Rapid, experience-dependent expression of synaptic NMDA receptors in visual cortex in vivo

Transcription

1 artiles Rapid, experiene-dependent expression of synapti NMDA reeptors in visual ortex in vivo Elizaeth M. Quinlan 1, enjamin D. Philpot 1, Rihard L. Huganir 2 and Mark F. ear 1 1 Howard Hughes Medial Institute, Department of Neurosiene, rown University, Providene, Rhode Island 2912, USA 2 Howard Hughes Medial Institute, Department of Neurosiene, The Johns Hopkins University Shool of Mediine, altimore, Maryland 2, USA The first two authors ontriuted equally to this paper. Correspondene should e addressed to M.F.. (mark_ear@rown.edu) Sensory experiene is ruial in the refinement of synapti onnetions in the rain during development. It has een suggested that some forms of experiene-dependent synapti plastiity in vivo are assoiated with hanges in the omplement of postsynapti glutamate reeptors, although diret evidene has een laking. Here we show that visual experiene triggers the rapid synapti insertion of new NMDA reeptors in visual ortex. The new reeptors have a higher proportion of suunits and, as a onsequene, different funtional properties. This effet of experiene requires NMDA reeptor ativation and protein synthesis. Thus, rapid regulation of postsynapti glutamate reeptors is one mehanism for developmental plastiity in the rain. Changes in NMDA reeptor expression provide a mehanism y whih rief sensory experiene an regulate the properties of NMDA reeptor-dependent plastiity in visual ortex. The neuronal response to glutamate released at exitatory synapses depends on the omplement of glutamate reeptors in the postsynapti memrane. Changes in reeptor numer, type and moleular omposition an sustantially alter the properties of synapti transmission. During postnatal development, exitatory synapti transmission is readily modified y sensory experiene. Thus, it has een suggested that some forms of experiene-dependent synapti plastiity are assoiated with rapid hanges in the omplement of postsynapti glutamate reeptors. Most attention has foused on hanges in the synapti expression of AMPA reeptors. However, NMDA reeptors (NMDARs) are ritial in triggering experiene-dependent synapti modifiations, and hanges in synapti expression of these reeptors ould have a large effet on the properties of synapti plastiity during development. In the present study, we addressed the possiility that experiene rapidly regulates the synapti expression of NMDARs in visual ortex in vivo. NMDARs in vivo are heteromeri ion hannels omposed of and NR2 suunits. The sutype (A D) of the NR2 suunit onfers distint funtional properties to the reeptor 1 7. Reeptors ontaining NR2 predominate in the neonatal forerain, and over the ourse of development, these are replaed or supplemented with -ontaining reeptors 2,8. This suunit swith alters the hannel properties suh that the synapti NMDARmediated urrents shorten in duration 2,9. The developmental shortening of NMDAR urrents in visual ortial neurons is postponed when animals are deprived of vision 1, suggesting that the / suunit omposition of synapti NMDARs differs etween dark-reared animals and light-reared ontrols. We tested this hypothesis and then examined the effets of rief light exposure in dark-reared animals. Our results show that new NMDARs with a higher / ratio are inserted into synapti memrane within one hour of the onset of visual experiene. RESULTS To examine NMDAR suunit omposition, we used quantitative immunolotting for, NR2 and in synaptoneurosomes prepared from the visual orties of postnatal day (P) light-reared and dark-reared Long-Evans rats. Synaptoneurosomes are a iohemial fration that is enrihed for synapti proteins 11. This iohemial analysis showed that synaptoneurosomal protein levels were signifiantly higher in light-reared visual ortex than in dark-reared ortex (mean optial density (OD) ± s.e.m., light reared, 177 ± 224; dark reared, 743 ± 152, paired t-test, p <.5, n = 6). In ontrast, levels of NR2 and proteins were not affeted y dark rearing (NR2 light reared, 197 ± 1; dark reared, 1777 ± 133, paired t-test, p >.5, n = 6; light reared, 1656 ± 22; dark reared, 1656 ± 222, paired t-test, p >.1, n = 8). Thus, as predited, a lower / ratio (Fig. 1) is orrelated with slower NMDAR urrent kinetis 1 in dark-reared visual ortex. The effet of dark rearing in visual ortex seems to e a speifi onsequene of sensory deprivation, as there were no detetale differenes in NMDAR protein levels in the hippoampi of dark-reared or lightreared animals ( light reared, 9 ± 184 ; dark reared, 167 ± 161, paired t-test, p >.1, n = 6; NR2 light reared, 873 ± 225; dark reared, 838 ± 221, paired t-test, p >.1, n = 6). To examine the effets of visual experiene on synaptoneurosomal NMDARs, we exposed dark-reared animals to a normal 12:12 light:dark yle for 24 or 48 hours. These experiments revealed that the redution in synaptoneurosomal ontent in darkreared animals ould e ompletely reversed within one day of 352 nature neurosiene volume 2 no 4 april 1999

2 artiles Fig. 1. Visual experiene regulates the omposition A of NMDARs in synaptoneurosomes from 15 visual ortex. Quantitative immunolotting for NMDAR suunit proteins of synaptoneurosomes prepared from visual orties of P21 23 rats raised in a normal (12:12) light:dark yle (LR), in omplete darkness (DR) or in darkness from irth followed y exposure to the normal light yle for 24 (+24) or 48 (+48) h. (a) protein * is redued in visual ortex synaptoneuro- 5 somes from dark-reared animals, and this redution is reversed y susequent light exposure LR DR (one-way ANOVA. p =.4, *signifiant differene versus all other groups in Tukey HSD post- ho omparison, p <.5). The O.D. from eah sample was normalized to the light-reared ontrol run on the same gel, and the summarized data are presented as perent of LR values (mean ± s.e.m.). Inset, representative immunolot for protein in synaptoneurosomes from LR, DR, DR + 24 and DR + 48 visual ortex. () Visual experiene does not alter the omplement of NR2 protein in synaptoneurosomes from visual ortex (one-way ANOVA, p =.93). The same synaptoneurosomes were proed as in (a). Inset, representative immunolot for NR2 protein in synaptoneurosomes from LR, DR, DR + 24 and DR + 48 visual ortex. () Visual experiene does not alter the omplement of protein in synaptoneurosomes from visual ortex (one-way ANOVA, p =.176). Immunolotting for protein was done simultaneously with the immunolotting for protein depited in (a). Inset, representative immunolot for protein in synaptoneurosomes from LR, DR, DR + 24 and DR + 48 visual ortex. Fig. 2. The ifenprodil sensitivity of NMDARmediated field potentials in visual ortex is experiene dependent. (a) The preparation. Slies of visual ortex were prepared from P21 28 rats, and synapti field potentials were reorded in layers 2/3 in response to layer-4 stimulation. () Experimental design. NMDAR-mediated synapti field potentials were isolated pharmaologially. To determine the ontriution of NR2-ontaining NMDARs to this response, we applied 3 µm ifenprodil. The remaining synapti field potential was ompletely loked y µm AP5 (DL-2-amino-5-phosphonovaleri aid), onfirming that the response reflets NMDARmediated urrents. (The residual AP5- insensitive negativity is non-synapti and is not affeted y the rearing history of the animal.) This representative example was reorded from a slie taken from a P26 dark-reared rat. The field potentials shown are averages of four onseutive responses taken at the indiated a A FP amplitude (mv) a Perent % of light-reared S light exposure (Fig. 1). Again, there were no signifiant effets of visual experiene on the levels of NR2 or. These iohemial data show that the ratio of / protein at the synapse is affeted y light deprivation and susequent visual experiene. However, it was ritial to determine whether these differenes in protein reflet differenes in the omposition of funtional NMDARs at visual ortial synapses. To address this question, we examined the ifenprodil sensitivity of synaptially evoked, NMDAR-mediated field potentials in slies of visual ortex from P21 28 rats. Ifenprodil seletively loks NR2-ontaining NMDARs 12, so enhaned ifenprodil sensitivity indiates a lower / ratio in the synapti NMDARs. The NMDAR omponent of layer 2/3 field potentials was pharmaologially isolated in artifiial ererospinal fluid (ACSF) ontaining redued Mg 2+ and lokers of AMPA and GAA A reeptors (Fig. 2). One a stale aseline was otained, we applied 3 µm ifenprodil, a onentration that seletively loks NR2ontaining NMDARs The perent inhiition y ifenprodil was alulated y omparing the aseline data with those olleted 9 minutes after the onset of drug appliation. In our initial experiments, slies from dark-reared and light-reared animals were studied with the experimenter lind to the rearing ondi µm ifenprodil µm AP Time from ifenprodil appliation (min) times. (Trae 1 is the AMPAR-dominated field potential reorded efore isolation of the NMDAR omponent.) Throughout the experiment, there was no hange in the non-synapti omponent of the evoked response. Sale ar,.2 mv and 5 ms. Dashed line, 3-minute average of the NMDARmediated field potential amplitude efore ifenprodil appliation. () NMDAR-mediated responses in dark-reared rats are more sensitive to ifenprodil than light-reared ontrols, and the effet of dark rearing is reversed y 24, 48 or 96 h in a normal light yle. Data are mean (± s.e.m.) redution after 9-minute ifenprodil appliation relative to the 3-minute aseline average. *Signifiant differene versus all other groups in Tukey HSD post-ho omparison, p <.5. (d) Partial lokade of NMDARs with AP5, whih does not distinguish etween NMDAR sutypes, reveals no differene etween dark-reared and light-reared groups. II/III IV WM 3 4 d Perent redution y ifenprodil Normalized NMDA FP Perent % of of light-reared C Perent % of of light-reared NR2 LR DR LR DR * LR DR LR DR 1 µm AP Time from AP5 appliation (min) nature neurosiene volume 2 no 4 april

3 light exposure on the ifenprodil sensitivity of NMDAR-mediated responses (Fig. 3d). Two hours of visual experiene (n = 1 slies from 6 rats) was suffiient to produe a signifiant redution in ifenprodil sensitivity ompared to dark-reared ontrols (n = 6 slies from 5 rats; t-test, p <.5), onsistent with a hange in the / ratio. Taken together, the data show that two hours of visual experiene is suffiient to signifiantly alter the moleular omposition (and therefore funtion) of synapti NMDARs in visual ortex. To our knowledge, this is the first demonstration that sensory experiene (versus deprivation) an indue a hange in the omplement of postsynapti glutamate reeptors in vivo. Experiene-dependent modifiations of synapti responses in visual ortex have een shown to require NMDAR ativation 16,17. We therefore were interested to know whether the experiene-indued regulation of NMDAR omposition was itself dependent on NMDAR ativation in the visual pathway. To address this question, we injeted dark-reared animals in the dark with the ompetitive NMDAR antagonist CPP (3-[2-aroxypiperazin-4-yl]-propyl-1-phosphoni aid, 1 15 mg per kg, i.p.) 3 minutes efore exposing them to light for two hours. Suseartiles Perent of % of dark-reared dark-reared tions. This experiment revealed a signifiant inrease in the ifenprodil sensitivity of NMDAR-mediated responses in darkreared visual ortex (n = 12 slies from 7 rats) as ompared to light-reared ontrols (n = 15 slies from 7 rats; t-test, p <.5). To onfirm the seletivity of this effet, we repeated the experiment using a susaturating onentration of AP5 (1 µm) that produes a partial lok of NMDAR responses, ut does not distinguish among the NMDAR suunits. Unlike ifenprodil, the AP5-sensitivity of NMDAR responses was omparale in the dark-reared (n = 7 slies from 4 rats) and light-reared (n = 8 slies from 4 rats) groups (t-test, p =.49). There were also no signifiant differenes in the asolute magnitude of the AMPAreeptor-dominated or NMDAR-mediated field potentials etween light-reared and dark-reared groups, or in the stimulation intensity required to eliit 8% of the maximal response. To test whether visual experiene ould reverse the effets of dark rearing, we examined ifenprodil sensitivity in a seond series of slies taken from animals that were light reared (n = 27 slies from 13 rats), dark reared (n = 26 slies from 13 rats, inluding those used for the lind portion of the study) or dark reared and then exposed to 24 (n = 14 slies from 8 rats), 48 (n = 15 slies from 7 rats) or 96 hours (n = 1 slies from 5 rats) of the normal light yle (Fig. 2). Exposing dark-reared rats to a regular light yle was suffiient to restore the light-reared phenotype (oneway ANOVA, F 4,91 = 4.85, p =.1). The data demonstrate that 24 or more hours of visual experiene is suffiient to aolish the effet of dark rearing. oth the iohemial and the eletrophysiologial results show that the synapti NMDARs have a lower / ratio in dark-reared visual ortex than in lightreared ortex. These findings were not entirely unexpeted, as they are onsistent with previous reports on the effet of light deprivation on the development of NMDAR response kinetis and ifenprodil sensitivity in the visual pathway 1,15. However, it was a surprise that NMDARs in dark-reared ortex ompletely reovered the light-reared phenotype with one day of light exposure. Thus, our next step was to explore the minimum visual experiene required to hange NMDAR omposition. Darkreared animals were exposed to.5, 1, 1.5 or 2 hours of a normal lighted environment, and hanges in, NR2 and protein were measured in visual ortial synaptoneurosomes. These experiments showed that light exposure for as little as one hour indued a signifiant inrease in protein (Fig. 3a and ). Again, the levels of NR2 and were not signifiantly hanged y light exposure. Thus, within two hours of light exposure, the / ratio reahed the light-reared value (Fig. 3). To address the question of whether the rapid hange in synaptoneurosomal protein reflets an alteration in the omposition of funtional synapti NMDARs, we examined the effets of rief aa C Perent of % of dark-reared dark-reared d D Normalized NMDA FP FP DR LR 8 // LR / // LR µm ifenprodil DR DR Time from ifenprodil appliation (min) Fig. 3. rief light exposure indues a rapid hange in synapti NMDAR omposition and funtion in visual ortex. (a) Representative immunolots for and proteins in synaptoneurosomes prepared from visual orties of rats raised in omplete darkness (DR), DR plus.5, 1, 1.5 or 2 h of light, or in a normal light yle (LR). () rief light exposure inreased protein from synaptoneurosomes of visual ortex (one-way ANOVA, p <.5). Eah sample is normalized to the dark-reared ontrol run on the same gel (dashed line), and the summarized data are presented as perent of DR values (mean ± s.e.m.). () rief light exposure inreases / from synaptoneurosomes of visual ortex (one-way ANOVA, p <.5). For eah sample, the signal for /NR2 is alulated efore normalizing to the dark-reared ontrol run on the same gel (dashed line). Summarized data are presented as perent of DR values (mean ± s.e.m.). (d) rief light exposure dereases the sensitivity of NMDAR potentials to ifenprodil. Normalized NMDAR-mediated field potentials in dark-reared rats (filled irles) or dark-reared rats exposed to light for two hours (DR + 2, open irles). Data are mean value averaged for the four pulses of proing stimulation. After two hours of light, NMDAR-mediated field potentials in visual ortex slies are less sensitive to ifenprodil than slies from dark-reared rats (t-test at 9 min after ifenprodil, p <.1). Error ars indiate s.e.m. Dashed line, normalized 3-minute aseline. 354 nature neurosiene volume 2 no 4 april 1999

4 artiles quent eletrophysiologial analysis of the ifenprodil sensitivity of NMDAR-mediated responses in visual ortex revealed no signifiant differenes etween the CPPinjeted, dark-reared group (n = 1 slies from 6 rats) and the CPP-injeted group reeiving light exposure (dark-reared +2; n = 13 slies from 8 rats; t-test, p =.47). Thus, CPP treatment ompletely prevented the experiene-dependent modifiation of synapti NMDARs (Fig. 4a). Immunolot analysis onfirmed that the light-indued inrease in protein was also ompletely loked in the visual ortex of CPP-injeted animals (Fig. 4; raw mean O.D. ± s.e.m., dark reared, 11 ± 116; dark reared + 2 h light, 936 ± 12; one-tailed t-test, p =.34). Thus, NMDAR ativity is neessary for the experiene-dependent insertion of ontaining synapti NMDARs. In a final series of experiments, we examined whether the experiene-dependent inrease in synapti requires protein synthesis. Animals were injeted in the dark with the mrna translation inhiitor yloheximide (1 mg per kg, i.p.) 3 minutes efore reeiving one hour of light exposure. This treatment ompletely loked the experiene-indued inrease in protein (Fig. 4; raw mean OD ± s.e.m., dark reared, 946 ± 46; dark reared + 1 h light, 941 ± 62; n = 4, onetailed t-test, p =.5). DISCUSSION 8 Studies in vitro have suggested that some developmental hanges in glutamate reeptors are ativity dependent. For example, growing neuronal ultures under onditions of heightened 18 or redued 19 ativity an ause a hange in the surfae expression of synapti AMPA reeptors. Likewise, NMDAR suunit expression 14,2 and lustering at postsynapti sites 21 in vitro are regulated y presynapti ativity and postsynapti NMDAR ativation. However, these hanges invarialy have required days for their expression, too slow to aount for rapid, ativity-dependent synapti modifiations 22. Thus, the finding that only one to two hours of visual experiene an alter the omplement of postsynapti glutamate reeptors in visual ortex in vivo is of onsiderale interest. Suh a hange is fast enough to ontriute to the rapid synapti modifiations that have een reported in the visual ortex of dark-reared animals exposed to light A model onsistent with the results of our study is that sensory experiene regulates, via NMDAR ativation, the synthesis and postsynapti surfae expression of -ontaining NMDARs in the visual ortex in vivo. This proposed mehanism must e viewed as tentative eause our manipulations of NMDARs and protein synthesis were systemi, and might have exerted their effets at sites other than the visual ortex. However, the idea that NMDAR suunit expression an e ontrolled y NMDAR ativity reeives strong support from studies using ultured ereellar slies. In this preparation, postsynapti a Normalized NMDA FP Perent % of of dark-reared C Perent % of of dark-reared DR + 2 DR 3 µm ifenprodil Time from ifenprodil appliation (min) D D D D D D D CTL CPP CTL CHX Fig. 4. Treatment of animals with CPP, a ompetitive antagonist of NMDARs, or yloheximide, an inhiitor of mrna translation, loks the experieneindued inrease in /. (a) Effet of ifenprodil on the magnitude of normalized NMDAR-mediated field potentials in visual ortex slies taken from rats injeted in the dark with CPP. Mean value averaged for the four pulses of proing stimulation in the DR (filled irles) and DR + 2 (open irles) groups. () Immunolots of and proteins of synaptoneurosomes prepared from visual orties of darkreared rats treated with CPP (D) and from visual ortex of animals exposed to light for 2 h (+2) starting 3 min after CPP treatment. Summary data show that light exposure did not hange the level of suunit following CPP treatment (solid line). For omparison, data from visual ortex in DR and DR + 2 without CPP (dashed line) are shown. Eah sample was normalized to the mean dark-reared value, and the summarized data are presented as perent of DR values (mean ± s.e.m.). () Immunolots of and proteins of synaptoneurosomes prepared from visual orties of darkreared rats treated with yloheximide (D) and from visual ortex of animals exposed to light for 1 h (+1) starting 3 min after yloheximide (CHX) treatment. Light exposure produed no hange in the level of suunit following CHX treatment (solid line). For omparison, data from visual ortex in DR and DR + 1 without CHX (dashed line) are shown. NMDAR ativation (for several days and oupled with neuregulin) drives expression of the NR2C suunit, whih, in turn, regulates the phenotype of NMDARs in granule ells 2. There is also a preedent for experiene-dependent regulation of mrna translation in the visual ortex. Speifially, rief visual experiene stimulates the rapid polyadenylation and translation of α- CaMKII mrna 26. The loalization of and NR2 mrna to neurites of ultured neurons 27 raises the intriguing possiility that experiene-dependent regulation of synthesis ould our in dendrites 28 in response to synapti ativation. What are the funtional onsequenes of hanging the suunit omposition of NMDARs in visual ortex? In visual ortex, as elsewhere, the amount of alium passing through ativated NMDARs an determine whether a synapse undergoes long-term potentiation (LTP) or long-term depression (LTD) 29. The experiene-dependent inrease in the / ratio and the onomitant shortening of synapti NMDAR urrents are likely to have a signifiant impat on the properties of synapti plastiity. Shortening NMDAR urrents would e expeted to alter the LTD-LTP modifiation threshold (θ m ), making LTD more likely and LTP less likely in response to a given amount of synapti ativation 3. Indeed, studies of LTD and LTP in visual ortex reveal preisely this hange in dark-reared animals exposed to nature neurosiene volume 2 no 4 april

5 artiles light 31. Theoretial investigations suggest that these ativitydependent adjustments of θ m are ruially involved in ortial development y maintaining the network of modifiale synapses within a useful dynami range 32. We suggest that one moleular mehanism for suh experiene-dependent modifiations of synapti plastiity in visual ortex is the regulation of NMDAR suunit omposition. METHODS Immunolot analysis. Male and female P21 23 Long-Evans rats (Charles River) were anesthetized with methoxyflurane vapor either in the dark (dark-reared) or the light (light-reared and light-exposure groups) and were deapitated following disappearane of orneal reflexes in ompliane with the U.S. Department of Health and Human Servies and rown University guidelines. Light exposure was egun 5 7 h into the light yle. Synaptoneurosomes were prepared using a proedure adapted from ref. 11. The primary visual ortex was rapidly disseted in ieold dissetion uffer (212.7 mm surose, 2.6 mm KCl, 1.23 mm NaH 2 PO4, 26 mm NaHCO 3, 1 mm dextrose, 1 mm MgCl 2,.5 mm CaCl 2,.2 mm CNQX and.1 mm AP5, saturated with 95% O 2 and 5% CO 2 ) and immediately homogenized in ie-old homogenization uffer (1 mm HEPES, 1 mm EDTA, 2 mm EGTA,.5 mm DTT,.1 mm PMSF, 1 mg per liter leupeptin, 5 mg per liter soyean trypsin inhiitor and nm miroystin). Tissue was homogenized in a glass glass tissue homogenizer (Kontes, Vineland, New Jersey), and the homogenate was passed sequentially through two -µm-pore nylon mesh filters, followed y a 5-µm-pore filter, and entrifuged at g for 1 min. The resulting pellets were resuspended in oiling 1% SDS and stored at 8 C. Equal amounts of synaptoneurosome protein, determined using the CA assay (Piere, Rokford, Illinois), were resolved on 7.5% polyarylamide gels, transferred to nitroellulose and proed with either anti- or anti-nr2 polylonal antiodies (1:; ref. 6) or anti- monolonal antiody (1:, lone 54.1, Pharmingen, San Diego, California), followed y the appropriate seondary antiody oupled to horseradish peroxidase (1:35, Sigma, St. Louis, Missouri) in Tris-uffered saline, ph 7.3, ontaining 1% ovine serum alumin and.1% Triton X- (Sigma). Visualization of immunoreative ands was produed y enhaned hemiluminesene (Amersham ECL) aptured on autoradiography film (Amersham Hyper ECL). Digital images produed y densitometri sans of autoradiographs on a SanJet IIx (Hewlett Pakard) with DeskSan II software (Hewlett Pakard) were quantified using NIH Image 1.6 software. The intensity of eah and was determined relative to a aseline immediately aove and elow the and within the same lane, and normalized to light-reared or dark-reared ontrols run on the same gel. Eletrophysiology. Slie eletrophysiology experiments were done as desried 33. riefly, rats were deeply anesthetized with inhalation anestheti methoxyflurane and deapitated. The rain was removed, disseted and slied in dissetion uffer as desried aove, with the exeptions that the uffer ontained 3 mm MgCl 2, 1 mm CaCl 2 and 5 1 mm kynureni aid instead of CNQX and AP5. Slies were allowed to reover for 1 2 h at room temperature in artifiial ererospinal fluid (ACSF) ontaining 124 mm NaCl, 5 mm KCl, 1.25 mm Na 2 PO 4, 26 mm NaHCO 3, 1 mm MgCl 2, 2 mm CaCl 2 and 1 mm dextrose, saturated in 95% O 2, 5% CO 2. For reording, slies were plaed in a sumersion reording hamer, maintained at 3 C and perfused with ACSF at a rate of 2 ml per min. Extraellular eletrodes (filled with ACSF; 1. MΩ) were used to monitor field potentials evoked with a stimulating eletrode (onentri ipolar tungsten). The magnitude of responses was monitored y the amplitude of the field potential. Stale aseline responses were eliited two per min at 8% of maximal response. NMDAR-mediated responses were pharmaologially isolated in artifiial ererospinal fluid (ACSF) ontaining 3 mm CaCl 2,.1 mm MgCl 2,.1% DMSO, 2 µm CNQX, 1 µm glyine and.5 µm iuulline methiodide. Proing stimulation onsisting of 4 pulses delivered at 3-seond intervals was given every 1 min to assess NMDAR-mediated field potentials. eause of studies demonstrating inhiition of -ontaining NMDARs y zin 34, we did pilot studies to demonstrate that there was no toni inhiition of NMDARs y zin in our slie preparation (data not shown). NMDAR lokade was ahieved y the ath appliation of µm DL- 2-amino-5-phosphonovaleri aid (AP5; Sigma). Ifenprodil (3 µm; RI) was used to lok NR2-ontaining reeptors. This onentration of ifenprodil is in a range that produes nearly maximal inhiition of NR2ontaining NMDARs, has little effet on -ontaining NMDARs and does not affet voltage-dependent alium hannels Dose response urves done for this study also determined that this onentration is speifi to antagonism of the NMDAR (data not shown). As noted previously 15, we found that wash-out of ifenprodil is slow and inomplete. Therefore, to alulate the perent inhiition of the NMDARmediated field potential y ifenprodil, we ompared the 3-minute average immediately efore adding the drug with the average of 4 onseutive sweeps olleted 9 minutes after drug appliation, when the ifenprodil effet had ompletely equilirated. A similar differene etween lightreared and dark-reared ortex was found using another NR2-seletive antagonist, CP-11,66-27 (5 µm, Pfizer; data not shown). Drug injetions. To lok NMDA reeptors or protein synthesis, we injeted animals in the dark (i.p.) with either CPP or CHX. CPP-injeted animals were awake and alert. However, a fration of the animals displayed slightly redued loomotor ativity. CHX-injeted animals were awake and alert, and displayed no overt illness or distress 1.5 h after injetion. However, we did note in pilot studies that animals were visily sluggish at 2.5 h after the injetion. Therefore, we restrited analysis to the earlier time point. Synapti transmission and neuronal exitaility remain normal for many hours following the lokade of protein synthesis with CHX 35. ACKNOWLEDGEMENTS The authors thank D. Olstein, A. Sekhar, E. Sklar and S. Meagher for assistane. This work was supported in part y grants from the Human Frontiers Siene Program and the National Eye Institute. RECEIVED 11 DECEMER 1998, ACCEPTED 24 FERUARY Wyszynski, M. et al. Differential regional expression and ultrastrutural loalization of alpha-atinin-2, a putative NMDA reeptor-anhoring protein, in rat rain. J. Neurosi. 18, (1998). 2. Monyer, H., urnashev, N., Laurie, D. J., Sakmann,. & Seeurg, P. H. Developmental and regional expression in the rat rain and funtional properties of four NMDA reeptors. Neuron 12, (1994). 3. Ishii, T. et al. Moleular haraterization of the family of the N-methyl-Daspartate reeptor suunits. J. iol. Chem. 268, (1993). 4. Rostas, J. A. et al. Enhaned tyrosine phosphorylation of the 2 suunit of the N-methyl-D-aspartate reeptor in long-term potentiation. Pro. Natl. Aad. Si. USA 93, (1996). 5. Gingrih, M.., Traynelis, S. F., Conn, P. J. & Zheng, F. Tyrosine kinase potentiates NMDA reeptor urrents y reduing toni zin inhiition. Nat. Neurosi. 1, (1998). 6. Lau, L. F. & Huganir, R. L. Differential tyrosine phosphorylation of N- methyl-d-aspartate reeptor suunits. J. iol. Chem. 27, (1995). 7. Kohr, G. & Seeurg, P. H. Sutype-speifi regulation of reominant NMDA reeptor-hannels y protein tyrosine kinases of the sr family. J. Physiol. (Lond.) 492, (1996). 8. Sheng, M., Cummings, J., Rolden, L. A., Jan, Y. N. & Jan, L. Y. Changing suunit omposition of heteromeri NMDA reeptors during development of rat ortex. Nature 368, (1994). 9. Flint, A. C., Maish, U. S., Weishaupt, J. H., Kriegstein, A. R. & Monyer, H. suunit expression shortens NMDA reeptor synapti urrents in developing neoortex. J. Neurosi. 17, (1997). 1. Carmignoto, G. & Viini, S. Ativity-dependent derease in NMDA reeptor responses during development of the visual ortex. Siene 258, (1992). 11. Hollingsworth, E.. et al. iohemial haraterization of a filtered synaptoneurosome preparation from guinea pig ereral ortex: yli adenosine 3 :5 -monophosphate-generating systems, reeptors, and enzymes. J. Neurosi. 5, (1985). 12. Williams, K., Russell, S. L., Shen, Y. M. & Molinoff, P.. Developmental swith in the expression of NMDA reeptors ours in vivo and in vitro. Neuron 1, (1993). 13. Churh, J., Flether, E. J., axter, K. & MaDonald, J. F. lokade y ifenprodil of high voltage-ativated Ca 2+ hannels in rat and mouse ultured hippoampal pyramidal neurones: omparison with N- methyl-d-aspartate reeptor antagonist ations. r. J. Pharmaol. 113, (1994). 356 nature neurosiene volume 2 no 4 april 1999

6 artiles 14. Gottmann, K., Mehrle, A., Gisselmann, G. & Hatt, H. Presynapti ontrol of suunit omposition of NMDA reeptors mediating synapti plastiity. J. Neurosi. 17, (1997). 15. Ramoa, A. S. & Prusky, G. Retinal ativity regulates developmental swithes in funtional properties and ifenprodil sensitivity of NMDA reeptors in the lateral geniulate nuleus. Dev. rain. Res. 11, (1997). 16. Roerts, E.., Meredith, M. A. & Ramoa, A. S. Suppression of NMDA reeptor funtion using antisense DNA loks oular dominane plastiity while preserving visual responses. J. Neurophysiol. 8, (1998). 17. ear, M. F., Kleinshmidt, A., Gu, Q. & Singer, W. Disruption of experienedependent synapti modifiations in striate ortex y infusion of an NMDA reeptor antagonist. J. Neurosi. 1, (199). 18. Lissin, D. V. et al. Ativity differentially regulates the surfae expression of synapti AMPA and NMDA glutamate reeptors. Pro. Natl. Aad. Si. USA 95, (1998). 19. Turrigiano, G. G., Leslie, K. R., Desai, N. S., Rutherford, L. C. & Nelson, S.. Ativity-dependent saling of quantal amplitude in neoortial neurons. Nature 391, (1998). 2. Ozaki, M., Sasner, M., Yano, R., Lu, H. S. & uonanno, A. Neuregulin-eta indues expression of an NMDA-reeptor suunit. Nature 39, (1997). 21. Rao, A. & Craig, A. M. Ativity regulates the synapti loalization of the NMDA reeptor in hippoampal neurons. Neuron 19, (1997). 22. Craig, A. M. Ativity and synapti reeptor targeting: the long view. Neuron 21, (1998). 23. uisseret, P., Gary-oo, E. & Imert, M. Oular motility and reovery of orientational properties of visual ortial neurones in dark-reared kittens. Nature 272, (1978). 24. uisseret, P., Gary-oo, E. & Imert, M. Plastiity in the kitten s visual ortex: effets of the suppression of visual experiene upon the orientational properties of visual ortial ells. Dev. rain Res. 4, (1982). 25. Fox, K., Daw, N., Sato, H. & Czepita, D. Dark-rearing delays the loss of NMDA-reeptor funtion in kitten visual ortex. Nature 35, (1991). 26. Wu, L. et al. CPE-mediated ytoplasmi polyadenylation and the regulation of experiene-dependent translation of α-camkii mrna at synapses. Neuron 21, (1998). 27. Miyashiro, K., Dihter, M. & Eerwine, J. On the nature and differential distriution of mrnas in hippoampal neurites: impliations for neuronal funtioning. Pro. Natl. Aad. Si. USA 91, (1994). 28. Steward, O. mrna loalization in neurons: a multipurpose mehanism? Neuron 18, 9 12 (1997). 29. ear, M. F. & Malenka, R. C. Synapti plastiity: LTP and LTD. Curr. Opin. Neuroiol. 4, (1994). 3. Gold, J. I. & ear, M. F. A model of dendriti spine Ca 2+ onentration exploring possile ases for a sliding modifiation threshold. Pro. Natl. Aad. Si. USA 91, (1994). 31. Kirkwood, A., Rioult, M. G. & ear, M. F. Experiene-dependent modifiation of synapti plastiity in visual ortex. Nature 381, (1996). 32. ienenstok, E. L., Cooper, L. N. & Munro, P. W. Theory for the development of neuron seletivity: Orientation speifiity and inoular interation in visual ortex. J. Neurosi. 2, (1982). 33. Kirkwood, A., Dudek, S. M., Gold, J. T., Aizenman, C. D. & ear, M. F. Common forms of synapti plastiity in the hippoampus and neoortex in vitro. Siene 26, (1993). 34. Paoletti, P., Asher, P. & Neyton, J. High-affinity zin inhiition of NMDA - reeptors. J. Neurosi. 17, (1997). 35. Otani, S., Marshall, C. J., Tate, W. P., Goddard, G. V. & Araham, W. C. Maintenane of long-term potentiation in rat dentate gyrus requires protein synthesis ut not messenger RNA synthesis immediately post-tetanization. Neurosiene 28, (1989). nature neurosiene volume 2 no 4 april