The Slack Channel Regulates Anxiety-Like Behaviors via Basolateral Amygdala Glutamatergic Projections to Ventral Hippocampus

Abstract

Anxiety disorders are a series of mental disorders characterized by anxiety and fear, but the molecular basis of these disorders remains unclear. In the present study, we find that the global Slack KO male mice exhibit anxious behaviors, whereas the Slack Y777H male mice manifest anxiolytic behaviors. The expression of Slack channels is rich in basolateral amygdala (BLA) glutamatergic neurons and downregulated in chronic corticosterone-treated mice. In addition, electrophysiological data show enhanced excitability of BLA glutamatergic neurons in the Slack KO mice and decreased excitability of these neurons in the Slack Y777H mice. Furthermore, the Slack channel deletion in BLA glutamatergic neurons is sufficient to result in enhanced avoidance behaviors, whereas Kcnt1 gene expression in the BLA or BLA-ventral hippocampus (vHPC) glutamatergic projections reverses anxious behaviors of the Slack KO mice. Our study identifies the role of the Slack channel in controlling anxious behaviors by decreasing the excitability of BLA-vHPC glutamatergic projections, providing a potential target for anxiolytic therapies.SIGNIFICANCE STATEMENT Anxiety disorders are a series of mental disorders characterized by anxiety and fear, but the molecular basis of these disorders remains unclear. Here, we examined the behaviors of loss- and gain-of-function of Slack channel mice in elevated plus maze and open field tests and found the anxiolytic role of the Slack channel. By altering the Slack channel expression in the specific neuronal circuit, we demonstrated that the Slack channel played its anxiolytic role by decreasing the excitability of BLA-vHPC glutamatergic projections. Our data reveal the role of the Slack channel in the regulation of anxiety, which may provide a potential molecular target for anxiolytic therapies.

Keywords: Slack channel; anxiety; basolateral amygdala; corticosterone; ventral hippocampus.

Figure 1.

Slack KO mice exhibit enhanced avoidance behaviors but with normal depression-related behaviors and normal fear memory. A, Schematic diagram of Slack knock-out strategy. B, PCR verification of targeted axons deletion in Slack KO mice. C–J, Slack KO mice display enhanced avoidance behaviors in EPM test (C–F) and OF test (G–J). Representative trajectories of WT mice (left) and Slack KO mice (right) in EPM exploration (C). Time spent in the anxiogenic open arms (D). Probability of open-arm entry (E). Total distance traveled in EPM (F). Two-tailed unpaired t test, n = 8, *p < 0.05 (D); n = 8, **p < 0.01 (E); n = 8, n.s. (F). Representative trajectories of OF exploration (G). Time spent in the OF center (H). Number of entries into the OF center (I). Velocity traveled in OF (J). Two-tail unpaired t test, n = 8, ***p < 0.001 (H); n = 8, ***p < 0.001 (I); n = 8, n.s. (J). Intraperitoneal injection of DZP (1 mg/kg) reduces avoidance behaviors of Slack KO mice in the EPM test (K–M) and light-dark box test (M, O). Time spent in the anxiogenic open arms (K). Probability of open-arm entry (L). Representative trajectories of EPM exploration (M). One-way ANOVA in K, L; n = 10–12, *p < 0.05, **p < 0.01. N, Latency to the first entry to the light box. O, Time spent in the dark box. One-way ANOVA (N, O), n = 10–12, *p < 0.01, **p < 0.05, ***p < 0.001. P–S, The Slack KO mice did not show significantly different depressive behavior from WT mice in FST (P), SPT 2 h and 24 h (Q, R), and TST (S). T, U, The Slack channel knock-out mice show normal contextual fear learning (T) or cued fear learning (U). Two-tail unpaired t test, n = 13–22 (P–U). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 2.

The Slack-Y777H gain-of-function mutant mice exhibit anxiolytic behaviors. A, Scheme of generating the Slack Y777H mutant strategy. B–H, Slack Y777H gain-of-function mutant mice exhibit anxiolytic behaviors in the EPM test (B–E) and OF test (B, F–H). Representative trajectories of EPM and OF exploration (B). Time spent in the open arms (C). Probability of open-arm entry (D). Total distance traveled in EPM (E). Two-tail unpaired t test, n = 8, *p < 0.05 (C); n = 8, **p < 0.01 (D); n = 8, n.s. (E). Time spent in the OF center (F). Number of entries into the center (G). Moving velocity (H). Two-tail unpaired t test, n = 8, *p < 0.05 (F); n = 8, **p < 0.01 (G); n = 8, n.s. (H). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 3.

Chronic CORT treatment decreases Slack expression in the BLA. A, Timeline of measuring body weight, chronic CORT exposure, and anxiety-like behavioral test. B, Body weight of mice before (day 0) and after (day 21) chronic CORT exposure. C, The Kcnt1 mRNA expression levels in different brain regions of vehicle-treated mice and CORT mice, normalized to the BLA Kcnt1 mRNA expression level in vehicle-treated mice. The two-tailed unpaired t test, n = 8 for Vehicle and n = 6 for CORT, n.s., ***p < 0.001 (B); n = 3, *p < 0.05 (C). D, Schematic diagram of Kcnt1 mRNA expression in anxiety-related brain regions. The number of green plots presents the expression level of Kcnt1 mRNA in the vehicle group. E, The Slack protein expression in the BLA of vehicle-treated mice and CORT mice; two-tailed unpaired t test, n = 4, **p < 0.01. F–M, Chronic CORT treatment increased avoidance behaviors of mice in the EPM test (F–I) and OF test (J–M). Representative trajectories of EPM exploration (F). Time spent in the open arms (G). Probability of open-arm entry (H). Total distance traveled in EPM (I). Representative traces trajectories of OF exploration (J). Time spent in the center (K). Number of entries into the center (L). Moving velocity (M); two-tailed unpaired t test, n = 7–8 for Vehicle and n = 6 for CORT mice, *p < 0.05 (G),*p < 0.05 (H), n.s. (I), **p < 0.01 (K), *p <0.05 (L), n.s. (M). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 4.

Compensatory expression of Slack channel in BLA neurons rescues the enhanced avoidance behaviors of the global Slack knock-out mice. A, Timeline of AAV injection, EPM test, scarification, and verification. B, Schematic diagram of a compensatory expression of the Slack channel by injection adeno-associated viral that carried the Flag-tagged Kcnt1 gene in the BLA of Slack KO mice (middle), Western blot bands (left) and fluorescent images confirmed the successful expression of the Slack channel in the BLA (right). Scale bar, 200 μm. C, Representative trajectories of EPM exploration. D, E, The Slack channel expression in BLA neurons decreased avoidance behaviors of Slack KO mice in the EPM test, shown as time spent in open arms (D) and probability of open-arm entry (E). Two-tailed unpaired t test, n = 8, *p < 0.05 (D); n = 8, ***p < 0.001 (E). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1.

Figure 5.

The Slack channel expresses in VGLUT-positive glutamatergic neurons in the basolateral amygdala and alters neural activities. A, Fluorescent images of in situ hybridization by RNA scope showing the distribution of Kcnt1 (red), VGLUT (cyan), VGAT mRNA (green), and DAPI (blue) in the BLA. Scale bar, 200 μm. B, Percentage of Kcnt1 mRNA-positive neurons that express VGLUT and VGAT. C, Percentage of VGLUT-positive (left) and VGAT-positive (right) neurons that express KCNT1 mRNA. D, Representative traces show the effect of injected current (−30 to 80 pA) induced action potentials in BLA CaMKII⁺ neurons from WT mice (left), Slack KO mice (middle), and Slack gain-of-function mutant (Y777H) mice (right). E, Sample patch of brain slice in mice with CaMKII⁺ neurons marked by mCherry fluorescence 21 d after injection of AAV-CaMKII-Cre-mCherry in the BLA. F, Numbers of APs induced by step current injection (−30 to 150 pA) in the CaMKII⁺ neurons of the WT, Slack KO, and Y777H mutant mice. Data were analyzed by two-way ANOVA; 18 cells for WT, 9 cells for KO, and 16 cells for Y777H from three to five mice in each genotype; *p < 0.05, **p < 0.01, ***p < 0.001. G, Average resting potentials in CaMKII⁺ BLA neurons of WT, Slack KO, and Y777H mutant mice. Data analyzed by one-way ANOVA, 16 cells for WT, 9 cells for KO, and 16 cells for Y777H, from three to five mice in each group, *p < 0.05. H, Average AHP in these neurons from WT, Slack KO, and Y777H mutant mice. One-way ANOVA, 16 cells for WT, 9 cells for KO, and 16 cells for Y777H from three to five mice each (G, H). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1.

Figure 6.

Specifically compensatory expression of the Slack channel in BLA glutamatergic neurons rescues the enhanced avoidance behaviors of the Slack knock-out mice. A, Timeline of AAV injection, behavioral tests, whole-cell recording, scarification, and verification. B, Schematic diagram of the specific compensatory expression of the Slack channel in the BLA CaMKII⁺ neurons of Slack channel knock-out mice. C, Left, Confocal images confirmed the mCherry fluorescence specifically expressed in the BLA. Scale bar, 100 μm. Right, Western blot confirmation of the Slack channel protein compensatory in the BLA 4 weeks after viral injection. D–K, Compensatory expression of the Slack channel in BLA glutamatergic neurons reverses the enhanced avoidance behaviors of Slack channel KO mice in EPM (D–G) and OF (H–K) tests. Representative trajectories of EPM exploration (D). Time spent in the anxiogenic open arms (E). Probability of open-arm entry (F). Total distance traveled in EPM (G). Representative trajectories of OF exploration (H). Time spent in the anxiogenic center (I). Number of entries into the center (J). Mice moving velocity (K). Two-tailed unpaired t test, n = 7, *p < 0.05 (E), *p < 0.05 (F), n.s. (G), *p < 0.05 (I), *p < 0.05 (J), n.s. (K). L–Q, The Slack knock-out mice with compensatory expression of the Slack channel in BLA do not exhibit significantly different depressive behavior from the mice with mCherry expression in FST (L), SPT 2 h, and 24 h (M, N) and TST (O). P–Q, Compensated expression of the Slack channel in glutamatergic neurons does not affect contextual fear learning (P) or cued fear learning (Q). Two-tailed unpaired t test, n = 7 (L–Q). R, Representative traces of action potentials in the BLA CaMKII⁺ neurons from Slack KO (mCherry group) and specifically compensatory (Kcnt1 group) mice in response to 90 pA current pulse lasting for 1 s. Numbers of action potentials induced by step current injection (−30 to 150 pA) in the BLA CaMKII+ neurons of the mCherry and Kcnt1 group mice (S). Data in S were analyzed by two-way ANOVA; 10 cells from three to five mice in both groups, *p < 0.05, **p < 0.01, ***p < 0.001. Average resting potentials of the BLA CaMKII⁺ neurons in mCherry and Kcnt1 group mice (T). Data analyzed by two-tailed unpaired t test, 10 cells from three to five mice in both groups, **p < 0.01. Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 7.

Conditional knock out of the Slack channel in BLA glutamatergic neurons increases the enhanced avoidance behaviors of the Slack CKO mice. A, The schematic diagram of the conditional knock-out strategy of the Kcnt1 gene. B, PCR verification of loxp insertion on target gene transcription in Slack CKO mice. The 271 bp fragment band is generated from WT alleles, and the 374 bp fragment band is the expected PCR result of the Slack CKO genotype. C, Timeline of AAV injection, OF and EPM tests, scarification, and verification. D, Schematic diagram of injection of AAV-CaMKII-Cre or AAV-CaMKII-mCherry into BLA in Slack CKO mice. E, Confocal fluorescent images confirm the expression of mCherry the BLA. A schematic of the viral expression for other subjects is provided in Extended Data Figure 7–1A. Scale bar, 100 μm. F, G, Quantification of knock down of Slack channels in mRNA (F) and protein levels (G) in the BLA of Slack CKO mice 3 weeks after AAV-CaMKII-Cre or AAV-CaMKII-mCherry injection. Two-tailed unpaired t test, n = 4, **p < 0.01 (F); n = 4, ***p < 0.001 (G). H–O, Conditional knock out of the Slack channel in BLA glutamatergic projection neurons increased avoidance behaviors of mice in the EPM (H–K) and OF (L–O) test. Representative traces of EPM exploration (H). Time spent in the anxiogenic open arms (I). Probability of open-arm entry (J). Total distance traveled in EPM (K). Representative trajectories of OF exploration (L). Time spent in the anxiogenic center (M). Number of entries into the center (N). Moving velocity (O). Two-tailed unpaired t test; n = 8, **p < 0.01 (I), **p < 0.01 (J), n.s. (K), **p < 0.01 (M), *p < 0.05 (N), n.s. (O). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 8.

Expression of the Slack channel in BLA–vHPC glutamatergic projection neurons relieves the enhanced avoidance behaviors of the Slack KO and CORT treatment mice. A, Timeline of AAV injection, anxiety-like behavioral test, and verification. B, Schematic diagram of injection of AAV-DIO-Kcnt1 or AAV-DIO-mCherry into BLA and AAV-Retro-CaMKII-Cre mixed with AAV-CaMKII-GFP into vHPC in Slack KO mice. C, Confocal fluorescent images confirmed the expression of GFP and mCherry in the vHPC or BLA, respectively. A schematic of the viral expression for other subjects is provided in Extended Data Figure 7-1B. Scale bar, 200 μm. D–K, Slack channel expressed in BLA–vHPC glutamatergic projection neurons reduces avoidance behaviors of Slack KO mice in EPM (D–G) and OF (H–K) test. Representative traces of EPM exploration (D). Time spent in the anxiogenic open arms (E). Probability of open-arm entry (F). Total distance traveled in EPM (G). Representative trajectories of OF exploration (H). Time spent in the anxiogenic center (I). Number of entries into the center (J). Moving velocity (K). Two-tailed unpaired t test; n = 6. *p < 0.05 (E), *p < 0.05 (F), n.s. (G), *p < 0.05 (I), *p < 0.05 (J), n.s. (K). L, Timeline of AAV injection, chronic CORT exposure, avoidance behavioral test, and verification. M, Scheme of injection of AAV-DIO-Kcnt1 or AAV-DIO-mCherry into BLA and AAV-Retro-CaMKII-Cre mixed with AAV-CaMKII-GFP into vHPC in Slack KO mice. N–U, Overexpression of Slack channel in BLA–vHPC glutamatergic projection neurons reduces chronic CORT treatment-induced enhanced avoidance behavior in EPM (M–Q) and OF (R–U) test. Representative traces of EPM exploration (N). Time spent in the anxiogenic open arms (O). Probability of open-arm entry (P). Total distance traveled in EPM (Q). Representative traces of OF exploration (R). Time spent in the anxiogenic center (S). Number of entries into the center (T). Moving velocity (U). Two-tailed unpaired t test; n = 8, *p < 0.05 (O), *p < 0.05 (P), n.s. (Q), *p < 0.05 (S), **p < 0.01 (T), n.s. (U). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 9.

The expression of the Slack channel in BLA–dmPFC glutamatergic projection neurons does not alter avoidance behaviors or fear memory of the Slack KO mice. A, Timeline of AAV injection, avoidance behavioral tests, fear conditioning tests, and verification. B, Schematic diagram of injection of AAV-DIO-Kcnt1 or AAV-DIO-mCherry into BLA and AAV-Retro-CaMKII-Cre mixed with AAV-CaMKII-GFP into dmPFC in Slack KO mice. C, Confocal fluorescent images confirmed the expression of GFP and mCherry in the dmPFC or BLA, respectively. A schematic of the viral expression for other subjects is provided in Extended Data Figure 7-1C. Scale bar, 200 μm. D–K, Slack channel expressed in BLA–dmPFC glutamatergic projection neurons does not alter avoidance behaviors of Slack KO mice in EPM (D–G) and OF (H–K) tests. Representative traces of EPM exploration (D). Time spent in the anxiogenic open arms (E). Probability of open-arm entry (F). Total distance traveled in EPM (G). Representative trajectories of OF exploration (H). Time spent in the anxiogenic center (I). Number of entries into the center (J). Moving velocity (K). L, M, Compensate expression of the Slack channel in BLA–dmPFC glutamatergic neurons does not affect contextual fear learning (L) or cued fear learning (M). Two-tailed unpaired t test, n = 7, n.s. (E–G, I–M). Data are presented as mean ± SEM. The details of the data are provided in Extended Data Figure 1-1. n.s., Not significant.

Figure 10.

Graphical abstract. Schematic showing the role of Slack channels in the regulation of anxiety-like behaviors via basolateral amygdala glutamatergic inputs to the ventral hippocampus.