## Abstract

Focal brain injuries are accompanied by processes of functional reorganization that partially compensate the functional loss. In a previous study, extracellular recordings at the border of a laser-induced lesion in the visual cortex of rats showed an enhanced synaptic plasticity, which was mediated by the activity of NR2B-contaning NMDA-receptors (NMDARs) shedding light on the potential cellular mechanisms underlying this reorganization. Given the potentially important contribution of NMDARs in processes of functional reorganization, in the present study, we used the same lesion model to further investigate lesion-induced changes in function and localization of NMDARs in the vicinity of the lesion. The most important finding was a lesion-mediated functional reexpression of nonpostsynaptic, but according to our data, presynaptic or peri-/extrasynaptic NMDARs (preNMDARs), which were undetectable in age-matched (>P21) sham-operated controls. Notably, preNMDARs were able to boost both spontaneous and evoked synaptic glutamatergic transmission. At the postsynaptic site, we also disclosed an increase in the decay time constant of NMDARs mediated currents, which was accompanied by a decreased NR2A/NR2B ratio, as revealed by Western blot analysis. All together these findings provide new insights into the role of NMDARs activity during processes of functional reorganization following a focal lesion in the cerebral cortex.

## Introduction

Several in vivo as well as in vitro experimental studies suggest that a focal neocortical injury is accompanied by processes of functional reorganization (Jenkins and Merzenich 1987; Brown et al. 2007; Keck et al. 2008; Palagina et al. 2009; Yamahachi et al. 2009), for review see Eysel (2009) that partially compensate the functional loss. Some authors reported a lesion-induced increase in neuroplasticity at the border of the injury in the visual cortex, in vivo (Eysel and Schweigart 1999; Zepeda et al. 2003) suggesting that a cortical lesion might confer a higher rewiring capability to the surrounding cortical networks. Furthermore, some in vitro studies performed in different experimental models of focal cortical lesion showed an enhanced induction of synaptic plasticity at the border of the damaged cortical tissue (Hagemann et al. 1998; Peters et al. 2004; Dohle et al. 2009). In search of the cellular mechanisms potentially underlying this increased plasticity, a previous study from our own laboratory reported an important contribution of NR2B-containing NMDA-receptors (NMDARs) to the lesion-induced increased synaptic plasticity (Huemmeke et al. 2004). These results suggested that changes in expression and/or functionality of NMDARs take place following a cortical lesion, and this could have a strong impact in processes of functional recovery. Nevertheless, it is still unclear how the lesion-induced alteration in the activity of NMDARs influences the function of neuronal networks. The answer to this question could be of fundamental importance to better understand the role of NMDARs in processes of functional recovery postlesion. Thus, in the present study, we evaluated changes in the functional expression of NMDARs and disclosed the impact of these alterations on the strength of synaptic transmission in neurons located in the vicinity of chronically induced focal laser lesions in rat visual cortex. Our main finding is a lesion-induced functional reexpression of nonpostsynaptic, presumably presynaptic, or peri-/extrasynaptic NMDARs (preNMDARs) at the border of the injury, which boosted both, spontaneous and evoked glutamatergic synaptic transmission. Evidence for preNMDARs came from previous electron microscopy studies at cortical (Aoki et al. 1994; DeBiasi et al. 1996; Corlew et al. 2007) and hippocampal synapses (Jourdain et al. 2007) (for review, see Corlew et al. 2008). Functionally, preNMDARs have been shown to regulate spontaneous synaptic transmission in entorhinal (Berretta and Jones 1996; Yang et al. 2006), visual (Sjostrom et al. 2003; Corlew et al. 2007; Li and Han 2007) and somatosensory cortex (Bender et al. 2006; Brasier and Feldman 2008), and in hippocampus (Mameli et al. 2005) as well as action potential–mediated transmitter release in visual (Sjostrom et al. 2003) and somatosensory cortex (Brasier and Feldman 2008). Importantly, under physiological conditions, the expression of preNMDARs in the visual cortex of rats and mice has been shown to be limited to the first 3 postnatal weeks (Corlew et al. 2007; Li and Han 2007). In accordance, the present study failed to detect functional preNMDARs in 4-week-old sham-operated control animals indicating that the reexpression of preNMDARs is an exclusive cellular mechanism taking place in the vicinity of a focally damaged visual cortex. This lesion-induced modulation in the function of NMDARs at glutamatergic synapses provides new insights into the role of glutamate receptors during processes of functional reorganization following a focal injury of the visual cortex.

## Materials and Methods

### Laser Lesion Model

All experiments were carried out in accordance with the guidelines published in the European Communities Council Directive of 24 November 1986 (86/609/EEC). Wistar rats (n = 125) at the age of 21–22 days were anaesthetized by intraperitoneal injection of chloral hydrate (4%, 0.1 mL/10 g). In addition, the local anesthetic Lidocaine (2 mg/kg b.w.; 0.08%) was subcutaneously injected into the skin above the cerebral cortical surface. The animals were fixed in a stereotaxic apparatus, and the skull was exposed and cautiously drilled thin above the visual cortex without touching the dura mater. Multiple, partially overlapping round lesions were made through the translucent wet bone under visual control with a 810-nm infrared diode laser (OcuLight SLX, Iris Medical) attached to a binocular operating microscope (Leica, Germany) to form elongated lesions of 1-mm mediolateral width and 4-mm anteroposterior length, 1-mm lateral to and parallel to the midline, starting anterior of the lambda suture in the visual cortex. After surgery, the skin above the lesion was closed with histoacryl blue glue (Braun, Melsungen, Germany). Age-matched sham-operated animals were treated similarly. However, here the skull was only thin drilled and no laser lesions were induced.

### Slice Preparation

Acute brain slices were prepared from the visual cortex 2–6 days after induction of the lesion. Within this specific time window, we have previously observed an enhanced synaptic plasticity at the border of the lesion (Mittmann and Eysel 2001; Huemmeke et al. 2004). The rats were deeply anaesthetized by ether inhalation and decapitated. Coronal slices (350 μm) containing the visual cortex were prepared from the lesioned hemisphere by use of a vibratome (LEICA, VT-1000-S, Germany). Typically, the vibratome produced 4–5 acute coronal cortical slices located in the region Bregma −5.8 mm to Bregma −7.5 mm (Paxinos and Watson 1986), where the visual cortical areas 18, 17, and 18a extend up to 7–8 mm lateral from the midline to the temporal cortex. The slices were subsequently incubated at room temperature in artificial cerebrospinal fluid (ACSF) containing (in mM): 125 NaCl, 25 NaHCO3, 2.5 KCl, 1.5 MgCl2, 2 CaCl2, 1.25 NaH2PO4, and 25 D-glucose and bubbled with 95% O2 and 5% CO2 (pH = 7.4).

### Histology of the Lesion

For immunohistochemical analysis, some slices were briefly incubated in phosphate-buffered saline (PBS, pH 7.3, 4 °C) and subsequently fixated in 4% paraformaldehyde, diluted in 0.1 M PBS for 24 h at 4 °C. For cryoprotection, the slices were immersed in 30% sucrose (in 0.1 M PBS) for at least 72 h at 4 °C. Coronal sections of 30-μm thickness were cut on a freezing microtome and collected in PBS. Some of the slices were used for standard NISSL and biocytin staining reactions. In some slices, double immunofluorescence staining was performed by use of the primary antibodies “anti-NeuN monoclonal” (1:100; MAB 377, CHEMICON) as neuronal cell marker and “anti-glial fibrillary acidic protein (GFAP) polyclonal” (1:250; Z334, DakoCytomation) as marker of reactive astrocytes. After overnight incubation, sections were incubated in a mixture of biotinylated-horse-anti-mouse (1:200; BA-2001, Vectorstain) and CY2 conjugated goat-anti-rabbit (1:200; 111-225-003, Jackson immunoresearch via Dianova, Hamburg, Germany) antibodies for 90 min. The immunoreactivity was visualized with CY3 conjugate streptavidin (1:500; 016-160-084, Jackson Immunoresearch via Dianova).

### Electrophysiological Recordings

#### Whole-Cell Patch Clamp Recordings

After incubation in oxygenated ACSF at room temperature for at least 1 h, individual slices were transferred to a recording chamber mounted on an upright microscope (Olympus BX50-WI, Olympus, Japan) and bubbled with oxygenated standard ACSF at 31 ± 1 °C. For experiments with nominally zero extracellular [Mg2+], we used a modified ASCF containing (in mM): 125 NaCl, 25 NaHCO3, 2.5 KCl, 0.1 MgCl2, 3.5 CaCl2, 1.25 NaH2PO4, and 25 D-glucose. Electrophysiological recordings were obtained under visual control by use of infrared differential interference contrast microscopy (IR-DIC). Recording pipettes were pulled from borosilicate glass capillaries (GB150F-8P, Science Products, Frankfurt, Germany) to reach resistances of 4–5 MΩ when filled with the intracellular solution containing (in mM): 130 Cesium gluconate, 8 KCl, 10 EGTA, 10 HEPES, 2 MgCl2, 2 CaCl2, 2 ATP-Mg and 0.3 GTP, and 5 QX-314 bromide (N-(2,6-Dimethylphenylcarbamoylmethyl) triethylammonium bromide). The pH was adjusted to pH = 7.3 with CsOH. Signals were acquired with an Axopatch 200B amplifier (Molecular Devices, Sunnyvale, CA), filtered at 2 KHz, and digitized at 5 KHz by use of a Digidata-1200B (Molecular Devices) before they were stored on a PC running Clampex 9 software (Molecular Devices). Access resistance (Rs) and input resistance (Ri) were controlled before and after each experimental sequence by applying a brief depolarizing voltage step. The experiment was discarded, if either Rs or Ri changed by more than 15%.

All recordings were performed from layers 2/3 pyramidal cells at distances of 2.5–4 mm from the border of lesion, corresponding to area 18a of the visual cortex (Paxinos and Watson 1986). Miniature excitatory postsynaptic currents (mEPSCs) were acquired at a holding potential of −80 mV in the presence of bath applied tetrodotoxin (TTX, 0.5 μM) to block fast and transient voltage-gated Na+-channel activity in the cortical network. AMPA-receptor (AMPAR)–mediated mEPSCs were pharmacologically isolated by bath applying the γ-aminobutyric acid (GABA)A receptor antagonist picrotoxin (PTX, 100 μM). In addition, a very low concentration of the competitive AMPAR antagonist DNQX (0.1 μM) was applied to the bath in order to prevent a potential hyperexcitability that could arise from the complete blockade of GABAA-mediated inhibition in the cortical slices.

Synaptically evoked excitatory postsynaptic currents (eEPSCs) were induced by current injection through a glass electrode (resistance 3 ± 1 MΩ) located in cortical layer IV underneath the recorded cell (Fig. 1A). The synaptic stimulation was achieved by delivering 100-μs long square wave pulses through the stimulation electrode. If not otherwise indicated the stimulus intensity was set to evoke a postsynaptic response of about 70 pA to prevent significant differences in the initial EPSC amplitude between the 2 experimental groups (Thomson et al. 1993).

Figure 1.

Histology of the laser-lesioned rat visual cortex. (A) Nissl-stained visual cortex slice of a 26-day-old rat at 3 days postinjury. The cartoon shows the location of the laser-induced lesion (black arrow) and the position of the stimulation and recording electrodes. (B) Magnification of the section of A containing the border area of the lesion C (D) Double immunofluorescence staining for NeuN (red labeling) and GFAP (green labeling). Note the limitation of gliosis reaction to an area close to the lesion center, while NeuN staining showed a normal-like distribution of neurons starting from >200 μm away from the border of the lesion.

Figure 1.

Histology of the laser-lesioned rat visual cortex. (A) Nissl-stained visual cortex slice of a 26-day-old rat at 3 days postinjury. The cartoon shows the location of the laser-induced lesion (black arrow) and the position of the stimulation and recording electrodes. (B) Magnification of the section of A containing the border area of the lesion C (D) Double immunofluorescence staining for NeuN (red labeling) and GFAP (green labeling). Note the limitation of gliosis reaction to an area close to the lesion center, while NeuN staining showed a normal-like distribution of neurons starting from >200 μm away from the border of the lesion.

To calculate the kinetics of postsynaptic NMDARs, NMDAR-mediated currents were pharmacologically isolated by bath application of PTX (100 μM) and DNQX (20 μM), while the neurons were clamped at a holding potential of +40 mV. The mean decay time constant of eEPSCs was analyzed by use of a single standard exponential fitting function according to the following equation: $f(t)=Σi=1nAi*e−t/Ti+C$. Summation of NMDAR-mediated currents was tested during high-frequency synaptic stimulations at 40 Hz with half-maximal stimulation intensity.

In order to disclose a potential activity of nonpostsynaptic, likely preNMDARs, we used a method initially introduced in the entorhinal cortex by Berretta and Jones (1996) and later successfully adapted by others (Yang et al. 2006; Li et al. 2008). This approach is based on the intracellular application of the noncompetitive NMDAR blocker dizocilpine maleate (MK-801, 1 mM) through the recording pipette, thereby blocking all activity from postsynaptically located NMDARs (Fig. 3A). To facilitate the diffusion of MK-801 from the cell body to dendrite, the recorded neurons were depolarized to −10 mV for 10 s per minute over a period of 10 min after reaching whole-cell configuration. The diffusion of MK-801 into the soma and dendrites was verified through recordings of EPSCs at a holding potential of −80 mV and +40 mV. The slow outward component of the signal was completely abolished at a holding potential of +40 mV after a diffusion time of 20 min, indicating that all NMDAR-mediated currents were blocked by MK-801 (Fig. 3B). A previous study using the same method failed to show any action of MK-801 on NMDAR function at other neurons located in the neighborhood of the recorded cell, which indicated an unlikely diffusion or leakage of the intracellularly applied MK-801 into the extracellular medium (Bender et al. 2006). We therefore assumed that the drug exerted its action exclusively on postsynaptic NMDARs.

### Biochemical Fractions and Immunoblot Analysis

Three biochemical fractions were prepared from visual cortices of lesion-treated (n = 18) and sham-operated (n = 18) rat brains according to the method recently described by Yashiro et al. (2005). Briefly, the first “postnuclear supernatant” (PNS) fraction was derived from homogenates, which were centrifuged twice at 1000 × g for 10 min to eliminate the nuclei. From these, PNS fractions crude synaptic pellets could be harvested after centrifugation at 10 000 × g for 20 min followed by suspending in HEPES-buffered sucrose. The resulting pellets were lysed in hypoosmotic buffer (4 mM HEPES, pH 7.4) by use of a homogenizer and mixed for 30 min. To obtain 2nd “lysed synaptosomal membrane” (LSM) fractions the lysates were centrifuged at 25 000 × g for 20 min, and the resulting pellets were suspended in HEPES-buffered sucrose. Next, density centrifugation (150 000 × g for 2 h) was performed on the LSM fractions (150 000 × g for 2 h) with a gradient of 0.8, 1.0, and 1.2 M sucrose in 4 mM HEPES, pH 7.4. The 1.0–1.2 M interface disclosed the synaptic plasma membrane fractions. After centrifugation (150 000 × g for 30 min), the resulting pellets were rotated and suspended in HEPES (50 mM, pH 7.4) for 15 min, centrifuged at 32 000 × g for 20 min, resuspended in 0.5% Triton X-100-containing buffer, rotated for another 15 min, and centrifuged at 200 000 × g for 20 min to finally obtain the third “postsynaptic density” (PSD) fraction. This fraction was diluted in 50 mM HEPES containing 0.2% SDS. The fractions were stored at −80 °C. Since the amount of cortical tissue harvested from each animal was relatively small (1 mm in diameter throughout the 6 cortical layers), and the protocol required different steps of centrifugation to separate the 3 fractions (PNS, LSM, PSD), we pooled together cortical tissue from 6 rats. As a consequence, each experiment (blot) consisted of tissue from 6 sham-operated and 6 lesion-treated animals. Three independent series of such experiments (3 × 6 rats) were performed in each experimental group to receive a valuable statistics. Protein concentrations were measured by use of Coomassie Plus reagent (Pierce, Rockford, IL). To minimize the variability between the preparations, the comparisons were made from fractions run in parallel. For Western blots, 10 μg of each PNS, LSM, and fractions of PSD were resolved by 8% SDS-PAGE and transferred to nitrocellulose membranes (Protran Schleicher & Schuell, PelkinElmer). The following primary antibodies were used anti-NR2A (Santa Cruz, Santa Cruz), anti-NR2B (Santa Cruz), anti-PSD95 (Chemicon, Temecula), and anti-β-tubulin (Chemicon). Protein bands were detected by incubating the membrane with ECL-plus and exposure to X-ray film. The signals were quantified by densitometry with a Chemilimager v5.5 (Alpha Innotech Corporation). The expression level of NR2A and NR2B were then normalized to the expression level of tubulin.

### Statistical Analysis

Parametric Student t test was performed for statistical evaluation of the data. Results are presented as mean ± standard error of the mean. Differences were considered as significant, if P < 0.05.

## Results

### Histology

The present lesion model (see Materials and Methods) led to reproducible focal injuries in the visual cortex of the rats (Fig. 1). The elongated, partially overlapping lesions of 1-mm mediolateral width and 4-mm anteroposterior length were located 1-mm lateral and parallel to the midline, starting anterior to the lambda suture in the visual cortex. The necrotic tissue damage measured 0.9–1.2 mm in mediolateral extent and reached down to cortical layer IV as shown by the Nissl staining (Fig. 1A,B). Immunohistochemistry was performed at 3–8 days postlesion to evaluate the morphological alterations surrounding the lesion. Double-immunofluorescence staining was performed by use of antibodies against the glial marker GFAP and the neuronal marker “NeuN” in the lesion-treated visual cortex. This experiment revealed a strong gliosis reaction (GFAP, green) limited to the close vicinity of the injury (up to 100–200 μm from the border of the lesion) (Fig. 1C,D). Outside this narrow GFAP positive band, the visual cortex showed a control-like population of NeuN labeled neurons (red). This indicated that the present focal injury model induced a spatially and sharply restricted zone of damage in the visual cortex, while the population of neurons located outside this band appeared control-like (>1.5 mm from the border of the lesion). All electrophysiological recordings were performed at a distance of 2–4 mm from the lesion, an area, where the tissue appeared histologically intact. This region of interest was specifically chosen, since previous electrophysiological and calcium imaging data derived from the same lesion model indicated that the highest level of enhanced synaptic plasticity (Dohle et al. 2009) associated with a moderate increase in the intraneuronal resting calcium concentration (Barmashenko et al. 2003) to be present at 2–4 mm distance from the border of the lesion. The visual cortex of sham-operated rats did not show any histological tissue damage.

### Functional Properties of Postsynaptic Glutamate Receptors

Recently, we have reported that there are no changes of the intrinsic membrane properties in pyramidal neurons in layers II/III in the same lesion model (Imbrosci et al. 2010). In the present work, we investigated potential lesion-induced changes in the functional properties of postsynaptic NMDARs. Pharmacologically isolated NMDAR-mediated EPSCs were evoked by a single presynaptic stimulation in layer 4 by setting the stimulus intensity to 1.5–2-fold of the absolute threshold to evoke an EPSC. Interestingly, NMDAR-mediated EPSCs recorded in cells from lesion-treated animals showed a strong increase in the decay time constant (186.5 ± 19.4 ms, 16 cells from 8 rats) when compared with controls (90.9 ± 8.6 ms, 14 cells from 9 rats, P < 0.001) (Fig. 2A,B).

Figure 2.

Changes in the functional properties of NMDARs recorded by evoked EPSCs. (A) Representative traces of evoked NMDAR-mediated EPSCs recorded at a holding potential of +40 mV from a sham-operated rat (gray line) and from a lesion-treated animal (black trace). (B) Summary bar diagram showing a prolonged mean decay time constant of NMDAR-mediated EPSCs in recordings from lesion-treated rats. (C) Representative traces of NMDAR-mediated EPSCs in response to presynaptic stimulation at 40 Hz (dark trace = lesion treated group, gray trace = sham-operated group). Note the stronger summation of EPSCs postlesion. (D) Summary graph showing the mean normalized amplitudes of NMDAR-mediated EPSCs during high-frequency stimulation as shown in C.

Figure 2.

Changes in the functional properties of NMDARs recorded by evoked EPSCs. (A) Representative traces of evoked NMDAR-mediated EPSCs recorded at a holding potential of +40 mV from a sham-operated rat (gray line) and from a lesion-treated animal (black trace). (B) Summary bar diagram showing a prolonged mean decay time constant of NMDAR-mediated EPSCs in recordings from lesion-treated rats. (C) Representative traces of NMDAR-mediated EPSCs in response to presynaptic stimulation at 40 Hz (dark trace = lesion treated group, gray trace = sham-operated group). Note the stronger summation of EPSCs postlesion. (D) Summary graph showing the mean normalized amplitudes of NMDAR-mediated EPSCs during high-frequency stimulation as shown in C.

When synapses were challenged with a high-frequency presynaptic stimulation (HFS, 40 Hz) NMDAR-mediated EPSCs recorded from lesion-treated animals showed a much stronger summation in comparison to sham-operated animals (amplitude of 15th pulse/1st pulse: postlesion: 1.26 ± 0.04, 12 cells from 10 rats; sham: 1.18 ± 0.07, 12 cells from 10 rats, P < 0.05) (Fig. 2C–D). This result could be expected, since the duration of individual EPSCs have been shown to tightly correlate with the temporal summation of NMDAR-mediated currents (Philpot et al. 2001). The prolonged decay time constants might enhance Ca2+-influx into the postsynaptic cell upon activation of NMDARs, thereby promoting synaptic plasticity and other Ca2+-dependent cell functions.

### Experimental Approach to Evaluate the Functional Expression of Nonpostsynaptic NMDARs (preNMDARs)

The enhanced temporal summation of NMDARs postlesion might be also mediated, at least in part, by an increased presynaptic glutamate release. In this regard, there is accumulating evidence for an important role of functional preNMDARs in the modulation of the spontaneous glutamate release (Corlew et al. 2007; Li and Han 2007). Thus, we investigated whether potential lesion-induced alterations in the function and/or expression of these presynaptic receptors could lead to changes in glutamatergic transmission. In order to disclose a potential activity of nonpostsynaptic, likely preNMDARs, we recorded AMPAR-mediated EPSCs at −80 mV in presence of the noncompetitive NMDAR blocker MK-801 into the patch pipette to block all activity of postsynaptically located NMDARs at the recorded neuron (Fig. 3). As a result, all observable mEPSCs were generated by the activity of postsynaptically located AMPARs. An alteration in the frequency of mEPSCs by the subsequent bath application of the specific NMDARs blocker, D-AP5 would therefore suggest the presence of functional nonpostsynaptic preNMDARs in the cortical network.

Figure 3.

Experimental approach to study the function of nonpostsynaptic, likely preNMDARs. (A) Cartoon showing the experimental setup. The recordings were performed in the whole-cell voltage-clamp configuration at a holding potential of −80 mV in presence of intracellular applied dizocilpine maleate (MK801, 1 mM). This will block all activity from NMDARs located at the recorded neuron, while at the same time, the activity of potential preNMDARs is not affected. Under these conditions, postsynaptic mEPSCs are purely mediated by AMPARs. The presence of nonpostsynaptic likely preNMDARs, which could facilitate the release of glutamate, was disclosed by measuring the frequency of mEPSCs, before and after bath application of the specific NMDAR antagonist D-AP5. This would block all remaining nonpostsynaptic NMDARs. (B) The original voltage traces show evoked EPSCs at one representative recorded neuron to verify the intracellular blockage of NMDARs through MK801. Twenty minutes of intracellular perfusion with MK801 (right column of traces) led to a complete blockade of all NMDAR-mediated EPSCs at a holding potentials of +40 mV. Note the appearance of NMDAR-mediated EPSCs in the absence of MK-801 (left column of traces).

Figure 3.

Experimental approach to study the function of nonpostsynaptic, likely preNMDARs. (A) Cartoon showing the experimental setup. The recordings were performed in the whole-cell voltage-clamp configuration at a holding potential of −80 mV in presence of intracellular applied dizocilpine maleate (MK801, 1 mM). This will block all activity from NMDARs located at the recorded neuron, while at the same time, the activity of potential preNMDARs is not affected. Under these conditions, postsynaptic mEPSCs are purely mediated by AMPARs. The presence of nonpostsynaptic likely preNMDARs, which could facilitate the release of glutamate, was disclosed by measuring the frequency of mEPSCs, before and after bath application of the specific NMDAR antagonist D-AP5. This would block all remaining nonpostsynaptic NMDARs. (B) The original voltage traces show evoked EPSCs at one representative recorded neuron to verify the intracellular blockage of NMDARs through MK801. Twenty minutes of intracellular perfusion with MK801 (right column of traces) led to a complete blockade of all NMDAR-mediated EPSCs at a holding potentials of +40 mV. Note the appearance of NMDAR-mediated EPSCs in the absence of MK-801 (left column of traces).

### Evidence for a Functional Expression of PreNMDARs at the Border of the Lesion

As described above, mEPSCs were recorded in the presence of intracellularly applied MK-801 (1 mM) before and after bath application of the NMDAR antagonist D-AP5. In presence of 25-μM D-AP5, a significant (P < 0.001) reduction in the frequency of mEPSCs was observed, but only in neurons from lesion-treated animals (Fig. 4A,B, Table 1). The mean amplitude of the signals was not affected (P > 0.05) by the NMDAR antagonist in both groups (Fig. 4A,C, Table 1), (lesion: 10 cells from 6 rats, sham: 10 cells from 6 rats). A possible interpretation of this experiment is that the lesion induced a functional reexpression of nonpostsynaptic, likely preNMDARs in the neighboring cortical networks, which modulate the release of glutamate from presynaptic terminals. However, this effect might also be explained by a different sensitivity of NMDARs to Mg2+-ions postlesion, which may facilitate the removal of the Mg2+-block. Since we could mimic this effect in sham-operated controls by removing the Mg2+-ions from the extracellular ACSF, we tested this hypothesis by repeating the experiment in presence of nominally zero extracellular magnesium [Mg2+]e. Under these conditions, a similar reduction in frequency (P < 0.001) but not in amplitude of mEPSCs was observed in lesion-treated rats, while frequency and amplitude of mEPSCs were unchanged (P > 0.05) in sham-operated controls (Fig. 4D,E; Table 1), (lesion: 14 cells from 8 rats, sham: 14 cells from 8 rats). These data strongly suggest that the functional expression of preNMDARs is limited to the postlesional visual cortex and that a facilitated glutamate release postlesion is mediated by the activity from nonpostsynaptic NMDARs.

Table 1

Lesion induced reexpression of functional pre-NMDARs

 Rats >p21 ACSF + MK801 ACSF + MK801 + D-AP5 0 [Mg2+]e + MK801 0 [Mg2+]e + MK801 + D-AP5 Sham mEPSC Frequency (Hz) 8.6 ± 0.9 8.1 ± 0.5 10.6 ± 0.8 9.1 ± 0.8 Amplitude (pA) 10.2 ± 0.7 9.9 ± 0.7 11.2 ± 0.7 9.3 ± 0.8 Lesion mEPSC Frequency (Hz) 11.6 ± 0.4 8.9 ± 0.7* 12.2 ± 0.9 8.0 ± 0.8* Amplitude (pA) 11.4 ± 0.8 10.6 ± 0.5 11.4 ± 0.7 10.6 ± 0.5
 Rats >p21 ACSF + MK801 ACSF + MK801 + D-AP5 0 [Mg2+]e + MK801 0 [Mg2+]e + MK801 + D-AP5 Sham mEPSC Frequency (Hz) 8.6 ± 0.9 8.1 ± 0.5 10.6 ± 0.8 9.1 ± 0.8 Amplitude (pA) 10.2 ± 0.7 9.9 ± 0.7 11.2 ± 0.7 9.3 ± 0.8 Lesion mEPSC Frequency (Hz) 11.6 ± 0.4 8.9 ± 0.7* 12.2 ± 0.9 8.0 ± 0.8* Amplitude (pA) 11.4 ± 0.8 10.6 ± 0.5 11.4 ± 0.7 10.6 ± 0.5

*P < 0.05.

Figure 4.

Functional expression of nonpostsynaptic likely preNMDARs at the border of the lesion in visual cortex. (A) Representative traces of AMPAR-mediated mEPSCs recorded in presence of intracellular applied MK-801 (1 mM) as introduced in Figure 3 in neurons from sham-operated animals (gray traces) and lesion-treated animals (black traces) before (top traces) and after (bottom traces) bath application of D-AP5. (B) Summary bar diagram representing the mean frequency of AMPAR-mediated mEPSCs before and after bath application of D-AP5. Note the significant reduction in the mean frequency of mEPSCs following D-AP5 application exclusively in lesion-treated animals. (C) Summary bar diagram representing the mean amplitude of AMPAR-mediated mEPSCs. (D) and (E) The same experiment as in B and C repeated after removing Mg2+ from the bathing solution. Under these conditions, the frequency of the EPSCs was reduced by D-AP5 again only in lesion-treated animals.

Figure 4.

Functional expression of nonpostsynaptic likely preNMDARs at the border of the lesion in visual cortex. (A) Representative traces of AMPAR-mediated mEPSCs recorded in presence of intracellular applied MK-801 (1 mM) as introduced in Figure 3 in neurons from sham-operated animals (gray traces) and lesion-treated animals (black traces) before (top traces) and after (bottom traces) bath application of D-AP5. (B) Summary bar diagram representing the mean frequency of AMPAR-mediated mEPSCs before and after bath application of D-AP5. Note the significant reduction in the mean frequency of mEPSCs following D-AP5 application exclusively in lesion-treated animals. (C) Summary bar diagram representing the mean amplitude of AMPAR-mediated mEPSCs. (D) and (E) The same experiment as in B and C repeated after removing Mg2+ from the bathing solution. Under these conditions, the frequency of the EPSCs was reduced by D-AP5 again only in lesion-treated animals.

### Postnatal Functional Expression of PreNMDARs

In addition to the lesion-induced expression of pre-NMDARs, these receptors seem to be also developmentally regulated: a physiological expression of functional preNMDARs has been reported in the cerebellum (Glitsch and Marty 1999) and in the visual cortex (Li and Han 2007) but only during early postnatal development (before p21) (Corlew et al. 2007). In order to verify these results in our experimental conditions, we repeated the experiment described in Figure 4 in untreated rats at postnatal days 13–17. According to the other studies, and similar to the lesion-treated older rats, neurons from untreated younger animals showed a decrease in frequency of mEPSCs after application of D-AP5 (P < 0.001) (Fig. 5A–C, Table 2). Once again, the amplitude of these signals was not affected (Fig. 5D,E; Table 2), (18 cells from 6 rats). Together with the results from Figure 4, these data suggest that a focal lesion in rat visual cortex induced a reexpression of functional preNMDARs in animals older than p21. This reexpression indicates that the focal lesion in the visual cortex induced a developmental shift backward to a functional earlier age, where, for example, synaptic plasticity is generally more enhanced.

Table 2

Expression of functional pre-NMDARs containing the NR2B-subunit in untreated relatively young (P13–P16) rats

 Rats >p13–17 ACSF + MK801 ACSF + MK801 + D-AP5 mEPSCs Frequency (Hz) 12.6 ± 1.2 10.8 ± 0.9* Amplitude (pA) 10.1 ± 1.5 9.2 ± 1.3
 Rats >p13–17 ACSF + MK801 ACSF + MK801 + D-AP5 mEPSCs Frequency (Hz) 12.6 ± 1.2 10.8 ± 0.9* Amplitude (pA) 10.1 ± 1.5 9.2 ± 1.3
Figure 5.

Neurons recorded from a population of relatively young, untreated rats (p13–p16). Note the reduction in the mEPSCs frequency in presence of D-AP5 similar to the lesion data in Figure 4, but here recorded in nonlesioned, younger visual cortex. (A) Representative traces of AMPAR-mediated mEPSCs recorded in presence of intracellular applied MK-801 (1 mM) as introduced in Figure 3 in neurons from younger (p13–p16), untreated rats before (top, black traces) and after (bottom, gray traces) bath application of D-AP5. (B) Cumulative probability plots of the mEPSC intervals revealed a shift to the right after application of D-AP5. (C) The summary bar diagram revealed a decrease in the mean frequency of mEPSCs in presence of D-AP5. (D, E) The cumulative plots as well as the summary bar plot of the amplitude of mEPSC showed no effects between the 2 groups.

Figure 5.

Neurons recorded from a population of relatively young, untreated rats (p13–p16). Note the reduction in the mEPSCs frequency in presence of D-AP5 similar to the lesion data in Figure 4, but here recorded in nonlesioned, younger visual cortex. (A) Representative traces of AMPAR-mediated mEPSCs recorded in presence of intracellular applied MK-801 (1 mM) as introduced in Figure 3 in neurons from younger (p13–p16), untreated rats before (top, black traces) and after (bottom, gray traces) bath application of D-AP5. (B) Cumulative probability plots of the mEPSC intervals revealed a shift to the right after application of D-AP5. (C) The summary bar diagram revealed a decrease in the mean frequency of mEPSCs in presence of D-AP5. (D, E) The cumulative plots as well as the summary bar plot of the amplitude of mEPSC showed no effects between the 2 groups.

### Postlesional Expression of PreNMDARs Containing the Subunit NR2B

Next, we tested the subunit composition of the preNMDARs, since recent studies disclosed a fundamental role of NR2B-containing preNMDARs in the modulation of glutamate release (Brasier and Feldman 2008; Li et al. 2009; Larsen et al. 2011). Thus, we repeated the experiment of Figure 4 in presence of bath applied Ro25-6981 (1 μM), a specific antagonist of NR2B-containing NMDARs. Similar to the effect of D-AP5, Ro25-6981 reduced the frequency of mEPSCs exclusively in lesion-treated animals (P < 0.001) (Fig. 6A,B, Table 3) with no change in the mEPSCs amplitude in both groups (P > 0.05, Fig. 6A,C, Table 3) (lesion: 7 cells from 6 rats, sham: 8 cells from 6 rats). To test if these NR2B-containing preNMDARs can also modulate action potential-dependent glutamate release, we additionally measured the amplitude of AMPAR-mediated evoked EPSCs. Electrical stimulation was applied to afferent fibers in cortical layer 4 before and after bath application of ifenprodil (10 μM), another specific blocker of NR2B-containing NMDARs. Again, the activity of postsynaptic NMDARs was blocked by intracellular application of MK-801. The application of ifenprodil led to a drastic decrease in the amplitude of these AMPAR-mediated events only in cells from lesion-treated animals, while no change was observed in cells from sham-operated animals (lesion: before ifenprodil 68.1 ± 3.5 pA, after ifenprodil 39.7 ± 3.5 pA, 7 cells from 3 rats; sham: before ifenprodil 70 ± 4.8 pA, after ifenprodil 78.8 ± 9 pA, 7 cells from 3 rats, P < 0.001) (Fig. 6D–F).

Table 3

Lesion induced reexpression of functional pre-NMDARs containing the NR2B subunit

 Rats >p21 ACSF + MK801 ACSF + MK801 + Ro25-6981 Sham mEPSC Frequency (Hz) 10.7 ± 0.7 9.8 ± 0.9 Amplitude (pA) 11.9 ± 1.2 11.0 ± 1.3 Lesion mEPSC Frequency (Hz) 12.6 ± 0.7 9.10 ± 0.6* Amplitude (pA) 11.2 ± 1.1 10.9 ± 1.3
 Rats >p21 ACSF + MK801 ACSF + MK801 + Ro25-6981 Sham mEPSC Frequency (Hz) 10.7 ± 0.7 9.8 ± 0.9 Amplitude (pA) 11.9 ± 1.2 11.0 ± 1.3 Lesion mEPSC Frequency (Hz) 12.6 ± 0.7 9.10 ± 0.6* Amplitude (pA) 11.2 ± 1.1 10.9 ± 1.3

*P < 0.05.

Figure 6.

Functional nonpostsynaptic, likely preNMDARs in the surround of the lesion contained the subunit NR2B. (A) Representative traces of AMPAR-mediated mEPSCs recorded in presence of intracellular applied MK-801 (1 mM) in neurons from sham-operated animals (gray traces) and lesion-treated animals (black traces) before (top traces) and after (bottom traces) bath application of the specific antagonist of NR2B-containing NMDARs Ro-256981. (B) Summary bar diagram representing the mean frequency of AMPAR-mediated mEPSCs before and after bath application of Ro-256981. (C) Summary bar diagram representing the mean amplitude of AMPAR-mediated mEPSCs. (D) The peak amplitude of evoked AMPAR-mediated currents is reduced in presence of another antagonist of NR2B-containing NMDARs (ifenprodil) but only in neurons at the border of the lesion. (E) Summary bar diagram representing the mean amplitude of AMPAR-mediated eEPSCs before and after bath application of ifenprodil in the 2 experimental groups. Note the drastic reduction in the amplitude of AMPAR-mediated responses exclusively in lesion-treated animals.

Figure 6.

Functional nonpostsynaptic, likely preNMDARs in the surround of the lesion contained the subunit NR2B. (A) Representative traces of AMPAR-mediated mEPSCs recorded in presence of intracellular applied MK-801 (1 mM) in neurons from sham-operated animals (gray traces) and lesion-treated animals (black traces) before (top traces) and after (bottom traces) bath application of the specific antagonist of NR2B-containing NMDARs Ro-256981. (B) Summary bar diagram representing the mean frequency of AMPAR-mediated mEPSCs before and after bath application of Ro-256981. (C) Summary bar diagram representing the mean amplitude of AMPAR-mediated mEPSCs. (D) The peak amplitude of evoked AMPAR-mediated currents is reduced in presence of another antagonist of NR2B-containing NMDARs (ifenprodil) but only in neurons at the border of the lesion. (E) Summary bar diagram representing the mean amplitude of AMPAR-mediated eEPSCs before and after bath application of ifenprodil in the 2 experimental groups. Note the drastic reduction in the amplitude of AMPAR-mediated responses exclusively in lesion-treated animals.

### Biochemical Fractionation Disclosed an Enhanced Expression of NMDARs Containing the NR2B Subunit at Nonpostsynaptic Locations in Lesion-Treated Rat Slices

In a methodically independent approach to evaluate the location and expression level of different subunits of NMDARs postlesion, we produced 3 biochemical fractions from the cortical tissue surrounding the lesion and from the homotopic cortical areas in control animals. The PNS fraction contained both cytoplasmic and cell membrane, the LSM fractions contained synaptic and extrasynaptic components of the plasma membrane but no cytosol and finally the PSD fraction contained a fraction of membrane highly enriched in PSD (see Materials and Methods). In particular, the difference between the LSM and PSD fraction could give an estimation of the localization of extrasynaptic NMDARs. Western blot analysis (Fig. 7A) revealed a lesion-induced downregulation in the expression of NR2A-containing NMDARs in all tested fractions (PNS, lesion: 0.015 ± 0.002, sham: 0.04 ± 0.003; LSM, lesion: 0.056 ± 0.01, sham: 0.16 ± 0.01; PSD, lesion: 0.47 ± 0.02, sham: 0.95 ± 0.02, P < 0.001; Fig. 7A,B). In parallel, the expression of NR2B-containing NMDARs in the vicinity of the lesion was found to be enhanced in all the fractions (PNS, lesion: 0.06 ± 0.009, sham: 0.02 ± 0.02, P < 0.001; LSM, lesion: 0.4 ± 0.03, sham: 0.2 ± 0.008, P < 0.001; PSD, lesion: 0.927 ± 0.03, sham: 0.85 ± 0.008, P < 0.05; Fig. 7A,C). This led to a drastic reduction in the ratio between the expression of NR2A- and NR2B-containing NMDARs postlesion (PNS, lesion: 0.25 ± 0.03, sham: 1.9 ± 0.07; LSM, lesion: 0.12 ± 0.006, sham: 0.73 ± 0.04; PSD, lesion: 0.51 ± 0.03, sham: 1.12 ± 0.03, P < 0.001; Fig. 7D). Interestingly, while the expression of NMDARs-NR2A was similarly reduced in all 3 biochemical fractions postlesion, the altered expression of the NR2B-subunit postlesion in the PSD fraction was relatively minor (9.06 ± 4.72%) when compared with the dramatic increase in PNS (200 ± 20.42%) and LSM (100 ± 5.63%) fractions (Fig. 7E). These results supported our electrophysiological data that the lesion led to an expression of NMDARs predominantly containing the NR2B subunit at a mainly nonpostsynaptic but presynaptic or peri-/extrasynaptic location.

Figure 7.

Lesion-induced changes in the expression of the NMDARs subunits NR2A and NR2B in the PNS, LSM, and PSD biochemical fractions. (A) Western blot with antibodies against NR2A, NR2B, PSD-95, and β-tubulin showing the level of expression of these proteins in each biochemical fraction in sham-operated (left) and lesion-treated (right) animals. Note the gradual (from PNS to PSD fraction) enrichment both in NMDAR subunits and in the postsynaptically located protein PSD-95 and the concomitant reduction of the nonsynaptic protein β-tubulin in the samples of both lesion-and sham-treated visual cortex. (B and C) Summary bar diagrams showing the densitometric quantification of the subunit NR2A (B) and NR2B (C) normalized to β-tubulin in the 3 fractions. Note the decreased expression of NR2A and the concomitant increased expression of NR2B in all 3 fractions in the visual cortex postlesion as compared with controls. (D) Bar diagram showing the significant reduced ratio of the NMDAR-subunits NR2A/2B in all fractions postlesion. (E) Bar diagram displaying the relative changes in the level of expression of NR2A and NR2B postlesion. Note the stronger increase of the NR2B subunit in the LSM fraction (containing both, pre- and postsynaptic structures) as compared with the PSD postlesion (containing purely postsynaptically located proteins).

Figure 7.

Lesion-induced changes in the expression of the NMDARs subunits NR2A and NR2B in the PNS, LSM, and PSD biochemical fractions. (A) Western blot with antibodies against NR2A, NR2B, PSD-95, and β-tubulin showing the level of expression of these proteins in each biochemical fraction in sham-operated (left) and lesion-treated (right) animals. Note the gradual (from PNS to PSD fraction) enrichment both in NMDAR subunits and in the postsynaptically located protein PSD-95 and the concomitant reduction of the nonsynaptic protein β-tubulin in the samples of both lesion-and sham-treated visual cortex. (B and C) Summary bar diagrams showing the densitometric quantification of the subunit NR2A (B) and NR2B (C) normalized to β-tubulin in the 3 fractions. Note the decreased expression of NR2A and the concomitant increased expression of NR2B in all 3 fractions in the visual cortex postlesion as compared with controls. (D) Bar diagram showing the significant reduced ratio of the NMDAR-subunits NR2A/2B in all fractions postlesion. (E) Bar diagram displaying the relative changes in the level of expression of NR2A and NR2B postlesion. Note the stronger increase of the NR2B subunit in the LSM fraction (containing both, pre- and postsynaptic structures) as compared with the PSD postlesion (containing purely postsynaptically located proteins).

## Discussion

The present study investigated the effect of a focal laser lesion in the rat visual cortex on the functional properties of NMDARs and its consequences on glutamatergic neurotransmission in the vicinity of the injury. Our interest in the function of NMDARs under this pathological condition mainly relies on the significance of these receptor molecules for neuronal network function and plasticity (for review, see Bear 1996; Nakazawa et al. 2004; Seely 2009). First, we focused our attention on the physiological properties of postsynaptic NMDARs. In a first series of experiments, we observed a strong lesion-induced change in the kinetics of pharmacologically isolated postsynaptic NMDAR-mediated currents. In particular, the decay time constant of the currents was increased in lesion-treated animals (Fig. 2A,B). Interestingly, NMDAR-mediated currents with relatively long decay time constants were also found in the untreated visual cortex at an earlier postnatal age (Monyer et al. 1994) and under experimental conditions of dark rearing (Kirkwood et al. 1996). The latter study used an animal model of experience-dependent plasticity, in which the visual cortex temporally received no regular sensory input through the eyes leading to a deprived development of the visual cortex. The data of these 2 studies are compatible with the hypothesis that the present focal lesion model in visual cortex might switch the developmental wheel functionally backward to reach a status of higher plasticity, a level that is normally characteristic for younger rats. On the cellular level, the increased decay time constant of the NMDAR-mediated currents implicated a prolonged open state of the channel upon binding to glutamate most likely leading to an elevated Ca2+ influx into the postsynaptic neuron. The NMDAR-mediated increase of intracellular Ca2+ concentration could strengthen processes of synaptic plasticity, thereby potentially supporting functional reorganization in the vicinity of the lesion (Eysel 2009). A recent study from our laboratory used the same lesion model and reported a similar but weaker prolongation in NMDAR-mediated currents postlesion (Imbrosci et al. 2010). In that study, the stimulation electrode was placed in layers 2/3 beside the recorded neuron in order to stimulate horizontal fibers located in the supragranular layers. The present study disclosed an even stronger prolongation of NMDAR-mediated EPSCs at synapses of ascending fibers projecting onto pyramidal neurons in layers 2/3. This might indicate that a focal lesion more strongly affected synapses formed by ascending fibers onto neurons in layers 2/3, and this would emphasize the importance of a cortical column-specific functional reorganization. However, such interpretation requires recordings from ascending and horizontal inputs in the same slice, which has not been done so far. In summary, our data together with the previously reported enhanced synaptic plasticity at ascending cortical fibers postlesion (Mittmann and Eysel 2001; Huemmeke et al. 2004) suggests that the surviving ascending fibers at the border of the lesion might facilitate the transport of incoming meaningful visual activity into the partially deafferentiated visual cortex.

### Lesion-Induced Functional Expression of Nonpostsynaptic NMDARs

The major finding of this study was a lesion-induced increase in the frequency of AMPAR-mediated mEPSCs due to the activity of nonpostsynaptic, presumably preNMDARs (Fig. 4A,B). The presence of tonically active nonpostsynaptic NMDARs, which are likely located at presynaptic or peri-/extrasynaptic sites, can boost the glutamate release in the tissue surrounding the injury, thereby facilitating synaptic inputs from ascending fibers. Functional preNMDARs have been already found to modulate glutamatergic synaptic transmission (Li and Han 2007), however the action of these presynaptic auto-receptors in the untreated healthy visual cortex seemed to be limited to the first 3 postnatal weeks, before the onset of the critical period (Corlew et al. 2007). In line with these findings, we never observed a modulatory effect of preNMDARs on glutamate release in sham-operated control animals at the age of 23–29 days, while this effect was clearly visible in untreated younger animals at the age of 13–16 days (Fig. 5A–C). Therefore, the cortical laser lesion induced a reexpression of presumably functional preNMDARs, which are able to modulate presynaptic glutamate release in 4-week-old animals. It should be noted that our data on AMPAR-mediated mEPSCs in presence of intracellular MK801 and D-AP5 did not allow us to specify the exact location of the disclosed nonpostsynaptic NMDARs. While many previous studies using this method suggested a presynaptic location of these NMDARs (Berretta and Jones 1996; Yang et al. 2006; Li and Han 2007), other studies mentioned rather a general nonpostsynaptic (Brasier and Feldman 2008) or peri-/extrasynaptic location (Yashiro et al. 2005). Thus, we conclude that our laser-lesion model led to a functional expression of nonpostsynaptic, likely presynaptic preNMDARs.

From our data, it is not clear, if such lesion-induced reexpression of preNMDARs would also be present in more adult animals older than 35 days, an age beyond the critical period for postnatal plasticity in the visual cortex. Yet, it has been shown in a lithium–pilocarpine epilepsy model in 4- to 6-month-old rats that chronic epileptic conditions led to a reexpression of preNMDARs in these animals (Yang et al. 2006). A similar observation has been made in the visual cortex of adult mice (P 68) following 10 days of visual deprivation (Yashiro et al. 2005). This indicates that the adult visual cortex might also have the potential to reexpress pre-NMADRs following a laser lesion. This modulation of spontaneous glutamate release requires that preNMDARs are tonically activated by an agonist and devoid of the voltage sensitive Mg2+ block. Thus, we tested whether this change in NMDAR function was due to a different sensitivity to Mg2+-ions and/or to a lesion-induced depolarization of presynaptic terminals, which would facilitate the removal of the Mg2+-block. If the pre-NMDAR-dependent modulation of glutamate release postlesion was due to a reduction in the voltage-sensitive Mg2+ block, we could expect to mimic this effect in sham-operated controls by removing the Mg2+-ions from the extracellular ACSF. We tested this hypothesis by repeating the experiments in absence of extracellular Mg2+ [Mg2+]e. However, even under these conditions we exclusively observed a nonpostsynaptic, likely preNMDAR-dependent increase of mEPSCs in the lesion-treated group, not in sham-operated controls (Fig. 4D). This result strongly indicated that a reduced Mg2+ block of preNMDRs cannot account for the observed effects.

### NR2B-Containing Nonpostsynaptic, Likely PreNMDARs Contribute to the Glutamatergic Transmission Postlesion

Recently, one type of modulation of AMPAR-mediated synaptic transmission was linked to activity from preNMDARs that contained the subunit NR2B (Brasier and Feldman 2008; Li et al. 2009; Larsen et al. 2011). In line with these observations, we observed a lesion-induced increase in the frequency of AMPAR-mediated mEPSCs, which was mediated by preNMDARs containing the subunit NR2B, as shown by its sensitivity to application of the specific NR2B subunit containing NMDAR antagonist, Ro-256981 (Fig. 6A,B). In another experimental approach, we tested, whether a reexpression of preNMDARs containing the NR2B-subunit would also affect the strength of evoked excitatory currents (eEPSCs) at the border of the lesion. We recorded synaptically evoked AMPAR-mediated currents (eEPSCs) before and after the application of ifenprodil (another specific blocker of NR2B-containing NMDARs). Here, the significant reduction in the amplitude of evoked AMPAR-mediated responses in presence of ifenprodil, exclusively visible in lesion-treated rats, was a robust indication of a likely NR2B-containing preNMDARs-dependent reinforcement of glutamatergic synaptic transmission postlesion (Fig. 6D–F). Although we expected a reduction in the amplitude of eEPSCs after application of ifenprodil postlesion, the rather drastic decrease of the signal amplitude was surprising. This might be explained by a cumulative effect. The eEPSCs were obtained with extracellular stimulation, which will inevitably activate several presynaptic terminals. It is possible that preNMDARs exert a relatively weak modulation of glutamate release on each presynaptic terminal. Nevertheless, the small effect on each terminal will be summative and produce a larger effect when several terminals are activated together. Interestingly, the physiological function of preNMDAR (during the first 3 postnatal weeks) is not restricted to a facilitation of glutamate release. In the visual and somatosensory cortex, preNMDARs were associated with the induction of one type of long-term synaptic plasticity, called spike timing–dependent long-term depression, which is expressed before the onset of the critical period (Sjostrom et al. 2003; Bender et al. 2006). This form of plasticity is believed to play an important role in the development of visual cortical circuits and of their receptive field properties (Dan and Poo 2006). Therefore, we speculate that a lesion-induced reexpression of functional preNMDARs might offer the surviving cortical networks at the border of the injury the capability to undergo a functional reorganization. Such functional reorganization has been described following different cortical injury models (Jenkins and Merzenich 1987; Eysel and Schweigart 1999; Zepeda et al. 2003).

### Western Blot Analysis on the Subunit Composition of NMDARs Postlesion

Analysis of enriched biochemical fractions disclosed a lesion-induced change in the subunit composition of post and nonpostsynaptic NMDARs. This independent methodical approach was useful to determine, whether a lesion-induced change in expression or localization of the NMDAR subunits correlated with alterations in the function of NMDARs revealed by our electrophysiological data. It is well documented that NR2B-containing NMDARs have a much longer decay time constant as compared with NR2A-containing NMDARs (for review, see Cull-Candy and Leszkiewicz 2004). In this regard, the Western blots of the fraction of the enriched PSD fraction disclosed a strong reduction in the expression of the NR2A subunit plus a slight increase in the expression of the NR2B-subunit postlesion (Fig. 7A–C), which strongly indicate that this altered receptor subunit composition at the PSD might be largely responsible for the altered kinetics of the NMDAR-mediated responses. These findings are in good agreement with a previous study from our laboratory, in which we used the same lesion model and observed a reduced ratio of NR2A versus NR2B subunits of NMDARs on the mRNA level postlesion (Rumpel et al. 2000). While a relatively low ratio of NR2A/NR2B is typical for the early postnatal age of the animals, it gradually increases during further development (Monyer et al. 1994; Sheng et al. 1994). Together, these data strengthen our hypothesis of a lesion-induced rejuvenation of the surrounding cortical tissue, which might be beneficial for the functional reorganization postinjury. Another interesting result of the biochemical analyses was the net increase in the expression of the NR2B subunit in the fraction enriched in the plasma membrane (LSM fraction) (Fig. 7A,C). This enhancement was even stronger when compared with the slight increase observed in the PSD fraction (Fig. 7A,E) indicating that the NR2B-containing NMDARs overexpressed after the injury may be preferentially expressed at an extrasynaptic/presynaptic location. Finally, the even more strongly expressed NR2B subunit in the PNS fraction postlesion (Fig. 7A,E) indicates that part of the oversynthetized NR2B subunits could be still retained in the cytosol. On the basis of all these findings, we conclude that a lesion-induced reexpression of functional NR2B-containing NMDARs at nonpostsynaptic sites might play an important role in the modulation of the glutamatergic transmission in the vicinity of the injury. However, additional factors should be considered as well. For example, a potential increase in the extracellular concentration of glutamate, as observed in a rat model of fluid percussion injury (Faden et al. 1989), could additionally promote or boost a tonic activation of potential presynaptic NMDARs. So far it is still not fully understood, how the above reported alterations in glutamatergic neurotransmission mediate a functional neocortical recovery postlesion. Enhanced glutamatergic transmission and prolonged postsynaptic NMDAR-mediated currents might also be deleterious in a structurally relatively developed mature cortical network. In particular, activity from NMDARs was classically discussed to contribute to processes of secondary brain damage following traumatic as well as ischemic brain injuries by promoting excitotoxic neuronal death (Simon et al. 1984; Faden et al. 1989). However, more recent studies pointed out that activity from NMDARs can promote both, neuronal health and death. In particular, there is emerging evidence that NMDARs are able to activate prosurviving or prodeath intracellular signaling pathway dependent on their cellular location. Here, the activity of synaptically located NMDARs seem to be neuroprotective, while extrasynaptic NMDARs may promote cell death (for review, see Hardingham and Bading 2010). In light of these findings, the preferentially extrasynaptic localization of the overexpressed NR2B-containing NMDARs postlesion suggested by our biochemical analyses (Fig. 7E) should raise some concern about the potential involvement of these extrasynaptic receptors in processes of neuronal damage. In this context, it should be noted that the present lesion-model failed to show an increased cell death in the vicinity of the injury between lesion-treated rats and sham-operated rats (Dohle et al. 2009; Imbrosci et al. 2010). Future studies should investigate in more detail, whether the functional alterations presented here might have a significant impact on the neuronal survival of specific subtypes of visual cortical neurons after such brain injuries. The prolonged activation of postsynaptic NMDARs upon binding to glutamate revealed by the increased decay time constant of NMDAR-mediated currents postlesion (Fig. 2A,B) should have a double-beneficial role because it is supposed to facilitate plastic processes and promote neuroprotection. Most importantly, the reported modifications in the function of NMDARs in a direction to a more juvenile developmental state seem to characterize a visual cortex that is regaining its capability to undergo different forms of synaptic plasticity. Since plasticity on the synaptic level has been shown to play a central role in the plasticity of cortical maps (Keck et al. 2008), we suggest that these changes could strongly contribute to the functional reorganization of the cortical networks surrounding the lesion.

## Funding

DFG (SFB 509: TP C4 to U.E. and T.M.; grant to T.M.: MI 432-1).

We thank Petra Küsener for excellent technical assistance. Conflict of Interest : None declared.

## References

Aoki
C
Venkatesan
C
Go
C-G
Mong
JA
Dawson
TM
Cellular and subcellular localization of NMDA-R1 subunit immunoreactivity in the visual cortex of adult and neonatal rats
J Neurosci
,
1994
, vol.
14
(pg.
5202
-
5222
)
Barmashenko
G
Eysel
UT
Mittmann
T
Changes in intracellular calcium transients and LTP in the surround of visual cortex lesions in rats
Brain Res
,
2003
, vol.
990
(pg.
120
-
128
)
Bear
MF
A synaptic basis for memory storage in the cerebral cortex
Proc Nat Acad Sci U S A
,
1996
, vol.
93
(pg.
13453
-
13459
)
Bender
VA
Bender
KJ
Brasier
DJ
Feldman
DE
Two coincidence detectors for spike timing-dependent plasticity in somatosensory cortex
J Neurosci
,
2006
, vol.
26
(pg.
4166
-
4177
)
Berretta
N
Jones
RS
Tonic facilitation of glutamate release by presynaptic N-methyl-D-aspartate autoreceptors in the entorhinal cortex
Neuroscience
,
1996
, vol.
75
(pg.
339
-
344
)
Brasier
DJ
Feldman
DE
Synapse-specific expression of functional presynaptic NMDA receptors in rat somatosensory cortex
J Neurosci
,
2008
, vol.
28
(pg.
2199
-
2211
)
Brown
CE
Li
P
Boyd
JD
Delaney
KR
Murphy
TH
Extensive turnover of dendritic spines and vascular remodeling in cortical tissues recovering from stroke
J Neurosci
,
2007
, vol.
27
(pg.
4101
-
4109
)
Corlew
R
Brasier
DJ
Feldman
DE
Philpot
BD
Presynaptic NMDA receptors: newly appreciated roles in cortical synaptic function and plasticity
Neuroscientist
,
2008
, vol.
14
(pg.
609
-
625
)
Corlew
R
Wang
Y
Ghermazien
H
Erisir
A
Philpot
BD
Developmental switch in the contribution of presynaptic and postsynaptic NMDA receptors to long-term depression
J Neurosci
,
2007
, vol.
27
(pg.
9835
-
9845
)
Cull-Candy
SG
Leszkiewicz
DN
Role of distinct NMDA receptor subtypes at central synapses
Sci STKE
,
2004
, vol.
2004
pg.
re16

Dan
Y
Poo
MM
Spike timing-dependent plasticity: from synapse to perception
Physiol Rev
,
2006
, vol.
86
(pg.
1033
-
1048
)
DeBiasi
S
Minelli
A
Melone
M
Conti
F
Presynaptic NMDA receptors in the neocortex are both auto- and heteroreceptors
Neuroreport
,
1996
, vol.
7
(pg.
2773
-
2776
)
Dohle
CI
Eysel
UT
Mittmann
T
Spatial distribution of long-term potentiation in the surround of visual cortex lesions in vitro
Exp Brain Res
,
2009
, vol.
199
(pg.
423
-
433.
)
Eysel
UT
Squire Larry
R
Encyclopedia of neuroscience. Waltham (MA):
,
2009
Elsevier
(pg.
141
-
147
)
Eysel
UT
Schweigart
G
Increased receptive field size in the surround of chronic lesions in the adult cat visual cortex
Cereb Cortex
,
1999
, vol.
9
(pg.
101
-
109
)
AI
Demediuk
P
Panter
SS
Vink
R
The role of excitatory amino acids and NMDA receptors in traumatic brain injury
Science
,
1989
, vol.
244
(pg.
798
-
800
)
Glitsch
M
Marty
A
Presynaptic effects of NMDA in cerebellar Purkinje cells and interneurons
J Neurosci
,
1999
, vol.
19
(pg.
511
-
519
)
Hagemann
G
Redecker
C
Neumann-Haefelin
T
Freund
HJ
Witte
OW
Increased long-term potentiation in the surround of experimentally induced focal cortical infarction
Ann Neurol
,
1998
, vol.
44
(pg.
255
-
258
)
Hardingham
GE
H
Synaptic versus extrasynaptic NMDA receptor signalling: implications for neurodegenerative disorders
Nat Rev Neurosci
,
2010
, vol.
11
(pg.
682
-
696
)
Huemmeke
M
Eysel
UT
Mittmann
T
Lesion-induced enhancement of LTP in rat visual cortex is mediated by NMDA receptors containing the NR2B subunit
J Physiol
,
2004
, vol.
559
(pg.
875
-
882
)
Imbrosci
B
Eysel
UT
Mittmann
T
Metaplasticity of horizontal connections in the vicinity of focal laser lesions in rat visual cortex
J Physiol.
,
2010
, vol.
588
(pg.
4695
-
4703
)
Jenkins
WM
Merzenich
MM
Seil
FJ
Herbert
E
Carlson
BM
Reorganization of neocortical representations after brain injury: a neurophysiological model of the basis of recovery from stroke
Progress in brain research
,
1987
New York
Elsevier Science Publishers B. V., Biomedical Division
(pg.
249
-
266
)
Jourdain
P
Bergersen
LH
Bhaukaurally
K
Bezzi
P
Santello
M
Domercq
M
Matute
C
Tonello
F
Gundersen
V
Volterra
A
Glutamate exocytosis from astrocytes controls synaptic strength
Nat Neurosci
,
2007
, vol.
10
(pg.
331
-
339
)
Keck
T
Mrsic-Flogel
TD
Vaz
AM
Eysel
UT
Bonhoeffer
T
Hubener
M
Massive restructuring of neuronal circuits during functional reorganization of adult visual cortex
Nat Neurosci
,
2008
, vol.
11
(pg.
1162
-
1167
)
Kirkwood
A
Rioult
MC
Bear
MF
Experience-dependent modification of synaptic plasticity in visual cortex
Nature
,
1996
, vol.
381
(pg.
526
-
528
)
Larsen
RS
Corlew
RJ
Henson
MA
Roberts
AC
Mishina
M
Watanabe
M
Lipton
SA
Nakanishi
N
Perez-Otano
I
Weinberg
RJ
, et al.  .
NR3A-containing NMDARs promote neurotransmitter release and spike timing-dependent plasticity
Nat Neurosci
,
2011
, vol.
14
(pg.
338
-
344
)
Li
YH
Han
TZ
Glycine binding sites of presynaptic NMDA receptors may tonically regulate glutamate release in the rat visual cortex
J Neurophysiol
,
2007
, vol.
97
(pg.
817
-
823
)
Li
YH
Han
TZ
Meng
K
Tonic facilitation of glutamate release by glycine binding sites on presynaptic NR2B-containing NMDA autoreceptors in the rat visual cortex
Neurosci Lett
,
2008
, vol.
432
(pg.
212
-
216
)
Li
YH
Wang
J
Zhang
G
Presynaptic NR2B-containing NMDA autoreceptors mediate gluta-matergic synaptic transmission in the rat visual cortex
Curr Neurovasc Res
,
2009
, vol.
6
(pg.
104
-
109
)
Mameli
M
Carta
M
Partridge
LD
Valenzuela
CF
Neurosteroid-induced plasticity of immature synapses via retrograde modulation of presynaptic NMDA receptors
J Neurosci
,
2005
, vol.
25
(pg.
2285
-
2294
)
Mittmann
T
Eysel
UT
Increased synaptic plasticity in the surround of visual cortex lesions in rats
Neuroreport
,
2001
, vol.
12
(pg.
3341
-
3347
)
Monyer
H
Burnashev
N
Laurie
DJ
Sakmann
B
Seeburg
PH
Developmental and regional expression in the rat brain and functional properties of four NMDA receptors
Neuron
,
1994
, vol.
12
(pg.
529
-
540
)
Nakazawa
K
McHugh
TJ
Wilson
MA
Tonegawa
S
NMDA receptors, place cells and hippocampal spatial memory
Nat Rev Neurosci
,
2004
, vol.
5
(pg.
361
-
372
)
Palagina
G
Eysel
UT
Jancke
D
Strengthening of lateral activation in adult rat visual cortex after retinal lesions captured with voltage-sensitive dye imaging in vivo
Proc Natl Acad Sci U S A
,
2009
, vol.
106
(pg.
8743
-
8747
)
Paxinos
G
Watson
C
The rat brain in stereotaxic coordinates
,
1986
London
Peters
O
Redecker
C
Hagemann
G
Bruehl
C
Luhmann
HJ
Witte
OW
Impaired synaptic plasticity in the surround of perinatally acquired dysplasia in rat cerebral cortex
Cereb Cortex
,
2004
, vol.
14
(pg.
1081
-
1087
)
Philpot
BD
Sekhar
AK
Shouval
HZ
Bear
MF
Visual experience and deprivation bidirectionally modify the composition and function of NMDA receptors in visual cortex
Neuron
,
2001
, vol.
29
(pg.
157
-
169
)
Rumpel
S
Hoffmann
H
Hatt
H
Gottmann
K
Mittmann
T
Eysel
UT
Lesion-induced changes in NMDA receptor subunit mRNA expression in rat visual cortex
Neuroreport
,
2000
, vol.
11
(pg.
4021
-
4025
)
Seely
RJ
Squire Larry
R
Nutrient sensing: cellular metabolism
Encyclopedia of neuroscience. Waltham (MA):
,
2009
Elsevier
(pg.
1303
-
1306
)
Sheng
M
Cummings
J
Roldan
LA
Jan
YN
Jan
LY
Changing subunit composition of heteromeric NMDA receptors during development of rat cortex
Nature
,
1994
, vol.
368
(pg.
144
-
147
)
Simon
RP
Griffiths
T
Evans
MC
Swan
JH
Meldrum
BS
Calcium overload in selectively vulnerable neurons of the hippocampus during and after ischemia: an electron microscopic study in the rat
J Cereb Blood Flow Metab
,
1984
, vol.
4
(pg.
350
-
361
)
Sjostrom
PJ
Turrigiano
GG
Nelson
SB
Neocortical LTD via coincident activation of presynaptic NMDA and cannabinoid receptors
Neuron
,
2003
, vol.
39
(pg.
641
-
654
)
Thomson
AM
Deuchars
J
West
DC
Large, deep layer pyramid-pyramid single axon EPSPs in slices of rat motor cortex display paired pulse and frequency-dependent depression, mediated presynaptically and self-facilitation, mediated postsynaptically
J Neurophysiol
,
1993
, vol.
70
(pg.
2354
-
2369
)
Yamahachi
H
Marik
SA
McManus
JN
Denk
W
Gilbert
CD
Rapid axonal sprouting and pruning accompany functional reorganization in primary visual cortex
Neuron
,
2009
, vol.
64
(pg.
719
-
729
)
Yang
J
Woodhall
GL
Jones
RS
Tonic facilitation of glutamate release by presynaptic NR2B-containing NMDA receptors is increased in the entorhinal cortex of chronically epileptic rats
J Neurosci
,
2006
, vol.
26
(pg.
406
-
410
)
Yashiro
K
Corlew
R
Philpot
BD
Visual deprivation modifies both presynaptic glutamate release and the composition of perisynaptic/extrasynaptic NMDA receptors in adult visual cortex
J Neurosci
,
2005
, vol.
25
(pg.
11684
-
11692
)
Zepeda
A
Vaca
L
Arias
C
Sengpiel
F
Reorganization of visual cortical maps after focal ischemic lesions
J Cereb Blood Flow Metab
,
2003
, vol.
23
(pg.
811
-
820
)