Social defeat stress causes selective attenuation of neuronal activity in the ventromedial prefrontal cortex

Abstract

The ventromedial prefrontal cortex (vmPFC) plays key roles in higher cognitive abilities, including mental representations and the regulation of emotion. Previous studies have reported that vmPFC activity is altered in depressed human patients, highlighting this subregion as a major site of dysfunction in neuropsychiatric diseases. To examine how neuronal activity at spike levels in the vmPFC is altered by social defeat stress, we performed electrophysiological multiunit recordings along the dorsoventral axis of the mPFC of freely moving mice. Chronic social defeat stress-susceptible mice showing an impairment in social interaction exhibited significant reductions in the overall spike frequencies of neurons in the vmPFC, but not in the dorsal mPFC. Analysis of local field potentials revealed that the vmPFC generated spatially constrained 20-40 Hz events lasting hundreds of milliseconds, with an average event frequency of 0.05 Hz; during these events, a subset of neurons were transiently inhibited. The frequency of 20-40 Hz events in the vmPFC was reduced in defeated stress-susceptible animals, and this decrease was reversed by systemic ketamine administration. The novel neurophysiological correlates of stress-induced changes in the vmPFC advance the understanding of the neural basis of stress-induced dysregulation of social behavior.

Figure 1

The vmPFC, but not the dmPFC, shows pronounced decreases in neuronal spike rates. (A) Schematic illustration of the social interaction (SI) test showing the interaction zone (IZ) and the other region. (B) Time periods in which individual mice stayed in the IZ. Each line shows each mouse (n = 14 non-SD mice and 73 SD mice). (C) For each animal, an SI ratio was computed as the ratio of occupancy time in the IZ region in the target session to that in the no-target session. Each dot represents an animal. Non-SD mice with a SI ratio of more than 1 are labeled as the orange region, termed as control mice, and SD mice with a SI ratio of less than and more than 1 were classified as stress-susceptible (magenta) and stress-resilient (cyan) mice, respectively. (D) (Top) A 32-site silicon probe was implanted along the dorsoventral axis of the mPFC. Recording sites (interval = 100 μm) are represented by a red line. (Bottom) Histological verification of the dorsal part of a recording site indicated by the arrow. (E) Representative simultaneous recordings of LFP signals obtained from the dmPFC and vmPFC using the silicon probe. Unfiltered and bandpass-filtered (20–40 Hz) LFP traces, including 20–40 Hz events (arrows), are superimposed and are shown in black and magenta, respectively. The shaded regions are magnified in Fig. 2A. The horizontal scalebar represents 500 ms. (F) Histological confirmation of tetrode recording locations in the ACC, dmPFC and vmPFC (indicated by arrows). (G) Comparison of neuronal firing rates in individual subcortical areas during SI tests in control and stress-susceptible mice (ACC, n = 11 and 13 cells; dmPFC, n = 32 and 15 cells; vmPFC, n = 15 and 24 cells). *P < 0.05, Mann-Whitney U test.

Figure 2

Physiological characteristics of vmPFC 20–40 Hz events in control mice. (A) LFP power spectrogram constructed from LFP traces magnified from the shaded regions in Fig. 1E. (B) Simultaneous LFP recordings from the dmPFC and vmPFC during the occurrence of a vmPFC 20–40 Hz event. Each panel shows typical unfiltered traces (top), 20–40 Hz-filtered LFP traces (middle), and the corresponding z-scored RMS amplitudes (bottom). (C) The average frequency of 20–40 Hz events during SI tests plotted against variable detection thresholds of z-scored amplitudes (n = 18 electrodes from 11 mice (vmPFC), 15 electrodes from 11 mice (dmPFC), and 11 electrodes from 8 mice (ACC)). *P < 0.05, Mann–Whitney U test followed by Bonferroni correction, vmPFC versus dmPFC and vmPFC versus ACC. (D) Distribution of vmPFC 20–40 Hz event intervals between two neighboring 20–40 Hz events (n = 352 intervals). (E,F) Distribution of duration (E) and oscillation frequency (F) of individual 20–40 Hz events (n = 382 events).

Figure 3

Modulation of vmPFC neuronal spikes by vmPFC 20–40 Hz events in control mice. (A) Two sample cells showing significant reductions in their firing rates during vmPFC 20–40 Hz events. (Top) A raster plot showing the spike patterns of a vmPFC neuron aligned to the centers of individual vmPFC 20–40 Hz events. Each dot indicates a spike. (Bottom) Average firing rate changes corresponding to the raster plots. (B) Relationship between baseline firing rate and percentage change in firing rate during vmPFC 20–40 Hz events (n = 14 cells). Each dot represents a vmPFC neuron. The magenta and blue dots in Fig. 3A show cell #1 and cell #2, respectively.

Figure 4

Reduction in vmPFC 20–40 Hz events in stress-susceptible mice. (A) Typical animal’s trajectories (gray) with vmPFC 20–40 Hz event locations superimposed (red dots). (B) Comparison of vmPFC 20–40 Hz events in control and stress-susceptible mice (n = 19 electrodes from 11 control mice and 28 electrodes from 17 stress-susceptible mice). *P < 0.05, Mann-Whitney U test.

Figure 5

Mice showing a ketamine-induced increase in social interaction behavior exhibit a restoration of vmPFC 20–40 Hz events. (A) Changes in the SI ratios of stress-susceptible mice by saline (left, n = 8 mice) or ketamine (right, n = 13 mice) administration. Each line shows each mouse. Six mice (magenta) showed marked increases in SI ratios after ketamine administration (termed “restored”), whereas the other seven mice showed no changes in SI ratio (termed “non-restored”). (B) (Left) Increases in vmPFC 20–40 Hz event rates in the restored mice (n = 9 electrodes from 4 mice). (Right) No significant changes in vmPFC 20–40 Hz event rates were observed in the 5 non-restored mice that showed no changes in SI ratios after ketamine treatment (n = 10 electrodes from 5 mice). *P < 0.05, paired t-test.