81:204-215, 1999. J NeurophysiolS.N.M. Reid and D. Czepita N. W. Daw, B. Gordon, K. D. Fox, H. J. Flavin, J. D. Kirsch, C. J. Beaver, Q.-H. Ji,

You might find this additional information useful...

31 articles, 20 of which you can access free at: This article cites http://jn.physiology.org/cgi/content/full/81/1/204#BIBL

13 other HighWire hosted articles, the first 5 are: This article has been cited by

  [PDF]  [Full Text]  [Abstract]

, October 5, 2005; 25 (40): 9266-9274. J. Neurosci.A. Kalatsky and M. G. Frank S. K. Jha, B. E. Jones, T. Coleman, N. Steinmetz, C.-T. Law, G. Griffin, J. Hawk, N. Dabbish, V.

Sleep-Dependent Plasticity Requires Cortical Activity 

[PDF]  [Full Text]  [Abstract], December 7, 2005; 25 (49): 11433-11443. J. Neurosci.

Kirkwood S.-Y. Choi, J. Chang, B. Jiang, G.-H. Seol, S.-S. Min, J.-S. Han, H.-S. Shin, M. Gallagher and A.

CortexMultiple Receptors Coupled to Phospholipase C Gate Long-Term Depression in Visual 

[PDF]  [Full Text]  [Abstract], March 1, 2006; 95 (3): 1718-1726. J Neurophysiol

S. D. Faulkner, V. Vorobyov and F. Sengpiel Experience

Visual Cortical Recovery From Reverse Occlusion Depends on Concordant Binocular 

[PDF]  [Full Text]  [Abstract], May 21, 2008; 28 (21): 5513-5518. J. Neurosci.

S. Gais, B. Rasch, U. Wagner and J. Born Antagonists

Visual-Procedural Memory Consolidation during Sleep Blocked by Glutamatergic Receptor 

[PDF]  [Full Text]  [Abstract], June 16, 2009; 106 (24): 9860-9865. PNAS

B.-J. Yoon, G. B. Smith, A. J. Heynen, R. L. Neve and M. F. Bear Essential role for a long-term depression mechanism in ocular dominance plasticity

on the following topics: http://highwire.stanford.edu/lists/artbytopic.dtlcan be found at Medline items on this article's topics

Chemistry .. Iontophoresis Veterinary Science .. Visual Cortex Psychology .. Visual Processing Neuroscience .. Visual Response Cell Biology .. Histograms Biochemistry .. Aspartate

including high-resolution figures, can be found at: Updated information and services http://jn.physiology.org/cgi/content/full/81/1/204

can be found at: Journal of Neurophysiologyabout Additional material and information http://www.the-aps.org/publications/jn

This information is current as of August 9, 2009 .  

http://www.the-aps.org/.Physiological Society. ISSN: 0022-3077, ESSN: 1522-1598. Visit our website at by the American Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2005 by the American

publishes original articles on the function of the nervous system. It is published 12 times a year (monthly)Journal of Neurophysiology

on August 9, 2009

jn.physiology.orgD

ownloaded from

Injection of MK-801 Affects Ocular Dominance Shifts More ThanVisual Activity

N. W. DAW, B. GORDON, K. D. FOX, H. J. FLAVIN, J. D. KIRSCH, C. J. BEAVER, Q.-H. JI, S.N.M. REID,AND D. CZEPITADepartment of Ophthalmology and Visual Science, Yale University Medical School, New Haven, Connecticut 06520-8061

Daw, N. W., B. Gordon, K. D. Fox, H. J. Flavin, J. D. Kirsch, (MK-801), also directly into the visual cortex (RauscheckerC. J. Beaver, Q.-H. Ji, S.N.M. Reid, and D. Czepita. Injection et al. 1990).of MK-801 affects ocular dominance shifts more than visual activ- Do NMDA antagonists affect ocular dominance shiftsity. J. Neurophysiol. 81: 204–215, 1999. Kittens were given intra- simply because of a reduction in visually driven activity inmuscular injections of the N-methyl-D-aspartate (NMDA) antago- the visual cortex? Ocular dominance shifts are a sensory-nist MK-801 twice daily (morning and midday) during the peak

driven phenomenon, and the sensory signals are carried toof the period of susceptibility for ocular dominance changes. Theythe visual cortex by electrical activity. A substantial reduc-were then exposed to light with one eye closed for 4 h after eachtion in activity might therefore prevent ocular dominanceinjection. The ocular dominance of these kittens was shifted sig-shifts, just as infusion of tetrodotoxin (TTX) into the eyesnificantly less than that of kittens injected with saline and exposed

to light over the same period at the same age. After recording a blocks ocular dominance segregation during normal devel-sample of cells for an ocular dominance histogram, the kittens opment (Stryker and Harris 1986) and infusion of TTX intowere injected with the same dose of MK-801 that was used during the visual cortex prevents ocular dominance shifts from MDrearing to observe its effect on the activity of single cells in the (Reiter et al. 1987). Compared with TTX, APV produces avisual cortex. In the majority of cells (7/13) there was no signifi- less dramatic decrease in activity. Iontophoresis of APVcant change in activity. Positive evidence for a reduction in activity

reduces activity in the visual cortex, and this effect is greaterwas seen in only a minority (3/13) of cells. In a separate seriesin young animals (Tsumoto et al. 1987). APV is effectiveof experiments, dose-response curves were measured for cells inin all layers before 3 wk of age, but the effect becomesthe visual cortex in response to iontophoresis of NMDA or a-

amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), restricted primarily to layers II and III after 6 wk of ageand the effect of an injection of MK-801 on these curves was (Fox et al. 1989). Infusion of 50 mM APV into the visualmeasured. MK-801, at doses similar to those used in the ocular cortex at 1 ml /h ( the procedure used by Kleinschmidt et al.dominance experiments, had a significant effect on the dose-re- 1987) leads to a substantial depression of activity in allsponse curves for NMDA, but little effect on the dose-response layers, even in adult animals (Miller et al. 1989).curves for AMPA, or the visual responses of the cells. We conclude

As a control for their experiments on ocular dominancethat ocular dominance shifts can be reduced significantly by achanges, Bear et al. (1990) recorded from cells in the cortextreatment that has little effect on the level of activity of cells inof their animals 2 days after starting infusion of 50 mMthe visual cortex but does specifically affect the responses of theAPV. They found that the percentage of visually drivencells to NMDA as opposed to the responses to AMPA.cells in the area from which they obtained ocular dominancehistograms was normal, although the response was de-

I N T R O D U C T I O N pressed. They did not quantify the extent of the depression.Moreover, the animals tested for the effect of APV on activ-Permanent irreversible changes occur in the visual cortexity in the visual cortex were different from those tested forwhen the visual input is restricted early in life (Wieselthe effect of APV on ocular dominance shifts.1982). The clearest example is the effect of monocular dep-

We therefore decided to measure the effect of an NMDArivation (MD). After MD the cortex becomes effectivelydisconnected from the deprived eye. N-Methyl-D-aspartate antagonist on ocular dominance shifts and also the effect of(NMDA) receptors are believed to play an important role the same NMDA antagonist at the same dose on visual activ-in this process. The prime evidence for this conclusion is ity in the same animals. We injected animals intramuscularlythat infusion of the NMDA antagonist D-2-amino-5-phos- with MK-801, which crosses the blood–brain barrier, keptphonovaleric acid (APV) into the visual cortex reduces the one eye closed for several days, recorded an ocular domi-ocular dominance shift (Bear et al. 1990; Kleinschmidt et nance histogram, and then tested the effect of the same doseal. 1987). A similar result was obtained with infusion of the of MK-801 on the activity of a single cell in the visualNMDA channel blocker (/)-5-methyl-10,11-dihydro-5H- cortex. Results in these animals were compared with controldibenzo [a, d] cyclohepten - 5, 10 - imine hydrogen maleate animals injected with vehicle solution, but otherwise treated

identically. A brief summary of the first series of experimentsThe costs of publication of this article were defrayed in part by the was included in a review article (Daw 1994). As a controlpayment of page charges. The article must therefore be hereby marked

for whether the MK-801 had a specific effect in visual cortex,‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734 solely toindicate this fact. we then observed the effect of MK-801 on dose-response

204 0022-3077/99 $5.00 Copyright q 1999 The American Physiological Society

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

NMDA RECEPTORS AFFECT PLASTICITY MORE THAN ACTIVITY 205

TABLE 1. Conditions for monocular deprivation

EyelidSuture Deprivation Recording Weighted Ocular

Animal (Days of Age) Treatment (Days of Age) (Days of Age) Dominance

198A 34 MK-801 35–39 42 0.58198B 35 MK-801 35–39 43 0.515117 35 MK-801 36–40 41 0.70203A 35 Saline 36–40 43 0.71203B 35 Saline 36–40 44 0.684383 27 Saline 30–37 38 0.954382 27 MK-801 30–37 40 0.754385 36 MK-801 36–43 44 0.704387 37 Saline 38–45 46 0.894386 38 MK-801 39–46 47 0.474374 29 MK-801 30–37 38 0.704372 31 Saline 32–39 40 0.974376 30 Saline 30–37 38 0.83

anesthetic (Lidocaine). After surgery, the animal was paralyzed bycurves plotted from iontophoresis of NMDA or a-amino-3-intravenous infusion of pancuronium bromide at 0.6–1.5 mg/h (El-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA).kins-Sinn, Cherry Hill, NJ). Body temperature was maintained at37.57C with a heating pad controlled by a rectal thermometer. Heart

M E T H O D S rate and end tidal CO2 were monitored continuously, and CO2 wasmaintained at 3.5–4.2% by adjusting the respirator.General procedure

Treatment was started at 4–51/2 wk of age (Table 1). The eyelids Recordings for determination of ocular dominanceof the right eye were sutured closed. The animal was then exposed

The lid margins of the closed eye were separated, and both eyesto light for 8 h/day for 5 days in the first series of experimentswere focused on a tangent screen at 57 in. by lenses of zero powerand for 8 days in the second. It was kept in the dark with its motherand appropriate curvature. The nictitating membrane was with-at all other times until the day of recording.drawn by a drop of 10% phenylephrine hydrochloride (Neosyneph-Seven animals were treated with MK-801, and six were treatedrine) , and the pupils were dilated with a drop or two of 0.01%with an equal volume of saline. The first injection in the morningatropine. A tungsten electrode (Hubel 1957) or a carbon fiberwas given by using infrared light and a viewer, or, occasionally,microelectrode (Armstrong-James and Millar 1979) was insertedbrief room light to find the animal. The lights were turned on 10–into the cortex for recordings from single cells. The left cortex,15 min after the injection, when the MK-801 started to have anipsilateral to the open eye and contralateral to the sutured eye, waseffect. The dose was Ç0.1 mg/kg im. It was adjusted so that thechosen because there is a slight dominance in the cortex by theanimal was slightly ataxic but awake and playful as much as thecontralateral eye. Therefore, ocular dominance changes are morecontrol littermates. On three occasions the dose was slightly toonoticeable when the cortex ipsilateral to the open eye is recordedlarge, and the animal was sedated for Ç30 min, but the next dose(Daw et al. 1992).was then adjusted to make the animal ataxic but not sleepy. In

Receptive fields were characterized with a light projector movedgeneral, small increases in the dose were necessary over the daysby hand. The preferred width and length of stimulus, preferredof treatment to get a constant effect. A second injection was givenorientation, and velocity and direction of movement were all deter-4 h after the lights were turned on. Four hours after the secondmined with moving stimuli. If the cell responded to all directionsinjection, the lights were turned off.of movement, as judged by listening to the audio monitor, it wascalled omnidirectional (see Daw and Ariel 1980). If the cell re-MDsponded to movement along one axis, but substantially less alongthe perpendicular axis, it was called bidirectional. If the cell re-Eyelids were sutured together under anesthesia with ketamine (20–sponded to movement in one direction, but substantially less to the30 mg/kg im) and xylazine (0.5 mg/kg im). One drop of propara-opposite direction, it was called unidirectional. The width of tuningcaine hydrochloride (Alcaine) was dropped onto the cornea. The lidfor a moving bar was judged by moving the stimulus through themargins were then cut. Polymyxin b-bacitracin–neomycin (Neo-receptive field and repeating this with angles further and furthersporin) eye ointment was placed between the eyelids, which wereaway from the preferred direction of movement until the responsethen sutured together with 4–0 thread, leaving a small aperture at thewas no longer audible. Finally, the responses in the two eyes weremedial edge for drainage. Sutures were inspected daily to make surecompared by using the preferred stimulus. Two observers indepen-that they were healing and that no holes appeared.dently assigned an ocular dominance according to the seven-pointscale introduced by Hubel and Wiesel (1962).Preparation for physiology

About four penetrations were made in each cortex, spaced 0.5–1 mm apart between AP0 and P4. In six animals (3 experimentalAnimals were sedated with acepromazine, 0.1 mg/kg im (Fer-

mentia Animal Health, Kansas City, MO), and given a preanesthetic and 3 control in the second series) , the electrode was inserted ina parasaggital plane, angled at 15–207 to the vertical, to sampledose of atropine, 0.04 mg/kg im. Anesthesia was induced with 4%

halothane in a mixture of 67% nitrous oxide-33% oxygen and main- as many layers and columns as possible. In the other seven animals,the electrode was inserted in a coronal plane, again angled at 15–tained with 0.5–0.9% halothane. After tracheotomy and insertion of

an intravenous cannula into the femoral vein, the skull was opened 207 to the vertical, to sample as many layers and columns aspossible. Two or three lesions (3 mA DC for 10 s) were made inover the lateral gyrus, and a small hole was made in the dura for

insertion of the electrode. All wound margins were treated with local each penetration to determine the layers for the cells recorded.

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

DAW ET AL.206

current to accumulate a series of firing rates between 0 and 50Recordings for determination of the effect of MK-801 onHz. In an ideal experiment, we recorded visual and iontophoreticactivityresponses for two or three cells, then injected MK-801, and then

After determining the ocular dominance of a sample of cells, held the cell recording two or three visual responses and an ionto-a computer-controlled bar of light was set up to stimulate the phoretic response once an hour. Four hours after the first injectioncell in the dominant eye with the preferred orientation, direction of MK-801, a second injection of MK-801 was given, and weof movement, velocity, length, and width of the stimulus. The continued to record for another 4–6 h. Because of the difficultybar of light remained stationary on one side of the receptive of holding a cell for 10 h, we did not always obtain this idealfield for 1 s and then swept across the receptive field, remained result, in which case a new cell would be isolated and recorded withstationary for 1 s on the far side, swept back, remained stationary the same procedure. Visual responses, iontophoretic responses, andfor 1 s more, and was then turned off. This procedure was re- spontaneous activity were calculated off-line by our ASYST pro-peated four times every minute; the average of these four re- gram and a straight line fitted to the linear portion of the dose-sponses constituted one group of records. Spikes were discrimi- response curve by Slide Write (Advanced Graphics Software,nated through a voltage window and monitored for amplitude Carlsbad, CA).and time course on a storage oscilloscope. The computer wasalso used to store spike discharge times and to construct a peri-stimulus time histogram (PSTH) on-line as the spikes came in. R E S U L T SThe first PSTH was displayed in red, and each subsequent PSTHwas displayed in green, updated each minute, so that the effect Effect of MK-801 on ocular dominance shiftsof the injection of MK-801 could be monitored as it occurred.Spike discharge times together with details of the stimulation In the first series of experiments, we compared ocularparameters were stored on hard disk for subsequent off-line dominance histograms from three animals treated with MK-analysis with custom-written ASYST programs (Asyst Soft- 801 and two animals treated with saline. All animals wereware, Rochester, NY) . deprived for a total of 40 h, 8 h/day over 5 days and then

kept in the dark until recorded. Cells in the control animalsHistology were driven by the open eye more often than cells in the

animals treated with MK-801 (Fig. 1) . Weighted ocularOn completion of the physiological recordings, the animaldominances for the three animals treated with MK-801 werewas deeply anesthetized with 4% halothane and then perfused0.58, 0.51, and 0.70 compared with weighted ocular domi-through the heart with 100 ml of Lactated Ringer, followed by

350– 450 ml of 4% paraformaldehyde. The lateral gyrus was nances of 0.68 and 0.71 for the two control animals. In otherremoved and allowed to sink in a 30% sucrose solution con- words, treatment with MK-801 reduced the ocular domi-taining 4% paraformaldehyde. Frozen sections were cut at 60 nance shift, but not greatly.mm and then stained with thionin ( Nissl stain ) . Lesions were However, the ocular dominance shift was not very large in50 –100 mm in diameter. The electrode tracks were recon- the control animals, so it was not possible to reduce it substan-structed, and the layer in which each cell was recorded was tially. This may have been due to the shortness of the perioddetermined according to the layering criteria described by

of deprivation (40 h) or it may have been due to the 2–3 daysKelly and Van Essen ( 1974 ) .that the animals were kept in the dark between deprivation andrecording. Consequently, we decided to do a second series ofAnalysis of data animals in which the length of deprivation was increased to

For analysis of the effect of MK-801 on the activity of a cell, 64 h (8 h/day over 8 days) and the time between deprivationfiring rates, expressed as spikes/s, were averaged over the entire and recording was kept as short as possible.response duration for one direction of stimulus presentation (typi- Ocular dominance histograms from four animals treatedcally 1–2 s) . Visual responses were expressed as average firing with MK-801 were compared with ocular dominance histo-rate while the stimulus was in the receptive field minus spontaneous grams from four animals treated with saline. The histogramsactivity. If the cell was unidirectional, the firing rate was measured from the control animals were substantially more shiftedwhile the stimulus was moving in the preferred direction. If the

than in the first series of experiments (Fig. 2A) . Treatmentcell was bidirectional, the firing rate was considered to be thewith MK-801 caused a large decrease in this shift (Fig. 2B) .average of the firing rates in both directions.Weighted ocular dominances in the control animals wereWeighted ocular dominance (WOD) was calculated from the0.95, 0.89, 0.97, and 0.83, whereas weighted ocular domi-ocular dominance histograms according to the formulanances in the treated animals were 0.75, 0.70, 0.47, and 0.70.1/6N2 / 2/6N3 / 3/6N4 / 4/6N5 / 5/6N6 / N7

N1 / N2 / N3 / N4 / N5 / N6 / N7Considering each animal as one observation, these numbersare significantly different ( two-tailed t-test, P õ 0.05).

where Ni is the number of cells in ocular dominance group i (Kasa- Three animals (4385, 4387, and 4386) were a few daysmatsu et al. 1981). older than the others, but all animals were deprived during

the most susceptible part of the critical period, which is 4–Effect of MK-801 on dose response curves for NMDA or 6 wk of age (Hubel and Wiesel 1970; Olson and FreemanAMPA 1980), and there was no difference between the older ani-

mals and younger ones that received the same treatment.Fourteen animals were recorded, aged 37–43 days of age. AIn the first series of experiments we noticed that the mostcell was isolated, and the preferred parameters for stimulation were

substantial reduction in ocular dominance occurred in upperestablished. Visual response was measured, and then the responseslayers. Consequently, we arranged penetrations in the secondto iontophoresis of NMDA or AMPA were measured. For ionto-series of experiments to record mostly from upper layers.phoresis, the cycle was NMDA on for 15 s, off for 57 s, AMPA

on for 15 s, off for 57 s, and so on, varying the iontophoretic The histological reconstructions of the penetrations showed

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

NMDA RECEPTORS AFFECT PLASTICITY MORE THAN ACTIVITY 207

FIG. 1. Ocular dominance histograms from 3 ani-mals treated with MK-801 and deprived for 40 h com-pared with ocular dominance histograms from 2 animalsinjected with saline and deprived over the same periodof time.

that ú85% of the cells came from layers II, III, and IV. The of unidirectional cells was 48% in MK-801 animals com-pared with 60% in controls. A similar result was obtaineddifference between ocular dominance histograms recorded

from MK-801 treated animals and ocular dominance histo- in the first series ( see Daw 1994) . However, the angleover which the cell responded was not significantly differ-grams recorded from control animals was seen in all layers

(Fig. 3) . Weighted ocular dominances were 0.90 in layers ent: 86 { 247 for MK-801 animals compared with 88 {297 for controls. In the 13 cells where we had quantitativeII /III, 0.92 in layer IV, and 0.91 in layers V/VI for control

animals,and 0.66 in layers II /III, 0.67 in layer IV, and 0.67 records, the firing rate in animals treated with MK-801(9.3 { 3.2 spikes / s, means { SD) was not significantlyin layers V/VI for MK-801–treated animals. However, the

sample from layers V and VI was not large enough to make different from the firing rate in control animals (6.3 {3.2 spikes / s ) : if anything, the firing rate in the treatedit a meaningful result in those layers.

Receptive fields of cells in the animals treated with MK- animals was higher.801 were slightly less specific than those in the controlanimals. The percentage of omnidirectional cells was 9% Effect of MK-801 on activity of single cells in the cortexin MK-801 animals compared with 4% in controls, thepercentage of bidirectional cells was 43% in MK-801 ani- Thirteen cells were analyzed for the effect of MK-801 on

their activity. The dose of MK-801 given was the averagemals compared with 36% in controls, and the percentage

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

DAW ET AL.208

FIG. 2. Ocular dominance histograms from 4 ani-mals treated with MK-801 and deprived for 64 h com-pared with ocular dominance histograms from 4 animalsinjected with saline and deprived over the same periodof time.

of the doses given during the rearing paradigm for animals response decreased to Ç25% of control after 30 min andrecovered to 60% after 60 min ( 198A unit A ) , in one casetreated with MK-801 during rearing. The dose given to con-

trol animals was the same as that given to treated littermates the response decreased to 60% after 20 min and recoveredto 80% after 45 min ( 198B unit A ) , and in one case theduring rearing. Two cells were recorded in the first series

of experiments, and 11 cells were recorded in the second. response decreased to Ç50% and recovered after 10 min(4373 unit B ) . The largest of these decreases is illustratedEight were recorded from MK-801–treated animals, and five

were recorded from control animals. Details of the age of in Fig. 4C .In two cells, there was a long-term decrease in the re-the animal, MK-801 dose, and the layer of the cell recorded

are given in Table 2. sponse, which did not recover over the period of recording.It is difficult to know how to interpret this result becauseIn seven cells, there was little or no change in the visual

response after the injection of MK-801. Some of these seven the response of a cell quite frequently decays over time,particularly when the cell is recorded for a period of ¢1 h.cells had a fairly constant, or even increased, response (Fig.

4A) . In other cases, the response of the cell varied consider- Consequently the decrease in these two cases could havebeen due to small shifts in the distance of the electrode fromably over time, and any change was not statistically signifi-

cant (Fig. 4B) . the cell, as a result of which amplitude of the action potentialfell below the threshold of the window used to discriminateIn three cells, there was a decrease in the response,

correlated with the injection of MK-801. In one case the the action potential from noise.

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

NMDA RECEPTORS AFFECT PLASTICITY MORE THAN ACTIVITY 209

FIG. 3. Ocular dominance histograms from the 2ndseries of animals analyzed by layer. More than 50% ofthe cells were in group 7 in all layers in the control ani-mals, and õ20% of the cells were in group 7 in all layersin MK-801–treated animals.

Effect of MK-801 on responses of cells to NMDA andTABLE 2. Effect of MK-801 on activity of cells in animals AMPAtested for ocular dominance shifts

We obtained results in 10 animals for the effect of MK-801 on the cortical responses to NMDA and AMPA as wellMK-801 Dose,as on the visual response. An example is shown in Fig. 5.Cell Age mg/kg Layer EffectThe cell responded to movement in both directions for a bar

198A unit A 42 0.1 II/III Decrease over 30 min of light moved at 47 /s. The visual response 4.5 h after the198B unit A 43 0.1 II/III Decrease over 20 min first injection of MK-801 and 30 min after the second was4383 unit A 38 0.09 III/IV Decrease, but unstable

close to the response before injection of MK-801 (Fig. 5,4382 unit A 40 0.09 IV Little change4385 unit C 44 0.1 Unknown Little change top) . AMPA and NMDA, both iontophoresed at 20 nA, gave4387 unit A 46 0.13 IV Long-term decrease substantial responses before injection of MK-801 (Fig. 5,4386 unit A 47 0.16 Unknown Little change left) ; 4.5 h later, after two injections of MK-801, the currents4374 unit A 38 0.15 IV Little change

required were higher, but AMPA iontophoresed at 33 nA4372 unit A 40 0.09 IV Long-term decreasegave a substantial response, whereas NMDA iontophoresed4373 unit A 41 0.13 IV Little change

4373 unit B 41 0.13 II Decrease over 10 min at 35 nA gave very little response (Fig. 5, right) .4376 unit A 38 0.14 II/III Little change Dose-response curves were measured for most cells. Re-4377 unit A 40 0.14 III Little change sults from a layer II /III cell recorded from a 42-day-old

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

DAW ET AL.210

60–70 nA; 3.5 h after the injection of MK-801, the AMPAresponse was again similar, and the NMDA response showedsubstantial recovery.

Dose-response curves were measured for 40 cells in 10animals. In general, the slope of the line fitted to the NMDAdose-response curve was equal to the slope of the line fittedto the AMPA curve before MK-801 was injected (ratio0.98 { 0.64, n Å 12). Cells recorded after the first injectionof MK-801 had an NMDA dose-response curve that wasmuch flatter in relation to the AMPA curve (ratio 0.39 {0.37, n Å 11). For cells recorded after the second injectionof MK-801, the NMDA curve was reduced further (ratio0.28 { 0.27, n Å 15). Recovery was seen in three cases,and not in three others. However, it was rarely possible tohold a cell for the 10 h required for two injections of MK-801 and recovery after the second injection.

One of the few cases where we were able to hold a cellfor a long period is illustrated in Fig. 7. In this animal,four cells were recorded, and the fourth was held for 11 h.Immediately before the first injection of MK-801, theNMDA and AMPA curves had similar slopes. For 3 h afterthe first injection of MK-801, the slope of the NMDA curvedropped to 1/4 the slope of the AMPA curve (Fig. 7A) .Then the NMDA response recovered, just before the secondinjection of MK-801. After the second injection of MK-801,the slope of the NMDA curve dropped to õ1/10 the slopeof the AMPA curve. Finally, 8 h after the first injection ofMK-801, and 4 h after the second injection, the response toNMDA again recovered, in part. The visual response duringthis time was relatively constant (Fig. 7B) .

Visual responses were compared before and after the firstinjection of MK-801 for eight cells (Table 3). In generalthere was little change, some increases and some decreases,but no overall change. This was true for doses of MK-801in the range of those used in animals tested for ocular domi-nance changes (0.12–0.15 mg/kg) and for rather higherdoses (0.2 mg/kg). Visual responses were also comparedafter the second injection of MK-801 for seven cells (Table3). There was little change at doses used in the ocular domi-nance experiments but a substantial decrease in two of threecases where a rather higher dose was used.

D I S C U S S I O N

MK-801 produced a significant decrease in the oculardominance shift that normally occurs after MD. This oc-curred with doses that did not have a substantial effect onFIG. 4. A : cell for which injection of MK-801 had no effect. Record

shows the visual response averaged over 16 trials spaced over 4 and 2 min spontaneous activity or visually evoked activity of most cor-before and 2 min after the time shown on the horizontal axis. MK-801 was tical neurons. The decrease in the ocular dominance shiftinjected at time 0 . B : cell for which injection of MK-801 had no effect, occurred in all layers and was most obvious in layers II, III,but the response of the cell over time showed considerable variability. C :

and IV.cell for which injection of MK-801 decreased the response to 25% over 20The injection of MK-801 reduced the ocular dominancemin, with recovery to 60% after 60 min.

shift but did not abolish it altogether. Weighted ocular domi-nance in the MK-801–treated animals was 0.597 { 0.096animal are illustrated in Fig. 6. Initially the cell responded

up to 40 spikes/s to AMPA iontophoresed between 20 and in the first series and 0.655 { 0.016 in the second series.This compares with 0.43 in normal animals (Daw et al.30 nA, and NMDA iontophoresed between 25 and 35 nA;

2.2 h after the injection of MK-801, the responses to AMPA 1992). Bear et al. (1990) and Rauschecker et al. (1990) alsofound that the ocular dominance shift was not completelywere very similar to the responses seen before injection of

MK-801. The responses to NMDA, however, were consider- abolished by their injections of APV and MK-801 directlyinto the cortex. Of the many treatments that affect ocularably decreased. There was little or no response up to an

iontophoretic current of 50 nA and only a small response at dominance shifts (see Daw 1994 for a list) , only complete

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

NMDA RECEPTORS AFFECT PLASTICITY MORE THAN ACTIVITY 211

FIG. 5. Visual response, response to iontophoresis of a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)and response to iontophoresis of N-methyl-D-aspartate (NMDA), measured in a layer II /III cell before injection of MK-801and 30 min after the 2nd injection of MK-801.

abolition of activity in the cortex with TTX was proved to ocular dominance experiments (0.1–0.15 mg/kg) did not.Thus our conclusion is limited; there is a dose of MK-801lead to a complete abolition of the ocular dominance shift

(Reiter et al. 1986). that affects ocular dominance shifts without affecting visualresponses, but we would not expect this to occur with allWe should emphasize that one could almost certainly use

a higher dose of NMDA antagonist and affect both visual doses.Bear et al. (1990) found a substantial loss in selectivityresponses and ocular dominance shifts. The effect of NMDA

antagonists on visual responses decreases with age in layers for orientation and direction, recording cells in animals withAPV infused into the visual cortex. We found a much smallerIV, V, and VI, but there is a distinct effect on visual re-

sponses in layers II and III at all ages (Fox et al. 1989; loss of selectivity. This may have been due to the age of theanimals. Treatment in the animals used by Bear et al. startedMiller et al. 1989; Tsumoto et al. 1987). We agree with all

of these authors that NMDA receptor blockers can decrease at 3–5 wk of age, compared with 30–39 days for our ani-mals. The critical period for direction selectivity ends earliervisual responses, but we found that the dose is critical. We

found that higher doses of MK-801 (0.2 mg/kg) affected than the critical period for ocular dominance changes (Ber-man and Daw 1977; Daw and Wyatt 1976), and so probablyactivity of visual cortex cells, but the doses used in our

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

DAW ET AL.212

FIG. 6. Dose-response curves for iontophoretic injection of NMDA and AMPA on a visual cortex cell before and afterinjection of MK-801. n and – – – : data for response to AMPA with line fitted; / and : data for responses to NMDAand line fitted to them.

does the critical period for orientation changes (Chapman due to the effect of MK-801 in retina, lateral geniculate, orcortex. However, we found that there was little effect ofand Stryker 1993; Kim and Bonhoeffer 1993). Consequently

interventions early in the critical period should affect orienta- MK-801 on activity in cortex, so possible effects on retinaand lateral geniculate are not of much concern.tion, direction, and ocular dominance, whereas interventions

late in the critical period should affect primarily ocular domi- Could we have obtained the reduction in ocular dominanceshift because MK-801 had an overall anesthetic effect? Annance. Alternatively, the difference could have been due to

the difference in drugs (APV vs. MK-801) and routes of overall anesthetic effect could be manifested in variousways. The animals might change their behavior so that theyadministration.

Bear et al. (1990) also found a reduction in response exhibit reduced activity or more sleep. More sleep would,of course, result in more time spent with the eyes closed.quality in their animals infused with APV. Cells with a brisk

and reliable response were given a high response quality, We monitored the dose so that no change in behavior oc-curred, other than a mild ataxia, and lowered the dose imme-whereas cells with a sluggish response were given a low

response quality. We did not assess response quality in the diately after the three occasions on which we noticed somesleepiness. In addition, the signals reaching the visual cortexneurons recorded to construct the ocular dominance histo-

grams because we were primarily concerned with getting as from nonvisual sources, such as the locus coeruleus, raphenuclei, basal forebrain, or dopaminergic nuclei, mightlarge a sample as possible. Our impression, however, was

that cells from both MK-801–treated animals and control change. Any change in signals reaching the visual cortexfrom nonvisual sources would probably have shown up asanimals responded equally vigorously. In addition, we did

have a quantitative measure of the visual response in those a change in spontaneous or visual activity after the injectionsof MK-801, if it were significant.neurons where we assessed the effect of injection of MK-

801 on the visual response; in those neurons there was little One of the questions posed by these results is how is itpossible for MK-801 to affect ocular dominance plasticitydifference between control and treated animals. Thus we did

not see a decline in response quality after treatment with without affecting visual responses? If visual activity is in-structive for ocular dominance plasticity and yet is unaf-MK-801. This also could be due to a difference in the age

of the animals. fected by MK-801, why does ocular dominance plasticitynot proceed normally? One possibility is that MK-801 actsWe found that the ocular dominance shift was reduced in

layer IV as well as in layers II and III. This contrasts with to block a permissive process, for example, by acting on anonvisual input to the cortex. The intralaminar input, forthe NMDA contribution to the visual response at the same

age, which is small in layer IV but substantial in layers II example, is an ascending nonsensory input that activatescortical NMDA receptors (Fox and Armstrong-James 1986)and III ( the NMDA contribution to the visual response is

defined as reduction of the visual response by iontophoresis and was implicated in ocular dominance plasticity (Singerand Rauschecker 1982). Another possibility is that MK-801of APV near the cell body) (Fox et al. 1989).

Our procedure of injecting MK-801 im could affect activ- does affect NMDA receptors engaged by visual inputs butthat the receptors are only blocked by MK-801 when theity in the retina and lateral geniculate nucleus as well as

activity in the visual cortex. If we had found that the activity visual inputs cause particularly sustained or strong levelsof depolarization. It is known that MK-801 blocks NMDAmeasured from single cells in the visual cortex was signifi-

cantly reduced, we would not have known whether this was receptors in an activity-dependent manner (McDonald et al.

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

NMDA RECEPTORS AFFECT PLASTICITY MORE THAN ACTIVITY 213

FIG. 7. A : slope of NMDA dose-responsecurve relative to slope of AMPA dose-re-sponse curve before and after 2 injections ofMK-801 spaced 4 h apart. B : visual responsesand spontaneous activity for cells recordedfrom the same animal over the same periodof time.

1981), so it is conceivable that, at a particular dose of MK- receptors, suggesting that calcium entry is the more crucialproperty (Komatsu 1991; see also Nicoll and Malenka801, heightened levels of visual activity are blocked but

normal visual processing is not. 1995). Moreover, calcium can activate various second ef-fector enzymes such as calcium/calmodulin-dependent pro-How else could long-term MK-801 treatment cause a

change in plasticity without causing much change in the tein kinase II and protein kinase C, which play a role in LTP(Malenka et al. 1989; Malinow et al. 1989) and may playvisual response of cortical cells? Activation of the NMDA

receptor has two effects, depolarization of the neuron and a role in plasticity in the visual cortex. We suggest that ourMK-801 treatment blocked calcium entry enough to affectentry of calcium into the cell (Dingledine 1983). The visual

response is affected primarily by depolarization, whereas plasticity, and the depolarization block was not enough toaffect the response of the cells significantly.plasticity could be affected by both depolarization and/or

calcium entry. Indeed, some forms of long-term potentiation Could MK-801 have a minor effect on depolarizationwhile it has a significant effect on plasticity through calcium(LTP) depend on calcium entry but not activation of NMDA

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

DAW ET AL.214

TABLE 3. Effect of MK-801 on activity of cells in animals used for iontophoresis

Response Response AfterCell Dose, mg/kg Layer Before MK-801 MK-801 Change, %

Effect of first dose

819 Mar D 0.12 II/III 12.3 { 1.9 (12) 14.7 { 3.2 (20) /20806 Apr D 0.12 II/III 13.2 { 1.7 (12) 13.3 { 3.3 (18) /1702 Dec B 0.15 II/III 7.3 { 2.0 (7) 6.7 { 1.2 (5) 08827 Apr D 0.15 III/IV 31.7 { 4.1 (18) 28.1 { 9.2 (18) 011619 Jul B 0.2 IV? 11.3 { 5.4 (13) 7.1 { 1.3 (16) 037622 Jul F 0.2 II/III 9.8 { 5.7 (16) 9.8 { 2.3 (13) 0624 Jul B 0.2 II/III? 12.6 { 1.4 (6) 9.5 { 2.0 (16) 025625 Jul B 0.2 II/III? 14.6 { 1.6 (12) 19.7 { 2.8 (20) /35

Effect of second dose

819 Mar E 0.12 V 20.6 { 3.4 (19) 22.7 { 3.7 (8) /10806 Apr D 0.12 II/III 18.7 { 2.5 (12) 17.6 { 6.3 (16) 06702 Dec C 0.15 II/III 12.8 { 3.2 (19) 20.1 { 6.3 (8) /57827 Apr E 0.15 IV 10.6 { 1.8 (18) 10.9 { 2.6 (18) /3622 Jul F 0.2 II/III 9.8 { 2.3 (13) 0 (12) 0100624 Jul C 0.2 II/III? 7.0 { 0.9 (6) 1.8 { 0.5 (16) 074625 Jul D 0.2 II/III? 24.0 { 3.8 (12) 22.7 { 2.4 (15) 05

Values are means { SD, number of subjects within parentheses.

monocular and direction deprivation in cats. J. Physiol. (Lond.) 265:entry? This is certainly possible. The relevant factor may be249–259, 1977.calcium concentrations in individual spines rather than the

CHAPMAN, B. AND STRYKER, M. P. Development of orientation selectivityoverall activity in the cell as a whole (Zador et al. 1990). in ferret visual cortex and effects of deprivation. J. Neurosci. 13: 5251–It is difficult to detect a change in activity of õ5–10%. A 5262, 1993.

DAW, N. W. Mechanisms of plasticity in the visual cortex. Invest. Ophthal-relatively minor change in activity if this magnitude couldmol. Vis. Sci. 35: 4168–4179, 1994.nevertheless be accompanied by a large change in the

DAW, N. W. AND ARIEL, M. Properties of monocular and directional depri-amount of calcium entering the cell at the dendritic spinevation. J. Neurophysiol. 44: 280–294, 1980.

caused by the nonlinearities in the relationship between de- DAW, N. W., FOX, K., SATO, H., AND CZEPITA, D. Critical period for monoc-polarization and intraspinal calcium concentration. ular deprivation in the cat visual cortex. J. Neurophysiol. 67: 197–202,

1992.In summary, we have shown that an intramuscular doseDAW, N. W. AND WYATT, H. J. Kittens reared in a unidirectional environ-of MK-801 that produces mild ataxia without sleepiness re-

ment: evidence for a critical period. J. Physiol. (Lond.) 257: 155–170,duces ocular dominance plasticity substantially and has a1976.

substantial effect on the responses of cells in the visual cor- DINGLEDINE, R. N-Methyl aspartate activates voltage-dependent calciumtex to NMDA, whereas the visual responses of the cells are conductance in rat hippocampal pyramidal cells. J. Physiol. (Lond.) 343:

385–405, 1983.not affected very much. Because the effect on the visualFOX, K. AND ARMSTRONG-JAMES, M. The role of the anterior intralaminarresponse was minor, we suggest that the effect on plasticity

nuclei and N-methyl-D-aspartate receptors in the generation of sponta-occurred primarily through calcium entering the cell at sub-neous bursts in rat cortical neurones. Exp. Brain Res. 63: 505–518,

threshold levels of depolarization. 1986.FOX, K., SATO, H., AND DAW, N. W. The location and function of NMDA

receptors in cat and kitten visual cortex. J. Neurosci. 9: 2443–2454,We thank J. Peters for help during the first series of experiments.1989.This work was supported by National Institutes of Health Grants RO1

HUBEL, D. H. Tungsten microelectrode for recording from single units.EY-00053 to N. W. Daw and RO1 EY-04050 to B. Gordon.Science 125: 549–550, 1957.Present address: B. Gordon, Institute of Neuroscience, University of Ore-

HUBEL, D. H. AND WIESEL, T. N. Receptive fields, binocular interaction andgon, Eugene, OR 97403; K. D. Fox, School of Molecular and Medicalfunctional architecture in the cat’s visual cortex. J. Physiol. (Lond.) 160:Bioscience, University of Wales, Cardiff CF1 3US, UK; S.N.M. Reid, Jules106–154, 1962.Stein Eye Institute, University of California, Los Angeles, CA 90095; D.

HUBEL, D. H. AND WIESEL, T. N. The period of susceptibility to the physio-Czepita, 1st Dept. Ophthalmology, Pomeranian Medical Academy, 70-111logical effects of unilateral eye closure in kittens. J. Physiol. (Lond.)Szczecin, Poland.206: 419–436, 1970.Address for reprint requests: N. W. Daw, Dept. of Ophthalmology and

KASAMATSU, T., PETTIGREW, J. D., AND ARY, M. Cortical recovery from theVisual Science, New Haven, CT 06520-8061.effect of monocular deprivation: acceleration with norepinephrine and sup-

Received 26 May 1998; accepted in final form 28 September 1998. pression with 6-hydroxydopamine. J. Neurophysiol. 45: 254–266, 1981.KELLY, J. P. AND VAN ESSEN, D. C. Cell structure and function in the visual

cortex of the cat. J. Physiol. (Lond.) 238: 515–547, 1974.REFERENCESKIM, D. S. AND BONHOEFFER, T. Chronic observation of the emergence of

iso-orientation domains in kitten visual cortex. Soc. Neurosci. Abstr. 19:ARMSTRONG-JAMES, M. AND MILLAR, J. Carbon fibre microelectrodes. J.1800, 1993.Neurosci. Methods 1: 279–287, 1979.

KLEINSCHMIDT, A., BEAR, M. F., AND SINGER, W. Blockade of ‘‘NMDA’’BEAR, M. F., KLEINSCHMIDT, A., GU, Q., AND SINGER, W. Disruption ofreceptors disrupts experience-dependent plasticity of kitten striate cortex.experience-dependent synaptic modifications in striate cortex by infusionScience 238: 355–358, 1987.of an NMDA receptor antagonist. J. Neurosci. 10: 909–925, 1990.

BERMAN, N.E.J. AND DAW, N. W. Comparison of the critical period for KOMATSU, Y., NAKAJIMA, S., AND TOYAMA, K. Induction of long-term

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from

NMDA RECEPTORS AFFECT PLASTICITY MORE THAN ACTIVITY 215

potentiation without participation of N-methyl-D-aspartate receptors in RAUSCHECKER, J. P., EGERT, U., AND KOSSEL, A. Effects of NMDA antago-kitten visual cortex. J. Neurophysiol. 65: 20–32, 1991. nists on developmental plasticity in kitten visual cortex. Int. J. Dev.

MACDONALD, J. F., BARTLETT, M. C., MODY, I., PAPAHILL, P., REYNOLDS, Neurosci. 8: 425–435, 1990.J. N., SALTER, M. W., SCHNEIDERMAN, J. H., AND PENNEFATHER, P. S. REITER, H. O., WAITZMAN, D. M., AND STRYKER, M. P. Cortical activityActions of ketamine, phencyclidine and MK-801 on NMDA receptor blockade prevents ocular dominance plasticity in the kitten visual cortex.currents in cultured mouse hippocampal neurons. J. Physiol. (Lond.) Exp. Brain Res. 65: 182–188, 1987.432: 483–508, 1991. SINGER, W. AND RAUSCHECKER, J. P. Central core control of develop-

MALENKA, R. C., KAUER, J. A., PERKEL, D. J., MAUK, M. D., KELLY, P. T., mental plasticity in the kitten visual cortex. II. Electrical activation ofNICOLL, R. A., AND WAXHAM, M. N. An essential role for postsynaptic mesencephalic and diencephalic projections. Exp. Brain Res. 47: 223–calmodulin and protein kinase activity in long-term potentiation. Nature 233, 1982.340: 554–557, 1989. STRYKER, M. P. AND HARRIS, W. A. Binocular impulse blockade prevents

MALINOW, R., SCHULMAN, H., AND TSIEN, R. W. Inhibition of postsynapticthe formation of ocular dominance columns in cat visual cortex. J. Neu-PKC or CaMKII blocks induction but not expression of LTP. Sciencerosci. 6: 2117–2133, 1986.245: 862–866, 1989.

TSUMOTO, T., HAGIHARA, K., SATO, H., AND HATA, Y. NMDA receptorsMILLER, K. D., CHAPMAN, B., AND STRYKER, M. P. Visual responses inin the visual cortex of young kittens are more effective than those ofadult visual cortex depend on N-methyl-D-aspartate receptors. Proc. Natl.adult cats. Nature 327: 513–514, 1987.Acad. Sci. USA 86: 5183–5187, 1989.

WIESEL, T. N. Postnatal development of the visual cortex and the influenceNICOLL, R. A. AND MALENKA, R. C. Contrasting properties of two forms ofof environment. Nature 299: 583–591, 1982.long-term potentiation in the hippocampus. Nature 377: 115–118, 1995.

ZADOR, A., KOCH, C., AND BROWN, T. H. Biophysical model of a HebbianOLSON, C. R. AND FREEMAN, R. D. Profile of the sensitive period for monoc-ular deprivation in kittens. Exp. Brain Res. 39: 17–21, 1980. synapse. Proc. Natl. Acad. Sci. USA 87: 6718–6722, 1990.

J393-8/ 9k30$$ja33 12-29-98 23:09:51 neupa LP-Neurophys

on August 9, 2009

jn.physiology.orgD

ownloaded from