Characterization of Neurensin-2 knockout mice: insights into stress-resilience mechanisms

Abstract

Major depressive disorder (MDD) affects millions worldwide, yet its pathophysiology remains poorly understood. While some individuals are susceptible to developing depression, others show resilience that protects them from developing MDD. Understanding the resilience-associated mechanisms will likely result in novel therapies for MDD. We have recently reported that the vesicular protein Neurensin-2 mediates depression and that its deletion confers profound resilience to chronic stress. Nonetheless, the behavioral and molecular adaptations that underlie the stress resilience in Neurensin-2 knockout mice are still unknown. In this study, we aimed to comprehensively characterize the basal behavioral effects of Neurensin-2 deletion in mice. We used Neurensin-2 knockout male and female mice to examine how Neurensin-2 deletion affects cognitive, emotional, and motor performance in mice. In addition, we examined the impact of Neurensin-2 deletion on body weight, analyzed the stress-induced molecular changes, and tested how these changes affect the excitatory/inhibitory balance. We found that while Neurensin-2 deletion confers basal anxiolysis and weight reduction, no discernible cognitive, social, or motor impairments were detected. Furthermore, we found that Neurensin-2 knockout mice have impaired hippocampal inhibitory transmission, which is resilient to the stress-evoked excitatory/inhibitory imbalance seen in wild-type mice. Our findings suggest that Neurensin-2 deletion confers basal anxiolysis, and shifts the hippocampal excitatory/inhibitory balance. These effects are not accompanied by impaired cognitive function or weight gain. Thus, we suggest Neurensin-2 inhibition as an exciting potential strategy for developing treatments for depression and anxiety disorders as well as for promoting stress resilience.

Fig. 1. Study design to explore the resilience phenotype of Nrsn2-KO mice and anxiolytic baseline effect of Nrsn2 deletion.

a WT or Nrsn2-KO males were subjected to social defeat stress followed by the social interaction (SI) test. The interaction time of the SI test is presented. n = 22, 15. *p = 0.0445. IZ – interaction zone. b Scheme of the research plan to explore the resilience phenotype seen in Nrsn2-KO mice by high SI ratio. c, d WT or Nrsn2-KO male and female mice were subjected to behavioral tests to examine their anxiety like behavior c Open field (OF) test. n = 13–35/group *p = 0.0111, **p = 0.0076. d Light-dark test. n = 12–30 ****p < 0.0001, **p = 0.0050. Unpaired t-test.

Fig. 2. Nrsn2-KO mice have normal social, cognitive and motor behaviors.

a experimental scheme. WT or Nrsn2-KO male and female mice were subjected to social performance tests, cognitive tests, and motor assessments. b Three chamber sociability test. n = 13–30/group. p_males = 0.5523 p_Females = 0.4700 c Three chamber social novelty test. n = 12–30/group. p_males = 0.4261 p_Females = 0.8283 d Tube dominance test. n = 12–15/group. **p = 0.0082 p_Females > 0.999. e Y-maze test. n = 12–34. p_males = 0.5935 p_Females = 0.7005. f Object recognition test n = 13–35/group. p_males = 0.2279 p_Females = 0.8900 g Contextual fear conditioning test. n = 13–33/group. p_males = 0.3357 p_Females = 0.3611. h Distance traveled in the open field arena. n = 13–35/group. ****p < 0.0001 p_Females = 0.7894. Unpaired t-test; ns, not significant. The dashed grey line represents the 50% preference threshold.

Fig. 3. Nrsn2-KO mice have reduced body weight and fat mass.

Monitoring of body weight in a female and b male WT or Nrsn2-KO mice. n = 10–40/group; Repeated measures within subject. Two-way ANOVA. Females, Age F(15,517) = 287.4 p < 0.0001; Genotype F(1,83) = 43.16 p < 0.0001; interaction F(15,517) = 2.534 p = 0.0012. Males, Age F(15.00, 590.0) = 538.7 p < 0.0001; Genotype F(1, 91) = 8.426 p = 0.0046; interaction F(15, 590) = 2.922 p = 0.0002. Holm-Šídák post hoc method *p < 0.05, **p 0.01 to 0.001, ***p 0.001 to 0.0001, ****p < 0.0001. c fat, d lean, and e water percentages based on body MRI. n = 10–13/group. **p_fat = 0.0075 ***p_lean = 0.0005 ***p_water = 0.0001. Unpaired t-test. ns, not significant.

Fig. 4. Identification of stress-responsive pathways in WT and Nrsn2-KO mice.

a Scheme of the experiment. Hippocampi were harvested from stress-naïve (control) mice and from mice after ten days of social defeat stress. RNA was extracted, sequenced and analyzed to identify differentially expressed genes. n = 5–7 mice/group b Venn diagram summary of differentially expressed genes resulting from stress in WT mice (blue), Nrsn2-KO mice (yellow), and overlapping genes (green). c Heat map representation of transcripts that were differentially expressed in socially defeated WT (WTSD) vs. WT mice and socially defeated Nrsn2-KO (KOSD) vs. KO mice. d, g Volcano plots. Red and blue dots indicate significantly upregulated and downregulated genes by 1.5 fold, respectively. e, h gene ontology and f, i KEGG pathway analysis of differentially expressed genes in WT and KO mice following social defeat. Blue marked bars represent pathways with significant Benjamini values (<0.05). Grey line represents p-value significance threshold.

Fig. 5. Electrophysiological changes in Nrsn2-KO and socially defeated mice.

a Scheme of the experiment. WT or Nrsn2-KO male mice were subjected to chronic social defeat stress (CSDS) followed by the social interaction (SI) test. Acute hippocampal slices were obtained, and dentate gyrus granular cells were patched. b Representative traces of mIPSCs in granule cells from WT and KO stress-naïve (control) mice and after CSDS. c average amplitudes. d, e Bar graphs of mIPSC d amplitudes and e frequency in granule cells. n = 8–13 cells/group, patched from n = 3–5 mice/group. Two-way ANOVA, d Amplitude: Genotype F(1,33) = 1.222 p = 0.277; Stress F(1,33) = 2.746 p = 0.107; Interaction F(1,33) = 5.732 p = 0.0225. *p_adjusted = 0.0489 **p_adjusted = 0.0094. e Frequency: Genotype F(1,32) = 0.01415 p = 0.9061; Stress F(1,32) = 0.6933 p = 0.4112; Interaction, F(1,32) = 1.404 p = 0.2448 Holm-Šídák post hoc method.