ARHGEF6-dependent cytoskeletal regulation underlies a conserved program of forebrain interneuron development

Article activity feed

1. 

   eLife

   May 29, 2026

   eLife Assessment

   The study presents valuable findings regarding the impact of ARHGEF6 deletion, a RhoGTPase regulator linked to X-linked intellectual disability (XLID46), in the development of interneurons. The evidence supporting the observed cellular and developmental phenotypes collected in both mouse and human iPSC models is convincing, although further work would strengthen the mechanistic interpretation and clarify the specificity of the findings. This work offers new insights into ARHGEF6 function and the potential contribution of its dysfunction to neurodevelopmental disorders.

   Read the original source
2. 

   eLife

   May 29, 2026

   Reviewer #1 (Public review):

   Summary:

   The manuscript has several strengths, including a technically comprehensive approach that combines mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models, providing a rich and multifaceted dataset. Cross-species validation through the parallel use of mouse and human systems strengthens the generality of the observed phenotypes and increases relevance to human neurodevelopment.

   Consistent phenotypic observations across systems show that ARHGEF6 loss affects migration, neurite morphology, growth cone structure, and neuronal survival, supporting a coherent role in cytoskeletal regulation.

   There is clear evidence for developmental defects, including reduced interneuron numbers, increased apoptosis in the ganglionic eminences, and migration deficits, all well supported …

   Reviewer #1 (Public review):

   Summary:

   The manuscript has several strengths, including a technically comprehensive approach that combines mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models, providing a rich and multifaceted dataset. Cross-species validation through the parallel use of mouse and human systems strengthens the generality of the observed phenotypes and increases relevance to human neurodevelopment.

   Consistent phenotypic observations across systems show that ARHGEF6 loss affects migration, neurite morphology, growth cone structure, and neuronal survival, supporting a coherent role in cytoskeletal regulation.

   There is clear evidence for developmental defects, including reduced interneuron numbers, increased apoptosis in the ganglionic eminences, and migration deficits, all well supported by quantitative analyses. Also, there is a high-quality electrophysiological characterization that demonstrates reduced firing in interneurons, providing a well-controlled functional phenotype.

   Strengths:

   The manuscript has several strengths, including a technically comprehensive approach that combines mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models, providing a rich and multifaceted dataset. Cross-species validation through the parallel use of mouse and human systems strengthens the generality of the observed phenotypes and increases relevance to human neurodevelopment.

   Consistent phenotypic observations across systems show that ARHGEF6 loss affects migration, neurite morphology, growth cone structure, and neuronal survival, supporting a coherent role in cytoskeletal regulation.

   There is clear evidence for developmental defects, including reduced interneuron numbers, increased apoptosis in the ganglionic eminences, and migration deficits, all well supported by quantitative analyses. Also, there is a high-quality electrophysiological characterization that demonstrates reduced firing in interneurons, providing a well-controlled functional phenotype.

   Weaknesses:

   Despite the strengths mentioned above, the study has some conceptual and experimental weaknesses that reduce its impact. The mechanistic insight is limited, as the research does not directly establish how ARHGEF6 regulates downstream signaling pathways.

   Also, there is insufficient evidence for interneuron specificity; although the central claim is that ARHGEF6 plays a selective role in interneurons, the data do not adequately exclude the possibility that the observed effects reflect broader neuronal defects. The study lacks critical controls across cell types, as several phenotypes observed in organoids and progenitors, including apoptosis, reduced neuronal output, and altered morphology, could also affect multiple neuronal populations without being directly tested. Furthermore, the data are predominantly descriptive, with many results remaining correlative and failing to establish causal relationships.

   Some more comments:

   (1) Given that ARHGEF6 is a guanine nucleotide exchange factor for Rac1 and Cdc42, the absence of direct measurements of GTPase activity or downstream signaling represents a significant gap. The interpretation that the observed phenotypes are mediated through specific cytoskeletal pathways, therefore, remains inferential.

   (2) The manuscript repeatedly interprets the findings as interneuron-specific. However, several key observations are not demonstrated to be restricted to IN. Without direct comparison to excitatory neurons or other cell types, it is difficult to conclude that ARHGEF6 plays a selective role in interneurons rather than a more general role in neuronal development. The well-done analysis of the transcriptomic dataset is not sufficient to claim IN specificity. This issue is particularly important for the interpretation of the human organoid experiments, where reductions in SOX2⁺ progenitors and NEUN⁺ neurons, as well as increased apoptosis, could reflect global developmental defects. Similarly, in the mouse experiments, the reduction in GAD67⁺ cells is compelling, but it is not shown whether other neuronal populations are also affected.

   (3) The study provides a strong phenotypic description but limited causal resolution. For example, migration defects, altered growth cone morphology, and reduced branching are all consistent with impaired cytoskeletal regulation, but the links between these phenotypes are not directly established. Likewise, while the electrophysiological data convincingly show reduced firing in interneurons, the connection between altered cytoskeletal dynamics and intrinsic excitability is not explored.

   (4) Several aspects of data presentation could be improved. In multiple figures (e.g., Figure 1A, D; Figure 4 and Video S1, 2), the images are difficult to interpret due to high cellular density, limited magnification, or lack of clear annotation. In some cases, it is not fully clear how quantifications were performed or which regions were analyzed. Improving the visual clarity with arrows, boxes, and high-magnification inserts of the data would strengthen confidence in the conclusions.

   Read the original source
3. 

   eLife

   May 29, 2026

   Reviewer #2 (Public review):

   The authors investigate the impact of the deletion of the small GTPase regulator ARHGEF6 on the development and physiology of interneurons. Using public databases, they first show that ARHGEF6 is enriched in interneurons or in areas that give rise to them, both in development and adulthood, in humans and mice. Using a complete KO mouse previously reported, and using a GAD67-GFP reporter mice line, they show that in the adult mouse cortex and hippocampus, there is a notorious reduction GFP+ cells. These mice show increased apoptotic cells at different timepoints and areas of the brain during development. In the developing cortex of ARHGEF6-KO mice, there are fewer IN in all layers of the developing cortex, and cells present processes not correctly oriented. IN from the hippocampus in culture show reduced …

   Reviewer #2 (Public review):

   The authors investigate the impact of the deletion of the small GTPase regulator ARHGEF6 on the development and physiology of interneurons. Using public databases, they first show that ARHGEF6 is enriched in interneurons or in areas that give rise to them, both in development and adulthood, in humans and mice. Using a complete KO mouse previously reported, and using a GAD67-GFP reporter mice line, they show that in the adult mouse cortex and hippocampus, there is a notorious reduction GFP+ cells. These mice show increased apoptotic cells at different timepoints and areas of the brain during development. In the developing cortex of ARHGEF6-KO mice, there are fewer IN in all layers of the developing cortex, and cells present processes not correctly oriented. IN from the hippocampus in culture show reduced excitability and impaired neurite branching. The authors then established isogenic hiPSCs lines to study ARHGEF6 deletion in human cells and differentiated ventral forebrain neurons, to find interneuron-related and non-related phenotypes. Most importantly, human interneurons grown in organoids show reduced branching and altered growth cone morphology. The authors claim that the novel interneuron phenotypes found in these models can explain, in part, the human intellectual disabilities associated with mutations in this protein. The study is well conducted and opens new avenues of research not only for the role of small GTPases regulation in early nervous system development, but also for how interneuron deficiencies impact a wider range of intellectual disability syndromes found in humans.

   However, most conclusions of the present version would be strengthened after considering the following comments:

   Major comments

   (1) The reported biological processes evaluated at different developmental stages may be directly or indirectly related to ARHGEF6 function itself. As a model of a hereditary disease, full organism gene deletion is valid, since the human patients suffer from that condition as well. However, to investigate the roles of a protein, complete deletions may not be very accurate since they can give rise to phenotypes that are only indirectly related to the protein function itself. Most conclusions of the present manuscript should either be discussed in this regard or add evidence for a direct role of the protein. One such evidence is typically performed with acute knockdowns in culture, or in developing brains by in utero electroporation. For example, Figure 1C shows that the principal excitatory neurons in the hippocampus do not express ARHGEF6. However, most electrophysiological and behavioral evidence of defects in ARHGEF6-KO mice arises from evaluating these cells (Remakers et al., 2012). I am not suggesting that either previous or actual evidence is wrong. But I believe readers would benefit from a clear distinction (or add caution notes) between a functional consequence of the deletion (that can be months away and in other cells than the actual molecular defect) and a true cell biological function of the protein under study. In favor of the authors, this is a concern with most conclusions derived from KO organisms.

   (2) Figure 1E-G H I. All conclusions are made with a GAD67-GFP reporter, which is a very powerful and reliable tool for large-scale screening. All the conclusions of the paper would be strengthened if some immunohistochemical staining in the same areas of specific markers for interneurons would be added as supporting complementary evidence.

   (3) Cell death in development: It is surprising that the high amount of TUNEL staining during development does not translate into gross histological changes in the adult brain (studied elsewhere). Can authors discuss possible explanations?

   (4) Section 4 (Figures 2F-J) - The authors present this staining as an analysis of migration. Normally, migration studies are performed with a "pulse-chase" paradigm, where a single cohort is labeled and then followed over time (normally by in utero electroporation of a fluorescent protein). Tissue is then fixed at different time points, and migration can be followed. On the contrary, the evidence is from a single point, in an experimental setting in which all Gad67 IN are stained, and hence, one cannot imply a defect in migration. The differences between WT and ARHGEF6-KO are obvious and interesting; it is just that they cannot be solely attributed to a problem in migration.

   Also, a true phenotype of migration in the current setting should have found that the cells that failed to migrate are accumulated in deeper layers. My impression is that the changes in IN per layer are easier explained by total cell number, rather than migration. Perhaps evaluating earlier timepoints could clarify this.

   (5) It is known that ARHGEF6 deletion produces severe F-actin phenotypes in neurons. Have the authors confirmed in their hippocampal cultures GAD67 cells ALSO have these phenotypes? Stress fibers in somas, growth cones, and actin patches along neurites.

   (6) Section 4. The authors present data for deficient migration of the GFP-labeled interneurons. Is it possible to assess, in the same sections, whether other cell types are also affected? Although the hypothesis that ARHGEF6 deletion will have an impact in IN is well rooted in expression data, by assessing other cell types, one can even include a positive control or evidence for a cell-autonomous phenotype.

   (7) ARHGEDF6 deletion has an important impact on organoid development (size, shape, etc). Have the authors analysed whether these organoids produced fewer interneurons?

   (8) In assembloids, the differences in migration parameters are very small between WT and ARHGEF6-KO, which reinforces that perhaps what is observed in the different layers of cortex during mouse development is likely not entirely due to migration, as concluded.

   (9) To properly weigh the present evidence -interneuron deficits- using the ARHGEF6-KO model, authors should include a deeper discussion in light of much work that has been done using these mice. How does the finding of a diminished IN population in the brain of these mice explain the large amount of electrophysiological and behavioral evidence produced before with these animals? Perhaps the most important work to discuss these aspects is the initial ARHGEF6-KO report by Ramakers and colleagues (2012), but there are others.

   Minor comments

   (1) Figure 1A. It looks clear that the GE shows the highest expression of ARHGEF6; however, the reader needs the reference levels where the log2 expression is calculated. What are the reference levels?

   (2) Have the authors compared the number of GAD67-eGFP cells in the hippocampal cultures between WT and ARHGEF6-KO mice?

   (3) Section 3, as a caution note, authors should mention that it is not possible to know from the evidence provided which cells are dying.

   (4) In the dorsal-ventral assembloids, it is expected that the ventral organoid would contain lots of GFP expression compared to the dorsal, but in the image shown (Figure 5A) both parts of the assembloid seem to have the same amount and distribution of GFP. How is that possible?

   Read the original source
4. 

   eLife

   May 29, 2026

   Reviewer #3 (Public review):

   Summary:

   ARHGEF6 is a RAC1/CDC42 guanine nucleotide exchange factor that has been proposed to be associated with X-linked intellectual disability, but its relevance to the pathology is not well established. ARHGEF6 has been assigned a role in spine density and plasticity of hippocampal pyramidal neurons, but nothing is known about its role in interneuron development. Here, the authors show that ARHGEF6 is expressed early in development in the inhibitory lineage during the peak of interneuron generation and migration. The aim of the study is therefore to investigate whether, in addition to its role in pyramidal neurons, ARHGEF6 could play a role in inhibitory neuron development. Using both ARHGEF6-KO mice and organoids from ARHGEF6-KO hiPSCs, the authors show that ARHGEF6 plays a critical role in interneuron …

   Reviewer #3 (Public review):

   Summary:

   ARHGEF6 is a RAC1/CDC42 guanine nucleotide exchange factor that has been proposed to be associated with X-linked intellectual disability, but its relevance to the pathology is not well established. ARHGEF6 has been assigned a role in spine density and plasticity of hippocampal pyramidal neurons, but nothing is known about its role in interneuron development. Here, the authors show that ARHGEF6 is expressed early in development in the inhibitory lineage during the peak of interneuron generation and migration. The aim of the study is therefore to investigate whether, in addition to its role in pyramidal neurons, ARHGEF6 could play a role in inhibitory neuron development. Using both ARHGEF6-KO mice and organoids from ARHGEF6-KO hiPSCs, the authors show that ARHGEF6 plays a critical role in interneuron development and function

   Strengths:

   The major strength of the paper is the very detailed analysis of the role of ARHGEF6 using two different systems: ARHGEF6-KO mice and deletion of ARHGEF6 in human iPSC-derived organoids. Strikingly, deletion of ARHGEF6 in both systems induces similar defects such as an increase in apoptosis, reduced neuronal output, impaired neuronal morphology, and disrupted migratory dynamics. This compelling evidence demonstrates that ARHGEF6, in addition to its already well-described role in spine formation and plasticity, is playing a crucial role during embryonic development through its function in interneurons.

   Weaknesses:

   (1) In Figure 1, the authors show that ARHGEF6 is expressed in different regions of the brain, including the interneuron lineage, and that depletion of ARHGEF6 reduces the number of GABAergic neurons in the adult cortex and hippocampus. To try to better characterize this defect, the authors in Figure 2 investigate whether deletion of ARHGEF6 affects interneuron migration and survival during embryonic development. To do so, ARHGEF6 ko mice were crossed with the GAD67-eGFP reporter line to follow the inhibitory lineage. The authors analyse apoptosis using TUNEL staining, and show that it is significantly increased in the ganglion eminence of ARHGEF6-KO E14.5 embryos. The authors claim that this is not the case in the cortex. However, the image shown in Figure 2A really suggests that staining is increased. Which part of the neocortex is analysed for quantification? This should be clarified.

   (2) In Figure 2F-J, the authors investigate the migration of interneurons by analysing the GAD67-eGFP staining, and clearly show that the migratory abilities of the depleted neurons are reduced. However, the authors do not discuss the fact that, because depletion of ARHGEF6 increases apoptosis, there are fewer neurons available for migration. This is important for the interpretation of the data. This point should be clarified.

   (3) In Supplementary Figure S2, the authors describe the establishment of the ARHGEF6-KO human iPSC line and test the ability of these cells to undergo correct development, especially for the generation of neural progenitor cells. I was wondering why the authors do not present the data of both control and ARHGEF6-KO cells.

   (4) At the molecular level, how ARHGEF6 depletion could affect neuronal survival is missing. In addition, as ARHGEF6 is a GEF for RAC1 and Cdc42 amongst other GEFs, I would have expected that the authors test how RAC1 activity (and Cdc42) is affected in ARHGEF6-depleted brains and in ARHGEF6-KO organoids. The measure of phalloidin staining and the anisotropy index are not really meaningful.

   (5) The authors show that ARHGEF6-KO forebrain organoids were markedly smaller compared to their isogenic controls, and their study suggests that ARHGEF6 expression impacts progenitor maintenance and neurogenesis. Despite representing only a minority of the total neuronal population, I was wondering whether ARHGEF6-KO mice present brain morphology defects such as microcephaly.

   Read the original source
5. 

   eLife

   May 29, 2026

   Author response:

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   The manuscript has several strengths, including a technically comprehensive approach that combines mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models, providing a rich and multifaceted dataset. Cross-species validation through the parallel use of mouse and human systems strengthens the generality of the observed phenotypes and increases relevance to human neurodevelopment.

   Consistent phenotypic observations across systems show that ARHGEF6 loss affects migration, neurite morphology, growth cone structure, and neuronal survival, supporting a coherent role in cytoskeletal regulation.

   There is clear evidence for developmental defects, including reduced interneuron numbers, increased apoptosis in the ganglionic eminences, and …

   Author response:

   Public Reviews:

   Reviewer #1 (Public review):

   Summary:

   The manuscript has several strengths, including a technically comprehensive approach that combines mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models, providing a rich and multifaceted dataset. Cross-species validation through the parallel use of mouse and human systems strengthens the generality of the observed phenotypes and increases relevance to human neurodevelopment.

   Consistent phenotypic observations across systems show that ARHGEF6 loss affects migration, neurite morphology, growth cone structure, and neuronal survival, supporting a coherent role in cytoskeletal regulation.

   There is clear evidence for developmental defects, including reduced interneuron numbers, increased apoptosis in the ganglionic eminences, and migration deficits, all well supported by quantitative analyses. Also, there is a high-quality electrophysiological characterization that demonstrates reduced firing in interneurons, providing a well-controlled functional phenotype.

   Strengths:

   The manuscript has several strengths, including a technically comprehensive approach that combines mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models, providing a rich and multifaceted dataset. Cross-species validation through the parallel use of mouse and human systems strengthens the generality of the observed phenotypes and increases relevance to human neurodevelopment.

   Consistent phenotypic observations across systems show that ARHGEF6 loss affects migration, neurite morphology, growth cone structure, and neuronal survival, supporting a coherent role in cytoskeletal regulation.

   There is clear evidence for developmental defects, including reduced interneuron numbers, increased apoptosis in the ganglionic eminences, and migration deficits, all well supported by quantitative analyses. Also, there is a high-quality electrophysiological characterization that demonstrates reduced firing in interneurons, providing a well-controlled functional phenotype.

   We thank the reviewer for their positive and thoughtful assessment of our manuscript. We appreciate their recognition of the technical breadth of the study, including the integration of mouse genetics, electrophysiology, live imaging in assembloids, and human organoid models. We are also grateful that the reviewer highlights the value of our cross-species approach, as a major goal of the study was to determine whether ARHGEF6 loss produces convergent developmental and cellular phenotypes in both mouse and human systems.

   Weaknesses:

   Despite the strengths mentioned above, the study has some conceptual and experimental weaknesses that reduce its impact. The mechanistic insight is limited, as the research does not directly establish how ARHGEF6 regulates downstream signaling pathways.

   We appreciate the reviewer’s constructive comment. We agree that, although our data establish a phenotypic link between ARHGEF6 loss and interneuron development, they do not directly dissect the molecular mechanisms underlying the observed defects. Our interpretation that the mutant phenotype involves dysregulation of cytoskeletal dynamics is based on the directly observed defects in actin polymerization and organization in neural progenitor cells and neuronal growth cones respectively, and is consistent with the abnormalities observed in neurite morphology and neuronal migration. This interpretation is further supported by the established role of Arhgef6 as a regulator of the small Rho GTPases Rac1 and Cdc42. Previous evidence shows that Arhgef6 loss reduces the activity of both GTPases and deregulates the expression of the cytoskeletal regulators Pak1–3, Limk1, and Cofilin in the mouse brain (Ramakers et al., 2012). Moreover, spine abnormalities in Arhgef6-knockdown ex vivo slice cultures can be rescued by expressing the active form of Pak3, a downstream effector of Rac1 and Cdc42 (Node-Langlois et al., 2006). Together, these findings support a model in which the loss of the protein affects development through cytoskeletal dysregulation, likely involving altered Rho GTPase signalling. We nevertheless agree that further experiments would be required to establish a direct causal relationship between ARHGEF6 loss, Rho GTPase activity, cytoskeletal dysregulation, and the interneuron phenotypes described here. We will therefore revise the manuscript to clarify that this mechanistic link remains an interpretation supported by our data and the literature, rather than a direct demonstration within the present study.

   Also, there is insufficient evidence for interneuron specificity; although the central claim is that ARHGEF6 plays a selective role in interneurons, the data do not adequately exclude the possibility that the observed effects reflect broader neuronal defects. The study lacks critical controls across cell types, as several phenotypes observed in organoids and progenitors, including apoptosis, reduced neuronal output, and altered morphology, could also affect multiple neuronal populations without being directly tested.

   We agree that the current data do not exclude the possibility of alterations in other neuronal lineages, specifically the excitatory lineage. With regard to this, we would like to emphasize that the investigation of excitatory cell phenotypes was beyond the scope of the present study, as this aspect has previously been examined by Ramakers et al., 2012 and Node-Langlois et al., 2006, particularly in the context of hippocampal pyramidal cells, which are among the few cell types showing consistent expression of the gene in the adult mouse brain (Allen Brain Atlas; Yao et al., 2021). In this context, it is interesting to note that, in Ramakers et al., 2012 (Figure S1), MAP2 immunostaining of hippocampal formations revealed comparable distribution and intensity of neuronal cell bodies and dendrites throughout the hippocampus of both wild-type and Arhgef6-KO animals. With regard to morphological maturation of excitatory cells, whereas we observe a simplification of interneuron morphology in both mouse and human models, Ramakers et al., 2012 reported increased dendritic arborization complexity in hippocampal pyramidal cells. With regard to migration, a direct comparison with excitatory neurons would be intrinsically difficult, as excitatory and inhibitory neurons undergo highly distinct migratory processes and are therefore not directly comparable. We greatly appreciate the reviewer’s comment, as it gives us the opportunity to better discuss the relationship between our findings and previous studies in the Discussion. We will revise the manuscript and avoid implying that the phenotype observed is exclusive to interneurons.

   Furthermore, the data are predominantly descriptive, with many results remaining correlative and failing to establish causal relationships.

   We agree that our study primarily establishes a phenotypic framework and does not fully resolve the causal hierarchy among altered survival, migration, cytoskeletal morphology, and intrinsic excitability. We will revise the manuscript to make this limitation explicit, avoiding statements that imply direct causality beyond the data presented.

   Some more comments:

   (1) Given that ARHGEF6 is a guanine nucleotide exchange factor for Rac1 and Cdc42, the absence of direct measurements of GTPase activity or downstream signaling represents a significant gap. The interpretation that the observed phenotypes are mediated through specific cytoskeletal pathways, therefore, remains inferential.

   We appreciate the comment. The interpretation that our phenotype involves dysregulated cytoskeletal dynamics is based on the observed defects in actin polymerization and F-actin organization in neuronal growth cones and is consistent with the abnormalities in neurite morphology and neuronal migration. We will explicitly state in the Discussion that, since we did not directly measure Rac1 and Cdc42 activity levels in our models, our hypothesis regarding the involvement of this molecular pathway in the establishment of the observed phenotype therefore remains inferential, despite being supported by the current literature.

   (2) The manuscript repeatedly interprets the findings as interneuron-specific. However, several key observations are not demonstrated to be restricted to IN. Without direct comparison to excitatory neurons or other cell types, it is difficult to conclude that ARHGEF6 plays a selective role in interneurons rather than a more general role in neuronal development. The well-done analysis of the transcriptomic dataset is not sufficient to claim IN specificity. This issue is particularly important for the interpretation of the human organoid experiments, where reductions in SOX2⁺ progenitors and NEUN⁺ neurons, as well as increased apoptosis, could reflect global developmental defects. Similarly, in the mouse experiments, the reduction in GAD67⁺ cells is compelling, but it is not shown whether other neuronal populations are also affected.

   As previously mentioned, we understand the reviewer’s concern regarding the specificity of the observed phenotypes in interneurons and agree that the claims should be tempered. However, it is important to note that the interpretation of the human organoid experiments should be reconsidered. The use of specifically ventralized MGE-like organoids allowed us to assess the cell-autonomous nature of defects such as the reduction in inhibitory progenitors’ neuronal output, the increased apoptosis, and the morphological abnormalities of inhibitory neurons. We will acknowledge in the Discussion the limitations of the study with regard to assessing the cell-autonomous nature of the observed migration defects.

   (3) The study provides a strong phenotypic description but limited causal resolution. For example, migration defects, altered growth cone morphology, and reduced branching are all consistent with impaired cytoskeletal regulation, but the links between these phenotypes are not directly established. Likewise, while the electrophysiological data convincingly show reduced firing in interneurons, the connection between altered cytoskeletal dynamics and intrinsic excitability is not explored.

   The observed migration defects, altered growth-cone morphology, and reduced branching are consistent with impaired cytoskeletal regulation. However, we acknowledge that the mechanistic links among these phenotypes remain to be directly demonstrated. Similarly, although our electrophysiological data show reduced firing in ARHGEF6-KO interneurons, the present study does not provide direct evidence linking impaired excitability to altered cytoskeletal dynamics. In the latter case, we think that the underlying mechanisms should be further investigated at the subcellular level, particularly with respect to cytoskeleton-mediated intracellular trafficking and localization and distribution of ion channels. One limitation of the present study, which may have masked electrophysiological alterations associated with differences in membrane composition (current Figure S1D–H), is that different interneuron subtypes with distinct intrinsic properties were pooled together in the analysis. We will expand the Discussion to address these limitations.

   (4) Several aspects of data presentation could be improved. In multiple figures (e.g., Figure 1A, D; Figure 4 and Video S1, 2), the images are difficult to interpret due to high cellular density, limited magnification, or lack of clear annotation. In some cases, it is not fully clear how quantifications were performed or which regions were analyzed. Improving the visual clarity with arrows, boxes, and high-magnification inserts of the data would strengthen confidence in the conclusions.

   We would like to thank the reviewer for pointing this out. We agree that some images and videos would benefit from clearer annotation. In the revised manuscript, we will add high-magnification insets, arrows or boxes highlighting the relevant regions/cells, and clearer descriptions of the quantified regions. We will also improve legends and video labels to indicate genotype, region, and tracked cells.

   Reviewer #2 (Public review):

   The authors investigate the impact of the deletion of the small GTPase regulator ARHGEF6 on the development and physiology of interneurons. Using public databases, they first show that ARHGEF6 is enriched in interneurons or in areas that give rise to them, both in development and adulthood, in humans and mice. Using a complete KO mouse previously reported, and using a GAD67-GFP reporter mice line, they show that in the adult mouse cortex and hippocampus, there is a notorious reduction GFP+ cells. These mice show increased apoptotic cells at different timepoints and areas of the brain during development. In the developing cortex of ARHGEF6-KO mice, there are fewer IN in all layers of the developing cortex, and cells present processes not correctly oriented. IN from the hippocampus in culture show reduced excitability and impaired neurite branching. The authors then established isogenic hiPSCs lines to study ARHGEF6 deletion in human cells and differentiated ventral forebrain neurons, to find interneuron-related and non-related phenotypes. Most importantly, human interneurons grown in organoids show reduced branching and altered growth cone morphology. The authors claim that the novel interneuron phenotypes found in these models can explain, in part, the human intellectual disabilities associated with mutations in this protein. The study is well conducted and opens new avenues of research not only for the role of small GTPases regulation in early nervous system development, but also for how interneuron deficiencies impact a wider range of intellectual disability syndromes found in humans.

   We appreciate the reviewer’s positive evaluation of our manuscript and their recognition of this work’s potential to expand the focus of intellectual disability research on the development and function of the inhibitory system. We are particularly encouraged that the reviewer highlights the strength of our combined mouse and human cellular models, as well as the relevance of the interneuron-related phenotypes we identify across systems.

   However, most conclusions of the present version would be strengthened after considering the following comments:

   Major comments:

   (1) The reported biological processes evaluated at different developmental stages may be directly or indirectly related to ARHGEF6 function itself. As a model of a hereditary disease, full organism gene deletion is valid, since the human patients suffer from that condition as well. However, to investigate the roles of a protein, complete deletions may not be very accurate since they can give rise to phenotypes that are only indirectly related to the protein function itself. Most conclusions of the present manuscript should either be discussed in this regard or add evidence for a direct role of the protein. One such evidence is typically performed with acute knockdowns in culture, or in developing brains by in utero electroporation. For example, Figure 1C shows that the principal excitatory neurons in the hippocampus do not express ARHGEF6. However, most electrophysiological and behavioral evidence of defects in ARHGEF6-KO mice arises from evaluating these cells (Ramakers et al., 2012). I am not suggesting that either previous or actual evidence is wrong. But I believe readers would benefit from a clear distinction (or add caution notes) between a functional consequence of the deletion (that can be months away and in other cells than the actual molecular defect) and a true cell biological function of the protein under study. In favor of the authors, this is a concern with most conclusions derived from KO organisms.

   We agree with the reviewer that phenotypes observed in constitutive knockout models may, in some contexts, reflect indirect or compensatory consequences of long-term gene loss. Conditional and/or inducible knockout or knockdown approaches can certainly help dissect the nature of the observed defects and better define the effects of gene ablation at different developmental stages or in specific cell types. However, in the context of our study, it is important to note that the experiments performed in ventralized MGE-like organoids allowed us to assess the cell-autonomous nature of very early developmental defects in the inhibitory lineage, in isolation from other cell types. These defects include reduced neuronal output from inhibitory progenitors, increased apoptosis, and morphological abnormalities in inhibitory neurons. Therefore, the phenotypes reported here are less likely to reflect effects originating in, or indirectly caused by, cell types that do not express Arhgef6.

   With regard to Figure 1C, we state in the Results that “among excitatory populations, only CA3 pyramidal neurons and mossy cells exhibited expression levels comparable to those observed in inhibitory clusters (Figure 1D, Table S2),” thereby not neglecting the potential effect of the lack of a functional protein in these populations.

   (2) Figure 1E-G H I. All conclusions are made with a GAD67-GFP reporter, which is a very powerful and reliable tool for large-scale screening. All the conclusions of the paper would be strengthened if some immunohistochemical staining in the same areas of specific markers for interneurons would be added as supporting complementary evidence.

   We appreciate the insightful comment of the reviewer. Additional validation using established interneuronal markers will further strengthen the GAD67-eGFP analysis. We will perform complementary stainings (e.g., PVALB and CCK) and quantifications and include these data as a Supplementary Figure.

   (3) Cell death in development: It is surprising that the high amount of TUNEL staining during development does not translate into gross histological changes in the adult brain (studied elsewhere). Can authors discuss possible explanations?

   We appreciate the thoughtful consideration of our findings. We think that possible explanations include partial compensatory mechanisms during development, which may mitigate the long-term anatomical consequences of increased cell death. In addition, the phenotype may be restricted to specific neuronal populations or developmental windows, thereby producing functional alterations without necessarily resulting in overt macroanatomical defects. Thus, although increased developmental cell death may contribute to altered circuit assembly and neuronal output, it may not be sufficient to produce gross histological changes detectable at the adult brain level.

   (4) Section 4 (Figures 2F-J) - The authors present this staining as an analysis of migration. Normally, migration studies are performed with a "pulse-chase" paradigm, where a single cohort is labeled and then followed over time (normally by in utero electroporation of a fluorescent protein). Tissue is then fixed at different time points, and migration can be followed. On the contrary, the evidence is from a single point, in an experimental setting in which all Gad67 IN are stained, and hence, one cannot imply a defect in migration. The differences between WT and ARHGEF6-KO are obvious and interesting; it is just that they cannot be solely attributed to a problem in migration.

   Also, a true phenotype of migration in the current setting should have found that the cells that failed to migrate are accumulated in deeper layers. My impression is that the changes in IN per layer are easier explained by total cell number, rather than migration. Perhaps evaluating earlier timepoints could clarify this.

   We appreciate the reviewer’s suggestion to implement an additional time point in the in vivo migration analysis. Since an earlier in vivo time point would most likely not reveal migration-related defects, as most cells would still be confined to the ganglionic eminence (Liaci et al., 2022), we will include analyses performed at a later developmental time point as supplementary evidence. We will also revise the wording to clarify that the fixed-tissue data show altered distribution and orientation of GAD67-eGFP-positive interneurons, which are consistent with impaired migratory behavior when considered together with the in vitro live-imaging data. At the same time, we will acknowledge that reduced interneuron survival and/or neuronal output may also contribute to the observed phenotype.

   (5) It is known that ARHGEF6 deletion produces severe F-actin phenotypes in neurons. Have the authors confirmed in their hippocampal cultures GAD67 cells ALSO have these phenotypes? Stress fibers in somas, growth cones, and actin patches along neurites.

   We did not directly assess F-actin organization in GAD67-eGFP murine primary cultures. Direct analyses of F-actin organization, growth-cone morphology, and cytoskeletal organization were performed only in the human system. To further assess this phenotype, we will perform phalloidin staining on GAD67-eGFP brain sections to evaluate F-actin organization in interneurons in vivo.

   (6) Section 4. The authors present data for deficient migration of the GFP-labeled interneurons. Is it possible to assess, in the same sections, whether other cell types are also affected? Although the hypothesis that ARHGEF6 deletion will have an impact in IN is well rooted in expression data, by assessing other cell types, one can even include a positive control or evidence for a cell-autonomous phenotype.

   We thank the reviewer for their thoughtful suggestions. We agree that extending the analysis to additional cell types would provide further insight into the specificity of the phenotype; however, a comprehensive evaluation of all neuronal populations falls beyond the scope of this research. The use of ventralized MGE-like organoids enabled us to examine whether key defects were cell-autonomous, including the reduced neuronal output of inhibitory progenitors, increased apoptosis, and abnormal inhibitory-neuron morphology.

   (7) ARHGEDF6 deletion has an important impact on organoid development (size, shape, etc). Have the authors analysed whether these organoids produced fewer interneurons?

   We would like to clarify that the organoids analyzed in the study are ventral MGE-like organoids and therefore the reduction in neuronal output (current Figure 4K) primarily reflects the ventral/interneuron lineage in this model.

   (8) In assembloids, the differences in migration parameters are very small between WT and ARHGEF6-KO, which reinforces that perhaps what is observed in the different layers of cortex during mouse development is likely not entirely due to migration, as concluded.

   We agree that the migration parameters in assembloids should not be interpreted in isolation. We will revise the text to emphasize that the reduction in the number of interneurons observed in the adult brains is part of a broader pattern that also includes altered neuronal output and reduced viability.

   (9) To properly weigh the present evidence -interneuron deficits- using the ARHGEF6-KO model, authors should include a deeper discussion in light of much work that has been done using these mice. How does the finding of a diminished IN population in the brain of these mice explain the large amount of electrophysiological and behavioral evidence produced before with these animals? Perhaps the most important work to discuss these aspects is the initial ARHGEF6-KO report by Ramakers and colleagues (2012), but there are others.

   We appreciate the reviewer’s emphasis on the importance of framing our findings within the broader context of the existing literature. We will expand the Discussion to better integrate previous work on ARHGEF6-KO mice. Specifically, we will discuss how reduced interneuron number and altered interneuronal function may contribute to previously reported electrophysiological and behavioral phenotypes, acting in concert with previously described alterations in excitatory neurons and synaptic plasticity (Ramakers et al., 2012).

   Minor comments:

   (1) Figure 1A. It looks clear that the GE shows the highest expression of ARHGEF6; however, the reader needs the reference levels where the log2 expression is calculated. What are the reference levels?

   We would like to thank the reviewer for pointing this out. We will clarify in the caption that the log2(RPKM+1) expression values are shown as absolute values and are not relative to a reference condition.

   (2) Have the authors compared the number of GAD67-eGFP cells in the hippocampal cultures between WT and ARHGEF6-KO mice?

   We did not rely on total GAD67-eGFP counts in dissociated hippocampal cultures because differences could reflect initial plating composition, survival, and maturation. In our experience, the MGE-like organoid system provides a more controlled in vitro context to assess neuronal output in the ventral lineage.

   (3) Section 3, as a caution note, authors should mention that it is not possible to know from the evidence provided which cells are dying.

   We agree with the reviewer and will add a cautionary statement noting that TUNEL staining alone does not identify the precise dying cell type. We will clarify that increased cell death in the ganglionic eminence and MGE-like organoids is consistent with a prominent involvement of the ventral/inhibitory lineage, while acknowledging the limits of the assay.

   (4) In the dorsal-ventral assembloids, it is expected that the ventral organoid would contain lots of GFP expression compared to the dorsal, but in the image shown (Figure 5A) both parts of the assembloid seem to have the same amount and distribution of GFP. How is that possible?

   We appreciate the thoughtful comment of the reviewer. After two weeks of fusion, a considerable number of interneurons are expected to have migrated from the ventral to the dorsal compartment of the assembloid (Birey et al., 2017; Sloan et al., 2018). In terms of distribution, we think that current Figure 5A shows a gradient of eGFP-positive cells within the dorsal compartment, with the number of labeled cells decreasing as the distance from the fusion interface between the two organoids increases. By contrast, a comparable gradient is not evident in the ventral compartment, where several labeled neurons remain present even in regions distal to the fusion site.

   Reviewer #3 (Public review):

   Summary:

   ARHGEF6 is a RAC1/CDC42 guanine nucleotide exchange factor that has been proposed to be associated with X-linked intellectual disability, but its relevance to the pathology is not well established. ARHGEF6 has been assigned a role in spine density and plasticity of hippocampal pyramidal neurons, but nothing is known about its role in interneuron development. Here, the authors show that ARHGEF6 is expressed early in development in the inhibitory lineage during the peak of interneuron generation and migration. The aim of the study is therefore to investigate whether, in addition to its role in pyramidal neurons, ARHGEF6 could play a role in inhibitory neuron development. Using both ARHGEF6-KO mice and organoids from ARHGEF6-KO hiPSCs, the authors show that ARHGEF6 plays a critical role in interneuron development and function

   Strengths:

   The major strength of the paper is the very detailed analysis of the role of ARHGEF6 using two different systems: ARHGEF6-KO mice and deletion of ARHGEF6 in human iPSC-derived organoids. Strikingly, deletion of ARHGEF6 in both systems induces similar defects such as an increase in apoptosis, reduced neuronal output, impaired neuronal morphology, and disrupted migratory dynamics. This compelling evidence demonstrates that ARHGEF6, in addition to its already well-described role in spine formation and plasticity, is playing a crucial role during embryonic development through its function in interneurons.

   We thank the reviewer for this positive assessment of our work and for highlighting the strength of our combined in vivo and human iPSC-derived organoid approaches. We are pleased that the reviewer recognizes the consistency of the phenotypes observed across both systems and acknowledges that our findings support a crucial role, during early stages of embryonic development, for a protein previously thought to be relevant primarily in the synaptic context.

   Weaknesses:

   (1) In Figure 1, the authors show that ARHGEF6 is expressed in different regions of the brain, including the interneuron lineage, and that depletion of ARHGEF6 reduces the number of GABAergic neurons in the adult cortex and hippocampus. To try to better characterize this defect, the authors in Figure 2 investigate whether deletion of ARHGEF6 affects interneuron migration and survival during embryonic development. To do so, ARHGEF6 ko mice were crossed with the GAD67-eGFP reporter line to follow the inhibitory lineage. The authors analyse apoptosis using TUNEL staining, and show that it is significantly increased in the ganglion eminence of ARHGEF6-KO E14.5 embryos. The authors claim that this is not the case in the cortex. However, the image shown in Figure 2A really suggests that staining is increased. Which part of the neocortex is analysed for quantification? This should be clarified.

   We would like to thank the reviewer for pointing this out. The region analyzed was the same as that used to assess GAD67-eGFP-positive cells in Figure 2F. We will clarify the exact neocortical region used for TUNEL quantification and revise the figure and legend to make the analyzed area explicit. We will also analyze additional animals to improve the accuracy of the analysis.

   (2) In Figure 2F-J, the authors investigate the migration of interneurons by analysing the GAD67-eGFP staining, and clearly show that the migratory abilities of the depleted neurons are reduced. However, the authors do not discuss the fact that, because depletion of ARHGEF6 increases apoptosis, there are fewer neurons available for migration. This is important for the interpretation of the data. This point should be clarified.

   We appreciate this comment and believe that it is particularly relevant to the interpretation of the data shown in Figure 2F–G. We will clarify the limited interpretation of this specific analysis in the Results section. The altered directionality observed in vivo, together with evidence of impaired migratory behavior obtained through in vitro live imaging, supports the possibility that altered migratory dynamics contribute to the phenotype, although increased apoptosis and reduced neuronal output may also contribute.

   (3) In Supplementary Figure S2, the authors describe the establishment of the ARHGEF6-KO human iPSC line and test the ability of these cells to undergo correct development, especially for the generation of neural progenitor cells. I was wondering why the authors do not present the data of both control and ARHGEF6-KO cells.

   We thank the reviewer for pointing this out. All staining reported in the organoids and assembloids in this paper shows that the WT ATCC-DYS0100 cell line, as well as the mutant, efficiently differentiates into neuronal tissue. The Supplementary Figure was intended to validate the impact of the mutation on the ability of the iPSC line to retain its differentiation capacity as a preliminary step before proceeding with organoid differentiation. We will integrate stainings for NPC markers on the WT line in the Supplementary Figure.

   (4) At the molecular level, how ARHGEF6 depletion could affect neuronal survival is missing. In addition, as ARHGEF6 is a GEF for RAC1 and Cdc42 amongst other GEFs, I would have expected that the authors test how RAC1 activity (and Cdc42) is affected in ARHGEF6-depleted brains and in ARHGEF6-KO organoids. The measure of phalloidin staining and the anisotropy index are not really meaningful.

   We appreciate the thoughtful comment of the reviewer. Previous evidence already shows that Arhgef6 loss reduces the activity of both GTPases and deregulates the expression of the cytoskeletal regulators Pak1–3, Limk1, and Cofilin in the mouse brain (Ramakers et al., 2012). Regarding organoids, we agree that direct RAC1/CDC42 activity measurements would have strengthened the molecular mechanism. We will revise the manuscript to avoid implying that our phalloidin-based measurements alone establish the underlying dysregulated molecular pathway.

   (5) The authors show that ARHGEF6-KO forebrain organoids were markedly smaller compared to their isogenic controls, and their study suggests that ARHGEF6 expression impacts progenitor maintenance and neurogenesis. Despite representing only a minority of the total neuronal population, I was wondering whether ARHGEF6-KO mice present brain morphology defects such as microcephaly.

   We appreciate the comment. We did not perform a morphometric analysis for microcephaly in the present study. We will add this limitation to the Discussion and note that gross brain morphology changes were not reported in the previously published ARHGEF6-KO mouse characterization (Ramakers et al., 2012). We will also clarify that the smaller organoid phenotype may reflect developmental defects that may reflect developmental defects that are not fully compensated in a reductionist in vitro model and therefore do not necessarily imply overt microcephaly in vivo.

   References

   Allen Institute for Brain Science. Allen Mouse Brain Atlas: Arhgef6 ISH data. Available from: Allen Brain Map.

   Birey, F., Andersen, J., Makinson, C. D., Islam, S., Wei, W., Huber, N., Fan, H. C., Metzler, K. R. C., Panagiotakos, G., Thom, N., O’Rourke, N. A., Steinmetz, L. M., Bernstein, J. A., Hallmayer, J., Huguenard, J. R., & Pașca, S. P. (2017). Assembly of functionally integrated human forebrain spheroids. Nature, 545(7652), 54–59. https://doi.org/10.1038/nature22330

   Liaci, C., Camera, M., Zamboni, V., Sarò, G., Ammoni, A., Parmigiani, E., Ponzoni, L., Hidisoglu, E., Chiantia, G., Marcantoni, A., Giustetto, M., Tomagra, G., Carabelli, V., Torelli, F., Sala, M., Yanagawa, Y., Obata, K., Hirsch, E., & Merlo, G. R. (2022). Loss of ARHGAP15 affects the directional control of migrating interneurons in the embryonic cortex and increases susceptibility to epilepsy. Frontiers in Cell and Developmental Biology, 10, 875468. https://doi.org/10.3389/fcell.2022.875468

   Nodé-Langlois, R., Muller, D., & Boda, B. (2006). Sequential implication of the mental retardation proteins ARHGEF6 and PAK3 in spine morphogenesis. Journal of Cell Science, 119(23), 4986–4993. https://doi.org/10.1242/jcs.03273

   Pelkey, K. A., Chittajallu, R., Craig, M. T., Tricoire, L., Wester, J. C., & McBain, C. J. (2017). Hippocampal GABAergic inhibitory interneurons. Physiological Reviews, 97(4), 1619–1747. https://doi.org/10.1152/physrev.00007.2017

   Ramakers, G. J. A., Wolfer, D., Rosenberger, G., Kuchenbecker, K., Kreienkamp, H.-J., Prange-Kiel, J., Rune, G., Richter, K., Langnaese, K., Masneuf, S., Bösl, M. R., Fischer, K.-D., Krugers, H. J., Lipp, H.-P., van Galen, E., & Kutsche, K. (2012). Dysregulation of Rho GTPases in the αPix/Arhgef6 mouse model of X-linked intellectual disability is paralleled by impaired structural and synaptic plasticity and cognitive deficits. Human Molecular Genetics, 21(2), 268–286. https://doi.org/10.1093/hmg/ddr457

   Sloan, S. A., Andersen, J., Pașca, A. M., Birey, F., & Pașca, S. P. (2018). Generation and assembly of human brain region-specific three-dimensional cultures. Nature Protocols, 13(9), 2062–2085. https://doi.org/10.1038/s41596-018-0032-7

   Yao, Z., Nguyen, T. N., van Velthoven, C. T. J., Goldy, J., Sedeno-Cortes, A. E., Baftizadeh, F., Bertagnolli, D., Casper, T., Chiang, M., Crichton, K., Ding, S.-L., Fong, O., Garren, E., Glandon, A., Gouwens, N. W., Gray, J., Graybuck, L. T., Hawrylycz, M. J., Hirschstein, D., … Zeng, H. (2021). A taxonomy of transcriptomic cell types across the isocortex and hippocampal formation. Cell, 184(12), 3222–3241.e26. https://doi.org/10.1016/j.cell.2021.04.021

   Read the original source
6. 

   Version published to 10.64898/2026.03.09.710568 on bioRxiv

   Mar 10, 2026