Distinct roles of segregated transmission of the septo-habenular pathway in anxiety and fear

Abstract

The posterior septum consisting of the triangular septum (TS) and the bed nucleus of the anterior commissure (BAC) is predominantly linked with the medial habenula (MHb) and has been implicated in the control of anxiety and fear responses. However, its anatomical and functional linkage has largely remained elusive. We established a transgenic mouse model in which the TS and BAC projection neurons were visualized by GFP fluorescence and selectively eliminated by immunotoxin-mediated cell targeting. The linkage between the TS/BAC and the MHb constitutes two parallel pathways composed of the TS-ventral MHb, the core part of the interpeduncular nucleus (IPN), and the BAC-dorsal MHb, the peripheral part of the IPN. Ablation of the TS and BAC projection neurons selectively impaired anxiety and enhanced fear responses and learning, respectively. Inputs from the TS and BAC to the MHb are thus segregated by two parallel pathways and play specialized roles in controlling emotional behaviors.

Figure 1. Specific Ablation of Cell Bodies and Their Projections in IT-Injected Transgenic Mice

GFP fluorescence in the region covering the TS, BAC, stria medullaris (sm) (A), TS (B), BAC and sm (C), and MHb (D) of IT-uninjected transgenic mice (tg) is shown for comparison with that in the IT-injected TS (F), IT-injected BAC (G), and the MHb following IT injection into the TS (H) and BAC (I) of transgenic mice (IT-tg). Note that the GFP-positive sm in the BAC-ablated mice (G) is derived from GFP-positive TS cells. GFP-positive glomerular structures are shown by magnifying the view of the MHb of the transgenic mouse (E). Scale bars, 1 mm (A); 500 μm (B and F); 200 μm (C, D, and G–I); 50 μm (E). Densities of GFP fluorescence in the TS (J), BAC (L), vMHb (K) and dMHb (M) were measured following IT injection into the TS (J and K) or BAC (L and M) of transgenic mice and compared with those of the corresponding IT-uninjected transgenic mice (n = 5 each). Columns and error bars represent the mean ± SEM, respectively; two-tailed paired Student’s t-test between IT-injected and uninjected transgenic mice, p < 0.0001. A restricted overlap of ChAT immunoreactivity (O and P) and SP hybridization signals (R and S) at the GFP-positive vMHb (N and P) and dMHb (Q and S) of transgenic mice, respectively. Scale bar, 200 μm. Separate projections of GFP-positive, cholinergic (T) and SPergic neurons (U) of the MHb into the IPN were identified by the VAChT and SP immunoreactivities, respectively. Scale bar, 200 μm. See also Figures S4 and S5.

Figure 2. Anterograde Tracing of the TS and BAC into the vMHb and dMHb, Respectively

AAV-CAG-mCherry (red) injected into the TS (A–C) and BAC (G–I) separately labeled the GFP-positive vMHb (D–F) and dMHb (J–L), respectively. Scale bars, 500 μm (A–C); 200 μm (D–L). See also Figures S1–S3.

Figure 3. Retrograde Tracing of the MHb into the TS and BAC

Unilateral injection of a retrograde tracer (red) into the MHb (A) showed prominent labeling at the ipsilateral side of the GFP-positive TS and BAC of transgenic mice (B–D). Arrows in C and D indicate false staining of ependymal cells lining the ventricles, which resulted from leakage of a tracer into the ventricles close to the MHb during removal of a tracer-filled syringe from the brain. Magnified views of labeling of the TS and BAC are shown in E–G and H–J, respectively. Scale bars, 500 μm (A); 1 mm (B–D); 200 μm (E–J). (K) Schematic drawing focusing on the TS-vMHb-cIPN and the BAC-dMHb-pIPN pathways. In the left drawing, solid arrows, predominant projections; dashed arrows, rare projections. See also Figures S1–S3.

Figure 4. Impairments of Anxiety-Related Behaviors of the TS-Ablated Mice, but not of Those of the BAC-Ablated Mice

(A and F) Total mobility and percentages of the distance traveled in the central square were measured for the TS-ablated (n = 9; A), BAC-ablated (n = 10; F) and IT-injected wt mice (n = 8; A or n = 9; F) in the open-field test. (B and G) In the elevated plus maze test, the total number of entries from a central square to the open and closed arms, percentages of entering an open arm, and time staying in the open arm were measured for the TS-ablated (n = 10; B) and BAC-ablated mice (n = 9; G) and compared with those for the IT-injected wt mice (n = 10; B or n = 8; G). (C and H) The number of buried marbles was measured for the TS-ablated (n = 10; C), BAC-ablated (n = 9; H), and IT-injected wt mice (n = 12; C or n = 9; H). Columns and bars represent the mean ± SEM, respectively. Statistical analysis between IT-injected transgenic and wt mice was conducted by the unpaired two-tailed Student’s t-test. *p < 0.05, **p < 0.01, n.s., not significant. (D and E) Typical examples of trajectories in the elevated plus maze test (D) and marble-burying test (E) are shown. In the elevated plus maze, dashed and solid lines indicate open and closed arms, respectively.

Figure 5. Impairments of Fear-directed Behaviors of the BAC-Ablated Mice, but not Those of the TS-Ablated Ones

(A) Percentages of freezing were determined for a 1-min period after giving electric shocks at 3, 4, and 5 min (arrows). Marks and bars represent the mean ± SEM, respectively; TS-ablated (n = 9), BAC-ablated (n = 9), and IT-injected wt mice (n = 8). Two-way ANOVA was conducted between time point and genotype, followed by Bonferroni post-hoc test; *p<0.05, ***p < 0.001. (B) Avoidance learning was tested by use of the one-trial inhibitory avoidance task performed by the TS-ablated (n = 9), BAC-ablated (n = 7), and IT-injected wt mice (n = 9 or 10). Columns and bars represent the mean ± SEM, respectively. Statistical analysis between IT-injected transgenic and wt mice was conducted by two-way ANOVA followed by Bonferroni post hoc test. *p < 0.05, n.s., not significant.