Bidirectional modulation of anxiety-related and social behaviors by amygdala projections to the medial prefrontal cortex

Abstract

The basolateral amygdala (BLA) and the medial prefrontal cortex (mPFC) modulate anxiety and social behaviors. It remains to be elucidated, however, whether direct projections from the BLA to the mPFC play a functional role in these behaviors. We used optogenetic approaches in behaving mice to either activate or inhibit BLA inputs to the mPFC during behavioral assays that assess anxiety-like behavior and social interaction. Channelrhodopsin-2 (ChR2)-mediated activation of BLA inputs to the mPFC produced anxiogenic effects in the elevated plus maze and open field test, whereas halorhodopsin (NpHR)-mediated inhibition produced anxiolytic effects. Furthermore, activation of the BLA-mPFC pathway reduced social interaction in the resident-intruder test, whereas inhibition facilitated social interaction. These results establish a causal relationship between activity in the BLA-mPFC pathway and the bidirectional modulation of anxiety-related and social behaviors.

Keywords: anxiety disorders; fear; infralimbic; optogenetics; prelimbic; stress.

Copyright © 2015 The Authors. Published by Elsevier Ltd.. All rights reserved.

Figures

Fig. 1

Photostimulation of BLA terminals in the mPFC increased anxiety-like behavior. (A) Illustration of infusion of viral vectors allowing expression of either ChR2-eYFP or eYFP alone into the BLA and optical fiber placement over the mPFC for photostimulation (ChR2 group n = 9, eYFP group n = 8). (B) Coronal confocal images (at 20X) show expression of ChR2-eYFP in BLA somata, as well as in BLA terminals within the prelimbic (PL) and infralimbic (IL) subregions of the mPFC (blue, DAPI; red, cfos; green, ChR2-eYFP). (C) Elevated-plus maze (EPM) testing consisted of 3-min epochs with alternating laser manipulation (OFF-ON-OFF). Heat maps show time spent at each location within the maze or chamber for a representative ChR2-mouse during the initial OFF epoch and the ON epoch (cooler shades represent less time and warmer shades represent more time spent at that location). (D) ChR2-mice spent significantly less time in the open arms of the EPM during the ON epoch, relative to eYFP-mice and relative to the ChR2 group during the OFF epoch. (E) Photostimulation also reduced the probability to enter the open arms of the EPM. (F) The open-field test (OFT) also consisted of 3-min epochs with alternating laser treatment (OFF-ON-OFF). Heat maps representing the time spent at each location are shown for a ChR2-mouse during the first OFF epoch and the ON epoch. (G) Average time mice spent exploring the center of the OFT arena. The two OFF epochs are combined on the main bar graph, and illustrated individually in the line plot inset. ChR2-mice spent significantly less time in the center of the arena during the ON epoch, relative to eYFP-mice and the OFF epochs. (H) No significant effects were detected in the total distance travelled by mice in the OFT. In all figures, data are illustrated as mean ± SEM. Numbers within bars indicate the n’s per group. ***p < 0.001, **p < 0.01, *p < 0.05, corrected for multiple comparisons.

Fig. 2

Photostimulation of BLA terminals in the mPFC reduced social interaction. (A) Illustration of the resident-juvenile-intruder paradigm. Two 3-min sessions separated by a 24-hr interval were conducted with counterbalanced laser treatment (ChR2 group n = 12, eYFP group n = 10). (B) Average time resident mice spent engaged in social-related behaviors. ChR2-mice spent significantly less time socializing with the juvenile intruders during the laser-ON session than eYFP-mice. (C) No significant differences were detected in self-grooming behavior. (D) A small but significant increase was observed in homecage exploration. (E) No significant differences were observed in freezing/immobilization behavior. (F) Distribution of specific social and non-social behaviors for the entire 3 min epoch (s).

Fig. 3

Activation of BLA inputs was sufficient to activate mPFC neurons. Mice underwent a 3-min photostimulation session in their homecage ~90 min prior to being sacrificed. Immunoreactivity to cfos was used as a proxy for neuronal activity. (ChR2 group n = 9, eYFP group n = 8). (A) 40X confocal images of the BLA from representative ChR2 and eYFP mice (blue: DAPI+ cells, green: eYFP+ cells, red: cfos+ cells). (B) Percentage of eYFP+ and cfos+ cells in the BLA, relative to the total DAPI+ cell counts. No significant differences were detected between the ChR2 and eYFP-control groups. (C) 40X images of PL sub-regions of the mPFC. (D) Percentage of eYFP+ and cfos+ cells in the PL sub-region of the mPFC, relative to DAPI counts. (E) 40X images of IL sub-regions of the mPFC. (D) Percentage of eYFP+ and cfos+ cells in the IL subr-region of the mPFC, relative to DAPI counts. As expected, almost no mPFC cell was eYFP+. The proportion of cfos+ mPFC (Both PL and IL) cells was significantly higher in the ChR2 group than the eYFP-control group, suggesting that photostimulation of BLA inputs facilitate mPFC activity.

Fig. 4

Location of viral infusion in the BLA and optical fiber placement in the mPFC for the ChR2 experiments. (A) Coronal drawings across multiple levels of the BLA, relative to bregma, show the center point of viral infusion for each animal (ChR2-mice: purple circles, eYFP-mice: grey circles). (B) Optical fiber tip in the mPFC for each animal (ChR2-mice: purple crosses, eYFP-mice: grey crosses).

Fig. 5

Photoinhibition of BLA terminals in the mPFC reduced anxiety-like behavior and facilitated social interaction. (A) Brain schematics illustrate the bilateral photoinhibition approach. Representative confocal images (at 40X) are shown for a mouse in the NpHR group (blue: DAPI, green: NpHR-eYFP). (B) Heat maps representing time spent at each location in the OFT for an NpHR-mouse during the initial laser-OFF epoch and the laser-ON epoch (NpHR group n = 10, eYFP group n = 9). (C) Mean time in the center zone of the OFT. The two laser-OFF epochs are combined on the main bar graph and illustrated individually in the inset. NpHR-mice spent significantly higher time in the center zone during the laser-ON epoch, relative to eYFP-mice and to the laser-OFF epoch. (D) No significant differences were observed in the total distance travelled by mice in the OFT. (E) Experimental design for the social task (NpHR group n = 11, eYFP group n = 12). (F) NpHR-mice spent significantly more time engaged in social interaction during photoinhibition, relative to eYFP-mice. (G–I) No significant differences were observed on self-grooming (G), homecage exploration (H), or freezing behavior (I). (J) Distribution of specific social and non-social behaviors for the entire epoch.

Fig. 6

Location of viral infusion in the BLA and optical fiber placement in the mPFC for the NpHR experiments. (A) Center point of viral infusion in the bilateral BLA (NpHR-mice: green circles, eYFP-mice: grey circles). (B) Optical fiber tip in the bilateral mPFC (NpHR-mice: green crosses, eYFP-mice: grey crosses).