Molecular Psychiatry (2007) 12, 620–629 & 2007 Nature Publishing Group All rights reserved 1359-4184/07 $30.00 www.nature.com/mp FEATURE REVIEW New genes associated with schizophrenia in neurite formation: a review of culture experiments A Bellon Menninger Department of Psychiatry & Behavioral Sciences, Baylor College of Medicine, Houston, TX, USA

New genes consistently associated with schizophrenia include NRG1, Akt, DISC-1 and dysbindin-1. Since these genes participate in neurotransmission, neuroplasticity and neurodevelopment it has not been easy to elucidate which of these roles are abnormal in patients with schizophrenia. Neurite formation is identified as a crucial stage in development, and it is proposed that a defect in neurite formation originating from abnormally encoded proteins by these new genes could be at least an in vitro marker that reflects the most consistent molecular and neuroanatomical findings in schizophrenia. A systematic review of the literature linking the process of neurite formation to genes with replicated evidence that supported their association with schizophrenia was conducted. In addition, an outline of the process of neurite formation was included. Neurite formation was shown to be induced by neuregulins, the product of the gene NRG1. The activation of Akt, a serine/threonine kinase, promoted neurite formation in six independent studies. Conversely, two studies found that Akt inhibits neurite outgrowth. Stronger evidence supporting an association with the new genes related to schizophrenia and neurite formation comes from DISC-1. Defects in DISC-1 protein were shown to directly alter the process of neurite formation. Dysbindin-1 has not yet been directly implicated in neurite outgrowth. These findings suggest that the proteins encoded by NRG1, Akt and DISC-1 are implicated in the process of neurite formation in cellular models as well as, at least in part, animal models during development. Abnormalities in this process could have potential etiologic implications for schizophrenia. Direct evidence, however, of abnormal neurite formation in patients with schizophrenia is still missing. Limitations to this model are identified. Molecular Psychiatry (2007) 12, 620–629; doi:10.1038/sj.mp.4001985; published online 17 April 2007 Keywords: neuropil; development; NRG1; Akt; DISC-1; dysbindin-1

Introduction recognized that a decrease in brain volume accom- panied by ventricular enlargement could result from Mounting evidence indicates that schizophrenia is a decreased and .25 This theory is developmental disease that originates from a complex known as the ‘reduced neuropil hypothesis’.25–28 combination of genetic information and environ- Evidence that supports this line of thinking comes mental factors.1–6 Multiple genes have recently been from findings of increased density of in the associated with schizophrenia. NRG1, Akt, DISC-1 cortex of patients with schizophrenia.25,28 This in- and dysbindin-1 are new genes with replicated data creased density of neurons is accompanied by supporting their association with this illness in reduced interneuronal space without changes in the different populations.7–12 The contribution of these number of neurons.26,27 Since the interneuronal space genes, however, to the pathophysiology of schizo- is occupied by dendrites and axons known together as phrenia is yet to be determined13–16 since NRG1, neuropil, a decreased interneuronal space indicates Akt, DISC-1 and dysbindin-1 participate in several that there is a decreased neuropil.25–28 Further neuronal mechanisms. evidence that is consistent with the ‘reduced neuropil The most consistent findings in schizophrenia are hypothesis’ comes from measurements of neuronal ventricular enlargement and decreased brain volume size as it correlates with the extension of the involving the hippocampus and cortex.17–24 These dendritic tree and the length.25,29,30 There are changes can result from white matter loss or even replicated findings of reduced neuronal soma size in fluid shifts that affect cellular volume. It is also the hippocampus of patients with schizophrenia.31–33 Molecular findings and direct measurements of Correspondence: Dr A Bellon, Menninger Department of Psychia- neuronal processes also support this association try & Behavioral Sciences, Baylor College of Medicine, One Baylor between schizophrenia and a reduction in axons Plaza, MS: BCM350, Houston, TX 77030, USA. and dendrites. The dendritic markers MAP2 and E-mail: [email protected] Received 15 August 2006; revised 22 February 2007; accepted 27 MAP1B were found to be significantly reduced in the February 2007; published online 17 April 2007 subiculum and entorhinal cortex in studies of post- New genes associated with schizophrenia in neurite formation A Bellon 621 mortem patients with schizophrenia.34 In addition, have been implicated in neurite formation consist loss of MAP2 was also reported in the medial mainly of growth factors. Among these are nerve prefrontal cortex (PFC) of patients with this same growth factor (NGF) and insulin-like growth factor disorder.35 A decrease in dendritic length was (IGF).44–46 Hormones like estrogens, cortisol, growth encountered by a 3D image analysis on the PFC of hormone, melatonin, T3 and T4 also participate in patients with schizophrenia36 and similar results neuritogenesis as extrinsic factors as well.47–51 The were evidenced in the hippocampus.37 extracellular environment influences neuronal differ- Reduced interneuronal space, decreased dendritic entiation through tenascins, hyaluronan and the markers, smaller neuronal soma sizes and a reduction adhesion molecules collagen and laminin among in dendritic length are all findings that support the others.52–54 ‘reduced neuropil hypothesis’.25 This hypothesis Several types of receptors such as Trk interact with could provide a neuroanatomical explanation for the the environment of the , inducing cell mem- aberrant connectivity between structures that has brane changes mediated by increases in calcium and been proposed as the origin of psychotic symp- other second messenger events55,56 that result in toms.38–41 What has not been explained is how this neurite protrusion. Da Silva and Dotti57 proposed aberrant connectivity comes about. that neurite sprouting might be determined by the One possibility for this aberrant disconnectivity, to presence of microdomains or ‘hot spots’ on the be explored in this paper, is that the genes recently plasma membrane. The difference between these found to be associated with schizophrenia encode portions of the membrane and the rest of the bilipid defective proteins that alter the process of neurite layer could reside in the type of receptors and formation and that this could constitute an etiological associated proteins that they express. factor for schizophrenia. Therefore, the first goal of The structural changes begin with the formation at this paper is to review the process of neurite the edge of the neuron of a lamellipodium,58 a sheet- formation exposing downstream cascades that could like extension that consists of an meshwork be potentially affected by proteins encoded by the (Figure 1). This lamellipodium produces the neces- genes recently associated with this illness. The sary adhesion and tension for the neuron to start the second and principal goal is to determine whether process of neurite sprouting.57 inside there is evidence in the literature that these new the immature cell rearrange in direction of this genes, namely NRG1, Akt, dysbindin-1 and DISC-1, newly formed lamellipodium.59,60 A neurite shaft is are implicated in this fundamental process of neuro- formed when the lamellipodium is separated from nal development. the membrane cell by the microtubules behind it (Figure 1). A growth cone is formed once microtu- bules, carrying with them mitochondria, vesicles and Methods other invade the lamellipodium.61 At the A number of biomedical databases were searched periphery of the growth cones long thin protrusions including MEDLINE and Cochrane. The main search composed of F-actin bundles are formed.57,62,63 These terms used included neurite formation, NRG1, Akt, outgrowth bundles are known as fillopodium and Dysbindin-1, DISC-1 and schizophrenia. The search their role is to sense and respond to environmental was limited to studies published in the English cues as well as to provide guidance for language. No restrictions were applied for type of growth inside the growth cone.64 Actin polymeriza- model utilized such as animal or cellular or year of tion at the leading edge of the growth cone moves the publication. In addition, to provide an overview neurite forward.65,66 Myosin on the other hand moves about neurite formation, articles related to this actin filaments to the center of the growth cone67 process were sought and included when applicable. where they are depolymerized into actin monomers Finally, the reference lists of selected articles were that can be reutilized for actin polymerization and scanned for additional references. further formation of fillopodia. Neurite elongation prevails as long as actin polymerization continues Neurite formation and microtubules invade the periphery of the growth Function of the central nervous system (CNS) relies cone.42 on neuronal structure, which allows interconnections Actin polymerization is regulated by the Rho family among cells and therefore the transmission of in- of small GTPases, which mainly consists of RhoA, formation. For neurons to be connected they need to Rac and Cdc42.63 Each small GTPase has been develop an intricate structure based on neuronal associated with actin polymerization that leads to processes known as dendrites and axons. These different structures. For instance, Rac has been neuronal processes originate from immature exten- related to lamellipodium formation68 whereas Cdc42 sions called neurites. was found to induce fillopodium.69 The process of neurite formation is influenced by the The normal development of a neurite requires the genetic information that the cell carries, considered interaction between formed by actin intrinsic factors. Extrinsic factors, which constitute the and microtubules formed by . How this interactions of the cell with its environment, also interaction takes place remains to be elucidated, contribute to this process.42,43 The extrinsic factors that although microtubule-associated proteins (MAPs)

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 622 a development while MAP1A is more abundant in mature neurons.78 microtubule In summary, the process of neurite formation involves several structures including the cytoskele- ton, and some of its component parts such as microfilaments, microtubules, MAPs like MAP1, MAP2 and tau. This process appears to be highly b lamellipodium regulated with a machinery that we are just beginning to understand.

NRG1 Five studies done independently with different populations of patients have found a relationship between NRG1, which is a protein structurally related 79 c to the epidermal growth factor family and schizo- fillopodium phrenia.80–84 Even though no functional mutation has been identified,80–86 a recent study reported abnorm- growth cone alities in NRG1 mRNA isoforms in the PFC of schizophrenic patients,87 suggesting a defect on the actin NRG1 gene. Moreover, erbB3, one of the three receptors currently identified as transducers of neur- egulin (NRG) signaling,79 is decreased by almost d 50% in patients with chronic schizophrenia.88 Abnor- malities in NRG1 and erbB could compromise the NRG–erbB signaling, and as a result the process in which this signaling takes place.89 Among the variety of functions NRGs posses in the CNS, their role in neurite formation stands out since it is a crucial stage in neuronal development. Without neurites, which are immature forms of dendrites and axons, neurons Figure 1 Cytoskeletal dynamics during neurite formation. (a) Neuroblast. (b) Lamellipodium formation through actin are unable to accomplish their ultimate goal of reorganization at the edge of the neuroblast. (c) Micro- receiving and transmitting information. The process tubules inside the neuroblast rearrange toward the newly of neurite formation is complex and the exact role of formed lamellipodium while actin-formed filopodia appear NRGs is still to be determined. There are however, at the edge of the cell. (d) Microtubules and actin several studies that link these molecules to the polymerization push the lamellipodium away form the cell process of neurite formation, some of them during body leaving behind a neurite shaft. The lamellipodium early phases of development. once separated by a neurite shaft forms the edge of the cell is NRGs present in a cultured-sciatic nerve condi- known as growth cone. tioned medium induced a 20% increased in short neurites and a 40% increased in long neurites of PC12 pheochromocytoma cells.90 NRG1 and NRG3 were the are thought to be the link between these two suggested isoforms of NRGs involved in this pro- cytoskeletal components.61 The specific role of MAPs cess.90 A different study, also done in PC12 cells (including tau, a low-molecular weight MAP) remains which is a widely accepted model for neurodevelop- unclear but their presence appears to be vital for ment, showed that the activation of receptors erbB2 neuritogenesis as depletion of MAP2 and tau has been and erbB3 by NRG promoted neurite outgrowth.91 found to inhibit neurite formation in cerebellar Erb4, another NRG receptor has also been implicated neurons.70,71 In addition, blocking the expression of in neurite outgrowth through the activation of the MAP2 decreases growth in vivo72 while mitogen-activated protein kinase (MAPK) as well as generating low levels of tau can inhibit axon growth.73 protein kinase C (PKC) in PC12 cells.92 The down- It is known that MAP2 and tau promote and stabilize stream cascade of MAPK and PKC was not tested microtubules polymerization,42 which is a fundamen- although there is evidence from other studies that the tal process for neurite formation.61 Transfection of activation of Cdc42 by PKC-induced fillopodium non-neuronal cells with the genes that express MAP2 formation.93 Activation of Rac by this same protein induces the formation of neurite-like processes74 kinase resulted in actin polymerization leading to the while the transfection of non-neuronal MAPs like formation of lamellipodium.94 Both filopodium and MAP4 does not result in any kind of outgrowth.75 lamellipodium are actin structures that precede MAP1 also appears to be a key regulator of neurite neurite formation.58,61 outgrowth76 as it also promotes microtubule polymer- NRGs also increased neuronal processes length in ization.77 MAP1B appears to be directly involved in cerebellar granule cells and it was suggested that this neuritogenesis since its expression is highest during process could be regulated in vivo through the

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 623 activation of the erbB family receptors.95 The activa- Defining the true relationship between schizophre- tion of erbB2, erbB3 and erbB4 by NRGs has been nia and Akt is still a work in progress. In an attempt to shown to induce neurite formation in cell cultures of open new avenues for research this section will focus embryonic and newborn rat retinal cells,96 supporting solely on the association of this serine/threonine a possible role of these molecules in the regulation of kinase and the process of neurite formation.112–119 neurite outgrowth in vivo. This by no means should be considered as an It has also been reported, form a study done with exclusion of all the other possible roles of Akt in pluripotential cells, that NRGs inhibit neurite out- different cell functions which have potential implica- growth.97 Arguments have been raised criticizing tions in the pathophysiology of schizophrenia. these results on the basis that the model did not The earliest study directly involving Akt in the allow for the exclusion of other molecules in the process of neurite formation suggested that Akt putative NRG inhibition of neurite outgrowth.95 The promoted neurite elongation in vitro and in vivo112 use of pluripotential cells could induce a wide variety through possible downstream mechanisms involv- of phenotypes making the results uncertain. ing cAMP response element-binding protein120 and In summary, despite NRGs either inducing or GSK-3.121 In addition, it was determined that this inhibiting neurite formation the studies presented serine/threonine kinase promotes axon regeneration here suggest that NRGs may have a role in the pro- in vivo in adult rats.112 cess of neurite formation, particularly during CNS Overexpression of Akt in embryonic dorsal root development. ganglion neurons resulted in an enlarged axon caliber and in an increased number of axon branches.113 AKT1 These changes were explained by a possible interac- Convergent evidence recently obtained led to Emamian tion between Akt and Rac 1.122 Rac 1 is a small Rho et al.14 to point to Akt1 as another gene associated GTPase capable of regulating the actin with schizophrenia. Since then three independent during neurite outgrowth.63 This Rho GTPase was groups have replicated this evidence in different found to induce lamellipodium formation upon populations.98–100 Furthermore, two of the studies activation.123 Inhibition of GSK-3 by Akt124 was also found associations with schizophrenia and the same suggested as a possible mechanism of microtubule haplotype composed of SNP2/SNP3/SNP4 (TCG) in polymerization resulting in axon branching.125 patients with European origin.14,99 There are however, The IGF-1 and the NGF have been associated with three other studies with Asian populations that found the activation of the PI3/Akt pathway leading to no association between Akt1 and schizophrenia.101–103 neurite outgrowth.114,116 The activation of this path- There is still controversy about how promising Akt1 way in hippocampal cells induced membrane assem- is as a susceptibility gene, although Akt1 has been bly at the axonal growth cone, a key parameter included in several recent reviews that concern the of axonal growth.126–130 Whereas in PC12 cells the pathophysiology of schizophrenia.7,9,10,13 PI3/Akt signaling mechanism was accompanied by Akt1 protein serves as a serine/threonine kinase an increase in protein NF68 a marker implicated in a variety of cell functions.104–106 One of of neuronal differentiation.114,131,132 Three studies the main and most widely recognized functions of done with PC12 cells114,115,117 and one done with Akt is -induced cell survival through the hippocampal cells116 found that Akt activation was phosphorylation and deactivation of glycogen followed by GSK-3 phosphorylation which as a result synthase kinase-3 (GSK-3).104,105 Indeed abnormalities can lead to microtubule polymerization112 and con- altering the Akt/GSK-3 signaling cascade result in sequently, neurite formation. It has also been pro- increased of neurons including hippocam- posed that the PI3/Akt pathway activates Cdc42,116 a pal stem cells.107,108 Increased apoptosis due to member of the Rho GTPases family. Cdc42 activation abnormalities in Akt signaling could contribute to induces fillopodium formation,69 one of the structures the pathophysiology of schizophrenia.109 Further- that precedes neurite outgrowth. more, Akt has been linked to glutamatergic and Contrary to the studies presented above, two GABAergic (g-aminobutyric acid) neurotransmis- independent studies have reported that overexpres- sion.110,111 Abnormalities in these two neurotransmit- sion of Akt resulted in neurite outgrowth inhibition in ters have been described in postmortem studies of PC12 cells.133,134 Another study done with PC12 cells patients with schizophrenia and it is believed that also found that Akt activation resulted in inhibition of GABA and glutamate play a significant role in the neurite branching accompanied, however, by neurite pathophysiology of this illness.110,111 In addition to elongation.118 These results contrasted with previous Akt functions in cell survival, apoptosis and neuro- reports in which Akt activation was associated with transmission, this serine/threonine kinase is also increased branching in mouse sensory neurons.113 involved in cell proliferation, glucose metabolism, The authors suggested that the contradictory findings angiogenesis and cell motility.106 Due to the multiple occurred as a consequence of the multiple signals that core roles of Akt in different cell functions it is likely can result following the activation of Trk receptors that Akt’s implication in the pathophysiology of by NGF.118 schizophrenia, if any, would involve several aspects A theoretical explanation for this contradictory of cell dynamics. result can be explained by Jiang and colleagues

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 624 results.135 They suggested that GSK-3b inhibition by Further support of the interaction between DISC-1 Akt is necessary for axon development but not for and the cytoskeleton comes form a study by Miyoshi dendrite formation135 in a model of neuronal polarity et al.,144 who found that neuronal process formation with mammalian neurons from the hippocampus.136 was enhanced by the overexpression of DISC-1 protein Therefore, it is possible that studies reporting inhibi- in PC12 cells. The authors provided evidence that tion of neurite formation by Akt were measuring DISC-1 protein interacts with fasciculation and elonga- neurites that were to become dendrites while the tion protein zeta-1 (FEZ-1), which is a protein involved opposite can be suggested when the neurites ana- in axon elongation.146,147 This interaction was asso- lyzed were to become axons or that axon formation ciated with the actin cytoskeleton and its inhibition prevents dendrite development at least in some stages resulted in abnormal neurite outgrowth.144 It was of differentiation.58 Another possible explanation concluded that DISC-1 is a cytoskeletal-associated would be that the PC12 cells studied were in different protein highly expressed during rat development stages of development which directly influences the which mutation could compromise its interaction with type of neurite to be formed, whether axons or FEZ-1 and the actin cytoskeleton ultimately resulting dendrites.58 in psychiatric disorders like schizophrenia.144 In summary, Akt appears to be involved in the In summary, it appears that a mutation on DISC-1 regulation of neurite formation. GSK-3b inhibition could compromise neurite outgrowth since DISC-1 resulting in microtubule polymerization and the protein participates in this developmental process in activation of small Rho GTPases appears the most PC12 cells either through the interaction with the likely downstream signal for this serine/threonine actin cytoskeleton or via microtubular dynamics. kinase in this developmental process. The Akt role in the process of neurite formation is still to be Dysbindin-1 determined. This role would only be one of the many Certain haplotypes of single nucleotide polymorphisms cellular functions in which Akt is involved. in the dysbindin-1 (dystrobrevin-binding protein 1) gene have consistently been associated with schizo- DISC-1 phrenia by several independent groups within different A balanced translocation (1;11)(q42.1;q14.3) that populations.82,148–156 The role of dysbindin-1 protein segregates with schizophrenia and related disorders remains poorly understood but the association of this was recently found in a large Scottish family.137,138 protein with schizophrenia initiated an intensive This translocation affects the DISC-1 gene and results search for new mechanisms of action in addition to in a defective DISC-1 protein that lacks a carboxy- its possible role in the pathogenesis of muscular terminal.15,139 DISC-1 has also been identified as a dystrophy.157 potential etiological gene by linkage and association Dysbindin has been linked to the PI3K-Akt signal- studies in Finish families.140 Furthermore, a case- ing pathway. Numakawa et al.156 showed that control study of a North American white population decreased dysbindin protein in cortical neuronal encountered that variation at the DISC-1 locus cultures attenuated Akt signaling and as a result predisposes individuals to schizophrenia, schizoaf- increased neuronal death. It is still to be determined if fective and bipolar disorders.141 Finally, another Akt signaling disruption by dysbindin also affects study reported a frameshift mutation in DISC-1 in a neurite formation. Currently, there is no association family with schizophrenia and schizoaffective dis- between dysbindin-1 and the process of neurite order further supporting the evidence that DISC-1 can outgrowth. confer susceptibility to schizophrenia and other 139 thought disorders. Discussion The role of DISC-1 protein is not entirely under- stood but is thought to be involved in neuronal The recent discovery of several susceptibility genes migration, synaptogenesis and glutamatergic trans- implicated in schizophrenia opens new possibilities mission.142–145 In addition, three recent studies done for the understanding of this disorder. What has been independently have directly related DISC-1 to the shown in this paper is that proteins encoded by process of neurite formation.15,16,144 NRG1, Akt1 and DISC-1 are implicated in the process Transfection of mutant DISC-1 into PC12 cells of neurite formation in cellular and animal models of resulted in shorter neurites16 while decreasing the neurodevelopment. Establishing if there are any expression of DISC-1 by RNA interference-inhibited abnormalities in the process of neurite formation in neurite outgrowth in these same cells.15 Moreover, in patients with schizophrenia would be the next step in utero gene-transfer techniques have shown that DISC-1 determining any potential etiologic implications of loss induces shorter dendrites randomly oriented in this crucial step in development to schizophrenia. the of rat embryos. The proposed The genetic component of schizophrenia appears to mechanism of action that induced abnormalities in be rather complex and does not follow a common the process of neurite formation was DISC-1 failure to Mendelian pattern of heritability. A multilocus model interact properly with the cytoskeleton.16 Adeficient has been proposed to explain the pattern of herit- interaction between DISC-1 and the cytoskeleton ability of this enigmatic illness.158 This model appears to result in defective microtubular dynamics.15 proposes that a combination of multiple genetic

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 625 factors is necessary for schizophrenia to occur.158 prefrontal cortex (DLPFC).175 An explanation for this Therefore it is likely that more than one gene is excessive pruning in patients with schizophrenia is implicated in the development of schizophrenia. still lacking but is possible that it originates in Consistent with this model is the data currently defective neurite formation as abnormal neurites available that link the function of NRG with Akt as might be prone to extensive removal. There is not NRG appears to activate the Akt signaling cas- enough information to draw any definite conclusions cade.159,160 In addition, dysbindin has also been in the role of encoded proteins by the new genes linked to the activation of Akt156 and it was very associated with schizophrenia and the excessive recently shown that knockdown of DISC-1 resulted in pruning observed in these patients. Nevertheless, it inhibition of Akt phosphorylation.161 These signaling is noteworthy that the extensive removal of synaptic pathways that involve more than one of the new genes connections with potential etiologic implications for associated with schizophrenia have not yet been schizophrenia as proposed by Lewis and Gonzalez- related to the process of neurite formation. The Burgos occurs in the same area of the brain namely complexity, however, of the pathophysiology of the PFC where decreased levels of NRG and Akt have schizophrenia suggests that a process rather than a been found.14,87,175 It is still to be determined if this single defect or abnormality is altered in this illness excessive pruning occurs only in specific areas of the making a neurite-based pathophysiological model brain or if it affects just certain layers such as layer 3 worth exploring. of the DLPFC. The multiple loci model of heritability for schizo- Disconnectivity between brain structures has also phrenia also suggests that environmental variables been proposed as a potential factor implicated on the can impact the development of this illness.158 Identi- basis of psychosis.38,40,41 Abnormalities in the process fied environmental factors that have been linked to of neurite formation are consistent with this line of the development of schizophrenia such as maternal thinking as defective neurites might be unable to infection during pregnancy and stress also affect reach their targets, consequently affecting the estab- neurite outgrowth.162–164 Maternal infection during lishment of normal connections. Nevertheless, with pregnancy is associated with increased inflammatory the information currently available this possibility cytokines IL-1b, IL-6 and TNFa,165–167 which in turn could only stand as a hypothesis. Further research in have been shown to reduce dendrite development in this direction is needed to establish whether the cortical neurons of embryonic rats.168 Consequently, it process of neurite formation is at all implicated in the has been proposed that the higher incidence of disconnectivity between brain structures. schizophrenia observed in the offspring after prenatal In conclusion, the link between the proteins maternal infection162,163 could result from abnormal encoded by NRG1, DISC-1 and Akt and the process dendritic development due to these increased levels of neurite formation in patients with schizophrenia is of inflammatory cytokines.168 Stress has also been still to be determined, because the evidence presented implicated in defective neurite formation. Animal here comes mostly from cellular models only. No models of stress induce schizophrenia-like responses conclusions can be reached yet with the information in auditory gating.169 This animal model of stress was presented here as data obtained from in vitro models recently found to induce a decrease in dendritic do not necessarily apply to brain conditions in vivo.It length of pyramidal neurons from the hippocampus should also be appreciated that the neurite-based of rats.170 pathophysiological model presented here only pro- A delicate balance between growth and elimination vides one plausible etiological model for schizophre- of synaptic connections appears to take place nia; a thorough understanding of this complex illness throughout life in response to learning and experi- should come from different lines of thinking. Further ence.63 Elimination of synaptic connections peaks research needs to be pursued linking the new genes during adolescence as part of normal development associated with schizophrenia to the process of and is likely to refine brain performance.171 It is also neurite formation. First, abnormalities in this funda- adolescence when schizophrenia typically appears.172 mental process for development in patients with A link between excessive removal of neuritic connec- schizophrenia need to be determined. Second, the tions and auditory hallucinations was proposed by relative contribution of defective neurite formation to Hoffman and McGlashan171 based on their computer the pathophysiology of this illness has yet to be model of neuronal networks. Moreover, data coming established. from current neuroimaging techniques173,174 and postmortem studies25–28 suggest that there is in- Acknowledgments creased pruning in patients with schizophrenia. The process of pruning synaptic connections, however, is I thank Dr John H Coverdale for his encouragement still poorly understood. and thorough editorial comments. There have been several attempts to identify whether pruning affects certain areas of the brain or if specific tracts are targeted. It has been proposed that References intrinsic systems are affected to a greater extent than 1 Weinberger DR. The pathogenesis of schizophrenia: a neuro- associational systems in layer 3 of the dorsolateral developmental theory. In: Nasrallah HAW, Weinberger DR (eds).

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 626 The Neurology of Schizophrenia. Elsevier: Amsterdam, 1986, resonance imaging: a meta-analytic study. Arch Gen Psychiatry pp 397–406. 1998; 55: 433–440. 2 Weinberger DR. Implications of normal brain development for the 25 Selemon LD, Rajkowska G, Goldman-Rakic PS. Abnormally high pathogenesis of schizophrenia. Arch Gen Psychiatry 1987; 44: neuronal density in schizophrenic cortex: a morphometric 660–669. analysis of prefrontal area 9 and occipital area 17. Arch Gen 3 Weinberger DR. From neuropathology to neurodevelopment. Psychiatry 1995; 52: 805–818. Lancet 1995; 346: 552–557. 26 Selemon LD, Goldman-Rakic PS. The reduced neurophil hypoth- 4 Harrison PJ. The neuropathology of schizophrenia: a critical esis: a circuit based model of schizophrenia. Biol Psychiatry review of the data and their interpretation. Brain 1999; 122: 1999; 45: 17–25. 593–624. 27 Selemon LD. Increased cortical neuronal density in schizophre- 5 Marenco S, Weinberger DR. The neurodevelopmental hypothesis nia. Am J Psychiatry 2004; 161:169. of schizophrenia: following a trail of evidence from cradle to 28 Buxhoeveden D, Roy E, Switala A. Reduced interneuronal space grave. Dev Psychopathol 2000; 12: 501–527. in schizophrenia. Biol Psychiatry 2000; 47: 681–683. 6 Eastwood SL. The synaptic pathology of schizophrenia: is 29 Gilbert CD, Kelly JP. The projections of cells in different layers of aberrant neurodevelopment and plasticity to blame? Int Rev the cat’s visual cortex. J Comp Neurol 1975; 63: 81–106. Neurobiol 2004; 59: 47–72. 30 Lund JS, Lund RD, Hendrickson AE, Bunt AH, Fuchs AF. 7 Arnold SE, Talbot K, Hahn CG. Neurodevelopment, neuroplasti- The origin of efferent pathways from the primary visual cortex, city, and new genes for schizophrenia. Prog Brain Res 2005; 147: area 17, of the macaque monkey as shown by retrograde trans- 319–345. port of horseradish peroxidase. J Comp Neurol 1975; 164: 8 Owen MJ, Williams NM, O’Donovan MC. The molecular genetics 287–304. of schizophrenia: new findings promise new insights. Mol 31 Zaidel DW, Esiri MM, Harrison PJ. Size, shape, and orientation of Psychiatry 2004; 9: 14–27. neurons in the left and right hippocampus: investigation of 9 Norton N, Williams HJ, Owen MJ. An update on the genetics of normal asymmetries and alterations in schizophrenia. Am J schizophrenia. Curr Opin Psychiatry 2006; 19: 158–164. Psychiatry 1997; 154: 812–818. 10 Maier W, Zobel A, Kuhn KU. Clinical impact of recently detected 32 Arnold SE, Franz BP, Gur RC, Gur RE, Shapiro RM, Moberg PJ et susceptibility genes for schizophrenia. Dialog Clin Neurosci al. Smaller neuron size in schizophrenia in hippocampal 2006; 8: 79–84. subfields that mediate cortical–hippocampal interactions. Am J 11 Cannon TD, Hennah W, van Erp TGM, Thompson PM, Lonnqvist Psychiatry 1995; 152: 738–748. J, Huttunen M et al. Association of DISC1/TRAX haplotypes with 33 Benes FM, Sorensen I, Bird ED. Reduced neuronal size in schizophrenia, reduced prefrontal gray matter, and impaired posterior hippocampus of schizophrenic patients. Schizophr Bull short-and long-term memory. Arch Gen Psychiatry 2005; 62: 1991; 17: 597–608. 1205–1213. 34 Arnold SE, Lee VM-Y, Gur RE, Trojanowski JQ. Abnormal 12 Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, expression of two microtubule-associated proteins (MAP2 and Semple CAM et al. Disruption of two novel genes by a MAP5) in specific subfields of the hippocampal formation in translocation co-segregating with schizophrenia. Hum Mol Genet schizophrenia. Proc Natl Acad Sci USA 1991; 88: 10850–10854. 2000; 9: 1415–1423. 35 Jones LB, Johnson N, Byne W. Alterations in MAP2 immunocy- 13 Harrison PJ, Weinberger DR. Schizophrenia genes, gene expres- tochemistry in area 9 and 32 of schizophrenic prefrontal cortex. sion, and neuropathology: on the matter of their convergence. Psychiatry Res 2002; 114: 137–148. Mol Psychiatry 2005; 10: 40–68. 36 Kalus P, Mu¨ ller JT, Zuschratter W, Senitz D. The dendritic 14 Emamian E, Hall D, Birnbaum MJ, Karayiorgou M, Gogos JA. architecture of prefrontal pyramidal neurons in schizophrenic Convergent evidence for impaired AKT1-GSK3b signaling in patients. Neuroreport 2000; 11: 3621–3625. schizophrenia. Nat Genet 2004; 36: 131–137. 37 Rosoklija G, Toomayan G, Ellis SP, Keilp J, Mann JJ, Latov N et al. 15 Kamiya A, Kubo K, Tomoda T, Takaki M, Youn R, Ozeki Y et al. Structural abnormalities of subicular dendrites in subjects with A schizophrenia-associated mutation of DISC1 perturbs cerebral schizophrenia and mood disorders: preliminary findings. Arch cortex development. Nat Cell Biol 2005; 7: 1167–1178. Gen Psychiatry 2000; 57: 349–356. 16 Ozeki Y, Tomoda T, Kleiderlein J, Kamiya A, Bord L, Fuji K et al. 38 Friston KJ. The disconnection hypothesis. Schizophr Res 1998; Disrupted-in-schizophrenia-1 (DISC-1): mutant truncation pre- 30: 115–125. vents binding to NudE-like (NUDEL) and inhibits neurite 39 Friston KJ, Frith CD. Schizophrenia: a disconnection syndrome? outgrowth. Proc Natl Acad Sci USA 2003; 100: 289–294. Clin Neurosci 1995; 3: 89–97. 17 Van Horn JD, McManus IC. Ventricular enlargement in schizo- 40 Andreasen NC, O’Leary DS, Cizadlo T, Arndt S, Rezai K, Pronto LL phrenia. A meta-analysis of studies of the ventricle:brain ratio. et al. Schizophrenia and cognitive dysmetria: a positron-emission Br J Psychiatry 1992; 160: 687–697. tomography study of dysfunctional prefrontal-thalamic-cerebellar 18 Lawrie SM, Abukmeil SS. Brain abnormality in schizophrenia. circuitry. Proc Natl Acad Sci USA 1996; 93: 9985–9990. A systematic and quantitative review of volumetric magnetic 41 Bullmore ET, Woodruff PW, Wright IC, Rabe-Hesketh S, Howard resonance imaging studies. Br J Psychiatry 1998; 172: 110–120. RJ, Shuriquie N et al. Does dysplasia cause anatomical dis- 19 Andreasen NC, Flashman L, Flaum M, Arndt S, Swayze V II, connectivity in schizophrenia? Schizophr Res 1998; 30: 127–135. O’Leary DR et al. Regional brain abnormalities in schizophrenia 42 Goldberg JL. How does an axon grow? Genes Dev 2003; 17: measured with magnetic resonance imaging. JAMA 1994; 272: 941–958. 1763–1769. 43 McAllister AK. Cellular and molecular mechanisms of dendrite 20 Baare WFC, van Oel CJ, Hulshof Pol HE, Schnack H, Sitskoorn growth. Cereb Cortex 2000; 10: 963–973. MM, Kahn RX. Volumes of brain structures in twins discordant 44 Tucker KL, Meyer M, Barde YA. are required for for schizophrenia. Arch Gen Psychiatry 2001; 58: 33–40. nerve growth during development. Nat Neurosci 2001; 4: 29–37. 21 Gur RE, Mozley PD, Shtasel DL, Cannon TD, Glaacher F, Turetsky 45 Labelle C, Leclerc N. Exogenous BDNF, NT-3 and NT-4 differen- B et al. Clinical subtypes of schizophrenia: differences in brain tially regulate neurite outgrowth in cultured hippocampal and CSF volume. Am J Psychiatry 1994; 151: 343–350. neurons. Brain Res Dev Brain Res 2000; 123: 1–11. 22 Woods BT, Ward KE, Johnson EH. Meta-analysis of the time- 46 Mendell LM, Munson JB, Arvanian VL. Neurotrophins and course of brain volume reduction in schizophrenia: implications synaptic plasticity in the mammalian spinal cord. J Physiol for pathogenesis and early treatment. Schizophr Res 2005; 73: 2001; 533: 91–97. 221–228. 47 Clendinnen BG, Eayrs JT. The anatomical and physiological 23 Wright IC, Rabe-Hesketh S, Woodruff PWR, David AS, Murray effects of prenatally administered somatotrophin on cerebral RM, Bullmore ET. Meta-analysis of regional brain volumes in development in rats. J Endocrinol 1961; 22: 183–193. schizophrenia. Am J Psychiatry 2000; 157: 16–25. 48 Francon J, Fellous A, Lennon AM, Nunez J. Is thyroxine a 24 Nelson MD, Saykin AJ, Flashman LA, Riordan JH. Hippocampal regulatory signal for neurotubule assembly during brain devel- volume reduction in schizophrenia as assessed by magnetic opment? Nature 1977; 266: 188–190.

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 627 49 Nunez J. Microtubules and brain development: the effects of (MAPs) in Chinese hamster ovary cells: evidence for differing thyroid hormones. Neurochem Int 1985; 7: 959–968. roles of MAPs in microtubule organization. J Cell Biol 1994; 126: 50 Hargreaves A, Yusta B, Aranda A, Avila J, Pacual A. Triiodothyr- 1017–1029. onine (T3) induces neurite formation and increases synthesis of a 76 Drubin DG, Feinstein SC, Shooter EM, Krischner MW. Nerve protein related to MAP1B in cultured cells of neuronal origin. growth factor-induced neurite outgrowth in PC12 cells involves Brain Res 1988; 466: 141–148. the coordinate induction of microtubule assembly and assembly- 51 Benitez-King G, Huerto-Delgadillo L, Anton-Tay F. Melatonin promoting factors. J Cell Biol 1985; 101: 1799–1807. effects on the cytoskeletal organization of MDCK and neuroblas- 77 Drubin D, Kobayashi S, Kellogg D, Krischner M. Regulation toma N1E-115 cells. J Pineal Res 1990; 9: 209–220. of microtubule protein levels during cellular morphogenesis 52 Rauvala H, Peng HB. HB-GAM (heparin-binding growth-asso- in nerve growth factor-treated PC12 cells. J Cell Biol 1988; 106: ciated molecule) and heparin-type glycans in the development 1583–1591. and plasticity of neuron-target contacts. Prog Neurobiol 1997; 52: 78 Mann SS, Hammarback JA. Gene localization and developmental 127–144. expression of light chain 3: a common subunit of microtubule- 53 Joester A, Faissner A. The structure and function of tenascins in associated protein 1A (MAP1A) and MAP1B. J Neurosci Res the nervous system. Matrix Biol 2001; 20: 13–22. 1996; 43: 535–544. 54 Brose K, Tessier-Lavigne M. Slit proteins: key regulators of axon 79 Lemke G. Neuregulins in development. Mol Cell Neurosci 1996; guidance, axonal branching, and cell migration. Curr Opin 7: 247–262. Nuerobiol 2000; 20: 95–102. 80 Stefansson H, Sarginson J, Kong A, Yates P, Steinthorsdottir V, 55 Spitzer NC. Activity-dependent neuronal differentiation prior to Gudfinnson E et al. Association of neuregulin 1 with schizo- synapse formation: the functions of calcium transients. J Physicol phrenia confirmed in a Scottish population. Am J Hum Genet 2002; 96: 73–80. 2003; 72: 83–87. 56 Chang S, DeCamilli P. Glutamate regulates actin-based motility in 81 Stefansson H, Sigurdsson E, Steinthorsdottir V, Bjornsdottir S, axonal filopodia. Nat Neurosci 2001; 4: 783–793. Sigmundsson T, Ghosh S et al. Neuregulin 1 and susceptibility to 57 Da Silva JS, Dotti CG. Breaking the neuronal sphere: regulation of schizophrenia. Am J Hum Genet 2002; 71: 877–892. the actin cytoskeleton in neuritogenesis. Nat Rev Neurosci 2002; 82 Williams NM, Preece A, Spurlock G, Norton N, Williams HJ, 3: 694–704. Zammit S et al. Support for genetic variation in neuregulin 1 and 58 Dotti CG, Sullivan CA, Banker GA. The establishment of pola- susceptibility to schizophrenia. Mol Psychiatry 2003; 8: 485–487. rity by hippocampal neurons in culture. J Neurosci 1988; 8: 83 Yang JZ, Si TM, Ruan Y, Ling YS, Han YH, Wang XL et al. 1454–1468. Association study of neuregulin 1 gene with schizophrenia. Mol 59 Tang D, Goldberg DJ. Bundling of microtubules in the growth Psychiatry 2003; 8: 706–709. cone induced by laminin. Mol Cell Neurosci 2000; 15: 303–313. 84 Tang JX, Chen WY, He G, Zhou J, Gu NF, Feng GY et al. 60 Yu W, Ling C, Baas PW. Microtubule reconfiguration during Polymorphisms within 50 end of the neuregulin 1 gene are axogenesis. J Neurocytol 2001; 30: 861–875. genetically associated with schizophrenia in the Chinese popula- 61 Dehmelt L, Halpain S. Actin and microtubules in neurite tion. Mol Psychiatry 2004; 9: 11–12. initiation: are MAPs the missing link? J Neurobiol 2004; 58: 18–33. 85 Stefansson H, Steinthorsdottir V, Thorgeirsson TE, Gulcher JR, 62 Meyer G, Feldman EL. Signaling mechanisms that regulate actin- Stefansson K. Neuregulin 1 and schizophrenia. Ann Med 2004; based motility processes in the nervous system. J Neurochem 36: 62–71. 2002; 83: 490–503. 86 Iwata N, Suzuki T, Ikeda M, Kitajima T, Yamanouchi Y, Inada T 63 Luo L. Actin cytoskeleton regulation in neuronal morphogenesis et al. No association with the neuregulin 1 haplotype to Japanese and structural plasticity. Annu Cell Biol 2002; 18: 601–635. schizophrenia. Mol Psychiatry 2004; 9: 126–127. 64 Schaefer AW, Kabir N, Forscher P. Filopodia and actin arcs guide 87 Hashimoto R, Straub RE, Weickert CS, Hyde TM, Kleinman JE, the assembly and transport of two populations of microtubules Weinberger DR. Expression analysis of neuregulin-1 in the with unique dynamic parameters in neuronal growth cones. J Cell dorsolateral prefrontal cortex in schizophrenia. Mol Psychiatry Biol 2002; 158: 139–152. 2004; 9: 299–307. 65 Tilney LG, Bonder EM, DeRosier DJ. Actin filaments elongate 88 Hakak Y, Walker JR, Li C, Wong WH, Davis KL, Buxbaum JD et al. from their membrane-associated ends. JCellBiol1981; 90: 485–494. Genome-wide expression analysis reveals dysregulation of 66 Smith SJ. Neuronal cytomechanics: the actin-based motility of myelination-related genes in chronic schizophrenia. Proc Natl growth cones. Science 1988; 242: 708–715. Acad Sci USA 2001; 98: 4746–4751. 67 Lin CH, Espreafico EM, Mooseker MS, Forscher P. Myosin drives 89 Corfas G, Roy K, Buxbaum JD. Neuregulin 1-erbB signaling and retrograde F-actin flow in neuronal growth cones. Neuron 1996; the molecular/cellular basis of schizophrenia. Nat Neurosci 2004; 16: 782–796. 7: 575–580. 68 Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A. The 90 Villegas R, Villegas GM, Longart M, Hernandez M, Maqueira B, small GTP-binding protein rac regulates growth factor-induced Buonanno A et al. Neuregulin found in cultured-sciatic nerve membrane ruffling. Cell 1992; 70: 401–410. conditioned medium causes neuronal differentiation of PC12 69 Nobes CD, Hall A. Rho, rac and cdc42 GTPases regulate the cells. Brain Res 2000; 852: 305–318. assembly of multimolecular focal complexes associated with 91 Gamett DC, Greene T, Wagreich AR, Kim HH, Koland JG, Cerione stress fibers, lamellipodia and filopodia. Cell 1995; 81: 53–62. RA. Heregulin-stimulated signaling in rat pheochromocytoma 70 Caceres A, Mautino J, Kosik KS. Suppression of MAP2 in cells. J Biol Chem 1995; 270: 19022–19027. cultured cerebellar macroneurons inhibits minor neurite forma- 92 Vaskovsky A, Lupowitz Z, Erlich S, Pinkas-Kramarski R. Erb-4 tion. Neuron 1992; 9: 607–618. activation promotes neurite outgrowth in PC12 cells. J Neuro- 71 Caceres A, Kosik KS. Inhibition of neurite polarity by tau chem 2000; 74: 979–987. antisense oligonucleotides in primary cerebellar neurons. Nature 93 Shigeta M, Sanzen N, Ozawa M, Gu J, Hasegawa H, Sekiguchi K. 1994; 343: 451–463. CD151 regulates epithelial cell–cell adhesion through PKC-and 72 Harada A, Teng J, Takei Y, Oguchi K, Hirokawa N. MAP2 is Cdc42-dependent actin cytoskeletal reorganization. J Cell Biol required for dendrite elongation, PKA anchoring in dendrites, and 2003; 163: 165–176. proper PKA signal transduction. JCellBiol2002; 158: 541–549. 94 Pipel Y, Segal M. Activation of PKC induces rapid morphological 73 Liu CW, Lee G, Jay DG. Tau is required for neurite outgrowth and plasticity in dendrites of hippocampal neurons via Rac and growth cone motility of chick sensory neurons. Cell Motil Rho-dependent mechanisms. Eur J Neurosci 2004; 19: 3151–3164. Cytoskeleton 1999; 43: 232–242. 95 Rieff HI, Raetzman LT, Sapp DW, Yeh HH, Siegel RE, Corfas G.

74 Leclerc N, Baas CW, Garner CC, Kosik KS. Juvenile and mature Neuregulin induces GABAA receptor subunit expression and MAP2 isoforms induce distinct patterns of process outgrowth. neurite outgrowth in cerebellar granule cells. J Neurosci 1999; 19: Mol Biol Cell 1996; 7: 443–455. 10757–10766. 75 Barlow S, Gonzalez-Garay ML, West RR, Olmsted JB, Cabral F. 96 Bermingham-McDonogh O, McCabe KL, Reh TA. Effects of GGF/ Stable expression of heterologous microtubule-associated proteins neuregulins on neuronal survival and neurite outgrowth correlate

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 628 with erbB2/neu expression in developing retina. Development 119 Jiang H, Guo W, Liang X, Rao Y. Both the establishment and the 1996; 122: 1427–1438. maintenance of neuronal polarity require active mechanisms: 97 Pinkas-Kramarski R, Eilam R, Alroy I, Levkowitz G, Lonai P, critical roles of GSK-3b and its upstream regulators. Cell 2005; Yarden Y. Differential expression of NDF/neuregulin receptors 120: 123–135. ErbB-3 and ErbB-4 and involvement in inhibition of neuronal 120 Du K, Montminy M. CREB is a regulatory target for the protein differentiation. Oncogene 1997; 15: 2803–2815. kinase Akt/PKB. J Biol Chem 1998; 273: 32377–32379. 98 Ikeda M, Iwata N, Suzuki T, Kitajima T, Yamanouchi Y, Kinoshita 121 Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Y et al. Association of AKT1 with schizophrenia confirmed in a Inhibition of glycogen synthase kinase-3 by insulin mediated by Japanese population. Biol Psychiatry 2004; 56: 698–700. protein kinase B. Nature 1995; 378: 785–789. 99 Schwab SG, Hoefgen B, Hanses C, Borrmann Hassenbach M, 122 Kwon T, Kwon DY, Chun J, Kim JH, Kang SS. Akt protein kinase Albus M, Lerer B et al. Further evidence for association of inhibits Rac1-GTP-binding through phosphorylation at serine 71 variants in the AKT1 gene with schizophrenia in a sample of of Rac1. J Biol Chem 2000; 275: 423–428. European sib-pair families. Biol Psychiatry 2005; 58: 446–450. 123 Ridley AJ, Paterson HF, Johnston CL, Diekmann D, Hall A. The 100 Bajestan SN, Sabouri AH, Nakamura M, Takashima H, Keikhaee small GTP-binding protein rac regulates growth factor-induced MR, Behdani F et al. Association of AKT1 haplotype with the risk membrane ruffling. Cell 1992; 70: 401–410. of schizophrenia in Iranian population. Am J Med Genet 124 Fukumoto S, Hsieh CM, Maemura K, Layne MD, Yet SF, Lee KH (Neuropsychitr Genet) 2006; 141: 383–386. et al. Akt participation in the Wnt signaling pathway through 101 Ohtsuki T, Inada T, Arinami T. Failure to confirm association Dishevelled. J Biol Chem 2001; 18: 17479–17483. between AKT1 haplotype and schizophrenia in a Japanese case- 125 Zhou FQ, Zhou J, Dedhar S, Wu YH, Snider WD. NGF-induced control population. Mol Psychiatry 2004; 9: 981–983. axon growth is mediated by localized inactivation of GSK-3b and 102 Liu YL, Fann CS, Liu CM, Wu JY, Hung SI, Chan HY et al. functions of the microtubule plus end binding protein APC. Absence of significant associations between four AKT1 SNP Neuron 2004; 42: 897–912. markers and schizophrenia in the Taiwanese population. 126 Mitchison T, Krishcner M. Cytoskeletal dynamics and nerve Psychiatr Genet 2006; 16: 39–41. growth. Neuron 1988; 1: 761–772. 103 Masayuki I, Ohnishi T, Murayama M, Matsumoto I, Yamada K, 127 Bray D. Surface movements during the growth of single Iwayama Y et al. Failure to support a genetic contribution of explanted neurons. Proc Natl Acad Sci USA 1970; 65: 905–910. AKT1 polymorphisms and altered AKT signaling in schizophre- 128 Feldman EL, Axelrod D, Schwartz M, Heacock AM, Agranoff BW. nia. J Neurochem 2006; 99: 277–287. Studies on the localization of newly added membranes in 104 Brunet A, Datta SR, Greenberg ME. Transcription-dependent and growing neurites. J Neurobiol 1981; 12: 591–598. independent control of neuronal survival by the PI3-Akt 129 Pfenninger KH, Maylie´-Pfenninger MF. Lecitin labeling of signaling pathway. Curr Opin 2001; 11: 297–305. sprouting neurons. II. Relative movement and appearance of 105 Lawlor MA, Alessi DR. PKB/Akt: a key mediator of cell glycoconjugates during plasmalemmal expansion. J Cell Biol proliferation, survival and insulin responses? J Cell Sci 2001; 1981; 89: 547–559. 114: 2903–2910. 130 Craig AM, Wyborski RJ, Banker G. Preferential addition of newly 106 Brazil DP, Park J, Hemmings BA. PKB binding proteins. Getting in synthesized membrane protein at axonal growth cones. Nature on the Akt. Cell 2002; 111: 293–303. 1995; 375: 592–594. 107 Kalkman HO. The role of the phosphatidylinositide 3-kinase- 131 Xiao J, Zhou Q, Liu Y. Variant PC12 cell line that spontaneously protein kinase B pathway in schizophrenia. Pharmachol Ther differentiates and extends neuritic processes. J Neurosci Res 2006; 110: 117–134. 2002; 69: 104–109. 108 Fukunaga K, Kawano T. Akt is a molecular target for signal 132 Greene LA, Tischler AS. Establishment of a noradrenergic clonal transduction therapy in brain ischemic insult. J Pharmacol Sci line of rat adrenal pheochromocytoma cells which respond 2003; 92: 317–327. to nerve growth factor. Proc Natl Acad Sci USA 1976; 73: 109 Hallmayer J. Getting our AKT together in schizophrenia? Nat 2424–2428. Genet 2004; 36: 115–116. 133 Bang OS, Park EK, Yang S, Lee SY, Franke TF, Kang SS. 110 Kim AH, Yano H, Cho H, Meyer D, Monks B, Margolis B et al. Overexpression of Akt inhibits NGF-induced growth arrest Akt1 regulates a JNK scaffold during excitotoxic apoptosis. and neuronal differentiation of PC12 cells. J Cell Sci 2000; 114: Neuron 2002; 35: 697–709. 81–88. 111 Wassef A, Baker J, Kochan LD. GABA and schizophrenia: a 134 Piiper A, Dikic I, Lutz MP, Leser J, Kronenberger B, Elez R et al. review of basic science and clinical studies. J Clin Psychophar- Cyclic AMP induces transactivation of the receptors for epider- macol 2003; 23: 601–640. mal growth factor and nerve growth factor, thereby modulating 112 Namikawa K, Honma M, Abe K, Takeda M, Mansur K, Obata T activation of MAP kinase, Akt, and neurite outgrowth in PC12 et al. Akt/Protein kinase B prevents injury-induced motoneuron cells. J Biol Chem 2002; 277: 43623–43630. death and accelerates axonal regeneration. J Neurosci 2000; 20: 135 Jiang H, Guo W, Liang X, Rao Y. Both the establishment and 2875–2886. maintenance of neuronal polarity require active mechanisms: 113 Markus A, Zhong J, Snider WD. Raf and Akt mediate distinct critical roles of GSK-3b and its upstream regulators. Cell 2005; aspects of sensory axon growth. Neuron 2002; 35: 65–76. 120: 123–135. 114 Kim Y, Seger R, Babu S, Hwang SY, Yoo YS. A positive role of the 136 Craig AM, Banker G. Neuronal polarity. Annu Rev Neurosci 1994; PI3-K/Akt signaling pathway in PC12 cell differentiation. Mol 17: 267–310. Cells 2004; 18: 353–359. 137 Millar JK, Wilson-Annan JC, Anderson S, Christie S, Taylor MS, 115 Riese U, Ziegler E, Hamburger M. Militarinone A induces Semple CAM et al. Disruption of two novel genes by a differentiation in PC12 cells via MAP and Akt kinase signal translocation co-segregating with schizophrenia. Hum Mol Genet transduction pathways. FEBS Lett 2004; 577: 455–459. 2000; 9: 1415–1423. 116 Laurino L, Wang XX, de la Houssaye BA, Sosa L, Dupraz S, 138 Blackwood DHR, Fordyce A, Walker MT, St Clair DM, Proteous Caceres A et al. PI3K activation by IGF-1 is essential for the DJ, Muir WJ. Schizophrenia and affective disorders-cosegregation regulation of membrane expansion at the nerve growth cone. with a translocation at chromosome 1q42 that directly disrupts J Cell Sci 2005; 118: 3653–3663. brain-expressed genes: clinical and P300 findings in a family. Am 117 Ooms LM, Fedele CG, Astle MV, Ivetac I, Cheung V, Pearson RB J Hum Genet 2001; 69: 428–433. et al. The inositiol polyphosphate 5-phosphatase, PIPP, is a novel 139 Sachs NA, Sawa A, Holmes SE, Ross CA, DeLisi LE, Margolis RL. regulator of phosphoinositide 3-kinase-dependent neurite elon- A frameshift mutation in disrupted in schizophrenia 1 in an gation. Mol Biol Cell 2006; 17: 607–622. American family with schizophrenia and schizoaffective dis- 118 Higuchi M, Onishi K, Masuyama N, Gotoh Y. The phosphatidy- order. Mol Psychiatry 2005; 10: 758–764. linositol-3 kinase (PI3K)-Akt pathway suppresses neurite 140 Cannon TD, Hennah W, Theo GM, van Erp TGM, Thompson PM, branch formation in NGF-treated PC12 cells. Genes Cells 2003; Lonnqvist J et al. Association of DISC1/TRAX haplotypes with 8: 657–669. schizophrenia, reduced prefrontal gray matter and impaired

Molecular Psychiatry New genes associated with schizophrenia in neurite formation A Bellon 629 short- and long-term memory. Arch Gen Psychiatry 2005; 62: susceptibility gene for schizophrenia. Hum Mol Genet 2004; 13: 1205–1213. 1699–1708. 141 Hodgkinson CA, Goldman D, Jaeger J, Persaud S, Kane JM, Lipsky 157 Benson MA, Newey SE, Martin-Rendon E, Hawkes R, Blake DJ. PH et al. Disrupted in schizophrenia 1 (DISC1): association with Dysbindin, a novel coiled-coil-containing protein that interacts schizophrenia, schizoaffective disorder and bipolar disorder. with the dystrobervins in muscle and brain. J Biol Chem 2001; Am J Hum Genet 2004; 75: 862–872. 276: 24232–24241. 142 Porteous DJ, Thomson P, Brandon NJ, Millar JK. The genetics and 158 Risch N. Linkage strategies for genetically complex traits. I. biology of DISC1 and emerging role in psychosis and cognition. Multilocus models. Am J Hum Genet 1990; 46: 222–228. Biol Psychiatry 2006; 60: 123–131. 159 Li BS, Ma W, Jaffe H, Zheng Y, Takahashi S, Zhang L et al. Cyclin- 143 Ishizuka K, Paek M, Kamiya A, Sawa A. A review of disrutped-in- dependent kinase-5 is involved in neuregulin-dependent activa- schizophrenia-1 (DISC1): neurodevelopment, cognition and tion of phosphatidylinositol 3-kinase and Akt activity mediating mental conditions. Biol Psychiatry 2006; 59: 1189–1197. neuronal survival. J Biol Chem 2003; 278: 35702–35709. 144 Miyoshi K, Honda A, Baba K, Taniguchi M, Oono K, Fujita T et al. 160 Gambarotta G, Garzotto D, Destro E, Mautino B, Giampietro C, Disrupted-in-schizophrenia 1, a candidate gene for schizophre- Cutrupi S et al. ErbB4 expression in neural progenitor cells nia, participates in neurite outgrowth. Mol Psychiatry 2003; 8: (ST14A) is necessary to mediate neuregulin-1beta1-induced 685–694. migration. J Biol Chem 2004; 279: 48808–48816. 145 Morris JA, Kandpal G, Ma L, Austin CP. DISC1 (disrupted-in- 161 Hashimoto R, Numakawa T, Ohnishi T, Kumamaru E, Yagasaki Y, schizophrenia 1) is a -associated protein that interacts Ishimoto T et al. Impact of the DISC1 Ser704Cys polymorphism with MAP1A, MIPT3, ATF475 and NUDEL: regulation and loss of on risk for major depression, brain morphology and ERK interaction with mutation. Hum Mol Genet 2003; 12: 1591–1608. signaling. Hum Mol Genet 2006; 15: 3024–3033. 146 Bloom L, Horvitz HR. The Caenorhabditis elegans gene unc-76 162 McGrath J, Murray R. Risk factors for schizophrenia: form and its human homologous define a new gene family involved in conception to birth. In: Hirsh SR, Weinberger DR (eds). Schizo- axonal outgrowth and fasciculation. Proc Natl Acad Sci USA phrenia. Blackwell: Malden, MA, 1995, pp 232–250. 1997; 94: 3414–3419. 163 Bagalkote H, Pang D, Jones PB. Maternal influenza and schizo- 147 Kuroda S, Nakagawa N, Tokunaga C, Tatematsu K, Tanizawa K. phrenia in offspring. Int J Ment Health 2001; 39: 3–21. Mammalian homologue of the Caenorhabditis elegans UNC-76 164 Nuechterlein KH, Dawson ME. A heuristic vulnerability/ protein involved in axonal outgrowth is a protein kinase stress model of schizophrenic episodes. Schizophr Bull 1984; C x-interacting protein. J Cell Biol 1999; 144: 403–411. 10: 300–312. 148 Straub RE, Jiang Y, MacLean DJ, Ma Y, Webb BT, Myakishev MV 165 Hillier SL, Witkin SS, Krohn MA, Watts DH, Kiviat NB, et al. Genetic variation in the 6p22.3 gene DTNBP1, the human Eschenbach DA. The relationship of amniotic fluid cytokines ortholog of the mouse dysbindin gene, is associated with and preterm delivery, amniotic fluid infection, histologic schizophrenia. Am J Hum Genet 2002; 71: 337–348. chorioamnionitis, and chorioamnion infection. Am J Obstet 149 Datta SR, Rizig MA, McQuillin A, Kalsi G, Lawrence J, Quested D Gynecol 1993; 81: 941–948. et al. Tests of linkage disequilibrium between schizophrenia and 166 Fortunato SJ, Menon RP, Swan KF, Menon R. Inflammatory genetic markers at the G72 and dysbindin loci. In: Barden N, Cote cytokines (interleukins 1,6,8 and tumor necrosis factor-a) release M, DeLisi L, Gill M, Kelosoe J, Schalling M (eds). XIth World form cultured fetal membranes in response to endotoxic Congress of Psychiatric Genetics. Quebec, 2003, pp 131–172. lipopolysaccharide mirrors amniotic fluid. Am J Obstet Gynecol 150 Van den Oord EJ, Sullivan PF, Jiang Y, Walsh D, O’Neill FA, 1996; 174: 1855–1862. Kendler KS et al. Identification of a high-risk haplotype for the 167 Yoon BH, Romero R, Moon J, Chaiworapongsa T, Espinoza J, Kim dystroberdin binding pretion 1 (DTNBP1) gene in the Irish study YM. Differences in the fetal interleukin-6 response to microbial of high-density schizophrenia families. Mol Psychiatry 2003; 8: invasion of the amniotic cavity between term and preterm 499–510. gestation. J Matern Fetal Neonatal Med 2003; 13: 32–38. 151 Bakker SC, Hoogendoom MLC, Sinke RJ, Verzibergen KF, Kusters 168 Gilmore JH, Jarskog LF, Vadlamudi S, Lauder JM. Prenatal KA, Otten HG et al. Association study of the neuregulin, infection and risk for schizophrenia: IL-6, IL-Ib and TNFa inhibit dysbindin and G72 genes in a large sample of Dutch schizo- cortical neuron dendrite development. Neuropsychopharma- phrenic patients. In: Barden N, Cote M, DeLisi L, Gill M, Kelosoe cology 2004; 29: 1221–1229. J, Schalling M (eds). XIth World Congress of Psychiatric Genetics. 169 Stevens KE, Johnson RG, Rose GM. Rats reared in social isolation Quebec, 2003, p 19. show schizophrenia-like changes in auditory gating. Pharmacol 152 Van Den Bogaert A, Schimacher J, Schulze TG, Otte AC, Ohlraun Biochem Behav 1997; 58: 1031–1036. S, Kovalenko S et al. The DTNBP1 (dysbindin) gene contributes 170 Silva-Gomez AB, Rojas D, Juarez I, Flores G. Decreased dendritic to schizophrenia, depending on family history of the disease. Am spine density on prefrontal cortical and hippocampal pyramidal J Hum Genet 2003; 73: 1438–1443. neurons in postweaning social isolation rats. Brain Res 2003; 983: 153 Shi YY, Zhao XZ, Tang JX, Gu NF, Feng GY, Zhu SM et al. Non- 128–136. family based association study of DBTNBP1 in 6p22.3 and 171 Hoffman RE, McGlashan TH. Synaptic elimination, neurodeve- schizophrenia in geographically and genetically structured lopment, and the mechanism of hallucinated ‘voices’ in schizo- Chinese Han population. In: Barden N, Cote M, DeLisi L, Gill phrenia. Am J Psychiatry 1997; 154: 1683–1689. M, Kelosoe J, Schalling M (eds). XIth World Congress of 172 Waddington JL. Schizophrenia: developmental neuroscience and Psychiatric Genetics. Quebec, 2003, p 102. pathobiology. Lancet 1993; 341: 531–536. 154 Tang JX, Zhou J, Fan JB, Li XW, Shi YY, Gu NF et al. Family-based 173 McGlashan TH, Hofman RE. Schizophrenia as a disorder of association study of DTNBP1 in 6p22.3 and schizophrenia. Mol developmentally reduced synaptic connectivity. Arch Gen Psychiatry 2003; 8: 717–718. Psychiatry 2000; 57: 637–648. 155 Yamada K, Detera-Wadleigh SD, Iwayama-Shigeno Y, Toyota T, 174 Thompson PM, Vidal C, Giedd JN, Gochman P, Blumenthal J, Corona W, Hattori E et al. Suggestive evidence of association Nicolson AW et al. Mapping adolescent brain change reveals between the DTNBP1 gene and schizophrenia in the Japanese dynamic wave of accelerated gray matter loss in very early- samples. In: Barden N, Cote M, DeLisi L, Gill M, Kelosoe J, onset schizophrenia. Proc Natl Acad Sci USA 2001; 98: Schalling M (eds). XIth World Congress of Psychiatric Genetics. 11650–11655. Quebec, 2003, p 102. 175 Lewis DA, Gonzalez-Burgos G. Intrinsic excitatory connections in 156 Numakawa T, Yagasaki Y, Ishimoto T, Okada T, Suzuki T, Iwata N the prefrontal cortex and the pathophysiology of schizophrenia. et al. Evidence of novel neuronal functions of dysbindin, a Brain Res Bull 2000; 52: 309–317.

Molecular Psychiatry