Phenotypical Screening on Neuronal Plasticity in Hippocampal-Prefrontal Cortex Connectivity Reveals an Antipsychotic with a Novel Profile

Abstract

Dysfunction in the hippocampus-prefrontal cortex (H-PFC) circuit is a critical determinant of schizophrenia. Screening of pyridazinone-risperidone hybrids on this circuit revealed EGIS 11150 (S 36549). EGIS 11150 induced theta rhythm in hippocampal slice preparations in the stratum lacunosum molecular area of CA1, which was resistant to atropine and prazosin. EGIS 11150 enhanced H-PFC coherence, and increased the 8−9 Hz theta band of the EEG power spectrum (from 0.002 mg/kg i.p, at >30× lower doses than clozapine, and >100× for olanzapine, risperidone, or haloperidol). EGIS 11150 fully blocked the effects of phencyclidine (PCP) or ketamine on EEG. Inhibition of long-term potentiation (LTP) in H-PFC was blocked by platform stress, but was fully restored by EGIS 11150 (0.01 mg/kg i.p.), whereas clozapine (0.3 mg/kg ip) only partially restored LTP. EGIS 11150 has a unique electrophysiological profile, so phenotypical screening on H-PFC connectivity can reveal novel antipsychotics.

Keywords: EGIS 11150; S 36549; dysconnectivity; ketamine; phencyclidine; schizophrenia; theta rhythm.

Figure 1

EGIS 11150 selectively generates a theta rhythm de novo in CA1 of hippocampal slices in vitro. (A). (i) Extracellular recordings from CA3 stratum radiatum show no effect of EGIS 11150. (ii) In stratum lacunosum molecular area of CA1, EGIS 11150 produces a persistent theta frequency rhythm. Scale bars 50 µV 200 ms. (iii) Laminar profile of CA1 theta, recorded in the layers indicated with extracellular electrode shows a highly localised occurrence in stratum lacunosum. Aalveus (Alv), Stratum oriens (SO), Stratum pyramidale (SPyr), Stratum radiatum (SR), Molecular layer dentate gyrus (ML), Granule cell layer dentate gyrus (GCL). (B). (i) EGIS 11150-generated theta rhythms were blocked by GBZ (2 µM) and ZD7288 (10 µM). (ii) Spectogram of the EGIS 11150-generated gamma rhythms. (C) EGIS 11150 induced theta rhythm is not affected by a muscarinic component in hippocampal region CA1 in vitro. Extracellular recordings from CA1 stratum lacunosum show 200 nM EGIS 11150 generates a persistent theta frequency rhythm, which is unaffected by the muscarinic agent, atropine, at a concentration of 100 nM. Scale bars 50 µV 200 ms.

Figure 2

Power spectra (total power) in the prefrontal and sensorimotor cortex with increasing doses of EGIS 11150. Panel (A) 0.002 mg/kg EGIS 11150 induces high theta bands (8–10 Hz) in the prefrontal cortex. Panel (B) 0.01 mg/kg EGIS 11150 maximally induces high theta bands in the prefrontal cortex. Panel (C) 0.05 mg/kg EGIS 11150 maximally induce high theta bands in the sensorimotor cortex (n = 6).

Figure 3

Comparison of the effects of EGIS-11150 on the EEG power spectrum with those of risperidone, olanzapine and clozapine. Panel (A) Composite graph of power spectra (total power) of the prefrontal cortex after administration of 0.3 mg/kg i.p. risperidone 5 mg/kg i.p. olanzapine, 0.3 mg/kg s.c. clozapine and 0.01 mg/kg i.p. EGIS-11150. Panel (B) Dose-response relationship of the effects of antipsychotics on total EEG power in the prefrontal cortex (n = 6).

Figure 4

Interaction of antipsychotic compounds with NMDA antagonists in the prefrontal cortex. Panel (A) EEG power spectra induced by increasing doses of PCP. Panel (B) Composite graph of power spectra (total power) of 1 mg/kg s.c. PCP, as well as 0.3 mg/kg s.c. clozapine, 0.5 mg/kg i.p. haloperidol, 0.3 mg/kg i.p. risperidone, 0.01 mg/kg i.p. EGIS 11150, 0.05 mg/kg i.p. EGIS-11150 in the presence of PCP. Panel (C) Power spectra (total power) of the prefrontal and sensorimotor cortex of 10 mg/kg i.p. ketamine and 0.05 mg/kg i.p. EGIS-11150 in the presence of ketamine (n = 6).

Figure 5

Effects of EGIS-11150 on EEG power in prefrontal cortex, in hippocampus and on hippocampal to prefrontal cortex coherence. Note the marked increase in hippocampal-prefrontal cortex coherence. Atropine (5 mg/kg i.p.) did not modify the effects of EGIS 11150 (0.05 mg/kg) indicating that the effects of EGIS 11150 on theta were presumably not of cholinergic origin; note however, that there was a small increase at 4 Hz in the presence of EGIS 11150 and atropine. Effects on sensorimotor cortex were considerably less marked. The 95% confidence interval corresponds to the error bars in each figure, n = 6.

Figure 6

Comparison of the effects of EGIS 11150 and clozapine on stress-induced impairment of hippocampal-prefrontal cortex LTP in vivo. Panel (A) EGIS 11150 administered at 0.01 mg/kg and clozapine at 0.3 mg/kg after stress, 40 min prior to HFS enables the induction of LTP in stressed rats when compared to saline-pretreated stressed rats. Values are mean ± SEM of the normalized postsynaptic potential (PSP) amplitude. Panel (B) LTP in saline, EGIS 11150 and clozapine pretreated stressed rats represented at different time periods. PSP amplitudes were analysed using A/Dvance software, expressed as a percentage change of the mean response over a 30 min baseline period and presented in figures as the mean ± SEM for 2 min epochs. Electrophysiological data were averaged in consecutive 30 min periods (T0–30 min, T30–60 min, T60–90 min, T90–120 min) after LTP induction. Statistical comparisons were carried out using analysis of variance [19]. The first and following groups of columns represent 30 min periods of mean ± SEM of the normalized PSP amplitude before, and after HFS. ANOVA: * p < 0.05 ** p < 0.01 and *** p < 0.001 vs. saline-treated stressed rats, ANOVA: ^# p < 0.05, ^## p < 0.01, vs. saline-treated stressed rats.

Figure 7

Haloperidol does not fully prevent stress-induced impairment of hippocampal-prefrontal cortex LTP in vivo. Panel (A) LTP in stressed rats treated with haloperidol is not significantly different from saline-treated stressed rats. Values are mean ± SEM of the normalized postsynaptic potential (PSP) amplitude. HFS is indicated by arrows. Panel (B) LTP in saline and haloperidol treated stressed rats represented at different time periods. The first and following groups of columns represent respectively 30 min periods of mean ± SEM of the normalized PSP amplitude before, and after HFS. ANOVA: * p < 0.05 vs. saline-treated stressed rats.

Figure 8

Application of E11150 to mouse hippocampal slices enhances submaximal and stress impaired LTP recorded from CA1 pyramidal neurones. Panel (A) The histogram illustrates that the magnitude of LTP (expressed as the percentage increase of the fEPSP slope) induced by the delivery of a 3-pulse TBS is significantly less than that resulting from delivery of a 4-pulse TBS protocol. The bath application of E11150 (100 nM and 300 nM) significantly enhanced the magnitude of LTP induced by the 3-pulse TBS. Each column represents the mean ± SEM of 4–11 individual experiments (* p < 0.05). Panel (B) A prior episode of acute stress to mice decreased the magnitude of ex vivo CA1 pyramidal neuron LTP induced by a 4-pulse TBS, in comparison to that induced by a 4-pulse TBS, recorded from equivalent CA1 neurones obtained from control mice. The bath application of E11150 (300 nM) significantly enhanced the magnitude of stress impaired LTP. Each column represents the mean ± SEM of 4–12 individual experiments * p < 0.05).

Figure 9

Impaired reversal learning by C57/Bl6 mice subjected to sub-chronic treatment with PCP is ameliorated by acute administration of Egis-11150 (0.03 and 0.1 mg/kg p.o.). Error scores of four acquisition sessions (left panel) and four reversal sessions (right panel) are shown. Data are mean ± SEM n = 10/group. Two-way ANOVA gave significant interaction drug treatment and trial number: F (9, 136) = 2.998, p = 0.0027. Dunnett’s post-hoc test: ^$ p < 0.001 for both 0.03 and 0.1 doses vs. the respective PCP treated group acutely receiving vehicle, ^§ p < 0.001 when compared to the group receiving sub-chronic as well as acute treatment with vehicle. (A = acquisition trials, R = reversal trials).