| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
* Department of Molecular Pharmacology and Pathology, Consorzio Mario Negri Sud, Istituto di Ricerche Farmacologiche ‘Mario Negri’, Santa Maria Imbaro, Italy;
I.N.M. Neuromed, Pozzilli, Italy; and
Brain Science Institute, RIKEN, Wako-Shi, Saitama, Japan
1Correspondence: Consorzio Mario Negri Sud, via Nazionale 66030 S. Maria Imbaro, Italy. E-mail: deblasi{at}cmns.mnegri.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Purkinje cells • mGlu1 receptor • receptor internalization • GRK • antisense knock down
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
.
Signal transduction mediated by GPCR must be properly regulated in
order to prevent overstimulation, achieve signal termination, and
render the receptor responsive to subsequent stimuli. One of the major
mechanisms of receptor regulation is homologous (i.e.,
agonist-dependent) desensitization, a process that occurs rapidly after
binding of the agonist to the receptor. GPCRs are also desensitized in
an agonist-independent manner (heterologous desensitization) by second
messenger-dependent protein kinases PKA and PKC (3
, 4)
.
Two types of proteins contribute to homologous desensitization of
GPCRs: 1) G-protein-coupled receptor kinases (GRKs), which
phosphorylate GPCR occupied by the agonist, and 2) their
functional cofactors, named (ß)-arrestin (4
5
6)
. The
multigene family of GRKs consists of six members, GRK1 to GRK6, which
are classified into three subfamilies on the basis of their sequence
homology. GRK1 (rhodopsin kinase) is the only member of the first
subfamily; GRK2 and GRK3 (previously known as ß-adrenergic receptor
kinases, ßARKs) form the second subfamily (‘ßARK subfamily’);
GRK4, GRK5, and GRK6 form the third subfamily (‘GRK4 subfamily’).
Human GRK4 has two sites of alternative splicing—one at the
amino-terminal domain (exon 2), the other at the carboxyl-terminal
domain (exon 15), resulting in four splice variants (7
, 8)
. The existence of GRK4 splice variants in rats, which differs
from those found in humans, has been reported (9)
. GRK2,
-3, -5, and -6 are widely distributed in different cell types and
tissues, where they mediate homologous desensitization of a variety of
GPCRs. In contrast, the expression of GRK1 and GRK4 is strictly
localized, suggesting effective receptor substrate selectivity for
these two kinases. GRK1 is expressed only in the retina, where it
regulates phototransduction (4
, 5)
. GRK4 expression has so
far only been described in testis, where it was found in germinal cells
and spermatozoa (8
, 10)
. However, GRK4 mRNA is also
detectable by reverse transcription-polymerase chain reaction (RT-PCR)
in other tissues, including brain and kidney (9
, 11)
. The
aim of the present study was to identify a novel cellular site of
expression and a physiologically relevant receptor target for GRK4.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
analysis. RP1 probe was generated from
full-length GRK4
in pBluescript II SK+ linearized with
EcoRI and transcribed with T7 RNA polymerase (bp 249 to
1772). The amino-terminal region of human GRK4
(bp -16 to 480) was
amplified by PCR and subcloned into the plasmid PcrII topo (Invitrogen,
San Diego, Calif.). This construct was linearized with EcoR
V and transcribed with SP6 RNA polymerase (sense control probe, RP2) or
was linearized by BamHI cutting and transcribed with T7 RNA
polymerase (antisense probe, RP3) transcription kit (Promega
Corporation, Madison, Wis.). Briefly, the digoxigenin labeling was
carried out using 1 µg of linearized cDNA, incubated for 2 h at
37°C with 1x transcription buffer containing the nucleotide mix (10
mM dATP, 10 mM CTP, 10 mM GTP, 6.5 mM UTP, 3.5 mM DIG-11-UTP, pH 7.5),
1 U/µl RNAsin, and 1 U/µl of the appropriate RNA polymerase into a
final volume of 20 µl. After transcription, the template DNA was
removed by digestion with 1 U/µl of DNAseI-RNase free (15 min at
37°C), whereas RNA was precipitated in 0.4 M LiCl, 75% of ethanol
and resuspended in RNase-free water. Thirty-day-old rats were
anesthetized with Nembutal (40 mg/Kg body weight, intraperitoneally),
perfused transcardiacally first with phosphate-buffered saline (PBS)
and then with 4% paraformaldehyde in PBS. The brain was excised
quickly from the skull and immersed overnight in the same fixative. The
brain was manually sectioned to obtain slices of 
1 mm in thickness.
The tissue was permeabilized by 30 min of digestion at room temperature
in 10 µg/ml of proteinase K. Digestion was blocked by incubation with
2 mg/ml of glycine in PBS plus 0.1% Tween-20 (PTW). The slices rinsed
in PTW were acetylated by incubation in 0.25% acetic anhydride in 100
mM triethanolamine, 150 mM NaCl pH 8, then washed again in PTW. After a
new fixation with paraformaldehyde, the sections were prehybridized
overnight at 60°C, in 50% formamide, 5x SSC pH 4.5, 1 mg/ml of
total yeast RNA, 100 µg/ml heparin, 1x Denhardt (0.02% Ficoll 400,
0.02% poly-vinyl-pyrrolidone, 0.02% BSA, 0.1% Tween-20, 0.1% CHAPS,
5 mM EDTA). The hybridization was carried out for 18 h at 60°C
in prehybridization solution containing the ribobrobe (RP1, RP2, or
RP3), at the concentration of 1 µg/ml. The specimens were then washed
at high stringency (0.2x SSC 60°C), saturated with 2% Boehringer
Mannheim Blocking Reagent, 20% sheep serum in maleic acid buffer
(MAB), (100 mM maleic acid, 150 mM NaCl pH 7.5), and incubated
overnight at 4°C with alkaline phosphatase-conjugated goat
anti-digoxigenin IgG 1:2000 in the same solution. The next day the
slices were washed abundantly with MAB and developed with
5-bromo-4-chloro-3-indolyl-phosphate and nitro-blue tetrazolium. The
staining was fixed with 100 mM MOPS pH 7.4, 2 mM EGTA, 1 mM
MgSO4, 3.7% formaldehyde, dehydrated with
ethanol, and photographed using a stereo microscope equipped with
camera (Yashica).
Northern and Western blot analysis
Northern blot was performed as described previously
(8)
. A large cDNA fragment generated by modified PCR
reaction (bp 815-1806) was used as a GRK4 probe. The reaction mixture
(10 µl final volume) contained 15 pmol of dATP, dGTP, and dTTP, 7.5
pmol of dCTP, 2.5 µCi (8.3 pmol) of 32P-dCTP
(3000 Ci/mmol), 200 ng of primers (forward 5'-CGA GCC ACA GGA AAA ATG
TA-3', and reverse 5'-GTG GGA CGA GAC GCT AAC A-3'), 1 µl of 10x PCR
buffer 5 ng of template, and 2 U of taq polymerase. The labeled probe
was purified over a Sephadex G-50 spun column and used for
hybridization. After hybridization, the membrane was washed at high
stringency [2x SSC, 1% sodium dodecyl sulfate (SDS) at 60°C for 30
min] and subjected to autoradiography at -80°C for 1–10 days. All
observations were confirmed in at least two separate experiments.
Western blot was performed as described previously (8)
,
using 0.2 µg/ml of anti-GRK4 (K-20), 1 µg/ml of
anti-mGlu1 antibodies, and developed with
alkaline phosphatase-conjugated goat anti-rabbit IgG (1:5000),
5-bromo-4-chloro-3-indolyl-phosphate, and nitro-blue tetrazolium.
Immunohistochemistry
Rats anesthetized as above were perfused transcardiacally with
PBS and then with 10% formalin in PBS. The brain and the testis were
immersed overnight in the same fixative, embedded in paraffin, and
serial sections (5 µm) were cut and used for immunohistochemistry
(13)
. Sections were pretreated with 0.3%
H2O2 in ethanol at 4°C
for 15 min to inhibit endogenous peroxidase activity and permeabilized
by washing with PBS containing 0.05% Tween-20. Samples were saturated
with 50% normal goat serum (NGS) for 10 min at 37°C and then
incubated overnight at 4°C with the specific antibody diluted in PBS
containing 1% NGS. Two different anti-GRK4 antibodies specific for the
GRK4 and GRK4ß (K-20), and for GRK4 and GRK4
(I-20), an
anti-GRK5 and an anti-GRK6 (all from Santa Cruz Biotechnology, Santa
Cruz, Calif.), were used at the concentration of 1 µg/ml.
Anti-mGlu1 from Upstate Biotechnology (Lake
Placid, N.Y.) was used at the concentration 12 µg/ml. The next day
sections were incubated with anti-rabbit IgG biotin-conjugated
(Calbiochem, San Diego, Calif.) (1:100) in 50% NGS for 30 min at room
temperature. Detection of immunoreactivity was accomplished using a
vectastain elite ABC kit (Vector Laboratories, Burlingame, Calif.).
After each incubation step, sections were carefully washed with PBS.
The immunolocalization was visualized using 0.04% of
3,5'-diaminobenzidine, 0.33 
H2O2 in PBS; tissues were
counterstained with methylene blue. Negative control sections were
processed in the same way using the primary antibody preadsorbed with
an excess of antigenic peptide or 1% NGS instead of the primary
antibody. Photomicrographs were taken using a Zeiss Axiophot microscope
(Carl Zeiss Inc., Jena, Germany).
Kinase-dead GRK4 mutant
A mutated GRK4 (named GRK4-(K216M, K217M), predicted to be
deficient in kinase activity (14)
, was prepared using a
PCR strategy to mutate two adjacent lysine residues present in GRK4 to
methionine. Briefly, two overlapping antiparallel oligos each bearing
the same three mutated bases were used to amplify GRK4 cDNA in pCMV
vector using the Pfu DNA polymerase (Stratagene, San Diego,
Calif.). The PCR product was digested with DpnI to destroy
the parental DNA template and transformed into competent XL1-Blue
E. Coli. The mutated bases were checked by specific
restriction analysis using AluI and sequencing of the new
clone. The lack of kinase activity of GRK4-(K216M, K217M) was assessed
by in vitro phosphorylation assay. Cytosolic proteins from
HEK293 cells transfected with GRK4 or GRK4-(K216M, K217M) were used to
phosphorylate rhodopsin from urea-treated, purified rod outer segments
(ROS) (8)
.
Transfection and measurement of IP levels
HEK293 cells were transfected as described (8)
. One
day after transfection, the cells were washed in PBS and incubated for
18 h with DMEM/Glutamax-1 (Life Technologies, Inc., Paisley,
U.K.), then washed and incubated overnight with MEM/Glutamax-1
containing 3 µCi/well of myo-[3H]inositol
(Amersham, Little Chalfont, U.K.). On the third day, inositol phosphate
(IP) production was measured as described (15)
. Briefly,
cells were washed twice and incubated for 1–2 h at 37°C in 1 ml of
HEPES-buffered saline (146 mM NaCl, 4.2 mM KCl, 0.5 mM
MgCl2, 0.1% glucose, 20 mM HEPES pH 7.4), washed
again with HEPES-buffered saline, and preincubated for 15 min in the
same buffer containing 10 mM LiCl, 1.8 U/ml glutamic pyruvic
transaminase, and 2 mM Na-pyruvate. The stimulus was carried out for 30
min with 100 µM of quisqualate, unless otherwise indicated. The
reaction was stopped by replacing the incubation medium with 1 ml of
ice-cold perchloric acid (5%). Inositol phosphates were separated by
an ion exchange chromatography column of Dowex AG1-X8 (formiate form)
(200–400 mesh, 350 µl bed volume). Usually 1 x
106 cells were cotransfected with 1 µg of
mGlu1 plasmid along with 5 µg of GRK or empty
vector. For mGlu1a, 5 µg of the plasmid
encoding the glutamate transporter EAAC1 (16)
was
included. Human mGlu1 cDNA in pcDNA 3 was kindly
provided by M. Corsi, Glaxo Wellcome, Verona. The human EAAC1 cDNA in
pRK was kindly provided by J. P. Pin, CNRS, Montpellier and
M. A. Hediger, Harvard Medical School (Boston, Mass.).
mGlu1a phosphorylation assay
Transiently transfected HEK293 cells were lysed in
immunoprecipitation buffer (50 mM Tris-HCl pH 8, 150 mM NaCl, 5 mM
EDTA, 1% Nonidet P-40, 0.1% SDS, 10 mM NaF, 10 mM disodium
pyrophosphate with protease inhibitors) for 90 min at 4°C with
continuous rocking, followed by centrifugation (200,000 g,
40 min). An equal amount of protein extracts (1,500 µg) were
incubated overnight with 2 µg of anti-mGlu1a
antibody plus 40 µl of protein A. After washing five times with
immunoprecipitation buffer, the immunoprecipitates were resolved in
SDS/8% polyacrylamide gel electrophoresis and transferred to
nitrocellulose. Blot was developed first with 0.2 µg/ml
anti-phosphoserine antibody (Biomol, Hamburg, Germany) and then with
the anti-mGlu1a to determine the amount of
immunoprecipitated receptor in each lane (17)
.
Purkinje cell rich primary culture
Cerebellar neurons were prepared from Wistar rats as described
previously (18
, 19)
, with minor modifications, to obtain a
Purkinje cell-rich culture. Seven-day-old pups were killed by cervical
dislocation and the cerebella was excised and minced with a scalpel.
The cerebellar cells were disagregated with 0.025% trypsin and 0.01%
DNAseI in Krebs Ringer plus 0.03% MgSO4, 0.3%
BSA for 15 min at 37°C. The cells were washed with the same buffer
containing 40 µg/ml of trypsin inhibitor and 0.01% DNAseI and
dissociated by repeated passage through a fine-tipped pipette. The cell
suspension was centrifuged at 400 g for 2 min and cells were
resuspended carefully in 2 ml of the same buffer. After 30–45 min, the
Purkinje cells are enriched from granules by gravity. The upper part of
the suspension (granules) was removed very carefully and the sediment
(Purkinje cells) was rinsed with culture medium. Recovered cells were
plated at a density of 20–25 x 104
cells/cm2 onto poly-D-lysine-coated chamber
slides in serum-free defined medium: Eagle’s medium supplemented with
1 mg/ml BSA, 10 µg/ml insulin, 0.1 nM L-thyroxin, 0.1 mg/ml
transferrin, 1 µg/ml aprotinin, 30 nM selenium, 100 µg/ml
streptomycin, and 100 U/ml penicillin. The cultures were maintained in
a humidified atmosphere of 5% CO2 in air at
37°C. The cultures, which consisted of 3–4% of Purkinje cells
(assessed by calbindin immunostaining), were used after 15–20 days
in vitro.
Confocal analysis of mGlu1a and GRKs
immunofluorescence
HEK293 cells transfected as above and 20-day-old Purkinje cell
primary cultures were fixed with 4% paraformaldehyde in PBS for 15 min
at room temperature. The autofluorescence was quenched by incubation
for 30 min in 50 mM NH4Cl, 50 mM glycine in PBS
and nonspecific interactions were blocked by treatment with blocking
solution (0.05% saponin, 0.5% BSA in PBS) for 30 min at room
temperature. Cells were incubated (13)
overnight at 4°C
with K-20 (1 µg/ml) or anti-GRK2/GRK3 (Upstate Biotechnology) (5
µg/ml) antibody in blocking solution. The chamber slides were then
incubated with blocking solution containing Alexa-488 anti-rabbit or
anti-mouse IgG (Molecular Probes, Eugene, Oreg.) (1:400) for 1 h
at room temperature. The anti-mGlu1 (Upstate
Biotechnology) conjugated to the fluorochrome Alexa-594 (Protein
labeling kit from Molecular Probes) was used at the concentration of 3
µg/ml. Each incubation step was carried out in the dark, followed by
careful washes with PBS (6 times/3 min each). After immunostaining the
coverslips were mounted on slides with Mowiol 4–88, and observed with
a Zeiss Axiophot (Carl Zeiss Inc.). Colocalization of the
mGlu1 and GRK antigens was assessed by INSIGHT
PLUS laser scanning confocal microscope system (Meridian, Oketos)
equipped with an Olympus IMT-2 inverted microscope. Ten groups of
optical Z-section serial slices from each experiment were taken with
0.5 µm Z-steps from the top to the bottom of the specimen.
Fluorescent images were recorded using a Dage CCD camera, and stored
directly on computer. Merging of both immunofluorescence generated the
colocalization maps by color. Image processing was performed on a
Macintosh computer using the public domain NIH Image program version
1.62 and NIH Image macros called ‘Confocal’ and ‘Measurement
macros’ (developed at the U.S. National Institutes of Health and
available on the Internet at ftp://codon.nih.gov/pub/nih-image).
Colocalization was quantified by measuring the number of pixels labeled
with the kinase, with the receptor and the pixels labeled by both
antigens; the results are expressed as percentage of the pixels of each
antigen.
A different set of experiments were performed using a Zeiss LSM 510 Laser Scanning Microscope equipped with an Axiovert 100 M-BP. Optical Z-sections from each experimental conditions were taken with 0.3 µm Z-steps from the top to the bottom of the cells. Using the ‘Display-Profile’ option of the LSM510 program, we traced two lines for each cell, which crossed the cells side by side. The intensity of fluorescence was measured along each line and expressed in arbitrary units (from 0 to 255). The internalization of mGlu1a receptor was quantified by calculating the percentage of immunofluorescence present in the cytosol vs. membrane. In this calculation membrane immunofluorescence is taken as 100%.
Antisense oligonucleotide treatment and calcium imaging
Cerebellar neurons were prepared and plated on coverslips
following a standard procedure (19)
, except that culture
medium contained 10% fetal calf serum. Cells were maintained in
vitro for 2–3 wk, at which time 2-end phosphorothiated
oligonucleotides at 1 µM final concentration for GRK4 and 2 µM for
GRK2 (refreshed daily) were added for 4–5 days. The GRK4 rat antisense
oligonucleotide and the scrambled mismatch control oligonucleotide had
the following sequence: 5'-GTTCTCCAGTTCCATGATCC-3' and
5'-GTTCCTGACAAGTTCTCTCC-3', respectively. The GRK2 rat antisense
oligonucleotide had the following sequence: 5'-CCG CTC GTT CAG AGC CAG
GG-3' (20)
. Purkinje cells primary culture treated with
antisense oligonucleotide were immunolabeled both with the GRK4
antibody (K-20) and a mouse anti-calbindin antibody (Sigma, St. Louis,
Mo.) using Alexa-594 anti-rabbit IgG and Alexa-488 anti-mouse secondary
antibodies (Molecular Probes). Immunostained cultures were imaged with
a cooled CCD attached to an Olympus fluorescence microscope. Background
subtracted fluorescence intensities were measured over regions
involving the cell body and proximal dendrites of individual Purkinje
cells. GRK4 immunofluorescence was normalized to calbindin
immunofluorescence and then compared between cultures. The effect of
antisense oligonucleotide treatments were assessed by measuring the
total immunofluorescence (cytosol plus membranes) of the GRK under
examination in control cells and in antisense treated Purkinje cells.
Imaging of intracellular calcium concentration was accomplished by
radiometric imaging of Fura-2 fluorescence. Cells were loaded with
Fura-2 by incubation with Fura-2 AM (5 µM; dissolved in DMSO at final
concentration of 0.1% v/v; 60 min at room temperature) in a modified
Hanks solution consisting of 137 mM NaCl, 2.5 mM KCl, 1 mM
MgSO4, 1.18 mM
NaH2PO4, 2 mM
CaCl2, 5.55 mM L-Glucose, 11.6 mM
NaHCO3. The fura-2AM containing solution was then
removed and cells were incubated in dye-free saline solution for at
least an additional 60 min to allow for complete cleavage of fura-2AM
into Fura-2. Coverslips with cerebellar cultures were mounted on an
upright fixed stage microscope and continuously superfused with the
above Hanks solution containing 1 µM TTX.
(S)-3,5-Dihydroxyphenylglycine (3,5-DHPG, Tocris) was bath-applied by
means of a two-way tap. Fluorescence of Fura-2 was excited by
epi-illumination via a 40x water immersion objective with light
provided by a monochromator and detected by a cooled CCD under control
of Axon Imaging Workbench software (Axon Instruments, Forster City,
Calif.). Fluorescence images were corrected for background fluorescence
(measured from image regions free of dye). Changes in
[Ca2+]i in regions
involving the cell body and proximal dendrites of individual Purkinje
cells were calculated from Fura-2 fluorescence excited alternatively at
340 nm and 380 nm (21)
.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Using the GRK4 cRNA probe RP1, which recognizes
all the GRK4 splice variants, we observed an intense staining inside
the bovine testis tubuli (Fig. 1a
, arrow); the intertubular spaces, including the Leydig
cells, were completely unlabeled. A similar pattern of hybridization
was found on rat testis (not shown), confirming previous results
(9)
. Identical results were obtained with the probe RP3,
which includes only the amino-terminal region of GRK4 (data not shown).
Specificity of this signal was confirmed by the following controls:
1) hybridization was prevented by pretreating the samples
with RNase H (Fig. 1b
), and 2) the use of the
sense cRNA probe RP2 gave no hybridization signal (not shown).
Immunohistochemical analysis of rat testis with the K-20 (Fig. 1c
) or I-20 (not shown) anti-GRK4 antibodies
(8)
showed a significant staining of sperm cells located
in the inner part of the seminiferous tubuli.
|
In the rat brain, in situ hybridization showed a substantial
expression of GRK4 mRNA in the cerebellar cortex (Fig. 1d
)
and moderate expression in other brain regions, including hippocampus
(not shown). Within the cerebellar cortex, GRK4 was selectively
expressed in the Purkinje cell layer (Fig. 1d
, arrow).
Labeling in the molecular and granular layer was negligible. The sense
cRNA probe RP2 gave no hybridization signal (Fig. 1e
). A
similar pattern of expression was observed in the bovine cerebellum
(not shown). Immunostaining of the rat cerebellum with the K-20
antibody showed very strong labeling throughout the molecular and
Purkinje cell layers whereas the granular cell layer was not decorated
(Fig. 1f
, 1g
). At higher magnification, we
observed a punctate staining of GRK4 immunoreactivity in the
somatodendritic region of Purkinje cells (Fig. 1f
).
Immunoreactivity was abolished by preabsorption of the K-20 antibody
with a 10-fold excess of its antigenic peptide (not shown). As opposed
to GRK4, GRK5 and GRK6, which are structurally closely related to GRK4,
are either not expressed (GRK5, Fig. 1h
) or are expressed at
relatively low levels (GRK6, Fig. 1i
) in the cerebellar
cortex. To rule out the possibility that the absence of significant
staining by anti-GRK5 and -6 antibodies could reflect a poor quality of
these antibodies in immunostaining assays, we analyzed the expression
of GRK5 and -6 in other brain regions and tissues. GRK5, which is
poorly expressed in the brain (22
, 23)
, was found in few
cell types, including small dendritic cells (perhaps oligodendrocytes)
located in the subcortical region of the cerebellum (Fig. 1j
) and in the midbrain (Fig. 1k
). In
addition, immunohistochemical analysis of rat heart using the same
antibody revealed a substantial expression of GRK5 (not shown). GRK6
was found in various brain regions, including cerebral cortex, where
this kinase was expressed at high levels in the second and third
neuronal layers (Fig. 1l
). GRK4 was not present in
these cells (Fig. 1m
).
The expression of GRK4 mRNA in the cerebellum was confirmed by Northern
blot analysis (Fig. 2
), although the level of expression was much lower than in the testis.
Taken collectively, these results show for the first time that GRK4 is
expressed in the CNS, particularly in cerebellar Purkinje cells.
|
mGlu1a receptors are expressed in sperm cells
Based on the expression of GRK4 in the somatodendritic region of
Purkinje cells, we hypothesized that the mGlu1
receptor could be a possible substrate for this kinase. We therefore
investigated whether this receptor is expressed in sperm cells, which
represent the classical site of expression of GRK4. Immunohistochemical
analysis of rat testis showed that the mGlu1a
receptor is expressed in the seminiferous tubuli (Fig. 1n
).
This pattern of expression was similar to that of GRK4 (Fig. 1o
). Western blotting and RT-PCR confirmed the expression of
mGlu1a receptor in rat testis (M. Storto, M.
Sallese, L. Salvatore, D. F. Condorelli, P. Dell’Albani, A. De.
Blasi, and F. Nicoletti, unpublished results).
Regulation of mGlu1 receptor by GRK in heterologous
expression systems
To test whether the mGlu1 receptor is
functionally regulated by GRK4, we measured IP production in HEK293
cells transiently transfected to express mGlu1
receptor alone or in combination with GRK4. This approach has been used
extensively to investigate the regulation of a variety of GPCRs by GRK
subtypes (24)
. For these experiments, we used the GRK4
isoform (8)
and two variants of the
mGlu1 receptor: mGlu1a and
mGlu1b. These two variants are generated from the
same gene by alternative splicing at the 3' terminus, resulting in
different carboxyl-terminal tails, with the 318 carboxyl-terminal amino
acids of the mGlu1a replaced by 20 different
amino acids in the mGlu1b subtype (1
, 2)
. Expression of the mGlu1a receptor in
HEK293 cells resulted in an increase of IP levels (222±18% of
untransfected cells, n=9; P<0.001) under basal
condition (i.e., in the absence of added agonist). This increase was
abolished by cotransfecting the cells with the glutamate transporter
EAAC1 (IP levels = 101±10% vs. untransfected cells,
n=7), which removes extracellular glutamate
(16)
. This suggests that under our experimental conditions
the increase in basal IP levels was due to the action of endogenous
glutamate rather than to the intrinsic activity of
mGlu1a receptors (25)
. Thus,
mGlu1a receptor was cotransfected with the EAAC1
in all the subsequent experiments in HEK293 cells. Transfection with
the mGlu1b receptor did not increase IP levels in
the absence of agonist, perhaps because this receptor has a lower
affinity for glutamate as compared to the mGlu1a
receptor (26)
.
Exposure of HEK293 cells expressing mGlu1
receptors to the agonist quisqualate increased IP production in a
dose-dependent manner (Fig. 3
). Coexpression of GRK4 significantly reduced quisqualate-stimulated IP
production, indicating that GRK4 is able to desensitize the
mGlu1 receptor-mediated response (Fig. 3)
. GRK4
expression did not affect the levels of mGlu1a
receptor, as assessed by immunoblot (not shown).
|
To assess the role of GRK4 kinase activity for
mGlu1 receptor desensitization, we generated a
kinase-dead mutant in which two lysines within the catalytic domain
were mutated into methionines (K216M and K217M). Mutation of the
corresponding amino acids of GRK6 was shown to disrupt its kinase
activity (14)
. We documented that these mutations
abolished the ability of the mutant GRK4-(K216M, K217M) to
phosphorylate purified ROS in vitro (Fig. 4a
). In HEK293 cells cotransfection of GRK4-(K216M, K217M) did
not affect mGlu1a receptor signaling, indicating
that GRK4 kinase activity is necessary for receptor desensitization
(Fig. 4a
). In parallel samples, wild-type GRK4, expressed to
similar levels (as assessed by immunoblot, Fig. 4b
)
significantly desensitized mGlu1a
receptor-mediated responses (Fig. 4a
). The levels of
receptor expression was similar in all these conditions (Fig. 4b
). Agonist-dependent phosphorylation of
mGlu1a receptor by GRK4 was also documented. We
immunoprecipitated mGlu1a receptor from
transfected HEK293 cells and receptor phosphorylation was revealed by
anti-phospho-serine antibodies (Fig. 4c
). Whereas the
phosphorylation of the receptor cotransfected with the empty vector was
modest (even in the presence of agonist), cotransfection with GRK4
resulted in a marked agonist-dependent phosphorylation of the
mGlu1a receptor.
|
The effect of other members of the GRK family on
mGlu1a and mGlu1b
receptor-mediated responses was also investigated (Table 1
). We found that GRK2, -4, -5, and -6 all significantly reduced
agonist-stimulated receptor signaling.
|
Immunofluorescence analysis of mGlu1a receptor and GRK
in heterologous expression systems
For a variety of GPCRs it has been demonstrated that after
GRK-dependent phosphorylation and uncoupling, the receptor is
internalized from the cell surface to intracellular compartments.
Arrestin bound to the GRK-phosphorylated receptor acts as a docking
protein between the receptor and membrane proteins (such as clathrin)
involved in the process of internalization (5)
. Hence, we
examined the agonist-induced internalization of
mGlu1a receptor and GRK4 in cotransfected HEK293
and cultured cerebellar Purkinje cells using a double
immunofluorescence confocal microscopy analysis.
In unstimulated HEK293 cells, the mGlu1a receptor
was mostly localized in the plasma membranes (Fig. 5a
), according to previous reports (27)
. After a
5 min exposure to quisqualate, the receptor was internalized in
intracellular vesicles (Fig. 5d
). GRK4, which was also
localized in the plasma membrane under basal conditions (Fig. 5b
), became internalized in intracellular vesicles after
quisqualate treatment (Fig. 5e
). The receptor and the kinase
appeared to be colocalized both in the plasma membranes (unstimulated
cells) and in intracellular vesicles (agonist-stimulated cells) (Fig. 5c
and f
, arrows). We determined the effect of
various GRKs cotransfected on mGlu1a receptor
internalization (Table 2
). In HEK293 cells transfected with mGlu1a
receptor alone (i.e., without GRK cotransfected), exposure to
quisqualate failed to internalize the receptor. Cotransfection of GRK4
increased the fraction of internalized receptors under basal conditions
(i.e., without agonist) by about twofold as compared to
mock-cotransfected cells. mGlu1a receptor
internalization was drastically increased when cells expressing GRK4
were exposed to agonist: the amount of intracellular immunofluorescence
was 156% of that present on the plasma membrane, about fivefold higher
than the control values. Cotransfection of mGlu1a
receptor with GRK2, GRK5, or GRK6 had only a modest effect on receptor
internalization: after agonist treatment, the amount of intracellular
immunofluorescence was 55 to 62% of that found on the membranes, less
than twofold the control values (Table 2)
.
|
|
Immunofluorescence analysis of mGlu1a receptor and GRK
in cerebellar Purkinje cells
We performed immunofluorescence analysis of
mGlu1a receptor internalization on Purkinje cells
in primary cultures. Also in this model, the
mGlu1a receptor was internalized from a membrane
location to intracellular vesicles after a 5 min exposure to
quisqualate (Fig. 5g
, j
).
mGlu1a receptor stimulation also resulted in a
redistribution of GRK4 (Fig. 5h
, k
), with this
kinase being colocalized with the receptor both under basal and
stimulated conditions (Fig. 5i
, 5l
, arrows). In
cultured Purkinje cells, we also examined the localization of the
mGlu1a receptor in relation to GRK2/GRK3 by using
a monoclonal antibody that recognizes an epitope common to both kinases
(Fig. 5m
, n
, o
, p
, q
, r
). It is known that under basal
conditions, GRK4 is almost completely associated to membranes, due to
the palmitoylation of its carboxyl terminus (7)
, whereas a
substantial fraction of GRK2 and GRK3 is cytosolic (4)
.
Accordingly, in unstimulated cells, the immunofluorescence of GRK2/GRK3
was localized predominantly in the cytosol of Purkinje cells (Fig. 5n
). As opposed to GRK4, GRK2/GRK3 were not colocalized with
mGlu1a receptors (Fig. 5o
, 5r
). In addition, GRK2/GRK3 were only minimally (if at all)
redistributed in response to quisqualate (Fig. 5q
). In all
these experimental conditions no labeling was seen in the absence of
the primary antibody (not shown). We performed a quantitative analysis
using an NIH image analysis program, which gives an estimate of
colocalized staining. In HEK293 cells, 84 ± 3.8%
(n=10) of the GRK4 and 66 ± 4.1% (n=10) of
the mGlu1a receptor staining were reciprocally
colocalized. In cultured Purkinje cells, 69 ± 5.6%
(n=10) of the GRK4 and 52 ± 3.4% (n=10) of
the receptor staining were colocalized. GRK2/GRK3 analysis in Purkinje
cells showed that only 28 ± 7.4% (n=10) of the kinase
and 25 ± 7.8 (n=10) of the receptor staining were
colocalized.
No specific staining of cultured Purkinje cells was observed with anti-GRK5 and anti-GRK6 antibodies (not shown).
GRK4 knock down affects mGlu1 receptor internalization
in cerebellar Purkinje cells
To assess whether GRK4 is functionally involved in the mechanism
of mGlu1a receptor internalization in Purkinje
cells, we measured the agonist-induced receptor internalization in
cultured Purkinje cells in which GRK4 was knocked down by antisense
oligonucleotide treatment. Four days of treatment of the cultures with
the antisense oligonucleotide reduced GRK4 immunolabeling by 67.7 ± 6.2% (n=12 cells), whereas the expression of GRK2/GRK3
was not affected (99±8% of the control value, n=10). The
agonist-induced internalization of mGlu1a
receptor was substantially blunted in the GRK4 antisense-treated
Purkinje cells (Table 3
), indicating that GRK4 plays a significant role in
mGlu1a receptor trafficking. To assess the
selectivity of this effect, we knocked down GRK2/GRK3 in primary
cultured Purkinje cells by using a specific antisense oligonucleotide.
Four days of treatment with the anti-GRK2 antisense oligonucleotide
reduced GRK2/GRK3 immunolabeling by 62 ± 5% (n=7)
whereas GRK4 was only slightly affected (89±12% of control value,
n=10). Knock down of GRK2 did not alter the agonist-induced
mGlu1a receptor internalization (Table 3)
. The
possible involvement of PKC in agonist-induced
mGlu1a receptor internalization was also
investigated. We found that the addition of two PKC inhibitors
(staurosporin and Ro-318220) did not affect agonist-induced
mGlu1a receptor internalization in Purkinje cells
(Table 3)
, indicating that PKC is not involved in this regulatory
mechanism.
|
GRK4 knock down affects mGlu1 receptor signaling in
cerebellar Purkinje cells
To assess whether GRK4 interacts with the activity of
mGlu1 receptors in Purkinje cells, antisense
oligonucleotides specifically targeted to the GRK4-encoding mRNA were
used to knock down GRK4 in cultured Purkinje cells. Untreated cultures
or cultures treated with a scrambled oligonucleotide were used as
controls. The efficacy of the antisense treatment was confirmed by
quantitative double immunolabeling with the anti-GRK4 K-20 antibody and
an anti-calbindin antibody. Calbindin staining allowed positive
identification of Purkinje cells and at the same time served to
normalize the level of fluorescent immunolabeling between cultures.
Antisense oligonucleotide treatment reduced GRK4 immunolabeling while
the scrambled oligonucleotide had no effect (Fig. 6a
, b
, c
, d
, e
, f
).
|
Activation of mGlu1 receptor induces an elevation
of intracellular calcium concentration
([Ca2+]i)
(28)
. In control cultures, superfusion with the
mGlu1 receptor agonist
(S)-3,5-dihydroxyphenylglycine (DHPG) resulted in a
[Ca2+]i rise with an
initial peak, followed by a sustained component that rapidly reversed
after washing out the agonist. The initial peak is mainly due to
liberation of calcium from internal stores whereas the sustained
component involves influx of calcium, because it is abolished when
external calcium is removed (Fig. 6
, and not shown). Treatment of
cultures with the antisense oligonucleotide resulted in a reduction of
the initial peak, whereas the calcium levels at the end of a 5 min
application of DHPG were higher (108.6±3.0 nM) than untreated cells
(65.1±3.0 nM) or cells treated with the scrambled oligonucleotide
(69.0±4.5 nM). The higher amplitude of the calcium response at the end
of agonist application indicates a reduced desensitization of
mGlu1 receptor in Purkinje cells with a reduced
GRK4 expression. In cultures treated with the GRK4 antisense
oligonucleotide, the sustained reduction in mGlu1
receptor desensitization may lead to a partial depletion of
intracellular Ca2+ stores, thus explaining the
lower initial Ca2+ peak in response to DHPG.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The activated subunit of the
Gq (Gq) can in turn be inhibited by RGS (regulators of G-protein
signaling) proteins (30)
. These RGS proteins work by
interacting with G and by increasing the intrinsic GTPase activity
of G, acting as GTPase-activating proteins (31
, 32)
.
The present study provides the first evidence that, in addition to
these mechanisms, mGlu1 receptors are regulated
by GRK4 in cerebellar Purkinje cells. This is the first demonstration
that GRKs can regulate receptors of the class III GPCRs.
The involvement of GRKs in the mechanism of homologous desensitization
is now considered a general phenomenon, which has been documented for
several GPCR. Using a variety of different approaches we now
demonstrate a functional role of GRK4 in the regulation of the
mGlu1a receptor signaling and internalization.
GRK4 desensitized the mGlu1a and
mGlu1b receptors when coexpressed in heterologous
expression systems. The kinase activity of GRK4 is necessary for
desensitizing mGlu1a receptor, since a
kinase-dead GRK4-(K216M, K217M) mutant failed to regulate receptor
signaling. Consistently, we documented that GRK4 phosphorylated
mGlu1a receptor in an agonist-dependent manner.
In transfected HEK293, all the GRKs tested were able to desensitize
mGlu1a receptor signaling. However, as discussed
in previous reports (14
, 33)
, this assay is sensitive but
not selective, since the overexpression of regulatory proteins may
force an interaction that does not normally occur at the physiological
expression level.
The analysis of mGlu1 receptor internalization
and colocalization with GRK4 strongly indicated a preferential
interaction between these proteins. In transfected HEK293 cells,
cotransfection with GRK4 significantly enhanced agonist-dependent
mGlu1a receptor internalization, whereas GRK2,
-5, and -6 affected much less receptor trafficking. These results are
similar to those obtained with follitropin receptors in transfected
HEK293 cells where different GRKs, which phosphorylate the receptor to
the same extent, have differential effects on receptor internalization
(14)
. According to Lazari et al. (14)
, we can
assume that phosphorylation of different domains, or even different
residues by various GRKs has a different effect on agonist-induced
receptor internalization.
In Purkinje cells primary culture we observed a marked colocalization
between mGlu1a and GRK4, and the kinase was also
redistributed into intracellular vesicles on stimulation with
mGlu1 receptor agonist. By contrast, GRK2/GRK3
immunoreactivity in Purkinje cells was not affected by stimulation of
mGlu1 receptor and was not colocalized with the
mGlu1a receptor under basal condition or after
agonist treatment. The experiments with GRK4 antisense oligonucleotides
in cerebellar Purkinje cells further indicate that GRK4 is involved in
the mechanism of mGlu1a receptor homologous
desensitization and internalization. Treatment with the GRK4 antisense
oligonucleotide, which knocked down the expression of GRK4 but did not
affect the expression of GRK2/GRK3 immunoreactivity, impaired the
desensitization of mGlu1-mediated response and
receptor internalization indicating that GRK4 is required for the
regulation of this receptor. Also, knocking down GRK2/GRK3 levels by
specific antisense oligonucleotides did not alter
mGlu1a receptor internalization. In cultured
Purkinje cells, inhibition of PKC did not affect agonist-induced
receptor internalization. This is consistent with the findings that
GPCR phosphorylation by GRK, but not by PKA and PKC, increases the
binding of (ß)-arrestins to the receptor (34)
, and this
step is important for receptor internalization. Taken together, these
data strongly indicate that GRK4 may be critical for
mGlu1a signaling and trafficking under
physiological conditions in cerebellar Purkinje cells.
In this study we have shown that GRK4, besides being expressed at high levels in sperm cells, is also present in cerebellar cortex. In addition, we have shown that mGlu1a receptor and GRK4 are expressed in the same cell types. In particular, they are both present in cerebellar Purkinje cells and in sperm cells, which are known to be major sites of expression of mGlu1a receptor and GRK4, respectively. This further suggests a regulatory role of this kinase on mGlu1a receptor. Since mGlu1b receptor is widely expressed in many brain regions that were not analyzed in detail in this investigation, further studies are necessary to document to what extent mGlu1b receptor and GRK4 are colocalized and the role of GRK4 in the regulation of mGlu1b in different neuronal cells.
mGlu1 receptors present in Purkinje cells have
been implicated in motor learning and motor coordination. Knockout mice
lacking mGlu1 receptor are ataxic and show a
selective impairment of long-term depression (LTD) in Purkinje cells, a
form of synaptic plasticity underlying cerebellar motor learning
(35
, 36)
. An impairment in LTD and in motor coordination
is also observed in knockout mice lacking the intracellular receptors
for inositol-1,4,5-trisphosphate (36)
or the Gq protein
(38)
, i.e., two elements that are downstream of the
mGlu1a receptor in the transduction cascade. The
critical role for GRK4 in regulating mGlu1a
receptor signaling implicates a major involvement of this kinase in the
physiology of Purkinje cells and in motor learning. In patients with
Hodgkin’s disease, paraneoplastic cerebellar ataxia has been
associated with autoantibodies directed against
mGlu1 receptors, implicating a role for
mGlu1 receptors in ataxic syndromes
(39)
. Hence, we propose GRK4 as a novel target for studies
aimed at identifying molecular determinants of cerebellar
disorders.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 14, 2000.
Revision received May 18, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1B-adrenergic and A1 adenosine receptor-mediated responses. Mol. Pharmacol. 56,316-324q. Proc. Natl. Acad. Sci. USA 94,14089-14094This article has been cited by other articles:
|
|
 |
|
  K. A. Neve Novel Features of G Protein-Coupled Receptor Kinase 4 Mol. Pharmacol., March 1, 2006; 69(3): 673 - 676. [Abstract] [Full Text] [PDF]
|
