Blockade of NR2B-containing NMDA receptors prevents BDNF enhancement of glutamatergic transmission in hippocampal neurons

Abstract

Application of brain-derived neurotrophic factor (BDNF) to hippocampal neurons has profound effects on glutamatergic synaptic transmission. Both pre- and postsynaptic actions have been identified that depend on the age and type of preparation. To understand the nature of this diversity, we have begun to examine the mechanisms of BDNF action in cultured dissociated embryonic hippocampal neurons. Whole-cell patch-clamp recording during iontophoretic application of glutamate revealed that BDNF doubled the amplitude of induced inward current. Coexposure to BDNF and the NMDA receptor antagonist AP-5 markedly reduced, but did not entirely prevent, the increase in current. Coexposure to BDNF and ifenprodil, an NR2B subunit antagonist, reproduced the response observed with AP-5, suggesting BDNF primarily enhanced activity of NR2B-containing NMDA receptors with a lesser effect on non-NMDA receptors. Protein kinase involvement was confirmed with the broad spectrum inhibitor staurosporine, which prevented the response to BDNF. PKCI19-31 and H-89, selective antagonists of PKC and PKA, had no effect on the response to BDNF, whereas autocamtide-2-related inhibitory peptide, an antagonist of CaM kinase II, reduced response magnitude by 60%. These results demonstrate the predominant role of a specific NMDA receptor subtype in BDNF modulation of hippocampal synaptic transmission.

Figure 1

BDNF enhancement of glutamate-induced current is strongly attenuated in the presence of AP-5. (A) Example traces showing the current prior to, and 20 min after, exposure to BDNF. Under control conditions, the peak amplitude of the current doubled. In this and all subsequent figures, the traces shown are an average of three individual currents. (B) Average time course of the response to BDNF (n = 9). Peak current amplitude was normalized to baseline. Note that in this and subsequent figures the y-axis starts at 75% with 100% indicating no response to BDNF. (Top) Bar shows when BDNF was applied. There was an early small increase in the current followed by a substantially greater response that starts at ∼7 min. (C) Example traces showing the effect of BDNF during bath application of 50 μm AP-5. The current showed a modest increase at 20 min. (D) Average time course of the response to BDNF in AP-5 (n = 4). (Top) Bars show when BDNF and AP-5 were applied. Although the response was much smaller, it proceeded along at the same general rate as under control conditions.

Figure 2

The effect of BDNF on glutamate-activated current can be reduced by an NR2B antagonist. (A) Superimposed example traces showing currents during the baseline period and after 20 min exposure to BDNF. Ifenprodil (3 μm) was present in the bath throughout the recording. BDNF elicited a modest increase in current. (B) Time course of the response to BDNF in the presence of ifenprodil (n = 4). The increase in the current was indistinguishable from that seen when AP-5 was used (cf. with Fig. 1D). (C) Superimposed example traces showing the block produced by AP-5 on the component of current increased by BDNF. (D) Time course of the response to BDNF followed by application of AP-5 at 25 min into the recording. Note that AP-5 almost entirely blocked the current that had been increased by BDNF.

Figure 3

Comparison of glutamate receptor antagonists shows NR2B dependence of BDNF effect. (A) Magnitude of glutamate-activated current after 20–25 min of exposure to BDNF. AP-5 and ifenprodil reduced the response to similar levels. (*) P < 0.005 vs. BDNF alone. (B) Percent block of the glutamate-activated current either prior to, or 25 min following application of BDNF. The far greater block by AP-5 after exposure to BDNF further demonstrates that it is NMDA receptor activity that is upregulated by BDNF. (**) P < 0.001 vs. Pre-BDNF.

Figure 4

Enhancement of glutamateinduced current is prevented by bath application of staurosporine. (A) Superimposed example traces showing the response to glutamate iontophoresis prior to, and 20 min after, application of BDNF in the presence of 1 μm staurosporine. The two traces are essentially indistinguishable. (B) Time course showing that bath application of staurosporine completely prevented the increase in the glutamate-activated current normally seen following exposure to BDNF (n = 4).

Figure 5

Comparison of the actions of PKC, PKA, and Ca/CAM kinase antagonists. Magnitude of glutamate-activated current after 20–25 min exposure to BDNF either alone (same data as in Fig. 3), in vehicle solution, or in the presence of different protein kinase inhibitors. The nonspecific inhibitor staurosporine completely prevented current enhancement by BDNF, (**) P < 0.001 vs. BDNF alone, whereas inhibitors of PKA and PKC were without effect. The CaMKII inhibitor, however, substantially, but not fully, blocked the action of BDNF. (^† P < 0.05 vs. BDNF alone and P < 0.05 vs. staurosporine. Note that the y-axis starts at 75%, with 100% indicating no response (horizontal broken line). Numbers in parentheses indicate number of recordings.