Differential signaling of dopamine-D2S and -D2L receptors to inhibit ERK1/2 phosphorylation
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Journal of Neurochemistry, 2007, 102, 1796–1804 doi:10.1111/j.1471-4159.2007.04650.x
Differential signaling of dopamine-D2S and -D2L receptors
to inhibit ERK1/2 phosphorylation
Irit Itzhaki Van-Ham, Behzad Banihashemi, Ariel M. Wilson, Kirsten X. Jacobsen1,
Margaret Czesak2 and Paul R. Albert3
Departments of Medicine and Cellular and Molecular Medicine, and Ottawa Health Research Institute (Neuroscience),
University of Ottawa, Ottawa, Ontario, Canada
Abstract fected GH4 cells, the D2L-SS mutant inhibited thyrotropin-
Although they have distinct functions, the signaling of dop- releasing hormone-induced ERK1/2 phosphorylation almost
amine-D2 receptor short and long isoforms (D2S and D2L) is as strongly as the D2S receptor. A D2S-triple mutant that
virtually identical. We compared inhibitory regulation of extra- eliminates PKC sites involved in D2S receptor desensitization
cellular signal-regulated kinases (ERK1/2) in GH4 pituitary also inhibited ERK1/2 activation. Similarly, in striatal cultures,
cells separately transfected with these isoforms. Activation of the D2-selective agonist quinpirole inhibited potassium-stimu-
rat or human dopamine-D2S, muscarinic or somatostatin lated ERK1/2 phosphorylation, indicating the presence of this
receptors inhibited thyrotropin-releasing hormone-induced pathway in neurons. In conclusion, the D2S and D2L receptors
ERK1/2 phosphorylation, while the D2L receptor failed to differ in inhibitory signaling to ERK1/2 due to specific residues
inhibit this response. In order to address the structural basis for in the D2L receptor alternatively spliced domain, which may
the differential signaling of D2S and D2L receptors, we exam- account for differences in their function in vivo.
ined the D2L-SS mutant, in which a protein kinase C (PKC) Keywords: cell proliferation, dopamine, G-protein, mitogen-
pseudosubstrate site that is present in the D2L but not D2S activated protein kinase, prolactin.
receptor was converted to a consensus PKC site. In trans- J. Neurochem. (2007) 102, 1796–1804.
There are five dopamine receptor subtypes, which are (PKC)-induced uncoupling (Liu et al. 1992) and agonist-
divided into two groups: D1-like (D1 and D5) and D2-like induced internalization (Ito et al. 1999), and their G-protein
receptors (D2, D3 and D4 receptors). D1-like receptors specificities differ in some cases (Liu et al. 1994; Guiramand
stimulate adenylyl cyclase activity while D2-like receptors et al. 1995; Albert and Robillard 2002).
inhibit cAMP production by coupling to pertussis toxin- Studies of knockout or transgenic mice indicate distinct
sensitive Gi/Go-proteins. Pharmacological and gene knock- roles for these D2 isoforms. Gene knockout studies of the D2L
out studies demonstrate that dopamine-D2 receptors mediate receptor indicate its preferential role as a post-synaptic
dopaminergic inhibition of prolactin (PRL) synthesis and
secretion, lactotroph proliferation and transformation Received December 7, 2006; revised manuscript received March 8,
(Missale et al. 1998; Ben-Jonathan and Hnasko 2001), and 2007; accepted April 1, 2007.
play key roles in dopaminergic control of movement and Address correspondence and reprint requests to Paul R. Albert, Ottawa
behavior. The short (D2S) and long (D2L) forms of the D2 Health Research Institute (Neuroscience), University of Ottawa, 451
Smyth Road, Ottawa, ON, Canada K1H 8M5.
receptor are generated by alternative splicing of exon VI, E-mail: palbert@uottawa.ca
which encodes 29 amino acids located in the third intracel- 1
Recipient of National Science and Engineering Research Council
lular (i3) domain of the D2L receptor (Civelli et al. 1993). Canadian Graduate Scholarship.
2
Although the i3 loop is implicated in receptor–G-protein Recipient of a CIHR Doctoral Research Award.
3
coupling, both receptor isoforms have virtually identical Recipient of the Novartis/Canadian Institutes of Health Research
(CIHR) Michael Smith Chair in Neurosciences.
pharmacology and share equivalent signaling pathways in Abbreviations used: ERK, extracellular signal-regulated kinase; PKC,
most cell types (Civelli et al. 1993; Albert 1994). The D2S protein kinase C; PRL, prolactin; PTX, pertussis toxin; TRH, thyrotro-
receptor, however, is more sensitive to protein kinase C pin-releasing hormone.
2007 The Authors
1796 Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–1804Differential dopamine-D2S and -D2L signaling 1797
receptor, while the D2S receptor is mainly pre-synaptic, Cell culture
located on dopamine neuron cell bodies (Khan et al. 1998; GH4-rD2S (Albert et al. 1990), GH4-hD2S and GH4-hD2L cells
Usiello et al. 2000). In transgenic mice separately expressing (Liu et al. 1994) and derivative clones were maintained in Ham’s
D2S or D2L receptors in lactotrophs, the D2S receptor F10 medium with 8% fetal bovine serum at 37C, 5% CO2. For
striatal primary cultures, the caudate and putamen were dissected
mediated inhibition of lactotroph proliferation, while the D2L
from 18-day-old rat embryos and mechanically dissociated by gently
receptor did not (Iaccarino et al. 2002). As the pharmacology
pipetting in Hank’s calcium-free medium by procedures approved
and signaling of D2 receptor isoforms are so similar, the by the University of Ottawa Animal Care Committee. An equal
mechanisms underlying their distinct roles in vivo are unclear. volume of Hank’s (plus calcium) medium was added and the cells
In this study, we have examined D2 receptor signaling to were collected by centrifugation at 750 g for 5 min. Cell pellets
the extracellular signal-regulated kinase (ERK1/2) cascade, a were resuspended in Neurobasal medium + B27 supplement (Invi-
pathway that stimulates PRL gene transcription (Wang and trogen, Burlington, ON, Canada) and 500 nmol/L L-glutamine and
Maurer 1999; Kievit et al. 2001) and may regulate lactotroph plated into six-well plates coated with 10 lg/mL poly-D-lysine. The
proliferation (Iaccarino et al. 2002). The coupling of D2 cells were cultured at 37C, 5% CO2 and were used after 14 days in
receptors to ERK1/2 regulation is cell type dependent. In culture.
non-neuronal cells, such as Balb/c-3T3 cells, dopamine-D2
Ligand binding
receptors couple via Gai2 and Gbc subunits to stimulate
Cell membranes were prepared from 15-cm dishes by replacing the
ERK1/2 activity and cell proliferation (Albert and Robillard
medium with hypotonic buffer (15 mmol/L Tris–HCl, pH 7.4,
2002). However, in neuroendocrine GH4-rD2S (GH4ZR7 2.5 mmol/L MgCl2, 0.2 mmol/L EDTA). After swelling for 5 min
clone) pituitary cells (Albert et al. 1990), dopamine-D2S on ice, the cells were scraped from the plate and centrifuged at 500 g
receptors mediated inhibition of basal or thyrotropin-releas- for 15 min and the pellet was resuspended in cold TME buffer
ing hormone (TRH)-stimulated ERK1/2 phosphorylation that (75 mmol/L Tris, pH 7.4, 12.5 mmol/L MgCl2, 1 mmol/L EDTA).
was mediated by Gao and Gai3, respectively (Banihashemi The cells were homogenized and centrifuged at 12 000 g for 30 min
and Albert 2002; Liu et al. 2002). Importantly, dopamine- at 4C and resuspended in TME. For binding assay, aliquots of
D2-mediated inhibition of ERK1/2 was replicated in normal 100 lg/tube membrane preparation were added to triplicate tubes
rat pituitary cells (Liu et al. 2002). containing 0.5 mL of TME + 0.1% ascorbic acid with [3H]-
To further elucidate dopamine-D2 signaling pathways, we spiperone (1 nmol/L) ± 10 lmol/L dopamine to determine total
and non-specific binding, respectively. Reactions were terminated
have stably transfected GH4C1 cells, which lack dopamine
after a 1-h incubation at 22C by the addition of 3-mL ice-cold
receptors, with the rat or human dopamine-D2S or human D2L
buffer, and bound ligand was separated from free by filtration onto
receptors to produce GH4-rD2S, GH4-hD2S and GH4-hD2L Whatman GF/A filter discs, using a vacuum manifold, and
cells (Albert et al. 1990; Liu and Albert 1991). Although radioactivity of the filter quantified by scintillation counting. The
activation of dopamine-D2S, as well as endogenous muscar- specific binding of [3H]-spiperone was [mean ± SE (n)] 63.3 ± 11.4
inic and somatostatin receptors, inhibited TRH-induced (3), 55.8 ± 2.9 (4) and 57.0 ± 3.6 (3) fmol/mg for GH4-rD2S, GH4-
ERK1/2 phosphorylation, activation of the dopamine-D2L hD2L and GH4-hD2LSS cells, respectively.
receptor failed to do so. By transfection of mutant D2S and
D2L receptors, we have defined a structural difference cAMP measurement
between two isoforms that determines signaling specificity Equal numbers of cells were plated in six-well plates and grown to
70–80% confluence and then incubated at 37C in 1 mL/well of
to inhibit ERK1/2 activation. In addition, we have found that
serum-free Dulbecco’s modified Eagle’s medium/20 mmol/L
in striatal cultures, D2 agonists couple to inhibition of ERK1/
HEPES, pH 7.0/100 lmol/L isobutylmethylxanthine, with or without
2, consistent with a role for this signaling pathway in vivo. experimental compounds. After 20 min, the media were recovered
and centrifuged at 12 000 g for 2 min at 4C to remove detached
cells. The supernatant was stored at )20C and thawed for specific
Materials and methods radioimmunoassay to measure cAMP level. Percent inhibition was
calculated as 100 ) [100(D ) C)/(S ) C)], where the cAMP level in
Materials control (C) forskolin-treated (S) cells, and forskolin/apomorphine,
Apomorphine, dopamine, quinpirole, EGTA, forskolin, pertussis carbachol or somatostatin-treated cells (D) was used. These values
toxin (PTX), TRH, somatostatin, and anti-b-actin were obtained were then normalized to control GH4-rD2S cells (=100%).
from Sigma (St Louis, MO, USA); [125I]-succinyl cAMP (2200 Ci/
mmol/L) and polyvinylidene difluoride membrane were from New Measurement of phospho-ERK1/2
England Nuclear Corp. (Boston, MA, USA); [3H]-spiperone Cells (3 · 105 cells/well) were plated in six-well plates and upon
(105 Ci/mmol) and enhanced chemiluminescence (ECL) detec- reaching 80% confluence transferred to serum-free Ham’s F10
tion kits were from Amersham Corp. (Arlington Heights, IL, USA). medium [1 or 16 h (overnight), 37C]. Cells were treated with the
Sera and media were obtained from BD Biosciences, Mississauga, indicated drugs at 37C and 15 min later the plates transferred on ice
ON, Canada. Anti-phospho-ERK1/2 antibody (T202/Y2040) and and washed two times with ice-cold phosphate-buffered saline. The
anti-ERK1/2 antibodies were from New England Biolabs (Boston, cells were lysed in 50 lL of 5· sodium dodecyl sulfate loading
MA, USA).
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–18041798 I. I. Van-Ham et al.
(a) Fig. 1 D2S, but not D2L receptors, inhibit TRH-induced ERK1/2
phosphorylation. (a) Pituitary GH4-rD2S and GH4-hD2L (stably trans-
fected with rat D2S or human D2L receptor cDNA, respectively) were
assayed for phospho-ERK1/2 (pERK1/2) levels by western blot using
a phospho-ERK1/2-specific antibody. Cells were incubated for 15 min
at 37C with no treatment or 1 lmol/L TRH, without or with added
dopamine agonist apomorphine (1 lmol/L). Above is shown a repre-
sentative western blot that was re-probed for b-actin (below) as a
loading control. Below is plotted the mean phospho-ERK1/2 level
quantified by densitometric analysis and expressed as percent of TRH
(1 lmol/L)-induced pERK1/2 for hD2S and rD2L clones (26 ± 7% and
108 ± 6%, respectively). Data represent mean ± SE of three inde-
pendent experiments. *p < 0.05 compared with TRH by paired two-
tailed t-test. (b) Human D2S receptor inhibits ERK1/2 activation.
A representative blot for pERK1/2 of triplicate samples of GH4-hD2S
cells (expressing the human D2S receptor) treated with 1 lmol/L
TRH ± apomorphine (1 lmol/L) as indicated for 15 min. Apomorphine
inhibited TRH-induced pERK1/2 level to 16 ± 4% (mean ± SE of three
independent experiments done in triplicate). Total ERK1/2 staining is
shown as loading control. (c) Quinpirole, a D2-selective agonist,
inhibits TRH-induced pERK1/2 in GH4-rD2S cells. A representative
blot for pERK1/2 in triplicate samples from GH4-rD2S cells treated for
15 min with TRH ± quinpirole (10 lmol/L) as indicated; shown below,
(b) the blot was re-probed for b-actin as a loading control. Quinpirole
reduced TRH-induced pERK1/2 to 61 ± 8%. Data represent
mean ± SE of three independent experiments, each done in triplicate.
Using separate GH4-rD2S and GH4-hD2L clones, we
compared the activity of these D2 receptor variants to
(c) regulate ERK1/2 phosphorylation (Fig. 1). The specific
binding of [3H]-spiperone was 63.3 ± 11.4 and
55.8 ± 2.9 fmol/mg for GH4-rD2S and GH4-hD2L cells,
respectively. Hence, the levels of D2 receptors in these clones
were similar and in the physiological range (50–100 fmol/
mg) (Assie et al. 2005), although somewhat less than
previously reported (Albert et al. 1990; Liu and Albert
1991), presumably due to decline in expression since the
buffer (500 mmol/L Tris, pH 6.8, 2% sodium dodecyl sulfate,
clones were first isolated. In GH4 cells, basal ERK1/2
40 lL/mL 2-mercaptoethanol, 0.1% bromophenol blue, 10% gly- phosphorylation was undetectable under our culture condi-
cerol) and subjected to western blot analysis as described previously tions, suggesting that the 1-h pre-incubation in serum-free
(Banihashemi and Albert 2002) using (1 : 1000) anti-phospho- medium substantially reduced basal ERK1/2 activity. As the
ERK1/2 to detect phospho-ERK1/2. Specific bands were detected basal level of phospho-ERK1/2 was low (Fig. 1a), all
by chemiluminescence. The corresponding band for p42 ERK2 and samples were treated with TRH, with or without dopamine
p44 ERK1 (collectively referred to as ERK1/2) was digitally agonists. TRH markedly increased phospho-ERK1/2 at
quantified using Adobe Photoshop 7, (Adobe Systems Inc., Seattle, 15 min of treatment (Fig. 1a), while the dopamine agonist
WA, USA). The membranes were re-probed with anti-ERK1/2 apomorphine did not affect basal phospho-ERK1/2 levels
(1 : 1000) or anti-b-actin (1 : 1000) as a loading control. The results
(data not shown). In cells expressing the D2S receptor,
were normalized to the control and were presented as mean ± SE.
apomorphine strongly reduced phospho-ERK1/2 to 26 ± 7%
of TRH-stimulated levels (Fig. 1a). In the same series of
Results experiments, apomorphine had no significant effect on
levels of phospho-ERK1/2 in GH4-hD2L cells (108 ± 6%,
Differential signaling of D2S and D2L receptors to ERK1/2 Fig. 1a). Co-addition of the D2 receptor antagonist spiperone
We have previously shown that, in GH4-rD2S cells, the rat (1 lmol/L) blocked D2S-mediated inhibition of ERK1/2
dopamine-D2S receptor mediates inhibition of TRH-induced phosphorylation (data not shown) as did pre-treatment with
ERK1/2 phosphorylation (Banihashemi and Albert 2002). PTX to inactivate Gi/Go-proteins (Banihashemi and Albert
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–1804Differential dopamine-D2S and -D2L signaling 1799
2002). Thus, in GH4 cells, dopamine-D2S receptors signaled (a)
via Gi/Go-proteins to inhibit TRH-induced ERK1/2 phos-
phorylation, while D2L receptors lacked this response.
Further experiments were done to verify the specificity of
D2S receptor-mediated inhibition of ERK1/2 activation.
Increasing pre-incubation time in serum-free medium from
1 h to overnight to reduce phospho-ERK1/2 levels did not
alter D2S receptor signaling, which strongly inhibited TRH-
induced ERK1/2 activation in GH4-rD2S cells (27 ± 6% and
23 ± 6% of TRH-induced level, respectively; data not (b)
shown). In GH4 cells stably transfected with the human
D2S receptor cDNA (GH4-hD2S cells), apomorphine
strongly inhibited ERK1/2 activation (Fig. 1b, 16 ± 4% of
TRH-stimulated level). The human D2S receptor therefore
couples to inhibition of ERK1/2 as effectively as the rat D2S
receptor in separately isolated GH4 clones. Quinpirole, a
D2-selective agonist, partially inhibited phospho-ERK1/2 to
61 ± 8% of TRH-stimulated ERK1/2 levels in GH4-rD2S
cells (Fig. 1c), suggesting that quinpirole may act as a partial
agonist for this response. Thus, D2S-mediated inhibition of
ERK1/2 activation was observed under different pre-incuba-
tion conditions, in GH4 clones separately transfected
with different D2S species homologs and using different
Fig. 2 Inhibition of TRH-induced ERK1/2 activation by somatostatin
D2 agonists. In contrast, dopamine-D2L receptor activation
and carbachol. GH4C1 (a) or GH4-hD2L (b) cells were treated with
did not alter ERK1/2 phosphorylation in GH4-hD2L cells.
1 lmol/L TRH alone or with carbachol (10 lmol/L) or somatostatin
(SST, 200 nmol/L) as indicated for 7 min and level of phosphorylated
Carbachol- and somatostatin-induced inhibition ERK1/2 was measured by western blot using a phospho-specific
of ERK1/2 antibody. GH4-hD2L cells were pre-treated without or with PTX (20 ng/
We further examined whether any other receptor could mL, 12 h) as indicated. The blot was re-probed for b-actin (a) or total
inhibit ERK1/2 phosphorylation. In parental GH4C1 cells, ERK1/2 (b) as a loading control. The blots shown are representative of
the muscarinic agonist carbachol inhibited TRH-induced three independent experiments.
phospho-ERK by 50% (Fig. 2a). Somatostatin also inhibited
TRH-induced ERK1/2 phosphorylation by 50% and the approximately 50% inhibition of forskolin-stimulated cAMP
somatostatin response was reversed by PTX pre-treatment formation in GH4-hD2L cells. As observed for inhibition of
(Fig. 2a), consistent with mediation by Gi/Go-proteins. ERK1/2 phosphorylation, these receptors mediated a smaller
Similarly in GH4-hD2L cells, somatostatin inhibited TRH- inhibition compared with D2 receptors. Pre-treatment with
induced ERK1/2 phosphorylation by 51 ± 6% and this effect PTX blocked inhibition of cAMP mediated by dopamine,
was blocked by PTX (Fig. 2b), as observed in GH4C1 cells. muscarinic and somatostatin receptors, implicating Gi/Go-
The somatostatin response indicates that the signaling proteins. Thus, differences in coupling between D2S and D2L
components required to mediate Gi/Go-dependent inhibition receptors are unlikely to be due to clonal variations as both
of ERK1/2 activation are present in GH4-hD2L cells. Hence, D2S- and D2L-expressing cells display similar cAMP
the lack of effect of D2L receptor activation on ERK1/2 responses and similar levels of receptor expression.
phosphorylation appears to represent restricted signaling of
the receptor. ERK1/2 signaling of D2L pseudosubstrate site mutant
D2L-SS
Inhibition of cAMP by multiple receptors In order to address the molecular basis for differential
In order to address whether other signaling properties of D2S, signaling of the D2S and D2L receptors, we focused on
D2L and other receptors differed in these cell lines, we residues near the 29-amino-acid alternatively spliced domain
examined receptor-mediated inhibition of forskolin-stimula- of the D2L receptor (Fig. 4). As we previously found that the
ted cAMP formation (Fig. 3). In GH4-rD2S and GH4-hD2L D2S receptor is more sensitive to PKC-induced receptor
cells, respectively, activation of dopamine D2S and D2L phosphorylation and uncoupling than the D2L receptor (Liu
receptors using apomorphine did not affect basal cAMP et al. 1992), we hypothesized that sites involved in the PKC
(not shown) but strongly inhibited forskolin-stimulated action on D2 receptors might influence D2 receptor coupling
cAMP. Similarly, carbachol and somatostatin mediated to inhibit ERK1/2 activity. In studies of the resistance of D2L
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–18041800 I. I. Van-Ham et al.
the role of these PKC regulatory sites on the differential
regulation of ERK1/2 by D2 receptor variants, we generated
GH4 clones separately transfected with two mutants: the
PKC pseudosubstrate-deficient hD2L-SS mutant in which
D2L-Ala271/272 residues were mutated to serine or the
PKC-resistant hD2S-triple mutant in which PKC sites at
Thr225-Ser228/229 were eliminated by substitution with
Ala225-Gly228/229 (Fig. 4). Clones were isolated that
expressed similar levels of D2 receptor binding and the
coupling of wild-type or mutant D2 receptors to inhibit TRH-
induced ERK1/2 activation was compared (Fig. 5). As
observed for the D2S receptor (wild-type), the D2S-triple
receptor mutant effectively reduced phospho-ERK1/2 levels
to 23 ± 6% of the TRH-induced level. Conversely, although
wild-type D2L receptor failed to inhibit ERK1/2 signaling,
Fig. 3 PTX-sensitive inhibition of forskolin-stimulated cAMP by the D2L-SS mutant receptor effectively reduced phospho-
dopamine-D2S, D2L, somatostatin, and muscarinic receptors. Cells ERK1/2 to the same extent as the D2S receptor (17 ± 9% of
were incubated for 20 min with no drug, forskolin (1 lmol/L) alone or TRH-stimulated level). These results indicate that mutation
with apomorphine (1 lmol/L) (in GH4-rD2S and GH4-hD2L cells), of the PKC pseudosubstrate domain in the 29-amino-acid
somatostatin (SST, 200 nmol/L) or carbachol (1 lmol/L) (in GH4- insert domain of the D2L receptor can switch the receptor to
hD2L), with or without pre-treatment with PTX (20 ng/mL, 12 h) as inhibit ERK1/2 activity.
indicated. Percent inhibition of forskolin action was calculated and
normalized to the maximum value for GH4-rD2S (D2S, 100%). The
D2-induced inhibition of ERK1/2 phosphorylation
data are expressed as mean ± SEM of three independent experi-
ments done in triplicate. In GH4-hD2L cells, basal and forskolin-sti-
in striatal neurons
mulated cAMP levels were not significantly different from In order to address whether D2 receptors mediate inhibition
corresponding levels in GH4-rD2S cells. of ERK1/2 in non-transformed cells, we studied D2 signaling
in primary striatal cultures. In several systems, depolarization
receptors to uncoupling by PKC, we identified an inhibitory induced by high potassium appears to activate ERK1/2
PKC pseudosubstrate site at Ala271/272 adjacent to the specifically via ERK1/2 kinase-dependent phosphorylation,
29-amino-acid domain, which when mutated to Ser residues including in adrenal chromaffin cells (Rosen et al. 1994),
enhanced D2L sensitivity to PKC-induced phosphorylation at striatal slices (Lindgren et al. 2002) and hippocampal cells
Ser228/229 and resultant desensitization of the receptor to (Impey et al. 1998). To mimic striatal activation in vivo, we
Gi/Go signaling (Morris et al. unpublished data). To address measured pERK1/2 levels following depolarization with high
Fig. 4 Model of human D2S and D2L receptors and mutants. Shown sequences of the D2S and D2L Ni3 region including the 29-amino-acid
above is the Kyte–Doolittle hydrophilicity plot (DNASTAR Protean alternatively spliced domain, point mutations generated by others, and
program, Madison, WI, USA) for the human dopamine-D2L recep- PKC phosphorylation and pseudosubstrate sites mutated in our studies
tor showing predicted hydrophobic transmembrane domains I–VII, (in bold) to generate the D2S-triple or D2L-SS mutant receptors,
and the hydrophilic third intracellular domain, amino- (Ni3) and respectively.
carboxyl-terminal (Ci3) segments. Below is shown the amino acid
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–1804Differential dopamine-D2S and -D2L signaling 1801
(a)
(b)
Fig. 5 Elimination of the PKC pseudosubstrate domain converts D2L
receptor to mediate inhibition of ERK1/2. GH4 cells transfected stably
with rD2S, hD2L, or mutant receptors were treated with 1 lmol/L TRH,
without or with apomorphine (1 lmol/L) as indicated, for 15 min and
probed for phospho-ERK1/2. The mutants were hD2S-triple (D2S
lacking Thr225/Ser228/229 PKC sites) and hD2L-SS (D2L lacking PKC
pseudosubstrate site). Above, a representative western blot of phos-
pho-ERK1/2 (pERK1/2) levels is shown. The blots were re-probed for
b-actin or total ERK1/2 as loading controls. Below, blots from at least
three independent experiments were quantified for phospho-ERK1/2 Fig. 6 Dopamine-D2 induced inhibition of potassium-stimulated
by densitometry and are presented as %TRH-induced response ERK1/2 phosphorylation in striatal cultures. (a) Striatal cultures were
(mean ± SE). incubated for 7 min without (control) or with KCl (40 mmol/L) and
ERK1/2 phosphorylation determined; the blot was re-probed for
b-actin as loading control. (b) Representative blot of phospho-ERK1/2
potassium. Addition of 40 mmol/L KCl markedly induced level, re-probed for total ERK1/2 as loading control. KCl (40 mmol/L)-
basal pERK1/2 activation (Fig. 6a) and this response was stimulated striatal cultures were treated without ()) or with apomor-
inhibited by dopamine agonist apomorphine (64 ± 3% of the phine (1 lmol/L, Apo) or quinpirole (10 lmol/L, Quin). Below, data
KCl response) or D2-selective agonist quinpirole (to from three independent experiments are presented as mean ± SE of
36 ± 10% of KCl response) (Fig. 6b). The basal level of the phospho-ERK1/2 level, normalized to potassium (KCl)-treated
ERK1/2 phosphorylation was low and was not stimulated by control (100%).
apomorphine (87 ± 23% basal level), but was inhibited by
quinpirole (32 ± 10% basal level). Thus, dopamine-D2 receptors differ in inhibitory signaling to ERK1/2 and have
receptors inhibited activation of ERK1/2 in striatal cells, as addressed the specific amino acid residues involved. Differ-
observed in pituitary cells. As striatal neurons express both ential signaling of D2 receptors to ERK1/2 may provide an
D2L and D2S receptors (although mainly D2L), these data explanation for differences in D2S versus D2L receptor
suggest that sufficient D2S receptors are present to mediate function in pituitary cells and neurons.
inhibition of ERK1/2.
D2S receptor-mediated inhibition of ERK1/2
In non-neuronal mesenchymal or glial cells, transfected
Discussion
dopamine-D2S and -D2L receptors couple to stimulation of
Studies of transgenic and knockout mice have shown that ERK1/2 (Luo et al. 1998; Ghahremani et al. 2000; Kim
dopamine-D2S and -D2L receptors have different functions in et al. 2004; Takeuchi and Fukunaga 2004). In contrast, we
the brain and pituitary (Usiello et al. 2000; Iaccarino et al. found that, in neuroendocrine GH4-rD2S cells, dopamine-
2002), but the underlying mechanism for these differences D2S receptors inhibited TRH-induced ERK1/2 activation
remains unclear as their signaling and pharmacology are (Banihashemi and Albert 2002). Furthermore, in GH4-rD2S
virtually identical. Here, we have found that D2S and D2L cells or cultured rat pituitary cells, D2 receptor agonists
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–18041802 I. I. Van-Ham et al.
inhibited basal ERK1/2 activity (Liu et al. 2002), although in the D2S receptor. Although these results suggest that the
our studies basal phospho-ERK was undetectable; hence, no di-alanine motif of the pseudosubstrate site may prevent
inhibition could be detected (Fig. 1a). However, D2S-medi- D2L-induced inhibition of ERK1/2, mutation to Ser creates a
ated inhibition of TRH-induced phospho-ERK1/2 was potential PKC site that could enhance coupling to Gai3.
observed under various pre-incubation conditions, using Alternately, mutation of the pseudosubstrate motif may have
different D2 agonists, and for both rat and human D2S conformational effects that are unrelated to PKC regulation.
receptors. In several independent clones, the D2S receptor The fact that the D2S-triple mutant retained coupling argues
mediated a nearly complete inhibition of TRH-induced that phosphorylation at Ser228/229 sites is not necessary for
ERK1/2 activation (Fig. 1). Hence, the D2S receptor medi- D2S-induced inhibition of ERK1/2. Further studies are
ates inhibition of ERK1/2 in GH4 cells. required to determine the precise actions of the di-alanine
By contrast, D2L receptors failed to inhibit TRH-induced motif in regulating G-protein coupling to inhibition of ERK1/
ERK1/2 activation in GH4-hD2L cells. The lack of coupling 2 signaling.
of D2L receptors to inhibit phospho-ERK1/2 was not due to Alternatively, subtype-selective coupling of D2 receptors
non-functional receptors, as the D2L receptor inhibited to ERK1/2 inhibition may result from differences in receptor
adenylyl cyclase activation (Fig. 3), as well as calcium entry, interactions with other proteins. Several proteins have been
PRL secretion and cell proliferation (Liu et al. 1994; Albert shown to interact with the N-terminal portion of the i3
2002). Furthermore, somatostatin inhibited TRH-induced domain (Ni3) of the D2 receptor, which is located proximal to
ERK1/2 activation in GH4-hD2L cells (Fig. 2b), indicating the site of alternative splicing (Fig. 4). For example, actin-
that Gi/Go-mediated inhibition of ERK1/2 can be triggered binding proteins filamin/ABP280 (Li et al. 2000; Lin et al.
in these cells. 2001) and protein 4.1N (Binda et al. 2002), and signaling
molecules calmodulin and PAR-4 (Bofill-Cardona et al.
Mechanism for differential D2S/D2L signaling 2000; Park et al. 2005), all interact with the D2 receptor at
The mechanism underlying coupling of D2S, but not D2L the Ni3 region and can affect receptor localization or
receptors, to inhibit ERK1/2 phosphorylation may be related coupling. Interestingly, the filamin/ABP280 interaction was
to differences in G-protein specificity between these recep- inhibited by the phosphomimetic mutation of the D2L PKC
tors (Albert and Robillard 2002). In GH4-rD2S cells, site Ser287. In most cases, however, interactions with D2S
apomorphine-induced inhibition of phospho-ERK1/2 and D2L were not compared or no difference was observed;
required Gao and Gai3, but was not dependent on Gai2 or hence, the role of these interactions in D2S versus D2L
Gbc subunits (Banihashemi and Albert 2002). Upon activa- signaling specificity remains to be addressed.
tion of D2S receptors, Gao couples to inhibit B-Raf-mediated
basal ERK1/2 activation, possibly via Rap-GAP (Jordan Roles for differential D2 coupling in vivo
et al. 1999), while Gai3 signals to inhibit TRH-stimulated Studies in D2 receptor knockout mice have demonstrated
cRaf-ERK1/2 kinase-ERK1/2 pathway. For inhibition of that the D2 receptor is the predominant inhibitory regulator
adenylyl cyclase and PRL secretion in GH4 cells, the D2L of lactotroph growth, PRL synthesis and secretion in vivo
receptor preferentially coupled to Gai2, while the D2S (Kelly et al. 1997; Saiardi et al. 1997). As the ERK1/2
receptor coupled to Gai3, Gai2 and Gao (Liu et al. 1994; pathway enhances PRL gene transcription (Kievit et al.
Albert 2002). Distinct G-protein selectivity may account for 2001; Liu et al. 2005), D2S-induced inhibition of ERK1/2
the differential signaling of the D2 receptor variants to activation could account for negative regulation of PRL
inhibition of ERK1/2 activation (Albert 2002). Mutations synthesis by D2 receptors in vivo. To address differential
in the 29-amino-acid insert domain that shift D2L receptor– roles of D2 receptors in vivo, transgenic mice over-
G-protein specificity from Gi2 to Gi3 or Go (e.g., S259/262A expressing D2S or D2L receptor cDNA driven by the
or D249V (Guiramand et al. 1995) may enhance D2L coupling PRL promoter were generated (Iaccarino et al. 2002). The
to inhibition of ERK1/2, but this remains to be determined. PRL-D2S mice displayed dramatic reductions in lactotroph
We identified a potential PKC pseudosubstrate site located cell number, PRL RNA and pituitary protein content.
adjacent to the 29-amino-acid insert of the dopamine-D2L Paradoxically, phosphorylation of ERK1/2 in anterior
receptor at Ala271/272 that influences PKC-mediated phos- pituitary was increased slightly, possibly due to ectopic
phorylation at Ser228/229, the major PKC phosphorylation activation as lactotroph number is greatly reduced in PRL-
sites of the D2 receptor (Namkung and Sibley 2004). D2S mice. By contrast, D2L-over-expressing pituitaries had
Mutation of the pseudosubstrate site from Ala271/272 to increased PRL content and slightly increased lactotroph cell
Ser in D2L-SS increases PKC-induced Ser228/229 phos- number compared with wild-type pituitaries but little
phorylation and uncoupling of the dopamine-D2L receptor change in ERK1/2 activity, consistent with the lack of
(Morris et al. unpublished data). This mutation of the D2-induced inhibition of ERK1/2 that we observed in GH4-
pseudosubstrate site resulted in a D2L receptor (D2L-SS) hD2L cells. Using GH4 cells that express only D2S or D2L
that coupled to inhibition of ERK1/2 activation as robustly as receptors, we were able to address the signaling of these D2
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–1804Differential dopamine-D2S and -D2L signaling 1803
receptor variants in isolation, and show a clear difference in Albert P. R., Neve K. A., Bunzow J. R. and Civelli O. (1990) Coupling
signaling to inhibit ERK1/2 phosphorylation, providing a of a cloned rat dopamine-D2 receptor to inhibition of adenylyl
cyclase and prolactin secretion. J. Biol. Chem. 265, 2098–2104.
possible mechanism for differential actions of D2S and D2L
Assie M. B., Consul-Denjean N., Koek W. and Newman-Tancredi A.
receptors in vivo. (2005) Differential in vivo inhibition of [3H]nemonapride binding
There is conflicting evidence as to whether D2 receptors by atypical antipsychotics in rat striatum, olfactory lobes, and
inhibit or stimulate ERK1/2 phosphorylation in vivo. In frontal cortex. Pharmacology 75, 63–68.
striatal slice cultures (Yan et al. 1999) or striatal or Banihashemi B. and Albert P. R. (2002) Dopamine-D2S receptor inhi-
bition of calcium influx, adenylyl cyclase, and mitogen-activated
mesencephalic primary cultures (Wang et al. 2005; Kim
protein kinase in pituitary cells: distinct Galpha and Gbetagamma
et al. 2006), D2 receptor agonists increased basal ERK1/2 requirements. Mol. Endocrinol. 16, 2393–2404.
phosphorylation. However, in intact striatum, acute treatment Ben-Jonathan N. and Hnasko R. (2001) Dopamine as a prolactin (PRL)
with a D2 receptor antagonist increased pERK1/2 staining inhibitor. Endocr. Rev. 22, 724–763.
in vivo (Gerfen et al. 2002). We examined primary rat Binda A. V., Kabbani N., Lin R. and Levenson R. (2002) D2 and D3
dopamine receptor cell surface localization mediated by interaction
embryonic striatal cultures for the regulation of depolariza-
with protein 4.1N. Mol. Pharmacol. 62, 507–513.
tion-dependent ERK1/2 phosphorylation by dopamine Bofill-Cardona E., Kudlacek O., Yang Q., Ahorn H., Freissmuth M. and
agonists. This model of activity-dependent dopaminergic Nanoff C. (2000) Binding of calmodulin to the D2-dopamine
regulation may be more representative of striatal neurons receptor reduces receptor signaling by arresting the G protein
in vivo than non-stimulated cultures. High potassium depo- activation switch. J. Biol. Chem. 275, 32672–32680.
Civelli O., Bunzow J. R. and Grandy D. K. (1993) Molecular diversity of
larization activates calcium influx through voltage-sensitive
the dopamine receptors. Annu. Rev. Pharmacol. Toxicol. 33,
calcium channels, which leads to ERK1/2 activation (Rosen 281–307.
et al. 1994). Depolarization of striatal cultures with elevated Gerfen C. R., Miyachi S., Paletzki R. and Brown P. (2002) D1 dopamine
KCl (40 mmol/L) increased ERK1/2 phosphorylation and receptor supersensitivity in the dopamine-depleted striatum results
activation of D2 receptors robustly inhibited stimulated, as from a switch in the regulation of ERK1/2/MAP kinase. J. Neu-
rosci. 22, 5042–5054.
well as basal ERK1/2 activation. D2 receptors can couple to
Ghahremani M. H., Forget C. and Albert P. R. (2000) Distinct roles
either inhibition (for D2S) or stimulation (both D2S and D2L) for Galpha(i)2 and Gbetagamma in signaling to DNA synthesis
of ERK1/2 phosphorylation and both pathways may be and Galpha(i)3 in cellular transformation by dopamine D2S
present in these heterogeneous primary cultures. In that case, receptor activation in BALB/c 3T3 cells. Mol. Cell. Biol. 20,
the effect of D2 receptor stimulation may depend on the ratio 1497–1506.
Guiramand J., Montmayeur J. P., Ceraline J., Bhatia M. and Borrelli E.
of D2S/D2L receptors, with inhibition of ERK1/2 observed
(1995) Alternative splicing of the dopamine D2 receptor directs
only in D2S-expressing cells. specificity of coupling to G-proteins. J. Biol. Chem. 270, 7354–
In summary, dopamine-D2S receptors inhibit TRH-stimu- 7358.
lated ERK1/2 phosphorylation in GH4 cells, while a Iaccarino C., Samad T. A., Mathis C., Kercret H., Picetti R. and Borrelli
di-alanine motif adjacent to the 29-amino-acid insert in the E. (2002) Control of lactotrop proliferation by dopamine: essential
role of signaling through D2 receptors and ERKs. Proc. Natl Acad.
D2L receptor prevents its signaling to inhibit ERK1/2
Sci. USA 99, 14530–14535.
activation. Similarly, dopamine-D2 receptor activation inhib- Impey S., Obrietan K., Wong S. T., Poser S., Yano S., Wayman G.,
ited basal and depolarization-induced ERK1/2 phosphoryla- Deloulme J. C., Chan G. and Storm D. R. (1998) Cross talk
tion in striatal cultures. Our finding that dopamine-D2S and between ERK and PKA is required for Ca2+ stimulation of CREB-
-D2L receptors differ in this important signaling pathway dependent transcription and ERK nuclear translocation. Neuron 21,
869–883.
may provide a novel target for subtype-selective inhibitors of
Ito K., Haga T., Lameh J. and Sadee W. (1999) Sequestration of dop-
D2 receptor signaling. amine D2 receptors depends on coexpression of G-protein-coupled
receptor kinases 2 or 5. Eur. J. Biochem. 260, 112–119.
Jordan J. D., Carey K. D., Stork P. J. and Iyengar R. (1999) Modulation
Acknowledgements of rap activity by direct interaction of Galpha(o) with Rap1
GTPase-activating protein. J. Biol. Chem. 274, 21507–21510.
Supported by grants from CIHR and Ontario Mental Health
Kelly M. A., Rubinstein M., Asa S. L. et al. (1997) Pituitary lactotroph
Foundation to PRA.
hyperplasia and chronic hyperprolactinemia in dopamine D2
receptor-deficient mice. Neuron 19, 103–113.
Khan Z. U., Mrzljak L., Gutierrez A., de la Calle A. and Goldman-Rakic
References
P. S. (1998) Prominence of the dopamine D2 short isoform in
Albert P. R. (1994) Heterologous expression of G protein-linked dopaminergic pathways. Proc. Natl Acad. Sci. USA 95, 7731–
receptors in pituitary and fibroblast cell lines. Vitam. Horm. 48, 7736.
59–109. Kievit P., Lauten J. D. and Maurer R. A. (2001) Analysis of the role of
Albert P. R. (2002) G protein preferences for dopamine D2 inhibition of the mitogen-activated protein kinase in mediating cyclic-adenosine
prolactin secretion and DNA synthesis in GH(4) pituitary cells. 3¢,5¢-monophosphate effects on prolactin promoter activity. Mol.
Mol. Endocrinol. 16, 1903–1911. Endocrinol. 15, 614–624.
Albert P. R. and Robillard L. (2002) G protein specificity. Traffic Kim S. J., Kim M. Y., Lee E. J., Ahn Y. S. and Baik J. H. (2004) Distinct
direction required. Cell. Signal. 14, 407–418. regulation of internalization and mitogen-activated protein kinase
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–18041804 I. I. Van-Ham et al.
activation by two isoforms of the dopamine D2 receptor. Mol. Luo Y., Kokkonen G. C., Wang X., Neve K. A. and Roth G. S. (1998)
Endocrinol. 18, 640–652. D2 dopamine receptors stimulate mitogenesis through pertussis
Kim S. Y., Choi K. C., Chang M. S., Kim M. H., Kim S. Y., Na Y. S., toxin-sensitive G proteins and Ras-involved ERK and SAP/JNK
Lee J. E., Jin B. K., Lee B. H. and Baik J. H. (2006) The dopamine pathways in rat C6-D2L glioma cells. J. Neurochem. 71, 980–
D2 receptor regulates the development of dopaminergic neurons 990.
via extracellular signal-regulated kinase and Nurr1 activation. Missale C., Nash S. R., Robinson S. W., Jaber M. and Caron M. G.
J. Neurosci. 26, 4567–4576. (1998) Dopamine receptors: from structure to function. Physiol.
Li M., Bermak J. C., Wang Z. W. and Zhou Q. Y. (2000) Modulation of Rev. 78, 189–225.
dopamine D(2) receptor signaling by actin-binding protein (ABP- Namkung Y. and Sibley D. R. (2004) Protein kinase C mediates phos-
280). Mol. Pharmacol. 57, 446–452. phorylation, desensitization, and trafficking of the D2 dopamine
Lin R., Karpa K., Kabbani N., Goldman-Rakic P. and Levenson R. receptor. J. Biol. Chem. 279, 49533–49541.
(2001) Dopamine D2 and D3 receptors are linked to the actin Park S. K., Nguyen M. D., Fischer A., Luke M. P., Affar el B., Dief-
cytoskeleton via interaction with filamin A. Proc. Natl Acad. Sci. fenbach P. B., Tseng H. C., Shi Y. and Tsai L. H. (2005) Par-4 links
USA 98, 5258–5263. dopamine signaling and depression. Cell 122, 275–287.
Lindgren N., Goiny M., Herrera-Marschitz M., Haycock J. W., Hokfelt Rosen L. B., Ginty D. D., Weber M. J. and Greenberg M. E. (1994)
T. and Fisone G. (2002) Activation of extracellular signal-regulated Membrane depolarization and calcium influx stimulate MEK and
kinases 1 and 2 by depolarization stimulates tyrosine hydroxylase MAP kinase via activation of Ras. Neuron 12, 1207–1221.
phosphorylation and dopamine synthesis in rat brain. Eur. J. Saiardi A., Bozzi Y., Baik J. H. and Borrelli E. (1997) Antiproliferative
Neurosci. 15, 769–773. role of dopamine: loss of D2 receptors causes hormonal dysfunc-
Liu Y. F. and Albert P. R. (1991) Cell-specific signaling of the 5-HT1A tion and pituitary hyperplasia. Neuron 19, 115–126.
receptor. Modulation by protein kinases C and A. J. Biol. Chem. Takeuchi Y. and Fukunaga K. (2004) Dopamine D2 receptor activates
266, 23689–23697. extracellular signal-regulated kinase through the specific region in
Liu Y. F., Civelli O., Grandy D. K. and Albert P. R. (1992) Differ- the third cytoplasmic loop. J. Neurochem. 89, 1498–1507.
ential sensitivity of the short and long human dopamine D2 Usiello A., Baik J. H., Rouge-Pont F., Picetti R., Dierich A., LeMeur M.,
receptor subtypes to protein kinase C. J. Neurochem. 59, 2311– Piazza P. V. and Borrelli E. (2000) Distinct functions of the two
2317. isoforms of dopamine D2 receptors. Nature 408, 199–203.
Liu Y. F., Jakobs K. H., Rasenick M. M. and Albert P. R. (1994) G Wang Y. H. and Maurer R. A. (1999) A role for the mitogen-activated
protein specificity in receptor–effector coupling. Analysis of the protein kinase in mediating the ability of thyrotropin-releasing
roles of Go and Gi2 in GH4C1 pituitary cells. J. Biol. Chem. 269, hormone to stimulate the prolactin promoter. Mol. Endocrinol. 13,
13880–13886. 1094–1104.
Liu J. C., Baker R. E., Sun C., Sundmark V. C. and Elsholtz H. P. Wang C., Buck D. C., Yang R., Macey T. A. and Neve K. A. (2005)
(2002) Activation of Go-coupled dopamine D2 receptors inhibits Dopamine D2 receptor stimulation of mitogen-activated protein
ERK1/ERK2 in pituitary cells. A key step in the transcriptional kinases mediated by cell type-dependent transactivation of receptor
suppression of the prolactin gene. J. Biol. Chem. 277, 35819– tyrosine kinases. J. Neurochem. 93, 899–909.
35825. Yan Z., Feng J., Fienberg A. A. and Greengard P. (1999) D(2) dopamine
Liu J. C., Baker R. E., Chow W., Sun C. K. and Elsholtz H. P. (2005) receptors induce mitogen-activated protein kinase and cAMP
Epigenetic mechanisms in the dopamine D2 receptor-dependent response element-binding protein phosphorylation in neurons.
inhibition of the prolactin gene. Mol. Endocrinol. 19, 1904–1917. Proc. Natl Acad. Sci. USA 96, 11607–11612.
2007 The Authors
Journal Compilation 2007 International Society for Neurochemistry, J. Neurochem. (2007) 102, 1796–1804You can also read
