A novel DISC1-interacting partner DISC1-Binding Zinc-finger ... - Nature

Molecular Psychiatry (2007) 12, 398–407 & 2007 Nature Publishing Group All rights reserved 1359-4184/07 $30.00 www.nature.com/mp

ORIGINAL ARTICLE

A novel DISC1-interacting partner DISC1-Binding Zinc-finger protein: implication in the modulation of DISC1-dependent neurite outgrowth T Hattori1,2,10, K Baba3,10, S Matsuzaki1,2,4,10, A Honda5, K Miyoshi6, K Inoue1,2, M Taniguchi1,2, H Hashimoto4,7, N Shintani7, A Baba7, S Shimizu1,2, F Yukioka1,2, N Kumamoto1,2, A Yamaguchi8, M Tohyama1,2,4 and T Katayama9 1 Department of Anatomy and Neuroscience, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan; 2The 21st Century Center of Excellence Program, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan; 3Department of Anatomy and Developmental Neurobiology, School of Medicine, Kobe University, Kobe, Japan; 4The Osaka-Hamamatsu Joint Research Center for Child Mental Development, Graduate School of Medicine, Osaka University, Suita, Osaka, Japan; 5 Pharmacology Research Laboratory, Tanabe Seiyaku Co. Ltd, Yodogawa-ku, Osaka, Japan; 6Department of Brain Science, Graduate School of Medicine and Dentistry, Okayama University, Okayama, Japan; 7Laboratory of Molecular Neuropharmacology, Graduate School of Pharmaceutical Sciences, Osaka University, Suita, Osaka, Japan; 8Department of Neurobiology, Graduate School of Medicine, Chiba University, Chiba, Japan and 9Department of Anatomy and Neuroscience, Hamamatsu University School of Medicine, Hamamatsu, Shizuoka, Japan

Disrupted-in-schizophrenia 1 (DISC1) is a gene disrupted by a (1;11) (q42.1;q14.3) translocation that segregates with major psychiatric disorders in a Scottish family. To investigate how DISC1 confers susceptibility to psychiatric disorders, we previously identified fasciculation and elongation protein zeta-1 and Kendrin as DISC1-interacting molecules in a yeast twohybrid screen of a human brain complementary DNA library. Here, we have further identified a novel DISC1-interacting protein, termed DISC1-Binding Zinc-finger protein (DBZ), which has a predicted C2H2-type zinc-finger motif and coiled-coil domains. DBZ was co-immunoprecipitated with DISC1 in lysates of PC12 cells and rat brain tissue. The domain of DISC1 interacting with DBZ was close to the translocation breakpoint in the DISC1 gene. DBZ messenger RNA (mRNA) was expressed in human brains, but not in peripheral tissues. In situ hybridization revealed high expression of DBZ mRNA in the hippocampus, olfactory tubercle, cerebral cortex and striatum in rats. Because this pattern of localization was similar to that of the pituitary adenylate cyclase (PAC1) receptor for pituitary adenylate cyclase-activating polypeptide (PACAP), which has recently been implicated in neuropsychological functions, we examined whether DISC1/DBZ interaction was involved in the PACAP signaling pathway. PACAP upregulated DISC1 expression and markedly reduced the association between DISC1 and DBZ in PC12 cells. A DISC1-binding domain of DBZ reduced the neurite length in PC12 cells after PACAP stimulation and in primary cultured hippocampal neurons. The present results provide some new molecular insights into the mechanisms of neuronal development and neuropsychiatric disorders. Molecular Psychiatry (2007) 12, 398–407. doi:10.1038/sj.mp.4001945; published online 23 January 2007 Keywords: pituitary adenylate cyclase-activating polypeptide (PACAP); pituitary adenylate cyclase-activating polypeptide receptor (PAC1); psychiatric disorders; schizophrenia; two-hybrid system; neutrite outgrowth

Introduction Schizophrenia is a devastating psychiatric disorder with a lifetime prevalence of about 1% of the

Correspondence: Dr S Matsuzaki, Department of Anatomy and Neuroscience, Graduate School of Medicine, Osaka University, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan. E-mail: [email protected] 10 These authors contributed equally to this work. Received 11 October 2005; revised 11 September 2006; accepted 11 December 2006; published online 23 January 2007

population worldwide, and it commonly has a chronic course. The underlying mechanisms are still largely unknown, but a growing body of evidence suggests that schizophrenia is a multifactorial disorder influenced by genetic, neurodevelopmental and social factors.1–6 Disrupted-in-schizophrenia 1 (DISC1) has been identified as a potential susceptibility gene for major psychiatric disorders. Disruption of this gene by a balanced (1;11) (q42.1;q14.3) translocation results in a predicted C-terminal truncation of the open reading frame, and this anomaly is segregated with schizophrenia, bipolar affective

DBZ/DISC1 interaction and neurites T Hattori et al

disorder and recurrent major depression in a large Scottish family.7,8 In addition, a frameshift mutation of DISC1 has been identified in an American family with schizophrenia and schizoaffective disorder,9 whereas the association of the single-nucleotide polymorphisms of DISC1 with schizophrenia, schizoaffective disorder and bipolar disorder has also been suggested.10 Although investigating the relationship of DISC1 to susceptibility to psychiatric disorders and brain function, we previously identified fasciculation and elongation protein zeta-1 (FEZ1)11,12 and Kendrin13 as DISC1-interacting molecules by yeast two-hybrid screening of a human brain complementary DNA (cDNA) library. DISC1 and FEZ1 were found to be colocalized in growth cones in association with F-actin in both SK-N-SH cells and cultured hippocampal neurons.11 The interaction of DISC1 with FEZ1 was markedly enhanced along with neurite extension in PC12 cells by nerve growth factor (NGF) stimulation.11,14 An association between singlenucleotide polymorphisms of the FEZ1 gene and schizophrenia has also been suggested in a Japanese population.15 Kendrin, also termed pericentrin-B, is a calmodulin-binding protein that is specifically localized to the centrosome, and we have previously demonstrated that Kendrin is involved in the centrosomal localization of DISC1.13 Pituitary adenylate cyclase-activating polypeptide (PACAP) is a neuropeptide expressed in the brain as well as in the neurons of a number of peripheral organs, and it has a role in various neurobiological functions, such as neurotransmission and neural plasticity, as well as a neurotrophic effect via three heptahelical G-protein-linked receptors, one of which is specific for PACAP (the PAC1 receptor) and two other receptors that are shared with vasoactive intestinal peptide (the VPAC1 and VPAC2 receptors).16,17 We have previously developed mice lacking the Adcyap1 gene encoding PACAP (Adcyap1/), which exhibited prominent behavioral abnormalities that included hyperactivity with impaired habituation to novel situations, increased novelty-seeking behavior and reduced anxiety, as well as evidence of neuronal dysfunction such as impairments of prepulse inhibition and hippocampal long-term potentiation (LTP).18–20 PAC1 receptor-deficient mice also demonstrated an increase of locomotor activity, reduced anxiety-like behavior and abnormal social behavior, as well as impairment of hippocampal LTP.21–23 These observations indicate that PACAP signaling mediated via the PAC1 receptor has a critical role in the development and/or functioning of neural pathways and suggest the potential clinical relevance of PACAP signaling dysfunction to neuropsychiatric disorders. In this study, we found that DBZ interacts with DISC1, and examined the distribution of DBZ transcripts in the brain and other tissues as well as the functional implications of its interaction with DISC1. A possible link between DISC1 and PACAP signaling was also explored.

399

Materials and methods Yeast two-hybrid screening assay The C-terminal domain of human DISC1 (amino acids 348–854) was subcloned into pAS2-1 (a GAL4 DNAbinding domain vector; Clontech, Palo Alto, CA, USA) and used as bait. Yeast strain AH109 was transformed with the bait plasmid, mated with strain Y187 that had been pretransformed with a human adult brain cDNA library (Clontech) and plated on quadruple dropout medium (-Ade,-His,-Leu,-Trp). The screening procedure and an a-galactosidase assay were performed according to the manufacturer’s instructions. Construction of the DISC1 and DBZ expression vectors We previously constructed the pcDNA3.1 ( þ ) expression vector (Invitrogen, Carlsbad, CA, USA) carrying the full-length human DISC1 cDNA encoding 854 amino acids, and a truncated sequence encoding amino acids 446–633 fused in-frame with the FLAG sequence at the 30 end11 (the resultant constructs were termed DISC1-FLAG and DISC1 (446–633)-FLAG, respectively). In this present study, human DISC1 cDNA was fused with the HA epitope sequence (DISC1-HA), and the DISC1 sequences encoding amino acids 348–597 and 598–854 were fused with the FLAG epitope sequence (DISC1 (348–597)-FLAG and DISC1 (598–854)-FLAG, respectively). The fulllength human DBZ cDNA encoding 407 amino acids was subcloned into pEGFP-N1 vector (Clontech) to generate DBZ with green fluorescent protein (GFP) fused to the C-terminal (DBZ-GFP). Full-length human DBZ (DBZ) and DBZ with FLAG fused at the 30 end (DBZ-FLAG) were subcloned into the expression vector pcDNA3.1 ( þ ). Rat DBZ cDNA encoding amino acids 152–301 that included the predicted coiled-coil domains, serving as a binding region with DISC1 as described later, was fused in-frame to the c-myc epitope sequence at the 30 end and subcloned into a bicistronic expression vector (pIRES2-EGFP; Clontech) to produce DBZ (152-301)-IRES-GFP. A recombinant adenovirus expressing GFP and human DBZ with GFP fused to the C-terminal (AdvGFP and Adv-DBZ-GFP, respectively) were generated using the ViraPower Adenoviral Expression system (Invitrogen) according to the manufacturer’s instructions, with the following oligonucleotide primers being used: human DBZ, 50 -ATG CAA CAG AAG GCT TTT GA-30 (sense) and 50 -TTC AGC ACT GCG ATC ATT TCC-30 (antisense). Cell culture and transfection HEK293T cells and PC12 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal calf serum (FCS) and DMEM containing 10% horse serum plus 5% FCS, respectively. Rat primary hippocampal neurons were prepared from embryonic day 18 Wistar rats with nerve cell culture system MB-X9901 (Sumitomo Bakelite Co. Ltd, Tokyo, Japan) as described in the Sumilon

Molecular Psychiatry


400

Protocol N-4.2. Neurons were plated on poly-L-lysinecoated chamber slides and cultured in DMEM with 10% FCS. HEK293T cells were transiently transfected with 5 mg of DISC1-FLAG, DISC1 (446–633)-FLAG, DISC1 (348–597)-FLAG, DISC1 (598–854)-FLAG, DBZ, DBZ-FLAG, DBZ-GFP or a GFP expression vector using Lipofectamine 2000 according to the manufacturer’s instructions (Invitrogen), and were subjected to immunoprecipitation and/or Western blot analysis. PC12 cells stably expressing DISC1HA and mock-transfected PC12 cells were generated as described previously,11 and were infected with Adv-DBZ-GFP or Adv-GFP at about 30 multiplicities of infection. PC12 cells and hippocampal neurons were transiently transfected with DBZ (152–301)IRES-GFP or a GFP expression vector using Lipofectamine 2000 (Invitrogen). Northern blot analysis A cDNA fragment of human DBZ was amplified by the reverse transcription-polymerase chain reaction (RT-PCR) using the following oligonucleotide primers – 50 -CCA TTT GAG GTG GTG GCA GAG AGG-30 (sense) and 50 -GAG GCC TGC TGG CTC TGA TAG TAC-30 (antisense) – and was used as templates for the synthesis of probes. Northern blot hybridization with a 32P-labeled DBZ cDNA fragment was performed on a human Multiple Tissue Northern Blot filter (Clontech) as described previously.24 In situ hybridization cDNA fragments of rat DBZ and rat PAC1 receptor were amplified by RT-PCR (the oligonucleotide primers were 50 -CTT TGC GCA GCT GAC TCA GAA-30 (sense) and 50 -TTC AGC ACT GCG ATC ATT TCC-30 (antisense) for rat DBZ, and 50 -CTT GTA CAG AAG CTG CAG TC-30 (sense) and 50 -GGT GCT TGA AGT CCA TAG TG-30 (antisense) for rat PAC1 receptor) and used as templates for probe synthesis. In situ hybridization of sagittal and coronal rat brain sections with 35S-labeled RNA probes was performed as described previously.25 Generation of anti-DBZ antibody The generation of a rabbit anti-rat DISC1 polyclonal antibody has been described previously.11 In this study, a rabbit polyclonal anti-DBZ antibody was similarly raised against a peptide (RKGNIRPKMAKK) derived from the human DBZ sequence, and was subjected to affinity purification. Immunoprecipitation and Western blot analysis Lysates of HEK293T cells PC12 cells and whole-rat brains were subjected to immunoprecipitation followed by Western blotting, or were directly used for Western blotting as described previously.11 The following antibodies were used: mouse monoclonal antibodies directed against the FLAG epitope (SigmaAldrich, St Louis, MO, USA), GFP (Santa Cruz Biotechnology, Santa Cruz, CA, USA), HA epitope (Sigma-Aldrich) and b-actin (Chemicon International,


Temecula, CA, USA), as well as rabbit polyclonal antibodies targeting rat DISC1 and human DBZ. A nonspecific rabbit IgG (Sigma-Aldrich) was used as the negative control for immunoprecipitation, and goat anti-rabbit or mouse IgG conjugated with horseradish peroxidase (Cell Signaling Technology, Beverly, MA, USA) were used as the secondary antibodies for Western blotting. Reaction products were visualized by detection of chemiluminescence using an enhanced chemiluminescence kit (Amersham Biosciences, Piscataway, NJ, USA). Quantitation of relative band densities was performed by scanning densitometry. All experiments were repeated independently at least three times. To promote neuronal differentiation, PC12 cells were cultured for 2 days at a low cell density, starved of serum for 4 h and then treated with 100 nM PACAP (PACAP-38) (Peptide Institute, Mino, Osaka, Japan). Cells were harvested at the indicated times from 0 to 24 h after PACAP stimulation. Immunocytochemistry, measurement of neurite outgrowth and assessment of cytotoxicity Immunocytochemical staining of the HA epitope in PC12 cells stably expressing DISC1-HA and BIIItubulin (a marker of immature neurons) in cultured hippocampal neurons (with anti-bIII-tubulin antibody; R&D Systems, Minneapolis, MN, USA) was performed as described previously.11 Confocal microscopy was performed using an LSM-510 laser scanning microscope (Carl Zeiss, Oberkochen, Germany). Mock or DISC1-HA stable PC12 cells were cultured for 2 days at a low density and the medium was exchanged with basal medium. Cells were infected for 24 h with Adv-GFP or Adv-DBZ-GFP, subjected to 4 h of serum starvation and then treated with 100 nM PACAP (PACAP-38). At 48 h after PACAP stimulation, cells were observed. PC12 cells were cultured for 2 days at a low density and the medium was exchanged with basal medium. Cells were transfected with DBZ (152-301)-IRES-GFP or GFP, subjected to 4 h of serum starvation and then treated with 100 nM PACAP. Observation was performed at 48 h after PACAP stimulation (PC12 cells) or 24 h after plating and transfection (cultured hippocampal neurons). To analyze neurite outgrowth, PC12 cells and cultured hippocampal neurons expressing GFP were used, and immunohistochemical detection of bIII-tubulin was also carried out in the case of cultured hippocampal neurons. The percentage of neurite-bearing cells, the neurite length and the number of neurites per cell were analyzed on randomly selected digital microscope images. Cells with neurites longer than the cell body diameter were defined as neurite-bearing cells and were counted. Cells in aggregates were not counted. The neurite length to cell body diameter ratio of PC12 cells was determined by tracing the length of the longest neurite of each cell in a field that had identifiable neurites that could be completely visualized. The total neurite length was determined for each individual hippocampal neuron using


LSM510 software (version 3.2; Carl Zeiss). To evaluate cytotoxicity and cell death under our experimental conditions, the TdT-mediated dUTP nick end labeling (TUNEL) assay was performed with an In situ Cell Death Detection Kit according to the manufacturer’s instructions (Roche Applied Science, Penzberg, Germany) at the same time as analysis of neurite outgrowth. Animals All animal experiments were carried out in accordance with a protocol approved by the institutional Animal Care and Use Committee of Osaka University.

Results Identification of DBZ as a novel DISC1-interacting partner DISC1 functions by interacting with various other molecules, including FEZ1, Kendrin and Nudel.11,13,26,27 To detect potential new partners for

DISC1, we performed yeast two-hybrid screening of a human brain cDNA library using the C-terminal region (amino acids 348–854) of human DISC1 as the bait (Figure 1b), and as a result we identified several positive clones. Among these, one clone was identical to human KIAA0844 (accession number AB020651), for which the mouse counterpart gene has been reported to show high expression in the brain.28,29 The open-reading frame encoded a putative 407 amino-acid protein, without significant homology to any other known protein that was predicted to have one C2H2 zinc-finger domain (amino acids 26–51) and several stretches of amino acids that could possibly form coiled-coil domains (amino acids 169–297) (Figure 1a). Therefore, we designated this protein as DISC1-Binding Zinc-finger protein (DBZ). Interaction between DISC1 and DBZ was confirmed in mammalian cells by an immunoprecipitation assay using transfected HEK293T cells (Figure 1c). HEK293T cells were co-transfected with expression vectors for DISC1-FLAG together with DBZ-GFP or GFP, and cell

401

Figure 1 The predicted structure of human DBZ and DISC1, and immunoprecipitation showing an association between DBZ and DISC1. (a) Predicted structure of human DBZ. DBZ has a C2H2-type zinc-finger motif in the N-terminal region (black box) and coiled-coil domains in the center (gray box). (b) Predicted structure of human DISC1. DISC1 has a globular Nterminal domain (hatched box) and a helical C-terminal region that contains three coiled-coil domains (gray boxes). The dashed line indicates the breakpoint of the balanced (1;11) (q42.1;q14.3) translocation at amino acid 598, which segregates with major psychiatric disorders.7,8 The bold lines indicate the DISC1 regions used as bait in the yeast two-hybrid assay, as well as the expression vectors encoding DISC1 (446–633)-FLAG, DISC1 (348–597)-FLAG and DISC1 (598–854)-FLAG. (c) Immunoprecipitation assay of lysates of HEK293T cells expressing recombinant DBZ and DISC1. HEK293T cells were transfected with DISC1-FLAG combined with DBZ-GFP or GFP. (d) Deletion analysis of the domain of DISC1 interacting with DBZ. HEK293T cells were transfected with DBZ-GFP, either alone or in combination with DISC1 (446–633)-FLAG, DISC1 (348–597)-FLAG or DISC1 (598–854)-FLAG. (e) Detection of DBZ protein by Western blot analysis (WB) with a specific antibody. Lysates of PC12 cells and HEK293T cells expressing DBZ or DBZ-FLAG were subjected to WB with anti-DBZ antibody (upper panel) or anti-FLAG antibody (lower panel). The arrow indicates DBZ. (f) Immunoprecipitation of lysates of PC12 cells and adult whole-rat brains. (c, d, f) Immunoprecipitates (IP) obtained by antibodies to tag proteins (FLAG or GFP) (c, d), anti-DBZ antibody (DBZ) or nonspecific rabbit IgG (IgG) (f), as well as 1% (rat brain) or 5% (HEK293T cells and PC12 cells) dilutions of each lysate, were subjected to WB with the antibodies indicated. Molecular Psychiatry


402

lysates were subjected to immunoprecipitation with anti-FLAG or anti-GFP antibodies, followed by Western blot analysis with a reciprocal antibody. DBZGFP was detected in the immunoprecipitates by antiFLAG antibody, while conversely DISC1-FLAG was detected in the immunoprecipitates by anti-GFP antibody. However, GFP was not co-immunoprecipitated with DISC1-FLAG and vice versa. These results indicated that DISC1 and DBZ are physically associated with each other, at least in transfected mammalian cells. To confirm the interaction between endogenous DBZ and DISC1, we produced a specific anti-DBZ antibody (Figure 1e), as described in Materials and methods, and this antibody was shown to recognize endogenous DBZ (left lane) overexpressed DBZ (middle lane) and overexpressed DBZ-FLAG (right lane). Co-immunoprecipitation experiments with antibodies for DISC and DBZ were performed in lysates of PC12 cells and adult whole-rat brains (Figure 1f). Western blot analysis with antiDISC1 and anti-DBZ antibodies showed the presence of DISC1 and DBZ, respectively, whereas immunoprecipitation and subsequent Western blot analysis showed an association between the two proteins in both PC12 cells and rat brain. DISC1 regions interacting with DBZ The DISC1 protein consists of an N-terminal globular domain and a long helical C-terminal tail domain that

contains three possible coiled-coil domains (Figure 1b).7 We have previously shown that amino acids 446–633 of DISC1, which overlap the region that is presumably lost owing to (1;11) (q42.1;q14.3) translocation, participate in the interaction with FEZ1.11 Here, we examined whether this DISC1 region also participates in the interaction with DBZ. As shown in Figure 1d, DISC1 (446–633)-FLAG was co-immunoprecipitated with DBZ-GFP, indicating that the DBZbinding region of DISC1 overlapped the FEZ1-binding region. We constructed two additional DISC1 deletion mutants, DISC1 (348–597)-FLAG and DISC1 (598– 854)-FLAG, and examined their interaction with DBZ. DISC1 (348–597)-FLAG was markedly co-immunoprecipitated with DBZ-GFP, whereas DISC1 (598– 854)-FLAG showed little co-immunoprecipitation (Figure 1d). These results suggested that the region of DISC1 near the translocation breakpoint, particularly the region encompassing amino acids 348–597, participates in interaction with DBZ. Distribution of DBZ mRNA in the brain and other tissues, and intracellular localization of DBZ in PC12 cells Figure 2a shows the tissue distribution of DBZ mRNA as determined by Northern blot analysis using a human Multiple Tissue Northern Blot filter and DBZ cDNA. DBZ mRNA (approximately 4 kb) was exclusively expressed in the brain and was not expressed

Figure 2 Tissue distribution of DBZ mRNA and intracellular localization of DBZ. (a) Northern blot hybridization of DBZ mRNA. A human multiple tissue Northern blot filter was hybridized with 32P-labeled DBZ cDNA fragments. Positions of RNA markers are indicated on the left. The arrowhead indicates a DBZ mRNA band of approximately 4 kb. (b, c) In situ hybridization of DBZ mRNA in the adult rat brain. Sections were hybridized with 35S-labeled antisense RNA probe for DBZ mRNA. As controls, adjacent sections were hybridized with 35S-labeled sense RNA probe (inset in b). Autoradiography (b) and bright-field photomicrographs (c) of sections that were emulsion dipped and counterstained with thionine are shown. Higher magnification of the section is shown for CA3 and the thalamus in the center (c-A) and right (c-B) panels of (c), respectively. Scale bars = 5 mm. (b), 500 mm (left panel of c) and 50 mm (A and B of c). Arrows indicate neurons merged with DBZ signals. (d) Intracellular localization of DBZ in transfected PC12 cells. PC12 cells stably expressing DISC1-HA were infected with a DBZ-GFP-expressing adenovirus (Adv-DBZ-GFP; upper panels) or a GFP-expressing adenovirus (Adv-GFP; lower panels), and were immunostained with anti-HA antibody followed by Alexa Fluor 596-labeled goat anti-mouse IgG. Confocal microscopic images are shown. Scale bar = 50 mm. Molecular Psychiatry


by other tissues such as the heart, placenta, lung, liver, muscle, kidney and pancreas. DBZ mRNA was localized in the rat brain by in situ hybridization using labeled antisense RNA. As shown in Figure 2b, expression of DBZ mRNA was prominent in the cerebral cortex, hippocampus, olfactory tubercle and striatum. Moderate expression was observed in the olfactory bulb, thalamus, inferior colliculus and pontine nucleus. A low level of expression was also observed in many other regions of the brain. In thionine-stained sections, some silver grains were observed in medium to large cells, indicating that expression of DBZ mRNA was localized to neurons, as shown in Figure 2c. In the hippocampus, strong DBZ mRNA signals were detected in the pyramidal cell layer of CA1-3 and the granule cell layer of the dentate gyrus (Figure 2b and c). No clear labeling was detected with a sense riboprobe (Figure 2b inset). We next examined the intracellular localization of DBZ and whether it was colocalized with DISC1 using PC12 cells stably expressing DISC1-HA. We established a recombinant adenovirus capable of expressing DBZ-GFP (Adv-DBZ-GFP). DBZ-GFP showed a diffuse distribution in the cytoplasm and a slightly punctate pattern of expression, as well as increased expression in the perinuclear region, irrespective of DISC1-HA expression (Figure 2d; data not shown). DISC1-HA also showed a similar pattern of localization pattern in PC12 cells infected with Adv-DBZGFP or Adv-GFP (Figure 2d). GFP (the negative control) was widely distributed in both the cytoplasm and the nucleus, revealing a different pattern from that of either DBZ-GFP or DISC1-HA. Effect of PACAP on DISC1/DBZ interaction Several lines of evidence have indicated that PACAP has both neurotrophic and neuroprotective actions and is involved in synaptic plasticity in the hippocampus as well as psychomotor function.16–23 In accordance with previous observations,30 in situ hybridization of adult rat brains showed intense expression of PAC1 mRNA in the periglomerular and granule cells of the olfactory bulb, the granule cells of the dentate gyrus, the cerebral cortex (especially layer II) and many regions of the hypothalamus (Figure 3a and b). Thus, prominent signals for both DBZ and PAC1 transcripts were detected in the dentate gyrus (Figure 3b upper panels). Most of the silver grains were localized in the granule cells of Figure 3b, lower panels, suggesting that DBZ and PAC1 could be colocalized in these cells. We next examined whether DISC1/DBZ interaction was involved in the PACAP signaling pathway in PC12 cells, which are known to express the PAC1 receptor.17 Exposure to PACAP (100 nM) increased the expression of endogenous DISC1 in PC12 cells by about 50% after 24 h, whereas it had no effect on DBZ expression (Figure 3c). Interestingly, PACAP had a marked influence on the interaction of DBZ with DISC1. The co-immunoprecipitation of DISC1 with

403

Figure 3 PACAP-induced increase of endogenous DISC1 expression and transient inhibition of the endogenous DISC1/DBZ interaction in PC12 cells. (a, b) In situ hybridization compares the expression of PAC1 mRNA and DBZ mRNA in the rat brain. Sections were hybridized with 35 S-labeled antisense RNA probes for PAC1 mRNA or DBZ mRNA. Autoradiography (a) and dark-field (upper panels; hippocampus) and bright-field (lower panels; dentate gyrus) photomicrographs (b) of sections that were emulsion dipped and counterstained with thionine are shown. Scale bars = 5 mm (a) and 50 mm (b). (c) Immunoprecipitation and Western blot (WB) analysis (with anti-DISC1 or antiDBZ antibodies) of PC12 cell lysates collected at the indicated times after stimulation with 100 nM PACAP by the same method. Immunoprecipitates (IP) obtained with anti-DBZ antibody (DBZ), as well as 5% of each lysate (5% input), were subjected to Western blot (WB) analysis with the antibodies indicated. Quantitation of relative band densities for DISC1 co-immunoprecipitated with DBZ, as well as for total DISC1 or DBZ protein, was performed by scanning densitometry. Data were expressed as the mean7s.e.m. of at least three independent experiments. *P < 0.05 vs control at time 0 (Student’s t-test).

DBZ in lysates was reduced by about 80% 1 h after treatment of PC12 cells with PACAP (100 nM). However, this reduction was transient and there was a gradual return to the control level by 24 h after treatment. Neurite growth in PC12 cells overexpressing DBZ and/or DISC1 Previously, we found that DISC1 has a role in neurite growth.11 Therefore, we examined neurite growth in PC12 cells overexpressing DBZ and/or DISC1. PC12 cells stably expressing DISC1-HA and mock-transfected cells were infected with Adv-DBZ-GFP or AdvGFP for 24 h. Immunoprecipitation and Western blot Molecular Psychiatry


404

analysis confirmed the overexpression of DBZ-GFP and DISC1-HA as well as the association between these two molecules (Figure 4a). After the transfected cells were treated with PACAP (100 nM) for 48 h, the number of cells bearing neurites were counted. Overexpression of both DBZ and DISC1 resulted in a significant decrease in the number of neurite-bearing PC12 cells, whereas overexpression of either DBZ or DISC1 alone (DBZ or DISC1 in conjunction with mock or GFP) did not alter the number of neurite-bearing cells (Figure 4b and c). No significant changes of neurite length or the number of neurites per cell were observed in either mock- or DISC1-HA-transfetced PC12 cells irrespective of the expression of DBZ (data not shown). We also performed the TUNEL assay to assess possible cytotoxicity of the experimental

conditions. Because no significant change of apoptosis was observed in either mock- or DISC1-HAtransfetced PC12 cells irrespective of the expression of DBZ (Figure 4d), we concluded that the DISC1/DBZ interaction might have an important role in neurite growth. To assess the specificity of this effect, we next examined neurite growth after NGF stimulation. Overexpression of both DBZ and DISC1 did not affect neurite growth after PC12 cells were stimulated by NGF (data not shown). On the other hand, NGF depressed the DISC1/DBZ interaction in a similar manner to PACAP, although its recovery to the control level occurred more slowly (data not shown). Effect of DISC1/DBZ interaction on the differentiation of PC12 cells and neurons To further investigate the functions of endogenous DBZ, we identified the region of rat DBZ (amino acids 152–301) that interacted with DISC1. This region was subcloned into a bicistronic expression vector, pIRES2-EGFP (DBZ (152–301)–IRES-GFP), and its interaction with DISC1 was confirmed by performing a co-immunoprecipitation assay in HEK293T cells (data not shown). PC12 cells were transiently transfected with DBZ (152–301)-IRES-GFP or with GFP alone, and were treated with PACAP (100 nM) for 48 h. Then the relative neurite length to cell body diameter of each GFP-positive cell was analyzed, revealing that cells expressing DBZ (152–301) had a shorter neurite length than cells expressing GFP alone (Figure 5a). Under these conditions, no significant change of apoptosis was detected (Figure 5b) and the number of transfected cells was similar (GFP: 35.7%; DBZ (152–301)-IRES-GFP: 35.1%). The effect of DBZ (152-301) on neurite growth was similarly examined by using primary cultured hippocampal neurons. To assess the effect on neurite extension rather than maintenance of neurite integrity, rat hippocampal neurons prepared from

Figure 4 Effect of DISC1 and DBZ overexpression on neurite growth in PC12 cells. PC12 cells stably expressing DISC1-HA or mock-transfected cells were infected with Adv-DBZ-GFP or Adv-GFP. After 24 h, the cells were starved of serum for 4 h, treated with 100 nM PACAP and cultured for 48 h. (a) Immunoprecipitation and Western blot (WB) analysis of PC12 cell lysates. Immunoprecipitates (IP) obtained with anti-HA antibody (HA), as well as 5% of each lysate (5% input), were subjected to WB analysis with the antibodies indicated. (b) Number of neurite-bearing cells. Cells with neurites longer than one cell body diameter were counted as neurite-bearing cells. Data were expressed as the mean7s.e.m. of at least three independent experiments. *P < 0.05 (Student’s t-test). (c) Morphology of PC12 cells on phase-contrast images. Arrows and arrowheads indicate cells with neurite (s) longer or shorter than one cell body diameter, respectively (upper panels). Low power images of Mock/Adv-GFP and DISC1-HA/Adv-DBZ-GFP expressing PC12 cells (lower panels). Scale bar = 200 mm. (d) GFP-positive cells were assessed for apoptosis by the TUNEL assay. Data were expressed as the mean7s.e.m. of atleast three independent experiments. Molecular Psychiatry


405

Figure 5 Neurite outgrowth was inhibited by the DISC1-binding domain of DBZ (DBZ (152–301)–IRES-GFP). (a) PC12 cells were transfected with DBZ (152–301)-IRES-GFP or GFP alone at 2 days after plating. After 24 h, cells were starved of serum for 4 h and treated with 100 nM PACAP for 48 h. Phase-contrast and fluorescence microscopy images are shown. Diagrams display the relative neurite length to cell body diameter of transfected PC12 cells. At least 400 cells were randomly counted in three independent experiments. (b) GFP-positive cells were assessed for apoptosis by the TUNEL assay. Data were expressed as the mean7s.e.m. of atleast three independent experiments. (c) Rat hippocampal neurons were plated on poly-Llysine-coated chamber slides and cultured in DMEM/10% FCS. At the time of plating, hippocampal neurons were transiently transfected with DBZ (152–301)-IRES-GFP or the GFP expression vector. Cells were immunostained with anti-bIII-tubulin antibody at 24 h after transfection. Low power (left) and high power (center) views of fluorescence microscopy images are shown. The diagrams show the total neurite length for transfected rat primary hippocampal neurons (right). Approximately 50 cells were randomly counted in three independent experiments. (a, c) Arrows indicate GFP-positive transfected cells. *P < 0.05 (Student’s t-test). Scale bars = 50 mm (a) and 20 mm (c). (b, d) GFP-positive cells were assessed for apoptosis by the TUNEL assay. Data were expressed as the mean7s.e.m. of at least three independent experiments.

E18 embryos were transfected with DBZ (152–301)IRES-GFP or with GFP alone immediately after plating, and the cells were immunostained for the immature neuronal marker bIII-tubulin to analyze the total neurite length 24 h after transfection. As in the case of PC12 cells, expression of DBZ (152–301) led to a significant decrease of neurite length without any change of cytotoxicity after transfection (Figure 5c

and d). There was also no significant difference of the transfection rate (GFP: 1.49%; DBZ (152-301)-IRESGFP: 1.32%).

Discussion DISC1 is a possible candidate gene for schizophrenia that is predominantly expressed in certain brain Molecular Psychiatry


406

regions, including the hippocampus.7,8,11 While investigating how DISC1 may confer susceptibility to psychiatric disorders, we identified DBZ as a novel DISC1-interacting protein. DBZ is identical to human KIAA0844, for which the mouse counterpart has been reported to show a high level of expression in the brain,28,29 but its function has not been characterized. The DBZ protein consists of 407 amino acids and is predicted to have a C2H2-type zinc-finger motif and coiled-coil domains. In mammalian cells (HEK293T and PC12 cells) and in rat brain extracts, DBZ interacted with DISC1, particularly with domains near the breakpoint of the (1;11) (q42.1;q14.3) translocation. Northern blotting and in situ hybridization revealed that DBZ mRNA was expressed in some human brain regions, and was prominently expressed in some rat brain regions including the cerebral cortex, hippocampus, olfactory tubercle and striatum. Possible involvement of DBZ in the formation of neurites and in the signaling pathway of PACAP, a neuropeptide regulating neuropsychological functions,18–23 was indicated by our present findings. DISC1 has been proposed to be a multifunctional protein that interacts with multiple proteins of the centrosome and cytoskeletal system at distinct domains.26 We have previously reported that DISC1 interacts and colocalizes with FEZ1 in growth cones in association with F-actin,11 and that a calmodulinbinding protein (Kendrin) anchors DISC1 to the centrosome.13 A co-immunoprecipitation study and a yeast two-hybrid assay suggested that amino acids 446–633 of DISC1 act as the FEZ1-binding region, whereas amino acids 348–597 show weak interaction with FEZ1.11 In addition, amino acids 446–533 of DISC1 were shown to be vital for interaction with Kendrin in a yeast two-hybrid assay.13 The present study demonstrated that both DISC1 (446–633)-FLAG and DISC1 (348–597)-FLAG could interact strongly with DBZ-GFP (Figure 1d). These results suggest that the regions of DISC1 near the translocation breakpoint participate in interactions with DBZ and FEZ1, and that DBZ, FEZ1 and Kendrin bind to overlapping (but different) regions of DISC1. DBZ (DBZ-GFP) showed a diffuse distribution in the cytoplasm and was mainly colocalized with DISC1 (DISC1-HA; Figure 2d). Taken together with the previous observation that DISC1 is colocalized with FEZ1 in growth cones and with Kendrin in the centrosome, the present results suggest that DISC1 interacts with different binding partners at various intracellular sites, thus further supporting its multifunctional nature. The interaction of DISC1 with FEZ1 in PC12 cells was markedly enhanced by exposure to NGF, along with neurite extension,11,14 whereas PACAP had no effect (our unpublished observation). On the other hand, PACAP caused a marked, but transient, decrease of association between DISC1 and DBZ in PC12 cells (Figure 3c) and NGF had a similar effect (data not shown). These results suggest that DISC1/DBZ interaction is regulated at least by PACAP, but further


investigation is needed to clarify the association between DISC1/DBZ interaction and other neurotrophic factors. In the adult rat brain, co-expression of DBZ and PAC1 mRNAs was detected in the granule cells of the dentate gyrus, where a high level of DISC1 mRNA expression is observed in both young and adult rat brains11 and PAC1 protein is localized in hippocampal mossy fiber terminals.21 Recently, we demonstrated that Adcyap1/ mice lacking PACAP exhibit marked behavioral abnormalities and other changes.18–20 Adcyap1/ mice are born in the expected Mendelian ratio, but show a high early mortality rate before weaning. The surviving mice display remarkable behavioral changes, including hyperlocomotion and jumping in an open field test, and increased noveltyseeking behavior, as well as defective hippocampal LTP and prepulse inhibition of the acoustic startle reflex. In addition, PAC1 receptor-deficient mice show increased locomotor activity and reduced anxiety-like behavior,22 abnormal social behavior23 and impaired LTP in the CA3 mossy fiber synapses.21 Our present observations imply a possible link between DISC1 and PACAP signaling via DBZ, which may suggest a new mechanism that contributes to psychiatric disorders. Although overexpression of either DBZ or DISC1 alone had no effect, overexpression of both proteins resulted in significant reduction in the number of neurite-bearing PC12 cells after PACAP stimulation (Figure 4), but not after NGF stimulation (data not shown). This result was unexpected, because NGF induced a similar change of DISC1/DBZ interaction to PACAP. The difference might have arisen because NGF not only influences DISC1/DBZ interaction but also DISC1/FEZ1 interaction. As coiled-coil domains are involved in interactions with other proteins or in oligomerization,31 we examined whether these domains of rat DBZ protein (amino acids 152–301) interact with DISC1 to effect neurite growth. Expression of DBZ (152–301)-IRES-GFP resulted in significantly shorter neurites of PC12 cells treated with PACAP and rat primary culture hippocampal neurons (DIV1) (Figure 5), whereas DISC1/DBZ interaction did not affect cell viability under our experimental conditions. These observations suggest that this region of DBZ affects the normal endogenous DISC1 interaction and that caused inhibition of neurite growth. Considering all the results, DISC1/DBZ interaction is dissociated at least by PACAP and it is necessary for PACAP-induced neurite growth. During the dissociation, DISC1 might change the binding partner(s) participating in neurite growth. DISC1 may be involved in several processes that are critical to neurodevelopment and neurological function, and because DBZ could have specific functions of its own, further studies are necessary to define the processes in which DBZ is involved. Studies of the proteins interacting directly with DISC1 (e.g. FEZ1, DBZ and Kendrin) as well as those with indirect interactions (e.g. PACAP and NGF) may open up a


new area of research into psychiatric disorders and provide additional insights into the pathogenesis of these conditions.

Acknowledgments We thank Ms Arakawa, Ms Moriya and Ms Ohashi for preparing our experiments. This research was partly supported by the 21st century COE program of the Ministry of Education, Culture, Sports, Science and Technology of Japan, by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science and by a grant from Sankyo Foundation of Life Science.

References 1 Weinberger DR. Implications of normal brain development for the pathogenesis of schizophrenia. Arch Gen Psychiatry 1987; 44: 660–669. 2 Lewis DA, Lieberman JA. Catching up on schizophrenia: natural history and neurobiology. Neuron 2000; 28: 325–334. 3 Frankle WG, Lerma J, Laruelle M. The synaptic hypothesis of schizophrenia. Neuron 2003; 39: 205–216. 4 Heinz A, Romero B, Gallinat J, Juckel G, Weinberger DR. Molecular brain imaging and the neurobiology and genetics of schizophrenia. Pharmacopsychiatry 2003; 36: S152–S157. 5 Mueser KT, McGurk SR. Schizophrenia. Lancet 2004; 363: 2063–2072. 6 Harrison PJ, Weinberger DR. Schizophrenia genes, gene expression, and neuropathology: on the matter of their convergence. Mol Psychiatry 2005; 10: 40–68. 7 Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, Semple CA et al. Disruption of two novel genes by a translocation co-segregating with schizophrenia. Hum Mol Genet 2000; 9: 1415–1423. 8 Millar JK, Christie S, Anderson S, Lawson D, Hsiao-Wei-Loh D, Devon RS et al. Genomic structure and localisation within a linkage hotspot of disrupted in schizophrenia 1, a gene disrupted by a translocation segregating with schizophrenia. Mol Psychiatry 2001; 6: 173–178. 9 Sachs NA, Sawa A, Holmes SE, Ross CA, DeLisi LE, Margolis RL. A frameshift mutation in disrupted in schizophrenia 1 in an American family with schizophrenia and schizoaffective disorder. Mol Psychiatry 2005; 10: 758–764. 10 Hodgkinson CA, Goldman D, Jaeger J, Persaud S, Kane JM, Lipsky RH et al. Disrupted in schizophrenia 1 (DISC1): association with schizophrenia, schizoaffective disorder, and bipolar disorder. Am J Hum Genet 2004; 75: 862–872. 11 Miyoshi K, Honda A, Baba K, Taniguchi M, Oono K, Fujita T et al. Disrupted-in-schizophrenia 1, a candidate gene for schizophrenia, participates in neurite outgrowth. Mol Psychiatry 2003; 8: 685–694. 12 Honda A, Miyoshi K, Baba K, Taniguchi M, Koyama Y, Kuroda S et al. Expression of fasciculation and elongation protein zeta-1 (FEZ1) in the developing rat brain. Brain Res Mol Brain Res 2004; 122: 89–92. 13 Miyoshi K, Asanuma M, Miyazaki I, Diaz-Corrales FJ, Katayama T, Tohyama M et al. DISC1 localizes to the centrosome by binding to kendrin. Biochem Biophys Res Commun 2004; 317: 1195–1199. 14 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 zetainteracting protein. J Cell Biol 1999; 144: 403–411.

15 Yamada K, Nakamura K, Minabe Y, Iwayama-Shigeno Y, Takao H, Toyota T et al. Association analysis of FEZ1 variants with schizophrenia in Japanese cohorts. Biol Psychiatry 2004; 56: 683–690. 16 Arimura A. Perspectives on pituitary adenylate cyclase activating polypeptide (PACAP) in the neuroendocrine, endocrine, and nervous systems. Jpn J Physiol 1998; 48: 301–331. 17 Vaudry D, Gonzalez BJ, Basille M, Yon L, Fournier A, Vaudry H. Pituitary adenylate cyclase-activating polypeptide and its receptors: from structure to functions. Pharmacol Rev 2000; 52: 269–324. 18 Hashimoto H, Shintani N, Tanaka K, Mori W, Hirose M, Matsuda T et al. Altered psychomotor behaviors in mice lacking pituitary adenylate cyclase-activating polypeptide (PACAP). Proc Natl Acad Sci USA 2001; 98: 13355–13360. 19 Tanaka K, Shintani N, Hashimoto H, Kawagishi N, Ago Y, Matsuda T et al. Psychostimulant-induced attenuation of hyperactivity and prepulse inhibition deficits in Adcyap1-deficient mice. J Neurosci 2006; 26: 5091–5097. 20 Matsuyama S, Matsumoto A, Hashimoto H, Shintani N, Baba A. Impaired long-term potentiation in vivo in the dentate gyrus of pituitary adenylate cyclase-activating polypeptide (PACAP) or PACAP type 1 receptor-mutant mice. NeuroReport 2003; 14: 2095–2098. 21 Otto C, Kovalchuk Y, Wolfer DP, Gass P, Martin M, Zuschratter W et al. Impairment of mossy fiber long-term potentiation and associative learning in pituitary adenylate cyclase activating polypeptide type I receptor-deficient mice. J Neurosci 2001; 21: 5520–5527. 22 Otto C, Martin M, Wolfer DP, Lipp HP, Maldonado R, Schutz G. Altered emotional behavior in PACAP-type-I-receptor-deficient mice. Brain Res Mol Brain Res 2001; 92: 78–84. 23 Nicot A, Otto T, Brabet P, Dicicco-Bloom EM. Altered social behavior in pituitary adenylate cyclase-activating polypeptide type I receptor-deficient mice. J Neurosci 2004; 24: 8786–8795. 24 Yamashita T, Shimada S, Guo W, Sato K, Kohmura E, Hayakawa T et al. Cloning and functional expression of a brain peptide/ histidine transporter. J Biol Chem 1997; 272: 10205–10211. 25 Matsuo N, Ogawa S, Takagi T, Wanaka A, Mori T, Matsuyama T et al. Cloning of a putative vesicle transport-related protein, RA410, from cultured rat astrocytes and its expression in ischemic rat brain. J Biol Chem 1997; 272: 16438–16444. 26 Morris JA, Kandpal G, Ma L, Austin CP. DISC1 (disrupted-inschizophrenia 1) is a centrosome-associated protein that interacts with MAP1A, MIPT3, ATF4/5 and NUDEL: regulation and loss of interaction with mutation. Hum Mol Genet 2003; 12: 1591–1608. 27 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. 28 Nagase T, Ishikawa K, Suyama M, Kikuno R, Hirosawa M, Miyajima N et al. Prediction of the coding sequences of unidentified human genes. XII. The complete sequences of 100 new cDNA clones from brain which code for large proteins in vitro. DNA Res 1998; 5: 355–364. 29 Gianfrancesco F, Esposito T, Ombra MN, Forabosco P, Maninchedda G, Fattorini M et al. Identification of a novel gene and a common variant associated with uric acid nephrolithiasis in a Sardinian genetic isolate. Am J Hum Genet 2003; 72: 1479–1491. 30 Hashimoto H, Nogi H, Mori K, Ohishi H, Shigemoto R, Yamamoto K et al. Distribution of the mRNA for a pituitary adenylate cyclaseactivating polypeptide receptor in the rat brain: an in situ hybridization study. J Comp Neurol 1996; 371: 567–577. 31 Rose A, Meier I. Scaffolds, levers, rods and springs: diverse cellular functions of long coiled-coil proteins. Cell Mol Life Sci 2004; 61: 1996–2009.

407
