(Received for publication, April 30, 1997, and in revised form, June 5, 1997)
From the Laboratory of Bioorganic Chemistry, NIDDK, National
Institutes of Health, Bethesda, Maryland 20892 and the
Departments of Medicine and Pharmacology, Gladstone
Institute of Cardiovascular Disease, University of California,
San Francisco, California 94141-9100
Characteristically, an individual member of the
superfamily of G protein-coupled receptors can interact only with a
limited number of the many structurally closely related G protein
heterotrimers that are expressed within a cell. Interestingly, the N
termini of two G protein subunits, G
q and
G
11, differ from those of other
subunits in that
they display a unique, highly conserved six-amino acid extension. To
test the hypothesis that this sequence element is critical for proper
receptor recognition, we prepared a G
q deletion mutant
(
6q) lacking these first six amino acids. The
6q construct (or wild
type G
q as a control) was coexpressed (in COS-7 cells)
with several different Gi/o- or Gs-coupled
receptors, and ligand-induced increases in inositol phosphate
production were determined as a measure of G protein activation.
Whereas these receptors did not efficiently interact with wild type
G
q, most of them gained the ability to productively
couple to
6q. Additional experiments indicated that the observed
functional promiscuity of
6q is not due to overexpression (as
compared with wild type G
q) or to a lack of
palmitoylation. We conclude that the N-terminal extension
characteristic for G
q/11 proteins is critical for
constraining the receptor coupling selectivity of these subunits,
indicative of a novel mechanism by which the fidelity of receptor-G
protein interactions can be regulated.
G protein-coupled receptors
(GPCRs),1 when activated by
extracellular ligands, interact with specific classes of heterotrimeric G proteins (consisting of ,
, and
subunits) which can then, in their activated forms, inhibit or activate various effector enzymes
and/or ion channels (1-5). Characteristically, a specific GPCR can
interact with only a limited subset of the many structurally similar G
proteins that are expressed within a cell. Molecular genetic and
biochemical studies have identified distinct intracellular regions (as
well as single amino acids contained within these domains) on the GPCR
proteins that play key roles in determining the fidelity of receptor-G
protein coupling (1-7). In addition, recent studies have shown that
residues at the extreme C terminus of the G protein
subunits are
also of fundamental importance for the selectivity of receptor-G
protein interactions (8-10). However, several lines of evidence
suggest that the C terminus of the G
subunits is clearly not the
only structural determinant on the G proteins that is critical for
dictating receptor-G protein coupling selectivity (2, 5).
Interestingly, two G protein subunits, G
q and
G
11, contain a unique six-amino acid extension that is
not found in other G
subunits (Fig.
1). This short sequence is highly
conserved among all vertebrate species from which these subunits have
been cloned so far (11-15), suggesting that it may be relevant for
some aspect of G
q/G
11 function.
Previous studies (16, 17) analyzing the biochemical properties of a
mutant Gq subunit lacking the N-terminal extension (hereafter referred to as
6q) failed to reveal any major functional differences between wild type G
q (WTq) and
6q. Both
studies showed that Gq/11-coupled receptors such as the NK2
neurokinin (16) and the m1 muscarinic receptor (17) were able to
activate
6q in a fashion identical to WTq. Similarly, other
functional properties, such as their affinity for
subunits and
their ability to activate downstream effectors (e.g.
phospholipase C
(PLC
)), were also found to be very similar for
the two G protein subunits (16, 17).
In this study, we tested the hypothesis that the N-terminal extension
in WTq may play a role in maintaining the selectivity of receptor
recognition, an issue that had not been addressed yet. Toward this
goal, the ability of several different Gi/o- and
Gs-coupled receptors to interact with WTq or 6q were
examined in cotransfected COS-7 cells. Whereas none of the receptors
(upon incubation with the appropriate agonist ligands) was able to
activate WTq to a significant extent, most of the receptors gained the ability to couple to
6q with considerable efficiency (measured biochemical response: stimulation of phosphatidylinositol (PI) hydrolysis). These data suggest that the N-terminal extension characteristic for G
q and G
11 subunits is
critical for constraining the receptor coupling selectivity of these
proteins.
To create a
construct coding for a mutant Gq subunit lacking the
first six amino acids (
6q), a pcDNAI-based expression plasmid coding for murine WTq (11, 18) was used. To generate the
6q expression plasmid, a 78-base pair synthetic
BamHI-FspI fragment containing the desired
deletion was used to replace the corresponding sequence in the wild
type plasmid. In both plasmids (WTq and
6q), the BamHI
site of the pcDNAI polylinker was immediately followed by the
initiating ATG codon. Both plasmids contained a short sequence coding
for an internal hemagglutinin (HA) epitope tag (DVPDYA), which replaced
WTq residues 125-130 (18). The presence of the epitope tag did not
affect the receptor and effector coupling properties of WTq (8, 9, 18).
The identity of the two G protein constructs was verified by dideoxy
sequencing (19).
COS-7 cells were grown in Dulbecco's modified
Eagle's medium supplemented with 10% fetal calf serum at 37 °C in
a humidified 5% CO2 incubator. For transfections, 1 × 106 cells were seeded into 100-mm dishes. About 24 h later, COS-7 cells were cotransfected with expression plasmids coding
for WTq or 6q (1 µg DNA/dish) and the indicated receptor cDNAs
(4 µg DNA/dish) by using a DEAE-dextran procedure (20). The following receptor expression plasmids were used: m2 muscarinic receptor in pcD
(21), D2 dopamine receptor (22) in pcDNAI,
-opioid receptor (23)
in pcDNA3, somatostatin SSTR1 receptor (24) in pCMV, A1 adenosine
receptor (25) in CDM7, D1 dopamine receptor (26) in pcDNAI, V2
vasopressin receptor in pcD-PS (27), and
2-adrenergic receptor (28)
in pSVL.
Approximately 24 h after transfections, cells were split into 6-well dishes (~0.4 × 106 cells/well) in culture medium supplemented with 3 µCi/ml [3H]myo-inositol (20 Ci/mmol; American Radiolabeled Chemicals Inc.). After a 24-h labeling period, cells were preincubated for 20 min at room temperature with 2 ml of Hanks' balanced salt solution containing 20 mM HEPES and 10 mM LiCl. Cells were then stimulated in the same buffer with the appropriate agonist ligands (1 h at 37 °C), and increases in intracellular inositol monophosphate (IP1) levels were determined by anion exchange chromatography as described (10).
In a subset of experiments, transfected cells were incubated with pertussis toxin (PTX; 500 ng/ml) for the last 18-24 h of culture.
[3H]Palmitate LabelingApproximately 48 h after transfections, COS-7 cells were metabolically labeled for
1 h with 800 µCi/ml of [9,10-3H]palmitate (60 Ci/mmol; American Radiolabeled Chemicals Inc.) in 5 ml of serum-free
medium supplemented with 1% (v/v) dimethyl sulfoxide as described
(29). The chemical nature of [3H]palmitate incorporation
into G protein subunits has been characterized earlier (18, 30) in
cells radiolabeled under similar conditions.
Cells were fractionated into particulate and soluble fractions as described (29).
Immunoprecipitation and ImmunoblottingThe 12CA5 mouse
monoclonal antibody (BAbCo) specific for the HA epitope was used for
immunoprecipitation and immunoblotting. Immunoprecipitation studies
were performed using equivalent amounts of protein (2.5 mg) from total
cell suspension in solubilization buffer consisting of 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1% (w/v) of
Nonidet P-40, 0.5% (w/v) sodium deoxycholate, and 0.1% (w/v) SDS in a
total volume of 1 ml. Cell suspensions were added to the solubilization
buffer and incubated at 4 °C on a rotator for 30 min followed by
centrifugation for 10 min at 14,000 rpm in an Eppendorf 5415 microcentrifuge to pellet insoluble material. Supernatants were
transferred to fresh Eppendorf tubes, and 5 µg of 12CA5 antibody and
20 µl of a 50% suspension of protein A-agarose (Sigma) were added
followed by an overnight incubation at 4 °C on a rotator.
Immunoprecipitates were recovered by centrifugation at 1,000 × g in a microcentrifuge, washed twice in 1 ml of
solubilization buffer containing of the original detergent
concentration, solubilized in 45 µl of gel sample buffer (Novex) in
the absence of reducing agents, boiled for 5 min, separated by SDS-PAGE
on 10% Tris-Gly gels (Novex), and prepared for fluorography using
Amplify (Amersham Corp.) according to the manufacturer's instructions.
Fluorograms were exposed for 4-6 weeks at
70 °C.
Immunoblotting was performed by separating equal amounts of protein
(100 µg) from subcellular fractions solubilized in gel sample buffer
(Novex) with 2.5% (v/v) -mercaptoethanol on 10% Tris-Gly gels
(Novex), transfer to nitrocellulose membranes, probing with the 12CA5
antibody conjugated to horseradish peroxidase (Boehringer Mannheim),
and development with enhanced chemiluminescence reagents (Amersham
Corp.). Protein concentrations were determined using the Bio-Rad
protein assay kit with IgG as the standard.
[Arg8]vasopressin, ()-isoproterenol,
and PTX were purchased from Sigma. All other ligands used in this study
were obtained through Research Biochemicals Inc.
Initially, a
series of receptors that are preferentially coupled to G proteins of
the Gi/o family (m2 muscarinic, D2 dopamine, -opioid,
SSTR1 somatostatin, and A1 adenosine) were coexpressed in COS-7 cells
with either WTq or
6q. Transfected cells were then incubated with the
appropriate agonist ligands, and the ability of the different receptors
to couple to the two G proteins was determined by measuring increases
in inositol phosphate production (due to WTq-mediated activation of
PLC
; Refs. 31 and 32). Coexpression of the different
Gi/o-coupled receptors with either vector DNA (pcDNAI)
or WTq, followed by ligand stimulation, resulted only in a rather small
increase in PLC
activity (Fig.
2A). As shown in Fig.
3 for the m2 muscarinic and D2 dopamine
receptors, this small increase in inositol phosphate production could
be almost completely blocked by pretreatment of cells with PTX (500 ng/ml). Consistent with previous findings (33, 34), this observation suggests that the m2 muscarinic and D2 dopamine receptors do not couple
to WTq to a significant extent and that the small increase in PI
hydrolysis seen after stimulation of these receptors is most likely due
to activation of PLC
by G protein
subunits released upon
receptor-mediated activation of endogenous Gi/o proteins (35, 36).
Interestingly, coexpression of the different Gi/o-coupled
receptors with 6q resulted in a significantly increased PI response (as compared with WTq) that was most pronounced in the case of the two
biogenic amine receptors (m2 muscarinic and D2 dopamine; Fig.
2A). These responses could only be partly blocked by PTX pretreatment (shown for the m2 muscarinic and D2 dopamine receptors in
Fig. 3), indicating that they were primarily due to receptor-mediated generation of activated
6q. Complete concentration-response curves for m2 muscarinic and D2 dopamine receptor-mediated activation of
6q
are given in Fig. 4. The EC50
values (means ± S.E. of five independent experiments, each
carried out in triplicate) for these responses amounted to 7.4 ± 0.3 µM in the case of the muscarinic agonist, carbachol
(Fig. 4A), and to 1.4 ± 0.2 µM in the
case of the dopaminergic ligand, (
)-quinpirole (Fig. 4B),
indicating that the interaction of the two biogenic amine receptors
with
6q was highly efficient.
We next examined the ability of three Gs-coupled receptors
(D1 dopamine, V2 vasopressin, and 2-adrenergic) to functionally interact with WTq or
6q. Consistent with their known coupling profiles (37), these receptors were unable to activate WTq to an
appreciable extent (Fig. 2B). However, two of the
investigated Gs-coupled receptors (D1 dopamine and
2-adrenergic) gained the ability to induce a significant increase in
inositol production when coexpressed with
6q (as compared with WTq;
Fig. 2B).
Two previous studies have shown that Gq/11-coupled
receptors such as the NK2 neurokinin (16) or the m1 muscarinic receptor (17) can activate 6q in a fashion identical to WTq. Our data therefore suggest that
6q, in contrast to qWT, can be activated by
receptors that are members of all three major functional classes of
GPCRs.
The maximum degree of PLC stimulation mediated by the bona fide
Gq/11-coupled m3 muscarinic receptor (7) amounted to
6-10-fold (data not shown), indicating that activation of 6q by
different Gi/o- and Gs-coupled receptors (most
of which mediated a 2-6-fold stimulation of PLC activity; Fig. 2) was
not optimal. This observation is consistent with the currently held
notion that efficient receptor-G protein coupling involves multiple
sites of contact between the receptor protein and the G protein
heterotrimer (1-7).
It should also be noted that coexpressed 6q did not improve coupling
of the V2 vasopressin receptor to PLC stimulation and that the absolute
magnitude of responses mediated by
6q upon coexpression with the
-opioid or the SSTR1 somatostatin receptor, respectively, was quite
small (Fig. 2), indicating that
6q is not generally promiscuous. One
possible reason for the observed differences in the ability of the
studied Gi/o- and Gs-coupled receptors to
interact with
6q may be that the relative functional importance of
individual receptor-G protein contact sites may vary among different
GPCRs (e.g. peptide receptors versus receptors for biogenic amines).
To exclude the
possibility that the promiscuous nature of 6q was simply due to
exceptionally high expression levels (as compared with WTq), the
subcellular distribution of the two G protein subunits was studied by
cell fractionation and immunoblotting. Both G
subunits were detected
with the monoclonal antibody, 12CA5, which recognizes the internal HA
epitope tag present in both proteins (see "Experimental
Procedures"). As shown in Fig.
5A, both G protein constructs
were found exclusively in the particulate fraction; however,
6q was
expressed at lower levels (approximately 10-20% of WTq, as determined
by scanning densitometry; data not shown). The precise reason for this
latter observation remains unclear; however, possible factors may
include reduced protein stability or translation efficiency. In any
case, these data exclude the possibility that the ability of
6q to be
activated by multiple classes of GPCRs is due to overexpression of this
subunit (as compared with WTq).
Palmitoylation Pattern of WTq and
Previous studies
(16-18, 38-40) have shown that G proteins of the Gq/11
family, like most other G
subunits, are palmitoylated at cysteine
residues located near the N terminus of the proteins (corresponding to
Cys9 and Cys10 in Fig. 1). To compare the
palmitoylation patterns of WTq and
6q, transfected COS-7 cells were
metabolically labeled with [3H]palmitic acid followed by
immunoprecipitation of WTq and
6q by the 12CA5 monoclonal antibody,
SDS-PAGE, and fluorography. As shown in Fig. 5B, WTq and
6q were the only immunoprecipitated proteins, because no labeled
proteins could be precipitated when cells were transfected with
"empty" vector DNA. Consistent with published results, both WTq
(16-18, 38, 39) and
6q (16) incorporated considerable amounts of
[3H]palmitate (Fig. 5B). The reduction in
signal strength seen with
6q (approximately 10-20% of WTq, as
determined by scanning densitometry; data not shown) correlated well
with the reduction in
6q levels revealed by immunoblotting (Fig.
5A). This observation suggests that
6q is palmitoylated to
an extent similar to that of WTq, as has been observed earlier in
transfected COS-7 cells (16). It is therefore unlikely that differences
in palmitoylation patterns are responsible for the different functional
properties of
6q and WTq.
We have shown that 6q, in contrast to WTq, can
productively interact with several different Gi/o- and
Gs-coupled receptors, suggesting that the six-amino acid
extension that is unique to WTq (as well as G
11) is
critical for constraining the receptor coupling selectivity of this G
protein subunit.
6q and WTq were found to be palmitoylated to a
similar extent, suggesting that the functional promiscuity of
6q is
not due to a lack of acylation. One possibility is that the N-terminal
extension characteristic for G
q/11 subunits constrains
the receptor coupling selectivity by preventing access of
Gi/o- and Gs-coupled receptors. Alternatively, it is conceivable that this six-amino acid sequence exerts an indirect
conformational effect on the structure of G
q/11 that is
crucial for maintaining the receptor selectivity of these subunits. Clearly, this issue needs to be addressed in future studies. In summary, our data suggest a novel mechanism by which receptor-G protein
coupling selectivity can be achieved and should contribute to a better
understanding of the molecular basis of receptor-G protein
interactions.
We thank all of the individuals who generously provided us with receptor expression plasmids.