Impaired Function of PLEKHG2, a Rho-Guanine Nucleotide-Exchange Factor, Disrupts Corticogenesis in Neurodevelopmental Phenotypes

Homozygosity of the p.Arg204Trp variation in the Pleckstrin homology and RhoGEF domain containing G2 (PLEKHG2) gene, which encodes a Rho family-specific guanine nucleotide-exchange factor, is responsible for microcephaly with intellectual disability. However, the role of PLEKHG2 during neurodevelopment remains unknown. In this study, we analyzed mouse Plekhg2 function during cortical development, both in vitro and in vivo. The p.Arg200Trp variant in mouse (Plekhg2-RW), which corresponds to the p.Arg204Trp variant in humans, showed decreased guanine nucleotide-exchange activity for Rac1, Rac3, and Cdc42. Acute knockdown of Plekhg2 using in utero electroporation-mediated gene transfer did not affect the migration of excitatory neurons during corticogenesis. On the other hand, silencing Plekhg2 expression delayed dendritic arbor formation at postnatal day 7 (P7), perhaps because of impaired Rac/Cdc42 and p21-activated kinase 1 signaling pathways. This phenotype was rescued by expressing an RNAi-resistant version of wildtype Plekhg2, but not of Plekhg2-RW. Axon pathfinding was also impaired in vitro and in vivo in Plekhg2-deficient cortical neurons. At P14, knockdown of Plekhg2 was observed to cause defects in dendritic spine morphology formation. Collectively, these results strongly suggest that PLEKHG2 has essential roles in the maturation of axon, dendrites, and spines. Moreover, impairment of PLEKHG2 function is most likely to cause defects in neuronal functions that lead to neurodevelopmental disorders.


Introduction
Rho family guanosine triphosphatases (GTPases) are known to regulate various cellular processes, such as morphology, gene transcription, proliferation, and migration through actin cytoskeletal rearrangement, and they play crucial roles in various cell types [1][2][3][4]. Similarly to other GTPases, Rho family proteins are binary switches that cycle between an inactive (GDP-bound) and an active (GTP-bound) conformational state. Interconversion of the two forms is controlled in a variety of ways. In general, guanine nucleotide exchange factors (GEFs) promote GDP/GTP exchange to activate Rho GTPases, which then initiate recognizing downstream effectors [5][6][7]. PLEKHG2 (Pleckstrin homology and RhoGEF domain containing G2), which is a GEF for Rac and Cdc42, interacts with Gβγ subunits of heterotrimeric G proteins and is activated upon co-expression of Gβγ in various cell types; whether activation is mediated directly by Gβγ remains to be clarified [8][9][10][11][12]. Rac and Cdc42 are crucial for the spatiotemporal activation of various effectors to control

Ethics Statement
We followed the fundamental guidelines for proper conduct of animal experiments and related activities at academic research institutions under the jurisdiction of the Ministry of Education, Culture, Sports, Science, and Technology (Tokyo, Japan). All protocols for animal handling and treatment were reviewed and approved by the Animal Care and Use Committee of Institute for Developmental Research, Aichi Developmental Disability Center (approval number: 2019-013).

In Utero Electroporation
Three-month-old specific-pathogen-free ICR pregnant mice (Japan SLC, Shizuoka, Japan) were housed singly in cage (197 mm × 340 mm × 165 mm) under a 12 h lightdark cycle with free access to autoclaved food (CE-2, CLEA Japan, Inc., Tokyo, Japan). Surgery on pregnant mice and embryo manipulations in the uterus were performed as previously described [21]. At E14, pregnant mice (Japan SLC, Shizuoka, Japan) were deeply anesthetized with a mixture of medetomidine (0.75 mg/kg), midazolam (4 mg/kg), and butorphanol (5 mg/kg). In utero electroporation was then performed in specificpathogen-free animal facilities as described [22,23]. Briefly, the indicated amounts of pCAG expression and RNAi vectors were injected in various combinations with pCAG-EGFP (0.5 µg each) into the lateral ventricle of embryos using a glass micropipette made from a microcapillary tube (GD-1; Narishige, Tokyo, Japan). Then, each embryo was placed between a tweezer-type disc electrode (5-mm in diameter) (CUY650-5; NEPA Gene, Chiba, Japan) and subjected to five electronic pulses (50 ms of 35 V) at 450 ms intervals using an electroporator (NEPA21; NEPA Gene). In this way, plasmids were introduced into the somatosensory area, which is part of the parietal lobe. Brains were fixed at indicated postnatal day, sectioned, and then analyzed. All experimental procedures were carried out during day time. No animals were excluded, and none died during experimentation.

Time-Lapse Imaging
Transfected neurons were cultured in an incubator chamber (5% CO 2 and 40% O 2 , at 37 • C) fitted onto an A1R HD25 confocal laser microscope (Nikon, Tokyo, Japan). Approximately 1-15 optical Z sections were acquired automatically every 1 min for 1 h, and about 10 focal planes (~30 µm-thickness) were merged to visualize the entire shape of the cells.

Immunofluorescence
Immunofluorescence analysis was conducted essentially as described [21]. As for cortical slice staining, brains were embedded in 3% agarose, cut into sections (100 µmthickness) using a vibratome, and photographed with an LSM-880 confocal laser microscope (Carl Zeiss, Oberkochen, Germany). Acquired images were analyzed using ImageJ to determine cell morphology and fluorescence intensity or using NeuronStudio to quantify dendritic spine density [24].

Characterization of Plekhg2 Variation
Since the p.Arg204Trp missense variation in human PLEKHG2 resides within the DH domain, which is responsible for activating Rac/Cdc42 by catalyzing the exchange of GDP for GTP, PLEKHG2 harboring this variation is likely to manifest abnormal GEF activity. We, therefore, examined whether this variation affects the GEF activity for Rac1, Rac3, and Cdc42, which have been reported to be important for the cortical neuron development. To this end, we utilized a COS7-cell transient expression system, in which Myc-tagged small GTPases and Gβγ were expressed in various combinations. When compared with wild-type mouse Plekhg2, i.e., Plek2-WT, the basal GEF-activity of Plekhg2 p.Arg200Trp, i.e., Plek2-RW, which is the mouse equivalent to human PLEKHG2 p.Arg204Trp, was only weakly affected ( Figure 1A-C, lane 1 and 3). Although co-expression with Gβγ enhanced GEFactivity of Plek2-WT as previously reported [8] ( Figure 1A-C, lane 1 and 2), Gβγ-mediated augmentation of Plek2-RW activity was decreased for the GTPases tested ( Figure 1A-C, lane 2 and 4). Similarly, when PAK1 activity was assessed by T423-phosphorylation of Myc-PAK1 expressed in COS7 cells, Gβγ-stimulated Plek2-RW significantly decreased the phosphorylation when compared to Plek2-WT [26] ( Figure 1D). Taken together, we assume that the disease-causative p.Arg204Trp variation impairs Gβγ-stimulated catalytic activity of PLEKHG2 on RAC and CDC42. However, further intensive in vivo examination is essential to determine whether this signaling defects indeed reflect the physiologically relevant substrate specificity of PLEKHG2.

Role of Plekhg2 in the Dendritic Arbor Development In Vivo
Rac/Cdc42-PAK1 signaling is most likely to be impaired in the cells expressing Plek2-RW from the results in Figure 1. Since the Rac/Cdc42-PAK1 signaling has been reported to be crucial for brain development [27][28][29][30][31], we analyzed the role of PLEKHG2 in corticogenesis by acute knockdown with an in utero electroporation method. To this end, we designed 2 RNAi vectors, sh-Plek2#1 and #2, against distinct regions in the mouse Plekhg2 coding sequence. First, we examined the role of Plekhg2 in the migration of newly generated cortical neurons, because organized neuronal migration is essential for brain development. pCAG-GFP was co-electroporated with 1.0 µg of sh-Luc (control), sh-Plek2#1, or sh-Plek2#2 into the ventricular zone (VZ) progenitor cells of E14.5 brains. When localization of transfected cells and their progeny was analyzed at P0, neurons were found to migrate normally to layers II-III of the cortical plate in the control and knockdown experiments (Figure 2A), indicating that Plekhg2-knockdown has little effects on cortical neuron migration. , either alone or in combination. Cell lysates were prepared, and the pull-down assay was conducted as described in Section 2 with GST-fused PAK1-PDB (5 μg). The GTP form was indicated by the amounts of each GTPase bound to GST-fused PDB (pulldown). Total cell lysates were also immunoblotted with anti-Myc for normalization (input). Relative band intensity was shown as boxplot when the value of co-expression of Plek2-WT with respective GTPases was taken as 1.0. Number of replicates, N = 3. Different letters above bars represent significant differences, p < 0.05, according to a Tukey's test. (D) COS7 cells were transfected with pCAG-Μyc-PAK1, -Plek2-WT, -Plek2-RW, pF5A-CMV-neo-Gβ1, and -Gγ2 (0.2 μg each), either alone or in combination. PAK1 activation was indicated by the amounts of phosphorylated PAK1 at T423, p-PAK1(T423). Total PAK1 was also immunoblotted with anti-Myc antibody for normalization, Myc (PAK1). Relative band intensity was shown as boxplot when the value of (-) was taken as 1.0. Number of replicates, N = 3. Different letters above bars represent significant differences, p < 0.05, according to a Tukey's test.

Role of Plekhg2 in the Dendritic Arbor Development In Vivo
Rac/Cdc42-PAK1 signaling is most likely to be impaired in the cells expressing Plek2-RW from the results in Figure 1. Since the Rac/Cdc42-PAK1 signaling has been reported to be crucial for brain development [27][28][29][30][31], we analyzed the role of PLEKHG2 in corticogenesis by acute knockdown with an in utero electroporation method. To this end, we designed 2 RNAi vectors, sh-Plek2#1 and #2, against distinct regions in the mouse Plekhg2 coding sequence. First, we examined the role of Plekhg2 in the migration of newly generated cortical neurons, because organized neuronal migration is essential for brain development. pCAG-GFP was co-electroporated with 1.0 μg of sh-Luc (control), sh-Plek2#1, or sh-Plek2#2 into the ventricular zone (VZ) progenitor cells of E14.5 brains. When localization of transfected cells and their progeny was analyzed at P0, neurons were found to migrate normally to layers II-III of the cortical plate in the control and knockdown experiments (Figure 2A), indicating that Plekhg2-knockdown has little effects on cortical neuron migration.
Given that aberrant synaptic network formation is linked to defective brain development and function, Plekhg2 may be involved in dendrite and/or axon growth during corticogenesis. When we examined the role of Plekhg2 in dendritic arbor formation, introduction of sh-Plek2#1 and #2 (1.0 μg) into the VZ cells at E14.5 resulted in highly suppressed dendritic arborization at P7 ( Figure 2B). Branch point number was significantly , either alone or in combination. Cell lysates were prepared, and the pull-down assay was conducted as described in Section 2 with GST-fused PAK1-PDB (5 µg). The GTP form was indicated by the amounts of each GTPase bound to GST-fused PDB (pulldown). Total cell lysates were also immunoblotted with anti-Myc for normalization (input). Relative band intensity was shown as boxplot when the value of co-expression of Plek2-WT with respective GTPases was taken as 1.0. Number of replicates, N = 3. Different letters above bars represent significant differences, p < 0.05, according to a Tukey's test. (D) COS7 cells were transfected with pCAG-Myc-PAK1, -Plek2-WT, -Plek2-RW, pF5A-CMV-neo-Gβ 1 , and -Gγ 2 (0.2 µg each), either alone or in combination. PAK1 activation was indicated by the amounts of phosphorylated PAK1 at T423, p-PAK1(T423). Total PAK1 was also immunoblotted with anti-Myc antibody for normalization, Myc (PAK1). Relative band intensity was shown as boxplot when the value of (-) was taken as 1.0. Number of replicates, N = 3. Different letters above bars represent significant differences, p < 0.05, according to a Tukey's test.
Given that aberrant synaptic network formation is linked to defective brain development and function, Plekhg2 may be involved in dendrite and/or axon growth during corticogenesis. When we examined the role of Plekhg2 in dendritic arbor formation, introduction of sh-Plek2#1 and #2 (1.0 µg) into the VZ cells at E14.5 resulted in highly suppressed dendritic arborization at P7 ( Figure 2B). Branch point number was significantly reduced in the deficient neurons compared to that in matching control cells, indicating that Plekhg2 participates in dendritic arbor formation ( Figure 2C). Rescue experiments were then conducted to rule out off-target effects. We prepared RNAi-resistant versions of wild type Plekhg2 and the variant, Plek2-WT-R and Plek2-RW-R, respectively, and confirmed that they were resistant to sh-Plek2#2 in COS7 cells ( Figure 2D). Although the suppressed dendritic arborization by silencing Plekhg2 was rescued by Plek2-WT-R, Plek2-RW-R could not rescue this abnormal phenotype ( Figure 2E,F). These results demonstrate the pathophysiological significance of the variation.
that Plekhg2 participates in dendritic arbor formation ( Figure 2C). Rescue experiments were then conducted to rule out off-target effects. We prepared RNAi-resistant versions of wild type Plekhg2 and the variant, Plek2-WT-R and Plek2-RW-R, respectively, and confirmed that they were resistant to sh-Plek2#2 in COS7 cells ( Figure 2D). Although the suppressed dendritic arborization by silencing Plekhg2 was rescued by Plek2-WT-R, Plek2-RW-R could not rescue this abnormal phenotype ( Figure 2E,F). These results demonstrate the pathophysiological significance of the variation.  Based on the obtained results in Figure 1, it is possible that the dendritic arbor development is regulated by Plekhg2-Rac/Cdc42-PAK1 signaling. We here focused on the roles of Rac3 and Cdc42, which were previously reported to be important for the development of cortical neurons [14]. Co-expression of wild type Rac3 and Cdc42, perhaps due to the increased basal activity in neurons, rescued the phenotype of Plekhg2-knockdown in terms of the branch number of dendrites ( Figure 3A,B). It should be noted that neuronal migration was not affected under the experimental conditions where 0.1 µg of wild type Rac3 or Cdc42 was electroporated. On the other hand, rescue effects by Rac1 were not characterized since simple overexpression of wildtype Rac1 impaired neuronal migration on its own (data not shown). Likewise, PAK1LF, an activated variant of PAK1, also rescued the phenotype ( Figure 3C,D). Given the crucial role of Rac/Cdc42-PAK1 signaling in actin cytoskeletal reorganization, impaired regulation of Plekhg2 should dysregulate actin dynamics and disrupt the function of the actin cytoskeleton, leading to defects in dendritic development.
Based on the obtained results in Figure 1, it is possible that the dendritic arbor devel opment is regulated by Plekhg2-Rac/Cdc42-PAK1 signaling. We here focused on the role of Rac3 and Cdc42, which were previously reported to be important for the developmen of cortical neurons [14]. Co-expression of wild type Rac3 and Cdc42, perhaps due to th increased basal activity in neurons, rescued the phenotype of Plekhg2-knockdown in terms of the branch number of dendrites ( Figure 3A,B). It should be noted that neurona migration was not affected under the experimental conditions where 0.1 μg of wild type Rac3 or Cdc42 was electroporated. On the other hand, rescue effects by Rac1 were no characterized since simple overexpression of wildtype Rac1 impaired neuronal migration on its own (data not shown). Likewise, PAK1LF, an activated variant of PAK1, also res cued the phenotype (Figure 3C,D). Given the crucial role of Rac/Cdc42-PAK1 signaling in actin cytoskeletal reorganization, impaired regulation of Plekhg2 should dysregulate actin dynamics and disrupt the function of the actin cytoskeleton, leading to defects in dendriti development.  Figure 2. (B) Quantification analyses of (A). More than 8 neurons were analyzed in slices from each of 2 brains obtained from two independent electroporation experiments. The sig nificance of difference between control (sh-Plek2#2), sh-Plek2#2 + Rac3, and sh-Plek2#2 + Cdc42 wa determined using Dunnett's test. *** p < 0.001, ** p < 0.01, * p < 0.05, +++ p < 0.001, ++ p < 0.01. (D Quantification analyses of (C). More than 8 neurons were analyzed in 2 slices from each of 2 brain obtained from two independent electroporation experiments. The significance of difference between control (sh-Plek2#2) and sh-Plek2#2 + PAK1LF was determined using Welch's t-test. ** p < 0.01, * < 0.05.  Figure 2. (B) Quantification analyses of (A). More than 8 neurons were analyzed in 2 slices from each of 2 brains obtained from two independent electroporation experiments. The significance of difference between control (sh-Plek2#2), sh-Plek2#2 + Rac3, and sh-Plek2#2 + Cdc42 was determined using Dunnett's test. *** p < 0.001, ** p < 0.01, * p < 0.05, +++ p < 0.001, ++ p < 0.01. (D) Quantification analyses of (C). More than 8 neurons were analyzed in 2 slices from each of 2 brains obtained from two independent electroporation experiments. The significance of difference between control (sh-Plek2#2) and sh-Plek2#2 + PAK1LF was determined using Welch's t-test. ** p < 0.01, * p < 0.05.

Role of Plekhg2 in the Dendritic Spine Formation
Since Plekhg2 was shown to be involved in dendrite network formation, we further explored whether Plekhg2 participates in the synapse formation and/or maintenance. It is also notable that ID, the main clinical features of the individuals with p.Arg204Trp in PLEKHG2, is associated with synaptic dysfunction. To this end, we assessed dendritic spine formation in cortical neurons following Plekhg2-knockdown in vivo. Notably, although Plekhg2-deficient neurons were differentiated with numerous dendrites at P14, the number appeared to be less than the control neurons ( Figure 4A). Under the conditions, spine density was significantly decreased in neurons transfected with sh-Plek2#2, and the phenotype was rescued by Plek2-WT-R, but not Plek2-RW-R ( Figure 4B,C). On the other hand, knockdown of Plekhg2 little affected the spine length and head diameter (Supplementary Figure S1). Further analyses are required to determine if the observed impaired spine formation is a primary phenotype or secondary to dendrite growth defects at this timing. In any case, the functional disruption in Plekhg2 is most likely to impair synaptic connectivity through the defects in dendritic spine morphology. The pathophysiology of ID with PLEKHG2 gene abnormalities may reflect these observed cellular phenotypes.
Since Plekhg2 was shown to be involved in dendrite network formation, we further explored whether Plekhg2 participates in the synapse formation and/or maintenance. It is also notable that ID, the main clinical features of the individuals with p.Arg204Trp in PLEKHG2, is associated with synaptic dysfunction. To this end, we assessed dendritic spine formation in cortical neurons following Plekhg2-knockdown in vivo. Notably, although Plekhg2-deficient neurons were differentiated with numerous dendrites at P14, the number appeared to be less than the control neurons ( Figure 4A). Under the conditions, spine density was significantly decreased in neurons transfected with sh-Plek2#2, and the phenotype was rescued by Plek2-WT-R, but not Plek2-RW-R ( Figure 4B,C). On the other hand, knockdown of Plekhg2 little affected the spine length and head diameter (Supplementary Figure S1). Further analyses are required to determine if the observed impaired spine formation is a primary phenotype or secondary to dendrite growth defects at this timing. In any case, the functional disruption in Plekhg2 is most likely to impair synaptic connectivity through the defects in dendritic spine morphology. The pathophysiology of ID with PLEKHG2 gene abnormalities may reflect these observed cellular phenotypes. . More than 8 neurons were analyzed in 2 slices from each of 2 brains obtained from two independent electroporation experiments. The significance of difference between sh-Luc (control) and Plekhg2-knockdown or each rescue condition was determined using Dunnett's test. ** p < 0.01.

Role of Plekhg2 in the Axon Growth In Vitro and In Vivo
As a next set of experiments, we examined the role of Plekhg2 in the axon growth. When control sh-Luc was introduced into primary cultured cortical neurons at div0 and analyzed at div2, cells were normally differentiated in a time-dependent manner ( Figure  5A,B, and Supplementary Video S1). In the meantime, Plekhg2-deficient neurons also appeared to be similarly differentiated in terms of axon elongation at this timing, perhaps due to the time-lag of the knockdown effects. Drastic phenotype was, however, observed when looking closer; the axonal growth cone of the deficient neurons was disrupted (Figure 5C,D, and Supplementary Video S2). The axon growth velocity of Plekhg2-silenced neurons was severely reduced when compared to control cells at div2 (Figure 5E), sug- . More than 8 neurons were analyzed in 2 slices from each of 2 brains obtained from two independent electroporation experiments. The significance of difference between sh-Luc (control) and Plekhg2-knockdown or each rescue condition was determined using Dunnett's test. ** p < 0.01.

Role of Plekhg2 in the Axon Growth In Vitro and In Vivo
As a next set of experiments, we examined the role of Plekhg2 in the axon growth. When control sh-Luc was introduced into primary cultured cortical neurons at div0 and analyzed at div2, cells were normally differentiated in a time-dependent manner ( Figure 5A,B, and Supplementary Video S1). In the meantime, Plekhg2-deficient neurons also appeared to be similarly differentiated in terms of axon elongation at this timing, perhaps due to the time-lag of the knockdown effects. Drastic phenotype was, however, observed when looking closer; the axonal growth cone of the deficient neurons was disrupted ( Figure 5C,D, and Supplementary Video S2). The axon growth velocity of Plekhg2-silenced neurons was severely reduced when compared to control cells at div2 (Figure 5E), suggesting that Plekhg2 is important for the growth cone-mediated axon guidance. We further analyzed the physiological significance of Plekhg2 in interhemispheric axon projections of cortical neurons in vivo. When Plekhg2 was silenced in the VZ cells at E14.5, axon bundles extended normally into the contralateral hemisphere at P7 (Supplementary Figure S2). However, such axons did not extend properly into the cortical layers on the contralateral side after reaching the contralateral white matter at P7 ( Figure 5F). These results suggest that Plekhg2 is involved in the axon pathfinding of cortical neurons. gesting that Plekhg2 is important for the growth cone-mediated axon guidance. We further analyzed the physiological significance of Plekhg2 in interhemispheric axon projections of cortical neurons in vivo. When Plekhg2 was silenced in the VZ cells at E14.5, axon bundles extended normally into the contralateral hemisphere at P7 (Supplementary Figure S2). However, such axons did not extend properly into the cortical layers on the contralateral side after reaching the contralateral white matter at P7 ( Figure 5F). These results suggest that Plekhg2 is involved in the axon pathfinding of cortical neurons.

Discussion
De novo RAC1 missense variants have been identified in neurodevelopmental disorders with global developmental delay/ID and brain size abnormalities as core phenotypes (Mental Retardation autosomal dominant 48, MRD48, OMIM #617751) [13]. The striking correlation between RAC1 activation levels and the head size of the affected individuals

Discussion
De novo RAC1 missense variants have been identified in neurodevelopmental disorders with global developmental delay/ID and brain size abnormalities as core phenotypes (Mental Retardation autosomal dominant 48, MRD48, OMIM #617751) [13]. The striking correlation between RAC1 activation levels and the head size of the affected individuals has been shown; the active and inactive variants tend to be associated with microcephaly and macrocephaly, respectively [13]. Interestingly, microcephaly and macrocephaly also have been described in individuals harboring distinct variants in TRIO, a RAC1-GEF [32][33][34]. Considering the relationship of microcephaly with the PLEKHG2 variation, impaired RAC1-signaling might be partially related to the clinical features of the patients. The present study showed that Plekhg2 plays essential roles in dendritic arbor development, axon pathfinding, and synapse formation/maintenance of cortical neurons during cortical development. It should be here noted that, unlike TRIO, Plekhg2 is not involved in neuronal migration (Figure 2A), suggesting that Plekhg2 appears to spatiotemporally activate Rac1 which functions in the "cell morphology machinery" rather than the "migration machinery". Given many RhoGEFs and RhoGTPases expressed in neurons [35,36], a variety of RhoGEFs including PLEKHG2 should coordinatedly modulate various Rho-mediated cellular events in spatiotemporally regulated manners during brain development.
The p.Arg204Trp variation in PLEKHG2 is reported to cause "postnatal" microcephaly; head circumference of the patients were normal at birth and fell in percentiles to approximately −3 S.D. at 4-6 months of age [19]. In addition, MRI analyses of the patients at 8 months to 5 years of age disclosed abnormal white matter with a diffuse high signal on T2 and atrophy apparently due to paucity of white matter [19]. Considering that Plekhg2knockdown had little effect on neuronal precursor proliferation (Supplementary Figure  S3), the postnatal microcephaly associated with the PLEKHG2 variation is supposed to be caused by defects in neurite growth after birth. On the other hand, given that ID is referred to as a "synaptic" disorder, impaired synaptic function due to defects in dendritic as well as spine development may underlie the pathogenesis and pathophysiology of ID of the patients.
In utero electroporation-based analyses confirmed that Plekhg2-knockdown had little effects on axon extension of cortical neurons, but suppressed axon pathfinding after reaching contralateral cortex ( Figure 5F). Additionally, Plekhg2-knockdown in dissociated cortical neurons induced axonal growth cone collapse ( Figure 5A-E). Although the apparently normal phenotype of the axon extension observed in vitro may be due to the time-lag of the short hairpin RNA expression, Plekhg2-mediated spatiotemporal activation of Rac/Cdc42 is possible to regulate axon pathfinding through the control of the growth cone rather than axon elongation.
It is notable that the International Mouse Phenotyping Consortium has reported that the Plekhg2-knockout mice displayed only decreased body length out of 75 parameters assessed (https://www.mousephenotype.org/data/genes/MGI:2141874 (accessed on 12 January 2022). On the other hand, we here observed that acute knockdown of Plekhg2 is involved in cortical architecture. We assume that conditional and acute knockdown by the combination of in utero electroporation and RNAi circumvents the compensatory effects of general gene-knockout approaches. Given a variety of GEFs for Rac and Cdc42 expressed in developing central nervous system, it is possible that yet unidentified GEF(s) may compensate for the function of Plekhg2 in the Plekhg2-knockout mice.
As for the molecular mechanism underlying the PLEKHG2-deficient phenotypes, it remains to be clarified whether coordinated interaction of PLEKHG2 with Gβγ is pivotal for brain development. In this context, it is notable that Arg204 of PLEKHG2 is positioned in DH domain which is responsible for interacting with both Rac/Cdc42 and Gβγ [8], indicating possible influence by the p.Arg204Trp variation on the interaction of PLEKHG2 with Gβγ. Indeed, p.Arg200Trp variation was shown to abrogate the ability of Gβγ to activate Plekhg2 (Figure 1A-C) and PAK1 ( Figure 1D) in COS7 cells. Considering that Gβγ is released from the heterotrimeric G proteins in responses to cell surface receptor stimulation and is associated with neurodevelopmental disorders [37][38][39], receptor-mediated signaling might play a role in neuronal morphology and synaptic plasticity. Further investigation is required to clarify the physiological significance of Gβγ-dependent Plekhg2 GEF activity for Rac/Cdc42 and subsequent PAK1 activation.
Although the spine structure is regulated by various molecular mechanisms, actin filaments provide the main foundation for spine shape, motility, and stability. We have previously reported that Plekhg2 localizes at both excitatory and inhibitory synapses [40], suggesting an essential role of Plekhg2 in synapse formation/maintenance. We assume that Plekhg2-Rac/Cdc42-PAK1 signaling also takes part in spine formation in vivo and reduced spine density observed in Figure 4 may be a primary phenotype. Indeed, the p.Arg200Trp variation in Plekhg2 was found to be essential for the impaired spine formation since reduced spine density of cortical neurons by Plekhg2-knockdown was rescued by wild type Plekhg2 but not by the variant (Figure 4). To elucidate the physiological and pathophysiological significance of Plekhg2 in spine formation, identification of the spatiotemporal signaling network involving Plekhg2 will be essential.
In the present study, we clarified that PLEKHG2 is crucial for brain development based on the results that Plekhg2-deficiency caused morphological defects in cortical neurons during brain development. Additionally, the p.Arg204Trp variation in PLEKHG2 is known to cause neurodevelopmental disorders with microcephaly and ID. As for an underlying mechanism in the clinical phenotypes, disrupted actin regulation caused by aberrant Rac/Cdc42-PAK1 signaling was supposed to be a core event; hampered actin reorganization gives rise to abrogated neurite growth and spine formation, which are relevant to the pathophysiology of ID and microcephaly in PLEKHG2-associated neurological disorder. Further intensive analyses of the molecular machinery should contribute to a better understanding of the pathophysiology of the disease.

Conclusions
The p.Arg204Trp variant in PLEKHG2 is responsible for postnatal microcephaly and ID. In this study, mouse Plekhg2 was shown to be essential for axon, dendrite, and synapse development in cortical neurons in vivo. Then, pathophysiological significance of the Plekhg2 variant p.Arg200Trp corresponding to p.Arg204Trp in PLEKHG2 was examined using a mouse model. As for an underlying mechanism of the pathogenesis, hampered actin reorganization by disrupted PLEKHG2-RAC/CDC42-PAK1 signaling appeared to be a core event. The obtained results indicate that PLEKHG2 has essential roles in brain development, and disruption of its function may cause defects in neuronal development, which lead to ID and microcephaly.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cells11040696/s1, Figure S1: Role of Plekhg2 in the dendritic spine morphology in vivo, Figure S2: Role of Plekhg2 in the axon extension in vivo, Figure S3: Estimation of the length of the cell cycle of Plekhg2-deficient cortical progenitors, Video S1: Timelapse imaging of control neuron in vitro, Video S2: Timelapse imaging of Plekhg2-knockdown neuron in vitro [41]. Informed Consent Statement: Not applicable.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding authors, upon reasonable request.