The female epilepsy protein PCDH19 is a new GABAAR-binding partner that regulates GABAergic transmission as well as migration and morphological maturation of hippocampal neurons

Abstract

The PCDH19 gene (Xp22.1) encodes the cell-adhesion protein protocadherin-19 (PCDH19) and is responsible for a neurodevelopmental pathology characterized by female-limited epilepsy, cognitive impairment and autistic features, the pathogenic mechanisms of which remain to be elucidated. Here, we identified a new interaction between PCDH19 and GABAA receptor (GABAAR) alpha subunits in the rat brain. PCDH19 shRNA-mediated downregulation reduces GABAAR surface expression and affects the frequency and kinetics of miniature inhibitory postsynaptic currents (mIPSCs) in cultured hippocampal neurons. In vivo, PCDH19 downregulation impairs migration, orientation and dendritic arborization of CA1 hippocampal neurons and increases rat seizure susceptibility. In sum, these data indicate a role for PCDH19 in GABAergic transmission as well as migration and morphological maturation of neurons.

Figure 1.

PCDH19 interacts with GABA_AR alpha 1 subunit. (A) Confocal images of HEK293T cells transfected with PCDH19-V5, GABA_AR myc-alpha 1 subunit, myc-alpha 1 plus beta 2 and gamma 2 subunits either alone (top panel) or in combination (middle and lower panel) and stained with anti-V5 and anti-myc antibody. PCDH19-V5 colocalizes with alpha 1 subunit, especially in the perinuclear region when alpha 1 is expressed alone (middle panel and relative inset magnifications) and mostly in vesicle-like clusters when alpha 1 is expressed together with beta 2 and gamma 2 subunits (lower panel and magnifications). The scale bar represents 10 μm. (B) CoIP in HEK293T cells expressing the GABA_AR subunit myc-alpha 1 (alone or in combination with beta 2 and gamma 2) and either full-length PCDH19 (PCDH19-V5), truncated PCDH19 (PCDH19Δ879) or PCDH9. Both full-length and truncated PCDH19, but not PCDH9, were immunoprecipitated by anti-myc. The asterisk indicates a non-specific signal in both the IgG and anti-myc lanes. (C) Representative images of cultured rat hippocampal neurons double-stained for PCDH19 and alpha 1 at DIV18. The fluorescence intensity profiles plot refers to the dendrite in the magnification insets. The images represent single Z-sections. The scale bar represents 10 μm. (D) Summary of colocalization experiments. Pearson’s (R, 0.662 ± 0.013) and Manders’ (M1 = fraction of PCDH19 overlapping alpha 1, 0.276 ± 0.020 and M2 = fraction of alpha 1 overlapping PCDH19, 0.299 ± 0.031) coefficients (top) and Van Steensel's CCF (bottom), obtained by shifting the alpha 1 channel in the X-direction relative to the PCDH19 channel and plotting R as a function of Δx (pixel shift) (see Supplementary Material, Table S1). The sharp correlation decrease excludes random colocalization (bottom). (E, F) CoIP of endogenous PCDH19 and alpha 1 in cultured neurons (E) and brain homogenate (cortex plus hippocampus) (F).

Figure 2.

Mapping of the PCDH19-GABA_AR-interacting region. (A) Schematic of the PCDH19 structure and C-terminus fragments that were fused to GST (CT, CT1, CT2, CT3) with relative amino acid extension. SP, signal peptide; TM, transmembrane domain; CM1 and CM2, conserved motifs 1 and 2. (B, C) GST pull-down assay showing that the PCDH19 constructs GST-CT, GST-CT1 and GST-CT3, but not GST-CT2, pulled-down myc-alpha 1 overexpressed in HEK293T cells (B) and endogenous alpha 1 in rat brain (C). (D) Schematic of the GABA_AR subunit alpha 1 and intracellular loop TM3-4 fragments that were fused to GST with relative amino acid extension. (E, F) GST pull-down assay showing that the alpha 1 constructs GST-TM, GST-TM-A and GST-TM-A1, but not GST-TM-B, GST-TM-C and GST-TM-A2, pulled-down PCDH19-V5 overexpressed in HEK293T cells (E) and PCDH19 endogenously expressed in rat brain (F). (G) GST pull-down in rat brain (cortex plus hippocampus) using PCDH19 GST-CT (left) and cartoon showing the GABA_AR subunit composition and structure and alignment of alpha 1 amino acids 334-343 with the corresponding amino acids in the other alpha subunits (right).

Figure 3.

The PCDH19 expression level affects inhibitory markers and surface GABA_AR expression. (A) Western blot of cultured neurons infected at DIV1 with lentiviruses expressing PCDH19 control shRNA (Scramble), shRNA, PCDH19-V5 [at two different concentrations: 1X (1) and 2X (2), where the 1X concentration is sufficient to infect approximately 80-100% of cells], or shRNA plus PCDH19-V5 (Rescue1 and 2, according to the PCDH19-V5 virus concentration used, either 1X or 2X) and lysed at DIV8. Anti-PCDH19 detects both endogenous and lentivirally expressed protein, while anti-V5 detects exclusively the lentivirally expressed protein. Scramble and shRNA vectors express GFP protein. (B) Quantification of blots in (A) showing that PCDH19 and alpha 1 levels, normalized on GAPDH, vary accordingly (see Supplementary Material, Table S3). (C) Quantification (± SEM) of synaptic markers (GABA_AR alpha 1, GAD65/67, gephyrin, NCAD and GluA2/3) normalized to GAPDH from western blots of neurons infected as in (A) with scramble, shRNA and PCDH19-V5 (2X concentration). PCDH19 overexpression increases inhibitory markers expression with respect to the control condition, but not NCAD and GluA2/3 expression. GAD65/67 expression was significantly reduced by PCDH19 shRNA (one-way ANOVA, post-hoc Tukey’s test, *P < 0.05, **P < 0.01, ***P < 0.001; Supplementary Material, Table S4). (D) Biotinylation experiments using hippocampal cultured neurons infected at DIV8 with control shRNA (Scramble), shRNA and PCDH19-V5 (2X) and lysed at DIV12. (E) Quantification (± SEM) of the GABA_AR subunits alpha 1 and 2 in whole lysate normalized to GAPDH (input) and the surface fraction normalized to the surface transferrin receptor (TfR) (biotinylated). The surface amount of both alpha 1 and 2 subunits varied according to the PCDH19 expression levels, with alpha1 showing the greatest variation (one-way ANOVA, post-hoc Tukey’s test, *P < 0.05, **P < 0.01, ***P < 0.001; Supplementary Material, Table S5).

Figure 4.

PCDH19 shRNA-mediated downregulation affects the miniature inhibitory post-synaptic currents (mIPSCs) frequency and kinetics. (A) Representative mIPSC recording (left) and population averages (right) from primary hippocampal pyramidal neurons expressing either PCDH19 control shRNA (Scramble), shRNA or shRNA plus PCDH19-V5 (Rescue). Neurons were transfected at DIV11 and recorded at DIV15. (B) Summary of mIPSC data (± SEM) showing that PCDH19 downregulation reduced the mIPSC frequency. Furthermore, mIPSCs of shRNA-expressing neurons were slower, as demonstrated by the increased decay time and area. Peak amplitude was not significantly affected by PCDH19 downregulation (one-way ANOVA, post-hoc Tukey’s test, *P < 0.05, **P < 0.01, ***P < 0.001; Supplementary Material, Table S6).

Figure 5.

PCDH19 is highly expressed during early postnatal development in the rat hippocampus. (A) Western blots (top) and relative quantification (± SEM, bottom) showing PCDH19 expression in different brain areas (BR, all brain regions except cerebellum; CX, cortex; HP, hippocampus; CB, cerebellum) at postnatal day (P)10 and in adult (2–2.5 months old) rats (Student’s t-test, *P < 0.05, **P < 0.01; Supplementary Material, Table S7). (B) Western blot (top) and relative quantification (± SEM, bottom) showing the temporal expression of PCDH19 in lysates of rat hippocampi at different embryonic (E) and postnatal (P) days (Supplementary Material, Table S8). (C) Representative IHC images of PCDH19 in P10 rat hippocampus. PCDH19 is shown in red, the nuclear marker DAPI in blue and the merge in purple. Magnification insets 1, 2 and 3 show dentate gyrus (DG), CA1 and CA3 regions with PCDH19-positive pyramidal neurons.

Figure 6.

PCDH19 downregulation in rats affects the migration, morphological maturation and orientation of hippocampal neurons and increases seizure susceptibility. (A) Confocal images of GFP fluorescence in coronal sections of rat hippocampus at P7 after in utero transfection (at E17.5) with pRNAT-U6.3/Hygro empty vector (Control), PCDH19 shRNA or shRNA plus PCDH19-V5 (Rescue). Slices were counterstained with the nuclear marker Hoechst for visualization of hippocampal layers (left). Scale bar, 50 μm. SO, stratum oriens; SP, stratum pyramidale; SR, stratum radiatum. (B) Quantification of the number of ectopic cells (± SEM) in control, shRNA and rescue conditions. Numbers are expressed as a percentage of the ectopic cells normalized to the total number of fluorescent cells in the same section. Asterisks: statistically significant difference (Student’s t-test, **P < 0.01; Supplementary Material, Table S9). In parenthesis: total number of animals (1 slice/animal). (C) Representative reconstructions of neurons electroporated with the control vector, PCDH19 shRNA or shRNA plus PCDH19-V5 (Rescue). Scale bar, 50 μm. (D) Quantification of the average total dendritic length (normalized to controls, ± SEM) (left panel) and of the average length of apical or basal dendrites (normalized to the total length of all dendritic processes in the same neuron, ± SEM) (right panel) of neurons as in A. Asterisks: statistically significant difference (Student’s t-test, **P < 0.01, ***P < 0.001; Supplementary Material, Table S9). In parenthesis, total number of cells (3 animals/group). (E) Confocal images of GFP fluorescence in pyramidal neurons electroporated with the control vector, PCDH19 shRNA or shRNA plus PCDH19-V5 (Rescue) (left) and higher magnification images (right) showing the orientation of the apical dendrites in the SR. The white lines in the control image show the criteria utilized to calculate the angle of the processes. Scale bar, 50 μm. (F) Quantification of the frequency of the angles (binned in four groups) of the first main branching of the apical dendrite in cells as in A. Data are presented as a percentage (± SEM) of all cells in the control, shRNA or rescue condition (one-way ANOVA, post-hoc Holm-Sidak, **P < 0.01, ***P < 0.001; 3–4 animals/group; Supplementary Material, Table S9). (G) Schematic graph of the experimental protocol. (H) Quantification of the time latency to induce the first generalized tonic-clonic seizure in P7 pups electroporated into the hippocampus with control vector, PCDH19 shRNA or shRNA plus PCDH19-V5 (Rescue). The circles represent data points from single animals, and histograms represent the average. Numbers are expressed as percentages normalized to controls (± SEM) (one-way ANOVA, Holm-Sidak post hoc test, *P < 0.05; Supplementary Material, Table S10).