Article

Arid1b haploinsufficiency disrupts cortical interneuron development and mouse behavior

  • Nature Neurosciencevolume 20pages16941707 (2017)
  • doi:10.1038/s41593-017-0013-0
  • Download Citation
Received:
Accepted:
Published:

Abstract

Haploinsufficiency of the AT-rich interactive domain 1B (ARID1B) gene causes autism spectrum disorder and intellectual disability; however, the neurobiological basis for this is unknown. Here we generated Arid1b-knockout mice and examined heterozygotes to model human patients. Arid1b-heterozygous mice showed a decreased number of cortical GABAergic interneurons and reduced proliferation of interneuron progenitors in the ganglionic eminence. Arid1b haploinsufficiency also led to an imbalance between excitatory and inhibitory synapses in the cerebral cortex. Furthermore, we found that Arid1b haploinsufficiency suppressed histone H3 lysine 9 acetylation (H3K9ac) overall and particularly reduced H3K9ac of the Pvalb promoter, resulting in decreased transcription. Arid1b-heterozygous mice exhibited abnormal cognitive and social behaviors, which were rescued by treatment with a positive allosteric GABAA receptor modulator. Our results demonstrate a critical role for Arid1b in interneuron development and behavior and provide insight into the pathogenesis of autism spectrum disorder and intellectual disability.

  • Subscribe to Nature Neuroscience for full access:

    $59

    Subscribe

Additional access options:

Already a subscriber?  Log in  now or  Register  for online access.

Additional information

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

References

  1. 1.

    Ellison, J. W., Rosenfeld, J. A. & Shaffer, L. G. Genetic basis of intellectual disability. Annu. Rev. Med. 64, 441–450 (2013).

  2. 2.

    Halgren, C. et al. Corpus callosum abnormalities, intellectual disability, speech impairment, and autism in patients with haploinsufficiency of ARID1B. Clin. Genet. 82, 248–255 (2012).

  3. 3.

    Santen, G. W. et al. Mutations in SWI/SNF chromatin remodeling complex gene ARID1B cause Coffin-Siris syndrome. Nat. Genet. 44, 379–380 (2012).

  4. 4.

    Hoyer, J. et al. Haploinsufficiency of ARID1B, a member of the SWI/SNF-a chromatin-remodeling complex, is a frequent cause of intellectual disability. Am. J. Hum. Genet. 90, 565–572 (2012).

  5. 5.

    Ronan, J. L., Wu, W. & Crabtree, G. R. From neural development to cognition: unexpected roles for chromatin. Nat. Rev. Genet. 14, 347–359 (2013).

  6. 6.

    López, A. J. & Wood, M. A. Role of nucleosome remodeling in neurodevelopmental and intellectual disability disorders. Front. Behav. Neurosci. 9, 100 (2015).

  7. 7.

    Marín, O. Interneuron dysfunction in psychiatric disorders. Nat. Rev. Neurosci. 13, 107–120 (2012).

  8. 8.

    Nelson, S. B. & Valakh, V. Excitatory/inhibitory balance and circuit homeostasis in autism spectrum disorders. Neuron 87, 684–698 (2015).

  9. 9.

    Ben-Ari, Y. The GABA excitatory/inhibitory developmental sequence: a personal journey. Neuroscience 279, 187–219 (2014).

  10. 10.

    Robertson, C. E., Ratai, E. M. & Kanwisher, N. Reduced GABAergic action in the autistic brain. Curr. Biol. 26, 80–85 (2016).

  11. 11.

    Han, S. et al. Autistic-like behaviour in Scn1a +/− mice and rescue by enhanced GABA-mediated neurotransmission. Nature 489, 385–390 (2012).

  12. 12.

    Moffat, J. J., Ka, M., Jung, E. M. & Kim, W. Y. Genes and brain malformations associated with abnormal neuron positioning. Mol. Brain 8, 72 (2015).

  13. 13.

    Skarnes, W. C. et al. A conditional knockout resource for the genome-wide study of mouse gene function. Nature 474, 337–342 (2011).

  14. 14.

    Stenman, J., Toresson, H. & Campbell, K. Identification of two distinct progenitor populations in the lateral ganglionic eminence: implications for striatal and olfactory bulb neurogenesis. J. Neurosci. 23, 167–174 (2003).

  15. 15.

    Xu, Q., Tam, M. & Anderson, S. A. Fate mapping Nkx2.1-lineage cells in the mouse telencephalon. J. Comp. Neurol. 506, 16–29 (2008).

  16. 16.

    Kim, D. I. et al. An improved smaller biotin ligase for BioID proximity labeling. Mol. Biol. Cell 27, 1188–1196 (2016).

  17. 17.

    Sommeijer, J. P. & Levelt, C. N. Synaptotagmin-2 is a reliable marker for parvalbumin positive inhibitory boutons in the mouse visual cortex. PLoS One 7, e35323 (2012).

  18. 18.

    Lucas, E. K. et al. PGC-1α provides a transcriptional framework for synchronous neurotransmitter release from parvalbumin-positive interneurons. J. Neurosci. 34, 14375–14387 (2014).

  19. 19.

    Cowell, R. M., Blake, K. R. & Russell, J. W. Localization of the transcriptional coactivator PGC-1alpha to GABAergic neurons during maturation of the rat brain. J. Comp. Neurol. 502, 1–18 (2007).

  20. 20.

    Tsurusaki, Y. et al. Mutations affecting components of the SWI/SNF complex cause Coffin-Siris syndrome. Nat. Genet. 44, 376–378 (2012).

  21. 21.

    Wonders, C. P. & Anderson, S. A. The origin and specification of cortical interneurons. Nat. Rev. Neurosci. 7, 687–696 (2006).

  22. 22.

    Ka, M., Chopra, D. A., Dravid, S. M. & Kim, W. Y. Essential roles for ARID1B in dendritic arborization and spine morphology of developing pyramidal neurons. J. Neurosci. 36, 2723–2742 (2016).

  23. 23.

    Zikopoulos, B. & Barbas, H. Altered neural connectivity in excitatory and inhibitory cortical circuits in autism. Front. Hum. Neurosci. 7, 609 (2013).

  24. 24.

    Stoner, R. et al. Patches of disorganization in the neocortex of children with autism. N. Engl. J. Med. 370, 1209–1219 (2014).

  25. 25.

    Reynolds, G. P., Zhang, Z. J. & Beasley, C. L. Neurochemical correlates of cortical GABAergic deficits in schizophrenia: selective losses of calcium binding protein immunoreactivity. Brain Res. Bull. 55, 579–584 (2001).

  26. 26.

    Lawrence, Y. A., Kemper, T. L., Bauman, M. L. & Blatt, G. J. Parvalbumin-, calbindin-, and calretinin-immunoreactive hippocampal interneuron density in autism. Acta Neurol. Scand. 121, 99–108 (2010).

  27. 27.

    Mariani, J. et al. FOXG1-dependent dysregulation of GABA/glutamate neuron differentiation in autism spectrum disorders. Cell 162, 375–390 (2015).

  28. 28.

    Gogolla, N. et al. Common circuit defect of excitatory-inhibitory balance in mouse models of autism. J. Neurodev. Disord. 1, 172–181 (2009).

  29. 29.

    Cellot, G. & Cherubini, E. GABAergic signaling as therapeutic target for autism spectrum disorders. Front Pediatr. 2, 70 (2014).

  30. 30.

    Wöhr, M. et al. Lack of parvalbumin in mice leads to behavioral deficits relevant to all human autism core symptoms and related neural morphofunctional abnormalities. Transl. Psychiatry 5, e525 (2015).

  31. 31.

    Saunders, J. A. et al. Knockout of NMDA receptors in parvalbumin interneurons recreates autism-like phenotypes. Autism Res. 6, 69–77 (2013).

  32. 32.

    Nagl, N. G. Jr., Wang, X., Patsialou, A., Van Scoy, M. & Moran, E. Distinct mammalian SWI/SNF chromatin remodeling complexes with opposing roles in cell-cycle control. EMBO J. 26, 752–763 (2007).

  33. 33.

    Chatterjee, N. et al. Histone H3 tail acetylation modulates ATP-dependent remodeling through multiple mechanisms. Nucleic Acids Res. 39, 8378–8391 (2011).

  34. 34.

    Naidu, S. R., Love, I. M., Imbalzano, A. N., Grossman, S. R. & Androphy, E. J. The SWI/SNF chromatin remodeling subunit BRG1 is a critical regulator of p53 necessary for proliferation of malignant cells. Oncogene 28, 2492–2501 (2009).

  35. 35.

    Vogel-Ciernia, A. et al. The neuron-specific chromatin regulatory subunit BAF53b is necessary for synaptic plasticity and memory. Nat. Neurosci. 16, 552–561 (2013).

  36. 36.

    Boyd, K., Woodbury-Smith, M. & Szatmari, P. Managing anxiety and depressive symptoms in adults with autism-spectrum disorders. J. Psychiatry Neurosci. 36, E35–E36 (2011).

  37. 37.

    Reid, K. A., Smiley, E. & Cooper, S. A. Prevalence and associations of anxiety disorders in adults with intellectual disabilities. J. Intellect. Disabil. Res. 55, 172–181 (2011).

  38. 38.

    Celen, C. et al. Arid1b haploinsufficient mice reveal neuropsychiatric phenotypes and reversible causes of growth impairment. eLife 6, e25730 (2017).

  39. 39.

    Dahlin, M. G., Amark, P. E. & Nergårdh, A. R. Reduction of seizures with low-dose clonazepam in children with epilepsy. Pediatr. Neurol. 28, 48–52 (2003).

  40. 40.

    Jung, E. M., Ka, M. & Kim, W. Y. Loss of GSK-3 causes abnormal astrogenesis and behavior in mice. Mol. Neurobiol. 53, 3954–3966 (2016).

  41. 41.

    Ka, M. & Kim, W. Y. Microtubule-actin crosslinking factor 1 is required for dendritic arborization and axon outgrowth in the developing brain. Mol. Neurobiol. 53, 6018–6032 (2016).

  42. 42.

    Ka, M., Kook, Y. H., Liao, K., Buch, S. & Kim, W. Y. Transactivation of TrkB by Sigma-1 receptor mediates cocaine-induced changes in dendritic spine density and morphology in hippocampal and cortical neurons. Cell Death Dis. 7, e2414 (2016).

  43. 43.

    Ka, M., Smith, A. L. & Kim, W. Y. MTOR controls genesis and autophagy of GABAergic interneurons during brain development. Autophagy 13, 1348–1363 (2017).

  44. 44.

    Ka, M., Condorelli, G., Woodgett, J. R. & Kim, W. Y. mTOR regulates brain morphogenesis by mediating GSK3 signaling. Development 141, 4076–4086 (2014).

  45. 45.

    Jung, E. M., An, B. S., Choi, K. C. & Jeung, E. B. Apoptosis- and endoplasmic reticulum stress-related genes were regulated by estrogen and progesterone in the uteri of calbindin-D(9k) and -D(28k) knockout mice. J. Cell. Biochem. 113, 194–203 (2012).

  46. 46.

    Ka, M., Jung, E. M., Mueller, U. & Kim, W. Y. MACF1 regulates the migration of pyramidal neurons via microtubule dynamics and GSK-3 signaling. Dev. Biol. 395, 4–18 (2014).

  47. 47.

    Ka, M., Moffat, J.J. & Kim, W.Y. MACF1 controls migration and positioning of cortical GABAergic interneurons in mice. Cereb. Cortex https://doi.org/10.1093/cercor/bhw319 (2016).

Download references

Acknowledgements

We thank S. Bonasera for helping with the behavioral analysis. Research reported in this publication was supported by an award from the National Institute of Neurological Disorders and Stroke of the National Institutes of Health under award number R01NS091220 and an Institutional Development Award (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under award number P20GM103471 to W.-Y.K.

Author information

Affiliations

  1. Department of Developmental Neuroscience, Munroe-Meyer Institute, University of Nebraska Medical Center, Omaha, Nebraska, USA

    • Eui-Man Jung
    • , Jeffrey Jay Moffat
    • , Channabasavaiah Basavaraju Gurumurthy
    •  & Woo-Yang Kim
  2. Department of Pharmacology, Creighton University, Omaha, Nebraska, USA

    • Jinxu Liu
    •  & Shashank Manohar Dravid

Authors

  1. Search for Eui-Man Jung in:

  2. Search for Jeffrey Jay Moffat in:

  3. Search for Jinxu Liu in:

  4. Search for Shashank Manohar Dravid in:

  5. Search for Channabasavaiah Basavaraju Gurumurthy in:

  6. Search for Woo-Yang Kim in:

Contributions

E.-M.J. and W.-Y.K. designed, performed and analyzed the experiments and wrote the paper. J.J.M., C.G. and J.L. performed the experiments. S.M.D. designed and analyzed the experiments. W.-Y.K. conceived and supervised the study.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to Woo-Yang Kim.

Integrated supplementary information

  1. Supplementary Figure 1 Generation of Arid1b +/– mice

    (a) A schematic drawing showing the strategy used to delete Arid1b in the mouse genome. Homologous recombination is marked with dotted lines. (b) Genotypes of Arid1b +/+, Arid1b +/- and Arid1b -/- mice were identified by PCR. (c) Immunofluorescence staining of Arid1b +/+, Arid1b +/- and Arid1b -/- mouse brains at E14.5. Arid1b is enriched in the nucleus, but was proportionately decreased in the mutant cells. DAPI was used as a counterstain. Scale bar= 4 μm. (d) RT-PCR analysis of Arid1b mRNAs in Arid1b +/+, Arid1b +/- and Arid1b -/- brains (top panel). Arid1b mRNAs were not detected in Arid1b -/- brains. Western blotting of ARID1B in mutant brains using an ARID1B antibody (bottom panel). (e) Quantification of d. Arid1b mRNA (F 2,51 = 421, ***P<0.0001 for Arid1b +/+ versus Arid1b +/-, ***P<0.0001 for Arid1b +/- versus Arid1b -/-, ### P<0.0001 for Arid1b +/- versus Arid1b -/-). ARID1B protein (F 2,51 = 654, ***P<0.0001 for Arid1b +/+ versus Arid1b +/-; ***P<0.0001 for Arid1b +/- versus Arid1b -/-; ### P<0.0001 for Arid1b +/- versus Arid1b -/-). N= 18 mice for each group. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test. (f) Body weights of Arid1b +/+ and Arid1b +/- mice were measured over a time course. P14 male (t 18= 3.045, **P= 0.0067). P21 male (t 18= 3.718, **P= 0.0016). P28 male (t 18= 3.209, **P= 0.0059). P28 female (t 18= 2.634, # P= 0.0232), P56 male (t 18= 2.654, *P= 0.0162), P63 male (t 18= 2.740, *P= 0.0135). N= 10 mice for each group. Statistical significance was determined between same age and sex mice by a two-tailed Student’s t test. Data shown are the mean ± SEM.

  2. Supplementary Figure 2 Numbers and distributions of cortical layer neurons and glia in Arid1b +/– cortices

    The numbers of cortical layer neurons, oligodendrocytes, and astrocytes were unaffected in Arid1b +/- cortices at P28 and P91. (a, c, e, g and i) Cortical sections from Arid1b +/+ and Arid1b +/- mice were immunostained with antibodies to the upper layer marker CUX1 (a), the lower layer marker TBR1 (c), the general neuron marker NeuN (e), the oligodendrocyte marker Olig2 (g) and the astrocyte marker GFAP (i). Scale bar= 100 μm. (b, d, f and h) Histograms show no changes in the numbers of the maker-stained cells in the cortex. N= 10 mice (8 males, 2 females) for Arid1b +/+ and 10 mice (7 males, 3 females) for Arid1b +/-. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. Data shown are the mean ± SEM.

  3. Supplementary Figure 3 Arid1b +/– mice have abnormal numbers of parvalbumin interneurons in the amygdala, thalamus, and hippocampus

    Arid1b +/- mice show reduced numbers of interneurons in the amygdala, thalamus, and hippocampus. (a) Parvalbumin-positive neurons were examined in the amygdala of Arid1b +/+and Arid1b +/- mice by immunostaining using an antibody to parvalbumin. The number of parvalbumin-positive cells was decreased in the Arid1b +/- amygdala at P28 (top panels) and P91 (bottom panels). Scale bar= 200 μm. (b) Quantification of a. P28: t 8= 3.045, ***P= 0.0006. P91: t 8= 3.686, **P= 0.0062. N= 5 mice (4 males, 1 females) for Arid1b +/+ and 5 mice (3 males, 2 females) for Arid1b +/-. Statistical significance was determined by a two-tailed Student’s t test. (c) Brain sections containing the thalamus of Arid1b +/+and Arid1b +/- mice were immunostained using an antibody to parvalbumin. Scale bar= 200 μm. The number of parvalbumin-positive cells was decreased in the Arid1b +/- thalamus at P28 (top panels) and P91 (bottom panels). (d) Quantification of c. P28: t 8= 4.419, **P= 0.0022. P91: t 8= 7.576, ***P<0.0001). N= 5 mice (4 males, 1 female) for Arid1b +/+ and 5 mice (3 males, 2 females) for Arid1b +/-. Statistical significance was determined by a two-tailed Student’s t test. (e) Brain sections containing the CA1 region from P28 and P91 Arid1b +/+and Arid1b +/- mice were immunostained using a parvalbumin antibody. Scale bar= 200 μm. (f) Quantification of parvalbumin-positive cells in CA1, CA2, CA3 and DG regions of P28 hippocampi (top panels). CA1: t 8= 6.758, ***P= 0.0001. CA2: t 8= 8.636, ***P<0.0001. CA3: t 8= 5.213, ***P= 0.0008. DG: t 8= 2.747, *P= 0.0252. P91 hippocampi (bottom panels). CA1: t 14= 2.544, *P= 0.0245. CA2: t 14= 3.469, **P= 0.0042. CA3: t 14= 2.613, *P= 0.0215. DG: t 14= 2.884, *P= 0.0128. P28 samples: N= 5 mice (3 males, 2 females) for Arid1b +/+ and 5 mice (3 males, 2 females) for Arid1b +/-. P91 samples: N= 8 mice for (4 males, 4 females) for Arid1b +/+ and 8 mice (5 males, 3 females) for Arid1b +/-. Statistical significance was determined by a two-tailed Student’s t test. Data shown are the mean ± SEM.

  4. Supplementary Figure 4 Somatostatin+, Calbindin+ and calretinin+ interneurons in the Arid1b +/– cortex

    The numbers of somatostatin-, calbindin- and calretinin-positive interneurons were not changed in the cortex of Arid1b +/- mice compared with Arid1b +/+ mice at P28. (a, c and e) Cortical sections were immunolabeled with somatostatin (a), calbindin (c) and calretinin (e) antibodies. Scale bar= 100 μm. (b, d and f) Quantification of a, c and e. Somatostatin staining: N= 7 mice (4 males, 3 females) for Arid1b +/+ and 7 mice (5 males, 2 females) Arid1b +/-. Calbindin staining: N= 5 mice (3 males, 2 females) for Arid1b +/+ and 5 mice (4 males, 1 female) for Arid1b +/-. Calretinin staining: N= 5 mice (4 males, 1 female) for Arid1b +/+ and 5 mice (3 males, 2 females) for Arid1b +/-. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. Data shown are the mean ± SEM.

  5. Supplementary Figure 5 Abnormal number of cortical interneurons in Arid1b conditional knockout mice

    The number of interneurons was decreased in Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP and Arid1b loxP/+ ; Nkx2.1-Cre cortices. (a) A schematic diagram of generating conditional Arid1b mice. (b) Cortical sections of P28 control (Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP) and Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice were immunostained using GABA and parvalbumin antibodies. Scale bar= 100 μm. (c and d) Quantification of b. GABA staining: total: t 14= 6.773, ***P<0.0001; layer II-III: t 14= 3.360, **P= 0.0047; layer IV: t 14= 5.451, ***P<0.0001; layer V: t 14= 10.75, ***P<0.0001. Parvalbumin staining: total: t 14= 11.78, ***P<0.0001; layer IV: t 14= 7.590, ***P<0.0001; layer V: t 14= 3.679, **P= 0.0025; layer VI: t 14= 6.548, ***P<0.0001. N= 8 mice (5 males, 3 females) for Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and 8 mice (6 males, 2 females) for Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP. Statistical significance was determined by a two-tailed Student’s t test. (e) Cortical sections of P28 control (Arid1b +/+ ; Nkx2.1-Cre) and Arid1b loxP/+ ; Nkx2.1-Cre mice were immunostained using GABA amd parvalbumin antibodies. Scale bar= 100 μm. (f and g) Quantification of e. GABA staining: total: t 12= 4.964, ***P=0.0003; layer II-III: t 12= 2.689, *P= 0.0197). Parvalbumin staining: total: t 12= 8.073, ***P<0.0001; layer IV: t 12= 4.223, **P= 0.0012; layer V: t 12= 2.4979, *P= 0.0281; layer VI: t 12= 2.701, *P= 0.0193. N= 7 mice (4 males, 3 females) for Arid1b +/+ ; Nkx2.1-Cre and 7 mice (4 males, 3 females) for Arid1b loxP/+ ; Nkx2.1-Cre. Statistical significance was determined by a two-tailed Student’s t test. (h) Body weight comparisons for control and Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP and Arid1b loxP/+ ; Nkx2.1-Cre mice. N= 7 mice (4 males, 3 females) for each group. NS: no significance. Data shown are the mean ± SEM.

  6. Supplementary Figure 6 The distribution of cortical interneurons in Arid1b +/– embryonic brains

    (a) Interneurons expressing EGFP were examined in E14.5 and E18.5 cortices from Arid1b +/+ ;Dlx5/6-cre-IRES-EGFP and Arid1b +/- ;Dlx5/6-cre-IRES-EGFP mice. Scale bar= 100 μm. CP, cortical plate; IZ, intermediate zone; VZ, ventricular zone. (b) Quantification of a. There was a reduction in the number of cortical interneurons in Arid1b +/- ;Dlx5/6-cre-IRES-EGFP mice when compared with Arid1b +/+ ;Dlx5/6-cre-IRES-EGFP mice. Histograms show the numbers of GFP-positive cells per 0.1 mm2. E14.5: t 8= 6.205, ***P= 0.0003. E18.5: t 8= 7.846, ***P<0.0001. N= 5 mice for each genotype. Statistical significance was determined by a two-tailed Student’s t test. Data shown are the mean ± SEM.

  7. Supplementary Figure 7 Altered gene expression in embryonic Arid1b knockout mice

    (a) Western blotting showed that the levels of β-catenin, Cyclin D1, H3K9ac, and H3K4me3 were decreased in a gene dose-dependent manner in Arid1b knockout mice at E15.5. The levels of phospho-β-catenin and H3K27me3 were increased in Arid1b knockout mice. The levels of CBP and PCAF showed no changes in Arid1b +/- cortices, but the PCAF level was decreased in Arid1b -/- samples. (b) Quantification of a.  N= 5 independent experiments using 5 mice for each genotype. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test. * p<0.05, ** p<0.01, *** p<0.001 vs. Arid1b +/+ mice. # p<0.05, ## p<0.01, ### p<0.001 vs. Arid1b +/- mice. (c) The mRNA levels of β-catenin target genes were assessed using real-time PCR in cortical tissues of E12.5 Arid1b +/+ and Arid1b +/- mice. N= 9 independent experiments using 9 mice for each genotype. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test. * p<0.05, ** p<0.01, *** p<0.001 vs. Arid1b +/+ mice. ### p<0.001 vs. Arid1b +/- mice. (d) The mRNA levels of Wnt family genes were assessed using real-time PCR in cortical tissues of E15.5 Arid1b +/+ and Arid1b +/- mice. The level of Gapdh mRNA was used for normalization. Data were shown as relative changes in Arid1b +/- versus Arid1b +/+ control samples. N= 9 independent experiments using 9 mice for each genotype. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test.* p<0.05, ** p<0.01 *** p<0.001 vs. Arid1b +/+ mice. (e) The mRNA levels of synaptic molecules, Gad1, Gad2, Gphn, Arhgef9, Slc17a7, Dlg4, Gabra1, Gabrb2 and Gabrg2 were assessed using real-time PCR in cortical tissues of E15.5 Arid1b knockout mice. The level of Gapdh mRNA was used for normalization. Data were shown as relative changes in Arid1b +/- or Arid1 b -/- versus Arid1b +/+ control samples. N= 6 independent experiments using 6 mice for each genotype. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test. * p<0.05, ** p<0.01, *** p<0.001 vs. Arid1b +/+ mice. # p<0.05, ## p<0.01, ### p<0.001 vs. Arid1b +/- mice. Data shown are the mean ± SEM.

  8. Supplementary Figure 8 Neurite development in Arid1b +/– interneurons

    Arid1b +/- interneurons show shorter neurites and less branches than Arid1b +/+ cells. (a) Interneuron neurite morphology was assessed in Arid1b +/+ ;Dlx5/6-Cre-IRES-EGFP and Arid1b +/- ;Dlx5/6-Cre-IRES-EGFP cortices at P91. Scale bar= 20 μm. (b) Representative traces of EGFP-expressing interneurons. (c) Quantification of the length and number of neurites. Neurite length: N= 23 neurites from 5 mice (3 males and 2 females) for each group. Neurite number: N= 10 cells from 5 mice for each group. Statistical significance was determined by a two-tailed Student’s t test. Neurite length: t 44= 3.208, **P= 0.0025. Primary neurite number: t 18= 4.569, ***P= 0.0002. Secondary neurite number: t 18= 4.358, ***P= 0.0004. Data shown are the mean ± SEM.

  9. Supplementary Figure 9 Number of excitatory synapses in Arid1b +/– mice

    (a) Cortical sections from Arid1b +/+ and Arid1b +/- mice were immunostained using antibodies to the excitatory presynaptic marker VGluT1 and the excitatory postsynaptic marker PSD95. Scale bar= 5 μm. Higher magnification images of white-boxed regions (merge panels) were shown in right panels. Scale bar= 0.5 μm. (b) Quantification of a.  N= 10 mice (5 males, 5 females) for each genotype. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. Data shown are the mean ± SEM.

  10. Supplementary Figure 10 Representative traces of postsynaptic currents in Arid1b +/– cortical slices

    (a) Representative whole-cell voltage-clamp recording showing mEPSCs as downward deflections in Arid1b +/+ and Arid1b +/- pyramidal neurons from cortical slices. N= 10 cells from 3 independent experiments using 3 mice for each condition. (b) Representative whole-cell voltage-clamp recording showing mIPSCs as upward deflections in Arid1b +/+ and Arid1b +/- neurons from cortical slices. N= 10 cells from 3 independent experiments using 3 mice for each condition.

  11. Supplementary Figure 11 ARID1B association with regulatory components of neural genes.

    (a) A schematic drawing of the Pvalb gene regions examined by ChIP-qPCR. (b) The ChIP-qPCR assay shows no binding of ARID1B to the Pvalb distal promoter (-1889 ~ -1645) and intron IV (+10265 ~ +10444) regions. After immunoprecipitating ARID1B-bound genomic DNAs from control and Arid1b +/- cortical lysates, the Pvalb distal promoter or intron IV regions shown in panel a were amplified by qPCR using appropriate primers. The sequences of the primers were described in Supplementary Table 4. (c) The ChIP-qPCR assay shows no changes in the strength of ARID1B binding to Slc17a7 or Dlg4 promoter regions. (d) Quantification of the bands shown in c. N= 3 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. (e) The levels of ARID1B binding to Gad1 or Slc32a1 promoters were decreased in Arid1b +/- cortices, compared with control levels. The ChIP-qPCR assay was performed using control and Arid1b +/- cortices. (f) Quantification of e. Gad1: t 4= 3.494, *P= 0.0025. Viaat: t 4= 3.499, *P= 0.0029. N= 3 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. (g) A schematic drawing of Ntrk2 gene promoter regions. (h) The ChIP-qPCR assay showed that the association of RNA polymerase II phospho-CTD (Ser5) with the Ntrk2 promoter was decreased in Arid1b +/- cortices compared with Arid1b +/+ samples. (i) Quantification of h. Region 1: t 4= 14.74, ***P<0.0001. Region 2: t 4= 4.041, *P= 0.0156. N= 3 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. Data shown are the mean ± SEM.

  12. Supplementary Figure 12 H3K9ac and phospho-CTD signatures in Syt2 and Ppargc1a promoters

    (a) The real-time PCR experiment shows no significant change in the levels of Syt2 and Ppargc1a mRNAs in P91 Arid1b +/+ and Arid1b +/- cortices. The level of Gapdh mRNA was used for normalization. Data were shown as relative changes in Arid1b +/- versus Arid1b +/+ control samples. N= 6 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. (b) A schematic drawing of the Syt2 gene regions examined by ChIP-qPCR. (c) Fragmented genomic DNAs in control and Arid1b +/- cortical lysates were immunoprecipitated using an H3K9ac antibody. Subsequently, the qPCR assay was performed to assess the strength of H3K9ac binding to Syt2 promoter regions. The sequences of the primers were described in Supplementary Table 4. (d) Quantification of the bands shown in c. N= 3 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. (e) A schematic drawing of the Ppargc1a gene regions examined by ChIP-qPCR. (f) H3K9ac binding to Ppargc1a promoter regions was examined by the ChIP-qPCR assay. (g) Quantification of the bands shown in f. N= 3 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. (h) The ChIP-qPCR assay was performed to assess the strength of phospho-CTD (Ser5) binding to Syt2 promoter regions. (i) Quantification of the bands shown in h. N= 3 independent experiments using 5 mice for each condition. Statistical significance was determined by a two-tailed Student’s t test. NS: no significance. Data shown are the mean ± SEM.

  13. Supplementary Figure 13 Conditional Arid1b knockout mice show abnormal behavior

    (a) The novel object recognition test was performed in Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice. Time exploring familiar and novel objects was not different in Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice (t 10= 5.842, ***P= 0.0002). N= 6 mice (4 males, 2 females) for Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and 6 mice (5 males, 1 female) for Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP. Statistical significance was determined between times exploring familiar and novel objects for each genotype and condition by a two-tailed Student’s t test. NS: no significance. (b) Using the three-chamber paradigm, the sociability test was performed in Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice. Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice had no preference for either chamber. In the social novelty test, Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice spent less time with the novel mouse (stranger II) than with the familiar mouse (Stranger I). Sociability test: t 11= 4.810, ***P= 0.0004. Social novelty test: t 11= 2.777, *P= 0.0148. N= 6 male mice for Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and 7 male mice for Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP. Statistical significance was determined between times exploring empty chamber and stranger I, or times exploring stranger I and stranger II for each genotype and condition by a two-tailed Student’s t test. NS: no significance. (c) Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP and Arid1b loxP/+ ; Nkx2.1-Cre mice showed increased grooming time compared with wild type mice. Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP: t 8= 10.84, ***P<0.0004. Arid1b loxP/+ ; Nkx2.1-Cre: t 8= 2.989, *P= 0.0174. N= 5 mice (4 males, 1 female) for Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP, 5 mice (4 males, 2 females) for Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP, 5 mice (3 males, 2 females) for Arid1b +/+ ; Nkx2.1-Cre, and 5 mice (3 males, 2 females) for Arid1b loxP/+ ; Nkx2.1-Cre. Statistical significance was determined by a two-tailed Student’s t test. (d) Representative tracing of mouse movement in the open field test.  (e) Quantification of the open field test shown in d. The time spent in the center and the number of entries into the center were examined in the open field. Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP mice showed reductions in both assessments compared with Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP mice. Open field center time: t 20= 3.832, **P= 0.001. Open field center entry: t 20= 2.385, *P= 0.0271). N= 11 mice (6 males, 5 females) for Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and 11 mice (7 males, 4 females) for Arid1b loxP/+ ; Dlx5/6-Cre-IRES-EGFP. Statistical significance was determined by a two-tailed Student’s t test. Data shown are the mean ± SEM.

  14. Supplementary Figure 14 Locomotor activities in Arid1b mutants, and clonazepam effects on postsynaptic currents in Arid1b mutant cortical slices

    (a-c) Quantification of the distance travelled by control and Arid1b mutant mice in the open field test. (a) wild-type control and Arid1b +/- mice were assessed (t 34= 5.885, ***P<0.0001). N= 16 mice (12 males, 4 females) for Arid1b +/+ and 20 mice (12 males, 8 females) for Arid1b +/-. Statistical significance was determined by a two-tailed Student’s t test. (b) Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and Arid1b +/- ; Dlx5/6-Cre-IRES-EGFP mice were assessed (t 20= 0.327, *P= 0.0296). N= 11 mice (6 males, 5 females) for Arid1b +/+ ; Dlx5/6-Cre-IRES-EGFP and 11 mice (7 males, 4 females) for Arid1b +/- ; Dlx5/6-Cre-IRES-EGFP. Statistical significance was determined by a two-tailed Student’s t test. (c) Arid1b mutant mice treated with clonazepam were assessed. F 3,28= 19.57, ***P<0.0001 for saline-treated Arid1b +/+ versus saline-treated Arid1b +/-; ### P<0. 0001 for saline-treated Arid1b +/- versus clonazepam-treated Arid1b +/-. N= 9 mice (6 males, 3 females) for saline-treated Arid1b +/+, 8 mice (5 males, 3 females) for clonazepam-treated Arid1b +/+, 9 mice (6 males, 3 females) for saline-treated Arid1b +/-, and 6 mice (6 males, 1 females) for clonazepam-treated Arid1b +/-. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test. NS: no significance. (d) A histogram showing the effect of clonazepam on mIPSC frequency and amplitude in Arid1b +/- cortical slices. Statistical significance was determined by one-way ANOVA with the Bonferonni correction test. F 3,36 = 5.427, *P<0.05 for vehicle control Arid1b +/+ versus vehicle control Arid1b +/-. N= 10 cells from 3 independent experiments using 3 mice for each condition. NS: no significance. Data shown are the mean ± SEM.

  15. Supplementary Figure 15 Full length Western blots

    Full length Western blots for cropped images in Fig. 3e, 5a, 5f, 5g, 5i, and Supplementary Fig. 1d and 7a.

Supplementary information