Phosphatidylinositol 3-kinase regulates the induction of long-term potentiation through extracellular signal-related kinase-independent mechanisms

Abstract

Inhibitors of both phosphatidylinositol-3-kinase (PI3-kinase) and MAPK/ERK (mitogen-activate protein kinase/extracellular signal-related kinase) activation inhibit NMDA receptor-dependent long-term potentiation (LTP). PI3-kinase inhibitors also block activation of ERK by NMDA receptor stimulation, suggesting that PI3-kinase inhibitors block LTP because PI3-kinase is an essential upstream regulator of ERK activation. To examine this hypothesis, we investigated the effects of PI3-kinase inhibitors on ERK activation and LTP induction in the CA1 region of mouse hippocampal slices. Consistent with the notion that ERK activation by NMDA receptor stimulation is PI3-kinase dependent, the PI3-kinase inhibitor wortmannin partially inhibited ERK2 activation induced by bath application of NMDA and strongly suppressed ERK2 activation by high-frequency synaptic stimulation. PI3-kinase and MEK (MAP kinase kinase) inhibitors had very different effects on LTP, however. Both types of inhibitors suppressed LTP induced by theta-frequency trains of synaptic stimulation, but only PI3-kinase inhibitors suppressed the induction of LTP by high-frequency stimulation or low-frequency stimulation paired with postsynaptic depolarization. Concentrations of PI3-kinase inhibitors that inhibited LTP when present during high-frequency stimulation had no effect on potentiated synapses when applied after high-frequency stimulation, suggesting that PI3-kinase is specifically involved in the induction of LTP. Finally, we found that LTP induced by theta-frequency stimulation was MEK inhibitor insensitive but still PI3-kinase dependent in hippocampal slices from PSD-95 (postsynaptic density-95) mutant mice. Together, our results indicate that the role of PI3-kinase in LTP is not limited to its role as an upstream regulator of MAPK signaling but also includes signaling through ERK-independent pathways that regulate LTP induction.

Fig. 1.

NMDA-induced increases in phosphorylated ERK2 are only partially PI3-kinase dependent. A, Average ± SEM results from nine experiments in which slices from the same animal were either left untreated (open bar) or exposed to 20 μmNMDA (5 min), 200 nm wortmannin (Wort) (≥40 min), or NMDA plus wortmannin. Phospho-ERK2 levels were significantly increased by NMDA in both control and wortmannin-treated slices (*p < 0.01 compared with untreated controls). The increase in phospho-ERK2 levels induced by NMDA in wortmannin-treated slices (250 ± 25.2% of untreated control slices) was significantly less than that induced by NMDA in slices bathed in ACSF (341 ± 30.4% of untreated control slices; #p < 0.05). Wortmannin did not have a significant effect on basal levels of phospho-ERK2. Total levels of ERK2 were unchanged in slices exposed to wortmannin alone, NMDA alone, or wortmannin plus NMDA (levels were 108 ± 5.5, 116 ± 6.4, and 108.3 ± 8.1% of that seen in untreated control slices, respectively). B, Results from individual experiments summarized in A. The plot shows the NMDA-induced increase in phospho-ERK2 levels in slices bathed in ACSF (control) versus ACSF plus 200 nm wortmannin (filled symbols) or 5 μm wortmannin (open symbols). C, Wortmannin inhibits NMDA-induced increases in phospho-Akt in hippocampal slices. Average ± SEM results from nine experiments. In the absence of wortmannin, NMDA (20 μm, 5 min) increased phospho-Akt levels to 214 ± 58% of baseline (*p < 0.05 compared with untreated control). Wortmannin reduced basal levels of phospho-Akt to 15.4 ± 6.2% of control (p < 0.05 compared with untreated controls) and completely blocked NMDA-induced increases in phospho-Akt (phospho-Akt levels were 9.3 ± 2% of control in wortmannin-treated slices exposed to NMDA). D, Representative immunoblots showing the effects of wortmannin on levels of phospho-ERK1/2 and phospho-Akt (Thr308) in untreated (UT) and NMDA treated (N) slices.

Fig. 2.

Blocking PI3-kinase with wortmannin inhibits HFS-induced ERK2 activation. A, Western immunoblots showing the effect of HFS delivered in the presence and absence of wortmannin (Wort) on phospho-ERK1/2 and total ERK1/2 levels in CA1 mini-slices. Lane 1, Untreated control (ACSF alone, no HFS); lane 2, HFS in ACSF; lane 3, 200 nm wortmannin alone; lane 4, HFS in wortmannin. Note that, although HFS delivered to slices bathed in normal ACSF induced a robust increase in phospho-ERK2 levels, it had little effect on phospho-ERK2 levels in wortmannin-treated slices.B, C, Average ± SEM results from seven experiments showing basal and stimulated (2.5–5 min after HFS) levels of phospho-ERK2 (B) and total ERK2 (C) in slices bathed in normal ACSF or ACSF containing 200 nm wortmannin. High-frequency stimulation delivered to slices bathed in ACSF increased phospho-ERK2 levels to 182 ± 19% of unstimulated controls (*p < 0.05 compared with unstimulated controls). High-frequency stimulation-induced ERK activation was significantly reduced in wortmannin-treated slices (levels were 125 ± 16% of unstimulated controls; #p < 0.05 compared with HFS in ACSF) and not significantly different from levels of phospho-ERK2 in unstimulated slices bathed in ACSF alone or ACSF plus wortmannin. None of the treatments had any effect on total ERK2 levels.

Fig. 3.

The PI3-kinase inhibitor LY294002 suppresses TPS-induced LTP. A, A 150 pulse train of TPS delivered at time 0 induced LTP in vehicle control experiments (0.1% DMSO; open symbols; n = 6) but had little lasting effect on synaptic transmission in slices treated with 20 μmLY294002 for at least 40 min before TPS (filled symbols;n = 6). LY294002 was present throughout the experiment. Inset shows fEPSPs recorded during baseline and 45 min after TPS in a control slice (left set of traces) and in a slice bathed in LY294002 (right set of traces). Calibration: 2 mV, 5 msec.B, LY294002 has no effect on synaptic responses evoked during TPS. Note that both the facilitation at the start of TPS and the depression of synaptic transmission at the end of the TPS train are similar in control (open symbols) and LY294002-treated slices (filled symbols). On average, fEPSPs elicited by pulse 2 to pulse 6 of the TPS train facilitated to 147.6 ± 6.1% of baseline in control experiments and were facilitated to 145.8 ± 4.8% of baseline in LY294002-treated slices (not significantly different;p = 0.82). In control experiments, fEPSPs were depressed to 37.1 ± 6.2% of baseline at the end of the TPS train (average of the last 5 stimulation pulses) and were depressed to 45.2 ± 5.8% of baseline in LY294002-treated slices (not significantly different; p = 0.363). Recordings were done using slices maintained in an interface chamber.

Fig. 4.

PI3-kinase inhibitors suppress the induction of LTP by long trains of TPS paired with β-adrenergic receptor activation and inhibit ISO-induced ERK2 activation. A, Nine hundred pulses of TPS (delivered at time 0) paired with a 10 min application of 1 μm ISO (indicated by the bar) induced significantly less potentiation in slices continuously bathed in 200 nm wortmannin (filled symbols; n = 14) compared with slices bathed in normal ACSF plus vehicle (0.1% DMSO; open symbols; n = 15). A long train of TPS delivered in the absence of ISO had little lasting effect on synaptic transmission in control and wortmannin-treated slices (data not shown).B, Wortmannin (Wort) inhibits ISO-induced ERK2 activation. Average ± SEM results from five separate experiments in which slices from the same animal were either left untreated (UT; open bar) or exposed to 1 μm ISO (5 min), 200 nm wortmannin (≥40 min), or ISO plus wortmannin. In vehicle control experiments (0.01–0.1% DMSO), phospho-ERK2 levels were increased to 250 ± 40.2% of control in ISO-treated slices (*p < 0.05 compared with untreated controls). A significantly smaller increase in phospho-ERK2 levels was induced by ISO in wortmannin-treated slices (phospho-ERK2 levels were increased to 164.1 ± 17.8% of control; #p < 0.05 compared with levels in slices treated with ISO alone). Total levels of ERK2 were unchanged in slices exposed to wortmannin alone, ISO alone, or wortmannin plus ISO (data not shown). The inset shows a representative immunoblot showing basal and ISO-stimulated levels of phospho-ERK2 in slices bathed in ACSF alone and ACSF plus 200 nm wortmannin. C, Wortmannin has no effect on the increase in phospho-S845 GluR1 levels induced by ISO. In slices bathed in ACSF, a 5 min application of 1.0 μm ISO increased phospho-GluR1 levels to 512 ± 69% of control (n = 5). In wortmannin-treated slices, phospho-GluR1 levels were 547.7 ± 90.4% of control. The inset shows a representative immunoblot showing basal and ISO-stimulated levels of phospho-S845 GluR1 in slices bathed in ACSF and ACSF plus wortmannin.

Fig. 5.

Low-frequency presynaptic fiber stimulation paired with postsynaptic depolarization induces an ERK-independent but PI3-kinase inhibitor-sensitive form of LTP. A, The MEK inhibitor U0126 (20 μm) blocks NMDA-induced ERK2 activation. Western immunoblot showing phospho-ERK2 levels (top) and total ERK levels (bottom) in a representative experiment. Lane 1, Untreated control (slices exposed to ACSF alone); lane 2, NMDA (20 μm, 5 min) in ACSF; lane 3, 20 μm U0126 alone (≥40 min); lane 4, U0126 plus NMDA. Note that U0126 strongly reduces basal levels of phospho-ERK2 and blocks the increase in ERK2 phosphorylation induced by NMDA. None of the treatments had an affect on total ERK2 levels. The same results were obtained in three separate experiments. B, Pairing-induced LTP is not blocked by U0126 but is inhibited by LY294002. Presynaptic stimulation pulses paired with postsynaptic depolarization to near 0 mV (at time 0) induced nearly identical levels of potentiation in vehicle control experiments (0.1–0.2% DMSO; open circles; EPSPs were potentiated to 267 ± 13% of baseline; n = 13) and in cells recorded from slices continuously bathed in ACSF containing 20 μm U0126 (triangles; EPSPs were potentiated to 275 ± 13% of baseline; n = 9). Significantly less LTP was induced, however, in cells in which LY294002 (100 μm) was included in the electrode filling solution to block postsynaptic PI3-kinase (filled circles; EPSPs were potentiated to 159 ± 22% of baseline; n = 8; p < 0.001 compared with vehicle control cells). The inset shows EPSPs (average of 3 responses) recorded during baseline (smaller responses) and 30 min after pairing. Calibration: 5 mV, 20 msec.

Fig. 6.

HFS-induced LTP is not blocked by MEK inhibitor SL327 but is suppressed by PI3-kinase inhibitors. A, Concentrations of the MEK inhibitor SL327 that block HFS-induced ERK2 activation have no effect on HFS-induced LTP. In experiments done in interface slice chambers, the amount of potentiation present 60 min after HFS (delivered at time 0) is the same in control and SL327-treated slices. In control experiments (open symbols; 0.1% DMSO), fEPSPs were potentiated to 173 ± 12% of baseline 60 min after HFS (n = 6) and were potentiated to 182 ± 12% of baseline in slices continuously bathed in 10 μm SL327 (filled symbols; n = 6). The inset shows example immunoblots probed with antibodies specific for phospho-ERK1/2 (top) and total ERK1/2 (bottom). Lane 1, Untreated control (ACSF alone); lane 2, HFS in ACSF; lane 3, SL327 alone (≥40 min); lane 4, HFS in SL327. Note that SL327 strongly reduced both basal levels of phospho-ERK2 and the increase in ERK2 phosphorylation induced by HFS. None of the treatments affected total ERK2 levels. The same results were obtained in three separate experiments. B, Inhibiting PI3-kinase activity with LY294002 significantly reduces HFS-induced LTP. In vehicle control experiments (0.1% DMSO; open symbols), fEPSPs were potentiated to 185 ± 16% of baseline 60 min after HFS; n = 6). In contrast, fEPSPs were potentiated to 138 ± 6% of baseline 60 min after HFS in slices continuously bathed in 20 μm LY294002 (filled symbols;n = 6; p < 0.05 compared with control). The inset shows fEPSPs recorded during baseline and 60 min after HFS (larger responses) in a control slice (left set of traces) and in a slice bathed in ACSF containing LY294002 (right set of traces). Calibration: 2 mV, 5 msec. Experiments were done using interface slice recording chambers.

Fig. 7.

PI3-kinase inhibitors suppress the induction but not the maintenance and expression of LTP. A, Wortmannin (Wort) inhibits HFS-induced LTP in submerged slices. Slices were continuously bathed in ACSF containing 0.02% DMSO (vehicle control; open symbols; n = 10) or ACSF containing 200 nm wortmannin (filled symbols; n = 10). One hour after HFS, fEPSPs were potentiated to 174 ± 11% of baseline in control slices but increased to only 132 ± 9% of baseline in wortmannin-treated slices (p < 0.01 compared with control). B, PI3-kinase (PI3K) inhibitors have no effect on established LTP. Wortmannin (200 nm; n = 3) or LY294002 (10 μm; n = 4) were applied starting 30 min after HFS (duration of inhibitors in the bath indicated by the bar). Neither wortmannin nor LY294002 inhibited potentiated synaptic transmission, and the results from all experiments were combined. All experiments were done using a submerged-slice recording chamber.

Fig. 8.

High concentrations of PI3-kinase inhibitors depress potentiated synaptic transmission but also inhibit transmission at unpotentiated synapses. A, Potentiated synaptic transmission is inhibited by a high concentration of LY294002 (100 μm) applied for 30 min starting 30 min after LTP induction (n = 5; presence of LY294002 in the recording chamber indicated by the bar). At this concentration, LY294002 also inhibited basal synaptic transmission in a separate series of experiments (open symbols; n = 5).B, A 30 min application of a high concentration of wortmannin (5 μm; Wort) also inhibited both potentiated (filled symbols; n = 5) and basal synaptic transmission (open symbols; n = 4). The bar indicates presence of wortmannin in the recording chamber. All experiments were done using a submerged-slice recoding chamber.

Fig. 9.

TPS-induced LTP is ERK independent but still PI3-kinase dependent in hippocampal slices from PSD-95 mutant mice.A, Both MEK and PI3-kinase inhibitors suppress the induction of LTP by a 150 pulse train of TPS in slices from wild-type animals. In vehicle control experiments (0.1- 0.2% DMSO; open circles), fEPSPs were potentiated to 205 ± 7% of baseline 45 min after TPS (n = 9). The amount of potentiation induced by TPS was significantly reduced (p< 0.001) in slices continuously bathed in ACSF containing either 200 nm wortmannin (Wort; filled circles; fEPSPs were 125 ± 11% of baseline; n = 6) or 20 μmU0126 (filled triangles; fEPSPs were 137 ± 9% of baseline;n = 11). B, The amount of LTP induced by a 150 pulse train of TPS was not reduced by U0126 in slices from PSD-95 mutant mice. In vehicle control experiments, fEPSPs were potentiated to 279 ± 15% of baseline (open circles;n = 14) and were potentiated to 270 ± 12% of baseline (n = 14) in slices continuously bathed in ACSF containing U0126 (triangles). LTP induced by TPS in PSD-95 mutant slices was inhibited by wortmannin. In slices from PSD-95 mutant mice that were continuously bathed in 200 nm wortmannin, fEPSPs were potentiated to 195 ± 13% of baseline (filled circles;n = 7; p < 0.05 compared with control levels of LTP in PSD-95 mutant slices). All experiments were done in an interface-type recording chamber.