Gatad2b, associated with the neurodevelopmental syndrome GAND, plays a critical role in neurodevelopment and cortical patterning

Abstract

GATAD2B (GATA zinc finger domain containing 2B) variants are associated with the neurodevelopmental syndrome GAND, characterized by intellectual disability (ID), infantile hypotonia, apraxia of speech, epilepsy, macrocephaly and distinct facial features. GATAD2B encodes for a subunit of the Nucleosome Remodeling and Histone Deacetylase (NuRD) complex. NuRD controls transcriptional programs critical for proper neurodevelopment by coupling histone deacetylase with ATP-dependent chromatin remodeling activity. To study mechanisms of pathogenesis for GAND, we characterized a mouse model harboring an inactivating mutation in Gatad2b. Homozygous Gatad2b mutants die perinatally, while haploinsufficient Gatad2b mice exhibit behavioral abnormalities resembling the clinical features of GAND patients. We also observed abnormal cortical patterning, and cellular proportions and cell-specific alterations in the developmental transcriptome in these mice. scRNAseq of embryonic cortex indicated misexpression of genes key for corticogenesis and associated with neurodevelopmental syndromes such as Bcl11b, Nfia and H3f3b and Sox5. These data suggest a crucial role for Gatad2b in brain development.

Fig. 1. Developmental expression of GATAD2B and gene enrichment analysis suggest GATAD2B has a critical neurodevelopmental role.

A Expression level of human GATAD2B measured in RPKM (reads per kilobase of exon model per million mapped reads) from postmortem human brain specimens at indicated developmental stages profiled by RNA sequencing (RNA-Seq) and exon microarray hybridization. Data extracted from BrainSpan Atlas of the Developing Human Brain (http://brainspan.org.) (Allen Institute for Brain Science et al. 2023). DFC dorsolateral prefrontal cortex, VFC ventrolateral prefrontal cortex, MFC medial prefrontal cortex, OFC orbital prefrontal cortex, M1C: primary motor cortex, S1C: primary somatosensory cortex, IPC posteroinferior parietal cortex, A1C primary auditory cortex, STC superior temporal cortex, ITC inferior temporal cortex, V1C primary visual (occipital) cortex, HIP: hippocampus, AMY amygdaloid complex, STR Striatum, MD mediodorsal nucleus of thalamus, CBC cerebellar cortex, pcw postconception weeks. White rectangles: no data. B Expression of mouse Gatad2b by qRT-PCR at indicated time points during development. C DisGeNET enrichment analysis. Phenotypic features reported as present in GAND are marked with a plus sign, question marks denote uncertainty.

Fig. 2. Gatad2b^stop/+ mice are haploinsufficient.

A Expression of Gatad2b mRNA as determined by qRT-PCR in brains of 1day old mice (n = 4 of each genotype, *p < 0.05). B Western blot for the detection of Gatad2b in cerebellum samples from WT (Gatad2b^+/+), heterozygous (Gatad2b^stop/+) and homozygous (Gatad2b^stop/stop) mice. C Immunostaining for Gatad2b in 40 μm sections derived from Gatad2b^+/+ E16.5 brain. Scale bar: 60 μm. Boxes D and E depict the CP and VZ/SVZ areas shown in D’, E”’, where immunostaining for Gatad2b (green) and Ki67 (red) is shown for Gatad2b^+/+ (D’, E’), Gatad2b^stop/+ (D”, E”) and Gatad2b^stop/stop (D”’, E”’) E 16.5 brain samples. DAPI (blue) was used as DNA marker to identify nuclei and showed merged with Gatad2b (green) and KI67 (red). Scale bar: 15 μm MZ marginal zone, CP cortical plate, SP subplate, IZ intermediate zone, VZ/SVZ ventricular zone and subventricular zone.

Fig. 3. Gatad2b haploinsufficient mice exhibit behavioral abnormalities.

A Spatial learning and memory was assessed by the Barnes maze task at 3 months of age. The time taken to identify the target hole (primary latency) is shown in the left panel for both the training phase as well as for the probe test day in which the escape box was removed. The distribution of time during the probe test, performed 1 day after the last training session, is shown in the right panel. Two-way repeated measures ANOVA showed main effects of day (F3,20 = 46.43, P < 0.0001) and genotype (F1,20 = 8.3, P = 0.0148), indicating that Gatad2b^stop/+ mice performed worse than the wild type at finding the target hole across the acquisition phase. In the probe test, Gatad2b^stop/+ exhibited differences in exploratory activity. *p < 0.05, post hoc comparisons using Tukey test. B Learning and memory of contextual fear was assessed by fear conditioning. Left: Contextual test: freezing responses on the contextual testing 24 h after conditioning. There was significantly less freezing in Gatad2b^stop/+ mice during contextual testing (**p < 0.01, unpaired Student’s t-test). Right: Cued test: freezing responses on the cued testing. There was significant difference between WT and Gatad2b^stop/+ mice in cued conditioning during pre-tone and tone (*p < 0.05, unpaired Student’s t-test). C Activity levels were measured in the open field test. Left: analysis of the ratio of distance traveled in the center of the open field to total distance traveled suggest abnormal anxiety-related responses in Gatad2b^stop/+ mice (*<p0.05, unpaired Student’s t-test). Right: total distance traveled showed novel environment-induced hyperactivity in Gatad2b^stop/+ mice (*p < 0.05, unpaired Student’s t-test). N = 9 male mice per genotype. Bars represent mean ± S.E.M.

Fig. 4. Gatad2b deficiency disrupts cortical neuronal development.

A Immunostaining of coronal sections of E16.5 brains (cortex) with antibodies against Satb2 (green), and Ctip2 (red). Scale bar, 60 μm. Quantification data obtained with ImageJ (n = 3 of each genotype) for the proportion of double Satb2+ Ctip2+ neurons is shown to the right of the image B Coronal sections from E18.5 cortices immunolabeled for Satb2 (green), Ctip2 (red) and Tbr1 (blue). Scale bar: 100 μm. The relative proportion of Tbr1+ cells (n = 3 of each genotype) in layer V is depicted on the right. *P < 0.05. Data are presented as the mean ± S.E.M. *p ≤ 0.05.

Fig. 5. Disorganization of midline structures in Gatad2b^stop/+ mice.

A Immunostaining of brain coronal sections just rostral to the fornix and the anterior commissure (which were not overtly affected in the Gatad2b^stop/+ brains by bright field microscopy), from P0 mice from the indicated genotypes for 2H3 (red, stains the pioneer axons of the cingulate cortex (CIC)) and NeuN (green). DAPI (blue) was used to identify nuclei. Most Gatad2b^stop/stop axons fail to reach the contralateral hemisphere, while significantly less Gatad2b^stop/+ axons cross the midline. Scale bar: 100 µm. B Area covered by 2H3 projections, quantified for each genotype (n = 3 Gatad2b^+/+, n = 2 Gatad2b^stop/+, n = 2 Gatad2b^stop/stop, all from the same litter). *P < 0.05 (Student’s t-test). Data are presented as the mean ± S.E.M. C Immunostaining for glial cells with antibodies for GFAP (teal) in E18 embryos shows less glial cells and disorganization of the indisium griseum (IG), glial wedge (GW) and midline zipper glia (MZG). Labeling of neuronal cytoskeleton with TuJ1 (green) and NeuN (red) confirms the reduction of midline crossing axons. i-iii are enhanced magnifications of the indicated areas. Scale bars: 90 µm (left panels) and 30 µm (right panels). Images are from single embryos of either sex (siblings) but representative of an additional set of single embryos of each genotype derived from another breeding pair.

Fig. 6. Cell clusters and cell identities of Gatad2b mutant mice.

A Uniform manifold approximation and projection (UMAP) plot of single cells from 3 Gatad2b^+/+, 4 Gatad2b^stop/+ and 3 Gatad2b^stop/stop embryos, with each cell colored according to cluster. B UMAP of the integrated scRNA-seq data colored for the visualization of Gatad2b expression. C UMAP visualization of scRNA-seq data from individual genotypes.

Fig. 7. Alterations of cell type proportions in cortex of Gatad2b mutant mice.

Plot of proportion of cells assigned to each cell type in Gatad2b^+/+ (WT), Gatad2b^stop/stop (het) and Gatad2b^stop/stop (ko) E16.5 developing cortex. cortical layers II-IV (Layet II-IV), interneurons (IN), cortical layers II-IV (Layer V-VI), ganglionic eminence (GE), striatal inhibitory (Str. Inh), radial glia (RG), subventricular zone (SVZ), migratory SVZ (SVZ mig), cortical layer I (Layer I), endothelial (endo) and microglia (micro) are presented as across all genotypes (% of total), and per genotype (Gatad2b^+/+: WT (avg%), Gatad2b^stop/+: Het (avg%), Gatad2b^stop/stop: KO (avg%)). Bolded values (last two columns) highlight significant p values (p < 0.05) in both comparisons, Gatad2b^+/+ vs. Gatad2b^stop/+: p value (Het vs. WT), and Gatad2b^+/+ vs. Gatad2b^stop/stop: p value (KO vs. WT).