Disrupted-In-Schizophrenia 1, a candidate gene for schizophrenia, participates in neurite outgrowth

Abstract

Disrupted-In-Schizophrenia 1 (DISC1) was identified as a novel gene disrupted by a (1;11)(q42.1;q14.3) translocation that segregated with schizophrenia in a Scottish family. Predicted DISC1 product has no significant homology to other known proteins. Here, we demonstrated the existence of DISC1 protein and identified fasciculation and elongation protein zeta-1 (FEZ1) as an interacting partner of DISC1 by a yeast two-hybrid study. FEZ1 and its nematode homolog are reported to represent a new protein family involved in axonal outgrowth and fasciculation. In cultured hippocampal neurons, DISC1 and FEZ1 colocalized in growth cones. Interactions of these proteins were associated with F-actin. In the course of neuronal differentiation of PC12 cells, upregulation of DISC1/FEZ1 interaction was observed as along with enhanced extension of neurites by overexpression of DISC1. The present study shows that DISC1 participates in neurite outgrowth through its interaction with FEZ1. Recent studies have provided reliable evidence that schizophrenia is a neurodevelopmental disorder. As there is a high level of DISC1 expression in developing rat brain, dysfunction of DISC1 may confer susceptibility to psychiatric illnesses through abnormal development of the nervous system.

Main

Schizophrenia is a debilitating mental disease that affects about 1% of the population. Like many other psychiatric disorders, schizophrenia is thought to involve the combined effects of multiple genetic components.1,2 Research, such as linkage analyses and association studies,3,4,5,6,7,8,9,10,11,12 has not yet identified definitive genes responsible for the disease.

In a large Scottish family, a balanced (1;11)(q42.1;q14.3) translocation that segregated with schizophrenia and affective disorders with an LOD score of 7.1 was found.13,14 Disrupted-In-Schizophrenia-1 (DISC1) on chromosome 1 was identified as a novel gene disrupted by this translocation.13 Family members exhibited no distinctive features by which the psychiatric phenotype could be distinguished from unrelated cases.14 Moreover, translocation carriers showed a significant reduction in the amplitude of the P300 event-related potential, which was also observed in unrelated patients with schizophrenia.14 These findings suggest that disruption of the function of this gene may confer susceptibility to these mental disorders. A linkage report15 also indicated 1q42 as a possible locus for schizophrenia in a study of Finnish families.

In the present study, we performed expression and functional analysis of DISC1 to elucidate the pathophysiological role of this candidate gene for schizophrenia. We first showed the enhanced level of DISC1 expression in rat brain at a developing stage, and then the existence of DISC1 protein was revealed by an antibody raised against the predicted amino-acid sequence. Furthermore, we identified fasciculation and elongation protein zeta-1 (FEZ1) as an interacting partner of DISC1. The interaction of DISC1 and FEZ1 was associated with actin cytoskeleton and upregulated during neurite outgrowth.

Materials and methods

In situ hybridization

A mouse homolog of DISC1 with a conserved surrounding genomic structure was identified on chromosome 8 and partial cDNAs of mouse and rat homolog were revealed. A fragment of rat cDNA that corresponds to 450–1174 nucleotides (nt) of human cDNA was obtained by PCR using the primers 5′-IndexTermGGACAGTGGTTGTCGGCAAGA and 5'-IndexTermAGGGCAGG-CGGCCTTTCTCCTGTTCCAG, and then subcloned into pGEM-T (Promega, Madison, WI, USA). Digoxigenin-labeled cRNA probes (antisense and sense) were generated by in vitro transcription using the cDNA fragment as a template in the presence of digoxigenin-labeled dUTP (Roche, Sydney, Australia). Hybridization and posthybridization procedures were performed as described.16

Yeast two-hybrid screening

The human DISC1 C-terminal domain (amino acids 348–854) was cloned into pAS2-1 (GAL4 DNA-binding domain vector, Clontech, Palo Alto, CA, USA) as bait. Yeast strain AH109 was transformed with the bait plasmid, mated with strain Y187 pretransformed with a human adult brain cDNA library (Clontech) and plated on a quadruple dropout medium (-Ade,-His,-Leu,-Trp). The screening procedure accompanied with an α-galactosidase assay was performed as described (Clontech Pretransformed MATCHMAKER Libraries User Manual). To determine the regions involved in the interaction, AH109 was cotransformed with truncated forms of human DISC1 and human FEZ1 subcloned into pAS2-1 or pACT2 (GAL4 activation domain vector, Clontech), respectively, and then assayed.

Plasmids

Full-length human DISC1 cDNA and its splicing variant form were cloned into pcDNA3.1(+). (Invitrogen, Carlsbad, CA, USA) and used in the Western blot analysis for the detection of endogenous DISC1. Human DISC1 cDNAs coding full protein, FEZ1-binding region (amino acids 446–633) and deleted protein that lacks the binding region were tagged with FLAG sequence at 3' end. Human FEZ1 cDNA was tagged with HA sequence at 3' end. These tagged constructs were cloned into pcDNA3.1(+) and used in the immunoprecipitation assay. Human DISC1 cDNA was also cloned into pEGFP-N1 (Clontech) and used in the immunocytochemical analysis. Human DISC1 cDNA coding FEZ1-binding region was also cloned into a bicistronic expression vector, pIRES2-EGFP (Clontech), and transfected to stable PC12 cells.

Cell culture and transfection

HEK293T cells, SK-N-SH cells and PC12 cells were cultured in DMEM containing 10% fetal calf serum (FCS), αMEM/10% FCS and DMEM/10% horse serum/5% FCS, respectively. Hippocampal neurons were prepared from embryonic day 18 Wistar rats as described17 and cultured in DMEM/10% FCS for 24 h. The medium was then replaced with DMEM/B27 supplement (Invitrogen). For the generation of PC12 cells stably expressing DISC1, FLAG-tagged human DISC1 cDNA in pcDNA3.1(+) was linearized by ScaI and transfected to PC12 cells. At 48 h after transfection, Geneticin (Invitrogen) was added at a concentration of 400 μg/ml. Several clones were picked, expanded in the selective medium and then checked for the expression. Mock-stable lines were also generated by the same procedures. For neuronal differentiation, PC12 cells were starved for serum for 4 h and then treated with nerve growth factor (NGF) at a concentration of 50 ng/ml. For the transfection of cells mentioned above, Lipofectamine 2000 (Invitrogen) was used according to the manufacturer's instructions.

Antibodies

Rabbit anti-human DISC1, anti-rat DISC1 and anti-FEZ1 polyclonal antibodies were raised against SCMTAGVHEAQA of human DISC1, RTPHPEE-EKSPLQ of rat DISC1 and KVPTLLTDYILKVL of human and rat FEZ1, respectively, and affinity-purified. Monoclonal anti-FLAG (Sigma-Aldrich, St Louis, MI, USA) and polyclonal anti-HA (Santa Cruz Biotechnology, Santa Cruz, CA, USA) antibodies were used in immunoprecipitation assays. Monoclonal anti-actin antibody (Chemicon, Temecula, CA, USA) was used in Western blot analysis.

Western blot analysis

Cells were homogenized in TNE buffer (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA) containing 1% NP40 in the presence of protease inhibitors, incubated on ice for 1 h and centrifuged at 15 000 g for 20 min. Lysates were boiled with SDS sample buffer for 5 min, subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to the PVDF membrane. After blocking with 5% membrane blocking agent (Amersham Biosciences, Buckinghamshire, UK), the membrane was incubated with the primary antibody for 12 h at 4°C. For the detection of human DISC1, rat DISC1 and FEZ1, antibodies raised against these proteins were used at 1 : 500, 1 : 500 and 1 : 250 dilutions, respectively. The membrane was then incubated with anti-rabbit or mouse IgG, HRP-linked antibody (Cell Signaling Technology, Beverly, MA, USA) at 1 : 10 000 dilution for 1 h at room temperature. Immunoblotting was visualized by chemiluminescence using the ECL kit (Amersham Biosciences).

Immunohistochemistry

Adult rats under deep pentobarbital anesthesia were perfused transcardially with 100–300 ml of Zamboni solution. Brain was removed and infused with 30% sucrose overnight at 4°C. Free-floating tissue sections were incubated in 3% hydrogen peroxide for 30 min to block endogenous peroxidase activity. After blocking with 5% bovine serum albumin, purified anti-rat DISC1 antibody was applied at a 1 : 100 dilution overnight at 4°C. Primary antibody was detected using the avidin–biotin methods (Vectastain ABC kit; Funakoshi, Tokyo, Japan) and visualized with diaminobenzidine (DAB). Stained sections were mounted on gelatine-coated glass slides, dehydrated in ethanol and observed by light microscopy.

Immunocytochemistry

Cells were fixed with 4% paraformaldehyde and permeabilized with 0.3% triton X-100. After blocking with 3% bovine serum albumin, purified anti-human DISC1, anti-rat DISC1 and anti-FEZ1 antibodies were applied at 1 : 100, 1 : 100 and 1 : 50 dilutions, respectively, for 24 h at 4°C. The secondary antibody (Alexa Flour 594-labeled goat anti-rabbit IgG, Molecular Probes, Eugene, OR, USA) was then applied at 1 : 500 dilution for 1 h at room temperature. For the detection of F-actin FITC-labeled phalloidin (Sigma-Aldrich) was used at 1 : 1000 dilution. Confocal microscopy was performed using an LSM-510 laser scanning microscope (Carl Zeiss, Germany).

Preparation of subcellular fractions

Subcellular fractionation for the separation of filamentous actin (F-actin) and globular actin (G-actin) was performed as described.18 SK-N-SH cells were lysed in 10 mM MOPS pH 7.0, 1.0% Triton X-100, 10% glycerol, 0.5 mM EDTA, 10 mM Na4P2O7, 50 mM NaF, 1 mM Na3VO4, 1 mM DTT in the presence of protease inhibitors. A measure of 2 mg/ml DNaseI was added to one-half of the lysate to depolymerize F-actin.19,20 After shaking in an ice bath for 15 min, the lysates were centrifuged at 16 000 g for 2 min to remove triton-insoluble cytoskeleton fibers, and the resulting supernatants were further centrifuged at 100000 g for 20 min. The pellets contain Triton-soluble F-actin, whereas the supernatants contain soluble G-actin. The pellets were washed three times and resuspended in lysis buffer (P fraction), raising the volume to correspond to that of the supernatant fraction (S fraction). The fractions were subjected to Western blot analysis using anti-actin antibody or anti-FEZ1 antibody.

Immunoprecipitation

HEK293T cells were transfected with human DISC1-FLAG and human FEZ1-HA, individually or in combination. FLAG-tagged truncated forms of DISC1 mentioned in Plasmids were also transfected in combination with FEZ-HA. Cells were lysed in TNE buffer/1% NP40. Prepared lysates were incubated with anti-FLAG antibody for 2 h at 4°C and then with rProtein G agarose (Invitrogen) for 1 h at 4°C. Agarose beads were then washed five times with TNE buffer. Immunoprecipitates were separated by SDS-PAGE and then FEZ-HA was detected in Western blot analysis using anti-HA antibody. Conversely, immunoprecipitates by anti-HA antibody were subjected to Western blot analysis using anti-FLAG antibody. PC12 cells were transfected with human FEZ1-HA or mock and then lysed in TNE buffer/1% NP40. Lysates were immunoprecipitated by anti-HA antibody. Immunoprecipitates were subjected to Western blot analysis using anti-actin antibody. Stably human DISC1-FLAG-expressing PC12 cells and mock-stable cells were treated with NGF (50 ng/ml) for 24 h and then lysed in TNE buffer/1% NP40. Lysates were immunoprecipitated by anti-FLAG antibody. Immunoprecipitates were subjected to Western blot analysis using anti-FEZ1 antibody. Cell lysis and blotting procedures were performed as described in Western blot analysis.

Results

Expression of DISC1 in rat brain

DISC1 is reported to be expressed throughout the body. 13 We first investigated the distribution of DISC1 mRNA in rat brain by in situ hybridization analysis (Figure 1). DISC1 was preferentially expressed in hippocampal, cortical, cerebellar and olfactor neurons in adult brain (Figure 1a). Specificity of hybridization signals was confirmed by an experiment using the sense cRNA probe. Since a relatively high level of expression was observed in the hippocampus, we investigated the DISC1 expression pattern in the hippocampal region at a developing stage. Signals in pyramidal cells of CA1-3 and granule cells of dentate gyrus were more potent at postnatal day 7 than in adulthood (Figure 1b).

Figure 1
figure1

In situ hybridization analysis of DISC1 mRNA in rat brain. (a) Sagittal sections of adult rat whole-brain hybridized by the antisense probe (left panel) and the sense probe (right panel). Signals were observed by incubation with the visualization reagent for 4 h at room temperature. High magnification views of the section hybridized by the antisense probe are shown for olfactor, hippocampal and cerebellar areas in lower panels, respectively. OB: olfactory bulb, CX: cerebral cortex, Hip: hippocampus, CB: cerebellum, TH: thalamus, SN: substantia nigra. (b) Coronal sections at the hippocampus region at postnatal day 7 (P7, left panel) and in adulthood (right panel). Signals were observed by incubation with the visualization reagent for 2 h at room temperature. High magnification views are shown for CA1 pyramidal cells in lower panels. DG: dentate gyrus, so: stratum oriens, sp: stratum pyramidale, sr: stratum radiatum.

DISC1 interacts with FEZ1

A putative protein of 854 amino acids encoded by the open reading frame in human DISC113 (Figure 2a) has no significant homology to other known proteins. The N-terminal region (amino acids 1–347) is predicted to consist of one or more globular domains.13 The helical C-terminal region (amino acids 348–854) is predicted to contain the translocation breakpoint and three stretches with coiled-coil-forming potential by interaction with other proteins.13 To confirm the existence of human DISC1 protein, we raised an antibody against the C-terminal 12 amino acids of the predicted sequence. This antibody detected two major bands (relative molecular mass, Mr, 105 kDa; expected size and 78 kDa) in lysates from SK-N-SH and HEK293T cells (Figure 2b). Compared to the sizes of overexpressed human DISC1 and its splicing variant that lacks 22 amino acids,13 endogenous DISC1 appeared to exist as a full-length form and its derivative. Partially revealed cDNA of rat homolog allowed us to raise an antibody against rat protein, which also detected two major bands in lysates from PC12 cells and brains of postnatal day 14 rats (Figure 2c). To investigate the distribution of DISC1 protein in adult rat brain, we performed an immunohistochemical study using anti-rat DISC1 antibody. A high level of expression was observed in granule cells of dentate gyrus and pyramidal cells throughout the CA1 to CA3 subfields in the hippocampus (Figure 2d). DISC1 positive signals were also detected in neuronal cells of olfactory bulb, cerebellum, cerebral cortex (data not shown).

Figure 2
figure2

Detection of DISC1 protein by specific antibodies. (a) Upper, predicted protein structure of human DISC1. DISC1 consists of the N-terminal globular region (black rectangle) and the helical C-terminal region (open rectangle) that contains three stretches with coiled-coil-forming potential (gray rectangles). Vertical line indicates position of the translocation breakpoint. Polyclonal antibody was raised against the underlined sequence. Lower, splicing variant of DISC1. Triangle indicates position of alternatively spliced 22 amino acids. (b) Western blot analysis using an antibody raised against human DISC1 with lysates from SK-N-SH and HEK293T cells. A protein of expected size (closed arrowhead) and one of smaller size (open arrowhead) were detected. Lysates from HEK293T cells transfected with human DISC1 or its splicing variant were also analyzed. (c) Western blot analysis using an antibody raised against rat DISC1 with lysates from PC12 cells and brains of postnatal day 14 rats. (d) The distribution of DISC1 protein in the hippocampus region of adult rat brain. A high magnification view is shown for CA1 subfield in the right panel. DG: dentate gyrus; so: stratum oriens, sp: stratum pyramidale, sr: stratum radiatum.

To identify interacting partners of DISC1 that might reflect some biological roles for this protein, we performed a yeast two-hybrid study. A human adult brain cDNA library was screened using the C-terminal region (amino acids 348–854) of human DISC1 as bait. One of the positive clones encoded a partial sequence of FEZ1 (amino acids 129–392), which is a mammalian homolog of the Caenorhabditis elegans UNC-76 protein involved in axonal outgrowth and fasciculation.21 UNC-76 and FEZ1 are not similar to any previously characterized proteins and represent a new protein family.21 Human FEZ1 protein is able to complement the function of UNC-76 in the nematode.21 Coexpression of FEZ1 and the constitutively active mutant of PKCζ induced PC12 cells to neuronal differentiation.22 From these observations, FEZ1 is assumed to play a crucial role in the axon guidance machinery in mammals, although the molecular mechanism involving FEZ1 is still unclear. To determine the regions in FEZ1 and DISC1 involved in their interaction, a yeast two-hybrid assay was performed using various shorter fragments of human FEZ1 and human DISC1 (Figure 3a). The C-terminal region of FEZ1 (amino acids 247–392), which is highly conserved with the nematode UNC-76, was required for interaction with DISC1. A DISC1 region (amino acids 446–633), containing two stretches with coiled-coil-forming potential and the translocation breakpoint, was shown to be critical for interaction with FEZ1. It is of note that in this assay a DISC1-truncated form (amino acids 348–597) lacking a C-terminus downstream of the translocation breakpoint interacted with FEZ1 weakly, because production of the truncated DISC1 protein would be possible in translocation carriers.13

Figure 3
figure3

DISC1 interacts with FEZ1. (a) Yeast two-hybrid assay to determine the binding regions of FEZ1 (upper panel) and DISC1 (lower panel). Open bars indicate the human FEZ1 clone obtained by the screening and human DISC1 clone used as bait. Closed bars indicate shorter fragments. Yeasts were cotransformed with the bait and one of the FEZ1 fragments, or with the FEZ1 C-terminal fragment (amino acids 247–392) and one of the DISC1 fragments. α-Galactosidase activity of yeast transformants was assayed. +++ indicates strongly positive as a positive control. ++, + and − indicate moderately positive, weakly positive and negative, respectively. (b) DISC1 coimmunoprecipitated with FEZ1. HEK293T cells were transfected with human DISC1-FLAG and human FEZ1-HA, individually or in combination. The FEZ1-binding region of DISC1 (amino acids 446–633, DISC1/BR) and a deleted DISC1 that lacks the binding region (DISC1/ΔBR) were also tagged with FLAG and tested in place of full-length DISC1-FLAG. Immunoprecipitates by anti-FLAG or anti-HA antibody were subjected to Western blot analysis using the reciprocal antibody.

The interaction between DISC1 and FEZ1 was confirmed by an immunoprecipitation assay using HEK293T cells (Figure 3b). The cells were transfected with FLAG-tagged human DISC1 and HA-tagged human FEZ1 expression vectors, individually or in combination. Cell lysates were prepared and immunoprecipitated by anti-FLAG or anti-HA antibody. HA-tagged FEZ1 was detected in the immunoprecipitates by anti-FLAG antibody in Western blot analysis. Conversely, FLAG-tagged DISC1 was detected in the immunoprecipitates by anti-HA antibody. FEZ1 also co-immunoprecipitated with a DISC1 fragment (amino acids 446–633), identified as FEZ1-binding region by the yeast two-hybrid assay, but not with a deleted DISC1 that lacks the binding region. These results demonstrated the interaction between DISC1 and FEZ1 in mammalian cells.

Intracellular localization of DISC1 and FEZ1

DISC1 has restricted structural similarities to structural proteins.13 On the other hand, in the course of characterization of FEZ1, a pull-down assay using rat brain lysate revealed that FEZ1 interacts with actin (TF and SK, unpublished data). We next examined the intracellular localization of DISC1 and FEZ1 from the viewpoint of cytoskeletal structure. We raised an antibody against FEZ1 which detected a protein of 46 kDa in the lysate from SK-N-SH cells (Figure 4p), the size revealed by in vitro synthesis.22 DISC1 exhibited a punctate distribution in the cytosol of SK-N-SH cells with the perinuclear high-density region (Figure 4a). DISC1 was also located on some filamentous structures which overlapped with F-actin as stress fibers detected by phalloidin staining (Figure 4a-c). FEZ1 was either punctate stained or distributed along organized filamentous structures, which remarkably overlapped with stress fibres, in the cytosol of SK-N-SH cells (Figure 4d-f). In cultured rat hippocampal neurons, colocalization of DISC1 and F-actin was observed in neurite growth cones (Figure 4g-i), where F-actin forms lamellipodia and filopodia, dynamic structures involved in axonal extension.23,24 FEZ1 also colocalized with F-actin in growth cones of cultured neurons (Figure 4j-l). Furthermore, we detected colocalization of transfected GFP-fused human DISC1 and endogenous FEZ1 in the growth cone (Figure 4m-o). These results suggest that the interaction of DISC1 and FEZ1 is associated with F-actin, presumably by direct binding of FEZ1 to actin. To investigate the association of FEZ1 and F-actin, we performed a subcellular fractionation study. Fractions for separation of F-actin and G-actin were prepared from SK-N-SH cell lysates as described18 and subjected to Western blot analysis. High-speed pellet fraction (P) and supernatant fraction (S) contain triton-soluble F-actin and soluble G-actin, respectively.18 FEZ1 was found in P fraction as well as S fraction (Figure 4q, lanes marked DNaseI−). When the lysate was treated with DNaseI to depolymerize F-actin19,20 before the fractionation, FEZ1 was not detectable in P fraction (Figure 4q, lanes marked DNaseI+). The depolymerization of F-actin was confirmed by the change of the actin population. This partitioning of FEZ1 into S fraction suggests that the protein, indeed, is associated with F-actin. The interaction between FEZ1 and actin was confirmed by immunoprecipitation assay (Figure 4r). PC12 cells were transfected with HA-tagged human FEZ1 or mock, and then cell lysates were immunoprecipitated by anti-HA antibody. Coimmunoprecipitation of actin and FEZ1 was detected by Western blot analysis using anti-actin antibody.

Figure 4
figure4

Intracellular localization of DISC1 and FEZ1. (a, d) SK-N-SH cells stained by anti-human DISC1 or anti-FEZ1 antibody, respectively. (b, e) Same cells stained by phalloidin for the detection of F-actin. (c, f) Merged images of (a) with (b), (d) with (e). Arrows in (c) indicate colocalization of DISC1 and F-actin. (g) Cultured rat hippocampal neuron stained by anti-rat DISC1 antibody. (h) Same neuron stained by phalloidin. (i) Merged image of (g) with (h). (j, m) Cultured rat hippocampal neurons stained by anti-FEZ1 antibody. (k, n) Same neurons stained by phalloidin or transfected with GFP-fused human DISC1, respectively. (l, o) Merged images of (j) with (k), (m) with (n). Arrows in (g–o) indicate growth cones. (p) Western blot analysis using an antibody raised against FEZ1 with lysate from SK-N-SH cells. (q) Subcellular fractionation study from SK-N-SH cell lysates. High-speed pellet fraction (P) and supernatant fraction (S) were prepared and subjected to Western blot analysis using anti-actin antibody or anti-FEZ1 antibody. P and S fractions contain triton-soluble F-actin and soluble G-actin, respectively. One-half of the lysate was treated with 2 mg/ml DNaseI to depolymerize F-actin before the fractionation. (r) Immunoprecipitation assay for the detection of the interaction between FEZ1 and actin. PC12 cells were transfected with human FEZ1-HA or mock, and then lysates were prepared. Immunoprecipitates by anti-HA antibody were subjected to Western blot analysis using anti-actin antibody.

DISC1 participates in neurite outgrowth through its interaction with FEZ1

FEZ1 is reported to be involved in axonal outgrowth and fasciculation,21,22 and we have shown that FEZ1 colocalizes with F-actin. Reorganization of the actin-based cytoskeletal structure is required for the neurite outgrowth of neurons.24 We evaluated the physiological role of DISC1/FEZ1 interaction in neuronal cells, especially at the stage of neurite outgrowth using PC12 cells. After stimulation with NGF, PC12 cells stop proliferation and begin to extend neurites. This feature is widely used as a model system for neuronal differentiation and neurite outgrowth. We established PC12 cell lines stably expressing FLAG-tagged human DISC1 and then examined the interaction between FLAG-tagged DISC1 and endogenous FEZ1 in the course of neuronal differentiation. As shown in Figure 5a, the amount of FEZ1 in the immunoprecipitates by anti-FLAG antibody was drastically increased upon NGF stimulation. As NGF stimulation did not alter the expression levels of endogenous FEZ1 (Figure 5a, lower panel) and FLAG-tagged DISC1 (data not shown), this result indicates that DISC1/FEZ1 interaction was upregulated during neuronal differentiation. In mock-stable cells, FEZ1 was not immunoprecipitated (Figure 5a). When treated with NGF, DISC1-stable lines exhibited enhanced neurite extension compared to mock-stable cells (Figure 5b-m).

Figure 5
figure5

DISC1 is involved in neurite outgrowth through its binding to FEZ1. (a) Interaction between DISC1 and FEZ1 was upregulated during neurite outgrowth. Stably human DISC1-FLAG-expressing PC12 cells were treated with NGF (50 ng/ml) for 24 h and collected. Lysates were immunoprecipitated by anti-FLAG antibody. Immunoprecipitated complexes were subjected to Western blot analysis using anti-FEZ1 antibody. (b–m) Stably DISC1-expressing PC12 cells (h–j and k–m showing line #1 and #2, respectively) exhibited enhanced extension of neurites upon the stimulation with NGF compared to mock-stable cells (b–d and e–g showing line #1 and #2, respectively). Cells were treated with NGF (50 ng/ml) for 0 h (b, e, h, k), 24 h (c, f, i, l), and 48 r (d, g, j, m) and their neurites were microscopically observed.

As shown in Figure 3, a DISC1 region (amino acids 446–633) is essential for the interaction with FEZ1 and therefore is expected to function as a dominant-negative form of DISC1 through the inhibition of binding between FEZ1 and full-length DISC1. This region was cloned into a bicistronic expression vector, and then DISC1-stable PC12 cells were transiently transfected with the construct or mock and treated with NGF for 48 h. Transfected cells were labeled by the expression of GFP (Figure 6a), and the lengths of the longest neurite of each labeled cell were measured. The cells expressing FEZ1-binding region of DISC1 displayed inhibited neurite extension compared to the mock-expressing cells (Figure 6b). The means ±SE of the lengths of the longest neurite were 80.9±24.5 and 32.8±27.7 μm for mock-expressing cells and binding region-expressing cells, respectively (P<0.001, Student's t-test, combined triplicate experiments). These results suggest that the interaction of DISC1 and FEZ1 plays a crucial role in neurite outgrowth.

Figure 6
figure6

Neurite outgrowth is inhibited by overexpressed FEZ1-binding region of DISC1. (a) Stably human DISC1-expressing PC12 cells were transiently transfected with mock or FEZ1-binding region of human DISC1 (amino acids 446–633) in pIRES2-EGFP vector, and then treated with NGF for 48 h. Transfected cells were labeled by expression of GFP. Corresponding phase-contrast images are also shown. (b) The lengths of the longest neurite of each labeled cell were measured. Cells with the longest neurite of 0–50, 50–100 or >100 μm were scored as a percentage of the total number of transfected cells. At least 70 cells were assessed in each experiment, and bars show means (± SE) of triplicate experiments.

Discussion

In the present study, we identified FEZ1 as an interacting partner of DISC1 by a yeast two-hybrid study. FEZ1 is a mammalian homolog of the C. elegans UNC-76 protein involved in axonal outgrowth and fasciculation.21,22 The interaction between DISC1 and FEZ1 was upregulated in PC12 cells during neuronal differentiation. Moreover, neurite outgrowth was enhanced by the overexpression of DISC1, and inhibition of the interaction between DISC1 and FEZ1 disturbed this enhanced neurite outgrowth. These results suggest that DISC1 participates in neurite extension machinery through its interaction with FEZ1. It should be noted that a DISC1-truncated form lacking a C-terminus downstream of the translocation breakpoint showed reduced potential for interaction with FEZ1 (Figure 3a), because production of the truncated DISC1 protein would be possible in translocation carriers.13 The DISC1 region that displayed a dominant-negative effect on neurite outgrowth consists of 188 amino acids, and our results do not exclude the possibility that this effect can be attributed to the interaction of the region with partner(s) other than FEZ1. A recent study25 has identified possible DISC1 interactors including cytoskeleton-related proteins by yeast two-hybrid screening using two portions of human DISC1 as bait, which partially overlap the region of the dominant-negative form. Further studies to determine restricted regions essential for the binding to interactors should be conducted.

In the Scottish family, the translocation was not associated with any physical disorders.14 It is unclear why the disruption of DISC1 by translocation causes psychiatric diseases selectively regardless of its expression throughout the body,13 while interacting partner(s) with the restricted expression to the brain might well explain this selectivity. The expression of FEZ1 is highly specific to the brain,22 and the elevated level of FEZ1 expression was observed in neurons of rat brain at embryonic day 18 and postnatal day 7 (unpublished data). In the study of nematode, the severe defects in newly hatched unc-76 mutant larvae suggest the importance of UNC-76 in nervous system development.21 Combined with the enhanced DISC1 expression in rat brain at a developing stage (Figure 1b), these findings imply that FEZ1/DISC1 interaction plays a crucial role in the development of the mammalian nervous system.

Recent studies have provided reliable evidence that schizophrenia is a neurodevelopmental disorder.3,26,27 Cytoarchitectual change in the hippocampus have been noteworthy among the various neuropathological abnormalities reported in schizophrenia.28,29,30 Decreased neuronal size and alterations in presynaptic and dendritic markers suggest abnormalities in the hippocampal neural circuitry in schizophrenia.30 In this regard, it is of note that the expression of DISC1 was abundant in hippocampal neurons especially at a developing stage (Figure 1a, b), suggesting the potential involvement of DISC1 in the formation of the hippocampal neural circuits. This raises the possibility that dysfunction of DISC1 may confer abnormal development of the nervous system, leading to susceptibility to psychiatric illnesses.

In the Western blot analysis, an antibody raised against human DISC1 detected at least two bands in lysates from human cell lines (Figure 2b). The band of 105 kDa fitted the expected size of human DISC1. As overexpressed 78 kDa protein was detected in the lysate from cells transiently transfected with full-length human DISC1 cDNA, this smaller form of DISC1 was thought to arise from post-translational modification. Further examination as to the origin of the 78 kDa protein is needed. A recent study25 has reported the existence of two forms of DISC1 in rat brain and the upregulated expression of the full-length form at developing stages, suggesting the importance of the full-length form with respect to the development of the nervous system. In the present study, an antibody raised against rat protein also revealed two forms in lysates from brains of postnatal day 14 rats (Figure 2c). The expression pattern of DISC1 protein revealed by immunohistchemical analysis lent support to the result in the in situ hybridization study.

The entire pathway involving DISC1 and FEZ1 is still unclear, although these components were identified as cytoskeletal-associated proteins by our findings. The possible implications of DISC1 in the cytoskeleton are also predicted by the recent study25 and preliminary reports.31,32 So far, the disruption of DISC1 by translocation has been found in one family from Scotland. Although dysfunction of DISC1 might account for the pathogenesis of a small subgroup of schizophrenia, elucidation of the pathophysiological role of DISC1 will provide a better understanding of the general etiology of schizophrenia.

References

  1. 1

    McGuffin P, Owen MJ, Farmer AE . Genetic basis of schizophrenia. Lancet 1995; 346: 678–682.

  2. 2

    Riley B, Williamson R . Sane genetics for schizophrenia. Nat Med 2000; 6: 253–255.

  3. 3

    Sawa A, Snyder SH . Schizophrenia: diverse approaches to a complex disease. Science 2002; 296: 692–695.

  4. 4

    Karayiorgou M, Gogos JA . A turning point in schizophrenia genetics. Neuron 1997; 19: 967–979.

  5. 5

    Riley BP, McGuffin P . Linkage and associated studies of schizophrenia. Am J Med Genet 2000; 97: 23–44.

  6. 6

    Berrettini WH . Are schizophrenic and bipolar disorders related? A review of family and molecular studies. Biol Psychiatry 2000; 48: 531–538.

  7. 7

    Brzustowicz LM, Hodgkinson KA, Chow EWC, Honer WG, Bassett AS . Location of a major susceptibility locus for familial schizophrenia on chromosome 1q21–q22. Science 2000; 288: 678–682.

  8. 8

    Straub RE, MacLean CJ, O'Neill FA, Burke J, Murphy B, Duke F et al. A potential vulnerability locus for schizophrenia on chromosome 6p24–22: evidence for genetic heterogeneity. Nature Genet 1995; 11: 287–293.

  9. 9

    Blouin JL, Dombroski BA, Nath SK, Lasseter VK, Wolyniec PS, Nestadt G et al. Schzophrenia susceptibility loci on chromosomes 13q32 and 8p21. Nat Genet 1998; 20: 70–73.

  10. 10

    Williams J, Spurlock G, McGuffin P, Mallet J, Nothen MM, Gill M et al. Association between schizophrenia and T102C polymorphism of the 5-hydroxytryptamine type 2a-receptor gene. Lancet 1996; 347: 1294–1296.

  11. 11

    Sklar P, Schwab SG, Williams NM, Daly M, Schaffner S, Maier W et al. Association analysis of NOTCH4 loci in schizophrenia using family and population-based controls. Nat Genet 2001; 28: 126–128.

  12. 12

    Anney RJ, Rees MI, Bryan E, Spurlock G, Williams N, Norton N et al. Characterisation, mutation detection, and association analysis of alternative promoters and 5' UTRs of the human dopamine D3 receptor gene in schizophrenia. Mol Psychiatry 2002; 7: 493–502.

  13. 13

    Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, Semple CAM et al. Disruption of two novel genes by a translocation co-segregating with schizophrenia. Hum Mol Genet 2000; 9: 1415–1423.

  14. 14

    Blackwood DHR, Fordyce A, Walker MT, St Clair DM, Porteous DJ, Muir WJ . Schizophrenia and affective disorders-cosegregation with a translocation at chromosome 1q42 that directly disrupts brain-expressed genes: clinical and P300 findings in a family. Am J Hum Genet 2001; 69: 428–433.

  15. 15

    Ekelund J, Hovatta I, Parker A, Paunio T, Varilo T, Martin R et al. Chromosome 1 loci in Finnish schizophrenia families. Hum Mol Genet 2001; 10: 1611–1617.

  16. 16

    Katayama T, Imaizumi K, Tsuda M, Mori Y, Takagi T, Tohyama M . Expressions of an ADP-ribosylation factor like gene, ARF4L, is induced after transient forebrain ischemia in the gerbil. Brain Res Mol Brain Res 1998; 56: 66–75.

  17. 17

    Neumann H, Cavalie A, Jenne DE, Wekerle H . Induction of MHC class I genes in neurons. Science 1995; 28: 549–552.

  18. 18

    Smith L, Parizi-Robinson M, Zhu MS, Zhi G, Fukui R, Kamm KE et al. Properties of long myosin light chain kinase binding to F-actin in vitro and in vivo. J Biol Chem 2002; 277: 35597–35604.

  19. 19

    Lehtonen S, Zhao F, Lehtonen E . CD2-associated protein directly interacts with the actin cytoskeleton. Am J Physiol Renal Physiol 2002; 283: F734–F743.

  20. 20

    Fox JE . Linkage of a membrane skeleton to integral membrane glycoproteins in human platelets. Identification of one of the glycoproteins as glycoprotein Ib. J Clin Invest 1985; 76: 1673–1683.

  21. 21

    Bloom L, Horvitz HR . The Caenorhabditis elegans gene unc-76 and its human homologs define a new gene family involved in axonal outgrowth and fasciculation. Proc Natl Acad Sci USA 1997; 94: 3414–3419.

  22. 22

    Kuroda S, Nakagawa N, Tokunaga C, Tatematsu K, Tanizawa K . Mammalian homologue of the Caenorhabditis elegans UNC-76 protein involved in axonal outgrowth is a protein kinase C ζ-interacting protein. J Cell Biol 1999; 144: 403–411.

  23. 23

    Hall A . Rho GTPases and the actin cytoskeleton. Science 1998; 279: 509–514.

  24. 24

    Luo L . Rho GTPases in neuronal morphogenesis. Nat Rev Neurosci 2000; 1: 173–180.

  25. 25

    Ozeki Y, Tomoda T, Kleiderlein J, Kamiya A, Bord L, Fujii K et al. Disrupted-in-Schizophrenia-1 (DISC-1): mutant truncation prevents binding to NudE-like (NUDEL) and inhibits neurite outgrowth. Proc Natl Acad Sci USA 2003; 100: 289–294.

  26. 26

    Weinberger DR . Implication of normal brain development for the pathogenesis of schizophrenia. Arch Gen Psychiatry 1987; 44: 660–669.

  27. 27

    Lewis DA, Levitt P . Schizophrenia as a disorder of neurodevelopment. Annu Rev Neurosci 2002; 25: 409–432.

  28. 28

    Harrison PJ . The neuropathology of schizophrenia.A critical review of the data and their interpretation. Brain 1999; 122: 593–624.

  29. 29

    Heckers S, Konradi C . Hippocampal neurons in schizophrenia. J Neural Transm 2002; 109: 891–905.

  30. 30

    Harrison PJ, Eastwood SL . Neuropathological studies of synaptic connectivity in the hippocampal formation in schizophrenia. Hippocampus 2001; 11: 508–519.

  31. 31

    Millar JK, James R, Christie S, Taylor MS, Devon RS, Hogg G et al. Abstract from Xth World Congress on Psychiatric Genetics, 2002.

  32. 32

    Kandpal G, Ma L, Acton P, Austin CP, Morris JA . Abstract from Society for Neuroscience 32nd Annual Meeting, 2002.

Download references

Acknowledgements

We acknowledge Drs A Sawa (Johns Hopkins University), N Mori and N Takei (Hamamatsu University School of Medicine), K Tanizawa and J Takeda (Osaka University) for critical reading of the manuscript and for valuable discussions and encouragement.

Author information

Correspondence to T Katayama.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Miyoshi, K., Honda, A., Baba, K. et al. Disrupted-In-Schizophrenia 1, a candidate gene for schizophrenia, participates in neurite outgrowth. Mol Psychiatry 8, 685–694 (2003). https://doi.org/10.1038/sj.mp.4001352

Download citation

Keywords

  • DISC1
  • FEZ1
  • mental disorders
  • schizophrenia
  • cytoskeleton
  • neurites

Further reading