references) and the truncated 329 mutant (9), which presents a
default of internalization after Ang II treatment but signals even
better than the wild-type receptor.
The peptide antagonist
[Sar1,Ile8]Ang II induces
the translocation of both the AT1A-EGFP receptor
and the Gq/11 protein without activating the receptor and
the signaling pathway (Fig. 6A). The nonpeptide
ligands Losartan (Fig. 6B) and L162,313 (Fig. 6C) are unable to induce
translocation either of the receptor or of the Gq/11-protein
independently of their agonist/antagonist status.
|
q/11 protein (Fig. 6D and
Table 2) in basal conditions. The translocation of both proteins is
similar to what is observed for the wild-type receptor in the presence
of 100 nM Ang II, despite the defect in receptor activation
and signal transduction (Fig. 6E and Table 2). This normal
internalization of the mutant is quantified by reductions of 90% and
67% of the S'/C' ratios of the
D74N-AT1A-EGFP- and
Gq/11-associated fluorescences, respectively, as compared with 90%
and 74% reductions for the wild-type receptor.
The 329-EGFP mutant stably expressed in HEK-293 cells presents the
same properties as the 329 mutant: it binds Ang II, induces an
increase in IP production, and has a default of internalization as
measured by the biochemical acid wash procedure (80% of Ang II-induced
internalization for the wild-type receptor and 40% for the 329-EGFP
mutant). The 329-EGFP mutant is colocalized at the cell surface with
the Gq/11 protein in basal conditions (Fig. 6F and Table 2). After a
100 nM Ang II treatment, the 329-EGFP mutant is partly
internalized as measured by a reduction of only 54% of the S'/C' ratio
(Fig. 6G, left, and Table 2) and despite Ang II-induced
receptor activation, the Gq/11 is still localized at the cell
surface, as measured by a reduction of only 8% of the S'/C' ratio
(Fig. 6G, middle, and Table 2).
The pharmacological data obtained with agonist and antagonist ligands
and those for the signaling- and the internalized-defective mutants
show that subcellular translocation of the Gq/11 protein is
dependent on receptor internalization but not on the activation state
of the AT1A-EGFP receptor.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
To achieve the morphological analysis of AT1A trafficking, we created and expressed a chimeric AT1A receptor with a C-terminal extension corresponding to a variant of the green fluorescent protein of the jelly fish Aequoria victoria. To be utilizable, this tool had to fulfil two main conditions: functionality and both direct and indirect specific detectability.
Despite the 27-kDa extension, the AT1A-EGFP
receptor is expressed at the plasma membrane, and its binding,
signaling, and internalization are comparable to that of the wild-type
receptor. Similar results have been obtained for other GPCR
C-terminally tagged with EGFP, such as the 1B-adrenergic (22), the
ß2-adrenergic (23), and the TRH receptors (24).
The immunoreactivity of this fusion protein was also verified. We found
that it could be specifically detected in transfected cells by Western
blot and immunoprecipitation, with an anti-EGFP antibody. The
heterogeneity of the bands observed (ranging from 70 to 90 kDa) is
reminiscent of what has been previously described for this receptor,
using either a polyclonal antibody raised against the C-terminal
segment of AT1A (10) or a short epitope
(hemagglutin)-tagged receptor (11). Finally, as shown in Fig. 3
of this paper the AT1A-EGFP receptor is
directly and again specifically visualized by confocal microscopy.
This functional and fluorescent AT1A-EGFP receptor is therefore the ideal tool for analyzing the interactions of this receptor with signaling and trafficking proteins and for the direct characterization of the internalization process.
The parallel development of a software for image analysis and quantification of the fluorescence in various cellular compartments was also of key importance for this study and was validated by a perfect correlation with the measurement of internalization by classical biochemical acid wash techniques. This direct morphological analysis of AT1A receptor internalization allows testing of a large variety of unlabeled ligands; extending the concept of previous studies (4, 25). Thus this analysis demonstrated that this process is dependent on binding to the peptide ligand binding site, but not to the nonpeptide ligand binding site, and is independent of coupling between G protein and the receptor since both peptide antagonist and G protein coupling-deficient mutants of the receptor undergo internalization. It also shows definitively and directly that the internalization of GPCR involves not only the disappearance of binding sites from the surface of the cell, but also the real intracellular translocation of the receptors contained in small vesicles with no contact with the cell surface. This model was also the perfect tool for evaluating the parallel traffic of the receptor and its cognate G proteins during receptor activation and signal transduction.
The fluorescent AT1A-EGFP receptor was
observed in cells and is mostly localized at the cell surface. The
opportunity to directly observe the receptor and the natural
variation of the receptor expression from one cell to another allowed
one interesting observation: the more the receptor is expressed at the
cell surface, the more the cell surface fraction of its cognate
Gq/11 protein increases. Such a recruitment could not be
demonstrated previously on cellular models where both the receptor and
the G proteins were overexpressed. Therefore, this observation is one
of the first experimental data that sustains the hypothesis that a GPCR
may recruit its G protein to the cell surface. This Gq/11
recruitment is not an artifact due to overexpression of the
AT1A-EGFP receptor, since it was made also for
the nontagged wild-type receptor and can be generalized to other
Gq/11-coupled receptors, such as the vasopressin V1b receptor (Fig. 4C). In the literature, G proteins are shown to be distributed in
both cytosolic and membrane compartments, with most G protein
immunoreactivity associated with the plasma membrane, microsomal, ER,
or Golgi membranes and the cytoskeleton (26, 27, 28, 29, 30). Thus, G proteins
can be addressed either to intracellular compartments or to the cell
surface, depending on the presence or absence of posttranslational
lipid modifications but also on the expression and targeting of its
associated proteins. A recent and interesting observation was made for
the Gß and Gz cellular distributions, indicating that the
Gß recruits the Gz to the proper cellular compartment (31).
Altogether, these data suggest strongly that the expression of both the
ß-subunits and of the cognate receptors induces the recruitment of
an adapted fraction of the G protein to the cell surface.
In the basal state (i.e. without any ligand) it is not
known how the receptor physically interacts with the G protein. Two
possibilities exist: either the GDP form of the G protein interacts
with the inactive forms of the receptor with a moderate affinity or
this G protein interacts only with the small fraction of the
receptor that is in its active state even in the basal conditions, but
this time with a strong affinity. The similar recruitment of Gq/11
by the wild-type and the D74N mutant, which
presents a major defect of activation, and inefficacy of Losartan to
modify the membrane recruitment of Gq/11 favor the hypothesis of an
association of the inactive forms of GPCR and Gq/11 rather than
association due to constitutive activity of the receptor.
The intracellular translocation of the Gq/11 protein after
interaction with the cognate receptor is another original and very
interesting observation of this study. This translocation is dependent
on receptor internalization but not on its functional coupling as
demonstrated by experiments with the coupling-defective
D74N-AT1A-EGFP and the
internalization-defective 329-EGFP mutants. This suggests that
intracellular translocation of the Gq/11 protein is not the
consequence of dissociation from the receptor and ß-subunits but
follows other cellular processes linked to internalization. Gq/11
translocation also occurs in a different cellular location compared
with that of the internalized receptor. Such intracellular
translocations of G proteins after receptor binding have been
demonstrated for Gs in response to activation of the ß2-adrenergic
receptor (26) and for Gq/11 in response to TRH binding to its
receptor (32, 33, 34). These studies were carried out in HEK-293 cells
transfected with both the receptor and its cognate G protein cDNAs.
The intracellular localizations were found either different, for the
ß2-adrenergic receptor and Gs protein, or similar for the TRH
receptor and Gq/11. For the AT1A receptor,
further studies using accurate markers of the different intracellular
compartments and their components should provide new information on the
fate of G proteins after their activation and intracellular
translocation.
In conclusion, this study demonstrates that endogenous Gq/11
proteins are recruited by the AT1A receptors from
an intracellular pool, depending on the level of expression of these
cognate receptors at the surface of the cell. Peptide binding to the
receptor induces a translocation of the G protein from the membrane to
the intracellular pools, simultaneously to the receptor
internalization. The translocation of Gq/11 is dependent on the
internalization rather than on the activation of the receptor. In the
near future, the in vivo labeling of proteins should make it
possible to follow in real time the kinetics of interactions and the
cellular translocation of these proteins during signal transduction.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
329-EGFP mutant was constructed by PCR. The sequence of the 329
was amplified using the 329 mutant (9) as a template and the primers
5'-(A A G C T T A C C A T G G C C C T T A C C T) and 3'-(C C C G G G C
C G A G T G G G A C T T G G C C T T T). The 5'-primer binds upstream
from the HindIII site of the 329 and the 3'- primer
introduces a XmaI site. The PCR fragment was digested with
HindIII and XmaI and inserted between the
HindIII and XmaI sites of pEGFP-N1
(CLONTECH Laboratories, Inc.); this construct was called
p329-EGFP. The sequences of the constructs were verified using
fluorescent dideoxynucleotide sequencing on an ABI Prism 377 sequencer
(Perkin-Elmer Corp., Norwalk, CT). The V1b-EGFP construct
was a gift from M. A. Ventura (M. A. Ventura and E.
Clauser, in preparation).
Cell Culture and Transfection
HEK-293 cells were obtained from ATCC (Manassas,
VA; F-14742, 1573-CRL) and were grown in DMEM supplemented with 7.5%
FCS, 0.5 mM glutamine, 100 U/ml penicillin, and 100 µg/ml
streptomycin (all from Life Technologies, Inc.,
Gaithersburg, MD). For stable and transient expression, HEK-293 cells
were tranfected with 1 µg/500,000 cells of the plasmid of interest,
using a liposomal transfection reagent (Dosper from Roche Molecular Biochemicals, Indianapolis, IN). Cell lines
stably expressing the AT1A-EGFP,
D74N-AT1A-EGFP,
329-EGFP, and V1b-EGFP receptors were selected for resistance to 750
µg/ml G418 (Life Technologies, Inc.) and cloned by
limiting dilution.
Pharmacological and Signaling Properties of the
AT1A-EGFP,
D74N-AT1A-EGFP,
329-EGFP, and V1b- EGFP Receptors
Binding experiments, including saturation and
displacement experiments with [125I]-labeled
Ang II, were performed on intact cells, essentially as previously
described (7). Binding data were analyzed by linear regression using
the Excel 5 program (Microsoft Corp., Bellvue,
WA).
The production of second messengers, inositol phosphates (IP), was determined by the extraction of IP and separation on a Dowex AG1-X8 (Bio-Rad Laboratories, Inc., Hercules, CA) column, after incubation of the cells with myo-[3H]inositol as previously described (7).
The signaling properties of the wild-type tagged receptor were also tested in an integrated biological assay, the stimulation of a reporter gene (luciferase) placed under the control of a minimal promoter and a multimer of a 12-O-tetradecanoylphorbol 13-acetate (TPA) (phorbol ester)-responsive element (CGTCA) (see Ref. 36 for details of the construction). Cells were transiently transfected with the various constructs. Two days after transfection, cells were seeded in 96-well white opaque plates (Packard Instruments, Meriden, CT) and stimulated for 24 h with various concentrations of Ang II, and then rinsed twice with PBS and assayed for luciferase activity by adding LucLite (Packard Instruments) substrate as described by the manufacturer. Luminescence was measured using a Top Count (Packard Instruments) in single photon counting mode.
Immunoprecipitation, Western-Blot Analyses of the
AT1A and the AT1A-EGFP
Receptors, and PNGase F Treatment
Transfected cells were scraped into
solubilization buffer (50 mM Tris-HCl, pH 7.4, 100
mM NaCl, 5 mM EDTA, 1% Triton X100, complete
protease inhibitor from Roche Molecular Biochemicals).
Cells (106) were incubated in 1 ml solubilization
buffer for 2 h at 4 C with slight agitation. Insoluble material
was removed by centrifugation, and the supernatant was incubated
overnight at 4 C with protein A sepharose coupled to 25 µg anti-GFP
antibody (anti-GFP antibody from Roche Molecular Biochemicals) per condition (corresponding to
106 cells). Immune complexes were collected by
centrifugation and washed four times in washing buffer (37) and then
eluted into Laemmli sample buffer containing 8% SDS and heated for 5
min at 95 C. Proteins were resolved by 10% SDS-PAGE and transferred to
a nitrocellulose membrane, which was processed as previously described
(37). The membrane was incubated overnight with 0.4 µg/ml of GFP
antibody, and then incubated with horseradish peroxidase-conjugated
goat antimouse (1:20,000; Amersham Pharmacia Biotech,
Arlington Heights, IL). Immune complexes were detected using
enhanced chemiluminescence (ECL) reagents (Amersham Pharmacia Biotech), and the membrane was placed against X-Omat film
(Eastman Kodak Co., Rochester, NY).
Deglycosylation of the AT1A-EGFP receptor was performed on immunoprecipitated material. This material was suspended in 25 µl of 1% SDS and heated at 100 C for 3 min. H2O (225 µl) was added, and the samples were heated again at 100 C for 1 min. After centrifugation, the supernatant containing the proteins was adjusted to 40 mM sodium phosphate, pH 7.5, 1% Triton, 20 mM EDTA, and 0.144 M ß-mercaptoethanol. After incubation with PNGase F (Roche Molecular Biochemicals) (2 U per sample) for 17 h at 37 C, samples were resuspended in Laemmli buffer containing 8% SDS and analyzed on a 10% SDS-PAGE.
Internalization Assays
Internalization of the AT1A-EGFP receptor
was measured by two different procedures using the stable
HEK-AT1A-EGFP cell line.
1. A biochemical acid wash procedure performed essentially as previously described (7) except that the acid wash was done in 0.2 M acetic acid, 0.5 M NaCl in binding buffer.
2. A confocal microscopy procedure: chambered coverglass with eight wells (Nunc, Roskilde, Denmark) was treated for 1 h with 0.1 mg/ml polyallylamine (Aldrich , Milwaukee, WI) and rinsed twice with water. Cells were seeded (50,000 cells per well), treated for 1 h at 37 C with 70 µM cycloheximide (Sigma, St. Louis, MO), and preincubated for 15 min at 4 C in Earle’s buffer (140 mM NaCl, 5 mM KCl, 1.8 mM CaCl2, 0.9 mM MgCl2·6 H2O, 25 mM HEPES, pH 7.6) supplemented with 0.1% BSA, 0.01% glucose, 70 µM cycloheximide, and 0.8 mM 1–10 phenantrolin. Cells were then incubated for 30 min at 4 C with various concentrations of ligand. At this point, as a control for the ligand specificity of internalization, some cells were subjected to an acid wash procedure (0.2 M acetic acid, 0.5 M NaCl in Earle’s buffer, 2 min at 4 C). Internalization was promoted by incubating the cells at 37 C for various periods of time in the case of Ang II and for 20 min at 37 C in the case of other peptide and nonpeptide ligands in Earle’s complete buffer. At the end of the incubation, cells were rinsed in ice- cold Earle’s buffer and fixed by incubation for 10 min in 100% methanol at 4 C.
Immunofluorescent Labeling of the G Proteins
Cells fixed and permeabilized by incubation in 100% methanol
for 10 min at 4 C were incubated for 30 min at room temperature with
5% normal goat serum (Sigma) in PBS + 0.1% BSA to
saturate nonspecific binding sites. Cells were then incubated overnight
at 4 C with an antibody directed against Gq/11 or Gs/olf
(Santa Cruz Biotechnology, Inc., Santa Cruz, CA) at a
concentration of 2 µg/ml in PBS + 0.1% BSA. Cells were washed (PBS +
0.1% BSA) four times for 10 min each at room temperature and incubated
with antirabbit IgG coupled to cyanine-3 (Cy3) (Sigma) at
a dilution of 1:800 for 60 min at room temperature in the dark. Cells
were washed a further four times and analyzed.
Confocal Microscopy
Cells were examined with a TCS NT confocal laser scanning
microscope (Leica Corp., Deerfield, IL) configured with a
Leica Corp. DM IRBE inverted microscope equipped with an
argon/krypton laser. EGFP fluorescence was detected after 100%
excitation at 488 nm using a spectrophotometer set with a window
between 530 and 560 nm. For double detection of EGFP and Cy3
fluorescences, after excitation at 488 nm (excitation set to 100%) and
568 nm (excitation set to 50%), respectively, the fluorescences were
detected in windows of 500–550 nm and 580–630 nm, respectively. Dual
excitation at 488 nm and 568 nm and the acquisition of the emission
were done simultaneously. Images of individual cells (1,024 x
1,024 pixels) were obtained using a 63x oil-immersion objective. Each
image was done on a crosssection through the cells. Laser output
power and photonmultipicator settings were kept at similar levels
throughout all experiments, so that the intensity values for different
experiments could be compared (see below). Lack of cross-talk
between the green and the red channels was verified as follows: 1)
an absence of red fluorescence was observed in
HEK-AT1A-EGFP cells unlabeled with Cy3; 2) in
HEKAT1A-EGFP cells labeled with Cy3, no loss
of red fluorescence intensity was observed when excitation at 488 nm
was set to 0%.
Quantification of the Subcellular Distribution of the
Fluorescence
Digital image analysis allowed us to measure the subcellular
distribution of the AT1A-EGFP receptor and the
Gq/11 protein, using a slightly modified version of a specific micro
software (40) derived from the public domain NIH Image program
(developed by the U.S. National Institute of Health and available on
the Internet at http://rsb.info.nih.gov/nih-image/). First, grayscale
conversion and median filtering of the confocal image were performed.
Then the mean density of total cellular fluorescence was determined for
each cell, and mean pixel grayscale density was automatically sampled
at 36 different locations on the cell plasma membrane, yielding the
mean surface fluorescence (S) value. The mean grayscale densities of
the cytoplasm (C) and the nucleus (N) were measured after manually
selecting the corresponding areas. The background fluorescence (N) was
subtracted from the C and S values to give the S' and C' values. The
S'/C' ratio provides reliable information about the level of cell
surface expression and the internalization state of the autofluorescent
receptor or membrane-bound proteins.
Statistics
Results are expressed as means ± SEM.
Statistical significance was assessed by Student’s
t-test.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
This study was supported by a grant from the Fond de Recherche H
scht
Marion Roussel (FR98CVS008) and also by the Institut national pour la
santé and la recherche médicale.
Received for publication November 29, 1999. Revision received October 4, 2000. Accepted for publication November 1, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1b-adrenoceptor with green
fluorescent protein. Mol Endocrinol 12:1099–1111. Mol Biol Cell 7:1225–1233[Abstract]
-subunits in rat pancreatic endocrine cells. J
Histochem Cytochem 47:289–302i-3 in foetal lung
fibroblasts. Eur Respir J 12:1137–1140[Abstract]
subunits. FEBS Lett 409:24–28[CrossRef][Medline]
targets G(Z). Proc Natl Acad
Sci USA 97:1085–1090: analysis of agonist regulation of
coexpressed G11 species variants. Mol Pharmacol 49:646–655[Abstract]
and
thyrotropin-releasing hormone receptors proceed on different time
scales. J Biol Chem 273:21699–21707/G11 induced by
agonist stimulation. Biochem J 340:529–538
This article has been cited by other articles:
 |
|
 |
|
  K. Thieme, D. M. N. Eguti, M. Mello-Aires, and M. Oliveira-Souza The effect of angiotensin II on intracellular pH is mediated by AT1 receptor translocation Am J Physiol Cell Physiol, July 1, 2008; 295(1): C138 - C145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Oliveira, C. M. Costa-Neto, C. R. Nakaie, S. Schreier, S. I. Shimuta, and A. C. M. Paiva The Angiotensin II AT1 Receptor Structure-Activity Correlations in the Light of Rhodopsin Structure Physiol Rev, April 1, 2007; 87(2): 565 - 592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-H. Feng, Y. Ding, S. Ren, L. Zhou, C. Xu, and S. S. Karnik Unconventional Homologous Internalization of the Angiotensin II Type-1 Receptor Induced by G-Protein-Independent Signals Hypertension, August 1, 2005; 46(2): 419 - 425. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. SPAT and L. HUNYADY Control of Aldosterone Secretion: A Model for Convergence in Cellular Signaling Pathways Physiol Rev, April 1, 2004; 84(2): 489 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Blondet, M. Doghman, M. Rached, P. Durand, M. Begeot, and D. Naville Characterization of Cell Lines Stably Expressing Human Normal or Mutated EGFP-Tagged MC4R J. Biochem., April 1, 2004; 135(4): 541 - 546. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. K. Lee, A. J. Lanca, R. Cheng, T. Nguyen, X. D. Ji, F. Gobeil Jr., S. Chemtob, Susan. R. George, and B. F. O'Dowd Agonist-independent Nuclear Localization of the Apelin, Angiotensin AT1, and Bradykinin B2 Receptors J. Biol. Chem., February 27, 2004; 279(9): 7901 - 7908. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Hunyady, A. J. Baukal, Z. Gaborik, J. A. Olivares-Reyes, M. Bor, M. Szaszak, R. Lodge, K. J. Catt, and T. Balla Differential PI 3-kinase dependence of early and late phases of recycling of the internalized AT1 angiotensin receptor J. Cell Biol., June 24, 2002; 157(7): 1211 - 1222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Holloway, H. Qian, L. Pipolo, J. Ziogas, S.-i. Miura, S. Karnik, B. R. Southwell, M. J. Lew, and W. G. Thomas Side-Chain Substitutions within Angiotensin II Reveal Different Requirements for Signaling, Internalization, and Phosphorylation of Type 1A Angiotensin Receptors Mol. Pharmacol., April 1, 2002; 61(4): 768 - 777. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Miserey-Lenkei, C. Parnot, S. Bardin, P. Corvol, and E. Clauser Constitutive Internalization of Constitutively Active Angiotensin II AT1A Receptor Mutants Is Blocked by Inverse Agonists J. Biol. Chem., February 15, 2002; 277(8): 5891 - 5901. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
) and G