Article Figures & Data
Figures
- Figure 1.
Timeline of the experimental protocols and immunohistochemical validation. A, Schematic timeline of the behavioral test battery. In case of VGluT3-Cre mice, experiments started 4 weeks after the stereotaxic surgery—which included 2 weeks of habituation to “reversed” day/night cycle—in order to ensure viral expression of DREADDs. Syringes indicate the occasions when VGluT3-Cre mice were injected intraperitoneally with CNO (1 mg/ml/kg) 30 min before the start of the experiment. Two to 3 d of break were kept between the experiments in order to eliminate the carry-over effect of CNO. WT and VGluT3 KO mice went under the same procedure, except for the surgery and CNO administration. B, Example of proper (left) and improper (right) viral expression in VGluT3-Cre mice. After the experiments, brains were collected from the animals, and Ni-DAB staining was done on midbrain slices against the RFP encoded in the AAV. Only the data of those animals that had proper viral expression in the MRR were included in the statistical analysis. On the right, there is an example of unacceptable staining in the DR. C, Representative figure of viral expression specificity in the MRR. Double fluorescent immunohistochemistry was done of midbrain slices against VGluT3 (Alexa-488, green) and RFP (Alexa-594, red) in control virus injected VGluT3-Cre mice. White arrows indicate example neurons where both markers were found. Scale bar, 100 μm. AAV, adeno-associated virus vector; Aq, Aqueduct cerebri; CNO, clozapine-N-oxide; DR, dorsal raphe; DREADD, designer receptor exclusively activated by designer drugs; KO, knock-out; MRR, median raphe region; Ni-DAB, nickel-3.3′-diaminobenzidine; RFP, red fluorescent protein; VGluT3, vesicular glutamate transporter 3; WT, wild-type.
- Figure 2.
Results of experiments on mobility and anxiety-like behavior. A, Schematic figure of sociability test, which started with a 5 min of OF test for mobility and anxiety-like behavior. Right after that, 5 min of object habituation followed when two identical wired cages were put into the arena. Right after that, one unfamiliar, juvenile conspecific was put under one of the wired cages for sociability phase. One day later, a 5 min SDT was conducted, during which the already familiar and a new unfamiliar conspecific were placed under the wired cages with interchanged positions compared with the previous day. B, Distance moved during OF. In dark, there was no difference between WT and VGluT3 KO mice (t test, NWT = 9; NKO = 11). On the other hand, inhibition of VGluT3+ neurons in the MRR resulted in decreased mobility (one-way ANOVA; Ncontrol = 8; Nexcitatory = 13; Ninhibitory = 14). C, Percentage of time in different zones during OF. Increased anxiety-like behavior (repeated-measures ANOVA) in VGluT3 KO mice is reflected by decreased time spent in the centrum (75% of the arena floor). D, Schematic representation of EPM apparatus. EPM consists of two closed arms and two open arms connected by a centrum area, elevated 67 cm above the ground. E, Open/total (%) is a locomotion independent parameter that reflects anxiety-like behavior on EPM. Increased anxiety-like behavior is reflected by decreased open/total (%) in VGluT3 KO mice (t test; NWT = 10; NKO = 11). On the other hand, the excitation of MRR-VGluT3+ neurons elicited anxiolytic effect (one-way ANOVA; Ncontrol = 8; Nexcitatory = 15; Ninhibitory = 15). Data are expressed as average ± SEM. Empty squares represent individual values. For more details see Extended Data Tables to Figure 2. EPM, elevated plus maze test; KO, knock-out; MRR, median raphe region; OF, open field test; VGluT3, vesicular glutamate transporter type 3; WT, wild-type. **p < 0.01 versus WT; (=)p = 0.08, ==p < 0.01 versus control; @p < 0.05 versus excitatory.
- Figure 3.
Results of experiments on social behavior. A, Percentage of time spent with empty cages and the cage containing an unfamiliar, juvenile conspecific. Both WT and VGluT3 KO mice preferred the social stimuli compared with the inanimate object, but KO animals spent more time with the conspecific (repeated-measures ANOVA; NWT = 10; NKO = 11). Similarly, all animals in the VGluT3-Cre groups preferred the social stimuli, but the excitatory group had decreased interest compared with the inhibitory group (repeated-measures ANOVA; Ncontrol = 8; Nexcitatory = 11; Ninhibitory = 15). B, Main differences between groups in SIT. VGluT3 KO mice exhibited more aggressive behavior compared with the WT animals (t test; NWT = 10; NKO = 11). On the other hand, friendly social behavior was only marginally changed in the VGluT3-Cre mice (one-way ANOVA; Ncontrol = 8; Nexcitatory = 13; Ninhibitory = 15). C, Schematic representation of SIT protocol: for 10 min two experimental animals from the same treatment group (e.g., KO vs KO, excitatory vs excitatory) was placed into a neutral, anxiogenic environment to freely interact. Under the figure Gantt diagrams visually represent the behavior of all experimental animals during SIT. Different behaviors are represented by different colors. D, Schematic representation of RIT protocol: for 10 min an unfamiliar, physically smaller, but adult conspecific was placed into the home cage of the experimental mouse to freely interact. Under the figure Gantt diagrams visually represent the behavior of all experimental animals during RIT. Different behaviors are represented by different colors. E, Frequency of aggressive behavior did not differ between WT and VGluT3 KO mice during RIT (t test; NWT = 10; NKO = 11). F, Frequency of friendly social behavior was decreased in the inhibitory group of VGluT3-Cre mice in RIT (one-way ANOVA; Ncontrol = 8; Nexcitatory = 13; Ninhibitory = 14). Data are expressed as average ± SEM. Empty squares represent individual values. For more details see Extended Data Tables to Figure 3. KO, knock-out; RIT, resident intruder test; SIT, social interaction test; VGluT3, vesicular glutamate transporter type 3; WT, wild-type. ##p < 0.01 versus mouse; **p < 0.01 versus WT; ==p < 0.01 versus control; @p < 0.05 versus excitatory; –p < 0.01 versus inhibitory.
- Figure 4.
Results of social and working memory experiments. A, SDT conducted 1 d after sociability test (Fig. 2A) showed a preference for the familiar, “old” mouse in VGluT3 KO mice (repeated-measures ANOVA; NWT = 10; NKO = 11). B, In SDT for VGluT3-Cre mice, there were no differences between the groups regarding their social memory (repeated-measures ANOVA), but the cumulative time spent with social stimuli showed an increased social interest in the inhibitory group (one-way ANOVA; Ncontrol = 8; Nexcitatory = 12; Ninhibitory = 15). C, Schematic figure of Y-maze test. Three identical arms placed perpendicular can be freely explored by the experimental animal for 5 min. Proper working memory is reflected by consecutive, different arm entries, called spontaneous alternation. D, All groups had intact working memory compared with random 50% except for the VGluT3-Cre control group (single sample t test; NWT = 10; NKO = 11; Ncontrol = 8; Nexcitatory = 13; Ninhibitory = 14). Data are expressed as average ± SEM. Empty squares represent individual values. For more details see Extended Data Tables to Figure 4. KO, knock-out; SDT, social discrimination test; VGluT3, vesicular glutamate transporter type 3; WT, wild-type. ##p < 0.01 versus “old” mouse; $$p < 0.01 versus random 50; ==p < 0.01 versus control; @@p < 0.01 versus excitatory.
- Figure 5.
c-Fos staining in the mouse brain and VGluT3 mRNA expression in healthy human samples. A–D, There are decreased numbers of c-Fos-positive cells in the KO (B,D) compared with WT (A,C) mice in the mPFC (A,B) and MS (C,D) after SIT (NWT = 5; NKO = 8). E,F, Functional DREADD excitation is reflected by increased c-Fos activity in the MRR of VGluT3-Cre mice (Ncontrol = 6; Nexcitatory = 3). Scale bar, 100 μm. G, Relative expression of VGluT3 mRNA to the measured housekeeping gene (GAPDH) in the investigated brainstem areas (LPB: N = 3; PRF: N = 10; PRN: N = 6; RN: N = 3). H, After RT-qPCR, samples containing VGluT3 cDNA were run with gel electrophoresis and sequencing (BIOMI) to validate the specificity of primers. Brain samples from pontine raphe nuclei, which contain the human equivalent of mouse MRR, all expressed VGluT3 (N = 6). I, Different cortical areas of healthy human samples were used as negative controls (N = 6). Data are expressed as average ± SEM. Empty squares represent individual values. For more details see Extended Data Tables to Figure 5. bp, base pair; DREADD, designer receptors exclusively activated by designer drugs; KO, knock-out; LPB, lateral parabrachial nucleus; mPFC, medial prefrontal cortex; MRR, median raphe region; MS, medial septum; PRF, pontine reticular formation; PRN, pontine raphe nucleus; RN, midbrain raphe nuclei; RT-qPCR, reverse transcription quantitative polymerase chain reaction; SIT, social interaction test; VGluT3, vesicular glutamate transporter type 3; WT, wild-type.
Tables
Experiment Knock-out MRR excitation MRR inhibition Mobility OF Ø Ø ↓ EPM Ø Ø Ø Y-maze Ø Ø Ø Anxiety OF ↑ Ø Ø EPM ↑ ↓ Ø Social behavior Social interest ↑ (sociability, tendency in SDT) ↓ (indirect during sociability) ↑ (in SDT) Friendly behavior Ø Ø ↓ in aggressive context (RIT) Aggressive behavior ↑ in anxiogenic context (SIT) Ø Ø Memory SDT ↓ 24 h SDT Ø Ø Y-maze Ø Ø Ø c-Fos activity ↓ in anterior cingulate, infralimbic cortex and medial septum ↑ in the MRR - The results are presented in comparison with their respective control groups: that is, VGluT3 wild-type littermates for knock-out mice and control group (expressing only RFP in the VGluT3+ neurons in the MRR) for the excitatory (Gq DREADD + RFP) and inhibitory (Gi DREADD + RFP) chemogenetically manipulated groups. Ø, no difference compared with respective control group; -, not applicable; DREADD, designer receptors exclusively activated by designer drugs; EPM, elevated plus maze; MRR, median raphe region; OF, open field; RFP, red fluorescent protein; RIT, resident intruder test; SDT, social discrimination test; SIT, social interaction test; VGluT3, vesicular glutamate transporter 3.
Figure 2-1
Results of open field test - VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-tests was 18. Data are expressed as mean ± SEM. WT: wild-type; KO: knock-out. ** p < 0.01 vs WT. Download Figure 2-1, DOCX file.
Figure 2-2
Results of elevated plus-maze test - VGluT3 WT-KO animals. WT-KO: Degree of freedom (df) for the two-sample t-test (all parameters) is 19. Marginal effects are in brackets (). Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out; RA: risk assessment; SAP: stretched attend posture. * p < 0.05 vs WT. Download Figure 2-2, DOCX file.
Figure 2-3
Results of open field test - VGluT3-Cre animals. Degree of freedom (df) for the one-way ANOVA was (2,32). Marginal effects are in brackets (). Data are expressed as mean ± SEM. Download Figure 2-3, DOCX file.
Figure 2-4
Results of elevated plus-maze test – VGluT3-Cre animals. Degree of freedom (df) for the one-way ANOVA for the frequency and time (%) spent in different zones is (2,32), while for the risk assessment behaviour is (2,31). Marginal effects are in brackets (). Data are expressed in mean ± SEM. RA: risk assessment; SAP: stretched attend posture. == p < 0.01 vs control; & p < 0.05 vs inhibitory Download Figure 2-4, DOCX file.
Figure 3-1
Results of sociability test – object habituation phase – VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-test (frequency and time [%] of ‘other’ behaviour) was 17. Degree of freedom in the repeated-measures ANOVA (frequency and time [%] of left vs right cage) was (1,17) for all effects. Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out. * p < 0.05 vs WT. Download Figure 3-1, DOCX file.
Figure 3-2
Results of sociability test -sociability phase – VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-test (frequency and time [%] of ‘other’ behaviour; SI) is 18. Degree of freedom in the repeated-measures ANOVA (frequency and time [%] of left vs right cage) is (1,18) for all effects. Marginal effects are in brackets (). Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out; SI: sociability index. ## p < 0.01 vs cage; ** p < 0.01, * p < 0.05 vs WT; $$ p < 0.01 vs random 50. Download Figure 3-2, DOCX file.
Figure 3-3
Results of social interaction test – VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-test (all parameters) is 19. Marginal effects are in brackets (). Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out. * p < 0.05 vs WT. Download Figure 3-3, DOCX file.
Figure 3-4
VGluT3 WT-KO contingency table. The table shows the numbers of animals who exhibited aggressive behaviour in social interaction test, followed by a Pearson Chi-square test. WT: wild-type; KO: knock-out. Download Figure 3-4, DOCX file.
Figure 3-5
Results of resident intruder test – VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-test (all parameters) is 19. Marginal effects are in brackets (). Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out. == p < 0.01 vs control. Download Figure 3-5, DOCX file.
Figure 3-6
Results of sociability test – object habituation phase – VGluT3-Cre animals. Degree of freedom (df) for the one-way ANOVA (frequency and time [%] of ‘other’ behaviour) was (2,31). Degree of freedom in the repeated-measures ANOVA (frequency and time [%] of left vs right) is (2,31) for the effect of manipulation and manipulation × choice interaction, while (1,32) for the effect of choice. Marginal effects are in brackets (). Data are expressed in mean ± SEM. Download Figure 3-6, DOCX file.
Figure 3-7
Results of sociability test – sociability phase – VGluT3-Cre animals. Degree of freedom (df) for the one-way ANOVA (frequency and time [%] of ‘other’ behaviour) is (2,32). Degree of freedom in the repeated-measures ANOVA (frequency and time [%] of mouse vs cage) is (2,32) for the effect of manipulation and manipulation × choice interaction, while (1,32) for the effect of choice. Data are expressed in mean ± SEM. SI: sociability index. ## p < 0.01 vs cage; - p < 0.05 vs inhibitory; $ p < 0.05 vs random 50. Download Figure 3-7, DOCX file.
Figure 3-8
Results of social interaction test – VGlut3-Cre animals. Degree of freedom (df) for the one-way ANOVA (all parameters) is (2,32). Marginal effects are in brackets (). Data are expressed in mean ± SEM. = p < 0.05 vs control; @ p < 0.05 vs excitatory Download Figure 3-8, DOCX file.
Figure 3-9
Results of resident intruder test – VGluT3-Cre animals. VGluT3-Cre: Degree of freedom (df) for the one-way ANOVA (all parameters) is (2,31). Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out. == p < 0.01 vs control; @ p < 0.05 vs excitatory. Download Figure 3-9, DOCX file.
Figure 4-1
Results of social discrimination test – VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-test (frequency and time [%] of ‘other’ behaviour; SI) is 18. Degree of freedom in the repeated-measures ANOVA (frequency and time [%] of ‘old’ vs ‘right’ mouse) is (1,18) for all effects. Marginal effects are in brackets (). Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out; SD: social discrimination index. * p < 0.05 vs WT; # p < 0.05 vs cage; $ p < 0.05 vs random 0. Download Figure 4-1, DOCX file.
Figure 4-2
Results of Y-maze test – VGluT3 WT-KO animals. Degree of freedom (df) for the two-sample t-test for locomotion is 19, for alteration is 16. Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out. $$ p < 0.01 vs random 50. Download Figure 4-2, DOCX file.
Figure 4-3
Results of social discrimination test – VGluT3-Cre animals. Degree of freedom (df) for the one-way ANOVA (frequency and time [%] of ‘other’ behaviour) is (2,32). Degree of freedom in the repeated-measures ANOVA (frequency and time [%] of mouse vs cage) is (2,32) for the effect of manipulation and manipulation × choice interaction, while (1,32) for the effect of choice. Marginal effects are in brackets (). Data are expressed in mean ± SEM. SD: social discrimination index. # p < 0.05 vs cage; == p < 0.01 vs control; @@ p < 0.01 vs excitatory. $ p < 0.05 vs random 0. Download Figure 4-3, DOCX file.
Figure 4-4
Results of Y-maze test – VGluT3-Cre animals. Degree of freedom (df) for the one-way ANOVA for locomotion is (2,32), while for alteration is (2,31). Data are expressed in mean ± SEM. $$ p < 0.01 vs random 50. Download Figure 4-4, DOCX file.
Figure 5-1
Results of c-Fos positive cell counting – VGluT3 WT-KO animals. VGluT3 KO mice showed decreased activity after social test in their anterior cingulate, infralimbic cortex and medial septum. Degree of freedom (df) for the two-sample t-test is 11. Benjamini-Hochberg (FDR, false discovery rate) post-hoc correction was used, thus a p-value was considered significant from 0.016. Data are expressed in mean ± SEM. WT: wild-type; KO: knock-out. * p < 0.05, ** p < 0.01 vs WT Download Figure 5-1, DOCX file.
Figure 5-2
Results of c-Fos positive cell counting – VGluT3-Cre animals. The excitatory group showed increased neuronal activity only in the viral injected target area, the median raphe region. Degree of freedom (df) for the two-sample t-test is 7. Benjamini-Hochberg (FDR, false discovery rate) post-hoc correction was used, thus a p-value was considered significant from 0.008. Data are expressed in mean ± SEM. ## p < 0.01 vs Control. Download Figure 5-2, DOCX file.
Figure 5-3
Subject information for the RT-PCR measurements in the brainstem of humans. Download Figure 5-3, DOCX file.
Figure 5-4
RT-PCR results of human samples. Four major brainstem nuclei were investigated, out of which all showed VGluT3 expression on an mRNA level. The results show averaged and normalized CT values. VGluT3: vesicular glutamate transporter type 3. Download Figure 5-4, DOCX file.
In this issue
