https://doi.org/10.1093/brain/awac106 BRAIN 2022: 145; 3308–3327 | 3308 Variant-specific changes in RAC3 function disrupt corticogenesis in neurodevelopmental phenotypes Marcello Scala,1,2,3,† Masashi Nishikawa,3,† Hidenori Ito,3,† Hidenori Tabata,3 Tayyaba Khan,4 Andrea Accogli,1 Laura Davids,5 Anna Ruiz,6 Pietro Chiurazzi,7,8 Gabriella Cericola,9 Björn Schulte,10 Kristin G. Monaghan,11 Amber Begtrup,11 Annalaura Torella,12,13 Michele Pinelli,12 Anne-Sophie Denommé-Pichon,14,15,16 Antonio Vitobello,14,15 Caroline Racine,15,16 Maria Margherita Mancardi,17 Courtney Kiss,18 Andrea Guerin,18 Wendy Wu,4,19 Elisabeth Gabau Vila,20 Bryan C. Mak,21 Julian A. Martinez-Agosto,21,22,23 Michael B. Gorin,21,24,25 Bugrahan Duz,26 Yavuz Bayram,27,28 Claudia M. B. Carvalho,29,30 Jaime E. Vengoechea,5 David Chitayat,31,32,33 Tiong Yang Tan,34 Bert Callewaert,35 Bernd Kruse,9 Lynne M. Bird,36,37 Laurence Faivre,14,16 Marcella Zollino,7,8 Saskia Biskup,10,38 Undiagnosed Diseases Network, Telethon Undiagnosed Diseases Program, Pasquale Striano,1,2 Vincenzo Nigro,12,13 Mariasavina Severino,39 Valeria Capra,40 Gregory Costain4,32,33,41 and Koh-ichi Nagata3,42 † These authors contributed equally to this work. Variants in RAC3, encoding a small GTPase RAC3 which is critical for the regulation of actin cytoskeleton and intra- cellular signal transduction, are associated with a rare neurodevelopmental disorder with structural brain anomalies and facial dysmorphism. We investigated a cohort of 10 unrelated participants presenting with global psychomotor delay, hypotonia, behavioural disturbances, stereotyped movements, dysmorphic features, seizures and musculoskeletal abnormalities. MRI of brain revealed a complex pattern of variable brain malformations, including callosal abnormalities, white matter thinning, grey matter heterotopia, polymicrogyria/dysgyria, brainstem anomalies and cerebellar dysplasia. These patients har- boured eight distinct de novo RAC3 variants, including six novel variants (NM_005052.3): c.34G > C p.G12R, c.179G > A p.G60D, c.186_188delGGA p.E62del, c.187G > A p.D63N, c.191A > G p.Y64C and c.348G > C p.K116N. We then examined the pathophysiological significance of these novel and previously reported pathogenic variants p.P29L, p.P34R, p.A59G, p.Q61L and p.E62K. In vitro analyses revealed that all tested RAC3 variants were biochemically and biologically active to variable extent, and exhibited a spectrum of different affinities to downstream effectors including p21-activated kinase 1. We then focused on the four variants p.Q61L, p.E62del, p.D63N and p.Y64C in the Switch II region, which is essential for the biochemical activity of small GTPases and also a variation hot spot common to other Rho family genes, RAC1 and CDC42. Acute expression of the four variants in embryonic mouse brain using in utero electroporation caused defects in cortical neuron morphology and migration ending up with cluster formation during corticogenesis. Notably, defective mi- gration by p.E62del, p.D63N and p.Y64C were rescued by a dominant negative version of p21-activated kinase 1. Our results indicate that RAC3 variants result in morphological and functional defects in cortical neurons during brain development through variant-specific mechanisms, eventually leading to heterogeneous neurodevelopmental phenotypes. Received August 05, 2021. Revised February 01, 2022. Accepted March 13, 2022. Advance access publication July 4, 2022 © The Author(s) 2022. Published by Oxford University Press on behalf of the Guarantors of Brain. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which per- mits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3309 1 Department of Neurosciences, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health, University of Genoa, Genoa, Italy 2 Pediatric Neurology and Muscular Diseases Unit, IRCCS Istituto Giannina Gaslini, Genoa, Italy 3 Department of Molecular Neurobiology, Institute for Developmental Research, Aichi Developmental Disability Center, 713-8 Kamiya, Kasugai 480-0392, Japan 4 Genetics and Genome Biology, Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada 5 Department of Human Genetics, Emory Healthcare, Atlanta, GA 30322, USA 6 Genetics Laboratory, UDIAT-Centre Diagnòstic, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de, Barcelona, Sabadell, Spain 7 Dipartimento Universitario Scienze della Vita e Sanità Pubblica, Sezione di Medicina Genomica, Università Cattolica Sacro Cuore, Rome, Italy 8 Genetica Medica, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy 9 Neuropediatric Department, Helios-Klinikum Hildesheim, Hildesheim, Germany 10 Praxis für Humangenetik, Tübingen, Germany 11 GeneDx, Gaithersburg, MD, USA 12 Telethon Institute of Genetics and Medicine, Pozzuoli, Naples, Italy 13 Department of Precision Medicine, University of Campania Luigi Vanvitelli, Naples, Italy 14 INSERM UMR1231 Génétique des Anomalies du Développement, Université de Bourgogne Franche-Comté, Dijon, France 15 Laboratoire de Génétique Moléculaire, UF Innovation en diagnostic génomique des maladies rares, Plateau Technique de Biologie, CHU de Dijon Bourgogne, Dijon, France 16 Centre de Génétique et Centre de Référence Anomalies du Développement et Syndromes Malformatifs de l’interrégion Est, FHU TRANSLAD, Hôpital d’Enfants, CHU de Dijon Bourgogne, Dijon, France 17 Unit of Child Neuropsychiatry, Department of Medical and Surgical Neuroscience and Rehabilitation, IRCCS Istituto Giannina Gaslini, Genova, Italy 18 Division of Medical Genetics, Department of Pediatrics, Queen’s University, Kingston, ON K7L 2V7, Canada 19 Queen’s University, Kingston, ON, Canada 20 Paediatric Unit, Parc Taulí Hospital Universitari, Institut d’Investigació i Innovació Parc Taulí I3PT, Universitat Autònoma de, Barcelona, Sabadell, Spain 21 Department of Human Genetics, David Geffen School of Medicine, University of California-Los Angeles, Los Angeles, CA, USA 22 Department of Pediatrics, David Geffen School of Medicine, University of California-Los Angeles, Los Angeles, CA, USA 23 Department of Psychiatry and Biobehavioral Sciences, David Geffen School of Medicine, University of California- Los Angeles, Los Angeles, CA, USA 24 Department of Ophthalmology, Jules Stein Eye Institute, David Geffen School of Medicine, UCLA, Los Angeles 90095, CA, USA 25 Brain Research Institute, UCLA, Los Angeles 90095, CA, USA 26 Haseki Training and Research Hospital, Istanbul, Turkey 27 Division of Genomic Diagnostics, Department of Pathology and Laboratory Medicine, Children’s Hospital of Philadelphia, Philadelphia, PA 19104, USA 28 Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA 29 Pacific Northwest Research Institute, Seattle, WA 98122, USA 30 Baylor College of Medicine, Houston, TX 77030, USA 31 The Prenatal Diagnosis and Medical Genetics Program, Department of Obstetrics and Gynecology, Mount Sinai Hospital, University of Toronto, Toronto, Ontario, Canada 32 Division of Clinical and Metabolic Genetics, Department of Paediatrics, The Hospital for Sick Children, Toronto, Ontario, Canada 33 Department of Molecular Genetics, University of Toronto, Toronto, Ontario, Canada 34 Victorian Clinical Genetics Services, Murdoch Children’s Research Institute, and Department of Paediatrics, University of Melbourne, Melbourne, VIC 3052, Australia 35 Center for Medical Genetics, Ghent University Hospital, Gent, Belgium 36 Department of Pediatrics, University of California San Diego, San Diego, CA, USA 37 Genetics/Dysmorphology, Rady Children’s Hospital San Diego, San Diego, CA, USA 38 CeGaT GmbH, Tübingen, Germany 39 Neuroradiology Unit, IRCCS Istituto Giannina Gaslini, Genoa, Italy 40 Medical Genetics Unit, IRCCS Istituto Giannina Gaslini, Genoa, Italy 41 Department of Paediatrics, University of Toronto, Toronto, Ontario, Canada 42 Department of Neurochemistry, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Nagoya 466- 8550, Japan 3310 | BRAIN 2022: 145; 3308–3327 M. Scala et al. Correspondence to: Koh-ichi Nagata Department of Molecular Neurobiology Institute for Developmental Research Aichi Human Service Center, 713-8 Kamiya Kasugai, Aichi 480-0392, Japan E-mail: knagata@inst-hsc.jp Correspondence may also be addressed to: Valeria Capra Medical Genetics Unit, IRCCS Istituto Giannina Gaslini, Genoa, Italy E-mail: valeriacapra@gaslini.org Correspondence may also be addressed to: Gregory Costain Division of Clinical and Metabolic Genetics Department of Pediatrics The Hospital for Sick Children Toronto, Ontario, Canada E-mail: gregory.costain@sickkids.ca Keywords: RAC3; small GTPase; brain development; axon guidance; neuronal migration Abbreviations: GAP = GTPase-activating protein; GEF = guanine nucleotide-exchange factor; GTPase = guanosine triphosphatase; NDD = neurodevelopmental disorder; NEDBAF = neurodevelopmental disorder with structural brain anomalies and dysmorphic facies; RBR = RAC-binding region; SVZ = subventricular zone Introduction listed ASD candidate genes are therefore directly involved in the sig- nalling pathways of Rho family GTPases. More specifically, the link Rho family guanosine triphosphatases (GTPases) are key regulators between the abnormal function of Rho GTPases per se and NDDs of cellular signalling and cytoskeletal dynamics, which influence has been recently highlighted as follows. As for the Rac subfamily, cell adhesion, morphology, migration and cell cycle progression, de novo missense RAC1 variants affecting protein function have and play crucial roles in various cell types.1 The Rac subfamily con- been associated with heterogeneous conditions characterized by sists of three proteins, Rac1–3, which share about 90% identity in their variable psychomotor delays and brain malformations,16 while varia- amino acid (aa) sequences, with the largest degree of divergence at tions in RAC2 have been identified in patients with different forms of their carboxy-terminal ends. While Rac1 is ubiquitously expressed primary immunodeficiencies.17 Recently, a novel NDD with struc- in most organs2 and Rac2 is specifically expressed in hematopoietic tural brain anomalies and dysmorphic facies (NEDBAF, OMIM cells,3 Rac3 is abundantly expressed both in the developing and adult 618577) has been reportedly in association with RAC3 variations.18,19 nervous system.4,5 Accordingly, Rac3 has been reported to participate This condition is characterized by global developmental delay, intel- in different aspects of neuronal development, such as neuritogen- lectual disability, abnormal muscle tone, heterogeneous neurological esis, axon and dendrites formation, synaptogenesis and regulation phenotypes and structural brain abnormalities.18,19 Although the re- of migration.6–10 Like other small GTPases, Rac3 is thought to act as ported variants appear to hamper RAC3’s basic functions, such as ac- a molecular switch cycling between GTP-bound (active) and tin cytoskeletal organization and signal transduction, the exact GDP-bound (inactive) states via conformational changes in the molecular mechanisms involved in the pathogenesis of NEDBAF re- Switch I and Switch II regions, which are highly conserved in small main to be elucidated. GTPases. The Switch II region determines the selective interaction In the present study, we investigated a cohort of 10 unrelated par- with functionally diverse guanine nucleotide-exchange factors ticipants presenting with a syndromic NDD and peculiar brain malfor- (GEFs) and GTPase-activating proteins (GAPs), which are the major mations, including one partially reported case.20 We identified six classes of regulators controlling the guanine nucleotide-binding novel de novo variants in RAC3, including five missense and one in- state.11 GEFs promote GTP-loading to activate small GTPases, while frame deletion. In vitro analyses revealed that the novel and previous- GAPs stimulate intrinsic GTP-hydrolysis to inactivate them. In add- ition to GEFs and GAPs, the Switch I region interacts with down- ly reported 11 variants were biochemically and biologically active to stream effector molecules whose activities are strictly regulated in variable extent and seemed to interact with various downstream ef- a spatiotemporal manner by activated small GTPases. Although fectors in variant-specific manners. In vivo analyses revealed that Rac1 and Rac3 share common downstream effectors, their distinct the four variants in the Switch II region caused morphological and mi- expression profiles strongly suggest isoform-specific functions. gration defects of excitatory neurons during corticogenesis, with the Accumulating evidence supports critical roles for Rho family involvement of a functional dysregulation of the downstream effector GTPases in neural development,1,12,13 and indicates that the dis- PAK1 in three out of four cases. In addition, the four variants pre- ruption of Rho GTPase signalling significantly contributes to the vented axonal development in vivo. Collectively, RAC3 was found to pathogenesis of neurodevelopmental disorders (NDDs) in hu- mans.14,15 In the Simons Foundation Autism Research Initiative play a crucial role in brain development and pathogenic variants af- (SFARI) database, 7/82 RhoGEFs (9%), 7/57 RhoGAPs (12%), and fecting its function are most likely to cause impaired corticogenesis, 6/73 downstream effectors (8%) are categorized as autism spec- leading to the neurodevelopmental phenotypes observed in patients trum disorders (ASD) candidate genes.14 At least 1.6% of all 1231 with NEDBAF associated with RAC3 abnormalities. RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3311 Materials and methods hybridization (aCGH) (Patients 1, 2, 4, 5, 6, 7, 8, 9 and 10) or CNV detection on exome sequencing data (Patient 3) (Supplementary Ethics statement material). The study involving human participants was conducted according to the guidelines of the Declaration of Helsinki. Ethical review and Variant identification and analysis approval were obtained locally and informed consent was obtained The identified variants were filtered according to minor allele fre- from the parents or legal guardians of all the enrolled participants. quency ≤0.001 in Genome Aggregation Database (gnomAD, https:// For the animal experiments, we followed the fundamental guide- gnomad.broadinstitute.org),23 conservation (Genomic Evolutionary lines for proper conduct of animal experiments and related activity Rate Profiling—GERP, http://mendel.stanford.edu/SidowLab/downloads/ in academic research institutions under the jurisdiction of the gerp/) and predicted impact on protein function. In silico prediction Ministry of Education, Culture, Sports, Science, and Technology tools were used for the interpretation of candidate variants, including (Japan). All protocols for animal handling and treatment were re- combined annotation dependent depletion (CADD, https://cadd.gs. viewed and approved by the Animal Care and Use Committee of washington.edu), Mutation Taster (http://www.mutationtaster.org), Institute for Developmental Research, Aichi Developmental Sorting Intolerant From Tolerant (SIFT, https://sift.bii.a-star.edu.sg) Disability Center (approval number: 2019-013). and Polyphen-2 (http://genetics.bwh.harvard.edu/pph2/). Ultimately, the American College of Medical Genetics and Genomics and the Patient enrolment and ascertainment Association for Molecular Pathology (ACMG-AMP) guidelines were Patients were recruited through international collaboration, also used to classify the candidate variants based on their predicted using GeneMatcher,21 from several clinical and research centres pathogenicity.24 All RAC3 variants are reported according to the in Canada, France, Germany, Italy, Spain and the USA (further de- NM_005052.3 transcript (NP_005043.1) (https://www.ncbi.nlm.nih. tails available in the Supplementary material). Patients were evalu- gov/nuccore/NM_005052.3). Further details are available in the ated by paediatricians with expertise in neurological disorders, Supplementary material. paediatric neurologists and geneticists with expertise in the field of neurogenetics. Vineland Adaptive Behavior Scale (VABS, third Plasmids edition) tests were also administered for adaptive behaviour as- RAC3 was amplified by PCR from a cDNA pool of human glioblast- sessment to Patients 4, 5 and 8. Clinical and molecular data of oma U251MG cells and cloned into the pCAG-Myc vector. The 11 Patient 6, who was partially described in a previous report,20 were variants, RAC3-G12R, -P29L, -P34R, -A59G, -G60D, -Q61L, -E62del, critically reviewed. Brain MRI scans, locally performed for patient -E62K, -D63N, -Y64C and -K116N, were generated by site-directed care, were reviewed by an expert paediatric neuroradiologist mutagenesis using KOD-Plus Mutagenesis kit (Toyobo Inc) with (M.S.). Previously published neuroimaging studies were also thor- pCAG-Myc-RAC3 as a template. RAC3 and the 11 variants were oughly reviewed for comparison and delineation of the neuroradio- also constructed into pTriEx-4 vector (Merck). pRK5-Myc-PAK1 logical spectrum. (p21-activated kinase 1)-KA, a kinase-negative variant with a single amino acid substitution p.K299A and pRK5-Myc-MLK2 (mixed lin- Previously reported cases assessment eage kinase 2/MAPKKK10)-KN, a kinase-negative variant lacking All the articles indexed in PubMed (https://pubmed.ncbi.nlm.nih. aa 139–183, were kindly provided by Prof. Alan Hall (University gov/?term=RAC3&sort=date) between April 2019, when RAC3 var- College London, UK), and subcloned into the pCAG-Flag vector. iants were first associated with NEDBAF by Costain et al.,18 and RAC-binding regions (RBRs) in human PAK1 (aa 67–150), human May 202122 were retrieved using the terms ‘RAC3’, ‘Rac3 GTPase’ MLK2 (aa 401–550), human IRSp53 (aa 2–228), rat N-WASP (aa and ‘neurodevelopmental syndrome’. All the articles concerning 191-270), mouse RTKN (Rhotekin) (aa 10–100) and human ROCK the molecular, clinical, and neuroradiological spectrum associated (Rho-associated coiled-coil containing protein kinase 1) (aa 67– with NEDBAF were thoroughly reviewed. Inclusion criteria for pre- 150) were amplified by PCR from a cDNA pool of U251MG cells, viously published patients included clinical data availability, un- rat brain or mouse brain, and constructed into the pGS21a ambiguous identification of pathogenic/likely pathogenic RAC3 vector (GenScript). For gene transcription analysis, pGL4.74[hRluc/ variants and lack of duplication from other previous reports. TK] (control reporter plasmid), pGL4.44[luc2P/AP1-RE/Hygro] Ambiguous clinical presentation not consistent with NEDBAF and (AP1-luciferase reporter plasmid) and pGL4.32[luc2P/NF-κB-RE/ inconclusive genetic testing were exclusion criteria. Hygro] (NFkB-luciferase reporter plasmid) were purchased by Promega. The SRF reporter element, 5′ -CAGACAGACGTGTTCTT Next generation sequencing AAGTCCATATTAGGACATCTACCATGTCCATATTAGGACATCTACT ATGTCCATATTAGGACATCTTGTATGTCCATATTAGGACATCTAAAA We investigated 10 individuals presenting with syndromic NDD TGTCCATATTAGGAC-3′ , was inserted into pGVB (Toyo, Inc) to yield and neuroimaging abnormalities. Genomic DNA was extracted SRF-luciferase reporter plasmid. All constructs were verified by from peripheral blood leucocytes using standard protocols DNA sequencing. (Supplementary material). Different next generation sequencing (NGS) approaches were employed, including trio-exome sequen- Antibodies and histochemical reagents cing (Trio-ES) (Patients 1–3, 8 and 10), singleton exome sequencing (Patients 4, 6 and 9), trio-genome sequencing and RNA sequencing The following antibodies were used: anti-GFP (Medical and (Patient 5) and a NGS panel including 480 intellectual disability Biological Laboratories Cat #598, RRID: AB_591819 or Nacalai genes (Patient 7). Sanger sequencing was performed for the valid- Tesque, Cat #04404-84, RRID: AB_10013361), anti-NeuN (Millipore, ation of the candidate variants and the parental segregation ana- Cat #MAB377, RRID:AB_2298772) and anti-Myc (Medical and lysis. The presence of copy number variants was investigated in Biological Laboratories, Cat #M047-3, RRID: AB_591112). Alexa all subjects, either through array comparative genomic Fluor 488 and 568 (Invitrogen) were used as secondary antibodies. 3312 | BRAIN 2022: 145; 3308–3327 M. Scala et al. 4′ ,6-diamidino-2-phenylindole (DAPI; Nichirei Bioscience) was Assay of SRF, AP1 and NFkB-mediated gene used for staining DNA. Rhodamine-phalloidin (Invitrogen) was transcription used for staining filamentous actin. Assays were performed as described.28 The control reporter vec- tor was co-transfected into COS7 cells seeded on 24-well plates with the indicated RAC3 expression plasmid (0.1 μg/well) to- GTP-hydrolysis and GTP/GDP-exchange assays gether with the SRF-, AP1- or NFkB-luciferase reporter plasmid. Preparation and purification of His-tag-fused RAC3 and the 11 var- After transfection, cells were washed once with phosphate- iants were carried out according to the manufacturer’s instruc- buffered saline (PBS) and lysed with the passive lysis buffer ac- tions. To assess the basal GTP/GDP-exchange reactions, release of cording to the manufacturer’s instructions. Luciferase activities methylanthraniloyl (mant)-GDP (Sigma-Aldrich) was measured.25 were determined with the dual-luciferase reporter assay system Briefly, loading of RAC3 proteins with mantGDP was performed by in- (Promega). cubation for 90 min at 20°C in mantGDP loading buffer [20 mM HEPES-NaOH (pH 7.5), 50 mM NaCl, 0.5 mM MgCl2, 5 mM EDTA, 1 mM dithiothreitol, 20-fold excess of mantGDP] and subsequently In utero electroporation adding 10 mM MgCl2 (final concentration) to stop the reaction. The surgery on pregnant mice and embryo manipulation in the Remaining non-bound nucleotide was removed from the incuba- uterus were performed as previously described.29,30 At E14, preg- tion mixture with a NAP-5 column (GE Healthcare Life Sciences). mant nant ICR mice provided by Japan SLC were deeply anaesthetized GDP loaded GTPase was incubated with GppNHp (Abcam), with a mixture of three drugs: medetomidine (0.75 mg/kg), midazo- which has 100 times higher concentration than that of RAC3 pro- lam (4 mg/kg) and butorphanol (5 mg/kg).31 In utero electroporation teins, and fluorescence (365/450 nm) change was measured using was then performed in the specific-pathogen-free animal facilities an MTP-800 microplate reader (Corona). Intrinsic GTP-hydrolysis as described.32 Briefly, 1 µl of solution containing 0.1 μg of activity was then undertaken by directly measuring changes in pCAG-Myc vector (control), pCAG-Myc-RAC3, -D63N, -E62del, the GTP concentration with a luciferase-coupled GTPase assay kit -Y64C or -Q61L was injected with pCAG-EGFP (0.5 μg) into the lateral (GTPase-Glo Assay Kit, Promega) according to the manufacturer’s ventricle of mouse embryos with a glass micropipette made from a instructions.26 In short, 5 μM GTP and serially diluted RAC3 proteins microcapillary tube (GD-1; Narishige). After the embryo in the were incubated in GTPase/GAP buffer for 60 min at room tempera- uterus was placed between the tweezers-type disc electrode ture. Once the GTP-hydrolysis reaction was completed, the remain- (5 mm in diameter) (CUY650-5; NEPA Gene), electronic pulses ing GTP was converted to ATP by adding GTPase-Glo Reagent. (50 ms of 35 V) were charged five times at 450 ms intervals with Subsequently, generated ATP was detected by the luciferase/ an electroporator (NEPA21; NEPA Gene). In this method, plasmids luciferin-based reagent using a Turner Designs TD-20/20 lumin- are introduced into the somatosensory area, which is included in ometer (BMG Labtech). the parietal lobe. Brains were fixed at indicated postnatal days, sec- tioned, and then analysed. As for the quantification of distribution of GFP-positive cells in brain slices, coronal sections of cerebral cor- Cell culture and transfection tices containing the labelled cells were classified into three bins. COS7 (monkey kidney fibroblast-like cell) and primary hippocam- The number of labelled cells in each region of at least three slices pal neurons derived from embryonic day (E) 16.5 mice were cul- per brain was counted. A graphical timeline of the study design is tured as described.27 Transient transfection was carried out using shown (Supplementary Fig. 1). All experimental procedures were polyethyleneimine ‘MAX’ reagent (for COS7 cells) (Polysciences carried out in the daytime. Animals were neither excluded nor Inc) or Neon transfection system (for primary neurons) (Invitrogen). died during experimentation. Time-lapse imaging Pull-down assay of GTP-bound RAC3 and the variants After in utero electroporation at E14, organotypic coronal slices (250 µm-thickness) were prepared at E16 from the interventricular Glutathione S-transferase (GST)-fused RBRs of PAK1, MLK2, foramen with a microtome, placed on an insert membrane (pore IRSp53, N-WASP, RTKN and ROCK were expressed in Escherichia size, 0.4 µm; Millipore), mounted in agarose gel, and cultured. The coli BL21 (DE3) strain. Recombinant GST-proteins were purified dishes were then mounted in an incubator chamber (5% CO2 and according to the manufacturer’s instructions. COS7 cells were 40%O2, at 37°C) fitted onto an FV1000 confocal laser microscope transfected with pCAG-Myc-RAC3, RAC3-G12R, -P29L, -P34R, (Olympus) and the primary somatosensory cortex was examined -A59G, -G60D, -Q61L, -E62del, -E62K, -D63N, -Y64C or -K116N as described.33 Approximately 10–15 optical Z sections were ac- (1.0 μg/60 mm-dish). After 24 h, cells were lysed with the pull- quired automatically every 15 min for 15 h, and about 10 focal down buffer (50 mM Tris-HCl, pH7.5, 150 mM NaCl, 5 mM MgCl2, planes (50 µm-thickness) were merged to visualize the entire 0.1% SDS, 1% Nonidet P-40 and 0.5% deoxycholate), and insoluble shape of the cells. materials were removed by centrifugation. The supernatant was then incubated for 30 min at 4°C with Glutathione Sepharose 4B beads (GE Healthcare Life Sciences) with which GST-fused RBR Quantitative analysis of axon growth of PAK1, MLK2, IRSp53, N-WASP, RTKN or ROCK was bound. The beads were washed twice with the pull-down buffer, and bound For estimation of axon growth, GFP signal intensity of the callosal proteins were analysed by western blotting. Images were captured axons was measured at P0 or P7 using ImageJ software in distinct with an LAS-4000 luminescent image analyser (GE Healthcare Life regions (bin 1–3 or bin 1–4). The relative intensities of bins were nor- Sciences). malized with bin 1 as 1.0 and compared using R software. RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3313 Immunofluorescence life to 10 months and semiology was variable, including infantile spasms, clonic, myoclonic and tonic seizures. EEG showed focal/ Immunofluorescence analysis was conducted essentially as de- multifocal epileptic discharges, sometimes with secondary gener- scribed.27 Images of cultured cells were captured with a BZ-9000 alization. Hypsarrhythmia or slowed background were also ob- microscope (Keyence) or an LSM-880 confocal laser microscope served in the most severely affected individuals (Patients 1 and 5). (Carl Zeiss). As for the cortical slice staining, brains were embedded In all individuals, seizures were responsive to antiseizure medica- in 3% agarose, cut into sections (100 μm-thickness) with a vibra- tions, e.g. valproate, vigabatrin, phenobarbital or levetiracetam. tome and photographed with an LSM-880 confocal laser micro- Progressive microcephaly was detected in one individual (Patient scope. Acquired images were analysed with ImageJ to determine 2), with others having abnormal cranial shape in the form of plagioce- cell morphological descriptor and fluorescence intensity. phaly (Patients 1 and 4), brachycephaly (Patient 5) and scaphocephaly (Patient 7). Additional syndromic features were also evident in most Statistical analysis individuals. Among the musculoskeletal abnormalities, scoliosis (Patients 1, 3 and 10), vertebral defects (Patient 1), pes planus For all cell imaging experiments, cell counting and traces were as- (Patients 3 and 5) and joint laxity (Patients 3 and 4) were recurrent. sessed in a blinded manner by a technical staff who were not aware Ocular involvement consisted of horizontal nystagmus (Patients 1 of experimental conditions. Statistical significance for multiple and 5), strabismus (Patients 1, 3 and 4) and astigmatism with hyper- comparisons was determined by Dunnett’s or Tukey’s test metropia (Patients 1 and 3). Some individuals presented with genito- (https://cran.r-project.org/web/packages/multcomp/multcomp.pdf). urinary abnormalities, such as genital hypoplasia (Patient 6), follicular Comparisons between two groups were performed with Welch’s cysts (Patient 5) and enuresis (Patients 3 and 4). Endocrinological fea- t-test.34 P < 0.05 was considered statistically significant. Statistical tures were less common and encompassed truncal obesity (Patient 4), analyses were performed using R (https://intro2r.com/citing-r.html; precocious puberty (Patient 5) and hyperthyroidism with advanced https://cran.r-project.org/doc/FAQ/R-FAQ.htmlCiting-R). bone age (Patient 10). Respiratory involvement was observed in four patients, who experienced recurrent respiratory tract infections Data availability (Patients 3 and 7) and apnoeic episodes (Patients 9 and 10). The data that support the findings of this study are available from Although clinical features of RAC3-related disorder have been the corresponding authors, upon reasonable request. well documented in infancy, limited follow-up information is avail- able and little is known about the natural history of the condition. Herein we report the oldest known affected individual (Patient 10, Results aged 29 years), who seems to confirm three aspects of NEDBAF. First, the psychomotor delay may mildly improve over time, where- Patient report as regression is unusual (only observed in Patient 1 from our cohort, The study cohort consisted of 10 patients presenting with a syndrom- and limited to phonological skills). Second, seizures are responsive ic NDD characterized by a moderate-to-severe psychomotor delay af- to antiepileptic treatment and have a generally benign course. fecting the achievement of all the developmental milestones and Third, premature death does not appear to be part of the core clinical resulting in global cognitive impairment (Supplementary Table 1). spectrum, although the apnoeic episodes reported in a minority of Facial dysmorphism was present in 8/10 individuals and recurrent subjects are potentially life-threatening. Additional studies report- features included wide forehead with frontal bossing and high anter- ing large case series will determine if these observations are valid. ior hairline, prominent eyes with upslanted palpebral fissures, arched As summarized in Supplementary Tables 2 and 4, affected indivi- eyebrows, long eyelashes, midface hypoplasia, broad nasal bridge duals present with hypotonia, weak cry or poor sucking in the neo- and anteverted nares (Fig. 1A). Gross motor development was im- natal period. Feeding difficulties are common and may fail to thrive paired in all individuals, who also showed a significant deficiency in (62.5%). During the first year of life, global developmental delay expressive language, with three patients (Patients 1, 7 and 9) being with a severe involvement of language skills becomes apparent, nonverbal. A mild improvement of motor and verbal skills over but psychomotor arrest or regression are not typical. Neurological time was common, although a regression limited to the phonological examination shows hypotonia (93.7%), dyspraxia (40%), or spasticity skills was observed in Patient 1. In Patients 4, 5 and 8, VABS adminis- (25%) (Supplementary Fig. 2). Seizures occur in 43.7% of individuals, tration revealed an overall low level of communication skills (scores as early as the newborn period in two patients. Behavioural disorders range 34–68), daily living skills (scores range 27–56) and social skills and stereotyped movements are seen (both in 27.3% of subjects). and relationships (scores range 40–64) (Supplementary material). Dysmorphic features are prominent (81.3%) and suggestive of a pecu- Behavioural abnormalities mainly consisted of ASD, aggressive/self- liar facial gestalt. Although microcephaly is rare (12.5%), an abnormal injurious behaviour and hypersensitivity. Feeding difficulties due to cranial shape is observed in 31.2% of individuals, but rarely reflects weak sucking or dysphagia were particularly relevant in the neonatal true craniosynostosis. Associated syndromic features are particular- period, leading to failure to thrive in most individuals. ly relevant in some patients and include musculoskeletal (43.7%), Neurological examination revealed hypotonia in all patients genitourinary (37.5%), ocular (25%) and endocrinological (18.7%) ab- (Fig. 1B), frequently of neonatal onset, whereas appendicular spas- normalities. Of note, respiratory problems are present in 31.2% of in- ticity was detected in only two patients (Patients 2 and 7). dividuals. The clinical features of 16 individuals reported in this and Additional neurological features included dyspraxia with poor previous studies were summarized in Supplementary Table 2. visuospatial coordination (Patients 1–4), sleep disorders (Patients 5 and 7), irritability (Patient 6) and eyelid clonus (Patient 9). Neuroimaging Stereotyped movements of the mouth and upper limbs were pre- sent in three patients (Patients 1, 2 and 9) (Supplementary Videos Callosal anomalies were present in 10/10 participants, including 1 and 2). Five out of 10 individuals (Patients 1, 5, 6, 9 and 10) experi- hypoplasia with prevalent posterior involvement (eight cases), enced recurrent seizures. Age at onset ranged from the first day of thick and short morphology (one case, Patient 3) and partial 3314 | BRAIN 2022: 145; 3308–3327 M. Scala et al. Figure 1 Clinical and neuroimaging features of individuals with RAC3-related disorder. (A) Dysmorphic features in subjects harbouring pathogenic RAC3 variants. Patient 1 shows wide forehead with frontal bossing, upslanted palpebral fissures, depressed nasal bridge, anteverted nares, prominent philtrum and upper lip, and thick earlobes. Patient 2 is microcephalic and presents with palpebral ptosis, synophrys, deep-set eyes, prominent nasal tip with hypoplastic nasal wings, large and prominent ears and pointed chin. A round face with facial asymmetry, frontal bossing, high anterior hairline, upslanted palpebral fissures, depressed nasal bridge, micrognathia and posteriorly rotated ears can be observed in Patient 4. Patient 5 has facial hypo- tonia, frontal bossing, bilateral ptosis, proptosis, hypertelorism, shallow orbits, midface hypoplasia and pointed chin. Skeletal features consist of ara- chnodactyly and pes planus. Patient 6 has prominent eyes, long eyelashes, hypertelorism, midface hypoplasia, short nose, wide nasal bridge, anteverted nares, long philtrum, micrognathia and dental anomalies. She also has clinodactyly and foetal finger and toe pads. Patient 7 displays sca- phocephaly with a wide forehead and a high anterior hairline, broad nasal base, anteverted nares, retrognathia and full lips. A prominent forehead with frontal bossing, epicanthal folds, deep philtrum, micrognathia with horizontal chin crease and posteriorly rotated ears are evident in Patient 9. Patient 10 shows down-slanted palpebral fissures, simplified ear helices, sharp tip of the nose, flat and long philtrum and pointed chin. (B) Graphic illustration of the most common clinical findings in the reported study participants: the bar graph shows the distribution of the cardinal clinical features, with the grey lines representing not available (NA) data. (C) Brain MRI, sagittal T1-weighted and axial T2-weighted images of Patients 1–4 and of a normal indi- vidual ( first row) for comparison. The corpus callosum is shorter and thinner in all patients, especially at the level of the splenium (empty arrows), with thick genu in Patients 3 and 4. Note the absence of the anterior commissure in all patients. The volume of the cerebral white matter is reduced in all patients, with marked ventricular enlargement (asterisks) and septal defect (black thin arrow) in Patient 1. Diffuse dysgyria is present in all subjects, associated with polymicrogyria in the posterior regions in Patients 3 and 4 (arrowheads). Bilateral cerebellar folia abnormalities are noted in the inferior portions of the cerebellum in all patients, especially on the left side (thick arrows). There is hypoplasia of the midbrain in Patients 1 and 4 (white thin arrows) and a small pons in Patients 1, 2 and 4 (empty arrowheads). Mild enlargement of the subarachnoid spaces is noted in all patients, while bra- chycephaly is present only in Patients 3 and 4. agenesis (one case, Patient 10) (Fig. 1C). White matter thinning Identification and analyses of RAC3 variants and enlarged subarachnoid spaces with global reduction of the brain volume were noted in 8/10 and 9/10 cases, respectively. Based on allele frequency (<0.001 in gnomAD), involvement of con- Malformations of cortical development were noted in 7/10 parti- served amino acid residues, and predicted impact on protein func- cipants, including polymicrogyria involving the posterior regions tion, variants in RAC3 were the most plausible candidate variants in (five cases), diffuse dysgyria (four cases), and small grey matter the study cohort. NGS led to the identification of seven distinct var- heterotopias dispersed in the cerebral white matter (one case, iants, including the previously reported c.184G>Ap.E62K in Patient Patient 1). Cerebellar dysplasia with foliation abnormalities in- 7 (ClinVar ID: 425149)18. Most of the novel variants were missense: volving the inferior cerebellar hemispheres was detected in 6/ c.187G>A p.D63N (Patients 1 and 9); c.191A>Gp.Y64C (Patient 3); 10 participants, while brainstem anomalies characterized by a c.179G>Ap.G60D (Patient 4); c.34G>Cp.G12R (Patient 5); small midbrain and/or pons were seen in 5/10 cases. Chiari I c.348G>Cp.K116N (Patient 8) (Supplementary Fig. 3). All these var- anomaly was not detected in this series, but mild (<5 mm) caudal iants are rare, affect conserved residues (GERP scores 2.5–3.91) descent of the cerebellar tonsils was noted in three participants. (Supplementary Fig. 4), and are predicted to have a high pathogenic Neuroimaging features of 16 cases reported in this and previous impact (CADD score 24.3–27.6) (Supplementary Table 3). The mis- studies were summarized in Supplementary Table 2. sense variant c.176C>Gp.A59G previously detected in Patient 620 is RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3315 also predicted to be pathogenic by several in silico tools with recombinant RAC3 proteins, the assay documented a variably (Supplementary Table 3). In Patient 2, the novel in-frame deletion increased or decreased activity. When compared with wild type c.186_188delGGAp.E62del was detected. This variant results in a RAC3, log10(kobs: observed rate constant) was especially high for protein coding length change through the deletion of the conserved RAC3-K116N and moderately high for RAC3-Q61L, -P29L, -Y64C, Glu62 (GERP score 3.79). No copy number variants with a pathogenic -E62del, -P34R and -D63N in this order, indicating an accelerated ex- potential could be detected in the reported subjects change reaction (Fig. 3A and Supplementary Fig. 6A). In contrast, a (Supplementary Table 1 and Supplementary material). statistically significant decrease of log10(kobs) was observed for A common structural and biochemical feature across the small RAC3-G60D, indicating suppression of the exchange reaction GTPase superfamily is the G domain, which is defined by five G (Fig. 3A and Supplementary Fig. 6A). The exchange activity of boxes with certain structurally conserved amino acid residues RAC3-G12R, -A59G and -E62K did not show statistical difference in (Supplementary Fig. 3). The G1 box/P-loop (GxxxxGKS/T, where x comparison to the wild type (Fig. 3A and Supplementary Fig. 6A). is any amino acid) recognizes the β-phosphate and Mg2+ ion of guan- Then, GTP-hydrolysis activity was assayed for each variant. This ine nucleotides. The G2 box (T) positioned in the Switch I region in- was highly suppressed by p.Q61L, p.G60D and p.G12R, while moder- teracts with the γ-phosphate and Mg2+, while the G3 box (DxxGQ/H/ ate to weak suppression was observed for p.E62del and p.A59G T) localizing in the Switch II region is responsible for GTP-hydrolysis. (Fig. 3B and Supplementary Fig. 6B). We consider that p.G12R, Eventually, the G4 (T/NKxD) and G5 (C/SAK/L/T) boxes make specific p.G60D and p.Q61L primarily activated the protein by eliminating contacts with the guanine base. The previously reported p.P34R and the GTP-hydrolysis activity like oncogenic RAS proteins. In contrast, p.P29L variants18,19 lead to substitutions of evolutionarily conserved the activity of RAC3-P34R was higher than the wild-type (Fig. 3B and amino acids in the Switch I region, with p.P34R being also included Supplementary Fig. 6B). Of note, RAC3-P29L, E62K, -D63N, -Y64C and in the effector binding region. It is of note that all variations identi- -K116N did not affect GTP-hydrolysis activity when compared to the fied in this study occurred at residues highly conserved across spe- wild-type (Fig. 3B and Supplementary Fig. 6B). cies (Supplementary Figs 3 and 4). Among the RAC3 variants Based on the obtained results summarized in Fig. 3C, it is possible observed in our cohort, seven variants (p.A59G, p.G60D, p.Q61L, to categorize the biochemical properties of the 11 variants into three p.E62del, p.E62K, p.D63N and p.Y64C) are mapped to the Switch II re- groups based on GTP/GDP-exchange activity. Group I encompasses gion, which is recognized as a variation hot spot among RAC1, RAC3 RAC3-P29L, -P34R, -Q61L, -E62del, -D63N, -Y64C and -K116N. These and CDC42.16,35,36 In addition, the p.G12R and p.K116N are the first variants demonstrated a higher GTP/GDP-exchange activity. While variants affecting the conserved residues in the G1 and G4 boxes, re- GTP-hydrolysis activity of RAC3-P29L, -D63N, -Y64C and -K116N was spectively (Supplementary Figs 3 and 4). comparable to that of wild-type, RAC3-Q61L and -E62del showed low- er hydrolysis activity. We assume that these seven variants are prone to be in GTP-bound active conformation, since they exhibited high In vitro biological and biochemical activities of the 11 GTP/GDP-exchange activity with GTP-hydrolysis activity equivalent disorder causative RAC3 variants to or lower than that of wild-type RAC3. On the other hand, although To get further insights into the pathophysiological significance of RAC3-P34R has a statistically high GTP-hydrolysis activity, we assume RAC3 variants associated with NDDs, activation states were first ex- that the variant is activated due to the accelerated GTP/GDP exchange amined in vitro for the eight variants identified in the enrolled pa- activity in the context of an intracellular environment in which the tients, together with the recently reported p.P34R19, p.P29L and GTP concentration is substantially higher than that of GDP. It is not- p.Q61L18 variants. Of note, the p.Q61L variant is best characterized able that the substitution of the corresponding Pro34 in RAC2 with as a GTPase-defective constitutively active version of RAC1 in- His was suggested to activate the protein based on crystal structure volved in tumour cell metastasis and invasion.37–39 modeling.40 Group II includes RAC3-G12R, -A59G and -E62K, all show- When control pCAG-Myc vector was expressed in primary cul- ing a GTP/GDP-exchange activity similar to the wild-type. Since tured hippocampal neurons, cells were normally differentiated in RAC3-G12R and -A59G displayed suppressed GTP-hydrolysis activity, a time-dependent manner (Fig. 2A and N). Likewise, neurons ex- they were likely to prefer GTP-binding. As for RAC3-E62K with normal pressing wild type RAC3 were normally differentiated with single GTP-hydrolysis and GTP/GDP-exchange activities in vitro, it appears axon and numerous dendrites, although its expression induced la- that abnormal interaction with unidentified GEFs and/or GAPs holds mellipodia at the growth cone of axon and cell body (Fig. 2B). These RAC3-E62K in an active state in primary neurons. In this context, results indicate that the basal activity of exogeneous wild type RAC2-E62K has been reported to be hyper-activated through altered RAC3 had little effect on cell differentiation in vitro. In contrast, neu- GEF specificity and impaired GAP function.41 Eventually, Group III in- rons transfected with the 11 studied variants displayed cell round- cludes RAC3-G60D, which may prefer GTP-binding because of its very ing with typical lamellipodia formation and less neurite extension, low GTP-hydrolysis activity even if the GTP/GDP-exchange rate is also resulting in a prominent increase in the solidity (Fig. 2C–N). While low. Based on the results in Figs 2 and 3, we concluded that all the 11 the median of solidity was shifted upward when each variant was variations confer a gain-of-function phenotype with respect to the expressed, the degree was variant-dependent because cells expres- RAC3 function. Biological and biochemical diversity among variations sing some variants such as RAC3-P29L, -A59G, -G60D and -K116N may variously dysregulate RAC3 function in vivo, leading to variation- occasionally showed neurite extension (Supplementary Fig. 5). specific clinical phenotypes. Taken together, all the 11 variants appeared to facilitate cytoskel- etal reorganization to form lamellipodia to greater or lesser de- Possible interaction of the RAC3 variants with grees. The respective variants should dysregulate intracellular various downstream effectors signalling in variant-dependent manners and adversely affect neuronal morphology and function. While the patients described in this study share core NDD pheno- We then performed biochemical analyses and measured GTP/ types associated with callosal and cortical malformations, clinical GDP-exchange and GTP-hydrolysis activity. When the effects of features peculiar to each individual may be attributable to the pos- these 11 variations on GTP/GDP-exchange activity were examined ition and type of the involved amino acids changes. Since the 11 3316 | BRAIN 2022: 145; 3308–3327 M. Scala et al. Figure 2 Effects of the disease-causative 11 RAC3 variants on neuronal morphology in vitro. (A–M) Dissociated hippocampal neurons from E16 mice were co-electroporated with pCAG-EGFP (0.1 μg) together with pCAG-Myc (−), pCAG-Myc-RAC3 (wild-type, WT), -G12R, -P29L, -P34R, -A59G, -G60D, -Q61L, -E62del, -E62K, -D63N, -Y64C and -K116N (0.3 μg each). Cells were fixed at 3 days in vitro (DIV), and co-stained with anti-GFP (green), rhodamine- phalloidin (red) and DAPI (blue). Scale bar = 20 μm. (N) Quantification of (A–M). The morphological descriptor (solidity) of GFP-positive neurons (≥200 cells each) was shown in violin plot with boxplot. ‘Solidity’ is the ratio of the area of a cell to the area of a convex hull of the cell. Values of the ratio were calculated by ‘Analyze Particles…’ method in ImageJ (https://imagej.nih.gov/ij/docs/menus/analyze.html). The significance of difference between (−) and each variant was determined using Dunnett’s test. WT versus (−), P = 1; G12R versus (−), P < 1 × 10−10; P29L versus (−), P < 1 × 10−10; P34R versus (−), P < 1 × 10−10; A59G versus (−), P < 1 × 10−10; G60D versus (−), P < 1 × 10−10; Q61L versus (−), P < 1 × 10−10; E62del versus (−), P < 1 × 10−10; E62K versus (−), P < 1 × 10−10; D63N versus (−), P < 1 × 10−10; Y64C versus (−), P < 1 × 10−10; K116N versus (−), P < 1 × 10−10. ***P < 0.001. RAC3 variants analysed in this study are considered to be different- displayed relatively strong affinity to RAC3-Q61L, -Y64C and -G12R ly activated based on cell biological and biochemical analyses (Figs in this order, while weak to moderate affinity was detected for all 2 and 3), we looked into the downstream signalling pathways inter- other variants but -P34R (Fig. 4A). The RBR of ROCK, a key regulator acting with these variants. To this end, a pull-down assay was per- of actin cytoskeleton and cell polarity,52,53 exerted strong binding to formed to examine the interaction of respective RAC3 variants with RAC3-G12R, -Q61L, -E62del and -Y64C (Fig. 4A). While moderate inter- recombinant RBRs of various effectors including PAK1, MLK2, action was shown for RAC3-A59G, -G60D and -E62K, other variants IRSp53, N-WASP, ROCK and RTKN. Consequently, the RAC3 var- except RAC3-P34R showed a weak affinity (Fig. 4A). RTKN has been iants were observed to interact with these RBRs in variant- shown to regulate the septin cytoskeleton.54 Although RTKN was first dependent manners (Fig. 4A). The RBR of PAK1, which is crucial reported as a Rho-specific effector,55 we have found that RTKN inter- for neuronal migration and variated in patients with NDDs,42–46 acts with activated RAC3 but not RAC1 through the Rho-binding do- was associated with all the variants except RAC3-P34R. High affin- main. We thus asked whether this domain, which is referred to as ity was displayed for RAC3-G12R, -Q61L, -E62del and -K116N in this RBR here, interacts with the 11 variants. We determined that order, and low affinity was observed for RAC3-D63N (Fig. 4A). The RTKN-RBR interacted with the 11 variants, with particularly strong RBR of MLK2, which activates the JNK-MAP kinase pathway down- affinity to RAC3-Q61L and -Y64C (Fig. 4A). Taken together, given stream of RAC,47 significantly interacted with RAC3-Q61L and then that the respective RAC3 variants are activated versions, they are -Y64C, and moderate to weak affinity was observed for all other var- strongly suggested to hyper-activate different sets of effectors in iants (Fig. 4A). The RBR of IRSp53, an adaptor protein acting at the variant-dependent manners and, thereby, dysregulate specific membrane-actin interface and related to the formation of filopodia downstream signalling pathways in context-dependent manners. and lamellipodia,48,49 showed relatively high affinity towards Full cropped data of western blotting were depicted in RAC3-Q61L, -Y64C and -G12R (Fig. 4A). Although RAC3-E62del and Supplementary Fig. 7. -K116N had moderate affinity, a very weak interaction was ob- We next examined signal transduction pathways underlying the served with other variants (Fig. 4A). The RBR of N-WASP, a regulator aberrant activation states caused by these variations. We focused on of actin polymer reorganization in the cytoplasm and nucleus,50,51 SRF-, NFkB- and AP1-mediated gene expression, since their related RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3317 moderate NFkB-activation (Fig. 4C). Then, when we looked into the effects on AP1-mediated gene expression, each RAC3 variant in- duced the expression similar to that of NFkB; the RAC3 variants other than RAC3-A59G, -G60D, -Q61L and -E62K might share common sig- nalling pathways in terms of NFkB and AP-1 (Fig. 4C and D). These re- sults further support the hypothesis that the disease-causing RAC3 variants may drive different and/or common downstream signalling pathways, leading to variant-dependent dysregulation of cellular processes and, eventually, common as well as patient-specific clinic- al phenotypes. In vivo effects of the four RAC3 variants in the Switch II region on neuronal migration and morphology during corticogenesis Based on the data obtained from in vitro analyses, the disease- causative RAC3 variants are most likely to be activated and induce abnormal neuronal morphology. Since cell morphology is closely associated with migration, we examined the effects of the RAC3 var- iants on the migration of newly generated excitatory cortical neu- rons in vivo. We especially focused on the Switch II region and selected the p.Q61L, p.E62del, p.D63N and p.Y64C, since the region is a variation hotspot among RAC1, RAC3 and CDC42. Using an in utero electroporation-mediated acute gene transfer method, pCAG- Myc-RAC3 or pCAG-Myc vector harbouring each variant was co-electroporated with pCAG-EGFP into the progenitor cells in the ventricular zone of E14.5 embryonic brains, and localization of Figure 3 Characterization of activation states of the disease-causative transfected cells and their progeny was observed at P0. Neurons ex- 11 RAC3 variants in vitro. (A) Measurement of GDP/GDP-exchange activ- pressing the control vector or pCAG-Myc-RAC3 migrated normally ity. Recombinant His-tag-fused RAC3 (wild-type, WT) and the 11 var- iants (RAC3-G12R, -P29L, -P34R, -A59G, -G60D, -Q61L, -E62del, -E62K, to the superficial layer (bin 3; layers II/III) of the cortical plate -D63N, -Y64C and -K116N) were preloaded with fluorescent mantGDP (Fig. 5A). In contrast, most cells transfected with pCAG-Myc- and incubated with unlabelled GTP. Intrinsic nucleotide exchange was RAC3-Q61L, -E62del, -D63N or -Y64C remained in the ventricular measured and mantGDP-dissociation rates of wild-type and respective and subventricular zones and the intermediate zone (bin 1) variants were calculated as observed rate constants [kobs (×10−5 s−1)] (Fig. 5A). Quantitative analyses confirmed that each variant exhib- from the results in Supplementary Fig. 5A. Number of replicates, n ≥ 4. The significance of difference between wild-type and each variant was ited statistically similar effects (Fig. 5B). The result that the expres- determined using Dunnett’s test. G12R versus WT, P = 0.8305; P29L ver- sion of the wild type did not statistically affect neuronal cell sus WT, P < 0.001; P34R versus WT, P < 0.001; A59G versus WT, P = positioning indicates that the basal activity of RAC3 had no effects 0.7655; G60D versus WT, P = 0.0135; Q61L versus WT, P < 0.001; E62del on neuronal cell migration and that physiological balance be- versus WT, P < 0.001; E62K versus WT, P = 0.1018; D63N versus WT, P = 0.0132; Y64C versus WT, P < 0.001; K116N versus WT, P < 0.001. *P < 0.05, tween GTP- and GDP-bound states of RAC3 should be essential ***P < 0.001. (B) Measurement of GTP-hydrolysis activity. Intrinsic for the establishment of cortical architecture during corticogen- GTP-hydrolysis activity was analysed by direct measuring of changes esis. Based on western blotting analyses with cortical neurons in the GTP concentration with GTPase-Glo assay kit. EC50 (half maximal where Rac3 proteins were electroporated, the expression level of effective concentration) was estimated from the sigmoidal fitting curve each protein was found to be comparable and lower than endogen- in Supplementary Fig. 5B. Number of replicates, n ≥ 3. P-value was calcu- lated as in (A). G12R versus WT, P < 1 × 10−04; P29L versus WT, P = 0.174; ous Rac3 (Supplementary Fig. 8). P34R versus WT, P < 1 × 10−04; A59G versus WT, P = 0.012; G60D versus Although cells expressing respective variants were dominantly WT, P < 1 × 10−04; E62del versus WT, P < 1 × 10−04; E62K versus WT, P = distributed at bin 1, a small portion of such neurons still reached 0.514; D63N versus WT, P = 0.743; Y64C versus WT, P = 0.952; K116N ver- the superficial layer of cortical plate (Fig. 5A and B), perhaps due sus WT, P = 0.833. *P < 0.05, ***P < 0.001. P.C. = positive control. (C) Summary of the results in A and B. → = no change; ↑ = activity in- to a relatively low amount of the expression vector incorporated. creased; ↓ = activity decreased. Given that transfection efficiency into each cell depends on the size of the cell surface area which is physically exposed to the ven- tricular lumen (CSF) where plasmids are injected, neurons incorp- signalling pathways include Rho-family proteins and MAP ki- orating low amount of the expression vector are supposed to nases.56–58 When effects of respective variants on SRF-dependent undergo partial effects of the variant. gene transcription were assessed, prominent transcriptional activa- When long-term effects of expression of the four RAC3 variants tion was observed in cells expressing RAC3-Q61L, while moderate ac- were examined, we noticed the formation of neuronal cell clusters tivation was induced by RAC3-G12R, -A59G, -E62del, -D63N and in the ventricular and subventricular zones in developing cerebral -Y64C (Fig. 4B). In contrast, RAC3-P29L, -P34L, -G60D and -K116N cortex at P7 (Fig. 5C). The cells incorporated in the cluster were posi- had marginal effects. Meanwhile, NFkB-dependent gene transcrip- tive for NeuN, indicating they were differentiated at abnormal posi- tion was highly increased by RAC3-E62del, -G12R and -Q61L tions. These cells also extended neurites in the cluster (Fig. 5C). The (Fig. 4C). Noteworthy, these three variants were observed to show results obtained indicate that the four variants prevent, rather than high affinity to PAK1-RBR (Fig. 4A). Other variants except for delay, cortical neuron migration, and that RAC3 plays a pivotal role RAC3-G60D and -K116N, which showed no effects, demonstrated in neuronal migration. 3318 | BRAIN 2022: 145; 3308–3327 M. Scala et al. Figure 4 Possible interaction of the disease-causative 11 RAC3 variants with various downstream effectors. (A) Interaction with the RBRs of PAK1, MLK2, IRSp53, N-WASP, ROCK and RTKN. COS7 cells were transfected with pCAG-Myc-RAC3 (wild-type, WT), -G12R, -P29L, -P34R, -A59G, -G60D, -Q61L, -E62del, -E62K, -D63N, -Y64C and -K116N (0.3 μg each). Cell lysates were prepared, and the pull-down assay was conducted as described in the ‘Materials and Methods’ section with respective GST-fused RBRs (5 μg each). Bound RAC3 proteins were detected by western blotting with anti-Myc. RAC3 activity was indicated by the amounts of RAC3 bound to respective GST-RBRs, and total cell lysates were also immunoblotted with anti-Myc for normalization (input). Relative band intensity was shown when the value of RAC3-Q61L was taken as 1.0. (B–D) Involvement of the RAC3 variations in SRF- (B), NFkB- (C) or AP1-dependent (D) gene transcription. COS7 cells were co-transfected with each luciferase expression vector together with pCAG-Myc-RAC3 (WT) or the expression vector for the 11 variants. Luciferase activity obtained with wild-type was taken as 1.0, and rela- tive activities are shown as box plot. Number of replicates, n = 4. The significance of difference between wild-type and each variant was determined using Dunnet’s test. (B) G12R versus WT, P < 0.001; P29L versus WT, P = 0.81326; P34R versus WT, P = 0.94861; A59G versus WT, P < 0.001; G60D versus WT, P = 0.99999; Q61L versus WT, P < 0.001; E62del versus WT, P = 0.00997; E62K versus WT, P = 0.27663; D63N versus WT, P < 0.001; Y64C versus WT, P < 0.001; K116N versus WT, P = 0.99999. (C) G12R versus WT, P < 1 × 10−04; P29L versus WT, P = 0.00159; P34R versus WT, P < 1 × 10−04; A59G versus WT, P < 1 × 10−04; G60D versus WT, P = 1.00000; Q61L versus WT, P < 1 × 10−04; E62del versus WT, P < 1 × 10−04; E62K versus WT, P < 1 × 10−04; D63N versus WT, P < 1 × 10−04; Y64C versus WT, P < 1 × 10−04; K116N versus WT, P = 1.00000. (D) G12R versus WT, P < 1 × 10−04; P29L versus WT, P = 0.00275; P34R versus WT, P < 1 × 10−04; A59G versus WT, P < 1 × 10−04; G60D versus WT, P = 0.00017; Q61L versus WT, P < 1 × 10−04; E62del versus WT, P < 1 × 10−04; E62K versus WT, P < 1 × 10−04; D63N versus WT, P < 1 × 10−04; Y64C versus WT, P < 1 × 10−04; K116N versus WT, P = 0.99999. **P < 0.01, ***P < 0.001. Time-lapse imaging of migration of cortical neurons subventricular zone and intermediate zone should be associated expressing the four RAC3 variants in the Switch II with impaired shape change into the bipolar status (multipolar- region bipolar transition). We thus carried out time-lapse imaging to further investigate the morphology of cells stuck in the subventricular zone Newborn cortical neurons generated at the ventricular zone primarily and intermediate zone during corticogenesis. To this end, ventricular exhibit multipolar shapes in the lower intermediate zone, where cells zone progenitor cells were co-electroporated with pCAG-EGFP to- show a slow and irregular movement termed multipolar movement for 24 h.59 Neurons then transform into a bipolar shape with a lead- gether with wild-type RAC3- or respective variant-expression vectors ing process and an axon in the upper intermediate zone, move into the at E14.5, and cell migration was monitored from E16 for 15 h in the cortical plate, and exhibit a saltatory movement termed radial migra- subventricular zone, intermediate zone and lower cortical plate. tion toward pial surface. Given the tight correlation between Consequently, time-lapse imaging revealed clear differences in mi- cell movement and morphology, the abnormal accumulation and gration profiles between control cells and those expressing the re- cluster formation of the RAC3 variant-expressing neurons in the spective variants. In the control experiments with wild-type RAC3, RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3319 Figure 5 Effects of the four RAC3 variants in the Switch II region on excitatory neuron migration during corticogenesis. (A and C) Migration defects of neurons expressing each variant. pCAG-EGFP (0.5 μg) was co-electroporated in utero with pCAG-Myc (−), pCAG-Myc-RAC3 (wild-type, WT), -RAC3-Q61L, -E62del, -D63N or -Y64C (0.1 μg each) into the ventricular zone progenitor cells at E14.5. Coronal sections were prepared at P0 (A) or P7 (C). Coronal slices were double-stained with anti-GFP (white) and DAPI (blue) for A or triple-stained with anti-NeuN (red), anti-GFP (green) and DAPI (blue) for C. Boxed areas in the left panels were magnified in right panels (C). Scale bars = 50 μm (A), 200 μm (C, left) and 30 μm (C, right). (B) Quantification of the distribution of GFP-positive neurons in distinct regions of the cerebral cortex (bin 1–3) for each condition in A. Number of replicates, n ≥ 5. The significance of dif- ference between control (−) and each variant was determined using Dunnett’s test and shown in box plot. (Bin 1) WT versus (−), P = 0.9498; Q61L versus (−), P = 0.0045; E62del versus (−), P = 0.0172; D63N versus (−), P = 0.0119; Y64C versus (−), P = 0.0356. (Bin 2) WT versus (−), P = 0.413; Q61L versus (−), P = 0.535; E62del versus (−), P = 0.185; D63N versus (−), P = 0.439; Y64C versus (−), P = 0.133. (Bin 3) WT versus (−), P = 0.212; Q61L versus (−), P < 0.001; E62del versus (−), P < 0.001; D63N versus (−), P < 0.001; Y64C versus (−), P < 0.001. ***P < 0.001, **P < 0.01, *P < 0.05. while a considerable number of GFP-positive cells were positioned in that migration distance of the variant-expressing neurons at the inter- the intermediate zone, many cells advanced slowly toward the pial mediate zone–cortical plate boundary was much shorter than that of surface (Fig. 6A and B and Supplementary Video 3). In stark contrast, the control cells (Fig. 6C). The average migration speed of neurons ex- cells expressing RAC3-Q61L, -E62del, -D63N or -Y64C were stuck in pressing each RAC3 variant was also reduced when compared to that the intermediate zone. When looking closer, cells expressing of control cells (Fig. 6D). RAC3-E62del or -D63N remained round and appeared not to obtain the multipolar status (Fig. 6A and B and Supplementary Videos 4 Role of PAK1 as a downstream effector of RAC3 in and 5). On the other hand, cells expressing RAC3-Q61L or -Y64C neuronal migration in vivo were observed to transform into the multipolar shape but then fail to undergo the multipolar-bipolar transition (Fig. 6A and B and Since PAK1 is a downstream effector for RAC3 and activating PAK1 Supplementary Videos 6 and 7). Quantification analyses revealed variants cause NDD,45 dysregulation of this kinase is most likely to 3320 | BRAIN 2022: 145; 3308–3327 M. Scala et al. Figure 6 Time-lapse imaging analyses of migration of cortical neurons expressing the four disease-causative RAC3 variants in the Switch II region. Electroporation was performed as in Fig. 5. Analyses were repeated three times for each case and representative results were shown in (A and B). (A) Tracing of neurons expressing Myc-RAC3 (wild-type, WT), -RAC3-Q61L, -E62del, -D63N or -Y64C in upper intermediate zone (IZ)–lower cortical plate (CP). Migratory tracks of over 100 cells were demonstrated as colour lines. (B) Time-lapse imaging of neurons, which express Myc-RAC3 (WT) or the variants, migrating in the IZ–CP boundary. (C) Migration distance of neurons expressing Myc-RAC3 or the four variants. Over 100 cells selected in (A) were analysed. The distance was shown in violin plot with box plot. The significance of difference between wild-type and each variant was deter- mined using Dunnett’s test. Q61L versus WT, P < 2 × 10−16; E62del versus WT, P < 2 × 10−16; D63N versus WT, P < 2 × 10−16; Y64C versus WT, P < 2 × 10−16. ***P < 0.001. (D) Migration velocity of cells expressing Myc-RAC3 or the four variants. Over 100 cells selected in (A) were analysed. The velocity was shown in violin plot with box plot. The significance of difference between wild-type and each variant was determined using Dunnett’s test. Q61L versus WT, P < 1 × 10−07; E62del versus WT, P < 1 × 10−07; D63N versus WT, P < 1 × 10−07; Y64C versus WT, P = 1.16 × 10−07. ***P < 0.001. RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3321 be involved in the pathogenesis of RAC3 variants. We thus investi- -D63N or -Y64C (Supplementary Fig. 10). In contrast, the phenotype gated the possible involvement of PAK1 in the migration defects by RAC3-Q61L was not rescued, as in the case of cortical neuron mi- caused by the four RAC3 variations localized in the Switch II region. gration (Supplementary Fig. 10). These results strongly suggest that When pCAG-EGFP was co-electroporated with pCAG-Myc-RAC3- abnormal PAK1 activation contributes to the impaired axon elong- D63N, -E62del or -Y64C, together with pCAG-Flag-PAK1KA encod- ation by p.E62del, p.D63N and p.Y64C variations, which leads to the ing a kinase-negative version of PAK1, the positional defects of corpus callosum dysgenesis, while yet unidentified effector(s) is in- GFP-positive cells were partially rescued at P0 (Fig. 7A–C). The ob- volved in the axon phenotype by p.Q61L variation. Further analyses served rescue effects by PAK1KA confirmed that these three var- are required to determine if the observed impaired axon growth is a iants cause a hyper-activation of PAK1, which is supposed to be a primary phenotype or secondary to migration defects. crucial pathogenic mechanism of RAC3-related disorder caused by these three variations. To confirm the variant-mediated activa- tion of PAK1, these three variants were expressed with PAK1 in COS7 cells, or without PAK1 in primary cultured cortical neurons. Discussion Consequently, activation of exogenous and endogenous PAK1 We identified seven distinct de novo missense variants (five novel, was observed in COS7 cells and cortical neurons, respectively one recurrent) and a novel de novo in-frame deletion in RAC3 in 10 pa- (Supplementary Fig. 9A, B, D and E). In addition, when these variants tients presenting with NEDBAF, which is a complex syndromic NDD were electroporated into embryonic mice brains, endogenous PAK1 characterized by a moderate to severe psychomotor delay leading to activation was again detected in cortical neurons (Supplementary intellectual disability, associated with peculiar neurological and Fig. 9C and F). On the other hand, PAK1KA had no effects on the mi- extra-neurological features. The potential contribution of pathogen- gration defects produced by RAC3-Q61L under the same experimen- ic copy number variants to the reported phenotypes was excluded in tal conditions (Fig. 7D). This result implicates the possibility that all subjects (Supplementary Table 1). Regarding the neuroimaging other downstream effector(s) is also involved in the pathogenicity features of RAC3 deficiency in this study, the most frequent findings of RAC3-related disorder. To answer this question, we selected an- were abnormalities in corpus callosum and cortical development, other kinase MLK2, an activator for broad MAP kinase cascades in- collectively noted in 100% and 90% of the individuals reported so cluding JNK (c-Jun N-terminal kinase), ERK (extracellular far, respectively (Supplementary Table 2). In particular, the corpus signal-regulated kinase) and p38,47 as the next candidate effector callosum was found to be frequently dysplastic, with thinner sple- molecule involved in RAC3-Q61L-mediated pathophysiological nium and/or a thick genu, indicating the presence of axonal growth mechanism. However, MLK2 failed to rescue the aberrant migration anomalies. On the other hand, cortical malformations included dif- phenotype by RAC3-Q61L (Fig. 7D), indicating that yet unidentified fuse dysgyria, polymicrogyria and small grey matter heterotopias, downstream effector(s) plays a crucial role in the variation-mediated showing remarkable evidence of abnormal neuronal migration. signalling dysregulation. Although disruption of corticogenesis has been presumed to be cru- cial for the pathogenesis and pathophysiology of RAC3-related dis- order, the underlying molecular mechanisms remained unknown. Effects of the four RAC3 variants in the Switch II We thus investigated the pathophysiological significance of RAC3 region on axon development during corticogenesis variants causing NEDBAF during cortical development in vitro and Anomalies of corpus callosum in subjects with RAC3-related dis- in vivo. order, such as hypoplasia and agenesis, reflect defects in axon We assume that the pathogenic effects of the seven variants in growth (Fig. 1C). Specifically, an anomaly at the level of the sple- the Switch II region (p.A59G, p.G60D, p.Q61L, p.E62K, p.E62del, nium strongly suggests impaired axon elongation of pyramidal p.D63N and p.Y64C) are related to the aberrant interactions with neurons in layer II/III and V of the temporal, parietal and occipital the regulatory proteins in variant- and context-dependent man- lobes. In this context, the in utero electroporation method is avail- ners, since the region is essential for the physical association able to examine axon elongation from pyramidal neurons in layer with GEFs and GAPs. Being positioned in the effector-binding loop II/III in the parietal lobe during mouse brain corticogenesis. When (aa32–41), which is overlapping with the Switch I region, the the wild-type of RAC3 (as a control), RAC3-Q61L, -E62del, -D63N or p.P34R variant might impact the interaction with downstream ef- -Y64C were introduced into the ventricular zone progenitor cells fectors as well as GEFs and GAPs. Indeed, when compared to other at E14.5 and axon bundle of corpus callosum was visualized at P0, variants, RAC3-P34R displayed weaker affinities to RBRs of all ef- neurons expressing RAC3-Q61L, -E62del and -Y64C did not project fector molecules tested (Fig. 4A). Notably, the patient harbouring axons under the conditions where control axon bundle reached p.P34R presented syntelencephaly, a rare brain malformation char- the midline [Fig. 8A(i) and (ii)]. On the other hand, axon elongation acterized by an abnormal midline connection of the cerebral hemi- was observed but resulted in significantly delayed in neurons ex- spheres, suggesting a role of RAC3 in the cleavage process of pressing RAC3-D63N [Fig. 8A(i) and (ii)]. When we further analysed prosencephalon.19 Meanwhile, p.G12R and p.K116N affect residues the long-term effects at P7, control neurons extended the axon bun- conserved in most small GTPases within the G1 and G4 boxes, re- dle into the contralateral white matter, whereas axon elongation spectively. In line with the different roles of these boxes in the could not be detected yet for cells expressing RAC3-Q61L, -E62del interaction with guanine nucleotides, these two variants may affect and -Y64C [Fig. 8B(i) and (ii)]. Notably, although axons from the biochemical properties distinctly. Indeed, while the p.G12R vari- hemisphere containing RAC3-D63N-expressing cells eventually ation activated the protein by suppressing GTP-hydrolysis activity, reached the contralateral white matter at P7, the bundle was ob- p.K116N facilitated GTP/GDP-exchange reaction. Collectively, bio- served to be thinner compared to the control neurons [Fig. 8(i)]. chemical analyses strongly suggest that the 11 RAC3 variants ana- Then, rescue effects by dominant negative PAK1 on the defects in lysed show ‘gain-of-function’ phenotypes; they preferentially bind axon growth were examined in vivo. When PAK1KA was co- GTP in variant type-specific modes and are hyper-activated in vari- expressed with RAC3-Q61L, -E62del, -D63N or -Y64C, the defective ous degrees. This hypothesis is consistent with the results that all phenotypes were at least partially rescued for RAC3-E62del, the RAC3 variants exhibited lamellipodia formation with cell 3322 | BRAIN 2022: 145; 3308–3327 M. Scala et al. Figure 7 Rescue effects by a dominantly negative PAK1 on migration defects caused by the four variants in the Switch II region. pCAG-Myc-RAC3-E62del (A), -D63N (B) or -Y64C (C) (0.1 μg each) was co-electroporated with pCAG-EGFP (0.5 μg) together with pCAG-Flag vector (1.0 μg, control) (left) or pCAG-Flag-PAK1KA (1.0 μg) (right). (D) pCAG-Myc-RAC3-Q61L (0.1 μg) was co-electroporated with pCAG-EGFP (0.5 μg) together with 1.0 μg each of pCAG-Flag vector (control) (left), pCAG-Flag-MLK2KN (middle) or pCAG-Flag-PAK1KA (right). Analysis was done as in Fig. 5. Quantification results of the distribution of GFP-positive neurons in distinct regions (bin 1–3) of the cerebral cortex were shown in boxplot for each con- dition. Number of replicates, n ≥ 5. The significance of difference between control and each rescue condition was determined using Welch’s t-test (A–C) or Dunnett’s test (D), and shown in boxplot. (A, bin 1) E62del versus +PAK1KA, P = 0.005591. (A, bin 2) E62del versus +PAK1KA, P = 0.0006076. (A, bin 3) E62del versus +PAK1KA, P = 0.01177. (B, bin 1) D63N versus +PAK1KA, P = 0.0007075. (B, bin 2) D63N versus +PAK1KA, P = 0.0009054. (B, bin 3) D63N versus +PAK1KA, P = 0.001021. (C, bin 1) Y64C versus +PAK1KA, P = 0.005963. (C, bin 2) Y64C versus +PAK1KA, P = 0.04256. (C, bin 3) Y64C versus +PAK1KA, P = 0.003158. (D, bin 1) Q61L versus +PAK1KA, P = 0.338; Q61L versus +MLK2KN, P = 0.807. (D, bin 2) Q61L versus +PAK1KA, P = 0.311; Q61L versus +MLK2KN, P = 0.993. (D, bin 3) Q61L versus +PAK1KA, P = 0.393; Q61L versus +MLK2KN, p = 0.673. ***P < 0.001, **P < 0.01, *P < 0.05; Scale bar = 50 μm. rounding when overexpressed in primary cultured hippocampal quantitatively different manner, which may contribute to the vari- neurons. able phenotypes observed in patients with RAC3-related disorder. Although the 11 RAC3 variants are supposed to be activated, it is To clarify the pathophysiological significance of RAC3 variations surprising that these variants displayed different affinities to RBRs in vivo, we focused on p.D63N, p.E62del, p.Y64C and p.Q61L in the of various downstream effectors, including PAK1, MLK2, IRSp53, Switch II region, a variation hotspot not only for RAC3 but also for N-WASP, ROCK and RTKN. Pull-down assay with PAK1-RBR has so RAC1 and CDC42. Although these four variants should affect differ- far been conducted to assess the ‘activation state’ of RAC by meas- ent signalling pathways in vitro, their expression in cortical neurons uring the amount of the GTP-bound form co-precipitated, under the resulted in similar phenotypes in vivo: severe defects in neuronal tacit recognition that the interaction with an RBR of any effector migration, mispositioning, and eventual cluster formation in the molecule precisely reflects the GTP-bound activated state. intermediate and subventricular zones, as well as defects in axon However, the results obtained here indicate that the RAC3-GTP extension to the contralateral hemisphere during corticogenesis. amount precipitated by the RBR of PAK1 or other effectors was As for the migration phenotypes, time-lapse imaging revealed dif- not necessarily correlated with the activation state of RAC3. We ferences among the analysed variants. Newborn neurons expres- thus suggest that disease-causative RAC3 variants even within sing RAC3-Q61L or -Y64C appeared to become multipolar but did the same structural/functional domains differently modulate the not transform into the bipolar status, whereas cells expressing activation state of RAC3, and show specific spectra in the inter- RAC3-E62del or -D63N failed to become even multipolar. Further action with effectors, leading to abnormal upregulation of relevant analyses are required to elucidate the pathophysiological meaning effectors in variant type and context-dependent manners. The hy- of this phenotypic difference. Cluster formation by mislocalized peractivation of certain signalling pathways may underlie the NeuN-positive neurons should account for the pathogenic me- pathogenesis and pathophysiology of the divergent clinical fea- chanisms underlying the heterotopia and, partially, polymicrogyria tures observed in each patient. Meanwhile, each variant is possible and dysgyria, which represent the main neuronal migration/posi- to not only activate particular downstream effectors differently but tioning abnormality observed in RAC3-related disorder. Also, con- also exert dominant negative effects on some other signalling path- sidering the corpus callosum anomalies observed in affected ways simultaneously. In any case, intracellular signalling networks individuals, it is a reasonable conclusion that the four RAC3 var- should be affected by each variant both in a qualitatively and iants suppress axon elongation in vivo. The partial suppression of RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3323 Figure 8 Role of the four disease-causative variants in the Switch II region in the axon growth in vivo. [A(i) and B(i)] pCAG-GFP was co-electroporated with pCAG-Myc-RAC3 (wild-type, WT), -RAC3-Q61L, -E62del, -D63N or -Y64C (0.1 μg each) into the ventricular zone progenitor cells at E14.5. Coronal sections were prepared at P0 [A(i)] or P7 [B(i)], and stained with anti-GFP (white). DAPI staining (blue) of a slice was also shown (top). Scale bars = 300 μm [A(i)] and 500 μm (C). [A(ii) and B(ii)] The GFP intensity of the callosal axon was measured at P0 [A(ii)] or P7 [B(ii)] in different regions [bins 1– 3 for A(i) and bins 1–4 for B(i)], and then the relative intensities of bins were normalized with bin 1 as 1.0. Number of replicates, n ≥ 4. Different letters above bars represent significant differences, P < 0.05, according to a Tukey’s test. [A(i), bin 2] Q61L versus WT, P < 0.001; E62del versus WT, P < 0.001; D63N versus WT, P < 0.001; Y64C versus WT, P < 0.001; E62del versus Q61L, P = 1.0000; D63N versus Q61L, P = 0.0205; Y64C versus Q61L, P = 0.9756; D63N versus E62del, P = 0.0444; Y64C versus E62del, P = 0.9835; Y64C versus D63N, P = 0.0235. [A(i), bin 3] Q61L versus WT, P < 0.001; E62del versus WT, P < 0.001; D63N versus WT, P < 0.001; Y64C versus WT, P < 0.001; E62del versus Q61L, P = 0.997; D63N versus Q61L, P = 0.646; Y64C versus Q61L, P = 0.997; D63N versus E62del, P = 0.537; Y64C versus E62del, P = 0.976; Y64C versus D63N, P = 0.936. [B(i), bin 2] Q61L versus WT, P < 0.001; E62del versus WT, P < 0.001; D63N ver- sus WT, P < 0.001; Y64C versus WT, P < 0.001; E62del versus Q61L, P = 0.1794; D63N versus Q61L, P = 0.0132; Y64C versus Q61L, P = 0.0755; D63N versus E62del, P < 0.001; Y64C versus E62del, P = 0.9784; Y64C versus D63N, P < 0.001. [B(i), bin 3] Q61L versus WT, P < 0.001; E62del versus WT, P < 0.001; D63N versus WT, P < 0.001; Y64C versus WT, P < 0.001; E62del versus Q61L, P = 0.31921; D63N versus Q61L, P = 0.38265; Y64C versus Q61L, P = 0.51491; D63N ver- sus E62del, P = 0.00505; Y64C versus E62del, P = 0.99767; Y64C versus D63N, P = 0.01394. [B(i), bin 4] Q61L versus WT, P < 0.001; E62del versus WT, P < 0.001; D63N versus WT, P = 0.01014; Y64C versus WT, P < 0.001; E62del versus Q61L, P = 0.48422; D63N versus Q61L, P = 0.34330; Y64C versus Q61L, P = 0.45377; D63N versus E62del, P = 0.00869; Y64C versus E62del, P = 0.99996; Y64C versus D63N, P = 0.00904. 3324 | BRAIN 2022: 145; 3308–3327 M. Scala et al. axon growth by RAC3-D63N may explain the absence of ‘white mat- inhibit TRIO function and thereby inactivate RAC1. In contrast, ter thinning’ in the individual harbouring this variant. missense variations in the seventh spectrin domain of TRIO, a Since activating PAK1 variants have been shown to cause variation hotspot, resulted in RAC1 hyperactivation, further sup- NDDs,45 dysregulation of PAK1 may be involved in the migration porting the relevance of the tight control of RAC1 signalling dur- defects associated with the RAC3 variants. Indeed, the abnormal ing brain development.64 phenotypes caused by p.D63N, p.E62del and p.Y64C were rescued Unlike the case of RAC1, we concluded that all the 11 tested by a kinase-negative version of PAK1, PAK1KA. We thus concluded RAC3 variants act as GTP-bound active versions in varying degrees. that these three variants hyper-activated PAK1 and provoked de- From the results obtained in this and previous studies, the variant- fects in neuronal migration during corticogenesis, irrespective of dependent spatiotemporal dysregulation (hyper-activation or in- their different affinities to PAK1 in vitro. On the other hand, despite hibition) of subsets of common downstream effectors appears to the strong affinity, PAK1 did not appear to be related to the cause diverse phenotypic features in RAC1- and RAC3-related disor- RAC3-Q61L-mediated migration defects, since PAK1KA did not ex- ders. Given that variants affecting RAC3 function cause polymicro- ert rescue effects. These results strongly suggest that PAK1 is not gyria, heterotopia and dysgyria, which are only occasional in the sole molecule regulating cortical neuron migration down- RAC1-related disorder,16 we assume that RAC3 plays a central role stream of RAC3. in neuronal migration in vivo, although Rac3 knockout mice dis- De novo RAC1 missense variants have also been identified in played little phenotype in terms of cortical architecture, perhaps NDDs with global developmental delay/intellectual disability and due to compensation by Rac1.65 In contrast, RAC1 variants are asso- brain size abnormalities as core phenotypes (Mental Retardation ciated with abnormalities in brain size (extraordinary spread from autosomal dominant 48, MRD48, OMIM 617751).16 While Rac3 is −5 to +4.5 standard deviation of occipital-frontal circumference), mainly, if not exclusively, expressed in developing and adult neu- which are rare among RAC3 patients, suggesting a major role of rons, Rac1 is ubiquitously expressed from the early embryonic RAC1 in neurogenesis and/or apoptosis.15,16 Since the disruption stage. Despite the different expression profiles, the precise func- of synaptic functions is involved in the pathogenesis of global de- tional difference between RAC1 and RAC3 has not been investi- velopmental delay/intellectual disability, RAC1 and RAC3 are sup- gated. Rather, RAC1–3 have seemingly been equated in various posed to cooperate to build up the neuronal signalling network analyses, since they show very similar phenotypes in vitro, especial- and functions. ly when overexpressed.60 RAC1 and RAC3 are primarily divergent in Another Rho family protein CDC42 is essential for the control of the last nine carboxy-terminal residues, which include a polybasic cell polarity, migration, endocytosis, and cell cycle.66 De novo mis- region and an adjacent CAAX box, where the C is a Cys, the two A sense variations in CDC42 have been reported in patients with residues are aliphatic amino acids and the X can be any residue.5 clinically heterogeneous but overlapping phenotypes (Takenouchi- The CAAX box is posttranslationally modified and is crucial, to- Kosaki syndrome, OMIM 616737).35,67,68 Biochemical assays re- gether with the polybasic region, for the subcellular localization vealed that all the variants were activated as in the case of RAC3 of the protein.61 Given the highly homologous structure and com- variants.35,36 Considering that CDC42 and RAC proteins are strictly mon effector molecules, the differences in the pathogenic mechan- linked in their physiological functions through a variety of shared isms and clinical phenotypes associated with RAC1 and RAC3 downstream effectors, it is plausible that the clinical features of variants may be attributable to diverse subcellular distribution, at CDC42-related disorder overlap with those of RAC1- and least partially. We assume that differentially distributed RAC1 RAC3-related conditions. The intersection of the pathogenic me- and RAC3 variants may spatiotemporally hyper-activate interact- chanisms underlying RAC1-, RAC3-, CDC42-, and TRIO-related dis- able effectors in context-dependent manners. orders is reflected in their overlapping clinical phenotypes The RAC1 missense variants associated with MRD48 exert (Supplementary Fig. 11). their pathophysiological functions differently from RAC3.16 The In the present study, we identified six novel de novo pathogenic p.Y64D variant was determined to be constitutively activated, variants in RAC3 in unrelated individuals with NEDBAF. A variable since its expression in fibroblasts resulted in a rounding shape degree of biochemically activated states was observed for the novel and formation of lamellipodia.16 The p.C18Y and p.N39S variants and previously reported RAC3 variants tested. Activation of these were instead considered as dominant negative alleles, based on variants was also confirmed biologically with primary hippocampal fibroblast morphology analyses and in vivo zebrafish analyses.16 neurons. It is noteworthy that each variant displayed different affin- Meanwhile, three additional variants (p.V51M, p.P73L and ities to a variety of effector molecules in vitro, which may dysregulate p.C157Y) were categorized as neither active nor inactive versions downstream cellular pathways in variant-specific manners and con- and considered to have context-dependent effects.16 Although tribute to the pathogenesis of patient-specific clinical features as GTP/GDP-binding states were not determined biochemically, well as common ones. We then performed in vivo analyses of four these results show that (i) the variant position may determine variants (p.D63N, p.E62del, p.Y64C and p.Q61L) affecting the Switch the activation state of RAC1, ranging from dominant negative to II region, a variation hot spot. They prevented neuronal migration neutral to constitutively active; and (ii) the precise control of during corticogenesis, through dysregulation of PAK1-mediated sig- RAC1 activity is crucial for correct brain development. In this con- nalling pathway in some cases. It is notable that the abnormally po- text, pathogenic variants in TRIO, encoding a GEF which activates sitioned neurons eventually formed clusters in the ventricular and RAC1 through the first GEF domain (GEFD1), were shown to cause subventricular zones in developing cerebral cortex, which may be neurodevelopmental impairment, behavioural disturbances, an underlying mechanism of heterotopia, polymicrogyria and dys- microcephaly or macrocephaly and skeletal features [autosomal gyria. These four variations also caused defective axon development dominant intellectual developmental disorder with microcephaly in vivo, explaining their involvement in corpus callosum hypoplasia/ (MRD44, OMIM 617061) or macrocephaly (MRD63, OMIM agenesis. Collectively, we conclude that all the RAC3 variants identi- 618825)].62,63 Affected individuals with variations in TRIO present fied so far are gain-of-function, and variation-dependent activation with variable neurodevelopmental phenotypes. For example, of downstream signalling pathways may contribute to the heteroge- missense changes affecting the GEFD1 domain were found to neous phenotypes observed in individuals with RAC3-related RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3325 disorder. The results obtained in this study contribute to the under- Appendix 1 standing of the molecular machinery underlying this complex condi- tion, providing insights for the development of novel targeted drugs A full list of members and their affiliations appears in the aimed at interfering with the disease pathogenesis. Supplementary material. Undiagnosed Disease Network Web resources Gabrielle Brown, Manish J. Butte, Esteban C. Dell’Angelica, ClinVar; https://www.ncbi.nlm.nih.gov/clinvar Naghmeh Dorrani, Emilie D. Douine, Brent L. Fogel, Irma Combined Annotation Dependent Depletion (CADD); http://cadd.gs. Gutierrez, Alden Huang, Deborah Krakow, Hane Lee, Sandra washington.edu K. Loo, Bryan C. Mak, Martin G. Martin, Julian A. Martínez-Agosto, DECIPHER; https://decipher.sanger.ac.uk Elisabeth McGee, Stanley F. Nelson, Shirley Nieves-Rodriguez, Gene Cards; https://www.genecards.org Christina G.S. Palmer, Jeanette C. Papp, Neil H. Parker, Genecee Gene Matcher; http://www.genematcher.org Renteria, Janet S. Sinsheimer, Jijun Wan, Lee-kai Wang and Genome Aggregation Database (GnomAD); http://gnomad. Katherine Wesseling Perry. broadinstitute.org Genomic Evolutionary Rate Profiling (GERP); http://mendel. Telethon Undiagnosed Diseases Program stanford.edu/SidowLab/downloads/gerp/ Mutation Taster; http://www.mutationtaster.org NeXtProt; https:// Vincenzo Nigro, Nicola Brunetti-Pierri, Giorgio Casari, Gerarda www.nextprot.org Cappuccio, Annalaura Torella, Michele Pinelli, Francesco Online Mendelian Inheritance in Man; https://www.ncbi.nlm.nih. Musacchia, Margherita Mutarelli, Diego Carrella, Giuseppina gov/Omim Vitiello, Valeria Capra, Giancarlo Parenti, Vincenzo Leuzzi, Angelo PubMed; https://www.ncbi.nlm.nih.gov/pubmed Selicorni, Silvia Maitz, Sandro Banfi, Marcella Zollino, Mario RefSeq; https://www.ncbi.nlm.nih.gov/refseq Montomoli, Donatelli Milani, Corrado Romano, Albina Tummolo, SIFT; https://sift.bii.a-star.edu.sg Daniele De Brasi, Antonietta Coppola, Claudia Santoro, Angela Simons Foundation Autism Research Initiative: SFARI; https:// Peron, Chiara Pantaleoni, Raffaele Castello and Stefano D’Arrigo. www.sfari.org UniProt; https://www.uniprot.org UCSC Human Genome Database; https://www.genome.ucsc.edu References Varsome; https://varsome.com 1. Stankiewicz TR, Linseman DA. Rho family GTPases: Key players Ensembl Variant Effect Predictor (VEP); https://www.ensembl.org/ in neuronal development, neuronal survival, and neurodegen- info/docs/tools/vep/index.html eration. Front Cell Neurosci. 2014;8:314. 2. Moll J, Sansig G, Fattori E, Van Der Putten H. The murine rac1 gene: cDNA cloning, tissue distribution and regulated expres- Acknowledgements sion of rac1 mRNA by disassembly of actin microfilaments. Oncogene. 1991;6(5):863–866. We thank Mses Noriko Kawamura and Nobuko Hane for technical 3. Shirsat N V, Pignolo RJ, Kreider BL, Rovera G. A member of the assistance. We also acknowledge the Telethon Undiagnosed ras gene superfamily is expressed specifically in T, B and mye- Disease Program (TUDP) for technical and scientific support. loid hemopoietic cells. Oncogene. 1990;5(5):769–772. 4. Malosio ML, Gilardelli D, Paris S, Albertinazzi C, De Curtis I. Differential expression of distinct members of rho family Funding GTP-binding proteins during neuronal development: Identification of RAC1B, a new neural- specific member of the This work was supported in part by Japan Society for the Promotion family. J Neurosci. 1997;17(17):6717–6728. of Science (JSPS) Grant-in-Aid for Scientific Research (KAKENHI) 5. Haataja L, Groffen J, Heisterkamp N. Characterization of RAC3, a (B) (Grant Number JP19H03629), Grant-in-Aid for Scientific novel member of the Rho family. J Biol Chem. 1997;272(33):20384– Research (C) (Grant Number JP19K07059), Grant-in-Aid for 20388. Research Activity Start-up (Grant Number JP20K22888), 6. Albertinazzi C, Gilardelli D, Paris S, Longhi R, De Curtis I. Grant-in-Aid for Early-Career Scientists (Grant Number Overexpression of a neural-specific Rho family GTPase, JP21K15895), a Grant-in-Aid of the Practical Research Project for cRac1B, selectively induces enhanced neuritogenesis and Rare/Intractable Diseases from Japan Agency for Medical neurite branching in primary neurons. J Cell Biol. 1998;142(3): Research and Development (AMED) (15ek0109040h0002), and sup- 815–825. ported in part by the National Human Genome Research 7. Bolis A, Corbetta S, Cioce A, De Curtis I. Differential distribution Institute (UM1HG006542; U01HG011758). of Rac1 and Rac3 GTPases in the developing mouse brain: Implications for a role of Rac3 in Purkinje cell differentiation. Eur J Neurosci. 2003;18(9):2417–2424. Competing interests 8. Corbetta S, Gualdoni S, Ciceri G, et al. Essential role of Rac1 and Rac3 GTPases in neuronal development. FASEB J. 2009;23(5): The authors report no competing interests. 1347–1357. 9. Pennucci R, Tavano S, Tonoli D, Gualdoni S, de Curtis I. Rac1 and Rac3 GTPases regulate the development of hilar mossy cells by Supplementary material affecting the migration of their precursors to the hilus. PLoS One. Supplementary material is available at Brain online. 2011;6(9):e24819. 3326 | BRAIN 2022: 145; 3308–3327 M. Scala et al. 10. Vaghi V, Pennucci R, Talpo F, et al. Rac1 and Rac3 GTPases con- 30. Hamada N, Ogaya S, Nakashima M, et al. De novo PHACTR1 mu- trol synergistically the development of cortical and hippocam- tations in West syndrome and their pathophysiological effects. pal GABAergic interneurons. Cereb Cortex. 2014;24(5):1247–1258. Brain. 2018;141(11):3098–3114. 11. Dvorsky R, Ahmadian MR. Always look on the bright site of Rho: 31. Kawai S, Takagi Y, Kaneko S, Kurosawa T. Effect of three types of Structural implications for a conserved intermolecular inter- mixed anesthetic agents alternate to ketamine in mice. Exp face. EMBO Rep. 2004;5(12):1130–1136. Anim. 2011;60(5):481–487. 12. Azzarelli R, Kerloch T, Pacary E. Regulation of cerebral cortex 32. Tabata H, Nagata KI. Decoding the molecular mechanisms of development by Rho GTPases: Insights from in vivo studies. neuronal migration using in utero electroporation. Med Mol Front Cell Neurosci. 2015;8:445. Morphol. 2016;49(2):63–75. 13. de Curtis I. The Rac3 GTPase in neuronal development, neuro- 33. Hamada N, Negishi Y, Mizuno M, et al. Role of a heterotrimeric developmental disorders, and cancer. Cells. 2019;8(9):1063. G-protein, Gi2, in the corticogenesis: possible involvement in 14. Guo D, Yang X, Shi L. Rho GTPase regulators and effectors in aut- periventricular nodular heterotopia and intellectual disability. ism spectrum disorders: Animal models and insights for thera- J Neurochem. 2017;140(1):82–95. peutics. Cells. 2020;9(4):835. 34. Welch BL. The significnace of the difference between two 15. Zamboni V, Jones R, Umbach A, et al. Rho GTPases in intellectual means when the population variances are unequal. Biometrika. disability: From genetics to therapeutic opportunities. Int J Mol 1938;29(3-4):350–362. Sci. 2018;19(6):1821. 35. Martinelli S, Krumbach OHF, Pantaleoni F, et al. Functional dys- 16. Reijnders MRF, Ansor NM, Kousi M, et al. RAC1 missense muta- regulation of CDC42 causes diverse developmental phenotypes. tions in developmental disorders with diverse phenotypes. Am J Am J Hum Genet. 2018;102(2):309–320. Hum Genet. 2017;101(3):466–477. 36. Hamada N, Ito H, Shibukawa Y, et al. Neuropathophysiological 17. Lougaris V, Baronio M, Gazzurelli L, Benvenuto A, Plebani A. significance of the c.1449T> C/p.(Tyr64Cys) mutation in the RAC2 and primary human immune deficiencies. J Leukoc Biol. CDC42 gene responsible for Takenouchi-Kosaki syndrome. 2020;108(2):687–696. Biochem Biophys Res Commun. 2020;529(4):1033–1037. 18. Costain G, Callewaert B, Gabriel H, et al. De novo missense var- 37. Krengel U, Schlichting I, Scherer A, et al. Three-dimensional iants in RAC3 cause a novel neurodevelopmental syndrome. structures of H-ras p21 mutants: Molecular basis for their Genet Med. 2019;21(4):1021–1026. inability to function as signal switch molecules. Cell. 1990;62(3): 19. Hiraide T, Kaba Yasui H, Kato M, Nakashima M, Saitsu H. A de 539–548. novo variant in RAC3 causes severe global developmental delay 38. Alan JK, Lundquist EA. Mutationally activated Rho GTPases in and a middle interhemispheric variant of holoprosencephaly. J cancer. Small GTPases. 2013;4(3):159–163. Hum Genet. 2019;64(11):1127–1132. 39. Kazanietz MG, Caloca MJ. The Rac GTPase in cancer: From 20. White JJ, Mazzeu JF, Coban-Akdemir Z, et al. WNT signalling per- old concepts to new paradigms. Cancer Res. 2017;77(20): turbations underlie the genetic heterogeneity of robinow syn- 5445–5451. drome. Am J Hum Genet. 2018;102(1):27–43. 40. Lougaris V, Chou J, Beano A, et al. A monoallelic activating mu- 21. Sobreira N, Schiettecatte F, Valle D, Hamosh A. GeneMatcher: A tation in RAC2 resulting in a combined immunodeficiency. J matching tool for connecting investigators with an interest in Allergy Clin Immunol. 2019;143(4):1649–1653.e3. the same gene. Hum Mutat. 2015;36(10):928–930. 41. Arrington ME, Temple B, Schaefer A, Campbell SL. The molecular 22. Pellegatta M, Berti C, Hurley E, et al. Rac1 and Rac3 have opposite basis for immune dysregulation by the hyperactivated E62K functions in Schwann cells during developmental myelination. mutant of the GTPase RAC2. J Biol Chem. 2020;295(34):12130– Neurosci Lett. 2021;753:135868. 12142. 23. Lek M, Karczewski KJ, Minikel EV, et al. Analysis of protein- 42. Causeret F, Terao M, Jacobs T, et al. The p21-activated kinase is coding genetic variation in 60,706 humans. Nature. 2016;536- required for neuronal migration in the cerebral cortex. Cereb (7616):285–291. Cortex. 2009;19(4):861–875. 24. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the 43. Ohori S, Mitsuhashi S, Ben-Haim R, et al. A novel PAK1 variant interpretation of sequence variants: a joint consensus recom- causative of neurodevelopmental disorder with postnatal mendation of the American College of Medical Genetics and macrocephaly. J Hum Genet. 2020;65(5):481–485. Genomics and the Association for Molecular Pathology. Genet 44. Harms FL, Kloth K, Bley A, et al. Activating mutations in PAK1, Med. 2015;17(5):405–424. encoding p21-activated Kinase 1, cause a neurodevelopmental 25. Kanie T, Jackson P. Guanine nucleotide exchange assay using disorder. Am J Hum Genet. 2018;103(4):579–591. fluorescent MANT-GDP. Bio Protoc. 2018;8(7):e2795. 45. Zhang K, Wang Y, Fan T, Zeng C, Sun ZS. The p21-activated ki- 26. Mondal S, Hsiao K, Goueli SA. A homogenous bioluminescent nases in neural cytoskeletal remodeling and related neurologic- system for measuring GTPase, GTPase activating protein, and al disorders. Protein Cell. 2022;13(1):6–25. guanine nucleotide exchange factor activities. Assay Drug Dev 46. Horn S, Au M, Basel-Salmon L, et al. De novo variants in PAK1 Technol. 2015;13(8):444–455. lead to intellectual disability with macrocephaly and seizures. 27. Nishikawa M, Ito H, Noda M, Hamada N, Tabata H, Nagata K. Brain. 2019;142(11):3351–3359. Expression analyses of PLEKHG2, a Rho family-specific guanine 47. Nagata Ki, Puls A, Futter C, et al. The MAP kinase kinase kinase nucleotide exchange factor, during mouse brain development. MLK2 co-localizes with activated JNK along microtubules and Med Mol Morphol. 2021;54(2):146–155. associates with kinesin superfamily motor KIF3. EMBO J. 1998; 28. Nishikawa M, Nakano S, Nakao H, et al. The interaction between 17(1):149–158. PLEKHG2 and ABL1 suppresses cell growth via the NF-κB signal- 48. Miki H, Yamaguchi H, Suetsugu S, Takenawa T. IRSp53 is an es- ling pathway in HEK293 cells. Cell Signal. 2019;61:93–107. sential intermediate between Rac and WAVE in the regulation 29. Tabata H, Nakajima K. Efficient in utero gene transfer system to of membrane ruffling. Nature. 2000;408(6813):732–735. the developing mouse brain using electroporation: visualiza- 49. Nakagawa H, Miki H, Nozumi M, et al. IRSp53 is colocalised with tion of neuronal migration in the developing cortex. WAVE2 at the tips of protruding lamellipodia and filopodia in- Neuroscience. 2001;103(4):865–872. dependently of Mena. J Cell Sci. 2003;116(12):2577–2583. RAC3 and neurodevelopmental disorders BRAIN 2022: 145; 3308–3327 | 3327 50. Burbelo PD, Drechsel D, Hall A. A conserved binding motif de- 60. Aspenström P, Fransson Å, Saras J. Rho GTPases have diverse ef- fines numerous candidate target proteins for both Cdc42 and fects on the organization of the actin filament system. Biochem J. Rac GTPases. J Biol Chem. 1995;270(49):29071–29074. 2004;377(2):327–337. 51. Tomasevic N, Jia Z, Russell A, et al. Differential Regulation of 61. Roberts PJ, Mitin N, Keller PJ, et al. Rho family GTPase modifica- WASP and N-WASP by Cdc42, Rac1, Nck, and PI(4,5)P2. tion and dependence on CAAX motif-signaled posttranslational Biochemistry. 2007;46(11):3494–3502. modification. J Biol Chem. 2008;283(37):25150–25163. 52. Soriano-Castell D, Chavero A, Rentero C, et al. ROCK1 is a novel 62. Ba W, Yan Y, Reijnders MRF, et al. TRIO loss of function is asso- Rac1 effector to regulate tubular endocytic membrane formation ciated with mild intellectual disability and affects dendritic during clathrin-independent endocytosis. Sci Rep. 2017;7(1):6866. branching and synapse function. Hum Mol Genet. 2016;25(5): 53. Amano M, Nakayama M, Kaibuchi K. Rho-kinase/ROCK: A key 892–902. regulator of the cytoskeleton and cell polarity. Cytoskeleton 63. Pengelly RJ, Greville-Heygate S, Schmidt S, et al. Mutations spe- (Hoboken). 2010;67(9):545–554. cific to the Rac-GEF domain of TRIO cause intellectual disability 54. Ito H, Iwamoto I, Morishita R, et al. Possible role of Rho/Rhotekin and microcephaly. J Med Genet. 2016;53(11):735–742. signalling in mammalian septin organization. Oncogene. 2005; 64. Barbosa S, Greville-Heygate S, Bonnet M, et al. Opposite modula- 24(47):7064–7072. tion of RAC1 by mutations in TRIO Is associated with distinct, 55. Reid T, Furuyashiki T, Ishizaki T, et al. Rhotekin, a new putative domain-specific neurodevelopmental disorders. Am J Hum target for Rho bearing homology to a serine/threonine kinase, Genet. 2020;106(3):338–355. PKN, and rhophilin in the Rho-binding domain. J Biol Chem. 65. Corbetta S, Gualdoni S, Albertinazzi C, et al. Generation and 1996;271(23):13556–13560. characterization of Rac3 knockout mice. Mol Cell Biol. 2005;25- 56. Hill CS, Wynne J, Treisman R. The Rho family GTPases RhoA, (13):5763–5776. Racl, and CDC42Hsregulate transcriptional activation by SRF. 66. Johnson DI. Cdc42: An essential Rho-type GTPase controlling Cell. 1995;81(7):1159–1170. eukaryotic cell polarity. Microbiol Mol Biol Rev. 1999;63(1):54–105. 57. Perona R, Montaner S, Saniger L, Sanchez-Perez I, Bravo R, Lacal 67. Takenouchi T, Kosaki R, Niizuma T, Hata K, Kosaki K. JC. Activation of the nuclear factor-kappaB by Rho, CDC42, and Macrothrombocytopenia and developmental delay with a de Rac-1 proteins. Genes Dev. 1997;11(4):463–475. novo CDC42 mutation: Yet another locus for thrombocytopenia 58. Coso OA, Chiariello M, Yu J-C, et al. The small GTP-binding pro- and developmental delay. Am J Med Genet A. 2015;167A(11):2822– teins Rac1 and Cdc42regulate the activity of the JNK/SAPK sig- 2825. nalling pathway. Cell. 1995;81(7):1137–1146. 68. Takenouchi T, Okamoto N, Ida S, Uehara T, Kosaki K. Further 59. Tabata H, Nakajima K. Multipolar migration: The third mode of evidence of a mutation in CDC42 as a cause of a recognizable radial neuronal migration in the developing cerebral cortex. J syndromic form of thrombocytopenia. Am J Med Genet A. 2016; Neurosci. 2003;23(31):9996–10001. 170A(4):852–855.