Bidirectional modulation of hippocampal synaptic plasticity by Dopaminergic D4-receptors in the CA1 area of hippocampus

Abstract

Long-term potentiation (LTP) is the persistent increase in the strength of the synapses. However, the neural networks would become saturated if there is only synaptic strenghthening. Synaptic weakening could be facilitated by active processes like long-term depression (LTD). Molecular mechanisms that facilitate the weakening of synapses and thereby stabilize the synapses are also important in learning and memory. Here we show that blockade of dopaminergic D4 receptors (D4R) promoted the formation of late-LTP and transformed early-LTP into late-LTP. This effect was dependent on protein synthesis, activation of NMDA-receptors and CaMKII. We also show that GABA_A-receptor mediated mechanisms are involved in the enhancement of late-LTP. We could show that short-term plasticity and baseline synaptic transmission were unaffected by D4R inhibition. On the other hand, antagonizing D4R prevented both early and late forms of LTD, showing that activation of D4Rs triggered a dual function. Synaptic tagging experiments on LTD showed that D4Rs act as plasticity related proteins rather than the setting of synaptic tags. D4R activation by PD 168077 induced a slow-onset depression that was protein synthesis, NMDAR and CaMKII dependent. The D4 receptors, thus exert a bidirectional modulation of CA1 pyramidal neurons by restricting synaptic strengthening and facilitating synaptic weakening.

Figure 1

D4 receptor inhibition augments LTP. (a) A schematic representation of a transverse hippocampal slice showing the position of electrodes in the CA1 region of hippocampus. S1 represents a stimulation electrode used to stimulate a single neuronal population. ‘Rec’ represents a recording electrode. (b) Induction of late-LTP by a STET protocol resulted in a long-lasting LTP which was stable for 3 h (n = 9, filled circles). Induction of early-LTP using a WTET protocol resulted in weaker form of LTP which was stable for 90 min (n = 8, open circles). (c) When late-LTP was induced in presence of the D4R antagonist, L-745, 870 (50 nM) the fEPSPs increased to a level significantly higher than in control late-LTP (n = 7, filled circles). (d) A similar experiment as in (c) but with the induction of early-LTP instead of late-LTP (n = 8, filled circles) which resulted in the reinforcement of early-LTP to late-LTP. (e) Late-LTP was induced using a STET protocol; 30 min after the induction of late-LTP, L-745, 870 was applied for 60 min. L-745, 870 did not affect the maintenance of late-LTP (n = 6, filled circles). (f) L-745, 870 did not affect basal fEPSPs (n = 7, filled circles). Here after recording a stable baseline, L-745, 870 was applied for 60 min. Representative fEPSPs at −30 (dotted lines), +30 (dashed line) and 180 min (solid line). WTET represents weak tetanisation to induce early-LTP and STET represents strong tetanisation to induce late-LTP. Horizontal rectangular bars indicate the duration of drug application.

Figure 2

Properties of D4R inhibition mediated reinforcement of early-LTP. (a) In order to study the properties of the reinforced early-LTP by D4R inhibition, we used anisomycin, a reversible protein synthesis inhibitor. L-745, 870 was co applied with anisomycin for 60 min and a WTET was applied 30 min after the L-745, 870 application. Reinforcement of early-LTP into late-LTP was prevented by anisomycin (n = 7, filled circles). (b) The same experiment was repeated with the exception that (b) AP5 (c) and KN93 were used instead of anisomycin. The NMDA-receptor antagonist, AP5 blocked the induction of LTP (n = 7, filled circles). (c) The CaMKII inhibitor KN93 reduced the induction and prevented the reinforcement of LTP (n = 7, filled circles). (d) Paired pulse facilitation (PPF) in control (n = 7, filled circles) and L-745 870 (n = 7, open circles) treated slices did not show any significant difference.

Figure 3

The role of GABA_A receptors in D4R action. (a) When picrotoxin, a GABA_A receptor antagonist was coapplied with L-745, 870 during the induction of late-LTP, the enhancement of LTP seen with L-745, 870 (as in Fig. 1c) was attenuated. (n = 6). (b) Similar experiments were repeated like in (a) with the exception that an early form of LTP was induced instead of late-LTP. When early-LTP was induced in presence of L-745, 870 and picrotoxin, it did not induce any significant changes in late-LTP (n = 6, filled circles). (c) To study whether picrotoxin when coapplied with L-745, 870 has any effect on baseline synaptic transmission, both drugs were coapplied for 60 min after recording a stable baseline for 30 min (n = 5, filled circles).

Figure 4

D4 receptors are critical for long-term depression. (a) Late-LTD can be induced by delivering SLFS, and an early-LTD can be induced using WLFS. SLFS induced a late-LTD which is stable for 3 h (filled circles, n = 8), and WLFS induced an LTD which is short lasting (open circles, n = 7). (b) L-745, 870 was applied 30 min before and 30 min after the induction of late-LTD after recording a stable baseline. L-745 870 blocked the late phase of LTD (n = 7, filled circles). (c) Next we studied the role of D4R inhibition during early-LTD. A similar experiment like (b) was repeated but with the induction of early-LTD (n = 6, filled circles). D4R inhibition blocked the maintenance of early-LTD. (d) In order to study whether D4R antagonist effect on LTD is activity dependent, L-745 870 was applied 30 min after the induction of late-LTD for 1 h. The D4R antagonist did not alter the maintenance of LTD which shows that D4R activity is not required for the maintenance phase of late-LTD (n = 6, filled circles). As early-LTD was also affected by D4R inhibition, we studied the role of D4R in synaptic tagging and capture. (e) Shows a schematic representation of a hippocampal slice with two stimulating electrodes in the CA1 area of hippocampal slice. S1 and S2 represent two stimulating electrodes to stimulate two independent synaptic inputs to a single neuronal population from where the recordings were made. ‘rec’ represents a recording electrode used to record the changes in synaptic activity. (f) After recording a stable baseline for 30 min, a SLFS was delivered to input S1. 30 min later L-745 870 was bath applied for 60 min, and 60 min after the induction of late-LTD in S1 another late-LTD was induced in synaptic input S2, but now in the presence of L-745, 870. Thus during the induction of late-LTD in S2 while D4Rs are blocked allowed the formation of synaptic tag and the capture of previously induced proteins transforming a declining fEPSP potentiation into a long-lasting one (n = 6, filled circles). SLFS-strong low frequency stimulation, WLFS-weak low frequency stimulation.

Figure 5

D4R activation induces synaptic depression. (a) The D4R agonist, PD 168077 induced a fEPSP depression that was stable for 3 h (n = 6, filled circles). PD 168077 was applied for 30 min after recording a stable baseline. (b) The same experiment was repeated as (a), but here Anisomycin was applied alone for 10 min before coapplication with PD 168077 for 30 min. Co application of PD 168077 with anisomycin prevented the fEPSP depression (n = 8, filled circles). (c) Here in order to study the role of NMDA-receptor activation, the experiment was done as in (b) but with AP5 instead of anisomycin and AP5 completely blocked the fEPSP depression (n = 6, filled circles). (d) The CaMKII inhibitor KN93 also prevented the fEPSP depression mediated by PD 168077 (n = 7, filled circles). (e) A D4R partial agonist, Ro-10-5824 was applied for 30 min and like PD, it induced a synaptic depression which was stable for 3 h (n = 7, filled circles). (f) The D4R agonist, PD 168077 was applied 5 min after the third tetanization (at 25 min) and was bath applied for 30 min. PD 168077 depotentiated late-LTP (n = 10).