R

Fc

HMa

9b

c

Ud

e

f

h

•••

a

ARRAA

KNSCCFK

1

saioc

S

h0

Neuroscience Letters 612 (2016) 18–24

Contents lists available at ScienceDirect

Neuroscience Letters

jo ur nal ho me p age: www.elsev ier .com/ locate /neule t

esearch paper

ilamin A interacting protein plays a role in proper positioning ofallosal projection neurons in the cortex

ideshi Yagi a,b, Yuichiro Oka a,c,d, Munekazu Komada a, Min-Jue Xie a,e, Koichi Noguchi f,akoto Sato a,c,d,e,∗

Division of Cell Biology and Neuroscience, Department of Morphological and Physiological Sciences, Faculty of Medical Sciences, University of Fukui, Fukui10-1193, JapanDepartment of Anatomy and Cell Biology, Hyogo College of Medicine, Hyogo 663–8501, JapanUnited Graduate School of Child Development, Osaka University, Kanazawa University, Hamamatsu University School of Medicine, Chiba University andniversity of Fukui, Osaka 565–0871, JapanDepartment of Anatomy and Neuroscience, Graduate School of Medicine, Osaka University, Osaka 565–0871, JapanResearch Center for Child Mental Development, University of Fukui, Fukui 910-1193, JapanDepartment of Anatomy and Neuroscience, Hyogo College of Medicine, Hyogo 663–8501, Japan

i g h l i g h t s

FILIP is involved in neuronal cell positioning.The localization of callosal neurons was disturbed by insufficiency of FILIP.Positioning of Plxnd1-expressing callosal neurons was altered in Filip-knockout mice.

r t i c l e i n f o

rticle history:eceived 13 August 2015eceived in revised form 7 November 2015ccepted 28 November 2015vailable online 2 December 2015

a b s t r a c t

The callosal connections between the two hemispheres of the neocortex are altered in certain psychi-atric disorders including schizophrenia. However, how and why the callosal connection is impaired inpatients suffering from psychiatric diseases remain unclear. Filamin A interacting protein (FILIP), whosealteration through mutation relates to schizophrenic pathogenesis, binds to actin-binding proteins andcontrols neurotransmission. Because cortical excitatory neurons, including callosal projection neurons,migrate to the cortical plate during development, with the actin-binding proteins playing crucial roles

eywords:euronal migrationchizophreniaallosal projection neuronortical layer formationILIP

during migration, we evaluated whether FILIP is involved in the development of the callosal projectionneurons by histological analysis of Filip-knockout mice. The positioning of the callosal projection neurons,especially those expressing Plxnd1, in the superficial layer of the cortex is disturbed in these mice, whichsuggests that FILIP is a key molecule that links callosal projections to the pathogenesis of brain disorders.

© 2015 Elsevier Ireland Ltd. All rights reserved.

nockout mouse

. Introduction

The callosal projection neurons connect the two cerebral hemi-pheres via the corpus callosum [1] and play a role in associativend cognitive functions [2,3]. Callosal connections are decreased

n the brains of patients with schizophrenia [4–6]. Developmentf the callosal projection neurons in the cerebral cortex requiresomplex and precise control of neuronal cell migration [7]. Excita-

∗ Corresponding author at: Department of Anatomy and Neuroscience, Graduatechool of Medicine, Osaka University, Osaka 565–0871, Japan. Fax: +81 6 6879 3229

E-mail address: makosato@anat2.med.osaka-u.ac.jp (M. Sato).

ttp://dx.doi.org/10.1016/j.neulet.2015.11.049304-3940/© 2015 Elsevier Ireland Ltd. All rights reserved.

tory neurons, including callosal projection neurons, are producedfrom neuronal progenitor cells in the ventricular zone and sub-ventricular zone and migrate radially toward their final position[8,9]. Cytoskeletal control plays an important role in the regulationof neuronal migration, and impairment of this neuronal migrationresults in a variety of disorders, including gross brain abnormal-ities such as periventricular heterotopia and lissencephaly [10].In addition to such disorders, migration impairment is thought tounderlie several psychiatric diseases. It has been argued that thevulnerability to or etiology of schizophrenia can be comprehended

in the context of neurodevelopmental processes including neuronalmigration [11].

nce Le

ifiWcbnfigFian

2

2

owEtme

FT(

H. Yagi et al. / Neuroscie

We previously identified a novel molecule, filamin A interact-ng protein (FILIP, or FILIP-1 in humans), as a potential candidateor control of radial migration from the ventricular zone throughnteraction with the actin-binding protein, filamin A (FLNA) [12].

e have also shown that FILIP binds to non-muscle myosin heavyhain IIb (Myosin-10) and functions as a modulator for such actin-inding proteins [13]. As FLNA and Myosin-10 are involved ineuronal migration [14–16], we considered that FILIP is important

or brain development and that the deletion of FILIP would resultn cytoarchitectural disorganization in the brain. In the analysis ofene disruption in schizophrenia, a de novo missense mutation inILIP (FILIP-1) was previously reported [17]. In the present study, wenvestigated how FILIP functions in vivo using Filip-knockout micend revealed a novel role for FILIP in the development of callosaleurons in the mouse cortex.

. Materials and methods

.1. Animals

Mice were maintained in the animal room at the Division of Lab-ratory Animal Resources of University of Fukui. All experimentsere conducted in accordance with the Guidelines for Animal

xperiments of University of Fukui. The Animal Research Commit-ee of University of Fukui approved the experiments. We ensured

inimal pain and discomfort of the animals. The day that the pres-nce of a vaginal plug was confirmed was defined as embryonic

ig. 1. No apparent abnormalities are recognized in Filip−/− mice (−/−) with Nissl staininhe sensorimotor (A) and occipital (B) cortices and corpus callosum (C) of control (Filip+/+ (A, C) and 100 �m (B).

tters 612 (2016) 18–24 19

day 0.5 (E0.5). The day of birth was designated P0. Details of thegeneration of Filip-knockout mice have been previously reported[13].

2.2. Histological examination

For conventional histological examination, whole brains werefixed with 4% paraformaldehyde (PFA) in 0.1 M phosphate buffer(pH 7.4) at 4 ◦C. Brains were cryoprotected with 30% sucrose in0.1 M phosphate buffer (pH 7.4) and frozen in powdered dry ice.Coronal sections were cut at 14-�m thickness with a cryostat.Sections were mounted on 3-aminopropyltriethoxysilane-coatedglass slides (Matsunami Glass Ind., Kishiwada, Japan) and stainedwith thionine. For immunohistochmical analyses, the sections wereincubated overnight at 4 ◦C with the antibody-dilution buffer con-taining the anti-Cux-1 antibody (1:50, Santa Cruz Biotechnology,Santa Cruz, CA). The signals were visualized with Alexa Fluor 488-conjugated anti-rabbit IgG (Life Technologies Corporation, GrandIsland, NY).

2.3. 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanineperchlorate (DiI) injection

On postnatal day 2, the mice were deeply anesthetized byhypothermia. After perfusion with 4% PFA in phosphate-bufferedsaline (PBS), the brain was dissected out and fixed in 4% PFA inPBS. DiI crystals were inserted into the left occipital cortex of the

g.+/ + ) and Filip +/− (+/−)) littermates and Filip−/− mice are shown. Scale bar = 200 �m

20 H. Yagi et al. / Neuroscience Letters 612 (2016) 18–24

Fig. 2. Cortical neurons are malpositioned in Filip−/− mice.The ventricular zone cells at E14.5 were electroporated with the tdTomato expression vector, and the brains were observed at 3 weeks after birth. The cortex was dividedinto ten bins, followed by counting of the labeled neurons located in each bin. (A, B) The distribution of the labeled neurons observed in the occipital cortex of Filip+/+ (A) andFilip−/− mice (B) is presented. Scale bar = 50 �m. (C) Red open circles, triangles, and rectangles show the distribution of the labeled neurons from 4 independent control mice.R ectangc omatb

bwwM

2

TewPco

2

rNpinplcf

ed closed square indicates the mean distribution. Blue open circles, triangles, and rlosed circle indicates the mean distribution. The x-axis shows the proportion of tdTy Shaffer’s modified sequentially rejective Bonferroni procedure).

rains, and the brains were stored at 56 ◦C for 2–3 weeks. The brainsere sectioned using a vibratome (50-�m thickness). Labeled cellsere observed using 2-photon microscopy (LSM710 NLO, Carl ZeissicroImaging, Jena, Germany).

.4. In utero electroporation

Electroporation was performed as previously described [15,18].he cells located in the ventricular zone of E14.5 embryos werelectroporated with EGFP and tdTomato expression vectors. At 3eeks after birth, the brains were dissected out and fixed with 4%

FA in PBS. To analyze cell position, 100-�m-thick sections wereut using a vibratome. EGFP- and tdTomato-expressing cells werebserved using 2-photon microscopy (LSM710 NLO).

.5. In situ hybridization

cDNA fragments encoding mouse Plxnd1 (5120–3064 bp of NCBIeference sequence; NM 026376.3) and Er81 (3009–3738 bp ofCBI Reference Sequence; NM 007960.4) were inserted into theBluescript plasmid (Agilent Technologies, Santa Clara, CA) for

n vitro transcription. T7 and T3 RNA polymerases (Roche Diag-ostic K.K., Tokyo, Japan) were used for the antisense and sense

robes, respectively. Brains of adult Filip-knockout mice and their

ittermates were fixed with 4% PFA in 0.1 M phosphate buffer andryosectioned at 14-�m thickness. In situ hybridization was per-ormed as described before [19].

les show the distribution of labeled neurons from 4 independent Filip−/− mice. Blueo positive cells that were present in each bin. (*p < 0.05, **p < 0.01, ANOVA followed

2.6. Statistical analysis

We used the Fisher’s exact test for statistical analyses of thesummarized and categorized data. To analyze each bin statistically,we used one-way ANOVA.

3. Results

3.1. Filip deletion resulted in disorganized localization ofexcitatory neurons in the cortex

We investigated the cortex of Filip-knockout mice (Filip−/−) his-tologically. We did not detect any apparent abnormalities in adultFilip−/− mice with analysis of Nissl stained slices (Fig. 1). As theexpression of FILIP in the adult cortex was observed in a limitedpopulation of neurons, especially in the superficial layer of the cor-tex, it is probable that effects of FILIP deletion are restricted tothis population [13]. To study the details of the cortical organiza-tion in Filip−/− mice, we investigated the localization of neuronstransfected with the tdTomato expression vector on E14.5 at post-natal week 3 (postnatal day 20–22 (P20–22)). Whereas 66.0 ± 8.3%(average ± standard deviation) of the tdTomato-positive cells weredistributed within the upper half of the cortex in control mice (from

wild-type (Filip+/+) and heterozygous Filip-knockout (Filip+/−) lit-termates), 41.6 ± 9.4% of the tdTomato-positive cells were locatedwithin this region in Filip−/− mice (Fig. 2, 4 control and 4 Filip−/−

mice, p = 0.008, Welch’s t test). The labeled cells were more widely

H. Yagi et al. / Neuroscience Letters 612 (2016) 18–24 21

Fig. 3. The localization of callosal projection neurons is disturbed in Filip−/− mice.DiI crystals were injected into the contralateral occipital cortex. The labeled callosal projection neurons in the occipital cortex were dispersed in Filip−/− mice at postnatald cipitaS e (red6

dt

3F

ia

ay 2 (P2). (A, B) Representative examples of the labeled neurons located in the ocummarized data of the distribution of the labeled neurons from three control mic. (D) The DiI injection site is shown (asterisk). Scale bar = 1 mm.

istributed in the vertical direction in the cortex of Filip−/− micehan in that of control mice (Fig. 2C).

.2. Distribution of callosal projection neurons was wider inilip−/− mice

We investigated the distribution of callosal projection neuronsn Filip−/− mice because many of the callosal projection neuronsre located in the upper layers (layers II/III) and these excitatory

l cortex of a Filip+/+ littermate (A) and a Filip−/− mouse (B). Scale bar = 100 �m. (C)) and five Filip−/− mice (blue). The dotted line indicates the border of bin 5 and bin

neurons arise late in development [20]. Callosal projection neuronswere labeled using DiI injection into the contralateral occipital cor-tex at P2 (Fig. 3D). The labeled callosal neurons were distributed ina vertically wider area in Filip−/− mice compared to control mice(Fig. 3A–C). Approximately 60% of the labeled neurons (65 out of

104 neurons) in control mice and 50% of the labeled neurons (97of 204) in Filip−/− mice were located in the upper half of the cortex(Fig. 3C, p = 0.016, Fisher’s exact test).

22 H. Yagi et al. / Neuroscience Letters 612 (2016) 18–24

Fig. 4. Layer markers indicate the alteration of the localization of neurons in the visual cortex of Filip−/− mice.(A–F) Expressions of Plxnd1 and Er81 were observed in the visual cortex of coronal sections of the adult mouse brain. Representative examples of Plxnd1 -labeled neuronslocated in the visual cortex of a Filip+/+ littermate (A) and a Filip−/− mouse (B). Representative examples of Er81-labeled neurons located in the visual cortex of a Filip+/+

littermate (D) and a Filip−/− mouse (E). (G–I) Expression of Cux1 was observed in the visual cortex of coronal sections of the mouse brain at P15. Representative examples ofCux1-labeled neurons located in the visual cortex of a Filip+/+ littermate (G) and a Filip−/− mouse (H). Scale bar = 100 �m. (C, F, I) Red open circles, triangles, and rectanglesshow the distribution of the labeled neurons from independent control mice. Red closed square indicates the mean distribution. Blue open circles, triangles, and rectanglesshow the distribution of labeled neurons from independent Filip−/− mice. Blue closed circle indicates the mean distribution. The x-axis shows the proportion of Plxnd1 (C),Er81 (F) or Cux1 (I) positive cell that were present in each bin. (*p < 0.05, ANOVA).

nce Le

3i

tcabtnwtaio(

4

ctP[ri

ttpliaztmpgpc

rlhitbmiFrpapcmorra

oroo

H. Yagi et al. / Neuroscie

.3. Subtle differences in the cortical architecture were observedn Filip−/− mice

Our data suggest that deletion of Filip resulted in disturbance ofhe localization of the callosal neurons. As diversity exists amongallosal projection neurons [21], we investigated the distribution of

subpopulation of callosal projection neurons in the adult mouserain. Plxnd1 has been observed in the callosal projection neuronshat were located in the layer V and layer II of the cortex [21,22]. Weoticed that the proportion of Plxnd1 mRNA-containing neuronsas smaller in the superficial layer neurons of the Filip−/− cortex

han in control neurons (Fig. 4A–C). We did not notice any apparentbnormalities in terms of the localization of Cux1-positive neuronsn the Filip−/− mice (Fig. 4G–I). No evident abnormalities were rec-gnized in terms of the localization of Er81-positive neurons, eitherFig. 4D–F).

. Discussion

Our results showed that FILIP plays a role in the positioning ofallosal projection neurons in the cortex. As DiI is rarely effective inhe mature brain [23], we labeled the callosal projection neurons at2 when the later-born neurons had not reached their final position24,25]. Our results for the expression of Plxnd1 suggest that FILIP iselated to the distribution of the callosal projection neurons locatedn the superficial layer.

As the DiI-labeled callosal projection neurons that migrated tohe deep layer region increased in Filip−/− mice at P2 comparedo the control mice, it is likely that the subpopulation of callosalrojection neurons predestined to be localized in the superficial

ayer, including Plxnd1-positive neurons, migrated to and localizedn the deep layers in Filip−/− mice. The superficial layer neuronsre mainly derived from progenitors located in the subventricularone, including the outer radial glial cells [21]. As FILIP controlshe subcellular distribution of Myosin-10 [13], which controls the

itotic somal translocation of the outer radial glial cells [26], it isrobable that the deletion of FILIP influenced where the outer radiallial cells go through cell division and the positioning of callosalrojection neurons, which are derived from the outer radial glialells.

FILIP mutation has been reported in schizophrenia [17], and ouresults give rise to the possibility that the subpopulation of cal-osal projection neurons is decreased in schizophrenia. In fact, itas been previously reported that callosal connections are altered

n the brain of patients with schizophrenia, especially via reduc-ion of the connection between the hemispheres [4]. It has alsoeen reported that the volume of the corpus callosum, where theyelinated axons of the callosal projection neurons are located,

s reduced in schizophrenia [27–29]. There is a possibility thatILIP is involved in the pathogenesis of schizophrenia through theeduction of cortico-cortical connections. Furthermore, we havereviously reported that the propagation of cortical excitation isltered in Filip−/− mice compared to control littermates [13]. It isrobable that the combination of neuroanatomical changes in theortical organization shown in the present study and altered spineorphology of neurons in Filip−/− mice [13] resulted in alteration

f cortical excitation. Although we suspect that FILIP function iselated to schizophrenia phenotype, further studies are needed toeveal the relationship between this alteration of cortical excitationnd the pathogenesis of schizophrenia.

Our data suggest that deletion of FILIP resulted in disturbance

f the radial migration in a limited population of excitatory neu-ons. We previously reported that FILIP accelerates the degradationf FLNA, and overexpression of FILIP results in the disturbancef neuronal migration in rats [12]. Based on its mutation in the

tters 612 (2016) 18–24 23

human migration disorder periventricular nodular heterotopia,FLNA is considered to be essential for neuronal migration in humans[14,30]. In the rat brain, FLNA knockdown results in the impair-ment of radial glia and in periventricular nodular heterotopia [31].However, deletion of FLNA does not result in migration arrest inmice; thus, the importance of FLNA, whose degradation is enhancedby FILIP, in radial migration in the mouse brain remains unclear[32,33]. We also reported that FILIP binds to Myosin-10 (non-muscle myosin heavy chain 2b) [13], which is also involved in thedevelopment of the cortex [26]. As FILIP controls the intracellu-lar distribution of Myosin-10 and the amount of FLNA [12,13], it isprobable that FILIP-FLNA and/or FILIP-Myosin-10 interactions areinvolved in cortical development. Regarding psychiatric disordersand de novo mutations of MYH10, the gene encoding Myosin-10has been identified in the genome of the patients diagnosed withschizophrenia [34,35] and autism spectrum disorders [27]. In addi-tion, copy number variation of FLNA is found in the genome ofpatients suffering from autism disorder [35]. FILIP-FLNA and/orFILIP-Myosin-10 interactions may thus represent one of the inter-connected protein networks involved in psychiatric disorders.

Expression of Plxnd1 was observed in layer V and layer II/III inthe cortex. Although Plxnd1 in layer V is a marker of callosal neuronswith collateral fibers projecting to the striatum [22], the details ofPlxnd1-expressing neurons in layer II/III are not known. It is spec-ulated that there are many types of callosal neurons in layer II/IIIbecause of differences in gene expression [21]. Further studies areneeded to reveal the mechanism underlying the decrease in cellsin layer II/III of Filip−/− mice.

In summary, our study indicates that FILIP plays a role in theproper positioning of callosal projection neurons. As alteration ofcallosal projection neurons is present in the brain of patients withpsychiatric diseases and mutation of FILIP is found in schizophre-nia, FILIP may link callosal projections to the pathogenesis of braindisorders.

Acknowledgments

We are grateful to H. Yoshikawa, S. Kanae, C.C. Wang, and H.Miyagoshi (University of Fukui) for technical assistance and T.Taniguchi (University of Fukui) for secretarial assistance. This workwas supported in part by the Competitive Allocation Fund and theMultidisciplinary Program for Elucidating the Brain Developmentfrom Molecules to Social behavior (Fukui Brain Project) of Univer-sity of Fukui, and a Grant-in-Aid for Scientific Research and theStrategic Research Program for Brain Sciences “Integrated researchon neuropsychiatric disorders” from the Ministry of Education, Cul-ture, Sports, Science, and Technology (MEXT) of Japan.

Author contributionsH.Y. performed most of the experiments and analyzed the data.

M.S. initially conceived the project and directed the research. Y.O.conducted the experiments for in situ hybridization. M.K. preparedthe samples for histological analyses and performed the birthdateanalysis. M.-J.X performed in utero electroporation. K. N. helpedothers to analyze the data. M.S. wrote the manuscript together withH.Y., and all authors contributed to the final version.

References

[1] L.J. Richards, C. Plachez, T. Ren, Mechanisms regulating the development ofthe corpus callosum and its agenesis in mouse and human, Clin. Genet. 66(2004) 276–289.

[2] I. Hasegawa, T. Fukushima, T. Ihara, Y. Miyashita, Callosal window betweenprefrontal cortices: cognitive interaction to retrieve long-term memory,

Science 281 (1998) 814–818.

[3] M.S. Gazzaniga, Forty-five years of split-brain research and still going strong,Nat. Rev. Neurosci. 6 (2005) 653–659.

[4] G.M. Innocenti, F. Ansermet, J. Parnas, Schizophrenia, neurodevelopment andcorpus callosum, Mol. Psychiatry 8 (2003) 261–274.

2 nce Le

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[M. Karayiorgou, De novo gene mutations highlight patterns of genetic andneural complexity in schizophrenia, Nat. Genet. 44 (2012) 1365–1369.

4 H. Yagi et al. / Neuroscie

[5] M. Kubicki, M. Styner, S. Bouix, G. Gerig, D. Markant, K. Smith, R. Kikinis, R.W.McCarley, M.E. Shenton, Reduced interhemispheric connectivity inschizophrenia-tractography based segmentation of the corpus callosum,Schizophr. Res. 106 (2008) 125–131.

[6] M. Ribolsi, Z.J. Daskalakis, A. Siracusano, G. Koch, Abnormal asymmetry ofbrain connectivity in schizophrenia, Front. Hum. Neurosci. 8 (1010) (2014).

[7] M. Valiente, O. Marín, Neuronal migration mechanisms in development anddisease, Curr. Opin. Neurobiol. 20 (2010) 68–78.

[8] S.C. Noctor, V. Martínez-Cerdeno, L. Ivic, A.R. Kriegstein, Cortical neurons arisein symmetric and asymmetric division zones and migrate through specificphases, Nat. Neurosci. 7 (2004) 136–144.

[9] T. Miyata, A. Kawaguchi, K. Saito, M. Kawano, T. Muto, M. Ogawa, Asymmetricproduction of surface-dividing and non-surface-dividing cortical progenitorcells, Development 131 (2004) 3133–3145.

10] J.J. LoTurco, J. Bai, The multipolar stage and disruptions in neuronal migration,Trends Neurosci. 29 (2006) 407–413.

11] C.A. Ross, R.L. Margolis, S.A. Reading, M. Pletnikov, J.T. Coyle, Neurobiology ofschizophrenia, Neuron 52 (2006) 139–153.

12] T. Nagano, T. Yoneda, Y. Hatanaka, C. Kubota, F. Murakami, M. Sato, FilaminA-interacting protein (FILIP) regulates cortical cell migration out of theventricular zone, Nat. Cell Biol. 4 (2002) 495–501.

13] H. Yagi, T. Nagano, M.J. Xie, H. Ikeda, K. Kuroda, M. Komada, T. Iguchi, R.M.Tariqur, S. Morikubo, K. Noguchi, K. Murase, M. Okabe, M. Sato, FilaminA-interacting protein (FILIP) is a region-specific modulator of myosin 2b andcontrols spine morphology and NMDA receptor accumulation, Sci. Rep. 4(2014) 6353.

14] J.W. Fox, E.D. Lamperti, Y.Z. Eks ioglu, S.E. Hong, Y. Feng, D.A. Graham, I.E.Scheffer, W.B. Dobyns, B.A. Hirsch, R.A. Radtke, S.F. Berkovic, P.R.Huttenlocher, C.A. Walsh, Mutations in filamin 1 prevent migration ofcerebral cortical neurons in human periventricular heterotopia, Neuron 21(1998) 1315–1325.

15] T. Nagano, S. Morikubo, M. Sato, Filamin A and FILIP (Filamin A-InteractingProtein) regulate cell polarity and motility in neocortical subventricular andintermediate zones during radial migration, J. Neurosci. 24 (2004) 9648–9657.

16] D.J. Solecki, N. Trivedi, E.E. Govek, R.A. Kerekes, S.S. Gleason, M.E. Hatten,Myosin II motors and F-actin dynamics drive the coordinated movement ofthe centrosome and soma during CNS glial-guided neuronal migration,Neuron 63 (2009) 63–80.

17] T. S. Gulsuner, A.C. Walsh, M.K. Watts, A.M. Lee, S. Thornton, C. Casadei, H.Rippey, V.L. Shahin, R.C. Nimgaonkar, R.M. Go, N.R. Savage, R.E. Swerdlow, D.L.Gur, M.C. King Braff, Spatial and temporal mapping of de novo mutations inschizophrenia to a fetal prefrontal cortical network, Cell 154 (2013) 518–529.

18] H. Tabata, K. Nakajima, Efficient in utero gene transfer system to thedeveloping mouse brain using electroporation: visualization of neuronalmigration in the developing cortex, Neuroscience 103 (2001) 865–872.

19] H. M.J. Xie, K. Yagi, C.C. Kuroda, M. Wang, H. Komada, A. Zhao, T. Sakakibara, K.

Miyata, Y. Nagata, T. Iguchi Oka, WAVE2-Abi2 complex controls growth coneactivity and regulates the multipolar-bipolar transition as well as theinitiation of glia-guided migration, Cereb. Cortex 23 (2013) 1410–1423.

20] B.J. Molyneaux, P. Arlotta, J.R. Menezes, J.D. Macklis, Neuronal subtypespecification in the cerebral cortex, Nat. Rev. Neurosci. 8 (2007) 427–437.

[

tters 612 (2016) 18–24

21] R.M. Fame, J.L. MacDonald, J.D. Macklis, Development, specification, anddiversity of callosal projection neurons, Trends Neurosci. 34 (2011) 41–50.

22] B.J. Molyneaux, P. Arlotta, R.M. Fame, J.L. MacDonald, K.L. MacQuarrie, J.D.Macklis, Novel subtype-specific genes identify distinct subpopulations ofcallosal projection neurons, J. Neurosci. 29 (2009) 12343–12354.

23] A. Vercelli, M. Repici, D. Garbossa, A. Grimaldi, Recent techniques for tracingpathways in the central nervous system of developing and adult mammals,Brain Res. Bull. 51 (2000) 11–28.

24] F.L. Rice, C. Gomez, C. Barstow, A. Burnet, P. Sands, A comparative analysis ofthe development of the primary somatosensory cortex: interspeciessimilarities during barrel and laminar development, J. Comp. Neurol. 236(1985) 477–495.

25] K. Fukumoto, T. Morita, T. Mayanagi, D. Tanokashira, T. Yoshida, A. Sakai, K.Sobue, Detrimental effects of glucocorticoids on neuronal migration duringbrain development, Mol. Psychiatry 14 (2009) 1119–1131.

26] B.E. Ostrem, J.H. Lui, C.C. Gertz, A.R. Kriegstein, Control of outer radial glialstem cell mitosis in the human brain, Cell Rep. 8 (2014) 656–664.

27] A.G. M. Walterfang, D.C. Wood, S.J. Reutens, J. Wood, D. Chen, P.D. McGorryVelakoulis, Morphology of the corpus callosum at different stages ofschizophrenia: cross-sectional study in first-episode and chronic illness, Br. J.Psychiatry 192 (2008) 429–434.

28] S.A. Mitelman, Y.K. Nikiforova, E.L. Canfield, E.A. Hazlett, A.M. Brickman, L.Shihabuddin, M.S. Buchsbaum, A longitudinal study of the corpus callosum inchronic schizophrenia, Schizophr. Res. 114 (2009) 144–153.

29] S.C. S.L. Collinson, P.S. Gan, C. Woon, M.Y. Kuswanto, G.L. Sum, J.M. Yang, Y.Y.Lui, W.L. Nowinski Sitoh, Corpus callosum morphology in first-episode andchronic schizophrenia: combined magnetic resonance and diffusion tensorimaging study of Chinese Singaporean patients, Br. J. Psychiatry 204 (2014)55–60.

30] S.P. Robertson, Filamin A: phenotypic diversity, Curr. Opin. Genet. Dev. 15(2005) 301–307.

31] A. Carabalona, S. Beguin, E. Pallesi-Pocachard, E. Buhler, C. Pellegrino, K.Arnaud, P. Hubert, M. Oualha, J.P. Siffroi, S. Khantane, I. Coupry, C. Goizet, A.B.Gelot, A. Represa, C. Cardoso, A glial origin for periventricular nodularheterotopia caused by impaired expression of Filamin-A, Hum. Mol. Genet. 21(2012) 1004–1017.

32] Y. Feng, M.H. Chen, I.P. Moskowitz, A.M. Mendonza, L. Vidali, F. Nakamura, D.J.Kwiatkowski, C.A. Walsh, Filamin A (FLNA) is required for cell–cell contact invascular development and cardiac morphogenesis, Proc. Natl. Acad. Sci. U. S.A. 103 (2006) 19836–19841.

33] A.W. Hart, J.E. Morgan, J. Schneider, K. West, L. McKie, S. Bhattacharya, I.J.Jackson, S.H. Cross, Cardiac malformations and midline skeletal defects inmice lacking filamin A, Hum. Mol. Genet. 15 (2006) 2457–2467.

34] B. Xu, I. Ionita-Laza, J.L. Roos, B. Boone, S. Woodrick, Y. Sun, S. Levy, J.A. Gogos,

35] S.R. Gilman, J. Chang, B. Xu, T.S. Bawa, J.A. Gogos, M. Karayiorgou, D. Vitkup,Diverse types of genetic variation converge on functional gene networksinvolved in schizophrenia, Nat. Neurosci. 15 (2012) 1723–1728.