Planar cell polarity proteins mediate ketamine-induced restoration of glutamatergic synapses in prefrontal cortical neurons in a mouse model for chronic stress

Abstract

Single administration of low-dose ketamine has both acute and sustained anti-depressant effects. Sustained effect is associated with restoration of glutamatergic synapses in medial prefrontal cortic (mFPC) neurons. Ketamine induced profound changes in a number of molecular pathways in a mouse model for chronic stress. Cell-cell communication analyses predicted that planar-cell-polarity (PCP) signaling was decreased after chronic administration of corticosterone but increased following ketamine administration in most of the excitatory neurons. Similar decrease of PCP signaling in excitatory neurons was predicted in dorsolateral prefrontal cortical (dl-PFC) neurons of patients with major depressive disorder (MDD). We showed that the basolateral amygdala (BLA)-projecting infralimbic prefrontal cortex (IL PFC) neurons regulate immobility time in the tail suspension test and food consumption. Conditionally knocking out Celsr2 and Celsr3 or Prickle2 in the BLA-projecting IL PFC neurons abolished ketamine-induced synapse restoration and behavioral remission. Therefore, PCP proteins in IL PFC-BLA neurons mediate synapse restoration induced by of low-dose ketamine.

Fig. 1. Single-cell transcriptomic and cell–cell communication analyses.

a Schematics of experimental design. The illustrations were created with BioRender. b UMAP plot of 17 Identified Cell Types. c Table of the top rankding pathways identified with CellChat that are inversely correlated between corticosterone-treated animals followed by ketamine treatment vs cortiosterone-treated animals without ketamine treatment. d CellChat diagrams made with a modified ChordDiagram where blue indicates reduced expression level in corticosterone ketamine treated mice and red indicates increased expression level in corticosterone ketamine treated mice. e, f Immunostaining of IL-1ß in microglia [F (2, 6) = 5.474, P = 0.044]. g, h Immunostaining of NFkB in microglia [F (2, 6) = 6.876, P = 0.028]. Each column represents the mean + S.E.M. of 3 animals. Statistical analysis was performed by one-way ANOVA followed by Fisher’s LSD. *p < 0.05 compared with the control group (BCD) treated with saline, and #p < 0.05 compared with the CORT treated with saline, Scale bar 50 µm.

Fig. 2. Changes of PCP signaling inferred by cell-cell communication analyses.

a CellChat Diagrams of PCP signaling pathway made with a modified ChordDiagram where blue indicates reduced expression level and red indicates increased expression level in mice. b CellChat Diagrams of PCP signaling pathway made with a modified ChordDiagram where blue indicates reduced expression level in MDD patients and red indicates increased expression level in MDD patients.

Fig. 3. IL PFC neurons projecting to the BLA control immobility time in the tail suspension test and food consumption in the food consumption test.

a Representative micrographs validating viral injections (repeated at least twice). The illustrations were created with BioRender. b Schematics of experimental design. The illustrations were created with BioRender. c Representative heatmaps of mouse movement in the TST. d Quantification of immobility time in the TST in (c) [T = 2.35, p = 0.038], mCh control n = 7, hM4Di-mCh n = 6. e Representative heatmaps of the open field test. f Quantification of locomotion in the open field test in (e) [T = 0.13, p = 0.89], mCh control n = 7, hM4Di-mCh n = 6. g Quantification of food consumption in the food consumption test [T = 5.92, p < 0.0001], mCh control n = 7, hM4Di-mCh n = 6. h Representative heatmaps of mouse movement in the TST. i Quantification of immobility time in the TST in (h) [T = 0.67, p = 0.51], mCh control n = 8, hM4Di-mCh n = 7. j Representative heatmaps of the open field test. k Quantification of locomotion in the open field test in (j) [T = 1.15, p = 0.27], mCh control n = 8, hM4Di-mCh n = 7. l Quantification of food consumption in the food consumption test [T = 0.61, p = 0.55], mCh control n = 8, hM4Di-mCh n = 7. Each column represents the mean + S.E.M. of n animals provided above. Statistical analysis was performed by Student’s t-test two-tailed. *p < 0.05 and ****p < 0.0001 compared with the control group (mCh).

Fig. 4. Verification of PCP gene expression changes by RNAscope.

a Schematics illustrating the experimental design. The illustrations were created with BioRender. b, e, h Representative micrographs of RNAscope. c Quantification of Celsr2 expression in excitatory neurons [F (2, 24) = 9.518, P = 0.0009], BCD/Sal control n = 10, CORT/Sal n = 8, CORT/Ket n = 9. d Quantification of Celsr3 expression in excitatory neurons [F (2, 21) = 11.36, P = 0.0005], BCD/Sal control n = 8, CORT/Sal n = 7, CORT/Ket n = 9. f Quantification of Celsr2 in inhibitory neurons [F (2, 30) = 5.829, P = 0.0073], BCD/Sal control n = 10, CORT/Sal n = 10, CORT/Ket n = 13. g Quantification of Celsr3 in inhibitory neurons [F (2, 24) = 11.88, P = 0.0003], BCD/Sal control n = 10, CORT/Sal n = 8, CORT/Ket n = 9. i Quantification of Pricke2 in excitatory neurons [F (2, 25) = 5.804, P = 0.0085], BCD/Sal control n = 11, CORT/Sal n = 9, CORT/Ket n = 8. j Quantification of Pricke2 in inhibitory neurons [F (2, 23) = 8.506, P = 0.0017], BCD/Sal control n = 8, CORT/Sal n = 10, CORT/Ket n = 8. Each column represents the mean + S.E.M. of n fields provided above. Statistical analysis was performed by one-way ANOVA followed by Tukey’s test. *p < 0.05 and **p < 0.01 compared with the control group (BCD) treated with saline, #p < 0.05, ##p < 0.01 and ###p < 0.001 compared with the CORT group treated with saline. Scale bar 10 µm; 50 µm; Magnification ×40.

Fig. 5. Celsr2/3 and PK2 are required for the sustained antidepressant effect of Ket in chronically stressed mice in the BLA-projecting IL PFC neurons.

a Schematics of experimental design. The illustrations were created with BioRender. b Representative heatmaps of mouse movement in the TST. c Quantification of immobility time in the TST in (a) [F (7, 101) = 13.20, P < 0.0001]. d Representative heatmaps of open field test. e Quantification of locomotion in the open field test in (d) [F (7, 101) = 1.321, P = 0.25]. f Quantification of food consumption in the food consumption test [F (7, 102) = 8.449, P < 0.0001]. Each column represents the mean + S.E.M. of 12–16 animals. Statistical analysis was performed by one-way ANOVA followed by Tukey’s test. *p < 0.05, **p < 0.01 compared with the control group (BCD/Control sgRNA) treated with saline, and ####p < 0.0001 compared with the CORT/Control sgRNA group treated with saline, $p < 0.05 and $$p < 0.01 compared with the CORT/Control sgRNA group treated with ketamine.

Fig. 6. Analyses of dendritic spine and excitatory synapses in the BLA-projecting IL PFC neurons.

a Schematics of experimental design. The illustrations were created with BioRender. b Representative micrographs showing dendritic spines and glutamatergic synapses in the BLA-projecting IL PFC neurons. c Quantification of dendritic spines [F (7, 16) = 56.68, P < 0.0001]. d Quantification of glutamatergic synapses [F (7, 16) = 24.27, P < 0.0001]. Each column represents the mean + S.E.M. of 3 animals (7-14 neurons/animal). Statistical analysis was performed by One-way ANOVA followed by Tukey’s test. *p < 0.05, ***p < 0.001 compared with the control group (BCD/Control sgRNA) treated with saline, ##p < 0.01 and ###p < 0.001 compared with the CORT/Control sgRNA group treated with saline, $p < 0.05 and $$$p < 0.001 compared with the CORT/Control sgRNA group treated with ketamine. Scale bar 3 µm. Green arrowheads: Bassoon puncta. Red arrowheads: PSD95 puncta. White arrowheads: Bassoon/PSD95 puncta.