Impact of Vortioxetine on Synaptic Integration in Prefrontal-Subcortical Circuits: Comparisons with Escitalopram

Abstract

Prefrontal-subcortical circuits support executive functions which often become dysfunctional in psychiatric disorders. Vortioxetine is a multimodal antidepressant that is currently used in the clinic to treat major depressive disorder. Mechanisms of action of vortioxetine include serotonin (5-HT) transporter blockade, 5-HT_1A receptor agonism, 5-HT_1B receptor partial agonism, and 5-HT_1D, 5-HT₃, and 5-HT₇ receptor antagonism. Vortioxetine facilitates 5-HT transmission in the medial prefrontal cortex (mPFC), however, the impact of this compound on related prefrontal-subcortical circuits is less clear. Thus, the current study examined the impact of systemic vortioxetine administration (0.8 mg/kg, i.v.) on spontaneous spiking and spikes evoked by electrical stimulation of the mPFC in the anterior cingulate cortex (ACC), medial shell of the nucleus accumbens (msNAc), and lateral septal nucleus (LSN) in urethane-anesthetized rats. We also examined whether vortioxetine modulated afferent drive in the msNAc from hippocampal fimbria (HF) inputs. Similar studies were performed using the selective 5-HT reuptake inhibitor [selective serotonin reuptake inhibitors (SSRI)] escitalopram (1.6 mg/kg, i.v.) to enable comparisons between the multimodal actions of vortioxetine and SSRI-mediated effects. No significant differences in spontaneous activity were observed in the ACC, msNAc, and LSN across treatment groups. No significant impact of treatment on mPFC-evoked responses was observed in the ACC. In contrast, vortioxetine decreased mPFC-evoked activity recorded in the msNAc as compared to parallel studies in control and escitalopram treated groups. Thus, vortioxetine may reduce mPFC-msNAc afferent drive via a mechanism that, in addition to an SSRI-like effect, requires 5-HT receptor modulation. Recordings in the LSN revealed a significant increase in mPFC-evoked activity following escitalopram administration as compared to control and vortioxetine treated groups, indicating that complex modulation of 5-HT receptors by vortioxetine may offset SSRI-like effects in this region. Lastly, neurons in the msNAc were more responsive to stimulation of the HF following both vortioxetine and escitalopram administration, indicating that elevation of 5-HT tone and 5-HT receptor modulation may facilitate excitatory hippocampal synaptic drive in this region. The above findings point to complex 5-HT receptor-dependent effects of vortioxetine which may contribute to its unique impact on the function of prefrontal-subcortical circuits and the development of novel strategies for treating mood disorders.

Keywords: cingulate cortex; escitalopram; lateral septum; nucleus accumbens; prefrontal cortex; serotonin; vortioxetine.

FIGURE 1

Between groups comparison of the effects of vortioxetine and escitalopram on medial prefrontal cortex (mPFC)-evoked responses recorded from anterior cingulate cortex neurons. (A–C) Representative traces of mPFC-evoked responses recorded from isolated anterior cingulate cortex (ACC) neurons (1000 μA stimulus intensity) in controls (A), and following vortioxetine (B), or escitalopram (C) administration. Ten consecutive overlaid responses are shown. There was no effect of vortioxetine or escitalopram on (D) spike probability, (E) onset latency, and (F) standard deviation (SD) of latency. Data are presented as Mean ± SEM and analyzed using one-way analysis of variance (ANOVA) (n = 32 cells for control, n = 21 cells for vortioxetine-treated, and n = 31 cells for escitalopram-treated groups).

FIGURE 2

Between groups comparison of the effects of vortioxetine and escitalopram on mPFC-evoked responses recorded from isolated lateral septal nucleus neurons. (A–C) Representative traces of mPFC-evoked responses recorded from isolated lateral septal nucleus (LSN) neurons (1000 μA stimulus intensity) in controls (A), and following vortioxetine (B), or escitalopram (C) administration. Ten consecutive overlaid responses are shown. (D) Escitalopram significantly increased spike probability at 800 μA [F(2,60) = 4.165, p = 0.0205] and 600 μA [F(2,60) = 6.358, p = 0.0032] stimulus intensities. There was also a trend toward increased spike probability at 1000 μA [F(2,60) = 2.917, p = 0.0618] stimulus intensities. Post hoc comparisons revealed a significant increase in spike probability following escitalopram administration compared with vortioxetine (^∗∗p < 0.01 at 600 μA, ^∗p < 0.05 at 800 μA) and control groups (^∗∗p < 0.01 at 600 μA, ^∗p < 0.05 at 800 μA, ^#p = 0.06). No changes in onset latency (E) and SD of latency (F) were observed with either vortioxetine or escitalopram as compared with controls. Data are presented as Mean + SEM and analyzed using one-way ANOVA with Tukey post hoc analysis (n = 26 cells for control, n = 20 cells for vortioxetine-treated, and n = 20 cells for escitalopram-treated groups).

FIGURE 3

Between groups comparison of the effects of vortioxetine and escitalopram on mPFC-evoked responses recorded from isolated nucleus accumbens shell projection neurons. (A–C) Representative traces of mPFC-evoked responses recorded from isolated msNAc neurons (1000 μA stimulus intensity) in controls (A), and following vortioxetine (B), or escitalopram (C) administration. Ten consecutive overlaid responses are shown. (D) Vortioxetine significantly reduced spike probability at 1000 μA [F(2,58) = 5.466, p = 0.0067] and 800 μA [F(2,58) = 3.627, p = 0.0328] stimulus intensities. There was also a trend toward a decrease in spike probability at 600 μA stimulus intensities [F(2,58) = 2.653, p = 0.0794]. Post hoc comparisons revealed a significant decrease in spike probability following vortioxetine administration compared with escitalopram (^∗p < 0.05 at 1000 μA and 800 μA) and control groups (^∗∗p < 0.01 at 1000 μA, ^∗p < 0.05 at 800 μA, ^#p = 0.08). No changes in onset latency (E) or SD of latency (F) were observed following either vortioxetine or escitalopram administration as compared with controls. Data are presented as Mean ± SEM and analyzed using one-way ANOVA with Tukey post hoc analysis (n = 21 cells for control, n = 20 cells for vortioxetine-treated, and n = 20 cells for escitalopram-treated groups).

FIGURE 4

Between groups comparison of fimbria-evoked responses recorded in isolated nucleus accumbens shell projection neurons. (A–C) Representative traces of fimbria-evoked responses recorded from isolated msNAc neurons in controls (A), and following vortioxetine (B), or escitalopram (C) administration. Ten consecutive overlaid responses are shown. Stimulus intensities were titrated to evoke a 50% response (approximately) to fimbria stimulation, and ranged from 200 to1300 μA (D). (E) Escitalopram induced a significant decrease in onset latency of fimbria-evoked responses compared with control and vortioxetine groups [F(2,17) = 7.030, p = 0.0060]. Post hoc comparisons revealed a significant decrease in onset latency following escitalopram administration compared with control (^∗∗p < 0.01) and vortioxetine-treated groups (^∗p < 0.05). (F) Vortioxetine and escitalopram significantly decreased the stimulus intensity required to evoke a 50% response [F(2,17) = 15.02, p = 0.0002]. Post hoc comparisons revealed a significant decrease in stimulus intensity post vortioxetine (^∗∗p < 0.01) and escitalopram (^∗∗∗p < 0.001) compared with controls. SD of latency was unchanged following drug treatment (data not shown). Data are presented as Mean ± SEM and analyzed using one-way ANOVA with Tukey post hoc analysis (n = 8 cells for control, n = 6 cells for vortioxetine-treated, and n = 6 cells for escitalopram-treated groups).