NMDA receptor-dependent long-term potentiation comprises a family of temporally overlapping forms of synaptic plasticity that are induced by different patterns of stimulation

Abstract

N-methyl-d-aspartate receptor (NMDAR)-dependent long-term potentiation (LTP) is extensively studied since it is believed to use the same molecular mechanisms that are required for many forms of learning and memory. Unfortunately, many controversies exist, not least the seemingly simple issue concerning the locus of expression of LTP. Here, we review our recent work and some of the extensive literature on this topic and present new data that collectively suggest that LTP can be explained, during its first few hours, by the coexistence of at least three mechanistically distinct processes that are all triggered by the synaptic activation of NMDARs.

Keywords: AMPA receptor; NMDA receptor; hippocampus; learning and memory; long-term potentiation; short-term potentiation.

Figure 1.

Different NMDAR subtypes mediate the induction of LTP and two forms of STP. (a) Pooled data to show LTP in response to TBS (four pulses at 100 Hz repeated 10 times at 5 Hz) under standard conditions (open circles) and with a 30 min pause in stimulation (filled circles) starting after the first four synaptic responses had been obtained (to assess the maximal level of NMDAR-dependent potentiation). Representative traces were obtained at the times indicated by the corresponding coloured numbers. (b(i)) Responses to a train of seven stimuli delivered at 12.5 Hz. (b(ii)) The graph plots the slope of each response normalized to the first response in the train to show depression in facilitation during either STP (red) or stored STP (green) but not LTP (blue) when compared with facilitation in the baseline (black). (c) Effects of NVP (0.1 µM) or UBP (10 µM) on STP and LTP. Note that UBP selectively blocks STP(2) whereas NVP selectively blocks LTP and STP(1). (d) The graphs plot the IC₅₀ for inhibition of LTP (d(i)) or STP(2) (d(ii)) versus the IC₅₀ for inhibition of the NMDAR-EPSC. Note the excellent correlation for LTP but lack of correlation for STP(2). Insets indicate the antagonists, shown in descending rank order of potency for inhibition of the form of synaptic plasticity studied. Adapted (with permission) from [15] (a,c and d) and [16] (b), respectively.

Figure 2.

A single HFS can induce a long-lasting PKA-independent form of LTP. (a) An example of LTP, induced by a single HFS (100 Hz, 1 s, test intensity) in the presence of a PKA antagonist (Rp-cAMPS; 100 µM). The traces were obtained at the times indicated by (a) and (b). (b) In interleaved experiments, the same concentration of Rp-cAMPS was able to fully block the induction of NMDAR-independent LTP in the mossy fibre (MF) pathway. EPSP, excitatory postsynaptic potential. (Adapted from [50] with permission.)

Figure 3.

The timing of the induction trigger determines the type of LTP induced. (a) LTP induced by spaced HFS (arrows: three trains, 100 Hz, 1 s, test intensity delivered with an inter-train interval of 10 min). The graph plots pooled data for five two-input experiments (filled symbols, test input; open symbols, control input). The inset shows superimposed traces (averages of four successive sweeps; scale bars, 0.5 mV/20 ms in this and the following figure) of baseline and LTP for a typical experiment obtained at the times indicated by (a) and (b). (b) LTP induced by compressed HFS (n = 6). The protocol was identical to that in (a) except that the inter-train interval was 10 s. (c) KT 5720 (1 µM) inhibits a component of spaced LTP (n = 5). The test and control inputs are shown as filled and open red symbols, respectively, and the control LTP is replotted from (a) for ease of comparison. (d) KT 5720 (n = 7, filled red) has no effect on compressed LTP, which is shown replotted from (b). (e) Anisomycin (20 µM) inhibits a component of spaced LTP (n = 5). The test and control inputs are shown as filled and open green symbols, respectively, and the control LTP is replotted from (a). (f) Anisomycin (filled green symbols, n = 7) has no effect on compressed LTP (replotted from b). Data for spaced LTP were quantified 2 h after the delivery of HFS and the levels of LTP were 175±12% of baseline, 127±8% (p<0.01) and 124 ± 9% (p<0.01) for the vehicle, PKA and anisomycin experiments, respectively. For compressed LTP data were quantified at both 2 and 5 h after HFS. The corresponding values were, after 2 h: 153±4%, 161±6% and 157±8% (p > 0.05 both cases; Students t-test) and after 5 h: 147 ± 4%, 136 ± 12 and 135 ± 10% (p > 0.05 both cases; Students t-test). Experiments, which were interleaved in a randomized manner, were performed as described previously [56] with data capture and analysis performed using WinLTP [57]. f-EPSP, field-EPSP.

Figure 4.

Paired-pulse analysis of LTP. (a) LTP induced by spaced HFS (n = 6). A plot of paired-pulse ratio (normalized to baseline) for the test (a(i), filled circles) and control (a(ii), open circles) inputs for these experiments. (b) LTP induced by compressed HFS (n = 5). A plot of paired-pulse ratio (normalized to baseline) for the test (b(i), filled circles) and control (b(ii), open circles) inputs for these experiments. There is a reduction in PPF during STP but no significant change thereafter in both sets of experiments. Experiments were performed as described in figure 3, except that paired stimuli (inter-stimulus interval of 50 ms) were delivered throughout the experiment to follow the time course of PPF.

Figure 5.

Schematic of different forms of NMDAR-dependent LTP. We suggest that there are multiple forms of LTP that differ in their expression mechanisms. LTPa is characterized by an increase in P(r). It can account for HFS-induced STP (or at least one major component of STP) and possibly for some other forms of LTP. LTPb is characterized by a change in AMPAR function; potentially both as an alteration in their single channel conductance properties (γ) and in the number of the receptors. LTPc may be due to synaptic growth, with changes in both the number of release sites (potentially associated with an increase in P(r)) and the number of AMPARs. LTPb corresponds to the PKA and protein synthesis-independent form of LTP (commonly referred to as e-LTP or LTP1 in [10]). LTPc corresponds to the PKA and protein synthesis-dependent form of LTP (commonly referred to as l-LTP or LTP2 in [10]).