Sex-specific perturbations of neuronal development caused by mutations in the autism risk gene DDX3X

Mossa, Adele; Dierdorff, Lauren; Lukin, Jeronimo; Garcia-Forn, Marta; Wang, Wei; Mamashli, Fatemeh; Park, Yeaji; Fiorenzani, Chiara; Akpinar, Zeynep; Kamps, Janine; Tatzelt, Jörg; Wu, Zhuhao; De Rubeis, Silvia

doi:10.1038/s41467-025-59680-8

Download PDF

Article
Open access
Published: 15 May 2025

Sex-specific perturbations of neuronal development caused by mutations in the autism risk gene DDX3X

Adele Mossa^1,2,3,4,5,
Lauren Dierdorff^1,2,3,4,5^na1,
Jeronimo Lukin ORCID: orcid.org/0000-0002-2284-7152^1,2,3,4,5^na1,
Marta Garcia-Forn^1,2,3,4,5,
Wei Wang⁶,
Fatemeh Mamashli⁷,
Yeaji Park^1,2,3,4,5,
Chiara Fiorenzani^1,2,3,4,5,
Zeynep Akpinar^1,2,3,4,5,8,
Janine Kamps ORCID: orcid.org/0000-0002-4930-3868⁷,
Jörg Tatzelt ORCID: orcid.org/0000-0001-5017-5528^7,9,
Zhuhao Wu ORCID: orcid.org/0000-0002-2471-0555⁶ &
…
Silvia De Rubeis ORCID: orcid.org/0000-0001-9383-6883^1,2,3,4,5

Nature Communications volume 16, Article number: 4512 (2025) Cite this article

10k Accesses
4 Citations
33 Altmetric
Metrics details

Subjects

Abstract

DDX3X is an X-linked RNA helicase that escapes X chromosome inactivation and is expressed at higher levels in female brains. Mutations in DDX3X are associated with intellectual disability (ID) and autism spectrum disorder (ASD) and are predominantly identified in females (DDX3X syndrome). Using cellular and mouse models, we show that Ddx3x mediates sexual dimorphisms in brain development at a molecular, cellular, and behavioral level. During cortical neuronal development, Ddx3x sustains a female-biased signature of enhanced ribosomal biogenesis and mRNA metabolism. Compared to male neurons, female neurons display larger nucleoli, higher expression of a set of ribosomal proteins, and a higher cytoplasm-to-nucleus ratio of ribosomal RNA. All these sex dimorphisms are obliterated by Ddx3x loss. Ddx3x regulates dendritic arborization complexity in a sex- and dose-dependent manner in both female and male neurons. Ddx3x modulates the development of dendritic spines but only in female neurons. Further, ablating Ddx3x conditionally in forebrain neurons is sufficient to yield sex-specific changes in developmental outcomes and motor function. Together, these findings pose Ddx3x as a mediator of sexual differentiation during neurodevelopment and open new avenues to understand sex differences in health and disease.

Cell type specific transcriptomic differences in depression show similar patterns between males and females but implicate distinct cell types and genes

Article Open access 22 May 2023

Sex differences in neuronal systems function and behaviour: beyond a single diagnosis in autism spectrum disorders

Article Open access 09 December 2021

Satellite DNA-mediated diversification of a sex-ratio meiotic drive gene family in Drosophila

Article 06 September 2021

Introduction

Genetic variation in the X-linked gene DDX3X is a leading cause of intellectual disability (ID) and autism spectrum disorder (ASD)^1,2,3,4,5,6 (DDX3X syndrome). Over 95% of individuals with DDX3X mutations are females, and they have de novo loss-of-function mutations leading to haploinsufficiency or missense mutations^2,3,5,7. The few identified males have missense variants typically inherited from apparently asymptomatic mothers and thought to act as hypomorphic alleles^2,3,5,8. Given that ASD and ID are disproportionally diagnosed and studied in males⁹, DDX3X syndrome provides a unique window into sex differences in neurodevelopment.

DDX3X encodes an ATP-dependent DEAD-box RNA helicase with broad functions in RNA metabolism^{3,10,11,12,13,14}, including regulating mRNA translation^10,11 and RNA-containing condensates and granules via liquid-liquid phase separation (LLPS)^{15,16,17,18,19}. Complex mechanisms of mRNA translation accompany and shape brain development^20,21,22,23. The functions of DDX3X in the brain are only beginning to emerge^3,12,24,25. While Ddx3x null mice die in utero^24,26, Ddx3x haploinsufficient females have delays in meeting postnatal milestones, followed by adult behavioral and neurological deficits²⁴. Evidence in this and other mouse models indicates that Ddx3x is indispensable for the development of the hindbrain²⁵ and the forebrain^3,12,24. The neocortex is particularly vulnerable, as shown by the congenital brain malformations often diagnosed in individuals with DDX3X syndrome^2,3,5, and the experimental evidence showing that the birth^3,12 and laminar subtype specification²⁴ of cortical glutamatergic neurons require Ddx3x. In addition to these functions ingrained in prenatal development, prior evidence suggest that Ddx3x also regulates neuronal morphogenesis²⁷, which might further alter the integration of cortical glutamatergic neurons into local circuitries.

DDX3X escapes X chromosome inactivation and is expressed from both alleles in women^28,29,30 and female mice^24,31, including in the brain. Despite only a few ancestral genes having survived on the Y chromosome, DDX3X has retained a Y homologous gene (DDX3Y)^32,33,34. Hence, DDX3X belongs to a class of X-linked genes with unique potential to drive sexual differentiation, through unbalanced expression and/or divergent functions of the Y homologue^33,35. However, the interplay between DDX3X and DDX3Y remains poorly understood. In brain-conditional mouse models, Ddx3x null females have more pronounced brain cytoarchitecture anomalies than Ddx3x null males^12,25, which also show upregulated Ddx3y mRNA^12,25, suggesting mechanisms of compensation. However, DDX3X ablation in males is incompatible with life in humans and mice^5,24,26. Also, DDX3Y deletions cause subfertility/infertility^36,37,38, not brain disorders, in line with the evolution of testis-restricting regulatory regions in this gene in primates³⁹.

DDX3X lies at the intersection of X chromosome biology, sexual differentiation, and risk for ID/ASD, and yet no studies have examined the differential effect of Ddx3x loss in female and male brains. Here, we address this gap by examining the influence of sex, Ddx3x gene dosage, and their interaction on neuronal development and behavioral outcomes translationally relevant for DDX3X mutations associated with ID and ASD. We show that Ddx3x mediates sexual dimorphisms during brain development, and that Ddx3x perturbations alter the development and function of cortical glutamatergic neurons in a sex-specific manner.

Results

Ddx3x defines the complexity of dendritic arborization in a dosage- and sex-dependent manner

To investigate the influence of sex and Ddx3x allelic dosage on neuronal development, we crossed Ddx3x^+/flox females²⁴ with wild-type males (Ddx3x^+/y) to obtain embryos of four genotypes: control females (Ddx3x^+/+), floxed heterozygous females (Ddx3x^+/flox), control males (Ddx3x^+/y), and floxed males (Ddx3x^flox/y). By crossing Ddx3x^+/flox females with Ddx3x^flox/y males, we also obtained floxed homozygous females (Ddx3x^flox/flox). Cortical neurons were isolated from embryos at embryonic (E) day 15 and transfected with a bicistronic construct carrying mCherry and CRE (pAAV-Ef1a-mCherry-IRES-Cre) at plating (day in vitro 0, DIV0) (Fig. 1A). Neurons were then cultured for 9 days, fixed, and imaged (Fig. 1A, B). As intended, this genetic manipulation resulted in sparsely labeled mCherry⁺ control female neurons (Ddx3x^+/+), haploinsufficient female neurons (Ddx3x^+/-), null female neurons (Ddx3x^-/-), control male neurons (Ddx3x^+/y), and null male neurons (Ddx3x^-/y) (Fig. 1A–C).

**Fig. 1: *Ddx3x* regulates dendritogenesis.**

In line with previous data that siRNA-mediated Ddx3x knockdown alters neurite outgrowth²⁷, Ddx3x deficiency reduced dendritic complexity: Ddx3x^+/-, Ddx3x^-/-, and Ddx3x^-/y neurons had simplified dendritic arborizations (Fig. 1B–D–H) and reduced length of the overall dendritic arbor (Fig. 1I). These cellular deficits were both dosage- and sex-dependent. Female Ddx3x^+/- and Ddx3x^-/- neurons differed, indicating a gene dosage effect (Fig. 1D, Fig. 1F–I). Female Ddx3x^+/- and male Ddx3x^+/y neurons had monoallelic Ddx3x expression and yet were dissimilar (Fig. 1D–G–I). Female Ddx3x^-/- and male Ddx3x^-/y null neurons showed no overt differences in their dendritic arborization (Fig. 1D), but Ddx3x^-/- neurons had significantly less secondary and tertiary dendrites than Ddx3x^-/y neurons (Fig. 1F, G).

To corroborate these findings in vivo, we took advantage of our Sox2-Ddx3x mouse model²⁴ and contrasted the dendritic morphology of cortical neurons from adult Sox2-Ddx3x^+/- and Sox2-Ddx3x^+/+ brains. To this end, layer V pyramidal neurons were in vivo labeled by injecting retrograde adeno-associated viral particles (AAV) carrying GFP (rAVV-CAG-GFP) in the pontine reticular nucleus and performing brain tissue clearing and light-sheet microscopy three weeks post-surgery (see Methods). We prioritized this cortical subtype based on evidence of alterations of layer V neurons and motor deficits in Sox2-Ddx3x^+/- mice²⁴ and clinical observations of pontine hypoplasia in individuals with DDX3X mutations^3,40. Retrograde viral tracing resulted in robust labeling of both controlateral and ipsilateral pyramidal neurons (Fig. 2A). Using Sholl analysis and focusing on neurons whose somata reside in the motor cortex, we found not overt genotype-driven changes in the overall dendritic arborization, but the Sox2-Ddx3x^+/- neurons had significantly fewer intersections at 600-700 μm from the soma (Fig. 2A, B). We also quantified the length of the proximal apical dendrite and the remaining dendritic tract that covers the distal apical dendrite and the terminal tuft (see Methods). We found that Sox2-Ddx3x^+/- neurons had longer proximal apical dendrite at the expense of the distal apical dendrite and tuft (Fig. 2C, D). Further, Sox2-Ddx3x^+/- neurons had simpler terminal tufts (Fig. 2E, F).

**Fig. 2: Corticopontine neurons from *Ddx3x* haploinsufficient mice have simplified dendritic arborizations.**

In sum, Ddx3x is indispensable for the development of proper dendritic arborization in a cell-autonomous manner in both female and male neurons, and this function depends on dosage and sex. The endogenous Ddx3y gene is unable to fully compensate for Ddx3x function in neuronal morphogenesis. Further, Ddx3x haploinsufficiency alters the dendritic arborization of corticopontine neurons residing in the motor cortex, with the greatest impact on processes extending to the upper layers, plausibly perturbing local circuitries.

Ddx3x regulates the development of dendritic spines in a sex-dependent manner

Following the previous reasoning, we sought to discriminate the effect of sex and Ddx3x dosage on the formation and maturation of dendritic spines. We adopted the same Cre-recombination strategy described above. To isolate the discrete function of Ddx3x in this process, sparse transfections were performed at DIV9 and neurons examined at DIV14 (Fig. 3A, B). This way, dendritogenesis proceeded unperturbed until the onset of synaptogenesis, and thus synaptic phenotypes were not confounded by earlier defects in dendritic development. 10 μm long segments at least 30 μm distant from the soma were then examined in DIV14 mCherry⁺ Ddx3x^+/+, Ddx3x^+/−, Ddx3x^−/−, Ddx3x^+/y, or Ddx3x^−/y neurons (Fig. 3A). The dendritic segments were examined across primary, secondary, and tertiary dendrites and in both thin and thick dendrites (Fig. S1).

**Fig. 3: *Ddx3x* regulates the formation and development of dendritic spines.**

Quantifications of dendritic density revealed that intact Ddx3x expression is indispensable for spines formation in females but not in males: Ddx3x^−/−, but not Ddx3x^+/− or Ddx3x^−/y neurons, had more dendritic spines than control neurons (Fig. 3B, C). When measuring spine density as a function of distance from the soma, we noted that the increased number of spines in Ddx3x^-/- neurons was most prominent at 60-70 μm from the soma (Fig. 3D). Consistently, when assessing the spine density by dendrite hierarchy or diameter, we observed that the overexuberant dendritic spines in Ddx3x^-/- neurons clustered to secondary and/or thin dendrites (Fig. S2).

To examine morphological characteristics, we measured the neck length and head width of individual dendritic spines. Ddx3x^+/- female neurons displayed increased neck length (Fig. 3F, G) and increased head width (Fig. 3H, I), when compared to Ddx3x^+/+ or Ddx3x^-/- female neurons. Using the neck length and head width measures, we then classified dendritic spines into six morphological types: branched, mushroom, stubby, thin, long thin, and filopodia (Fig. 3E) (see Methods). When considering spines cumulatively, only Ddx3x^+/- female neurons had a distinct profile, with a surplus of mushroom-like and long thin spines at the expense of thin spines (Fig. 3E). We note that these data are in contrast with prior observations of reduced spine density, excess of mushroom-like spines, and increase of filopodia after acute knockdown of Ddx3x, again without considering the variables of sex and allelic dosage²⁷.

Our data indicate that Ddx3x influences the development of dendritic spines in a cell-autonomous manner. This regulation is female-specific, as Ddx3x null male neurons are unaffected. The effect of dosage in female neurons is complex, as haploinsufficiency affects spines morphology but not density, and complete loss affects density but not morphology. These observations suggest that Ddx3x is required for both the formation and the maturation of dendritic spines: Ddx3x monoallelic expression is sufficient to repress excessive spines formation at an early phase and/or facilitate pruning at a later phase, but biallelic expression is required throughout spine maturation.

Ddx3x controls the female-biased expression of ribosomal proteins and mRNA translation regulators

We next investigated the molecular changes that accompany the cellular phenotypes resulting from Ddx3x mutations. To this end, we first transduced DIV0 cortical neurons from Ddx3x^+/+, Ddx3x^+/flox, Ddx3x^+/y or Ddx3x^flox/y embryos with AAV8-Ef1a-mCherry-IRES-Cre viral particles (Fig. 4A). With this strategy, we obtained homogenous populations of Ddx3x^+/- and Ddx3x^-/y neurons, and their respective controls (Fig. 4A). We then performed label-free quantitative proteomics at DIV9, restricting analyses to proteins identified with at least two unique peptides (see Methods), and proceeded to identify up- and down-regulated proteins (indicated in the figure and below with their gene names, Fig. 4B–D).

**Fig. 4: *Ddx3x* regulates the neuronal and synaptic proteome.**

When comparing Ddx3x^+/+ and Ddx3x^+/y genotypes, we found 193 dysregulated proteins (Supplementary Data 1A), specifically 114 proteins with higher expression in female neurons than male neurons, and 79 proteins with the opposite sex bias in expression (Fig. 4B). Ddx3x was one of the most differentially expressed proteins based on sex (Fig. 4B, Supplementary Data 1A), in agreement with our prior data²⁴ and human data (Fig. S3A). Interestingly, the only other X-linked protein detected, Ubl4a, also showed higher expression in female neurons, suggesting that this gene might escape X chromosome inactivation in murine neurons. However, no sex bias in expression is seen for UBL4A in human tissues²⁹ (Fig. S3A). Also, the female-biased expression for 4 genes in addition to Ddx3x (Tufm, Scfd2, Lrpprc, and Eif3i) is conserved in the human cortex, as shown by bulk transcriptomic data³⁰ (Fig. S3B) (see Supplemental Note).

When comparing Ddx3x^+/- and Ddx3x^+/+ genotypes, we found 182 dysregulated proteins (Supplementary Data 1B), specifically 59 up-regulated and 123 down-regulated proteins (including Ddx3x) (Fig. 4C). When comparing Ddx3x^-/y and Ddx3x^+/y genotypes, we found 98 dysregulated proteins (Supplementary Data 1C), specifically 39 up-regulated and 59 down-regulated proteins (including Ddx3x) (Fig. 4D). Previous analyses have detected Ddx3y mRNA in the brain^12,25. However, since DDX3Y has been shown to undergo post-transcriptional restriction in male germ cells^39,41, we analyzed our proteomics data in search of DDX3Y-specific peptides in male neurons. We identified a DDX3Y-specific peptide (KPILVATAVAAR in DDX3Y, Uniprot #Q62095; SPILVATAVAAR in DDX3X, Uniprot #Q62167) exclusively in the Ddx3x^-/y dataset. We also detected DDX3Y in Ddx3x^-/y cultures by Western blot (Fig. S4A). These data corroborate the expression of DDX3Y, albeit at low levels, in male neurons after ablation of Ddx3x.

We then interrogated biological processes annotated in Gene Ontology (GO) with the datasets of 193 (Fig. 4B), 182 (Fig. 4C), and 98 (Fig. 4D) proteins dysregulated in Ddx3x^+/+, Ddx3x^+/-, Ddx3x^-/y neurons, respectively. Using Fisher’s exact test with Bonferroni correction, we found that of the 193 proteins with a sex bias in expression, the 114 upregulated in female neurons converged on GO terms related to RNA processing and mRNA translation terms (Fig. 4E). Intriguingly, the GO results of the proteins dysregulated in Ddx3x^+/- when compared with Ddx3x^+/+ neurons offered a specular scenario, with an enrichment of mRNA translation terms for downregulated proteins in Ddx3x^+/- neurons (Fig. 4F). For example, proteins in the top enriched term “cytoplasmic translation” had higher expression in female neurons than male neurons, but reduced expression in female neurons with Ddx3x haploinsufficiency compared to control female neurons (Fig. 4E, F). Among the proteins with reduced expression in Ddx3x^+/- compared to Ddx3x^+/+ neurons, were 5 tRNA synthetases, 5 ribosomal proteins of the small 40S subunit, 3 initiation factors, 8 ribosomal proteins of the large 60S subunit (Fig. 4G), and 5 proteins involved in mitochondrial translation (Mrps18b, Mrpl24, Tufm, Mrps26, Mrps30). The reduced expression of two ribosomal proteins of the large subunit (Rpl36a; Rpl11), one protein of the small subunit (Rps5) and a translation initiation factor (Eif3k) in Ddx3x^+/- neurons was also validated by Western blot (Fig. S4B, C). The brain-specific form of the translation termination factor eRF3⁴², ERF3B (encoded by Gspt2), had enhanced expression in Ddx3x^+/- neurons (Fig. 4G). The observed reduction in ribosomal proteins aligns with findings in a DDX3X knockdown model in lymphoma cell lines⁴³. Strikingly, 7 of the 13 ribosomal proteins reduced in Ddx3x^+/- neurons are encoded by mRNAs that were found to interact with DDX3X based on iCLIP data in lymphoma cell lines ⁴³ (i.e., Rps3, Rps5, Rps8, Rps19, Rpl11, Rpl13, Rpl30), and 5 were found to have reduced translational efficiency in the DDX3X knockdown lymphoma model⁴³ (i.e., Rps3, Rps5, Rps19, Rpl11, Rpl13). These data suggest that in cortical neurons Ddx3x regulates the expression of core components of the protein synthesis machinery.

By contrast, we found no significant enrichment for the 98 proteins dysregulated in Ddx3x^-/y compared with Ddx3x^+/y neurons. Several of the proteins found dysregulated are indispensable for proper dendritic development, e.g. Dcx⁴⁴ and Plxnd1⁴⁵, or for neuronal migration, e.g., Rab23⁴⁶ and Fbxo45⁴⁷. This suggests that the overlapping morphogenesis defects observed in Ddx3x^+/- and Ddx3x^-/y neurons (Fig. 1) result from distinct molecular mechanisms. Further, the differentially expressed proteins in Ddx3x^-/y neurons (but not those dysregulated in Ddx3x^+/- neurons) were enriched for proteins encoded by targets of two RNA-binding proteins associated with ID/ASD, FMRP and RBFOX (Fig. S3D).

To gain insights on the disease relevance of our datasets (Fig. 4B–D), we examined the intolerance to loss-of-function genetic variation in their human orthologues using the well-established metrics of loss-of-function observed/expected upper bound fraction (LOEUF)⁴⁸ (Supplemental Note). Human orthologues of sex-biased proteins and proteins dysregulated in Ddx3x mutant neurons, either female or male, all displayed skewed distribution toward lower LOEUF deciles (Fig. S3C), in keeping with the high genomic constraints for genes expressed in brain. Also, human orthologues of the sex-biased proteins included genes associated with severe neurodevelopmental disorders (DDG2P⁴⁹ genes), ASD risk genes (ASC⁵⁰ genes), and/or epilepsy (EPI^51,52,53 genes), albeit with no statistically significant enrichment. For example, 18 proteins encoded by DDG2P genes had higher expression in female neurons (e.g., Ddx3x, Auh, Tufm) and another 8 had higher expression in male neurons (e.g., Napb, Shank2, Ank2) (Supplementary Data 1). 5 proteins encoded by ASC gene orthologues had a sex bias in expression, and they all had higher expression in male neurons (Adcy5, Coro1a, Ank2, Shank2, Cul3) (Supplementary Data 1). 30 proteins encoded by DDG2P gene orthologues had altered expression in Ddx3x^+/- neurons, specifically 10 up-regulated proteins (e.g., Shank2, Nrxn3) and 20 down-regulated proteins (e.g., Auh, Ccdc88a, Tufm). 13 proteins encoded by DDG2P gene orthologues had altered expression in Ddx3x^-/y neurons, specifically 7 up-regulated proteins (e.g., Dcx, Adsl, Shank1) and 6 down-regulated proteins (e.g., Rab23, Napb).

To explore the molecular correlates of the altered dendritic spines in Ddx3x^+/- neurons, we delineated the proteomic profile of cortical synaptosomes purified from Sox2-Ddx3x^+/- and Sox2-Ddx3x^+/+ female mice at postnatal day 21, a peak for synaptogenesis. As for prior experiments, we restricted analyses to proteins identified with at least two unique peptides (see Methods) and proceeded to identify up- and down-regulated proteins (indicated in the figure and below with the gene names, Fig. 4H). When comparing Ddx3x^+/+ and Ddx3x^+/- synaptosomes, we found 498 dysregulated proteins (Supplementary Data 2), specifically 147 proteins up-regulated and 351 down-regulated in Ddx3x^+/- synaptosomes (Fig. 4H). Ddx3x was found to be expressed at synapses in line with a prior synaptic proteomic dataset⁵⁴ and, as expected, reduced in expression in Ddx3x^+/- synaptosomes (Fig. 4H, Supplementary Data 2). As expected, the differentially expressed proteins were enriched in synaptic terms (Fig. 4I), as 114 of the 346 proteins dysregulated in Ddx3x^+/- synaptosomes were bona fide synaptic proteins as defined by the SynGO consortium⁵⁵. We then crossed the list of genes encoding the 498 differentially expressed proteins with the DDG2P genes⁴⁹, ASC genes⁵⁰, and EPI genes^51,52,53, and found a statistically significant enrichment for DDG2P genes (Fig. S3D). 84 were encoded by DDG2P genes, 9 by ASC genes (4 of which also captured by DDG2P; Psmd12, Nrxn1, Scn2a, and Plxna1), and 3 by EPI genes (all 3 captured by DDG2P genes, Ube3a, Eef1a2, and Scn2a). At the intersection of the differential expressed proteins, established SynGO genes, and disease risk genes are critical regulators of synaptic function, including Dlg3, Nrxn1, Nrxn2, Scn2a, Cntnap2 and Rac1. When crossing the list of dysregulated proteins in Ddx3x^+/- synaptosomes (Fig. 4H) and neurons (Fig. 4C), we found 19 shared proteins in addition to Ddx3x, including Rpl22a and Ccdc88a (also known as Girdin) (Fig. S3E).

These data map the molecular correlates of the deficits in neuronal morphogenesis (Figs. 1, 2) and dendritic spines (Fig. 3) caused by perturbations in Ddx3x, and show that there are sex-specific Ddx3x signatures, including alterations of ribosomal proteins and translational regulators exclusively in female neurons.

Ddx3x mediates sex dimorphisms in neuronal ribosome biogenesis

Compared to male neurons, female neurons have higher dosage requirements for Ddx3x, as it has evolved as an X chromosome escape gene^24,29,31. Female neurons also display enhanced expression of components of the protein synthesis machinery (Fig. 4B–E). Ddx3x^+/- neurons have dampened expression of ribosomal proteins compared to female control neurons (Fig. 4C–F). Therefore, we conjectured that Ddx3x might mediate the female-biased surplus of protein synthesis components.

To test this hypothesis, we verified if the higher expression of proteins in female neurons requires Ddx3x. We started by intersecting proteins with altered levels in Ddx3x^+/- neurons (Fig. 4C) with those showing differential expression based on sex (Fig. 4B) and identified 37 proteins (Supplementary Data 3). We then filtered for proteins that have at least 30% higher expression in female neurons compared to male neurons (Expression_Ddx3x+/+/Expression_Ddx3x+/y > 1.3) and identified 20 proteins. Of these 20, 17 have similar levels of expression when comparing female haploinsufficient neurons and male control neurons (Expression_Ddx3x+/-/Expression_Ddx3x+/y = 1 ± 0.15), indicating that Ddx3x deficiency obliterates their female bias in expression (Fig. 4J, Supplementary Data 3). Six of these proteins are downregulated in Sox2-Ddx3x^+/- synaptosomes (Auh, Tufm, Naxd, Acad8, Farsb, and Tfam). Six proteins are also encoded by genes whose human orthologues are DDG2P genes (TUFM, TFAM, NUP133, NAXD, FARSB, and AUH) and one (TUFM) shows higher expression in human cortices from female donors than tissues from male donors (Fig. S3B). The remaining three proteins (Rbm4b, Pds5b, Ssbp1) do not meet these stringent criteria but show an attenuation of their sex bias (Supplementary Data 3). Of the 17 proteins with female-biased expression, five are implicated in mRNA translation (Tufm, Rps3, Rpl11, Mrpl24, Farsb) and another four are RNA-binding proteins (Igf2bp1, Hnrnpf, Celf3, Auh). Rps3 and Rpl11 were also detected in a DDX3X iCLIP dataset⁴³. These data suggest that female neurons have higher expression of a set of ribosomal and translational regulators and that this female bias depends on the biallelic expression of Ddx3x.

We next sought to verify whether this signature resulted into changes in global protein synthesis rates. To this end, we performed a puromycin-based assay adapted from the SUnSET technique^56,57. We trasduced DIV0 cortical neurons from Ddx3x^+/+, Ddx3x^+/flox, or Ddx3x^+/y E15 embryos with AAV8-Ef1a-mCherry-IRES-Cre viral particles. At DIV9, we performed a 10-min puromycin pulse followed by a 30 min chase (Fig. S5A). This approach was effective in detecting newly synthesized proteins, as shown by abrogated puromycin incorporation in neurons treated with the protein synthesis inhibitor cycloheximide prior to the pulse (Fig. S5A, B). When comparing the three genotypes, we did not detect significant differences in the puromycin labeling (Fig. S5B, C), suggesting that there are no overt changes in global protein synthesis detectable with this approach.

We reasoned that the Ddx3x-dependent female bias might reflect more complex aspects of ribosome formation, which undergoes sophisticated forms of regulation in neural cells^21,58. Therefore, we measured the abundance and subcellular distribution of ribosomal RNA (rRNA). Using a monoclonal antibody raised against the 5.8S rRNA (clone Y10b^{58,59,60,61,62,63,64,65}), we quantified rRNA in the developing motor cortex of Sox2-Ddx3x^+/+, Sox2-Ddx3x^+/-, or Sox2-Ddx3x^+/y neonates (Fig. 5A). Based on previous observations of altered layer V cortical neurons expressing the transcription factor CTIP2 in Sox2-Ddx3x^+/- females²⁴, we conducted analyses in both CTIP2⁺ neurons and neighboring CTIP2^- cells (Fig. 5A–C). Irrespective of genotype, most rRNA was detected in the cytoplasm (Fig. 5B, C), as in previous reports^58,59,60,61. In CTIP2⁺ neurons, the cytoplasm-to-nucleus ratio of the rRNA was ~87:13 % in Sox2-Ddx3x^+/+ females, but dropped to ~75:25 % in both Sox2-Ddx3x^+/- and Sox2-Ddx3x^+/y mice (Fig. 5B). The ratio was detected at ~75:25 % in CTIP2^- cells, irrespective of genotype (Fig. 5C). These observations suggest that female neurons might have a different rate of ribosomal maturation specifically in CTIP2⁺ neurons, and that this sex bias depends upon Ddx3x. We note that Rpl11 is among the ribosomal proteins with higher expression in female neurons but reduced levels in Ddx3x haploinsufficient neurons (Fig. 4J, Fig. S4B, C) and that this protein has a critical role in pre-ribosome assembly^66,67.

**Fig. 5: *Ddx3x* mediates a female bias in ribosome biogenesis.**

When quantifying the mean rRNA expression, male neurons had overall higher rRNA levels than female neurons, for both CTIP2⁺ and CTIP2^- cells (Fig. 5D, E). Put together with the cytoplasm-to-nucleus ratio, these results suggest sex differences in ribosome biogenesis, where higher rates of ribosome export/maturation are preferred in CTIP2⁺ neurons in females and an overall higher production of ribosomes is more favorable in males. Further, the CTIP2⁺ neurons from Sox2-Ddx3x^+/- neonates have a ~ 15% net reduction in rRNA (Fig. 5D). These observations suggest that Ddx3x mutations in neurons are associated with altered rRNA expression and ribosomal deficits.

As canonical ribosome biogenesis takes place in the nucleolus, we conjectured that sex and Ddx3x dosage might alter the nucleoli. To test this hypothesis, we examined the number and volume of nucleoli in cortical neurons from the developing motor cortex of Sox2-Ddx3x^+/+, Sox2-Ddx3x^+/-, and Sox2-Ddx3x^+/y neonates using the nucleolar marker fibrillarin^59,62,68 (Fig. 5F). When compared to their male counterparts, CTIP2⁺ female neurons had on average larger nucleoli (Fig. 5G) and higher total nucleolar volume (Fig. 5H), without changes in the number of nucleoli (Fig. 5J). CTIP2⁺ female neurons also had larger nuclei than male neurons and female haploinsufficient neurons (Fig. 5I). The specificity of these differences for CTIP2⁺ neurons goes hand in hand with the Ddx3x-dependent female bias toward higher repartitioning of cytoplasmic ribosomes (Fig. 5B, C), corroborating that ribosome biogenesis differ in female and male neurons via pathways that depend on Ddx3x biallelic expression.

We examined nucleolar measures also in cultured cortical neurons, which are mostly expressing CTIP2 (Fig. S6A). We found that the Ddx3x-dependent female bias manifested in the number of nucleoli, and that Ddx3x^-/y neurons did not show changes in nucleoli compared with male controls (Fig. S6B–F). We note that although there is convergence in a Ddx3x-dependent female-biased nucleolar signature ex vivo (Fig. 5) and in vitro (Fig. S6B–F), the nature of this signature differs. These differences might result from the fact that the Ddx3x deletion in the cultured system ensues at E15, plausibly post-mitotically, while it starts at gastrulation (thus also in the neural progenitors) in Sox2-Ddx3x^+/- mice.

Altogether, these data show that biallelic expression of Ddx3x in female neurons is required to impart sexually dimorphic characteristics and that Ddx3x haploinsufficiency results in deficits in nucleolar properties and ribosome biogenesis.

DDX3X and its paralog DDX3Y have only partially redundant functions in neuronal development

The data so far show that Ddx3x shapes sexual differentiation during neuronal development. This might result from unbalanced expression with Ddx3y and/or divergent molecular properties. To eliminate dosage as a variable and formally assessed functional redundancy, we exogenously expressed human DDX3X (hDDX3X) or DDX3Y (hDDX3Y) in Ddx3x^+/- neurons (Fig. 6A). Further, to verify that potential hDDX3X-induced phenotypes could be circumscribed to its catalytic functions, we used an ATPase-deficient mutant (E348Q)^69,70,71 (Fig. 6A).

**Fig. 6: The functions of DDX3X in dendritogenesis and ribosome biogenesis depend on its catalytic activity and are partially redundant with DDX3Y.**

Cortical neurons were isolated from E15 Ddx3x^+/flox embryos and transfected with the pAAV-Ef1a-mCherry-IRES-Cre with or without expression vectors carrying tagged hDDX3X, hDDX3Y, or hDDX3X-E348Q (Fig. 6A). This genetic manipulation resulted in sparse mCherry⁺ haploinsufficient neurons (Ddx3x^+/-), mCherry⁺HA⁺ haploinsufficient neurons co-expressing hDDX3X (Ddx3x^+/hDDX3X), mCherry⁺HA⁺ haploinsufficient neurons co-expressing the ATPase-deficient hDDX3X (Ddx3x^+/E348Q), or mCherry⁺FLAG⁺ haploinsufficient neurons co-expressing hDDX3Y (Ddx3x^+/hDDX3Y) (Fig. 6A, B). We then examined dendritic development and nucleolar properties, in keeping with the design of our previous analyses (Figs. 1, 2, 5, Fig. S6B–F).

We found that hDDX3X or hDDX3Y alleviated the dendritic deficits observed in Ddx3x^+/- neurons (Fig. 6B, C). Co-expression of hDDX3X or hDDX3Y significantly increased nucleolar volume (Fig. 6D–F) but had divergent effects on the number of nucleoli (Fig. 6G). In fact, Ddx3x^+/hDDX3Y showed a number of nucleoli comparable to Ddx3x^+/- neurons, but Ddx3x^+/hDDX3X had a significant reduction compared to all other genotypes (Fig. 6G). The lack of effect of hDDX3Y on the number of nucleoli is compatible with the lack of a nucleolar phenotype in Ddx3x^-/y neurons (Fig. S6). Nuclear features remained unaltered across all genotypes (Fig. 6H). Co-expressing hDDX3X-E348Q had no effect on any of the measures (Fig. 6B, C). Considering that the nucleolus is a phase-separated organelle, our data align with prior evidence showing that human DDX3X and DDX3Y have distinct propensities to LLPS^15,18,19.

This body of findings demonstrates that intact catalytic activity is required for DDX3X to regulate neuronal morphogenesis and nucleolar dynamics, and that the balance between DDX3X and DDX3Y resides in both their expression and function.

Loss of Ddx3x in the forebrain leads to developmental delays and adult motor deficits

We then next sought to assess whether the interaction between Ddx3x dosage and sex observed at the molecular and cellular level translates into behavioral outcomes.

Complete knockout of Ddx3x is incompatible with life in male mice^24,26, while Sox2-Ddx3x^+/- females have postnatal motor delays, adult motor deficits, hyperactivity, anxiety-related behaviors, and cognitive impairments²⁴. To be able to compare female and male mutant mice, we selected a forebrain-conditional Ddx3x knock-out model, based on clinical observations^2,3,5 and evidence in mouse models^3,12,24 (including those presented here) showing that forebrain neurons are particularly vulnerable to DDX3X mutations. We crossed the Ddx3x^flox line with an Emx1^IRES-Cre driver line that expresses the CRE recombinase under the endogenous Emx1 promoter in the telencephalon as early as E10.5^72,73. We obtained four genotypes: Emx1-Ddx3x^+/+ control females, Emx1-Ddx3x^+/- haploinsufficient females, Emx1-Ddx3x^+/y control males, and Emx1-Ddx3x^-/y null males. As intended, Emx1-Ddx3x^-/y null males show loss of DDX3X in the forebrain, but not in the midbrain or hindbrain (Fig. 7A). Emx1-Ddx3x mutants were born at the expected mendelian ratio, did not show reduced viability from birth to the postnatal day 21, and survived to adulthood.

**Fig. 7: *Ddx3x* loss in the forebrain causes adult motor deficits.**

Emx1-Ddx3x^-/y null males showed delayed growth over the first four weeks of life (Fig. 7B) which then normalized in adulthood (Fig. 7C). Emx1-Ddx3x^-/y null males showed no major physical (Fig. S7), sensory (Fig. S8), or motor (Fig. S9) developmental delays, but had a statistically significant delay in the eruption of the top tooth (Fig. S7D), and the acquisition of motor abilities such as grip strength (Fig. S9E) and air righting (Fig. S9F) showed a distinct trajectory over time. The developmental trajectory of the Emx1-Ddx3x^+/- heterozygous females was mostly indistinguishable from that of their control littermates (Fig. 7B, Fig. S7–9), except for precocious development of the forelimb grasp reflex which then reached a plateau (Fig. S8B). Further, both Emx1-Ddx3x^-/y and Emx1-Ddx3x^+/- juvenile mice displayed subtle changes suggestive of a wider gait, appearing at P22 in Emx1-Ddx3x^+/- and only at P30 in Emx1-Ddx3x^-/y males (Fig. S10). This is in contrast with what we previously observed in Sox2-Ddx3x^+/- mice, which showed a reduction in horizontal and diagonal parameters at P22, resolved by P30²⁴. Collectively, these data indicate that the previously reported developmental delays observed in the global Sox2-Ddx3x model are not exclusively dependent on the forebrain, and that nevertheless there is an interaction between sex and genotype and/or a complex dosage effect, as some phenotypes are detected in haploinsufficient females but not in null males (Figs. S8B, S10).

Among the most clinically salient phenotypes observed in the Sox2-Ddx3x mouse model are motor delays and poor motor coordination and balance²⁴, reminiscent of the pervasive motor manifestations in individuals with DDX3X syndrome^2,3,5. Further, corticopontine neurons, which are engaged in critical motor circuits, have dendritic alterations when Ddx3x is mutated (Fig. 2). Therefore, we assessed motor function in the Emx1-Ddx3x mutant line. During handling one week prior to behavioral testing, we noted no differences in physical appearance or weight (Fig. 7C). In terms of baseline activity, we observed reduced inverted screen performance for both Emx1-Ddx3x^+/- and Emx1-Ddx3x^-/y mutants compared to controls (Fig. 7D), and reduced number of jumps during spontaneous activity monitoring for Emx1-Ddx3x^+/- females (Fig. S11B), but no other changes (Fig. S11A, S11C–E).

To assess neuromuscular strength, we employed a wire hanging test (Fig. 7E, F). We detected a significant effect of genotype on performance, with Emx1-Ddx3x^-/y null males showing reduced latency to fall from the wire when compared to male littermates (Fig. 7E), similar to the reduced endurance observed in Sox2-Ddx3x^+/- mice²⁴. To assess motor coordination and balance, we used a balance beam test and a vertical pole test. When walking on the balance beam, control females covered a larger distance on the beam (Fig. 7G) than male littermates, plausibly because of their lower body weight (Fig. 7C). In line with observations in Sox2-Ddx3x^+/- mice²⁴, Emx1-Ddx3x^+/- covered a shorter distance on the beam than control female littermates (Fig. 7G). No differences in the number of slips or latency were noted (Fig. S12A, B). No sex or genotype differences were observed when mice were climbing down a vertical pole (Fig. S12C, D), in contrast to previous findings of impairments in this motor task in Sox2-Ddx3x^+/- mice²⁴. To further test motor coordination and endurance and extend on motor learning, we used the accelerating rotarod test, with a design of four trials each day, over three days. In line with the outcomes of the balance beam test (Fig. 7G), control females outperformed male littermates (Fig. 7H). While Emx1-Ddx3x^+/- females were able to learn the task, they showed reduced performance compared to control females (Fig. 7H). Emx1-Ddx3x^-/y males showed both impaired learning and coordination when compared to control males (Fig. 7H).

These observations are consistent with the defects in corticopontine neurons (Fig. 2), indicate that Ddx3x deficiency in the forebrain is sufficient to produce motor sequelae reminiscent of those found in the patient population⁷⁴, and reaffirm the complex sex-by-dosage interactions for the Ddx3x locus.

Discussion

By virtue of its nature as an X chromosome inactivation escapee²⁹, DDX3X is one of the top 15 female-biased genes in the human cortex³⁰ (Fig. S3), and one of the top 5 proteins with higher expression in female cortical murine neurons (Fig. 4B, Supplementary Data 1). Despite the dramatic Y chromosome decay during evolution, DDX3X and another 18 escapee genes have retained Y homologues³², suggesting that the equilibrium between the X and Y partners (or the lack thereof) contributes to shape sex dimorphisms^33,35.

Our study provides evidence that Ddx3x acts as a mediator of sexual differentiation during brain development. First, we show that Ddx3x is necessary for the expression of a female-biased molecular and cellular signature associated with ribosome biogenesis (Figs. 4–6). Second, Ddx3x is indispensable for the formation and maturation of dendritic spines in female neurons only, as spines of male control or null neurons are indistinguishable (Fig. 3). Third, female mice outperform males in two motor tasks, but this sex advantage is lost in Ddx3x mutant females (Fig. 7). These data, combined with evidence in the literature, support a scenario whereby DDX3X and DDX3Y have evolved to have only partially redundant functions^15,18,19, in addition to a divergent expression pattern^37,39,41. In fact, human DDX3X and DDX3Y differ in their catalytic properties and propensity to LLPS^15,18,19. Coherently, they have distinct impact on nucleoli (Fig. 6). Further, unlike other X-Y pairs, DDX3X and DDX3Y can reciprocally regulate their dosage, likely as a result of inheriting high dosage sensitivity from their common autosomal ancestor⁷⁵. In fact, in line with previous data in mouse models^12,25, we detect DDX3Y protein expression in male neurons when Ddx3x is silenced (Fig. S4A, Supplementary Data 1). Nevertheless, DDX3Y is unable to compensate for DDX3X during neurodevelopment, as evidenced by male cases with pathogenic DDX3X mutations^2,3,5,8 and the molecular (Fig. 4), cellular (Fig. 1), and behavioral (Fig. 7) sequalae of Ddx3x loss in males.

Our observations add a layer of complexity to the molecular alterations resulting from DDX3X haploinsufficiency in neurons. DDX3X has been postulated to act at the step of translational initiation and facilitate the translation of a subset of mRNAs with 5’UTRs bearing complex secondary structures through its helicase activity^10,11. Our data in neurons show distinct molecular signatures based on sex. Ddx3x loss in male neurons results in a distinct and narrowed impact on proteins directly necessary for neurite outgrowth (Fig. 4), with no noticeable nucleolar alterations (Fig. S6B–F). Ddx3x haploinsufficiency in female neurons causes a downregulation of constitutive components of the protein synthesis machinery (Fig. 4, Fig. S4B, C), orthogonally corroborated by a reduction in rRNA and nucleoli volume (Fig. 5). Whether Ddx3x haploinsufficiency also impacts the processing and maturation of rRNAs remains to be examined. The ribosomal deficits might in turn alter the synthesis of specific proteins required to sustain dendritic growth, yielding the reduced dendritic branching resulting from Ddx3x perturbations (Figs. 1, 2).

The implications of our findings relating Ddx3x to ribosome biogenesis are twofold. First, ribosome biogenesis is dynamically regulated during cortical development^21,62,76 and protein synthesis rate is an inherent feature of cortical neuronal subtype²³. The evidence of a Ddx3x-dependent female bias in ribosome biogenesis in CTIP2-expressing cortical neurons introduces a new player (Ddx3x) and a new variable (sex). More broadly, our study adds on an ever-evolving landscape of mechanisms of protein synthesis in neurons, where ribosomes have distinct conformations⁷⁷, specific co-regulators⁷⁸, and non-canonical mechanisms of biogenesis⁵⁸. Second, a class of genetic conditions caused by mutations in proteins regulating LLPS and resulting in nucleolar alterations has emerged⁶⁸. Our findings suggest that DDX3X syndrome belongs to this group of disorders.

Our findings have direct relevance for our understanding of how neurodevelopment goes awry when DDX3X is mutated, by adding to the clinical^2,3,5 and functional^3,12,24 data indicating that forebrain neurons, particularly in the cortex, are exquisitely vulnerable to perturbations in DDX3X. We show that Ddx3x expression in the forebrain is required for intact development and function of neurons engaged in motor circuits, as Ddx3x-deficient corticopontine neurons have altered dendritic arborization (Fig. 2) and forebrain-conditional Ddx3x mutant mice have motor deficits, although with significant sex differences (Fig. 7). We also show that Ddx3x is needed for the development of cortical neurons post-mitotically, in line with a previous report²⁷, thus expanding the window of risk for DDX3X mutations beyond prenatal development^3,12. Synaptic function, in particular, has been pinpointed as a major nexus of risk for ASD⁷⁹, and genetically engineered mouse models for ASD have also shown deficits in synaptic development^80,81,82,83. Importantly, dendritic growth, de novo synapse formation, and synapse maturation continue to be plastically reshaped during postnatal periods, including in response to experience^84,85,86, thus offering a more feasible window for therapeutic intervention. Our previous data²⁴ lend support for potential benefits of intervening postnatally on Ddx3x mutant mice. In fact, we found that one-year old Sox2-Ddx3x^+/- mice have more profound motor deficits than naïve 4-month-old Sox2-Ddx3x^+/- mice, but this decline is no longer seen in one-year old mice pre-exposed to behavioral training at 4 months of age. These data suggest that there might be mechanisms maintaining the pathological status (perhaps through the synaptic cellular and molecular alterations we found) that could be manipulated postnatally. This might add to the growing body of evidence showing that restoring expression/function of some ASD risk genes in adult mice can alleviate their molecular and/or behavioral deficits^{87,88,89,90,91}. Future studies will be critical to understand how the molecular and cellular alterations identified in this study can be targeted for the development of novel therapeutics for DDX3X syndrome.

Methods

Mouse lines

All animal procedures were approved by the Institutional Animal Care and Use Committee of the Icahn School of Medicine at Mount Sinai. To generate the neuronal cultures, we crossed Ddx3x^flox/+ females²⁴ with wild-type Ddx3x^+/y males or Ddx3x^flox/y males. To generate the Sox2-Ddx3x mice, Ddx3x^flox/flox females were crossed with B6.Cg-Edil3^{Tg(Sox2-Cre)1Amc}/J males (Sox2-Cre/+)⁹² (The Jackson Laboratory, stock number #008454). To generate the Emx1-Ddx3x mice, Ddx3x^flox/flox females were crossed with B6.129S2-Emx1^tm1(cre)Krj/J (Emx1^IRES-Cre/+) hemizygous males^72,73 (The Jackson Laboratory, stock number #005628). The colony was maintained in a room on a 12/12 h light/dark cycle, with lights on at 7 A.M. at a constant temperature of 21-22 °C and 55% humidity. Standard rodent chow and potable water were available ad libitum. Animals were socially housed, with 3-5 mice per cage. Mice were weaned at P21. The colony was maintained on a C57BL/6 J background. Genotyping is described in the Supplemental Note.

Primary cortical cultures

Cultures were prepared from E15 single embryos with a modified protocol⁹³. Cortices were isolated in ice-cold Hank’s balanced salt solution (HBSS) and then digested with pre-warmed papain containing DNase (Worthington, #LK003150) at 37 °C and 5% CO₂ for 30 minutes. Tissue chops were washed 3 times with Neuronal Plating Medium and then centrifuged for 3 minutes at 400 g. Mechanical dissociation was then performed in 1 ml of pre-warmed Neurobasal Medium (Thermo Fisher Scientific, #21103049). Cell density was determined with the Countess Automated Cell Counter (Thermo Fisher Scientific), with a yield on the order of 3-4 ×10⁶ cells per embryo. Irrespective of timing of transfection/infection (see below), cells were maintained in Neuronal Plating Medium, and 90 minutes after plating the medium was replaced with pre-warmed Neurobasal Plus (Thermo Fisher Scientific, #A3582901) supplemented with B27 Plus (Thermo Fisher Scientific, #A3582801). At DIV4, cells were treated with 1 µM cytosine arabinoside (1-b-D-arabinofuranosylcytosine (Sigma-Aldrich Inc, #C1768). Genotyping is described in the Supplemental Note.

DNA transfections

For dendritic arborization analyses in Fig. 1, 2 x 10⁵ cells were incubated for 1 h in a 50 µl of Neurobasal containing 0.5 µl of Lipofectamine (Thermo Fisher Scientific, #11668019) and 500 ng of pAAV-Ef1a-mCherry-IRES-Cre (Addgene, #55632) in 1.5 ml conical tubes right before plating (DIV0) and then seeded onto poly-L-lysine-coated coverslips in 24-well plates containing Neuronal Plating Medium. Coverslips were fixed (see below) at DIV9. For dendritic spines analyses (Fig. 3), 1.5 ×10⁵ neurons were plated at DIV0 onto poly-L-lysine-coated coverslips in 24-well plates containing Neuronal Plating Medium and then transfected at DIV9 by adding 100 µl of Neurobasal containing 0.75 µl of Lipofectamine and 500 ng of pAAV-Ef1a-mCherry-IRES-Cre directly onto each well. Coverslips were fixed (see below) at DIV14. For analyses in Fig. 6, DIV0 2 ×10⁵ cells were incubated for 1 h in a 100 µl of Neurobasal containing 0.5 µl of Lipofectamine (Thermo Fisher Scientific, #11668019), and 500 ng of pAAV-Ef1a-mCherry-IRES-Cre (Addgene, #55632) with or without pCI-Neo-HA-DDX3X, pCI-Neo-HA-DDX3X-DQAD (E348Q), or pTWIST-CMV-FLAG-DDX3Y, in 1.5 ml conical tubes and then seeded onto poly-L-lysine-coated coverslips in 24-well plates containing Neuronal Plating Medium. Coverslips were fixed (see below) at DIV9. For nucleoli analyses in Fig. S6, DIV0 2 × 10⁵ cells were incubated for 1 h in a 100 µl of Neurobasal containing 0.5 µl of Lipofectamine (Thermo Fisher Scientific, #11668019), and 500 ng of pAAV-Ef1a-mCherry-IRES-Cre or 500 ng of a control pAAV-Ef1a-mCherry (Addgene, #114470). Coverslips were fixed (see below) at DIV9.

Adeno-associated virus transductions

For analyses in Fig. 4, 10⁶ cells were directly plated onto poly-L-lysine-coated 35 mm Petri dishes containing Neuronal Plating Medium, and cells were then transduced with 1 µl of AAV8-Ef1a-mCherry-IRES-Cre (Addgene, #55632-AAV8, 1 × 10¹³ vg/ml, multiplicity of infection (MOI) of 10,000). Cells were harvested at DIV9 (see below).

Immunostaining on cultured neurons

At DIV9 (Figs. 1, 6, Fig. S6A–F) or DIV14 (Fig. 3), coverslips were washed with pre-warmed 1X phosphate buffered saline (PBS) and fixed with a pre-warmed solution of 4% paraformaldehyde (PFA) /4% sucrose in 1X PBS at room temperature (RT) for 10 minutes. Coverslips were then washed once with 50 mM glycine in 1X PBS and twice with 1X PBS /0.1% Triton X-100 at RT for 5 minutes, and then treated with 1X PBS/0.3% Triton X-100 for 20 minutes to allow permeabilization. After blocking in 10% donkey serum/1X PBS /0.1% Triton X-100 for 1 h at RT, coverslips were incubated with goat anti-mCherry polyclonal antibody (SICgen antibodies, #AB0040, 1:200), a rabbit anti-DDX3X polyclonal antibody (Thermo Fisher Scientific, #A300-474A, 1:400), a rabbit anti-Fibrillarin polyclonal antibody (Abcam, #ab5821, 1:2,000), a mouse anti-HA-Tag (6E2) monoclonal antibody (Cell Signaling, #2367, 1:100), a mouse anti-FLAG M2 monoclonal antibody (Sigma-Aldrich, #F3165, 1:100), a rat anti-CTIP2/BCL11B monoclonal antibody (Millipore, #MABE1045, 1:300), and/or Hoechst dye (Thermo Fisher Scientific, #H3570, 1:1,000), diluted in blocking solution overnight at 4 °C. Coverslips were then washed 3 times with 1X PBS and incubated for 1 h at RT with a donkey anti-rabbit antibody conjugated with Alexa Fluor 647 (Thermo Fisher Scientific, #A-31573, 1:200), a donkey anti-goat antibody conjugated with Alexa Fluor 568 (Thermo Fisher Scientific, #A-11057, 1:200), a donkey anti-mouse antibody conjugated with Alexa Fluor 488 (Thermo Fisher Scientific, #A-21202, 1:200), or a donkey anti-rat conjugated with Alexa Fluor 488 (Abcam, #ab150153, 1:200), and/or Hoechst dye (Thermo Fisher Scientific, #H3570, 1:1,000), in 1X PBS/0.1% Triton X-100. Coverslips were mounted on 25 × 75 x 1.0 mm microscope glass slides, on a drop of ProLong Diamond Antifade Mountant (Thermo Fisher Scientific, #P36970).

Immunostaining on brain tissue

E18 (Fig. 7) or P0 (Fig. 5) mouse brains were fixed in 4% PFA in 1X PBS overnight at 4 °C, cryopreserved in 30% (g/vol) sucrose/0.05% sodium azide/100 mM glycine in 1X PBS, and embedded in Scigen Tissue-Plus O.C.T. Compound (Fisher Scientific, #23-730-571). 30 μm sections were obtained using a Leica CM1860 cryostat and mounted on slides. For CRE staining, slides were washed in 1X Tris-buffered saline (TBS) 3 times, 5 minutes each, and blocked for 1 h at RT in 10% donkey serum (Sigma-Aldrich Inc., #D9663)/0.3% Triton-X 100 in 1X TBS. Slides were incubated overnight with a guinea pig anti-CRE polyclonal antibody (Synaptic Systems, #257004, 1:100), a rabbit anti-DDX3X polyclonal antibody (Thermo Fisher Scientific, #A300-474A, 1:400), a rabbit anti-Fibrillarin polyclonal antibody (Abcam, #ab5821, 1:2000), a mouse monoclonal antibody raised against the 5.8S rRNA (Y10b clone, Abcam, #ab171119, 1:50), or a rat anti-CTIP2/BCL11B monoclonal antibody (Millipore, #MABE1045, 1:300) in incubation buffer (5% donkey serum/0.1% Triton-X 100 in 1X TBS). After washing, slides were incubated for 2 h at RT with a donkey anti-rabbit antibody conjugated with Alexa Fluor 647 (Thermo Fisher Scientific, #A-31573, 1:200) a donkey anti-mouse antibody conjugated with Alexa Fluor 488 (Thermo Fisher Scientific, #A-32787, 1:200), a donkey anti-guinea pig conjugated with Alexa Fluor 488 (Abcam, #ab150185, 1:200), a donkey anti-rat conjugated with Alexa Fluor 488 (Abcam, #ab150153, 1:200), and/or Hoechst dye (Thermo Fisher Scientific, #H3570, 1:1,000) in 0.1% Triton-X 100 in 1X TBS. Coverslips were mounted using Dapi-Fluoromount G (SouthernBiotech, #0100-20).

Confocal imaging

Images were acquired using a Leica TCS SP8 inverted microscope and Leica LAS X software. Excitation beams 561 and 633 nm or 405, 488, 561 or 633 nm were coupled, image size was set at 1024 ×1024 pixels, scan speed was selected at 600 Hz. 60 z-stacks of 300 nm step size (18 µm in total) were taken using a 20X air objective for dendrite analyses, and a 100X oil immersion objective for spine analyses. Z-stacks projections were prepared using the Fiji software version 2.9.0/1.53t⁹⁴. For nucleoli and nuclei imaging, as well as imaging of rRNA, 30 z-stacks of 300 nm step size (9 μm in total) were taken using a 100X oil immersion objective with a zoom 2.

Immunofluorescence analyses

For the nucleoli analyses (Fig. 5, Fig. S6A, B), nucleoli and nuclei volumes were reconstructed from the Z-stacks images using IMARIS 10.1.0. Size for each nucleolus, sum of nucleolar volume for each nucleus, size of the nuclei, and number of nucleoli in each nucleus were measured. For the rRNA analyses (Fig. 5), the percentage of rRNA in cytoplasm or nucleus was calculated as the value raw integrated density in the cytoplasm or nucleus over the total cell using Fiji. The mean rRNA immunofluorescence signal was measured as mean of the pixel gray values on binary images using Fiji. For the DDX3X expression analyses (Fig. 1C), the fluorescence was quantified as mean of the pixel gray values, after image segmentation, using Fiji. DDX3X expression levels were measured as ratio between the mean of mCherry⁺ neurons and the average of 5 neighboring mCherry^- within the same coverslip.

Retrograde viral tracing

5-week-old mice were anesthetized with a mixture of oxygen and isoflurane (5% induction and 1% maintenance) in an incubation chamber and placed in the stereotaxic apparatus for intracerebral injections in the reticular pontine nucleus (AP -4.00, ML 0.75, DV 4.50, based on the Allen Brain Atlas) of retrograde AAV-CAG-GFP viral particles (Addgene, #37825-AAVrg, 0.5 μl; rate of 0.1 μl/minute) using a 5 μl Hamilton microliter syringe. The needle was left in place for 5 minutes post-injection to ensure a proper diffusion of the virus into the tissue. Mice were perfused and brains collected for posterior analysis 3 weeks after the stereotaxic injection.

Whole brain clearing and imaging

Fixed adult mouse brains with retrograde viral tracing were delipidated with a modified Adipo-Clear protocol^95,96. Briefly, perfusion fixed brain samples were washed with B1n buffer (H2O/0.1% Triton X-100/0.3 M glycine, pH 7), then transferred to a methanol gradient series (20%, 40%, 60%, 80%) in B1n buffer, 4 mL for each brain, 1 h for each step; then 100% methanol for 1 h; then overnight incubation in 2:1 mixture of DCM:methanol and a 1.5 h incubation in 100% DCM the following day; then 100% methanol for 1 h three times, and reverse methanol gradient series (80%, 60%, 40%, 20%) in B1n buffer, 30 minutes for each step. Samples were then washed in B1n buffer for 1 h and overnight. The above procedures were done at RT with rocking to complete delipidation. The delipidated samples were then blocked in PTxwH buffer (PBS/0.1% Triton X100/0.05% Tween 20) with 5% DMSO and 0.3 M glycine for 3 h and overnight at 37 °C, then washed with PTxwH for 1 h, 2 h, and overnight at RT. For staining, brain samples were incubated in primary antibody (goat anti-GFP primary antibody, Rockland 600-101-215, 1:300) diluted in PTxwH for 2 weeks at 37 °C. After primary antibody incubation, samples were washed in PTxwH for 1 h, 2 h, 4 h, overnight, then 1 day, and then incubated in secondary antibody (Jackson ImmunoResearch 705-607-003, Alexa Fluor 647 donkey-anti-goat, 1:100) diluted in PTxwH at RT for 2 weeks. Samples were then washed in PTxwH for 1 h, 2 h, 4 h, overnight, and then 1 day. Samples were further fixed in 1% PFA at 4 °C overnight, washed in PTxwH at RT (1 h, 2 h, 4 h, overnight), then blocked in B1n at RT overnight, and washed in PTxwH at RT (1 h, 2 h, 4 h). Samples were then bleached in 0.3% H2O2 at 4 °C overnight, washed 3 times in 20 mM PB (16 mM Na2HPO4, 4 mM NaH2PO4 in H2O) at RT for 2 h. For further delipidation, samples were immersed in SBiP buffer four times (1 h, 2 h, 4 h, overnight, 1day X6). Next, brains were washed 2 times (2 h, 4 h) in 20 mM PB buffer and finally 2 times in PTS solution (25% 2,2’-thiodiethanol/10 mM PB) (2 h, overnight), then equilibrated with 75% histodenz buffer (Cosmo Bio USA AXS-1002424) with refractive index adjusted to 1.53 using 2,2’-thiodiethanol. Samples were stored at -20^oC until acquisition. The cleared brain samples were imaged horizontally with tiling using the LifeCanvas SmartSPIM lightsheet microscope. 647 nm lasers were used for the GFP viral labeling IHC imaging with the 3.6×/0.2 detection lens. Lightsheet illumination is focused with NA 0.2 lens, and axially scanned with electrically tunable lens coupled to the camera (Hamamatsu Orca Back-Thin Fusion) in slit mode. Camera exposure was set at fast mode (2 ms) with 16b imaging. The X/Y sampling rate was 1.866 μm and Z step at 2 μm.

Sholl analysis

mCherry⁺ neurons in Fig. 1 were reconstructed using the Simple Neurite Tracer (SNT) plugin of the Fiji software in a semi-automated fashion. Dendrites were classified in primary, secondary, tertiary, and quaternary based on their outgrowth from the soma and branching patterns. The length and diameter of each dendrite were also measured. Axonal projections were excluded from the analysis. Concentric rings 10 µm apart around the somata of SNT reconstructed neurons and radiating outward were defined on SNT reconstructed neurons, and the number of intersections between the dendritic arbors and the concentric rings was counted. mCherry⁺ neurons (Fig. 6) and in vivo labeled GFP⁺ cortical neurons (Fig. 2), were analyzed as segmented images, free of axonal projections, to run an automated Sholl analysis. GFP⁺ neurons were additionally analyzed with SNT to measure the length of apical trunks and tufts.

Dendritic spine analyses

Dendritic spines were annotated on mCherry+ neurons starting at 30 µm from the soma and on 10 µm-long segments on the dendrite. Spines were manually counted on an average of 16 10 µm-long segments per neuron, and spine count was then averaged per neuron. Dendritic spine density was measured as the total number of dendritic spines in 10 µm of dendritic length. For each spine, the head width (W) and neck length (L) was measured. Spines were classified into 6 morphological subtypes: branched (two-head spines), mushroom (W ≥ 0.7 µm; 1 < L > 3 µm), stubby (W ≤ 0.6 µm; no neck detected), thin (W ≤ 0.7 µm; L ≤ 1 µm), long thin (W ≤ 0.7 µm; 1 < L > 3 µm), and filopodia (no head detected; L ≥ 3 µm), based on previous literature^97,98.

Synaptosomes

Synaptosomes were isolated as described previously⁹⁸. P21 Sox2-Ddx3x^+/+ and Sox2-Ddx3x^+/- female mice were euthanized by cervical dislocation. Cortices were rapidly dissected on ice and homogenized in ice-cold homogenizing buffer (0.32 M sucrose, 1 mM EDTA, 1 mg/ml BSA, 5 mM HEPES pH=7.4) in a glass Teflon douncer, and centrifuged at 3000 × g for 10 minutes at 4 °C. The supernatants were recovered and centrifuged at 14,000 g for 12 minutes at 4 °C. The pellets containing synaptosomes were gently resuspended in Krebs-Ringer buffer (140 mM NaCl, 5 mM KCl, 5 mM glucose, 1 mM EDTA, 10 mM HEPES pH 7.4), subjected to a density gradient upon addition of Percoll Plus (Sigma-Aldrich) (final 45% v/v), and centrifuged at 18,800 × g for 2 minutes at 4 °C to enrich the synaptosomes at the surface of the gradient. The synaptosomes were recovered, washed in Krebs Ringer buffer, and then centrifuged at 18,800 × g for 30 seconds at 4 °C. The pellets containing purified synaptosomes were lysed as detailed in the proteomics section.

Sample preparation for liquid chromatography with tandem mass spectrometry (LC-MS/MS)

Cell or synaptosome pellets were suspended in 200 μl RIPA buffer contains protease and phosphatase inhibitor cocktail. Three bursts (5% amp, 1 second pulses) of sonication for 15 seconds each were used to lyse the cells/synaptosomes. Suspension was centrifuged at 20,500 g for 10 minutes at 4 °C, and 175 μl of the supernatant was aliquoted for chloroform:methanol:water protein precipitation using standard protocols. The dried protein pellet was resuspended in 10 μl 8 M urea containing 400 mM ammonium bicarbonate (ABC), reduced with dithiothreitol (DTT), alkylated with iodoacetamide, and dual enzymatic digestion with LysC and trypsin (carried out at 37 °C overnight or 7 h, respectively). Digestion was quenched by adding trifluoroacetic acid (to 0.5%) prior to peptide de-salting with C18 MiniSpin columns (The Nest Group). The effluents from the de-salting step were dried and re-dissolved in in MS loading buffer (2% acetonitrile, 0.2% trifluoroacetic acid). Protein concentrations (A260/A280) were measured with a Nanodrop 2000 UV-Vis Spectrophotometer (Thermo Fisher Scientific). Each sample was then further diluted with MS loading buffer to 0.06 µg/µl, with 300 ng (5 µl) injected for LC-MS/MS analysis. 1:10 dilution of 10X Pierce Retention Time Calibration Mixture (Thermo Fisher Scientific, #88321) was added to each sample to check for retention time variability during LFQ data analysis.

LC-MS/MS analyses

LC-MS/MS analysis was performed on a Thermo Scientific Q Exactive HF-X equipped with a Waters M-class UPLC system utilizing a binary solvent system (A: 100% water, 0.1% formic acid; B: 100% acetonitrile, 0.1% formic acid). Trapping was performed at 10 µl/min, 99% Buffer A for 3 minutes using a Waters nanoEase M/Z Symmetry C18 Trap Column (100 Å, 5 µm, 180 µm x 20 mm). Peptides were separated using a nanoEase M/Z Peptide BEH C18 Column (130 Å, 1.7 µm, 75 µm X 250 mm) (37 °C) by elution with linear gradients reaching 6% B at 2 minutes, 25% B at 175 minutes, 40% B at 195 minutes, and 90% B at 200 minutes. Column regeneration and up to three blank injections were carried out in between all sample injections. MS was acquired in profile mode over the 350-1,500 m/z range using 1 microscan, 60,000 resolution, AGC target of 3E6, and a maximum injection time of 100 ms. Data dependent MS/MS were acquired in centroid mode on the top 20 precursors per MS scan using 1 microscan, 30,000 resolution, AGC target of 1E5, maximum injection time of 100 ms, and an isolation window of 1.6 m/z. Precursors were fragmented by HCD activation with a collision energy of 28%. MS/MS were collected on species with an intensity threshold of 2.5E4, charge states 2-6, and peptide match preferred. Dynamic exclusion was set to 20 seconds. The LC-MS/MS data were processed with Progenesis QI software (Nonlinear Dynamics, version 4.2) with protein identification carried out using in-house Mascot search engine (2.4). The Progenesis QI software performs chromatographic/spectral alignment (one run is chosen as a reference for alignment of all other data files to), mass spectral peak picking and filtering (ion signal must satisfy the 3 times standard deviation of the noise), and quantitation of proteins and peptides. A normalization factor for each run was calculated to account for differences in sample load between injections as well as differences in ionization. The normalization factor was determined by calculating a quantitative abundance ratio between the reference run and the run being normalized, with the assumption being that most proteins/peptides are not changing in the experiment so the quantitative value should equal 1. The experimental design was setup to group multiple injections (technical and biological replicates) from each run into each comparison sets. The algorithm then calculates the tabulated raw and normalized abundances, ANOVA P-values for each feature in the data set. The MS/MS spectra was exported as mgf (Mascot generic files) for database searching. Mascot Distiller was used to generate peak lists, and the Mascot search algorithm was used for searching against the Swiss Protein database with taxonomy restricted to Mus musculus (17,174 sequences); and carbamidomethyl (Cys), oxidation of Met, Phospho (Ser, Thr, Tyr), acetylation (Lys and Protein N-term), and deamidation (Asn and Asp) were entered as variable modifications. Two missed tryptic cleavages were allowed, precursor mass tolerance was set to 10 ppm, and fragment mass tolerance was set to 0.02 Da. The significance threshold was set based on a False Discovery Rate (FDR) of 2%. The Mascot search results was exported as.xml files and then imported into the processed dataset in Progenesis QI software where peptides identified were synched with the corresponding quantified features and their corresponding abundances. Protein abundances (requiring at least 1 unique peptide with MOWSE score of >95% confidence) were then calculated from the sum of all non-conflicting peptide ion ID assignment for a specific protein on each run. A label free quantitation approach utilizing Progenesis QI (Nonlinear Dynamics)⁹⁹ was utilized to obtain quantitative information on peptides and proteins. Downstream analyses were restricted to proteins identified with at least two unique peptides. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE¹⁰⁰ partner repository with the dataset identifier PXD060033.

Puromycin-based assay

Details are described in the Supplemental Note.

Developmental milestones

Testing was performed as reported previously²⁴, using a battery of tests adapted from the Fox scale^101,102. Details are described in the Supplemental Note.

Gait analyses

Details are described in the Supplemental Note.

Adult motor testing

Testing started at 10 weeks ± one week. Behavioral testing was conducted during the light phase in sound-attenuated rooms. Mice were handled daily for one week prior to behavioral testing to assess general health, physical appearance, and spontaneous activity (Supplemental Note). Mice were habituated to the testing room for 30 minutes prior to testing. All surfaces and equipment were cleaned with Rescue Veterinary disinfectant between trials and nitrile gloves were used. The four genotypes were tested on the same day in randomized order by an experimenter blind to genotype. All data were scored and analyzed blind to the genotype. For accelerating rotarod, mice were tested on a motorized rod measuring 3 cm in diameter (Omnitech Electronics Inc) with gradual speed increase from 4 to 40 rpm over 5 minutes. Mice were presented with four 5 min-trials per day, 1 h apart, over three consecutive days. The latency to fall and speed was recorded. The balance beam test was conducted as described previously²⁴. Briefly, the beam consisted of a 70-cm long square prism with 1.3 cm face placed horizontally 40 cm above the bench. Each frame was 5-cm long. Mice were allowed to freely walk on the beam for two sessions (one training and one testing) of 2 minutes each, separated by 4 h. The number of slips and the distance covered during the testing session were analyzed. The vertical pole test was conducted as described previously²⁴. A wooden pole wrapped in tape to facilitate walking was used. The test consisted of two consecutive training days and a testing day (3 trials each day). Mice were placed facing upwards just below the top of the pole and the time to complete the turn and the time to climb down the pole were measured. The wire hanging test was conducted as described previously²⁴. Mice were led from the base of the tail and allowed to grasp a rod placed 40 cm above the bench by both forepaws. The task consisted of a 5 sec-training session followed by three 60 sec-testing sessions. The latency to fall and the distance traveled along the hanger were recorded, and the average of the testing sessions was used for analysis.

Statistical analyses

All statistics and plots were generated using the R packages stats and ggplot2. Statistical tests, number of animals in each experiment, and significance are indicated in each figure and/or corresponding legend. Data are shown as mean ± SEM, except for the plots showing empirical cumulative distribution function (ECDF). ECDF plots display data points from lowest to highest (X axis) against their percentiles (Y axis). Outliers were defined as data points below Q1-1.5xIQR or above Q3 + 1.5xIQR, where Q1 is the first quartile, Q3 is the third quartile, and the IQR is the interquartile range. Outliers, shown in plots with the ⊗ symbol, were removed from the calculations of the mean and SEM and for the statistical testing. Shapiro–Wilk test was used to assess normality and Bartlett’s test to assess homoscedasticity. One-way or two-way repeated measure ANOVA was used for repeated observations. In case of multiple comparisons, one-way, two-way, or three-way ANOVA test was employed for normally distributed data with equal variance, Welch ANOVA for normally distributed data with unequal variance, or Kruskal-Wallis test for non-parametric data. Groups comparisons were then conducted with Student’s t test for normally distributed data with equal variance, Welch Two Sample t test for normally distributed data with unequal variance, or Wilcoxon signed-rank test for not normally distributed data, with Benjamini-Hochberg correction for multiple comparisons. To compare empirical cumulative distribution function across groups, Kolmogorov–Smirnov test was used.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Source data are provided with this paper. The processed proteomic data generated in this study are available in Supplementary Data 1 and Supplementary Data 2. The corresponding raw proteomic data have been deposited in the ProteomeXchange under identifier PXD060033. Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Silvia De Rubeis (silvia.derubeis@mssm.edu). Source data are provided with this paper.

References

Deciphering Developmental Disorders, S. Large-scale discovery of novel genetic causes of developmental disorders. Nature 519, 223–228 (2015).
Article ADS Google Scholar
Tang, L. et al. Prospective and detailed behavioral phenotyping in DDX3X syndrome. Mol. Autism 12, 36 (2021).
Article PubMed PubMed Central Google Scholar
Lennox, A. L. et al. Pathogenic DDX3X Mutations Impair RNA metabolism and neurogenesis during fetal cortical development. Neuron 106, 404–420.e408 (2020).
Article CAS PubMed PubMed Central Google Scholar
Ruzzo, E. K. et al. Inherited and de novo genetic risk for autism impacts shared networks. Cell 178, 850–866.e826 (2019).
Article CAS PubMed PubMed Central Google Scholar
Snijders Blok, L. et al. Mutations in DDX3X are a common cause of unexplained intellectual disability with gender-specific effects on WNT signaling. Am. J. Hum. Genet. 97, 343–352 (2015).
Article CAS PubMed PubMed Central Google Scholar
Wang, X. et al. Phenotypic expansion in DDX3X - a common cause of intellectual disability in females. Ann. Clin. Transl. Neurol. 5, 1277–1285 (2018).
Article ADS PubMed PubMed Central Google Scholar
Ng-Cordell, E. et al. Social and emotional characteristics of girls and young women with DDX3X-associated intellectual disability: a descriptive and comparative study. J. Autism Dev. Disord. 53, 3208–3219 (2023).
Article PubMed Google Scholar
Kellaris, G. et al. A hypomorphic inherited pathogenic variant in DDX3X causes male intellectual disability with additional neurodevelopmental and neurodegenerative features. Hum. Genomics 12, 11 (2018).
Article PubMed PubMed Central Google Scholar
Nordahl, C. W. Why do we need sex-balanced studies of autism? Autism Res 16, 1662–1669 (2023).
Article PubMed PubMed Central Google Scholar
Soto-Rifo, R. et al. DEAD-box protein DDX3 associates with eIF4F to promote translation of selected mRNAs. EMBO J. 31, 3745–3756 (2012).
Article CAS PubMed PubMed Central Google Scholar
Calviello, L. et al. DDX3 depletion represses translation of mRNAs with complex 5’ UTRs. Nucleic Acids Res. 49, 5336–5350 (2021).
Article CAS PubMed PubMed Central Google Scholar
Hoye, M. L. et al. Aberrant cortical development is driven by impaired cell cycle and translational control in a DDX3X syndrome model. Elife 11, e78203 (2022).
Article CAS PubMed PubMed Central Google Scholar
Shih, J. W. et al. Critical roles of RNA helicase DDX3 and its interactions with eIF4E/PABP1 in stress granule assembly and stress response. Biochem J. 441, 119–129 (2012).
Article CAS PubMed Google Scholar
Fiorenzani, C., Mossa, A. & De Rubeis, S. DEAD/DEAH-box RNA helicases shape the risk of neurodevelopmental disorders. Trends Genet https://doi.org/10.1016/j.tig.2024.12.006 (2025).
Shen, H. et al. Sexually dimorphic RNA helicases DDX3X and DDX3Y differentially regulate RNA metabolism through phase separation. Mol. Cell 82, 2588–2603.e2589 (2022).
Article CAS PubMed PubMed Central Google Scholar
Hondele, M. et al. DEAD-box ATPases are global regulators of phase-separated organelles. Nature 573, 144–148 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Saito, M. et al. Acetylation of intrinsically disordered regions regulates phase separation. Nat. Chem. Biol. 15, 51–61 (2019).
Article ADS CAS PubMed Google Scholar
Yanas, A., Shweta, H., Owens, M. C., Liu, K. F. & Goldman, Y. E. RNA helicases DDX3X and DDX3Y form nanometer-scale RNA-protein clusters that support catalytic activity. Curr. Biol. 34, 5714–5727.e5716 (2024).
Article CAS PubMed Google Scholar
Owens, M. C. et al. Specific catalytically impaired DDX3X mutants form sexually dimorphic hollow condensates. Nat. Commun. 15, 9553 (2024).
Article CAS PubMed PubMed Central Google Scholar
Salamon, I. et al. Celf4 controls mRNA translation underlying synaptic development in the prenatal mammalian neocortex. Nat. Commun. 14, 6025 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Harnett, D. et al. A critical period of translational control during brain development at codon resolution. Nat. Struct. Mol. Biol. 29, 1277–1290 (2022).
Article CAS PubMed PubMed Central Google Scholar
Duffy, E. E. et al. Developmental dynamics of RNA translation in the human brain. Nat. Neurosci. 25, 1353–1365 (2022).
Article CAS PubMed PubMed Central Google Scholar
Borisova, E. et al. Protein translation rate determines neocortical neuron fate. Nat. Commun. 15, 4879 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Boitnott, A. et al. Developmental and behavioral phenotypes in a mouse model of DDX3X syndrome. Biol. Psychiatry 90, 742–755 (2021).
Article CAS PubMed PubMed Central Google Scholar
Patmore, D. M. et al. DDX3X suppresses the susceptibility of hindbrain lineages to medulloblastoma. Dev. Cell 54, 455–470 e455 (2020).
Article CAS PubMed PubMed Central Google Scholar
Chen, C. Y. et al. Targeted inactivation of murine Ddx3x: essential roles of Ddx3x in placentation and embryogenesis. Hum. Mol. Genet 25, 2905–2922 (2016).
CAS PubMed Google Scholar
Chen, H. H., Yu, H. I. & Tarn, W. Y. DDX3 modulates neurite development via translationally activating an RNA regulon involved in Rac1 activation. J. Neurosci. 36, 9792–9804 (2016).
Article CAS PubMed PubMed Central Google Scholar
Garieri, M. et al. Extensive cellular heterogeneity of X inactivation revealed by single-cell allele-specific expression in human fibroblasts. Proc. Natl Acad. Sci. USA 115, 13015–13020 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Tukiainen, T. et al. Landscape of X chromosome inactivation across human tissues. Nature 550, 244–248 (2017).
Article ADS PubMed PubMed Central Google Scholar
Oliva, M. et al. The impact of sex on gene expression across human tissues. Science 369, eaba3066 (2020).
Article CAS PubMed PubMed Central Google Scholar
Borensztein, M. et al. Contribution of epigenetic landscapes and transcription factors to X-chromosome reactivation in the inner cell mass. Nat. Commun. 8, 1297 (2017).
Article ADS PubMed PubMed Central Google Scholar
Bellott, D. W. et al. Mammalian Y chromosomes retain widely expressed dosage-sensitive regulators. Nature 508, 494–499 (2014).
Article ADS CAS PubMed PubMed Central Google Scholar
Arnold, A. P. X chromosome agents of sexual differentiation. Nat. Rev. Endocrinol. 18, 574–583 (2022).
Article CAS PubMed PubMed Central Google Scholar
Cortez, D. et al. Origins and functional evolution of Y chromosomes across mammals. Nature 508, 488–493 (2014).
Article ADS CAS PubMed Google Scholar
Lukin, J., Smith, C. M. & De Rubeis, S. Emerging X-linked genes associated with neurodevelopmental disorders in females. Curr. Opin. Neurobiol. 88, 102902 (2024).
Article CAS PubMed Google Scholar
Foresta, C., Ferlin, A. & Moro, E. Deletion and expression analysis of AZFa genes on the human Y chromosome revealed a major role for DBY in male infertility. Hum. Mol. Genet. 9, 1161–1169 (2000).
Article CAS PubMed Google Scholar
Dicke, A. K. et al. DDX3Y is likely the key spermatogenic factor in the AZFa region that contributes to human non-obstructive azoospermia. Commun. Biol. 6, 350 (2023).
Article CAS PubMed PubMed Central Google Scholar
Ramathal, C. et al. DDX3Y gene rescue of a Y chromosome AZFa deletion restores germ cell formation and transcriptional programs. Sci. Rep. 5, 15041 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Vogt, P. H., Rauschendorf, M. A., Zimmer, J., Drummer, C. & Behr, R. AZFa Y gene, DDX3Y, evolved novel testis transcript variants in primates with proximal 3 UTR polyadenylation for germ cell specific translation. Sci. Rep. 12, 8954 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Scala, M. et al. Three de novo DDX3X variants associated with distinctive brain developmental abnormalities and brain tumor in intellectually disabled females. Eur. J. Hum. Genet 27, 1254–1259 (2019).
Article PubMed PubMed Central Google Scholar
Ditton, H. J., Zimmer, J., Kamp, C., Rajpert-De Meyts, E. & Vogt, P. H. The AZFa gene DBY (DDX3Y) is widely transcribed but the protein is limited to the male germ cells by translation control. Hum. Mol. Genet 13, 2333–2341 (2004).
Article CAS PubMed Google Scholar
Chauvin, C. et al. Involvement of human release factors eRF3a and eRF3b in translation termination and regulation of the termination complex formation. Mol. Cell Biol. 25, 5801–5811 (2005).
Article CAS PubMed PubMed Central Google Scholar
Gong, C. et al. Sequential inverse dysregulation of the RNA helicases DDX3X and DDX3Y facilitates MYC-driven lymphomagenesis. Mol. Cell 81, 4059–4075.e4011 (2021).
Article CAS PubMed Google Scholar
Deuel, T. A. et al. Genetic interactions between doublecortin and doublecortin-like kinase in neuronal migration and axon outgrowth. Neuron 49, 41–53 (2006).
Article CAS PubMed Google Scholar
Sawada, M. et al. PlexinD1 signaling controls morphological changes and migration termination in newborn neurons. EMBO J. 37, e97404 (2018).
Article PubMed PubMed Central Google Scholar
Hor, C. H. H. & Goh, E. L. K. Rab23 regulates radial migration of projection neurons via N-cadherin. Cereb. Cortex 28, 1516–1531 (2018).
Article PubMed PubMed Central Google Scholar
Na, Y. et al. Fbxo45 Binds SPRY motifs in the extracellular domain of N-cadherin and regulates neuron migration during brain development. Mol. Cell Biol. 40, e00539–19 (2020).
Article CAS PubMed PubMed Central Google Scholar
Karczewski, K. J. et al. The mutational constraint spectrum quantified from variation in 141,456 humans. Nature 581, 434–443 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Thormann, A. et al. Flexible and scalable diagnostic filtering of genomic variants using G2P with Ensembl VEP. Nat. Commun. 10, 2373 (2019).
Article ADS PubMed PubMed Central Google Scholar
Fu, J. M. et al. Rare coding variation provides insight into the genetic architecture and phenotypic context of autism. Nat. Genet 54, 1320–1331 (2022).
Article CAS PubMed PubMed Central Google Scholar
Heyne, H. O. et al. De novo variants in neurodevelopmental disorders with epilepsy. Nat. Genet 50, 1048–1053 (2018).
Article CAS PubMed Google Scholar
Epi, K. C., Epilepsy Phenome/Genome, P. Ultra-rare genetic variation in common epilepsies: a case-control sequencing study. Lancet Neurol. 16, 135–143 (2017).
Article Google Scholar
Epi25 Collaborative. Electronic address, s. b. u. e. a. & Epi, C. Ultra-Rare genetic variation in the epilepsies: A whole-exome sequencing study of 17,606 Individuals. Am. J. Hum. Genet. 105, 267–282 (2019).
Mercaldo, V. et al. Altered striatal actin dynamics drives behavioral inflexibility in a mouse model of fragile X syndrome. Neuron 111, 1760–1775 e1768 (2023).
Article CAS PubMed Google Scholar
Koopmans, F. et al. SynGO: An evidence-based, expert-curated knowledge base for the synapse. Neuron 103, 217–234 e214 (2019).
Article CAS PubMed PubMed Central Google Scholar
Schmidt, E. K., Clavarino, G., Ceppi, M. & Pierre, P. SUnSET, a nonradioactive method to monitor protein synthesis. Nat. Methods 6, 275–277 (2009).
Article CAS PubMed Google Scholar
Hoeffer, C. A. et al. Inhibition of the interactions between eukaryotic initiation factors 4E and 4G impairs long-term associative memory consolidation but not reconsolidation. Proc. Natl Acad. Sci. USA 108, 3383–3388 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Fusco, C. M. et al. Neuronal ribosomes exhibit dynamic and context-dependent exchange of ribosomal proteins. Nat. Commun. 12, 6127 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Seo, S. S. et al. Excess ribosomal protein production unbalances translation in a model of Fragile X Syndrome. Nat. Commun. 13, 3236 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Scarnati, M. S., Kataria, R., Biswas, M. & Paradiso, K. G. Active presynaptic ribosomes in the mammalian brain, and altered transmitter release after protein synthesis inhibition. Elife 7, e36697 (2018).
Article PubMed PubMed Central Google Scholar
Akbalik, G. et al. Visualization of newly synthesized neuronal RNA in vitro and in vivo using click-chemistry. RNA Biol. 14, 20–28 (2017).
Article PubMed Google Scholar
Chau, K. F. et al. Downregulation of ribosome biogenesis during early forebrain development. Elife 7, e36998 (2018).
Article PubMed PubMed Central Google Scholar
Salehi, S. et al. Cytosolic Ptbp2 modulates axon growth in motoneurons through axonal localization and translation of Hnrnpr. Nat. Commun. 14, 4158 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Kondrashov, N. et al. Ribosome-mediated specificity in Hox mRNA translation and vertebrate tissue patterning. Cell 145, 383–397 (2011).
Article CAS PubMed PubMed Central Google Scholar
Metz, J. B. et al. High-throughput translational profiling with riboPLATE-seq. Sci. Rep. 12, 5718 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Calvino, F. R. et al. Symportin 1 chaperones 5S RNP assembly during ribosome biogenesis by occupying an essential rRNA-binding site. Nat. Commun. 6, 6510 (2015).
Article ADS CAS PubMed Google Scholar
Leidig, C. et al. 60S ribosome biogenesis requires rotation of the 5S ribonucleoprotein particle. Nat. Commun. 5, 3491 (2014).
Article ADS PubMed Google Scholar
Mensah, M. A. et al. Aberrant phase separation and nucleolar dysfunction in rare genetic diseases. Nature 614, 564–571 (2023).
ADS CAS PubMed PubMed Central Google Scholar
Valentin-Vega, Y. A. et al. Cancer-associated DDX3X mutations drive stress granule assembly and impair global translation. Sci. Rep. 6, 25996 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Toyama, Y. & Shimada, I. NMR characterization of RNA binding property of the DEAD-box RNA helicase DDX3X and its implications for helicase activity. Nat. Commun. 15, 3303 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Xiol, J. et al. RNA clamping by Vasa assembles a piRNA amplifier complex on transposon transcripts. Cell 157, 1698–1711 (2014).
Article CAS PubMed Google Scholar
Kast, R. J., Lanjewar, A. L., Smith, C. D. & Levitt, P. FOXP2 exhibits projection neuron class specific expression, but is not required for multiple aspects of cortical histogenesis. Elife 8, e42012 (2019).
Article PubMed PubMed Central Google Scholar
Gorski, J. A. et al. Cortical excitatory neurons and glia, but not GABAergic neurons, are produced in the Emx1-expressing lineage. J. Neurosci. 22, 6309–6314 (2002).
Article CAS PubMed PubMed Central Google Scholar
Levy, T. et al. DDX3X Syndrome: Summary of Findings and Recommendations for Evaluation and Care. Pediatr. Neurol. 138, 87–94 (2023).
Article PubMed Google Scholar
Rengarajan, S., Derks, J., Bellott, D. W., Slavov, N. & Page, D. C. Post-transcriptional cross- and auto-regulation buffer expression of the human RNA helicases DDX3X and DDX3Y. https://pubmed.ncbi.nlm.nih.gov/39794123/ (2024).
Sidhaye, J. et al. Integrated transcriptome and proteome analysis reveals posttranscriptional regulation of ribosomal genes in human brain organoids. Elife 12, e85135 (2023).
Article CAS PubMed PubMed Central Google Scholar
Anadolu, M. N. et al. Puromycin reveals a distinct conformation of neuronal ribosomes. Proc. Natl Acad. Sci. USA 121, e2306993121 (2024).
Article CAS PubMed PubMed Central Google Scholar
Kraushar, M. L. et al. Protein synthesis in the developing neocortex at near-atomic resolution reveals Ebp1-mediated neuronal proteostasis at the 60S tunnel exit. Mol. Cell 81, 304–322.e316 (2021).
Article CAS PubMed Google Scholar
De Rubeis, S. et al. Synaptic, transcriptional and chromatin genes disrupted in autism. Nature 515, 209–215 (2014).
Article PubMed PubMed Central Google Scholar
Clement, J. P. et al. Pathogenic SYNGAP1 mutations impair cognitive development by disrupting maturation of dendritic spine synapses. Cell 151, 709–723 (2012).
Article CAS PubMed PubMed Central Google Scholar
Cruz-Martin, A., Crespo, M. & Portera-Cailliau, C. Delayed stabilization of dendritic spines in fragile X mice. J. Neurosci. 30, 7793–7803 (2010).
Article CAS PubMed PubMed Central Google Scholar
Fazel Darbandi, S. et al. Enhancing WNT signaling restores cortical neuronal spine maturation and synaptogenesis in Tbr1 Mutants. Cell Rep. 31, 107495 (2020).
Article CAS PubMed Google Scholar
Spratt, P. W. E. et al. The autism-associated gene Scn2a contributes to dendritic excitability and synaptic function in the prefrontal cortex. Neuron 103, 673–685.e675 (2019).
Article CAS PubMed PubMed Central Google Scholar
Holtmaat, A. & Svoboda, K. Experience-dependent structural synaptic plasticity in the mammalian brain. Nat. Rev. Neurosci. 10, 647–658 (2009).
Article CAS PubMed Google Scholar
Richards, S. E. V. et al. Experience-dependent development of dendritic arbors in mouse visual cortex. J. Neurosci. 40, 6536–6556 (2020).
Article CAS PubMed PubMed Central Google Scholar
Jan, Y. N. & Jan, L. Y. Branching out: mechanisms of dendritic arborization. Nat. Rev. Neurosci. 11, 316–328 (2010).
Article CAS PubMed PubMed Central Google Scholar
Creson, T. K. et al. Re-expression of SynGAP protein in adulthood improves translatable measures of brain function and behavior. Elife 8, e46752 (2019).
Article PubMed PubMed Central Google Scholar
Robinson, L. et al. Morphological and functional reversal of phenotypes in a mouse model of rett syndrome. Brain 135, 2699–2710 (2012).
Article PubMed PubMed Central Google Scholar
Sztainberg, Y. et al. Reversal of phenotypes in MECP2 duplication mice using genetic rescue or antisense oligonucleotides. Nature 528, 123–126 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
Garg, S. K. et al. Systemic delivery of MeCP2 rescues behavioral and cellular deficits in female mouse models of Rett syndrome. J. Neurosci. 33, 13612–13620 (2013).
Article CAS PubMed PubMed Central Google Scholar
Mei, Y. et al. Adult restoration of Shank3 expression rescues selective autistic-like phenotypes. Nature 530, 481–484 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Hayashi, S., Lewis, P., Pevny, L. & McMahon, A. P. Efficient gene modulation in mouse epiblast using a Sox2Cre transgenic mouse strain. Mech. Dev. 119, S97–S101 (2002).
Article PubMed Google Scholar
Kaech, S. & Banker, G. Culturing hippocampal neurons. Nat. Protoc. 1, 2406–2415 (2006).
Article CAS PubMed Google Scholar
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
Article CAS PubMed Google Scholar
Chi, J., Crane, A., Wu, Z. & Cohen, P. Adipo-Clear: A tissue clearing method for three-dimensional imaging of adipose tissue. J. Vis. Exp., https://doi.org/10.3791/58271 (2018).
Hou, Y. et al. Topographical organization of mammillary neurogenesis and efferent projections in the mouse brain. Cell Rep. 34, 108712 (2021).
Article CAS PubMed Google Scholar
Risher, W. C., Ustunkaya, T., Singh Alvarado, J. & Eroglu, C. Rapid golgi analysis method for efficient and unbiased classification of dendritic spines. PLoS One 9, e107591 (2014).
Article ADS PubMed PubMed Central Google Scholar
De Rubeis, S. et al. CYFIP1 coordinates mRNA translation and cytoskeleton remodeling to ensure proper dendritic spine formation. Neuron 79, 1169–1182 (2013).
Article PubMed PubMed Central Google Scholar
Vidyadhara, D. J. et al. Dopamine transporter and synaptic vesicle sorting defects underlie auxilin-associated parkinson’s disease. Cell Rep. 42, 112231 (2023).
Article CAS PubMed PubMed Central Google Scholar
Perez-Riverol, Y. et al. The PRIDE database at 20 years: 2025 update. Nucleic Acids Res. 53, D543–D553 (2025).
Article PubMed Google Scholar
Fox, W. M. Reflex-ontogeny and behavioural development of the mouse. Anim. Behav. 13, 234–241 (1965).
Article CAS PubMed Google Scholar
Heyser, C. J. Assessment of developmental milestones in rodents. Curr. Protoc. Neurosci. Chapter 8, Unit 8 18, https://doi.org/10.1002/0471142301.ns0818s25 (2004).
Hempstead, E. https://BioRender.com/ndyh7n2 (2025).
Hempstead, E. https://BioRender.com/97593al (2025).

Download references

Acknowledgements

Research reported in this publication was supported by the Eunice Kennedy Shriver National Institute Of Child Health & Human Development of the National Institutes of Health under Award Number R21HD097561 (to SDR) and R01HD104609 (to SDR), and by the National Institute Of Mental Health of the National Institutes of Health under Award Number R01MH131537 (to SDR and ZW). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The work was also supported by the Beatrice and Samuel A. Seaver Foundation and a Distinguished Scholar Award from the Icahn School of Medicine at Mount Sinai to SDR. AM, LD, and JL were supported by fellowships from the Beatrice and Samuel A. Seaver Foundation. AM was also supported by the Friedman Brain Institute (Doft Family and Friedman Brain Institute Postdoctoral Innovator Award) and the 2023 Brain and Behavior Research Foundation Young Investigator Grant. LD was also supported by a traineeship from the NIDCR-Interdisciplinary Training in Systems and Developmental Biology and Birth Defects (NICHD, T32HD075735). ZA was supported by a Dean’s Undergraduate Research Fund Grant from New York University and Seaver Undergraduate Research Scholarship. JT is supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC 2033 – 390677874 – RESOLV and TA 167/6-3. We also thank the Keck MS & Proteomics Resource at Yale School of Medicine for providing the necessary mass spectrometers and the accompany biotechnology tools funded in part by the Yale School of Medicine and by the Office of The Director, National Institutes of Health (S10OD02365101A1, S10OD019967, and S10OD018034). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Microscopy was performed at the Microscopy and Advanced Bioimaging CoRE at the Icahn School of Medicine at Mount Sinai. We thank Dévina Ung for initial experimental setup, Kristi Niblo for technical support, and Olivia Pistone, Sylvia Maxwell and Mahmuda Hannan for support with data entry. The pCIneo-HA-DDX3X plasmids were kindly provided by Dr. Ricardo Soto-Rifo (Universidad de Chile). The pTWIST-CMV-FLAG-DDX3Y was kindly provided by Dr. Stephen Floor (University California San Francisco). This work is dedicated to the loving memory of Dr. Daniele Di Marino.

Author information

These authors contributed equally: Lauren Dierdorff, Jeronimo Lukin.

Authors and Affiliations

Seaver Autism Center for Research and Treatment, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, USA
Adele Mossa, Lauren Dierdorff, Jeronimo Lukin, Marta Garcia-Forn, Yeaji Park, Chiara Fiorenzani, Zeynep Akpinar & Silvia De Rubeis
Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, USA
Adele Mossa, Lauren Dierdorff, Jeronimo Lukin, Marta Garcia-Forn, Yeaji Park, Chiara Fiorenzani, Zeynep Akpinar & Silvia De Rubeis
The Mindich Child Health and Development Institute, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, USA
Adele Mossa, Lauren Dierdorff, Jeronimo Lukin, Marta Garcia-Forn, Yeaji Park, Chiara Fiorenzani, Zeynep Akpinar & Silvia De Rubeis
Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, USA
Adele Mossa, Lauren Dierdorff, Jeronimo Lukin, Marta Garcia-Forn, Yeaji Park, Chiara Fiorenzani, Zeynep Akpinar & Silvia De Rubeis
Alper Center for Neural Development and Regeneration Friedman Brain Institute, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, USA
Adele Mossa, Lauren Dierdorff, Jeronimo Lukin, Marta Garcia-Forn, Yeaji Park, Chiara Fiorenzani, Zeynep Akpinar & Silvia De Rubeis
Appel Alzheimer’s Disease Research Institute, Feil Family Brain and Mind Research Institute, Weill Cornell Medicine, New York, NY10021, USA
Wei Wang & Zhuhao Wu
Department Biochemistry of Neurodegenerative Diseases, Institute of Biochemistry and Pathobiochemistry, Ruhr University Bochum, Bochum, Germany
Fatemeh Mamashli, Janine Kamps & Jörg Tatzelt
Department of Biology, New York University, College of Arts and Science, New York, NY, 10003, USA
Zeynep Akpinar
Cluster of Excellence RESOLV, Bochum, Germany
Jörg Tatzelt

Authors

Adele Mossa
View author publications
Search author on:PubMed Google Scholar
Lauren Dierdorff
View author publications
Search author on:PubMed Google Scholar
Jeronimo Lukin
View author publications
Search author on:PubMed Google Scholar
Marta Garcia-Forn
View author publications
Search author on:PubMed Google Scholar
Wei Wang
View author publications
Search author on:PubMed Google Scholar
Fatemeh Mamashli
View author publications
Search author on:PubMed Google Scholar
Yeaji Park
View author publications
Search author on:PubMed Google Scholar
Chiara Fiorenzani
View author publications
Search author on:PubMed Google Scholar
Zeynep Akpinar
View author publications
Search author on:PubMed Google Scholar
Janine Kamps
View author publications
Search author on:PubMed Google Scholar
Jörg Tatzelt
View author publications
Search author on:PubMed Google Scholar
Zhuhao Wu
View author publications
Search author on:PubMed Google Scholar
Silvia De Rubeis
View author publications
Search author on:PubMed Google Scholar

Contributions

S.D.R., A.M., L.D., and J.L. conceptualize the study and designed the experiments. A.M., L.D., J.L., Y.P., C.F., Z.A., and M.G.F. conducted experiments and collected data. Z.W. and W.W. designed the whole-mount experiments and collected data. F.M., J.K., and J.T. provided input on biochemical experiments. S.D.R., A.M., L.D., and J.L. analyzed the data. S.D.R., A.M., L.D., and J.L. wrote the original draft and all the other authors edited and approved the manuscript.

Corresponding author

Correspondence to Silvia De Rubeis.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Description of Additional Supplementary Files

Dataset 1

Dataset 2

Dataset 3

Reporting Summary

Transparent Peer Review file

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Mossa, A., Dierdorff, L., Lukin, J. et al. Sex-specific perturbations of neuronal development caused by mutations in the autism risk gene DDX3X. Nat Commun 16, 4512 (2025). https://doi.org/10.1038/s41467-025-59680-8

Download citation

Received: 06 May 2024
Accepted: 29 April 2025
Published: 15 May 2025
DOI: https://doi.org/10.1038/s41467-025-59680-8

This article is cited by

Expanding the understanding of DDX3X-related neurodevelopmental disorder in males
- Silvia De Rubeis
European Journal of Human Genetics (2025)