Metabotropic glutamate and dopamine receptors co-regulate AMPA receptor activity through PKA in cultured chick retinal neurones: effect on GluR4 phosphorylation and surface expression


of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Metabotropic glutamate and dopamine receptors co-regulate AMPA receptor activity through PKA in cultured chick retinal neurones: effect on GluR4 phosphorylation and surface expression
  Metabotropic glutamate and dopamine receptors co-regulateAMPA receptor activity through PKA in cultured chick retinalneurones: effect on GluR4 phosphorylation and surface expression Andre´ R. Gomes,* Paulo Cunha,* ,1 Mutsuo Nuriya,   Carlos J. Faro,* ,   Richard L. Huganir,  Euclides V. Pires,* ,   Ana Luı´sa Carvalho* , § and Carlos B. Duarte* , § * Center for Neuroscience and Cell Biology, University of Coimbra, Portugal    Department of Neuroscience, The Johns Hopkins University School of Medicine, Baltimore, Maryland, USA   Department of Biochemistry, University of Coimbra, Portugal  §  Department of Zoology, University of Coimbra, Portugal  Abstract Glutamate receptor phosphorylation has been implicated inseveral forms of modulation of synaptic transmission. It hasbeen reported that protein kinase A (PKA) can phosphory-late the  a -amino-3-hydroxy-5-methyl-4-isoxazolepropionate(AMPA) receptor subunit GluR4 on Ser842, both  in vitro  and  in vivo  . Here, we studied the regulation of GluR4phosphorylation and intracellular trafficking by PKA and bymetabotropic receptors coupled to adenylyl cyclase (AC), incultured chick retinal amacrine-like neurones, which areenriched in GluR4. The regulation of AMPA receptor activityby PKA and by metabotropic AC-coupled receptors wasalso investigated by measuring the [Ca 2+ ] i  response tokainate in Na + -free medium. Stimulation of AC with forskolin(FSK), or using the selective agonist of dopamine D 1 receptors (±)-1-phenyl-2,3,4,5-tetrahydro-(1H)-3-benzaze-pine-7,8-diol (SKF38393), increased the [Ca 2+ ] i  response tokainate, GluR4 phosphorylation at Ser842 and GluR4 sur-face expression. Pre-incubation of the cells with(2S,2 ¢ R,3 ¢ R)-2-(2 ¢ ,3 ¢ -dicarboxycyclopropyl)glycine (DCG-IV),an agonist of group II metabotropic glutamate receptors(mGluR), which are coupled to inhibition of AC, inhibited theeffect of FSK and of SKF38393 on AMPA receptor activity,GluR4 phosphorylation and expression at the plasmamembrane. These results indicate that there is a functionalcross-talk between dopamine D 1  receptors and group IImGluR in the regulation of GluR4 phosphorylation andAMPA receptor activity. Our data show that GluR4 phos-phorylation at Ser842 by PKA, and its recruitment to theplasma membrane upon phosphorylation, is regulated bymetabotropic receptors. Keywords:  AMPA receptors, chick retina, GluR4, metabo-tropic receptors, protein kinase A, surface expression. J. Neurochem.  (2004)  90,  673–682. Glutamate is the major excitatory neurotransmitter in thecentral nervous system (CNS) and its receptors have beenshown to play an important role in synaptic plasticity (Bear and Abraham 1996; Malenka and Nicoll 1999), cell death(Choi 1988) and neural development (Constantine-Paton Received February 9, 2004; revised manuscript received March 18,2004; accepted March 19, 2004.Address correspondence and reprint requests to Carlos B. Duarte,Center for Neuroscience and Cell Biology, Department of Zoology,University of Coimbra, 3004–517 Coimbra, Portugal.E-mail:  1 The present address of Paulo Cunha is European Molecular BiologyLaboratory, Meyerhofstrasse 1, 69117 Heidelberg, Germany.  Abbreviations used  : AC, adenylyl cyclase; AMPA,  a -amino-3-hydroxy-5-methyl-4-isoxazolepropionate; BCA, bicinchoninic acid; BME, basalmedium of Eagle; BSA, bovine serum albumin; CaMKII, Ca 2+ / calmodulin-dependent protein kinase II; CHEL, chelerythrine chloride;db-cAMP, N 6 ,2 ¢ -O-dibutyryladenosine-3 ¢ ,5 ¢ -cyclic monophosphate,sodium salt monohydrate; DCG-IV, (2S,2 ¢ R,3 ¢ R)-2-(2 ¢ ,3 ¢ -dicarboxy-cyclopropyl)glycine; DOC, deoxycholic acid; FSK, forskolin; H-89, N-[2-(p-bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide. 2HCl;Indo-1/AM, indo-1/acetoxymethyl ester; LTP, long-term potentiation;mGluR, metabotropic glutamate receptors; NMG,  N  -methyl- D -glucam-ine; PKA, protein kinase A; PKC, protein kinase C; PVDF, polyvinylidene difluoride; RO-20-1724, 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone; Rp-cAMPs, Rp-adenosine-3 ¢ ,5 ¢ -cyclic monophosp-horothioate; SKF38393, (±)-1-phenyl-2,3,4,5-tetrahydro-(1H)-3-ben-zazepine-7,8-diol; SKF83566, (±)-7-bromo-8-hydroxy-3-methyl-1- phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hydrochloride; SDS,sodium dodecyl sulfate.  Journal of Neurochemistry , 2004,  90 , 673–682 doi:10.1111/j.1471-4159.2004.02519.x   2004 International Society for Neurochemistry,  J. Neurochem.  (2004)  90 , 673–682  673  et al  . 1990). Glutamate receptors belong to the ionotropicand metabotropic families of receptors (reviewed in Ozawa et al  . 1998). Ionotropic glutamate receptors are ligand-gatedion channels that mediate most of the rapid excitatorysynaptic transmission in the CNS. These receptors areclassified according to their pharmacological and electro- physiological properties as  N  -methyl- D -aspartate (NMDA), a -amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA)or kainate receptors (reviewed in Wisden and Seeburg1993; Hollmann and Heinemann 1994). Metabotropic glu-tamate receptors do not form ion pores or mediate fast synaptic transmission. These receptors act via secondmessengers and have a modulatory role over several cellular  processes, including ionotropic receptor response (Pin andBockaert 1995).Phosphorylation is perhaps the most common form of  post-translational modification of ion channels and ionotrop-ic receptors. In general, phosphorylation can modulate themain characteristics of the channel, including its open probability and mean open time, as well as intracellular receptor trafficking (reviewed in Carvalho  et al  . 2000; Barryand Ziff 2002; Malinow and Malenka 2002; Song andHuganir 2002; Gomes  et al  . 2003). Phosphorylation of non- NMDA receptors by protein kinase A (PKA), followingstimulation of adenylyl cyclase (AC) with forskolin (FSK),increases the amplitude and decay time of spontaneousexcitatory post-synaptic currents in cultured hippocampal pyramidal neurones (Greengard  et al  . 1991). This effect ismediated by an increase in the opening frequency and meanopen time of non-NMDA receptors. Further studies have pinpointed the consequences of GluR1 phosphorylation.Interestingly, phosphorylation of Ser831 by Ca 2+ /calmodu-lin-dependent protein kinase II (CaMKII) potentiates singlechannel conductance (Derkach  et al  . 1999), while phos- phorylation of Ser845 by PKA increases open channel probability (Banke  et al  . 2000). Furthermore, these different  phosphorylation sites acted upon by different kinases areinvolved in long-term potentiation (LTP) in a way that depends on the activation history of the synapse (Lee  et al  .2000). GluR1 phosphorylation is reversible, and several protein phosphatases are involved (Tavalin  et al  . 2002) in acomplex and highly regulated equilibrium involving PKAand CaMKII (Lee  et al  . 2000). Recently, it was shown that  phosphorylation of GluR1 is necessary for plasticity andlearning and memory, since mice with knockin mutations inthe GluR1 phosphorylation sites show deficits in LTP andlong-term depression (LTD) and have memory defects inspatial learning tasks (Lee  et al  . 2003).GluR4-containing AMPA receptors are expressed inspecific regions of the CNS, where they are responsible for signal transmission at high rates. In the rat hippocampus,GluR4 is expressed early in development and it is the first AMPA receptor subunit to target synapses (Zhu  et al  . 2000).The GluR4 subunit is expressed in most subpopulations of cat and rat retinal amacrine cells (Qin and Pourcho 1999;Grunder   et al  . 2000), and is the major AMPA receptor subunit present in cultures enriched in chick amacrine-likeneurones (Carvalho  et al  . 2002). GluR4 is also the mainAMPA receptor subunit expressed in auditory cells of thecochlear nucleus, where the fast kinetics of GluR4 is physiologically important in transmitting the signals neces-sary for sound localization (Raman  et al  . 1994). The spinalcord motor neurones, where rapid neurotransmission has been recorded (Smith  et al  . 1991), are also enriched inGluR4 (Tomiyama  et al  . 1996). It has been shown that PKCand PKA can phosphorylate GluR4 on Ser842, both  in vitro and  in vivo . However,  in vivo  PKA activation leads to asignificantly higher phosphorylation of Ser842 than PKCactivation (Carvalho  et al  . 1999). Furthermore, in hippo-campal slices, PKA phosphorylation of GluR4 on Ser842was shown to be necessary and sufficient for homomericGluR4 synaptic delivery (Esteban  et al  . 2003).Inthisstudy,wehaveevaluatedtheeffectofPKAactivationon the activity of native GluR4-containing AMPA receptors,on the phosphorylation of GluR4 and on GluR4 surfaceexpression in cultured chick retina cells. We also investigatedwhether metabotropic receptors coupled to AC can modulatethe activity of AMPA receptors, and the phosphorylation andsurface expression of GluR4. Our results show that PKAactivation or dopamine D 1  receptor activation leads toincreased AMPA receptor activity and increased GluR4Ser842 phosphorylation and GluR4 surface expression. Experimental procedures Materials Indo-1/acetoxymethyl ester (Indo-1/AM) was obtained fromMolecular Probes Inc. (Leiden, the Netherlands). Chelerythrinechloride (CHEL), ionomycin and N-[2-(p-bromocinnamylamino)-ethyl]-5-isoquinolinesulfonamide.2HCl (H-89) were purchased fromCalbiochem (San Diego, CA, USA). Forskolin (FSK), 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone (RO-20-1724), N 6 ,2 ¢ -O-dibu-tyryladenosine-3 ¢ ,5 ¢ -cyclic monophosphate, sodium salt monohy-drate (db-cAMP) and Rp-adenosine-3 ¢ ,5 ¢ -cyclic monophosphoro-thioate (Rp-cAMPs) were obtained from BIOMOL ResearchLaboratories, Inc. (Plymouth Meeting, PA, USA). (±)-1-Phenyl-2,3,4,5-tetrahydro-(1H)-3-benzazepine-7,8-diol (SKF38393) and2S,2 ¢ R,3 ¢ R-2-(2 ¢ ,3 ¢ -dicarboxycyclopropyl)glycine (DCG-IV) wereobtained from Tocris Cookson Inc. (Bristol, UK). Trypsin was fromGibco Life Technologies (Paisley, UK) and fetal calf serum wasfrom Biochrom KG (Berlin, Germany) or from Biowhittaker (Walkersville, MD, USA). Complete Mini protease inhibitor cocktail and microporous polyvinylidene difluoride (PVDF) mem- branes were purchased from Roche Diagnostics GmbH (Basel,Switzerland). The rabbit polyclonal anti-GluR4 antibody wasobtained from Upstate Biotechnology (Buckingham, UK) and therabbit polyclonal anti-GluR4 phosphorylated at Ser842 antibodywas produced against the chemically phosphorylated peptide(RNKARL S P ITGSV) corresponding to rat GluR4 C-terminal 12 674  A. R. Gomes  et al.   2004 International Society for Neurochemistry,  J. Neurochem.  (2004)  90 , 673–682  amino acids (836–847) phosphorylated at Ser842 (Esteban  et al  .2003). The alkaline phosphatase-conjugated anti-rabbit and anti-mouse secondary antibodies, and the ECF immunodetectionsubstrate, were obtained from Amersham Biosciences (Uppsala,Sweden). EZ-link Sulfo-NHS-SS-biotin, the UltraLink Plus Immo- bilized Streptavidin Gel and the bicinchoninic acid (BCA) proteinassay reagent kit were purchased from Pierce (Rockford, IL, USA).All the other reagents were obtained from Sigma (St Louis, MO,USA) or from Merck (Darmstadt, Germany). Stock solutions of Indo-1/AM, CHEL, RO-20-1724, ionomycin and FSK were made indimethyl sulfoxide. All the other chemicals were kept in aqueousstocks. Embryonic chick retina cell culture Monolayer cultures of chick retina amacrine-like cells were preparedas previously described (Duarte  et al  . 1992, 1996). Briefly, the retinas from 8-day-old chick embryos (White Leghorn) weredissected and digested with 0.1% trypsin, in Ca 2+ - and Mg 2+ -freeHank’s balanced salt solution, for 15 min at 37  C. The cells werecultured on poly D-lysine (0.1 mg/mL)-coated glass coverslips or tissue culture dishes in basal medium of Eagle (BME), buffered with25 m M  HEPES and 10 m M  NaHCO 3 , pH 7.4, and supplementedwith 5% heat-inactivated fetal calf serum, penicillin (100 U/mL) andstreptomycin (100  l g/mL). The cells were maintained at 37  C in ahumidified incubator with 95% air and 5% CO 2 , and used after 5 days in culture. Indo-1 loading and Ca 2+ measurements [Ca 2+ ] i  was measured by Indo-1 fluorescence, as described previously (Duarte  et al  . 1996; Carvalho  et al  . 1998), using acomputer-assisted Perkin-Elmer LS-5B luminescence spectrometer with fluorescence excitation at 335 nm and emission at 410 nm, andusing 5 nm slits. Monolayers of chick retina cells cultured oncoverslips, at a density of 0.6  ·  10 6 cells/cm 2 , were incubated with3  l M  Indo-1/AM in BME buffered with 25 m M  HEPES, for 45 minat 37  C, and further incubated in BME, for 15 min, in the absenceof the dye, to ensure complete hydrolysis of Indo-1/AM. The glasscoverslips were then mounted in a stirred, thermostatted (37  C)cuvette, and the fluorescence emission intensities were recordedevery 0.5 s. The [Ca 2+ ] i  was calculated from the emissionfluorescence values using the equation for single fluorescencemeasurements (Grynkiewicz  et al  . 1985; Bandeira-Duarte  et al  .1990). The calibration was performed at the end of each experiment,using 3  l M  ionomycin to obtain maximal fluorescence, followed byaddition of 2 m M  MnCl 2  to obtain autofluorescence. The dissoci-ation constant taken for the Indo-1/Ca 2+ complex was 250 n M (Grynkiewicz  et al  . 1985). Stimuli Cells were washed and stimulated as indicated in  N  -methyl- D -glucamine (NMG) medium (132 m M  NMG; 4 m M  KCl; 1 m M CaCl 2 ; 1.4 m M  MgCl 2 ; 6 m M  glucose; buffered with 10 m M HEPES; pH 7.4). A Na + -free medium was used in order to prevent cell depolarization following AMPA receptor activation, ruling out calcium entry through voltage-dependent calcium channels andtherefore isolating the component due to Ca 2+ influx throughactivated AMPA receptors. To allow correlation with the [Ca 2+ ] i data, NMG medium was also used in the experiments performed todetermine GluR4 phosphorylation on Ser842 and GluR4 surfaceexpression.In all experiments, including the controls, cells were incubatedwith a selective inhibitor of cAMP phosphodiesterase, RO-20-1724(1  l M ), throughout the experiment (5 min), to prevent cAMPdegradation. GluR4 phosphorylation After incubation with the indicated stimuli, cells were washed twicewith NMG medium, lysed with ice-cold RIPA buffer [150 m M  NaCl, 50 m M  Tris-HCl, 5 m M  EGTA, 1% Triton X-100, 0.5%deoxycholic acid (DOC), 0.1% sodium dodecyl sulfate (SDS); pH 7.5] supplemented with protease inhibitors (Complete Mini protease inhibitor cocktail) and phosphatase inhibitors (10 m M  Na 4 PO 7 ; 50 m M  NaF; 1 m M  Na 3 VO 4 ), and scraped off the culturedishes. The lysates were centrifuged (14 000  g  , 10 min, 4  C) andthe pellet was discarded. Protein concentration was measured usingthe BCA protein assay reagent kit (Pierce) and the proteinconcentration was equalized among samples. The samples werethen diluted 4 : 1 with sodium dodecyl sulfate-polyacrylamide gelelectrophoresis (SDS-PAGE) sample buffer [final concentration:0.125  M  Tris, 2% (w/v) SDS, 5% (v/v) glycerol, 5% (v/v)  b -mercaptoethanol] and boiled for 10 min. Biotinylation of plasma membrane-associated proteins Cells were washed twice with phosphate-buffered saline (PBS) withcalcium and magnesium (PBS/Ca 2+ /Mg 2+ : 137 m M  NaCl; 2.7 m M KCl; 1.8 m M KH 2 PO 4 ; 10 m M  Na 2 HPO 4 ; plus 0.5 m M MgCl 2 ; 1 m M CaCl 2 ;pH 7.4),andthenincubatedwith1 mg/mLNHS-SS-Biotinfor 30 min at 4  C under mild shaking. They were then rinsed three timeswith PBS/Ca 2+ /Mg 2+ supplemented with glycine (100 m M ) and afourth time with PBS supplemented with protease inhibitors. Cellswere then lysed with RIPA buffer supplemented with protease and phosphatase inhibitors, scraped off the plates and centrifuged at 14 000  g  for10 minat4  C.Thesupernatantfluidsweretransferredtoclean tubes and protein concentration was measured using the BCA protein assay reagent kit. Ultra-Link Streptavidin Plus was added toequal amounts of supernatant fluid (2  l g/10  l g total protein) andincubated for 2 h at 4  C with mild shaking (orbital shaker).Complexeswerethencentrifuged(2500  g  ,3 min)andthesupernatant fluid, corresponding to intracellular proteins, was collected. Com- plexeswerewashedfurtherwithRIPAbufferandcentrifuged(2500  g  ,3 min) four times. Proteins were eluted from streptavidin beads by boiling for 10 min in SDS-PAGE sample buffer. SDS-PAGE and immunoblotting The extracts obtained were resolved by SDS-PAGE in 10% polyacrylamide (Laemmli 1970) gels using 30–35  l g protein per well. This was followed by overnight electrotransfer to PVDFmembranes, at 40 V, complemented by 30 min at 200 V. Mem- branes were then blocked for 1 h with 1% (w/v) bovine serumalbumin (BSA) in Tris-buffered saline with 0.1% Tween-20 (TBS-T),and probed for 1 h with the primary anti-GluR4 (1 : 400) or anti-GluR4 phosphorylated at Ser842 (1 : 2500) antibodies. Followingfive washes (5 min) in 1% BSA/TBS-T, the membranes wereincubated for 1 h with alkaline phosphatase-conjugated secondaryanti-rabbit antibody (1 : 20 000). The membranes were then washedagain five times (5 min), incubated with chemifluorescence substrate Regulation of AMPA receptors by PKA  675   2004 International Society for Neurochemistry,  J. Neurochem.  (2004)  90 , 673–682  (ECF) for 5 min, and scanned with the Storm 860 scanner (Amersham Biosciences). The scanned digital images were quan-tified using the ImageQuant 5 software (Amersham Biosciences). Statistical analysis Results are presented as means ± SEM of the indicated number of experiments, carried out in different preparations. Statistical signi-ficance was determined by one-way  ANOVA  followed by theDunnett’s or Bonferroni’s test. Results Modulation of Ca 2+ influx through AMPA receptors byPKA In previous studies, we have shown that kainate increasesthe [Ca 2+ ] i  in cultured chick retina cells due to activation of AMPA receptors (Carvalho  et al  . 1998). These cells containCa 2+ -permeable AMPA receptors (Carvalho  et al  . 1998,2002), and activation of AMPA receptors in a Na + -containing medium leads to Ca 2+ entry through thereceptor-associated channels and through voltage-gatedCa 2+ channels activated by depolarization (Duarte  et al  .1996). To avoid cell depolarization upon receptor activa-tion, all experiments were performed in a Na + -free NMGmedium. Under these circumstances, receptor activationleads to Ca 2+ entry only through AMPA receptor-associatedchannels (Duarte  et al  . 1996; Carvalho  et al  . 1998). There-fore, the [Ca 2+ ] i  response to kainate in primary cultures of chick amacrine-like neurones is a good experimental modelfor studying the regulation of Ca 2+ -permeable, AMPA-typeglutamate receptor activity (Duarte  et al  . 1996; Carvalho et al  . 1998).To examine whether the activity of AMPA receptors isregulated by PKA, we measured the effect of FSK and of db-cAMP on the [Ca 2+ ] i  response to kainate (Fig. 1b). To better unravel the effect of activating PKA, a phosphodiest-erase inhibitor, RO-20-1724, was present in the experimentalmedium throughout the experiment. Representative traces of the effect of AC activation with FSK on the influx of Ca 2+ through AMPA receptors are shown in Fig. 1(a). Pre-incubation of the cells with 10  l M  FSK led to an increaseof 24.0 ± 4.5% on the [Ca 2+ ] i  rise evoked by kainaterelatively to the control response (Fig. 1b). FSK activatesAC, leading to an increase of cAMP and subsequent activation of PKA. A similar effect was observed whenPKA was activated with db-cAMP (1 m M ), a cell permeablecAMP analogue (35.3 ± 8.18% above control; Fig. 1b). Onthe other hand, pre-incubation of amacrine-like neuroneswith 500  l M  Rp-cAMPs, a selective competitive inhibitor of PKA, before activation of AMPA receptors with kainate,caused a decrease of 23.1 ± 5.6% on the [Ca 2+ ] i  response tothe agonist (Fig. 1b). Thus, endogenous PKA activelymodulates AMPA receptors in cultured retinal neurones. Effect of dopamine D 1  receptors and group II mGluR onAMPA receptor activity Since PKA modulates the kainate-evoked Ca 2+ influxthrough AMPA receptors, we looked for physiologicalstimuli that could act upstream of PKA. For this purpose,we tested whether AMPA receptor activity is changed byactivation of dopamine D 1  receptors, which activate AC viaG proteins, increasing the intracellular cAMP content andthereby activating PKA, and of group II mGluR that inhibit AC. Both types of metabotropic receptors are expressed inchick retinal neurones (de Mello  et al  . 1996; Caramelo  et al  .1999).Activation of dopamine D 1  receptors with the selectiveagonist SKF38393 (10  l M ) ( Noh and Gwag 1997)increased the [Ca 2+ ] i  response to kainate by 20.2 ± 6.8%compared with the control (Fig. 2a). This increase wastotally blocked by pre-incubation of the cells with a D 1 receptor antagonist (SKF83566; 10  l M ) ( Noh and Gwag1997) or Rp-cAMPs (a PKA inhibitor) before stimulationwith SKF38393 (Fig. 2a). This shows that the potentiationof Ca 2+ influx through AMPA receptors by SKF38393 wasdue to dopamine D 1  receptor activation and was mediatedvia PKA.Figures 2b and 2c show the effect of group II mGluR activation, with the selective agonist DCG-IV (1  l M )(Brabet   et al  . 1998), on the [Ca 2+ ] i  response to kainate.DCG-IV did not significantly affect the influx of Ca 2+ through AMPA receptors (0.3 ± 1.4% above control, Fig. 1  Effect of PKA activation or inhibition on the [Ca 2+ ] i  response tokainate in Na + -free extracellular medium, in cultured chick amacrine-like neurones. (a) Representative traces of potentiation of the kainate-evoked Ca 2+ influx by FSK. The cells were incubated with or without10  l M  FSK 3 min before stimulation of AMPA receptors with 100  l M kainate. (b) The cells were pre-incubated with the PKA activator db-cAMP (1 m M ) for 1 h, with FSK (10  l M ) for 3 min, or with PKA inhibitorRp-cAMPs (500  l M ) for 1 h, before stimulation with 100  l M  kainate.Data are expressed as percentage of control of the maximal [Ca 2+ ] i response ( D [Ca 2+ ] i ) to 100  l M  kainate and are means ± SEM of theindicated number of experiments carried out at least in duplicate indifferent preparations. * p <   0.05, ** p <   0.01, Dunnett’s test. 676  A. R. Gomes  et al.   2004 International Society for Neurochemistry,  J. Neurochem.  (2004)  90 , 673–682  Fig. 2b). However, an interesting effect was seen whenDCG-IV was added to the cells before FSK (Fig. 2b) or SKF38393 (Fig. 2c). DCG-IV inhibited the up-regulatoryeffects of FSK (3.1 ± 5.6% above control) and SKF38393( )  2.5 ± 1.1%) on AMPA receptor-mediated Ca 2+ influx.These results show that there is a functional cross-talk  between mGluR and dopamine receptors leading to modu-lation of AMPA receptor activity. Phosphorylation of GluR4 at Ser842 GluR4 is the most abundant AMPA-type glutamate receptor subunit present in cultured amacrine-like neurones (Carvalho et al  . 2002), and PKA can phosphorylate GluR4 on Ser842 in vitro  and in transfected cells (Carvalho  et al  . 1999). Toexamine the effect of PKA activation on the phosphorylationstate of native GluR4, we stimulated cultures enriched inretinal amacrine-like cells as indicated previously. Theextracts obtained were then submitted to SDS-PAGE,followed by immunoblot with a phosphorylation site-specificantibody against GluR4 phosphorylated on Ser842. Thisantibody is blocked by the phosphorylated antigenic peptideand does not show cross reactivity with unphosphorylatedGluR4 (Esteban  et al  . 2003). Equal amounts of protein wereapplied to all lanes and the quantification of phosphorylatedGluR4 was normalized according to the amount of totalGluR4 in each lane.In control conditions, phosphorylated GluR4 migrated as a band of approximately 110 kDa, showing that there was basal phosphorylation of GluR4 (Fig. 3).There was a clear increase in GluR4 phosphorylation after exposure of retinal neurones to 10  l M  FSK in the presence(83.17 ± 7.9% above control; Fig. 3a) or in the absence(95.6 ± 3.4% above control; data not shown) of the phos- phodiesterase inhibitor RO-20-1724. This potentiation wasnot significantly affected by pre-incubating the cells withCHEL (5  l M ), a specific protein kinase C (PKC) inhibitor (Herbert   et al  . 1990) (60.7 ± 13.9% above control; Fig. 3a).In contrast, incubation with 1  l M  H-89 (Davies  et al  . 2000),an inhibitor of PKA, decreased GluR4 phosphorylation to64.5 ± 9.2% of control (Fig. 3a), indicating that basal PKAactivity is important for phosphorylation of the receptors inresting conditions. This decrease in GluR4 phosphorylationafter incubation with H-89 was also observed when the cellswere stimulated with FSK after pre-incubation with the PKAinhibitor (69.1 ± 2.7% of control; Fig. 3a).The dopamine D 1  receptor agonist (SKF38393) alsoincreased the phosphorylation of GluR4, to 53.5 ± 3.8%above the control (Fig. 3b). Similar results were observed inexperiments where the cells were stimulated with the D 1 receptor agonist in the absence of the phosphodiesteraseinhibitor (62.5 ± 8.6% above control; data not shown),indicating that this effect was not due to a cell overloadwith cAMP following inhibition of phosphodiesterases.Moreover, the increase in GluR4 phosphorylation induced by SKF38393 was completely abolished by pre-incubatingthe cells with 1  l M  H-89 (72.3 ± 4.2% of control; Fig. 3b).These results indicate that GluR4 phosphorylation at Ser842,induced both by FSK and D 1  receptor activation withSKF38393, occurs via PKA.Activation of group II mGluR with the selective agonist DCG-IV (1  l M ) slightly decreased GluR4 phosphorylation,although this effect was not statistically significant (Fig. 3c).However, DCG-IV inhibited the effect of both FSK and Fig. 2  Effect of dopamine D 1  receptors and group II mGluR activationon Ca 2+ influx through AMPA-type glutamate receptors in response tokainate, in Na + -free medium. (a) Effect of dopamine D 1  receptoractivation was measured after pre-incubation with the D 1  receptorspecific agonist SKF38393 (10  l M ) for 3 min. When Rp-cAMPs(500  l M ) was used, the cells were pre-incubated with the kinaseinhibitor for 1 h before stimulation with SKF38393. Inhibition of dop-amine D 1  receptors was carried out with SKF83566 (10  l M ) for 4 min,before addition of the receptor agonist. (b) and (c) Effect of Group IImGluR activation with DCG-IV (1  l M ) for 4 min, and effect of pre-incubation with DCG-IV on the response to AC stimulation with FSK(b) or D 1  receptor activation with SKF38393 (c). The cells were incu-bated with FSK (10  l M ) or SKF38393 (10  l M ) 3 min before stimulationwith kainate. Data are expressed as percentage of control of themaximal [Ca 2+ ] i  response ( n [Ca 2+ ] i ) to 100  l M  kainate, and are means± SEM of the indicated number of experiments carried out at least induplicate in different preparations. * p <   0.05, ** p <   0.01, Bonferroni’stest. Regulation of AMPA receptors by PKA  677   2004 International Society for Neurochemistry,  J. Neurochem.  (2004)  90 , 673–682
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks