Subunit contribution to NMDA receptor hypofunction and redox sensitivity of hippocampal synaptic transmission during aging

Abstract

We examined the contribution of N-methyl-D-aspartate receptor (NMDAR) subunits in the redox-mediated decline in NMDAR function during aging. GluN2A and GluN2B selective antagonists decreased peak NMDAR currents to a similar extent in young and aged animals, indicating that a shift in diheteromeric GluN2 subunits does not underlie the age-related decrease in the NMDAR synaptic function. Application of dithiothreitol (DTT) in aged animals, increased peak NMDAR synaptic currents, prolonged the decay time, and increased the sensitivity of the synaptic response to the GluN2B antagonist, ifenprodil, indicating that DTT increased the contribution of GluN2B subunits to the synaptic response. The DTT-mediated increase in NMDAR function was inhibited by partial blockade of NMDARs, and this inhibition was rescued by increasing Ca²⁺ concentration in the recording medium. The results indicate that DTT-mediated potentiation requires Ca²⁺ influx through NMDAR activity. Finally, redox regulation of NMDAR function depends on the activity of Ca²⁺/calmodulin-dependent protein kinase II (CaMKII). The results indicate that activity-dependent NMDAR synaptic plasticity is suppressed by redox-mediated inhibition of CaMKII activation during aging. The redox regulation of NMDARs represents a suppression of a metaplasticity mechanism, which can disrupt synaptic plasticity and cognition associated with neurological or psychiatric diseases, and aging.

Keywords: CA1 pyramidal neurons; NMDA receptor current; aging; dithiothreitol; hippocampus; redox state.

Figure 1

Whole-cell patch clamp recording from CA1 hippocampal pyramidal neurons of aged and young animals demonstrating the current-voltage relationship and synaptic decay duration. (A) The current-voltage relationship was recorded from CA1 pyramidal neurons from young (11/4 cells/animals) and aged (9/4 cells/animals) animals. When cells are clamped at positive voltages, the currents are outward and larger currents are observed for young animals. The reversal potential is near 0 mV for both age groups. When cells are clamped at negative voltages, currents are inward and reduced, consistent with Mg²⁺ blockade of the NMDAR channel. Examination of peak amplitude and time to half-decay of the NMDAR EPSC during aging. The cells were voltage clamped at +40 mV. (B) Representative traces evoked by the eight different stimulation intensities and recorded from young (top) and aged animals (bottom). (C) A decrease in the peak NMDAR EPSC was observed across the range of stimulation intensities for CA1 pyramidal cells recorded from aged animals (filled circle, n = 26/14 cells/animals), relative to cells from young animals (open circle, n = 20/9 cells/animals). (D) The mean (±SEM) time for the EPSC to decay to 50% of the peak for the three highest stimulation intensities. The inset shows the time course of the EPSC, evoked by 40 V stimulation, across all CA1 pyramidal cells recorded from young (gray trace, n = 20/9 cells/animals) and aged (dark trace, n = 26/14 cells/animals) animals. For each cell, the response amplitude evoked by 40 V stimulation was normalized to the peak of the response.

Figure 2

The GluN2A and GluN2B selective antagonists attenuated the NMDAR EPSC amplitude to a similar extent in young and aged CA1 pyramidal neurons. For each cell, the peak response was normalized to the 5 min pre-drug baseline. (A) Time course of the decrease in the NMDAR EPSCs recorded from CA1 hippocampal pyramidal neurons 5 min before and 15 min after bath application of ifenprodil (5 µM, solid line) in young (open circle, n = 4/4 cells/animals) and aged (filled circle, n = 6/5 cells/animals) animals. For the control condition (gray circle, n = 7/6 cells/animals, young-aged combined) recordings were obtained before and after application of ethanol vehicle. (B) Bar graph demonstrates percentage decrease in NMDAR EPSCs for young and aged animals following application of ifenprodil or vehicle. Asterisks indicate a significant difference from baseline. The top panel provides representative traces illustrating the NMDAR EPSC at baseline (1) and at the end of a 15 min of ifenprodil application (2) recorded from a young (left) or aged (middle) cell, and for a cell recorded in the vehicle control condition (right). The GluN2A selective antagonist, NVP, attenuated the NMDAR EPSC to a similar extent in young and aged CA1 pyramidal neurons. For each cell, the peak response was normalized to the 5 min pre-drug baseline. (C) Time course of the decrease in the NMDAR EPSCs recorded from CA1 hippocampal pyramidal neurons 5 min before and 15 min after bath application of NVP (0.4 µM, solid line) in young (open circle, n = 4/4 cells/animals) and aged (filled circle, n = 5/5 cells/animals) animals. For the control condition (gray circle, n = 6/6 cells/animals, young-aged combined) recordings were maintained for the same duration in the absence of NVP application. (D) Bar graph demonstrates percentage decrease in NMDA EPSCs during the last 5 min of recording. Asterisks indicate a significant difference from baseline. Representative traces on the top illustrating the NMDAR EPSC at baseline (1) and at the end of a 15 min NVP application (2) recorded from a young (left) or aged (middle) cell, and for a cell in the control condition (right).

Figure 3

Input-output curves examining age-related differences in the peak NMDAR EPSCs under control conditions and in the presence of DTT (0.5 mM). The cells were voltage clamped at +40 mV. (A) Bath application of DTT (filled circle, n = 8/2 cells/animals) failed to increase NMDAR EPSCs for CA1 pyramidal cells recorded from young animals relative to the control condition (open circle, n = 25/11 cells/animals). (B) Across the range of stimulation intensities, DTT (filled circle, n = 7/3 cells/animals) significantly augmented NMDAR EPSCs in CA1 cells recorded from aged animals relative to the control condition (open circle, n = 26/14 cells/animals). (C) DTT increases the time to half-decay of the NMDAR synaptic response. The symbols represent the mean (±SEM) time of NMDAR-mediated EPSC to decay to 50% of the peak under control conditions and in the presence of DTT for the three highest stimulation intensities. The inset (left) shows time course of the EPSC, evoked by 40 V stimulation, across all CA1 pyramidal cells recorded from young animals under the control condition (gray trace, n = 20/9 cells/animals) and in the presence of DTT (black trace, n = 8/2 cells/animals). The inset (right) time course of the EPSC, evoked by 40 V stimulation, across all CA1 pyramidal cells recorded from aged animals under the control condition (gray trace, n = 26/14 cells/animals) and in the presence of DTT (black trace, n = 7/3 cells/animals). For each cell, the response amplitude evoked by 40 V stimulation was normalized to the peak of the response. (D) Increased contribution of the GluN2B subunit to the peak NMDAR EPSC following DTT-induced potentiation of NMDAR EPSCs in slices obtained from aged animals. The cells were voltage clamped at +40 mV and input-output curves of the peak NMDAR EPSCs were generated in presence of DTT (open circle, n = 7/3 cells/animals), DTT+NVP (gray circle, n = 5/2 cells/animals), and DTT+ifenprodil (filled circle, n = 5/3 cells/animals).

Figure 4

NMDAR activity and Ca²⁺ are required for the DTT-induced potentiation of NMDAR synaptic function. (A) Time course of mean (±SEM) NMDAR-fEPSP slope normalized to the baseline (dashed line) for the control condition (open circles), in the presence of ifenprodil or Ro 25-6981 in 2 mM Ca²⁺ recording medium (gray circles), and ifenprodil in 3 mM Ca²⁺ recording medium (filled circles). For clarity, the responses for the GluN2B antagonists (ifenprodil and Ro 25-6981) in 2 mM Ca²⁺ recording medium were combined. The arrow indicates the time of DTT (0.5 mM) application. The insert provides an example of the growth of the NMDAR-mediated fEPSP during baseline (1) and 60 min following application of the DTT (2) under the control condition. (B) Bar graph demonstrates the percent change in NMDAR-mediated fEPSP response during the last 5 min of recording, due to DTT application under the control condition (open bar, n = 31/26 slices/ animals), 2 mM Ca²⁺ + ifenprodil (light gray bar, n = 8/8 slices/animals), 2 mM Ca²⁺ + Ro 25-6981 (gray bar, n = 5/5 slices/animals), and 3 mM Ca²⁺ + ifenprodil (black bar, n = 8/4 slices/animals). (C) Time course of mean (±SEM) NMDAR-fEPSP slope normalized to the baseline (dashed line) for the control condition (open circles), in the presence of NVP in 2 mM Ca²⁺ recording medium (gray circles), and NVP in 3 mM Ca²⁺ recording medium (filled circles). The arrow indicates the time of DTT (0.5 mM) application. (D) Bar graph demonstrates the percent change in NMDAR-mediated fEPSP response during the last 5 min of recording, due to DTT application under the various conditions including control (open bar, n = 31/26 slices/animals), NVP (light gray bar, n = 7/6 slices/animals), and 3 mM Ca²⁺ + NVP (black bar, n= 8/4 slices/animals). For B & D, the asterisks indicate a significant difference relative to control.

Figure 5

The DTT-induced potentiation of the NMDAR-synaptic response is not due to zinc chelation. The panels A-C illustrate the time course for the NMDAR-fEPSP slope; each point represents the mean (±SEM), normalized to the baseline (dashed line). (A) The arrow indicates the time of bath application of DTT (0.5 mM) in presence of ZnCl₂ (1 µM). (B) The arrow indicates the time of bath application of ZX1 (100 µM). (C) The last ten min of NMDAR-fEPSP slope recording in presence of ZX1 was renormalized and DTT was added (arrow). (D) Bar graph represents the mean (+SEM) percent change in NMDAR-mediated fEPSP during the last 5 min of recording, in response to Zn²⁺ plus DTT (open bar, n = 3/3 slices/animals), ZX1 alone (gray bar, n = 11/6 slices/animals,) and ZX1 plus DTT (filled bar, n = 11/6 slices/animals). Asterisks indicate significant potentiation relative to baseline.

Figure 6

Decrease in NMDAR synaptic responses in young animals, under oxidizing conditions, depends on CaMKII activity. (A) Time course of normalized NMDAR-fEPSP slope following application of DTNB (0.5 mM, arrow) in the young animals. Each point represents the mean (±SEM), normalized to the baseline (dashed line), for slices in the control condition (open circles) or following pre-incubation with the CaMKII inhibitor, KN-62 (10 µM, filled circles). DTNB reduced NMDAR synaptic response. Pre-incubation with KN-62 blocked the decrease in the NMDAR response associated with DTNB application. (B) Quantification of the mean percent change in the NMDAR-fEPSP slope during the last 5 min of recording in the presence of vehicle control (open, n = 7/7 slices/animals) and KN-62 (, n = 7/7 slices/animals). Pound sign indicates a significant difference between the two groups. The waveforms represent examples of NMDAR-fEPSPs recorded during baseline (1) and 60 min following application of DTNB (2) in the control condition (left) and following pre-incubation in KN-62 (right).