SIL1 , a causative cochaperone gene of M arinesco‐ S jögren syndrome, plays an essential role in establishing the architecture of the developing cerebral cortex
2014; Springer Nature; Volume: 6; Issue: 3 Linguagem: Inglês
10.1002/emmm.201303069
ISSN1757-4684
AutoresYutaka Inaguma, Nanako Hamada, Hidenori Tabata, Ikuko Iwamoto, Makoto Mizuno, Yoshiaki Nishimura, Hidenori Ito, Rika Morishita, Motomasa Suzuki, Kinji Ohno, Toshiyuki Kumagai, Koh‐ichi Nagata,
Tópico(s)Plant Molecular Biology Research
ResumoResearch Article29 January 2014Open Access Source Data SIL1, a causative cochaperone gene of Marinesco-Sjögren syndrome, plays an essential role in establishing the architecture of the developing cerebral cortex Yutaka Inaguma Yutaka Inaguma Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Nanako Hamada Nanako Hamada Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Hidenori Tabata Hidenori Tabata Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Ikuko Iwamoto Ikuko Iwamoto Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Makoto Mizuno Makoto Mizuno Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Yoshiaki V Nishimura Yoshiaki V Nishimura Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Hidenori Ito Hidenori Ito Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Rika Morishita Rika Morishita Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Motomasa Suzuki Motomasa Suzuki Central Hospital, Aichi Human Service Center, Kasugai, Aichi, Japan Search for more papers by this author Kinji Ohno Kinji Ohno Division of Neurogenetics, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan Search for more papers by this author Toshiyuki Kumagai Toshiyuki Kumagai Central Hospital, Aichi Human Service Center, Kasugai, Aichi, Japan Search for more papers by this author Koh-ichi Nagata Corresponding Author Koh-ichi Nagata Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Yutaka Inaguma Yutaka Inaguma Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Nanako Hamada Nanako Hamada Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Hidenori Tabata Hidenori Tabata Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Ikuko Iwamoto Ikuko Iwamoto Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Makoto Mizuno Makoto Mizuno Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Yoshiaki V Nishimura Yoshiaki V Nishimura Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Hidenori Ito Hidenori Ito Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Rika Morishita Rika Morishita Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Motomasa Suzuki Motomasa Suzuki Central Hospital, Aichi Human Service Center, Kasugai, Aichi, Japan Search for more papers by this author Kinji Ohno Kinji Ohno Division of Neurogenetics, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan Search for more papers by this author Toshiyuki Kumagai Toshiyuki Kumagai Central Hospital, Aichi Human Service Center, Kasugai, Aichi, Japan Search for more papers by this author Koh-ichi Nagata Corresponding Author Koh-ichi Nagata Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan Search for more papers by this author Author Information Yutaka Inaguma1,‡, Nanako Hamada1,‡, Hidenori Tabata1, Ikuko Iwamoto1, Makoto Mizuno1, Yoshiaki V Nishimura1, Hidenori Ito1, Rika Morishita1, Motomasa Suzuki2, Kinji Ohno3, Toshiyuki Kumagai2 and Koh-ichi Nagata 1 1Department of Molecular Neurobiology, Institute for Developmental Research, Kasugai, Aichi, Japan 2Central Hospital, Aichi Human Service Center, Kasugai, Aichi, Japan 3Division of Neurogenetics, Nagoya University Graduate School of Medicine, Nagoya, Aichi, Japan ‡These authors contributed equally to this work. *Corresponding author. Tel: +81 568 88 0811; Fax: +81 568 88 0829; E-mail: [email protected] EMBO Mol Med (2014)6:414-429https://doi.org/10.1002/emmm.201303069 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract Marinesco-Sjögren syndrome (MSS) is a rare autosomal recessively inherited disorder with mental retardation (MR). Recently, mutations in the SIL1 gene, encoding a co-chaperone which regulates the chaperone HSPA5, were identified as a major cause of MSS. We here examined the pathophysiological significance of SIL1 mutations in abnormal corticogenesis of MSS. SIL1-silencing caused neuronal migration delay during corticogenesis ex vivo. While RNAi-resistant SIL1 rescued the defects, three MSS-causing SIL1 mutants tested did not. These mutants had lower affinities to HSPA5 in vitro, and SIL1-HSPA5 interaction as well as HSPA5 function was found to be crucial for neuronal migration ex vivo. Furthermore time-lapse imaging revealed morphological disorganization associated with abnormal migration of SIL1-deficient neurons. These results suggest that the mutations prevent SIL1 from interacting with and regulating HSPA5, leading to abnormal neuronal morphology and migration. Consistent with this, when SIL1 was silenced in cortical neurons in one hemisphere, axonal growth in the contralateral hemisphere was delayed. Taken together, abnormal neuronal migration and interhemispheric axon development may contribute to MR in MSS. Synopsis SIL1 mutations cause Marinesco-Sjögren syndrome. SIL1 is shown to be involved in neuronal morphology and migration and axon network formation. SIL1 mutations may thus contribute to abnormal corticogenesis during development leading to mental retardation A new SIL1 gene mutation was found to cause Marinesco-Sjögren syndrome. SIL1 is crucial for proper neuronal morphology, migration and axon growth. SIL1 may be essential for brain development. Defects in the SIL1-HSPA5 chaperone system may cause functional deficiency in brain leading to mental retardation. Introduction Marinesco–Sjögren syndrome (MSS; OMIM 248800) is an autosomal recessive disorder affecting various tissues. The clinical features of MSS are cerebellar ataxia, early-onset cataracts, progressive myopathy and mental retardation (MR). In addition, muscle atrophy, skeletal abnormalities, short stature, psychomotor delay, nystagmus, strabismus, hypotonia and dysarthria are frequent findings (Marinesco et al, 1931; Sjogren, 1950; Lagier-Tourenne et al, 2003). Brain magnetic resonance imaging (MRI) studies have revealed general cerebellar atrophy while dysplastic cytoarchitecture in the cerebral cortex was reported by an autopsy in one patient clinically diagnosed as MSS (Reinhold et al, 2003; Sakai et al, 2008). Recently, the MSS phenotypes have been found to be caused by mutations in the SIL1 gene on chromosome 5q31, resulting in loss of SIL1 function due to premature termination of translation, abnormal splicing of the transcript or single amino acid substitution (Anttonen et al, 2005, 2008; Senderek et al, 2005; Karim et al, 2006; Eriguchi et al, 2008; Riazuddin et al, 2009; Takahata et al, 2010). SIL1 is an ER-resident glycoprotein harboring an N-terminal endoplasmic reticulum (ER) targeting sequence, four armadillo repeats (ARMs), 2 N-linked glycosylation sites (Asn193 and Asn236) and a C-terminal putative ER retention tetrapeptide (Chung et al, 2002). SIL1 acts as an adenine nucleotide exchange factor for the ER homologue of Hsp (heat shock protein) 70 family, HSPA5 (also known as BiP and GRP78; Zoghbi, 2005). ER is the central site of protein synthesis and protein quality control, and ER chaperones such as HSPA5 associate with newly synthesized proteins to prevent their aggregation and help them fold and assemble correctly. SIL1 binds preferentially with ADP-bound HSPA5, catalyzes ADP release and subsequent release of HSPA5 from its substrates, and in turn allows ATP to bind to HSPA5 (Chung et al, 2002). SIL1 is therefore involved in proper folding of newly synthesized proteins and degradation of proteins that fail to mature properly, in a coordinated manner with HSPA5. Mutations in SIL1 gene are thus predicted to cause malfunction of HSPA5 leading to misfolding and dysfunction of HSPA5 substrates, resulting in loss of function of the substrates. While most SIL1 mutation-positive MSS patients show the hallmark clinical features such as MR, myopathy, cerebellar atrophy and ataxia, and cataracts, additional features and their severity vary from patient to patient, seemingly depending on the mutation type at least to some extent. Moreover, MSS is considered to be a clinically and genetically heterogeneous disorder, since SIL1 gene mutations are not found in approximately 40% of patients with classical MSS (Senderek et al, 2005). In the present study, we identified novel compound heterozygous mutations in a Japanese patient. By biochemical and cell biological characterization, 3 representative MSS-causing SIL1 mutants showed aberrant subcellular localization, cytoplasmic aggregate formation and decreased affinity to HSPA5. We then demonstrated that these SIL1 mutations induced neuronal migration delay during corticogenesis and that the SIL1-HSPA5 interaction is essential for migration. SIL1 was also found to play an important role in interhemispheric neuronal connection formation through axon development. Consequently, the SIL1 mutations are likely to cause defective cortical development leading to MR of MSS. Results Clinical investigations of MSS patients and mutation analyses All eight patients analyzed for SIL1 mutations presented key clinical features of MSS, including MR, cerebellar ataxia with cerebellar atrophy, bilateral cataracts and myopathy (Table 1). The family name and permanent domicile differed among the cases, and no information indicating kinship among them was obtained. Familial occurrence of MSS was found for none of them. Table 1. Clinical features of the patients Patients with SIL1 mutations Patients without SIL1 mutaions Clinical features Case 1 Case 2 Case 4 Case 6 Case 3 Case 5 Case 7 Case 8 Sex F M M M M M F M Age at last evaluation 34 5 7 3 30 17 17 3 Short stature + + − − − + + − Cataract + + + − + + + ( C (L457P) (Heterozygous) c.936dupG c.1230-1244del F, female; M, male; RV, rimmed vacuole; ND, not determined; +, mild; ++, moderate; +++, profound; −, absent. a Age at diagnosis. We identified mutations in the SIL1 gene in four patients (Table 1). Case#1 was homozygous for a G to A transition in intron 9 (c.1030-9G>A), leading to introduction of a novel splice acceptor site and frameshift (Anttonen et al, 2008). This mutation causes a frameshift at codon 345 and truncation of the protein after eight novel amino acids (p.Phe345AlafsX9). Case#2 was homozygous for a one-base insertion of G at the position 936_937 in exon 9 (c.936dupG) predicted to cause frameshift at codon 313 and premature truncation with 38 novel amino acids at the beginning of exon 10 (p.Leu313AlafsX39; Anttonen et al, 2008; Eriguchi et al, 2008). The mutant protein, designated here as SIL1-7G, lacks the C-terminal 2 ARMs, which are involved in the interaction with HSPA5, and the ER retention motif (Fig 1A). Case#4 was compound heterozygous with c.936dupG and a T to C transition c.1370T>C leading to substitution of leucine with proline (p.Leu457Pro). This mutant protein, SIL1-L457P, does not localize normally to ER but distributes in the cytoplasm with aggregate formation (Fig 1A; Anttonen et al, 2008). Case#6 was compound heterozygous with c.936dupG and a novel 15 bp-deletion at the position of 1230_1244 in exon10 (c.1230-1244del), leading to a deletion mutant, SIL1-15del, lacking 5 amino acids from aa411 to 415 (RYRQD; Fig 1A and B). c.1230-1244del was not registered in world-wide SNP (single nucleotide polymorphism) database of dbSNP137, 1000 genomes project, and NHLBI ESP database; or in Japanese SNP database of JSNP (http://snp.ims.u-tokyo.ac.jp/) and Japanese genetic variation consortium (http://www.genome.med.kyoto-u.ac.jp/SnpDB/). This mutated region was not a major site to interact with HSPA5, but the five amino acid-deletion may affect the SIL1 structure and/or function. Figure 1. Mutation analyses of MSS patients Description of wild type and the three MSS-causing SIL1 mutants analyzed in this study. Structural domains are abbreviated as follows: ETS, ER targeting sequence (aa1-30); ARMs, Armadillo repeats (aa170-341); NG, N-linked glycosylation site (Asn193 and Asn236); ERS, ER retention signal (aa458-461). Chromatograms showing SIL1 mutations in Case#6. Compound heterozygous mutations were identified for a 1-bp insertion of G at the position 936_937 in exon 9 (upper panels) and a novel in-frame 15 bp-deletion at the position of 1230_1244 in exon 10 (lower panels). The mutated nucleotides are marked. Download figure Download PowerPoint In a subset of patients (Case#3, #5, #7 and #8), no mutations were found in the exons and exon-intron boundaries of the SIL1 gene, even though the patients expressed the clinical criteria for classical MSS (Table 1). This supports the suggestion of genetic heterogeneity in MSS, and further MSS-causing gene(s) should be identified to unravel the molecular genetic basis of the disease. Aggregate formation of the MSS-causing SIL1 mutants expressed in COS7 cells SIL1-L457P was shown to distribute abnormally in the cytoplasm and form aggregates, strongly suggesting that the ER-retention motif is impaired in the mutant (Anttonen et al, 2008). When SIL1-7G and SIL1-15del were expressed in COS7 cells, they also formed cytoplasmic aggregates under the conditions where wild type SIL1 was preferentially localized to the ER (P = 0.0069 for SIL1-7G, P = 0.0312 for SIL1-L457P, and P = 0.0017 for SIL1-15del; Fig 2A). Although the ER retention motif is present in SIL1-15del, abnormal conformation of the mutant protein may mask it. Figure 2. Cell biological and biochemical characterization of MSS-causing SIL1 mutants Quantification of aggregate formation of SIL1 and the MSS-causing mutants exogenously expressed in COS7 cells. The percentage of cells with aggregates of SIL1 and the mutant proteins was scored. Numbers of cells used for each calculation are more than 150. Error bars indicate s.d. (n = 5); *P = 0.0312 (L457P), **P = 0.0017 (15del), **P = 0.0069 (7G) by Student's t-test. Immunoprecipitation of SIL1 and the MSS-causing mutants with HSPA5. Lysates from HEK293 cells expressing GFP-hHSPA5 alone or with Flag-hSIL1, -SIL1-15del, -SIL1-7G, -SIL1-L457P or -hHSPBP1 were immunoprecipitated with anti-Flag M2 (1 μg). A portion (20%) of the precipitated materials was subjected to western blotting with polyclonal anti-Flag or anti-GFP antibody (IP). Expression of each protein was confirmed with M2 and anti-GFP (Exp; 3% of total volume). Interaction of N-terminal half of SIL1-ARM region with HSPA5. HEK293 cells were transfected with Flag-hHSPA5, GFP-SIL1-ARM (aa141-350) and GFP-SIL1-ARM-N (aa141-260) in various combinations. Lysates were immunoprecipitated with polyclonal anti-GFP (1 μg) and the precipitated materials (20%) were analyzed by western blotting with a mixture of M2 and monoclonal anti-GFP antibodies (IP). Asterisk (*) shows non-specific signals. Expression of each protein was also confirmed with M2 and monoclonal anti-GFP (Exp; 3% of total volume). Source data are available for this figure. Source Data for Figure 2BC [emmm201303069-SourceData-Fig2BC.pdf] Download figure Download PowerPoint Interaction of MSS-causing SIL1 mutants with HSPA5 SIL1 exons 6–9 encode ARMs, which interact with HSPA5, and exon 10 also provides a minor interaction site for HSPA5 (Senderek et al, 2005). In this context, mutations in SIL1-7G, SIL1-15del and SIL1-L457P are located in exon 9 or 10, and thus may cause conformational change of SIL1 that affects its interaction with HSPA5. To test this possibility, interaction of these mutants with HSPA5 was analyzed in vitro by immunoprecipitation following co-expression in COS7 cells. As shown in Fig 2B, these mutations significantly diminished the binding capacity of SIL1 to HSPA5 when compared to wild type SIL1 and a SIL1-related protein HSPBP1 (Raynes & Guerriero, 1998). We next narrowed down the ARM region responsible for binding to HSPA5, and found that the N-terminally localized two ARMs were sufficient for the interaction (Fig 2C). The tissue expression pattern of endogenous SIL1 was very similar to that of HSPA5 in mouse brain by immunohistochemical analyses, consistent with their functional interaction (Anttonen et al, 2005). Since abnormal corticogenesis may contribute to the onset of MR, one of the key symptoms of MSS, we examined SIL1 expression in the cerebral cortex during the developmental stage. In western blotting, both SIL1 and HSPA5 were detected in cerebral cortex from E13.5 to P30 although their expression patterns were distinct; SIL1 increased dramatically after P8 and reached the maximum level around P15 while HSPA5 amount remained unchanged during the time period (Fig 3A and B). We next examined mRNA expression profiles of SIL1 and HSPA5 by in situ hybridization during brain development. As shown in Fig 3C, SIL1 and HSPA5 were expressed in the ventricular zone (VZ)/subventricular zone (SVZ) cells and neurons in CP at E15, E17, P0 and P8. Although HSPA5 mRNA showed relatively strong expression in VZ cells until P0, both SIL1 and HSPA5 were expressed in progenitor cells in VZ/SVZ and neurons in CP during corticogenesis. Together with western blotting results, SIL1 is likely to interact with HSPA5 in a spatiotemporally regulated manner during brain development. Figure 3. Expression profiles of SIL1 and HSPA5 in developing mouse brain A, B. Whole lysates (20 μg protein) of cerebral cortices at various developmental stages were subjected to western blotting with antibodies indicated. C. Coronal sections were examined for SIL1- or HSPA5-mRNA by in situ hybridization at E15, E17, P0 and P8. Sense control cRNA probes were used for P0 slices. Bars, 100 μm. Source data are available for this figure. Source Data for Figure 3AB [emmm201303069-SourceData-Fig3AB.pdf] Download figure Download PowerPoint Roles of SIL1 in neuronal migration during corticogenesis Since MSS-causing SIL1 mutations are thought to induce abnormal cytoarchitecture of the cerebral cortex, we performed RNAi experiments to examine the role of SIL1 in the migration of newly generated cortical neurons. We designed two RNAi vectors, pSUPER-mSIL1#1 and -mSIL1#2, against distinct regions in the mSIL1 coding sequence. Both vectors efficiently knocked down mSIL1 in COS7 cells (Fig 4A, left panels). When dissociated mouse cortical neurons were transfected with these RNAi vectors, endogenous SIL1 was reduced significantly (Fig 4B). Subsequently, the RNAi vectors and pCAG-EGFP were coelectroporated into progenitor cells in VZ of embryonic murine brains by the in utero electroporation method, and localization of transfected cells and their progeny was observed at P0. In control experiments, vector-transfected neurons migrated normally to the superficial layer (layers II–IV) of the cortical plate (CP; Fig 4C). In contrast, a considerable portion of cells transfected with pSUPER-mSIL1#1 or -mSIL1#2 remained in the lower zone of CP and intermediate zone (IZ; Fig 4C). As shown in Fig 4D, one-way ANOVA revealed significant effects of the RNAi vectors in layers II–IV (F2,6 = 599.694, P = 0.0001), layers V–VI (F2,6 = 49.576, P = 0.0002), IZ (F2,6 = 14.216, P = 0.0053 and SVZ/VZ (F2,6 = 15.128, P = 0.0045). Post-hoc tests detected significant neuronal migration defects by transfection of pSUPER-mSIL1#1 or -mSIL1#2 compared with control experiments. Consistently, neuron distribution in pSUPER-mSIL1#1- or -mSIL1#2-transfected brain slices revealed abnormal neuronal positioning when compared to control experiments (χ2 = 67.973, df = 3, P = 0.0001 for mSIL1#1) and (χ2 = 111.87, df = 3, P = 0.0001 for mSIL1#2). It is notable that many SIL1-deficient neurons reached the superficial layer of CP (Fig 4C and D). A possible explanation for these results is that neurons incorporating low amount of the RNAi vector were incompletely depleted of SIL1; transfection efficiency (and therefore knockdown) is dependent on the cell surface area physically exposed to the ventricular lumen (cerebrospinal fluid) where RNAi vectors are present. When the morphology of SIL1-deficient neurons was examined in the lower CP (layers V–VI), abnormal multipolar-shaped cells were frequently observed (Fig 4E). Figure 4. Role of SIL1 in neuronal migration during mouse brain development Characterization of pSUPER-mSIL1 vectors. pCAG-Myc-mSIL1 or –hSIL1 was cotransfected into COS7 cells with control pSUPER vector (−), pSUPER-mSIL1#1 or -mSIL1#2. After 48 h, cells were harvested and subjected to western blotting with anti-Myc. Anti-β-tubulin was used for loading control. The RNAi-target sequence of pSUPER-mSIL1#1 was shown with the corresponding human sequence. Different nucleotides were marked under the human sequence. Knockdown of endogenous SIL1 in cortical neurons. pCAG-EGFP was transfected with pSUPER vector, pSUPER-mSIL1#1 or -mSIL1#2 into dissociated neurons obtained from the cerebral cortices at E 17 and cultured in vitro for 48 h. After fixation, cells were immunostained with monoclonal anti-GFP (green) and polyclonal anti-SIL1 (red). Note that the number of EGFP/SIL1-double positive cells decreased significantly in pSUPER-mSIL1#1- or -mSIL1#2-transfection experiments. Bars in (B) and (E), 10 μm. Migration defects of SIL1-deficient cortical neurons at P0. pCAG-EGFP was coelectroporated with control pSUPER vector (a), pSUPER-mSIL1#1 (b) or -mSIL1#2 (c) into cerebral cortices at E14, followed by fixation at P0. Coronal sections were stained for GFP (white) and nuclei with DAPI (blue). Dotted lines represent the pial and ventricular surfaces. Bar, 100 μm. Quantification of the distribution of SIL1-deficient neurons in distinct parts of the cerebral cortex (layers II–IV, layers V–VI, IZ and SVZ/VZ) for each condition shown in (C). Error bars indicate s.d. (n = 3); *P = 0.0223 (layer IZ), **P = 0.004 (layers V–VI), **P = 0.0018 (layer IZ), **P = 0.0017 (layers SVZ/VZ), ***P = 0.0001 (layers II–IV), ***P = 0.0001 (layers V–VI) by Fisher's LSD. Representative images of control and SIL1-deficient neurons migrating in lower CP. pSUPER vector (control)- and pSUPER-mSIL1#1-transfected cells were shown at left and right panels, respectively. Source data are available for this figure. Source Data for Figure 4A [emmm201303069-SourceData-Fig4A.pdf] Download figure Download PowerPoint When we further analyzed the effects of SIL1-knockdown on the neuronal migration at E17, abnormal positioning of SIL1-deficient neurons was observed (supplementary Fig 1A and B). On the other hand, it should be noted that SIL1-deficient cells reached the target location (layers II–III) at P7, indicating that SIL1-silencing delayed, but did not prevent, radial migration of cortical neurons (supplementary Fig 1G and H). Next, since the SIL1-HSPA5 chaperone system is involved in cell stress and MSS is thought to be a neurodegenerative disease, we evaluated whether apoptosis takes place in SIL1-deficient neurons. When caspase3 activation was visualized during brain development, the activation was hardly detected at E17, P0 and P7 (supplementary Fig 2B–Da and b). We then examined whether cells that were abnormally positioned in the IZ and VZ/SVZ were differentiated to neurons or not. To this end, immunostaining was done at P0 for doublecortin (Dcx) and Tbr2, markers for neuronal precursor cells/immature neurons and basal progenitor cells, respectively. Consequently, these cells were Dcx-positive but only weakly positive for Tbr2, indicating that they were being committed to neurons (supplementary Fig 3A and Ba). We next performed rescue experiments and used hSIL1 for this purpose since it was resistant to pSUPER-mSIL1#1-mediated silencing (Fig 4A, right panels). When pCAG-EGFP and pSUPER-mSIL1#1 were coelectroporated along with pCAG-Flag-hSIL1, the positional defects caused by SIL1-knockdown were rescued at P0 (Fig 5Aa–c). One-way ANOVA revealed the significant effects of the RNAi-vector transfection in layers II–IV (F2,6 = 25.852, P = 0.0011), layers V–VI (F2,6 = 25.609, P = 0.0012) and SVZ/VZ (F2,6 = 16.052, P = 0.0039), but not in IZ (F2,6 = 2.769, P = 0.1406). Rescue effects of hSIL1 were confirmed statistically (χ2 = 44.684, df = 3, P = 0.0001; Fig 5B). In contrast, expression of Flag-SIL1-7G, -SIL1-L457P or -SIL1-15del was unable to rescue the migration defects (Fig 5Ac–f). As shown in Fig 5C, one-way ANOVA revealed the significant effects of the RNAi-vector transfection in layers II–IV (F3,8 = 8.89, P = 0.0063, layers V–VI (F3,8 = 6.295, P = 0.0168 and SVZ/VZ (F3,8 = 4.872, P = 0.0326), but not in IZ (F3,8 = 0.45, P = 0.7245). Post-hoc tests showed that abnormal neuronal migration was not rescued by SIL1-7G, SIL1-L457P or SIL1-15del when compared with hSIL1. Consistently, neuron distribution in hSIL1-7G-, hSIL1-L457P- or hSIL1-15del-expressing brain slices exhibited statistically abnormal neuronal positioning when compared to hSIL1-expressing ones (χ2 = 14.579, df = 3, P = 0.0022 for SIL1-7G, χ2 = 18.167, df = 3, P = 0.0004 for SIL1-L457P, and χ2 = 9.847, df = 3, P = 0.0199 for SIL1-15del). We confirmed that the SIL1 mutants as well as the wild type were comparably expressed in the cortical neurons, and HSPA5 level was not influenced by the transfection (Fig 5D). These results indicate that the MSS-causing SIL1 mutants could not substitute for SIL1, suggesting pathophysiological significance of these SIL1 mutations in the abnormal cortical neuron migration, which might result in the dysplastic cerebral cytoarchitecture in MSS. Figure 5. Rescue of SIL1-RNAi-induced migration defects Rescue of migration defects by an RNAi-resistant version of SIL1 (hSIL1) but not MSS-causing mutants. pCAG-EGFP was coelectroporated with pSUPER vector only (a) or pSUPER-mSIL1#1 together with pCAG vector (b), pCAG-Flag-hSIL1 (c), -SIL1-7G (d), -SIL1-L457P (e) or -SIL1-15del (f) into cerebral cortices at E14, followed by fixation at P0. Coronal sections were stained for GFP (white) and nuclei (blue). Dotted lines represent the pial and ventricular surfaces. Bar, 100 μm. Quantification of the distribution of neurons in distinct regions of the cerebral cortex for each condition in (A, a–c). Error bars indicate s.d. (n = 3); **P = 0.0019 (layers II–IV), **P = 0.0016 (layers V–VI), **P = 0.0019 (SVZ/VZ; control versus SIL1#1), **P = 0.0041 (SVZ/VZ; SIL1#1 versus SIL1#1+hSIL1), ***P = 0.0005 (layers II–IV), ***P = 0.0005 (layers V–VI) by Fisher's LSD. Quantification of the distribution of neurons in distinct regions of the cerebral cortex for each condition in (A, c–f). Error bars indicate s.d. (n = 3); *P = 0.012 (layers II–IV), *P = 0.0164 (layers V–VI), **P = 0.005 (layers II–IV; 7G), **P = 0.0012 (layers II–IV; L457P), **P = 0.0059 (layers V–VI; 7G), **P = 0.0059 (layers V–VI; L457P), **P = 0.0056 (layers SVZ/VZ) by Fisher's LSD. Expression of hSIL1 and the mutants in dissociated cortical neurons. Cortical neurons (E17) were electroporated with pSUPER-SIL1#1 only (a) or together with pCAG-Flag vector (b), pCAG-Flag-hSIL1 (c), -SIL1-7G (d), -SIL1-L457P (e) or -SIL1-15del (f), and cultured in vitro for 48 h. Cells were then subjected to SDS-PAGE (20 μg protein per lane) followed by western blotting with anti-Flag M2, anti-HSPA5 or anti-β-tubulin. Asterisks indicate M2-positive bands. Source data are available for this figure. Source Data for Figure 5D [emmm201303069-SourceData-Fig5D.pdf] Download figure Download PowerPoint When effects of the three MSS-causing SIL1 mutants on neuronal positioning were analyzed at E17 after electroporation at E14, their phenotypes resembled SIL1-deficient cells and migration defects were observed at this time point (supplementary Fig 1C–E). Ultimately, SIL1-deficient neurons expressing the hSIL1 mutants reached the target location (layers I
Referência(s)