Modulation of neuronal excitability by serotonin-NMDA interactions in prefrontal cortex

Abstract

Both serotonin and NMDA signaling in prefrontal cortex (PFC) are implicated in mental disorders, including depression and anxiety. To understand their potential contributions to PFC neuronal excitability, we examined the effect of co-activation of 5-HT and NMDA receptors on action potential firing elicited by depolarizing current injection in PFC pyramidal neurons. In the presence of NMDA, a low concentration of the 5-HT(1A) agonist 8-OH-DPAT substantially reduced the number of spikes, and a low concentration of the 5-HT(2A/C) agonist alpha-Me-5HT significantly enhanced it, while both agonists were ineffective when applied alone. The 8-OH-DPAT effect on firing was mediated by inhibition of protein kinase A (PKA), whereas the alpha-Me-5HT effect was mediated by activation of protein kinase C (PKC). Moreover, the extracellular signal-regulated kinase (ERK), a signaling molecule downstream of PKA and PKC, was involved in both 5-HT(1A) and 5-HT(2A/C) modulation of neuronal excitability. Biochemical evidence showed that 5-HT(1A) decreased, whereas 5-HT(2A/C) increased the activation of ERK in an NMDA-dependent manner. In animals exposed to acute stress, the enhancing effect of 5-HT(2A/C) on firing was lost, while the decreasing effect of 5-HT(1A) on firing was intact. Concomitantly, the effect of 5-HT(2A/C), but not 5-HT(1A), on ERK activation was abolished in stressed animals. Taken together, our results demonstrate that distinct 5-HT receptor subtypes, by interacting with NMDA receptors, differentially regulate PFC neuronal firing, and the complex effects of 5-HT receptors on excitability are selectively altered under stressful conditions, which are often associated with mental disorders.

Figure 1

In the presence of NMDA, the 5-HT_1A or 5-HT_2A/C agonist reduces or enhances PFC pyramidal cell excitability, respectively. A-D, Representative traces of action potential firing evoked by depolarizing current pulses illustrating the effect of 8-OH-DPAT (1 μM, A, B) or α-Me-5HT (0.8 μM, C, D) applied in the presence of NMDA (4 or 10 μM). Scale bars: 20 mV, 100 ms. E, Cumulative data (mean ± SEM) showing the percentage change of the firing rate by 8-OH-DPAT or α-Me-5HT in the absence (compared to ACSF) or presence of NMDA (compared to NMDA alone). Different currents were injected to generate 3-5, 6-8 or 9-11 spikes under the control condition (ACSF or NMDA alone). The number of cells tested in each condition is shown in each bar. *: p<0.01, ANOVA, compared to the agonist effects in the absence of NMDA. F, G, Summary I-V graphs (voltages in response to a series of current injections) depicting passive membrane properties in cells before and after 8-OH-DPAT (1 μM, F) or α-Me-5HT (0.8 μM, G) application in the presence of NMDA (4 μM).

Figure 2

5-HT_1A or 5-HT_2A/C receptors mediate the agonist effects on AP firing, which is dependent on co-activation of NMDA receptors. A, B, D, E, Representative AP firing traces showing the effect of 8-OH-DPAT (1 μM) or α-Me-5HT (0.8 μM) (with 4 μM of NMDA present) in the presence of the 5-HT_1A antagonist NAN-190 (1 μM, A), 5-HT_2A/C antagonist ketanserin (5 μM, D), or NMDA receptor antagonist APV (50 μM, B, E). Scale bars: 20 mV, 100 ms. Currents were injected to generate 6-8 spikes under the control condition (NMDA alone). C, F, Cumulative data (mean ± SEM) showing the percentage change of the firing rate by 8-OH-DPAT (C) or α-Me-5HT (F) (with NMDA present) in the absence or presence of various antagonists. *: p<0.01, ANOVA, compared to the effect in the absence of antagonists (-).

Figure 3

In the presence of NMDA, low doses of 5-HT reduce PFC pyramidal cell excitability. A, B, Representative AP traces showing the effect of 5-HT (1 μM, A) or 5-HT (10 μM, B) on the firing rate (with 4 μM of NMDA present). Scale bars: 20 mV, 100 ms. C, Cumulative data (mean ± SEM) showing the percentage change of the firing rate by different doses of 5-HT (with NMDA present).

Figure 4

The regulation of AP firing by 5-HT_1A-NMDA interaction involves PKA and ERK. A-D, Representative AP firing traces showing the effect of 8-OH-DPAT (1 μM) in the presence of NMDA (4 μM) in neurons dialyzed without (control, A) or with the PKA inhibitor PKI_6-22 (20 μM, B), PKC inhibitor calphostin (1 μM, C), or ERK inhibitor U0126 (20 μM, D). Scale bars: 20 mV, 100 ms. Currents were injected to generate 6-8 spikes under the control condition (NMDA alone). E, Cumulative data (mean ± SEM) showing the percentage decrease of the firing rate by 8-OH-DPAT (with NMDA present) in the absence or presence of various inhibitors. *: p<0.01, ANOVA, compared to the effect in the absence of kinase inhibitors (-).

Figure 5

The regulation of AP firing by 5-HT_2A/C-NMDA interaction involves PKC and ERK. A-E, Representative AP firing traces showing the effect of α-Me-5HT (0.8 μM) in the presence of NMDA (4 μM) in neurons dialyzed without (control, A) or with the PLC inhibitor U73122 (5 μM, B), PKC inhibitor calphostin (1 μM, C), PKA inhibitor PKI_6-22 (20 μM, D), or ERK inhibitor U0126 (20 μM, E). Scale bars: 20 mV, 100 ms. Currents were injected to generate 6-8 spikes under the control condition (NMDA alone). F, Cumulative data (mean ± SEM) showing the percentage increase of the firing rate by α-Me-5HT (with NMDA present) in the absence or presence of various inhibitors. *: p<0.01, ANOVA, compared to the effect in the absence of kinase inhibitors (-).

Figure 6

Acute stress attenuates the effect of 5-HT_2A/C, but not 5-HT_1A, on AP firing. A-B, Representative AP firing traces showing the effect of 8-OH-DPAT (1 μM, A) or α-Me-5HT (0.8 μM, B) in the presence of NMDA (4 μM) in PFC pyramidal neurons from non-stressed (control) vs. stressed animals. Scale bars: 20 mV, 100 ms. Currents were injected to generate 6-8 spikes under the control condition (NMDA alone). C, Cumulative data (mean ± SEM) showing the percentage change of the firing rate by 8-OH-DPAT or α-Me-5HT (with NMDA present) in PFC neurons from control vs. stressed rats. *: p<0.01, ANOVA, compared to the effect in control animals.

Figure 7

The 5-HT_2A/C, but not 5-HT_1A, effect on ERK activation is lost in stressed animals. A, Western blot analysis of ^{Thr202/Tyr204}phos-ERK and total ERK in PFC slices treated without or with 8-OH-DPAT (1 μM, 10 min) or α-Me-5HT (0.8 μM, 10 min) in the presence of NMDA (4 μM, 10 min pre-incubated) from control vs. stressed animals. B, Bar graphs showing the percentage change of p-ERK by 8-OH-DPAT or α-Me-5HT (with NMDA present) in PFC slices from control vs. stressed animals. C, Western blot analysis of p-ERK and total ERK in PFC slices from control animals treated without or with 8-OH-DPAT (1 μM, 10 min) or α-Me-5HT (0.8 μM, 10 min) in the absence of NMDA. *: p<0.01, ANOVA, compared to the effect in control animals.