Glucocorticoids, the most downstream effectors of the hypothalamus–pituitary–adrenal axis, are one of main mediators of the stress reaction. Indeed, exposure to high levels of stress-triggered glucocorticoids is detrimental to brain development associated with abnormal behaviors in experimental animals and the risk of psychiatric disorders in humans. Despite the wealth of this knowledge, the cellular and molecular mechanisms underlying the detrimental effects of glucocorticoids on brain development remain unclear. Here, we show that excess glucocorticoids retard the radial migration of post-mitotic neurons during the development of the cerebral cortex, and identify an actin regulatory protein, caldesmon, as the glucocorticoids’ main target. The upregulation of caldesmon expression is mediated by glucocorticoid receptor-dependent transcription of the CALD1 gene encoding caldesmon. This upregulated caldesmon negatively controls the function of myosin II, leading to changes in cell shape and migration. The depletion of caldesmon in vivo impairs radial migration. The overexpression of caldesmon also causes delayed radial migration during cortical development, mimicking the excessive glucocorticoid-induced retardation of radial migration. We conclude that an appropriate range of caldesmon expression is critical for radial migration, and that its overexpression induced by excess glucocorticoid retards radial migration during cortical development. Thus, this study provides a novel insight into the underlying mechanism of glucocorticoid-related neurodevelopmental disorders.
Dysregulation of the hypothalamus–pituitary–adrenal axis is implicated in certain psychiatric disorders, including anxiety, depression and schizophrenia.1, 2, 3, 4 The most downstream effectors of the hypothalamus–pituitary–adrenal axis, glucocorticoids (GCs), are one of the main mediators of the stress reaction. It is well documented that stress-triggered GCs modulate changes in synapse formation, dendritic arborization and hippocampal volume in adult life.5 Exposure of excessive stress/GCs also affects the structure and function of other adult brain regions, including the prefrontal cortex and amygdala.6, 7 Furthermore, elevation of GCs in response to stress is detrimental to the development and function of the brain.3, 8, 9 For instance, maternal stress can affect the structure of the developing brain, leading to altered behaviors in adult offspring.10 Prenatal stress is also associated with increased anxiety in adolescence and a greater incidence of psychiatric disorders in adulthood.1, 11, 12 Besides perinatal stress, GC treatment with neonatal rats transiently retards brain development.13 The single or repeated administration of a synthetic GC, betamethasone, to pregnant sheep also retards fetal brain growth.14 In humans, antenatal exposure to betamethasone results in infants having a reduced cortex convolution index and surface area.15 Thus, excessive stress/GC exposure affects brain development, and subsequently causes abnormal behavior in experimental animals and an increased risk of psychiatric disorders in humans. Despite these numerous studies, the underlying mechanism of the detrimental effects of GCs on brain development and function remains unclear.
Post-mitotic neurons in the cerebral cortex generated from neuronal precursor cells (NPCs) within the ventricular zone (VZ) radially migrate along radial glial fibers to their final destination in the cortical plate (CP), resulting in the formation of the cortical mantle.16, 17 Several human disorders that arise from defective neuronal migration have been identified. These disorders include periventricular nodular heterotopia and lissencephaly, which are caused by mutations in genes involved in neuronal migration.18 More commonly than gene mutations, however, environmental factors such as stress-triggered GCs as described above and thyroid hormones are implicated in inducing abnormalities of brain development. It has been demonstrated regarding thyroid hormones that early maternal hypothyroxinemia in rats impairs radial migration and cortical cytoarchitecture.19, 20
Here, we investigated the effects of excess GCs on brain development and found cellular and molecular bases for the detrimental actions of GCs on neuronal migration during cortical development.
Materials and methods
All of the procedures involving animals and their care were approved by the Animal Care Committee of Osaka University, and were carried out according to the guidelines for animal experiments of the Osaka University School of Medicine. Wistar rats were used (Japan SLC, Shizuoka, Japan).
Protocol for DEX administration to pregnant rats
Pregnant rats were housed singly and assigned randomly to the dexamethasone (DEX) or control (vehicle) groups. DEX (200 μg kg−1 day−1, dissolved in sesame oil containing 0.16% ethanol) was injected subcutaneously into pregnant rats once a day from E14.5 to E20.5.21, 22 A control group received injections of vehicle (sesame oil containing 0.16% ethanol) during the same time period.
Expression plasmids and transfection
We used highly efficient mammalian expression plasmids, pCAGGS. The coding regions for rat l-CaD, human l-CaD, the N terminus (1–263 amino acids) and C terminus (264–558 amino acids) of human l-CaD, human ezrin and human GR were amplified by PCR and subcloned into a highly efficient mammalian expression plasmids, pCAGGS. A HA tag sequence was fused to the 5′ end of the coding sequence of the rat l-CaD gene by PCR. For the GFP-tagged CaD constructs, the coding region of the EGFP gene was amplified from the pEGFP-C2 vector (Clontech, Mountain View, CA, USA) and ligated into the 5′ end of the CaD sequence. A constitutively active form of human GR (GRΔC) was constructed as described previously.23 The CaD microRNAs (miRNA) plasmids (CaD miRNA no. 1 and CaD miRNA no. 2) were constructed using the Block-iT Pol II miR RNAi Expression Vector kit (Invitrogen, Carlsbad, CA, USA). The single-stranded DNA oligos containing target sequences were designed as follows: CaD miRNA no. 1, 5-IndexTermtgctgtccgcttgccagatacatctcgttttggccactgactgacgagatgtatggcaagcgga-3′; CaD miRNA no. 2, 5′-IndexTermtgctgtgcattggtatattgctccaggttttggccactgactgacctggagcaataccaatgca-3′. For higher expression efficiency in NPCs, the coding regions of EmGFP and miRNAs were amplified from target sequence-inserted pcDNA6.2-GW/EmGFP-miR plasmids by PCR and subcloned into pCAGGS. The expression plasmids were transfected into NPCs or HEK 293T cells using TransIT-LT1 transfection reagent (Mirus Bio, Madison, WI, USA) 2 or 3 days before analysis.
The NPCs were transfected with short-interfering RNAs (siRNAs) against myosin IIA and/or myosin IIB using Lipofectamine RNAiMAX (Invitrogen) and cultured for 3 days before analysis. The siRNA sequences were as follows: myosin IIA no. 1, 5′-IndexTermgtcattaatccttataagaac-3′; myosin IIA no. 2, 5′-IndexTermgcaagcttacaaaggacttct-3′; myosin IIB no. 1, 5′-IndexTermcccacgttgcttcttcacaca-3′; myosin IIB no. 2, 5′-IndexTermcgcacgtttcatatcttttat-3′. In control experiments, AllStars Negative Control siRNA (Qiagen, Venlo, Netherlands) was used.
In utero electroporation
In utero electroporation was performed at E15.5 or E17.5, as described previously.24 In brief, pregnant rats were deeply anaesthetized, and 1–2 μl of plasmid solution including 0.25 mg ml−1 fast green (Sigma-Aldrich, St Louis, MO, USA) was administered into the intraventricular region of the embryonic brain, followed by electroporation. Electric pulses were generated by an Electro Square Porater ECM 830 (BTX, Holliston, MA, USA) and applied to the cerebral wall as five repeats of 50 V for 50 ms, with an interval of 500 ms. All the expression plasmids were used at a concentration of 2 μg μl−1.
NPC cultures and drug treatment
For monolayer cultures, NPCs were isolated from the cerebral cortex of E15.5 rat embryos. The cells were incubated with 0.025% trypsin (Sigma-Aldrich) in Ca2+/Mg2+-free Hank's Balanced salt solution (HBSS, Sigma-Aldrich) at 37 °C for 5 min. The trypsin was washed out, and the tissues were triturated in growth medium [1:1 Dulbecco's modified Eagle's medium and nutrient Mixture F-12 (DMEM/F12, Gibco, Carlsbad, CA, USA) containing supplements (5 μg ml−1 insulin, 100 μm putrescine and 30 nm selenium dioxide)] and 20 ng ml−1 recombinant human fibroblast growth factor basic (bFGF, R&D systems, Minneapolis, MN, USA). The dissociated cells were plated on culture dishes coated with 1.5 μg ml−1 poly-l-ornithine (Sigma-Aldrich) and 1 μg ml−1 fibronectin (Asahi Glass, Tokyo, Japan) at 6 × 104 cells per cm2. The growth medium was changed every day. The cells were passaged at 70–80% confluency, and replated at the appropriate cell density onto 12-mm coverslips (Matsunami, Osaka, Japan) or dishes freshly coated with poly-l-ornithine and fibronectin, as described above. The cells were cultured with or without bFGF and used within two passages. To observe the effects of drug treatment, cells were treated with indicated amounts of corticosterone (CORT) for 48 h or 10 μM blebbistatin for 24 h. DMSO (dimethyl sulfoxide) was used as the control reagent.
NPCs cultured on coverslips were fixed with 4% paraformaldehyde and 15% saturated picric acid in PBS for 15 min at room temperature, and then permeabilized with blocking solution (0.1% Triton X-100, 2% bovine serum albumin (BSA) and 10% normal goat serum in PBS) for 30 min at 37 °C. The fixed cells were incubated with the primary antibodies in the Can Get Signal Immunostain Immunoreaction Enhancer Solution (Toyobo, Osaka, Japan) for 2 h at 37 °C, followed by the appropriate Alexa 488- or 568-conjugated secondary antibodies in the blocking solution for 1 h at 37 °C. To visualize actin filaments, Alexa 568-conjugated phalloidin (Invitrogen) was added with the secondary antibody solution. The stained cells were mounted in GEL/MOUNT mounting medium (Biomeda, Foster City, CA, USA) and observed under a microscope equipped with a charge-coupled device camera (BZ-9000, Keyence, Osaka, Japan) and using an oil-immersion × 60 (NA 1.4) and a × 10 (NA 0.45) objective, at room temperature. Fluorescent images were contrast-enhanced using Adobe Photoshop (Adobe Systems, San Jose, CA, USA).
BrdU (50 mg kg−1, Sigma-Aldrich) was intraperitoneally administered by a single injection into pregnant rats on E15.5 or E17.5.25 The brains were then dissected on the indicated embryonic days (E17.5, E18.5, E19.5, E21.5) or postnatal day (P3 and P7), and stored overnight in 4% paraformaldehyde. The brains were then embedded in paraffin and cut into coronal sections (3–5 μm thick) on a microtome (Yamato Kohki Industrial, Saitama, Japan). The sections were deparaffinized using standard xylene and alcohol procedures and denatured in 4 N HCl (Wako, Tokyo, Japan) at room temperature for 30 min. The sections were digested with 0.1% pepsin (Wako) for 20 min, washed with water, and then rinsed with TBS buffer. The sections were blocked in 1% BSA at 37 °C for 30 min, incubated overnight with anti-BrdU antibodies (1:250) at 4 °C and labeled with streptavidin–biotin and by the enhanced tyramide signal amplification technique.26, 27 The frontal cortex was observed under a microscope equipped with a charge-coupled device camera (BZ-9000) and × 10 objective.
Tissue samples stained with anti-CaD antibody were fixed with modified Bouin’s fluid (15% saturated picric acid and 5% formaldehyde in PBS) at 4 °C for 6 hours. Other samples were fixed with 4% paraformaldehyde in PBS at 4 °C overnight. After fixation, the brains were cryoprotected in 25% sucrose in PBS and embedded in OCT compound (Sakura Finetek, Torrance, CA, USA). Blocks were cut into coronal sections (30-μm thick) on a Cryomicrotome (Leica, Wetzlar, Germany). The sections were blocked in 2% skim milk in PBS containing 0.5% Tween 20 for 1 h at room temperature, then incubated with anti-GFP (1:200) or anti-CaD (1:200) antibodies diluted in Can Get Signal Immunostain Solution. Secondary antibodies with or without propidium iodide (PI, 1:1000, Invitrogen) were applied for 1 h at room temperature. The fluorescent images were observed under a microscope (BZ-9000) using a × 10 (NA 0.45) objective. Confocal images were captured using an LSM 5 PASCAL laser-scanning microscope (Carl Zeiss, Oberkochen, Germany) with × 10 (NA 0.3) and × 63 (NA 1.4) objectives. In Figure 5e, the full-focused images were generated by Keyence application. The other images shown represent single optical sections. Fluorescent images were contrast-enhanced using Adobe Photoshop.
For time-lapse imaging, NPCs were passaged and plated onto a 35-mm culture dish or 12-mm coverslip. The culture dish was transferred to a heated stage and observed under an Axiovert 200 M microscope (Carl Zeiss). The images were captured for an hour at intervals of 2 or 3 min. To trace the cells’ migration, the frames showing migrating cells were captured manually, and the paths were analyzed using DIAS software (Soll Technologies, Iowa City, IA, USA).
Western blot analysis
The cell and whole cortex lysates were separated by SDS-PAGE, and then transferred onto a nitrocellulose membrane. The membrane was incubated with primary antibody diluted in 5–10% non-fat dry milk in TBS containing 0.1% Tween 20 (TBS-T), followed by incubation with the appropriate HRP-conjugated secondary antibody. Image J (NIH) software was used to quantify the band intensity, and the value was normalized to the expression level of α-tubulin or GAPDH in each sample.
The total RNAs were isolated using TRIzol reagent (Invitrogen), and the quality was determined using an RNA 6000 nano chip in a Bioanalyzer (Agilent, Santa Clara, CA, USA). The cDNAs were synthesized with the GeneChip T7-Oligo (dT) Promoter Primer Kit (Affymetrix, Santa Clara, CA, USA) and TaKaRa cDNA Synthesis Kit (Takara Bio, Shiga, Japan) from 3 μg total RNA. Biotinylated cRNAs were synthesized with the IVT Labeling Kit (Affymetrix). Following fragmentation, 10 μg of the cRNAs were hybridized for 16 h at 45 °C to the GeneChip Rat Genome 230 2.0 Array. The GeneChip slides contain 31 000 probe sets, which represent 28 000 substantiated rat genes. The GeneChips were washed and stained in the Affymetrix Fluidics Station 450, then scanned using a GeneChip Scanner 3000 7G. A Single Array Analysis was performed by Microarray Suite version 5.0 (MAS5.0) with the Affymetrix default setting, and global scaling was used as the normalization method. The trimmed mean target intensity of each array was arbitrarily set to 500. After normalization, t-tests from three independent experiments were carried out to identify genes with more than a 2.0-fold increase or 50% decrease in expression with P<0.05. The identified genes are summarized in Supplementary Table 1.
Quantitative real-time PCR (real-time qPCR) and RT-PCR
The total RNAs were extracted from NPCs using TRIzol reagent and reverse-transcribed with MultiScribe Reverse Transcriptase (Applied Biosystems, Carlsbad, CA, USA). The cDNAs were amplified with gene-specific primer pairs using SYBR GreenER qPCR SuperMix Universal reagent (Invitrogen) in the real-time qPCR, and using PrimeStar DNA polymerase (Takara Bio) in the RT-PCR. The quantities measured by real-time qPCR and RT-PCR were normalized to the expression level of GAPDH in each sample. The primer sequences used in the real-time qPCR are listed in Supplementary Table 3.
Data are presented as the mean±s.d. Statistical analyses were performed using Student's t-test, and differences were considered statistically significant at P<0.05.
GenBank accession numbers
The microarray data discussed in this publication have been deposited in NCBI's Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO Series accession number GSE 11478.
Dexamethasone-induced retardation of radial migration during development of the cerebral cortex
To investigate the effects of excess GC on brain development, we monitored neuronal migration of post-mitotic immature neurons in both the medial and lateral sides of the cortices by labeling NPCs with a mitotic marker, BrdU (Figure 1 and Supplementary Figure 1). BrdU was administered to pregnant rats on gestational day 15.5 or 17.5 (corresponding to embryonic day (E) 15.5 or 17.5 of embryos). We used a synthetic compound, DEX, for the in vivo experiments, because it has a high affinity for the GR and is resistant to degradation by placental 11-β-hydroxysteroid dehydrogenase type 2 (11βHSD2).28 As the GR mRNA is reported to be highly expressed throughout the brain from midgestation,29 DEX (200 μg kg−1)30 was injected into pregnant rats daily during the last week of gestation (from E14.5 to E20.5). In this period, cell proliferation of NPCs occurs in the VZ, and post-mitotic immature neurons actively migrate to the CP.17 When BrdU was administered on E15.5, the medial side of the cortex was observed on days E16.5 to E21.5. Overall, examination of the control (vehicle-injected) cortices showed that the BrdU-labeled cells migrated radially from the VZ to the CP during development (Figures 1a and b). Specifically, by E17.5, 46.5±13.2% of the BrdU-positive cells (n=7) had accumulated in the CP of the control animals, and this accumulation continued at later time points. In contrast, DEX treatment temporarily induced the retardation of radial migration on E17.5 (22.6±7.7% of the BrdU-positive cells in the CP, n=6).
When BrdU was administered on E17.5, DEX treatment induced a marked retardation of the radial migration from E21.5 to postnatal day (P) 3 (Figures 1c and d). This prolonged DEX-induced retardation might have been due to the growth of the cortical mantle during development. We further analyzed the lateral side of the cortex. The DEX-induced retardation of radial migration was also seen (Supplementary Figure 1). Together, these results indicate that DEX treatment temporarily retards radial migration at the boundary between the intermediate zone (IZ)/subventricular zone (SVZ) and the CP throughout prenatal and postnatal development of the cerebral cortex along the medio–lateral axis, rather than just during a limited window of development.
Identification of CaD as a main cytoskeletal protein among GC-responsive targets
To understand the mechanism of GC-induced retardation of radial migration, we examined the effect of GC on cultured NPCs. Almost all of the NPCs (>99%) were nestin-positive. As confirmed by the expression of neural markers, doublecortin and βIII-tubulin, about 60–70% of NPCs were able to differentiate into post-mitotic neurons during culture (data not shown). In the following cell culture experiments, we used CORT as a rodent GC. As detected by immunocytochemistry, 92.0±3.1% of NPCs (n=3) were GR-positive with varying degrees of staining intensity (Supplementary Figure 2). About 80% (78.5±3.6%, n=6) of the control (DMSO-treated) cells had a bipolar shape and migrated linearly (Figures 2a–c and Supplementary Movie 1). It is important to note that more than half of the CORT-treated cells (55.6±1.7%, n=6) showed phenotypic changes in the bipolar-to-multipolar transition and a random migratory path. In particular, the tips of the multipolar processes appeared to have dynamic neurite growth endings (webfoot-like structures) (Figures 2a–c and Supplementary Movie 1).
To identify the genes that induced these phenotypic changes, we screened for GC-responsive genes by microarray analysis, real-time qPCR and western blot. In the microarray analysis, genes were considered GC-responsive if their expression levels were at least 2-fold higher or 50% lower than in the absence of GC. Using these criteria, a genome-wide analysis revealed large changes in the gene expression profile of the CORT-treated cells, with 195 genes upregulated and 16 downregulated (Supplementary Table 1).
Among the GC-responsive genes, we selected those encoding cytoskeletal proteins and confirmed their expression levels by real-time qPCR. We found the upregulation of seven genes; they encoded myosin 1e, CaD, ezrin (villin 2), gelsolin, plastin 3, doublecortin-like kinase 1 and adducin γ (Supplementary Table 2). There are two different molecular weight (Mr) isoforms of CaD: high Mr (h-CaD) and low Mr CaD (l-CaD),31 which are generated from a single gene by alternative splicing.32 As NPCs and neurons only express l-CaD, we refer to l-CaD as CaD in this study. Western blot showed that the CaD (2.8±0.6 fold, n=3) and ezrin (1.9±0.5 fold, n=3) proteins were significantly upregulated by CORT treatment, whereas the changes in the protein levels of doublecortin-like kinase 1, gelsolin and adducin γ were less significant (Figure 2d). Neither the plastin 3 nor myosin 1e proteins were detected by western analysis (data not shown). Although the CaD and ezrin proteins were upregulated by CORT treatment, ezrin did not directly participate in neuronal migration in vivo, as shown later (Supplementary Figure 5). We therefore focused on CaD in the following experiments.
CORT treatment upregulated the CaD protein in a dose-dependent manner (Figure 2e). A half-maximal increase in CaD expression was achieved with about 200 nM CORT, which corresponds to the circulating levels of rodent CORT under severely stressful conditions.33 Thus, pathological levels of CORT can upregulate CaD expression.
Immunocytochemistry showed that CaD and F-actin in the control cells were co-localized in the cytosol and the tips of the leading but not trailing processes. In CORT-treated cells, both proteins were localized in the cytosol and concentrated in the neurite growth endings at the tips of the multipolar processes (Figures 2f and g). The immunostaining intensity of CaD was increased in the cells located in the VZ, IZ and CP of the cortex of DEX-treated animals compared with control animals (Figure 2h). In both cases, the blood vessels were densely stained with the anti-CaD antibody. Western blot showed that DEX treatment induced a small but significant increase in CaD protein expression in the lysates of the cerebral cortex (Supplementary Figure 4). As CaD is highly expressed in vascular smooth muscle cells and pericytes,31 this small increase in CaD protein expression may reflect a limited population of GC-responsive cells in addition to higher background levels of CaD protein by containing numerous blood vessels in the cerebral cortex. Higher magnification images of the IZ showed multipolar cells in both the control and DEX-treated cortices. Like the CORT-treated cells in culture (Figures 2f and g), multipolar cells in the DEX-treated but not control cortices had the similar neurite growth endings at the tips of their multipolar processes (Figure 2i). Thus GC treatment induced the same morphological changes in cell culture and in vivo.
GR-dependent transcription of the CALD1 gene
As excess GC treatment upregulated the CaD expression at both the mRNA and protein levels (Supplementary Table 2 and Figures 2d and e), we analyzed the GC-dependent transcription of the CaD-encoding gene, CALD1, using its rat promoter. There are two kinds of promoter regions in the rat CALD1 gene: the fibroblast- and HeLa-types. We recently found the GR-mediated transcription of the human CALD1 gene in lung cancer cells.23 Consistent with this finding, promoter analysis and semi-quantitative RT-PCR in NPCs showed that the fibroblast-type promoter yielded higher activity and more abundant transcripts than the HeLa-type one (Supplementary Figures 3a and b). We therefore examined the fibroblast-type promoter as follows. Within the promoter region, a CArG box and two GC-response element (GRE) sequences are highly conserved among mammalian species (Supplementary Figure 3c). We performed the promoter analysis of the CALD1 gene using a series of its promoter constructs (Supplementary Figure 3d). A constitutively active form of GR (GRΔC), in which the carboxy-terminal domain was deleted, markedly enhanced the promoter activity (4.6±0.6 fold, n=3) (Supplementary Figure 3e). Although the CArG box of the CALD1 promoter is critical for SRF-dependent transcription,34 it was dispensable for the GR-dependent transcription (Supplementary Figure 3e). Mutation of either GRE1 or GRE2 significantly reduced the GRΔC-enhanced promoter activity, and a double mutation (GRE1/GRE2) completely abolished it (Supplementary Figure 3e). In vitro DNA-binding assays showed that although both the biotinylated GRE1 and GRE2 oligo-DNAs bound to GR, GRE1 had a higher affinity for GR than GRE2 (Supplementary Figure 3f). These data indicate that GR transactivates the CALD1 gene by binding directly to the GRE sequences within the rat CALD1 promoter.
Mechanism of the GC-induced changes in cell shape and migration
To examine the mechanism of the CaD-induced changes in cell shape and migration, GFP-fused CaD (GFP–CaD) was overexpressed in cultured NPCs. Similar to the CORT-treated cells, the GFP–CaD-expressing cells exhibited the bipolar-to-multipolar transition with dynamic neurite growth endings at the tips of their multipolar processes, and they also migrated randomly (Figures 3a and b and Supplementary Movie 2). It is well documented that the carboxy-terminal domain of CaD, which negatively regulates interactions between actin and myosin, has actin-, tropomyosin- and calmodulin-binding sites.31 Overexpression of the GFP-fused carboxy-terminal (GFP-C-CaD), but not amino-terminal (GFP-N-CaD), region of CaD induced the same phenotypic changes as CORT treatment and GFP–CaD overexpression (Figures 3a and b). Treatment with blebbistatin, a selective inhibitor of myosin II ATPase, also induced the bipolar-to-multipolar transition accompanied by formation of the dynamic neurite growth endings (Figures 3c and d and Supplementary Movie 3). In the control- and CORT-treated cells, myosin IIA was localized in the cytosol in addition to the tips of the leading and multipolar processes, respectively, along with F-actin. Myosin IIB was diffusely localized in the cytosol, not concentrated in the tips of processes, in both types of cells (Figures 3e and f). To examine the roles of the myosin II isozymes, we designed two specific short-interfering RNAs against myosin IIA or IIB. These siRNAs efficiently reduced the expression of endogenous myosin IIA and IIB, respectively (Figure 3g). Compared with control siRNA-transfected cells, a population of multipolar cells with dynamic neurite growth endings at their multipolar processes increased in the myosin IIA- or IIB-depleted cells. These changes were further enhanced by their double depletion (Figures 3h and i). These data suggest that CaD affects the cell shape and migration of NPCs by downregulating myosin IIA and/or IIB functions.
A critical role of CaD in radial migration in vivo
We further examined the involvement of CaD in neuronal migration in vivo. In HEK293T cells, two expression plasmids, respectively encoding CaD miRNAs no. 1 and no. 2 fused with GFP, and effectively suppressed the expression of exogenous CaD protein (Figure 4a). Using in utero electroporation, these expression plasmids were introduced into NPCs located at the ventral surface of the rat brain at E15.5. Neuronal migration was monitored by the GFP-positive cells. Compared with the control miRNA alone, both CaD miRNAs no. 1 and no. 2 markedly impaired radial migration (Figures 4b and c). Furthermore, the GFP-positive cells transfected with CaD miRNAs no. 1 and no. 2, which remained in the IZ, showed a suppressed expression of endogenous CaD (Figure 4d). These results suggest that CaD is critical for radial migration in vivo.
Retardation of radial migration by CaD overexpression in vivo
To investigate the effect of CaD overexpression on neuronal migration in vivo, we also used in utero electroporation to introduce expression plasmids encoding GFP or GFP–CaD into NPCs of the rat brain at E15.5 (Figures 5a and b) or E17.5 (Figures 5c and d). When expression plasmids were introduced into the E15.5 brain, the GFP–CaD-expressing cells markedly retarded radial migration at E17.5-18.5. Nearly 60% of these cells were retained in the IZ (58.2±12.1% and 67.1±9.5% of the GFP–CaD-expressing cells in the IZ at E17.5 and E18.5, respectively). When expression plasmids were similarly introduced into the E17.5 brain, the retardation of radial migration was apparent around E21.5-P3. In both cases, the retardation was temporary, later returning to the normal rate (Figures 5a–d). Although both GFP- and GFP–CaD-expressing cells located in the IZ of the E17.5 and 18.5 cortices showed a multipolar shape, the GFP–CaD-expressing cells alone had the neurite growth endings at the tips of their multipolar processes (Figure 5e). These data indicate that CaD overexpression precisely mimics the DEX-induced retardation of radial migration as revealed by BrdU labeling (Figure 1 and Supplementary Figure 1). Although the ezrin protein expression in cultured NPCs was upregulated by CORT treatment (Figure 2d), overexpression of ezrin-GFP showed no changes in their radial migration in vivo (Supplementary Figure 5). Taken together, these data indicate that among the GC-responsive cytoskeletal proteins examined, CaD is the prime target involved in GC-induced abnormality of radial migration in vivo.
Despite numerous studies showing the detrimental effects of stress-triggered GCs on brain development as described in the Introduction section, its underlying mechanisms remain unclear. In this study, we showed that the exposure to excess GC induced the retardation of radial migration during brain development (Figure 1 and Supplementary Figure 1). We screened for GC-responsive genes encoding cytoskeletal proteins, and identified the CALD1 gene as a main target of GCs (Figure 2d, Supplementary Tables 1 and 2). Indeed, GC upregulated the CaD expression in the developing cerebral cortex in vivo (Figures 2h and i and Supplementary Figure 4). The upregulation of CaD was mediated by GR-dependent transactivation of the CALD1 genes (Supplementary Figures 2d–f). The overexpression of CaD in cultured NPCs also caused them to adopt GC-induced phenotypic changes in cell shape and migration (Figures 3a and b). GC-induced phenotypic changes depended on the CaD-mediated negative regulation of myosin II function (Figures 3c–i). In utero electroporation experiments revealed that CaD depletion by RNA interference impaired radial migration (Figure 4), indicating that CaD is critical for normal radial migration. CaD overexpression in vivo retarded radial migration during corticogenesis (Figure 5), mimicking the GC-induced retardation of radial migration. Taken together, these results suggest that excess GC-induced upregulation of CaD is critically involved in the retardation of radial migration throughout cortical development. Heine and Rowitch recently35 reported that hedgehog signaling antagonized the harmful effect of GCs, such as proliferation and apoptosis of neural stem/progenitor cells, by induction of 11βHSD2 in the developing mouse cerebellum. In their report, 11βHSD2 antagonized the effects of GCs, such as CORT, hydrocortisone and prednisolone, but not DEX. However, the mechanism of the action of GCs on the hedgehog-signaling pathway was not identified. As demonstrated here, DEX and CORT markedly induced the retardation of radial migration in vivo and the phenotypic changes of cultured NPCs in cell shape and migration. Our present study regarding the detrimental effects of excess GCs on radial migration may, therefore, be different from their report.
CaD is one of the major components of smooth muscle thin filaments,36 and it negatively regulates interactions between actin and myosin.31 In non-muscle cells, CaD also stabilizes actin filaments in addition to regulating actin–myosin interactions,31 implicating a vital role in cell motility in a wide variety of cells. It has been demonstrated that the transcription of the CALD1 gene in smooth muscle and non-muscle cells is SRF-dependent.34, 37 Although GCs were reported to induce changes in the expression of several cytoskeletal proteins, their expressional regulation by GCs remained unclear.38, 39, 40 Most recently, we found the GR-mediated transcription of the human CALD1 gene in lung carcinoma cells.23 In this study, using genome-wide analysis, we identified the CALD1 gene as a main target among GC-responsive genes encoding cytoskeletal proteins in NPCs (Supplementary Tables 1 and 2 and Figure 2). Similar to the human CALD1 gene in lung carcinoma cells,23 the GC-induced upregulation of CaD in NPCs could be mediated by the GR-dependent transcription of the rat CALD1 gene (Supplementary Figures 3e and f). Several studies performed genome-wide analysis for screening stress/GC-responsive genes in the brain.41, 42 There are, however, no indications about expression changes of the CALD1 gene, because these studies used the tissue samples. As demonstrated in Figure 2, expression changes of CaD are restricted to the population of GC-responsive cells and the tissue samples contain numerous blood vessels that express the exclusively high levels of CaDs (h- and l-CaD).31 When the tissue sample is used, GC-induced expression changes of CaD in the whole cerebral cortex are significant but small (Supplementary Figure 4). Thus, the NPCs used here (more than 90% of GR-positive cells; Supplementary Figure 2) are beneficial to identify the ubiquitously expressed genes using genome-wide analysis.
In non-neuronal cells including lung carcinoma cells, the upregulation of CaD via either GR- or SRF-dependent transcription enhances the formation of thick stress fibers and focal adhesion.34, 43 These studies suggest the stabilization of actin filaments by the upregulated CaD, which result in the suppression of cell migration. In contrast, GC-induced upregulation of CaD in NPCs exhibited different phenotypes with the bipolar-to-multipolar transition and random migration (Figures 2a–c). The similar phenotypic changes were also found in the developing cortex (Figures 2h and i). These changes were caused by the suppression of myosin II function by the upregulation of CaD in cell culture experiments (Figures 3c–i). It has been demonstrated in non-neuronal cells that myosin IIA plays an important role in cellular contractility and promoting microtubule dynamics mediated through the Rac/Tiam1 pathway.44 However, we were not able to detect the activation of Rac in both the CORT- and blebbistatin-treated NPCs (Supplementary Figure 6). Thus the GC-induced upregulation of CaD in NPCs might negatively regulate the function of myosin II in a Rac/Tiam1-independent manner.
Recent studies have demonstrated that during development of the cerebral cortex, migrating cells become multipolar within the IZ and SVZ, and then display the multipolar-to-bipolar transition just before migrating radially from the IZ/SVZ to the CP.45, 46 In addition, studies using in utero electroporation and overexpression of a plasmid or RNAi suggest that cytoskeletal proteins such as filamin A,47 LIS148 and doublecortin49 are, at least in part, involved in the multipolar-to-bipolar transition. Mutation of these genes causes periventricular nodular heterotopia and lisencephaly.18 Cdk5 phosphorylates a number of cytoskeletal proteins.50 Cortex-specific Cdk5 conditional knockout mice shows that it is required for proper multipolar-to-bipolar transition.51 Thus, the loss or gain of the functions of these proteins impairs both radial migration and the multipolar-to-bipolar transition in the developing cortex. As described here, the GC-induced upregulation of CaD and overexpression of GFP–CaD in vivo temporarily retarded the radial migration at the boundary between the IZ/SVZ and the CP (Figures 1, 2, 5 and Supplementary Figure 1). In this study, DEX-treated and GFP–CaD-overexpressing cortices were followed up until P7. In both the vehicle- and DEX-treated cortices, most migrating cells had similarly reached the CP on P7 (Figure 1 and Supplementary Figure 1). Compared with the GFP-transfected cortices, only a few GFP–CaD-overexpressing cells remained in the lower region of the cortex (P7 in Figures 5c and d). This may be due to the excessive expression of CaD. The neurite growth endings as detected in cell culture experiments (Figures 2 and 3) were also found in the multipolar cells retained in the IZ of the DEX-treated and GFP–CaD-overexpressing cortex in vivo (Figure 5e). Therefore, our present findings suggest that the CaD-linked negative regulation of myosin II is responsible for a delay in the multipolar-to-bipolar transition in vivo, resulting in the transient retardation of radial migration at the boundary between the IZ/SVZ and the CP.
In experimental animals, excessive stress/GC exposure at the perinatal stage impairs the brain development, and results in abnormal behavior in adult offspring.1, 8, 9, 11 Excessive GC exposure triggered by stress or medications during pregnancy in humans is also implicated in the etiology of psychiatric disorders.1, 2, 3, 12, 15 The mechanism underlying these persistent effects of GCs on the brain structure and function is, however, unclear, except for the epigenetic regulation of GR expression.52 Our present results, in which excessive GC-induced upregulation of CaD temporarily retards the radial migration during development of the cerebral cortex, may be a critical contributor to the abnormality of neural network formation, with life-long consequences. Thus, our findings provide a novel insight into the detrimental effects of GCs on brain development in vivo, particularly as they relate to the risk of psychiatric disorders, and may lead to the establishment of new therapies for stress/GC-induced neurodevelopmental disorders.
Conflict of interest
The authors declare no conflict of interest.
McEwen BS . Glucocorticoids, depression, and mood disorders: structural remodeling in the brain. Metabolism 2005; 54: 20–23.
Manji HK, Drevets WC, Charney DS . The cellular neurobiology of depression. Nat Med 2001; 7: 541–547.
Koenig JI, Kirkpatrick B, Lee P . Glucocorticoid hormones and early brain development in schizophrenia. Neuropsychopharmacology 2002; 27: 309–318.
De kloet ER, Joëls M, Holsboer F . Stress and the brain: from adaptation to disease. Nat Rev Neurosci 2005; 6: 463–475.
McEwen BS . Stress and hippocampal plasticity. Annu Rev Neurosci 1999; 22: 105–122.
Cerqueira JJ, Pêgo JM, Taipa R, Bessa JM, Almeida OF, Sousa N . Morphological correlates of corticosteroid-induced changes in prefrontal cortex-dependent behaviors. J Neurosci 2005; 25: 7792–7800.
Mitra R, Sapolsky RM . Acute corticosterone treatment is sufficient to induce anxiety and amygdaloid dendritic hypertrophy. Proc Natl Acad Sci USA 2008; 105: 5573–5578.
Bakker JM, Bel FV, Heijnen CJ . Neonatal glucocorticoids and the developing brain: short-term treatment with life-long consequences? Trends Neurosci 2001; 24: 649–653.
Stéphane VS, Borradori-Tolsa C, Vauthay DM, Lodygensky G, Lazeyras F, Hüppi PS . Impact of intrauterine growth restriction and glucocorticoids on brain development: insights using advanced magnetic resonance imaging. Mol Cell Endocrinol 2006; 254-255: 163–171.
Weinstock M . The long-term behavioural consequences of prenatal stress. Neurosci Biobehav Rev 2008; 32: 1073–1086.
Becker JB, Monteqqia LM, Perrot-Sinal TS, Romeo RD, Taylor JR, Yehuda R et al. Stress and disease: is being female a predisposing factor? J Neurosci 2007; 27: 11851–11855.
Phillips NK, Hammen CL, Brennan PA, Najman JM, Bor W . Early adversity and the prospective prediction of depressive and anxiety disorders in adolescents. J Abnorm Child Psychol 2005; 33: 13–24.
Flagel SB, Vázquez DM, Watson Jr SJ, Neal Jr CR . Effects of tapering neonatal dexamethasone on rat growth, neurodevelopment, and stress response. Am J Physiol Regul Integr Comp Physiol 2002; 282: R55–R63.
Huang WL, Beazley LD, Quinlivan JA, Evans SF, Newnham JP, Dunlop SA . Effect of corticosteroids on brain growth in fetal sheep. Obstet Gynecol 1999; 94: 213–218.
Modi N, Lewis H, Al-Naqeeb N, Ajayi-Obe M, Doré CJ, Rutherford M . The effects of repeated antenatal glucocorticoid therapy on the developing brain. Pediatri Res 2001; 50: 581–585.
Dehay C, Kennedy H . Cell-cycle control and cortical development. Nat Rev Neurosci 2007; 8: 438–450.
Ayala R, Shu T, Tsai LH . Trekking across the brain: the journey of neuronal migration. Cell 2007; 128: 29–43.
Gleeson JG, Walsh CA . Neuronal migration disorders: from genetic diseases to developmental mechanisms. Trends Neurosci 2000; 8: 352–359.
Lavado-Autric R, Ausó E, García-Velasco JV, Arufe Mdel C, Escobar del Rey F, Berbel P et al. Early maternal hypothyroxinemia alters histogenesis and cerebral cortex cytoarchitecture of the progeny. J Clin Invest 2003; 111: 1073–1082.
Ausó E, Lavado-Autric R, Cuevas E, Del Rey FE, Morreale De Escobar G, Berbel P . A moderate and transient deficiency of maternal thyroid function at the beginning of fetal neocorticogenesis alters neuronal migration. Endocrinology 2004; 145: 4037–4047.
Konno J, Yoshida S, Ina A, Ohmomo H, Shutoh F, Nogami H et al. Upregulated expression of neuropeptide Y in hypothalamic–pituitary system of rats by chronic dexamethasone administration. Neurosci Res 2008; 60: 259–265.
Wong EY, Herbert J . Roles of mineralocorticoid and glucocorticoid receptors in the regulation of progenitor proliferation in the adult hippocampus. Eur J Neurosci 2005; 22: 785–792.
Mayanagi T, Morita T, Hayashi K, Fukumoto K, Sobue K . Glucocorticoid receptor-mediated expression of caldesmon regulates cell migration via the reorganization of the actin cytoskeleton. J Biol Chem 2008; 283: 31183–31196.
Konno D, Yoshimura S, Hori K, Maruoka H, Sobue K . Involvement of the phosphatidylinositol 3-kinase/rac1 and cdc42 pathways in radial migration of cortical neurons. J Biol Chem 2005; 280: 5082–5088.
Li HP, Honma S, Miki T, Takeuchi Y, Kawano H . Multiple defects in the formation of rat cortical axonal pathways following prenatal X-ray irradiation. Eur J Neurosci 2005; 21: 1847–1858.
Giorno R . A comparison of two immunoperoxidase staining methods based on the avidin-biotin interaction. Diagn Immunol 1984; 2: 161–166.
Mishima T, Sakatani S, Hirase H . Intracellular labeling of single cortical astrocytes in vivo. J Neurosci Methods 2007; 166: 32–40.
Matthews SG . Antenatal glucocorticoids and the developing brain; mechanisms of action. Semin Neonatol 2001; 6: 309–317.
Diaz R, Brown RW, Seckl JR . Distinct ontogeny of glucocorticoid and mineralocorticoid receptor and 11β-Hydroxysteroid dehydrogenase types I and II mRNAs in the fetal rat brain suggest a complex control of glucocorticoid actions. J Neurosci 1998; 18: 2570–2580.
Slotkin TA, Kreider ML, Tate CA, Seidler FJ . Prenatal and postnatal periods for persistent effects of dexamethasone on serotonergic and dopaminergic systems. Neuropsychopharmacology 2006; 31: 904–911.
Sobue K, Sellers JR . Caldesmon, a novel regulatory protein in smooth muscle and nonmuscle actomyosin systems. J Biol Chem 1991; 266: 12115–12118.
Hayashi K, Yano H, Hashida T, Takeuchi R, Takeda O, Asada K et al. Genomic structure of the human caldesmon gene. Proc Natl Acad Sci USA 1992; 89: 12122–12126.
Bowers SL, Bilbo SD, Dhabhar FS, Nelson RJ . Stressor-specific alterations in corticosterone and immune responses in mice. Brain Behav Immun 2008; 22: 105–113.
Morita T, Mayanagi T, Sobue K . Dual roles of myocardin-related transcription factors in epithelial mesenchymal transition via slug induction and actin remodeling. J Cell Biol 2007; 179: 1027–1042.
Heine VM, Rowitch DH . Hedgehog signaling has a protective effect in glucocorticoid-induced mouse neonatal brain injury through an 11âHSD2-dependent mechanism. J Clin Invest 2009; 119: 267–277.
Sobue K, Muramoto Y, Fujita M, Kakiuchi S . Purification of a calmodulin-binding protein from chicken gizzard that interacts with F-actin. Proc Natl Acad Sci USA 1981; 78: 5652–5655.
Hayashi K, Nakamura S, Nishida W, Sobue K . Bone morphogenetic protein-induced MSX1 and MSX2 inhibit myocardin-dependent smooth muscle gene transcription. Mol Cell Biol 2006; 26: 9456–9470.
Castellino F, Heuser J, Marchetti S, Bruno B, Luini A . Glucocorticoid stabilization of actin filaments: a possible mechanism for inhibition of corticotropin release. Proc Natl Acad Sci USA 1992; 89: 3775–3779.
Antonow-Schlorke I, Schwab M, Li C, Nathanielsz PW . Glucocorticoid exposure at the dose used clinically alters cytoskeletal proteins and presynaptic terminals in the fetal baboon brain. J Physiol 2003; 547: 117–123.
Cereseto M, Reinés A, Ferrero A, Sifonios L, Rubio M, Wikinski S . Chronic treatment with high doses of corticosterone decreases cytoskeletal proteins in the rat hippocampus. Eur J Neurosci 2006; 24: 3354–3364.
Datson NA, Van der Perk J, De Kloet ER, Vreugdenhil E . Identification of corticosteroid-responsive genes in rat hippocampus using serial analysis of gene expression. Eur J Neurosci 2001; 14: 675–689.
Alfonso J, Pollevick GD, Van Der Hart MG, Flüqqe G, Fuchs E, Frasch AC . Identification of genes regulated by chronic psychosocial stress and antidepressant treatment in the hippocampus. Eur J Neurosci 2004; 19: 659–666.
Morita T, Mayanagi T, Yoshio T, Sobue K . Changes in the balance between caldesmon regulated by p21-activated kinases and the Arp2/3 complex govern podosome formation. J Biol Chem 2007; 282: 8454–8463.
Even-Ram S, Doyle AD, Conti MA, Matsumoto K, Adelstein RS, Yamada KM . Myosin IIA regulates cell motility and actomyosin-microtubule crosstalk. Nat Cell Biol 2007; 9: 299–309.
Tabata H, Nakajima K . Multipolar migration: the third mode of radial neuronal migration in the developing cerebral cortex. J Neurosci 2003; 23: 9996–10001.
Noctor SC, Martínez-Cerdeño V, Ivic L, Kriegstein AR . Cortical neurons arise in symmetric and asymmetric division zones and migrate through specific phases. Nat Neurosci 2004; 7: 136–144.
Nagano T, Yoneda T, Hatanaka Y, Kubota C, Murakami F, Sato M . Filamin A-interacting protein (FILIP) regulates cortical cell migration out of the ventricular zone. Nat Cell Biol 2002; 4: 495–501.
Tsai JW, Chen Y, Krieqstein AR, Vallee RB . LIS1 RNA interference blocks neural stem cell division, morphogenesis, and motility at multiple stages. J Cell Biol 2005; 170: 935–945.
Bai J, Ramos RL, Ackman JB, Thomas AM, Lee RV, LoTurco JJ . RNAi reveals doublecortin is required for radial migration in rat neocortex. Nat Neurosci 2003; 6: 1277–1283.
Xie Z, Tsai LH . Cdk5 phosphorylation of FAK regulates centrosome-associated miocrotubules and neuronal migration. Cell Cycle 2004; 3: 108–110.
Ohshima T, Hirasawa M, Tabata H, Mutoh T, Adachi T, Suzuki H et al. Cdk5 is required for multipolar-to-bipolar transition during radial neuronal migration and proper dendrite development of pyramidal neurons in the cerebral cortex. Development 2007; 134: 2273–2282.
Meaney MJ, Szyf M . Maternal care as a model for experience-dependent chromatin plasticity? Trends Neurosci 2005; 28: 456–463.
We thank Dr S Furukawa (Gifu Pharmaceutical University) for advice on the in utero electroporation technique. This research was supported by Grant-in-Aids for Scientific Research (20240038) from the Ministry of Education, Culture, Sports, Science and Technology of Japan (KS).
About this article
Cite this article
Fukumoto, K., Morita, T., Mayanagi, T. et al. Detrimental effects of glucocorticoids on neuronal migration during brain development. Mol Psychiatry 14, 1119–1131 (2009) doi:10.1038/mp.2009.60
- radial migration
- brain development
- in utero electroporation
Region-specific control of microglia by adenosine A2A receptors: uncoupling anxiety and associated cognitive deficits in female rats
Adverse neuropsychiatric development following perinatal brain injury: from a preclinical perspective
Pediatric Research (2019)
Frontiers in Cell and Developmental Biology (2019)
Neuroscience Letters (2019)
Brain and Behavior (2018)