P-Rex1 Overexpression Results in Aberrant Neuronal Polarity and Psychosis-Related Behaviors

Abstract

Neuronal polarity is involved in multiple developmental stages, including cortical neuron migration, multipolar-to-bipolar transition, axon initiation, apical/basal dendrite differentiation, and spine formation. All of these processes are associated with the cytoskeleton and are regulated by precise timing and by controlling gene expression. The P-Rex1 (phosphatidylinositol-3,4,5-trisphosphate dependent Rac exchange factor 1) gene for example, is known to be important for cytoskeletal reorganization, cell motility, and migration. Deficiency of P-Rex1 protein leads to abnormal neuronal migration and synaptic plasticity, as well as autism-related behaviors. Nonetheless, the effects of P-Rex1 overexpression on neuronal development and higher brain functions remain unclear. In the present study, we explored the effect of P-Rex1 overexpression on cerebral development and psychosis-related behaviors in mice. In utero electroporation at embryonic day 14.5 was used to assess the influence of P-Rex1 overexpression on cell polarity and migration. Primary neuron culture was used to explore the effects of P-Rex1 overexpression on neuritogenesis and spine morphology. In addition, P-Rex1 overexpression in the medial prefrontal cortex (mPFC) of mice was used to assess psychosis-related behaviors. We found that P-Rex1 overexpression led to aberrant polarity and inhibited the multipolar-to-bipolar transition, leading to abnormal neuronal migration. In addition, P-Rex1 overexpression affected the early development of neurons, manifested as abnormal neurite initiation with cytoskeleton change, reduced the axon length and dendritic complexity, and caused excessive lamellipodia in primary neuronal culture. Moreover, P-Rex1 overexpression decreased the density of spines with increased height, width, and head area in vitro and in vivo. Behavioral tests showed that P-Rex1 overexpression in the mouse mPFC caused anxiety-like behaviors and a sensorimotor gating deficit. The appropriate P-Rex1 level plays a critical role in the developing cerebral cortex and excessive P-Rex1 might be related to psychosis-related behaviors.

Keywords: Lamellipodia; Neurodevelopment; P-Rex1; Polarity; Psychosis-related behavior.

Fig. 1

P-Rex1 overexpression inhibited the radial migration of cortical neurons by impairing neuron polarity. A Representative images of E14.5 mouse cortices electroporated with P-Rex1:OE and Mock plasmids at P0. Representative coronal sections at P0 stained with antibodies to GFP (green) and counterstained with Hoechst (blue). Scale bar, 50 μm. B Percentages of GFP cells in each region [379 cells from 9 embryos (Mock); 238 cells from 7 embryos (P-Rex1:OE)]. Data are represented as the mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test. C, D Enlarged images of electroporated neurons after the overexpression of P-Rex1 that inhibited the formation of leading processes and changed neuronal polarity. Scale bars, 50 μm. E Representative images showing the morphology of neurons in layers II–III and V–VI of P2 cortex. Scale bar, 50 μm. F Representative images showing the Golgi apparatus localization of electroporated neurons in the IZ (arrows) (E14.5–E16.5). Scale bar, 50 μm. G Radially migrating cells defined as the percentage of cells with Golgi apparatus facing the cortical plate. Mock cells from 12 slices; P-Rex1:OE cells from 16 slices. *P < 0.05, **P < 0.01, ***P < 0.001, Pearson’s χ² test. Data represented as the mean ± SEM. H Representative images showing the migration of cortical neurons electroporated with NeuroD:: IRES-EGFP and NeuroD: P-Rex1-IRES-EGFP (E14.5-P0). Scale bars, 50 μm. I Percentages of GFP+ cells in each region (E14.5-P0) [93 cells from 5 embryos (NeuroD::IRES-EGFP); 182 cells from 6 embryos (NeuroD::P-Rex1-IRES-EGFP)]. *P < 0.05, **P < 0.01, ***P < 0.001, Mann-Whitney test. Data represented as the mean ± SEM.

Fig. 2

P-Rex1 overexpression impeded neurite initiation through cytoskeletal change. A Percentages of primary cortical neurons at different developmental stages (DIV0-3) [846 Mock cells and 782 P-Rex1:OE cells from 3 independent experiments analyzed using two-way ANOVA. Interaction effect: F (2, 36) = 82.340, P < 0.001. Data represented as the mean ± SEM. B, C Stage 1 neurons, and lamellipodia protrusion area in stage 1 neurons [59 Mock cells and 49 P-Rex1:OE cells from 3 independent experiments]. *P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test. Scale bar, 20 μm. Data presented as the mean ± SEM. D, E Stage 2 neurons, and numbers of primary neurites in these neurons. P-Rex1 overexpression regulated F-actin and influenced neurite initiation. Scale bar, 20 μm; 43 Mock cells and 52 P-Rex1:OE cells from 3 independent experiments; *P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test. Data presented as the mean ± SEM). F–H Stage 3 neurons, and quantification of other neurites and axon length in these neurons (scale bar, 20 μm; DIV0-3; 130 Mock cells and 97 P-Rex1:OE cells from 3 independent experiments: *P < 0.05, **P < 0.01, ***P < 0.001, Mann-Whitney test. Data presented as the mean ± SEM).

Fig. 3

P-Rex1 overexpression led to excess lamellipodia and abnormal spines in vitro. A Representative images of cortical neurons transfected with Mock and P-Rex1 at DIV 4 for 10 days (OE plasmids by calcium phosphate transfection; scale bar, 50 μm). B High magnification images showing marked lamellipodia formation in the dendrites of rat primary cortical neurons overexpressing P-Rex1 (DIV4–10; scale bar, 10 μm). C–E Numbers of dendrite branch points, length of the longest dendrites, and total dendritic length (DIV4–10; Scale bars, 10 μm; 59 Mock cells and 50 P-Rex1:OE cells; OE cells from 3 independent experiments; *P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test). Data presented as the mean ± SEM. F Sholl analysis showing numbers of dendritic intersections and total number of dendritic intersections (DIV4-10; cells from 3 independent experiments; main effect of group, F (1, 2676) = 357.411, P < 0.001, two-way ANOVA). Data presented as the mean ± SEM. G, H Dendritic spine morphology at DIV10–18 (scale bar, 5 μm; 216 Mock cells and 122 P-Rex1:OE cells from 3 independent experiments; *P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test). Data presented as the mean ± SEM.

Fig. 4

P-Rex1 overexpression in vivo caused abnormal spines and abnormal behaviors. A Virus overexpressing P-Rex1 injected into the mPFC (scale bar, 500 μm). B Western blot result of the virus injection site in the mPFC after the behavioral tests. C Morphology of dendritic spines in Mock and P-Rex1:OE group (scale bar, 5 μm). D Dendritic spine density, height, width, and head area were greater in the P-Rex1:OE group (135 mock cells and 114 P-Rex1:OE cells from 3 independent experiments; *P < 0.05; **P < 0.01; ***P < 0.001, Student’s t test). Data presented as mean ± SEM. E The open field results did not different between the Mock and P-Rex1:OE groups (*P < 0.05; **P < 0.01; ***P < 0.001, Student’s t test; Mock, n = 16, P-Rex1:OE, n = 19). Data presented as the mean ± SEM. F The time in open arms was significantly lower in the P-Rex1:OE group (*P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test; Mock, n = 13, P-Rex1:OE, n = 15). Data presented as the mean ± SEM. G The time in the light box was lower in the P-Rex1:OE group (*P < 0.05, **P < 0.01, ***P < 0.001, Student’s t test; Mock, n = 17, P-Rex1:OE, n = 17). Data presented as the mean ± SEM. H Performance of P-Rex1:OE mice was abnormal in PPI. P-Rex1:OE mice showed a significant reduction in the ability to gate the startle response, as indicated by decreased pre-pulse inhibition at three kinds of startle (70, 74, and 82 dB) (main effect of group: F (1, 60) = 16.709, P < 0.001, two-way ANOVA; Mock, n = 12; P-Rex1, n = 10). Data presented as the mean ± SEM.