J Mol Neurosci (2014) 52:353–365 DOI 10.1007/s12031-013-0156-8

Antibodies Directed to Neisseria gonorrhoeae Impair Nerve Growth Factor-Dependent Neurite Outgrowth in Rat PC12 Cells B. Reuss

Received: 25 September 2013 / Accepted: 17 October 2013 / Published online: 8 November 2013 # Springer Science+Business Media New York 2013

Abstract In children born from mothers with prenatal infections with the Gram-negative bacterium Neisseria gonorrhoeae , schizophrenia risk is increased in later life. Since cortical neuropil formation is frequently impaired during this disease, actions of a rabbit polyclonal antiserum directed to N. gonorrhoeae on neurite outgrowth in nerve growth factorstimulated PC12 cells were investigated here. It turned out that 10 μg/ml of the antiserum leads indeed to a significant reduction in neurite outgrowth, whereas an antiserum directed to Neisseria meningitidis had no such effect. Furthermore, reduction in neurite outgrowth could be reversed by the neuroleptic drugs haloperidol, clozapine, risperidone, and olanzapine. On the molecular level, the observed effects seem to include the known neuritogenic transcription factors FoxO3a and Stat3, since reduced neurite outgrowth caused by the antiserum was accompanied by a reduced phosphorylation of both factors. In contrast, restitution of neurite outgrowth by neuroleptic drugs revealed no correlation to the phosphorylation state of these factors. The present report gives a first hint that bacterial infections could indeed lead to impaired neuropil formation in vitro; however, the in vivo relevance of this finding for schizophrenia pathogenesis remains to be clarified in the future. Keywords Neurite outgrowth . PC12 cells . Nerve growth factor . Neuroleptic . FoxO3a . Stat3

Introduction Schizophrenia is a severe and debilitating neuropsychiatric disorder, affecting around 1 % of the general population once B. Reuss (*) Institute for Neuroanatomy, Universitary Medicine Göttingen, Kreuzbergring 36, 37075 Göttingen, Germany e-mail: [email protected]

during lifetime. Characteristic symptoms of this disease consist of delusion, hallucinations, and bizarre thinking, resulting in a severely impaired quality of life (Tandon et al. 2009). These behavioral changes have been shown to go along with a complex pattern of neuropathological changes of the brain (Harrison 1999); the underlying cellular and molecular mechanisms of which remain still largely enigmatic. Up to now, the only effective treatment of psychosis is the application of so-called neuroleptic drugs like haloperidol, risperidone, clozapine, and olanzapine. They all interact with aminergic neurotransmitter receptors, with the type 2 dopamine receptor, playing a prominent role. This forms the basis of the so-called dopamine hypothesis for the pathogenesis of schizophrenia (Meltzer and Stahl 1976). However, which causes lead to perturbed dopaminergic transmission in the schizophrenic brain is still not yet known. A possible explanation comes from the so-called neurodevelopmental hypothesis for the pathogenesis of schizophrenia (Murray et al. 1992), which suggests prenatal and/or obstetric complications leading to disease-relevant changes of neuronal brain architecture and function. Thus, pre- and perinatal maternal and/or fetal viral infections by influenza or herpes simplex as well as by eukaryotic parasites like Toxoplasma gondii (Khandaker et al. 2013) seem to result in perturbed brain development in children destined in later life to become schizophrenic patients. Much less is known on the role of bacterial infections as putative pathogenetic agents for schizophrenia. A type of bacterial infection that seems indeed to be linked to an increased risk for schizophrenic psychosis is that with the Gramnegative bacterium Neisseria gonorrhoeae (NG). As to my knowledge, at least two clinical studies demonstrate that prenatal gonococcal infections during the first trimester of pregnancy are associated with an increased risk for the offspring to suffer from schizophrenia in later life (Babulas et al. 2006; Sørensen et al. 2009). NG is a major cause for clinical and

354

subclinical reproductive tract infections in women and men (Edwards and Butler 2011), and antibodies directed to NGspecific epitopes are a common feature of the blood serum of NG-infected women (Hoffman et al. 1979). Therefore, crossreactivity of NG-specific antibodies with specific brain structures could be responsible for perturbed brain development following exposure during early pregnancy. A major morphological correlate of schizophrenia is the occurrence of enlarged ventricles and decreased cortical volume in the brains of these patients (Gur et al. 1994; Van Horn and McManus 1992; Woods et al. 2005; Wright et al. 2000). At the microscopical level, this correlates with an apparent increase of neuronal density in the prefrontal cortex and other brain regions, probably as a result of decreased neurite formation (Selemon et al. 1995; Black et al. 2004). These earlier findings raised for us the question: Whether antibodies directed to NG would be able to impair neurite formation during brain development and which could be the underlying cellular and molecular factors? In order to answer these questions, an in vitro approach was established by applying bacteria-specific antibodies (directed to NG) to PC12 cells, a widely used cell culture model for neuronal differentiation; and neurite outgrowth (Greene 1978; Levi et al. 1988). Differentiation of PC12 cells can be induced by the application of several growth factors such as FGF-2 (Jeon et al. 2010a), IGF-1 (Pugazhenthi et al. 1999), EGF (Kasai et al. 2005), and most prominent NGF (Greene 1978; Kaplan and Stephens 1994). With regard to the actions of NGF in PC12 cells and other cells, a set of transcription factors has already been identified as direct or indirect intracellular regulators for neurite outgrowth, and of which, FoxO3a and Stat3 are known to play an important role. Thus, both phosphorylations of FoxO3a at serine residue 253 (Wen et al. 2011) and of Stat3 at serine residue 727 (Zhou and Too 2011) are known to be induced during NGF-dependent neurite outgrowth in PC12 cells. The aim of the present study was to finally clarify whether antibodies directed to NG are able to impair neurite outgrowth in NGF-stimulated PC12 cells and whether such an effect could be reversed by the application of neuroleptic drugs. In addition, a putative role of the neurite outgrowth-related transcription factors FoxO3a and Stat3 for these effects was investigated.

J Mol Neurosci (2014) 52:353–365

transducer and activator of transcription 3 pSer727 (α-Stat-3pSer727; Antikoerper-online, Aachen, Germany; cat. nr. ABIN482587); rabbit anti-Forkhead box protein O3a (αFoxO3a; Antikoerper-online, Aachen, Germany; cat. nr. ABIN349449); rabbit anti-Forkhead box protein O3a pSer253 (α-FoxO3a-pSer253; Antikoerper-online, Aachen, Germany; cat. nr. ABIN349221); and mouse anti-β-actin (α-β-actin; Sigma-Aldrich, Steinheim, Germany; cat. nr. A5441). Cell Culture Rat PC12 pheochromocytoma cells (DSMZ, Braunschweig, Germany) were maintained in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10 % fetal calf serum (FCS), 5 % horse serum (HS), and penicillin/streptomycin (PS; all from Gibco/Invitrogen, Darmstadt, Germany). For maintenance, medium was exchanged twice per week. For differentiation, cells were quantified with a hematocytometer (Labor Optik Ltd., Lancing, UK), and 5,000 cells/well were seeded on a six-well cell culture plate (Sarstedt, Nümbrecht, Germany). Cells were first preincubated for 1 day with DMEM supplemented with 0.5 % FCS and 0.25 % HS and PS, and then, the medium was supplemented with 10 ng/ml NGF (Sigma-Aldrich, Steinheim, Germany), to induce neurite outgrowth. To analyze the effects of α-NG on neurite outgrowth, PC12 cells were treated in parallel with 10 μg/ml of α-NG, and Na azide of which has been removed by microdialysis using Amicon Ultra filter units (Merck-Millipore, Schwalbach, Germany). Cells were also treated with antipsychotic drugs haloperidol (HAL; 0.1 μmol/l), risperidone (RIS; 1 μmol/l), clozapine (CLZ; 0.1 μmol/l; all from SigmaAldrich, Steinheim, Germany), and olanzapine (OLA; 10 μmol/l; Lilly Pharma, Giessen, Germany). All substances were applied simultaneously with NGF, starting right from the beginning of cellular differentiation as monitored by neurite outgrowth. Cells were imaged with a DS digital camera system mounted on the Eclipse Inverted Microscope (both Nikon; Düsseldorf, Germany). Neurite length as well as neurite numbers per cell were photographically evaluated for every treatment and time point at 10 randomly chosen fields in eight independent experiments. General significance of the observed differences was analyzed first by one-way ANOVA which was then specified at more detail by Student’s two-tailed t test.

Materials and Methods Western Blot Antibodies The following antibodies were used in this study: rabbit antiN. gonorrhoeae (α-NG; Antikoerper-online, Aachen, Germany; cat. nr. ABIN285584); rabbit anti-signal transducer and activator of transcription 3 (α-Stat-3; Antikoerper-online, Aachen, Germany; cat. nr. ABIN1385070); rabbit anti-signal

Cells were harvested in electrophoresis sample buffer (0.5 mmol/l Tris–HCl, pH 6.8, 2 % sodium dodecyl sulfate (SDS) (w /w), 10 % glycerol (v/v)), and the protein concentration of the obtained total cell extract was then determined using a densitometric method (Henkel and Bieger 1994). Five micrograms of total cellular protein was then

J Mol Neurosci (2014) 52:353–365

electrophoretically size separated on an 8.5 % SDS polyacrylamide gel (Laemmli 1970). After tank-blot Western transfer onto polyvinyl difluoride membranes (PVDF; Roth, Karlsruhe, Germany), blocking occurred for 1 h at 4 °C in 3 % (w/v) nonfat dry milk in Tris-buffered saline with 0.01 % Tween 20 (TBST). Blots were then incubated overnight at 4 °C with primary antibodies (as listed above), diluted at a ratio of 1:2,000 in TBST with 0.1 % nonfat dry milk. After washing, depending on the primary antibodies used, either mouse- or rabbit-specific peroxidase-coupled secondary antibodies (Sigma-Aldrich, Steinheim, Germany) were applied for 90 min at room temperature, diluted at a ratio of 1:10, 000 in TBST, with 0.1 % nonfat dry milk. Subsequent visualization occurred by exposing SuperRX medical X-ray films (Fuji, Düsseldorf, Germany) during the application of a peroxidase chemiluminescence substrate (0.1 mol/l Tris–HCl (pH 8.6), 0.25 mg/ml luminol, 0.2 mg/ml p-hydroxycoumaric acid, and 0.1 % H2O2). For relative quantification, blots were stripped with 1 mol/l NaOH for 15 min and were reincubated with an antibody directed to the housekeeping protein β-actin. Blots were densitometrically evaluated using the noncommercial image analysis program IMAL (http://www.randombio. com/imal.html). Significant levels of the obtained results were analyzed with Student’s two-tailed t test. Immunocytochemistry Cells were washed with phosphate-buffered saline (PBS) and fixed for 10 min with 4 % paraformaldehyde (PFA). After washing with PBS, cells were permeabilized for 10 min with a mixture of acetone/methanol (1:1) at −20 °C. Following three washes with PBS, cells were blocked for 1 h with goat serum (1:50) in PBS (PBS-GS). Primary antibodies diluted at a ratio of 1:50 in PBS-GS were applied overnight at 4 °C, followed by three washes with PBS and a 90-min incubation period with biotin-coupled secondary antibodies, diluted at a ratio of 1:400 in PBS-GS at 37 °C. In order to fluorescently label the antibody-tagged proteins, after three washes with PBS, FITClabeled streptavidin, diluted at a ratio of 1:400 in PBS, was applied for 90 min at room temperature. After washing again with PBS, three drops of a commercially available fluorescent mounting medium (DAKO, Glostrup, Denmark) were applied, and cells were covered with standard cover slips. Imaging was performed using the Axiocam digital camera system, mounted on the Axiophot Microscope equipped with epifluorescence (Zeiss, Jena, Germany).

Results A decrease in neuropil by yet unknown mechanisms is a common feature in the frontal cortex of the schizophrenic brain (Selemon et al. 1995; Black et al. 2004). We tested here

355

whether a polyclonal antiserum from rabbit, directed to the Gram-negative bacterium N. gonorrhoeae (α-NG) was able to reduce the neurite formation in NGF-stimulated PC12 cells, an in vitro model for neuronal differentiation, and neurite outgrowth (Greene 1978). As shown in Fig. 1, this was indeed the case, since in NGF-stimulated PC12 cells, both neurite length as well as the number of neurites formed by each cell were significantly reduced. Before treatment with either NGF alone (Fig. 1a) or in combination with α-NG (Fig. 1b), cells revealed no neurite-like processes at all. After a 4-day treatment with 10 ng/ml NGF (Fig. 1c), phase-contrast image reveals several processes with a length from the cell body (arrowheads) to the tip (double arrows) of around 40 μm and more. In contrast, after a 4-day treatment with 10 ng/ml NGF together with 10 μg/ml α-NG (Fig. 1d), a distinct shortening of neurite-like processes in PC12 cells can be observed (arrowheads, double arrows). As shown in Fig. 1a–d, the diagram of the statistical evaluation of neurite length in a series of experiments demonstrates that the α-NG treatment leads to a significant decrease in neurite length in NGF-stimulated PC12 cells. Likewise, as shown in Fig. 1f, the diagram of the statistical evaluation of the neurite numbers per cell in a series of experiments as shown in Fig. 1a–d demonstrates a significant decrease in numbers of neurites per cell by α-NG in NGF-stimulated PC12 cells. An important question with regard to the effects of α-NG treatment on NGF-stimulated PC12 cells as shown in Fig. 1 was that whether this effect is really antibody specific and not only a result of an unspecific interaction. To answer this, the effects of α-NG were compared to the effects of an antiserum directed to N. meningitidis (α-NM), another member of the Neisseriaceae family of pathogenic bacteria and a leading cause for meningitis and sepsis in immune-competent patients (Rosenstein et al. 2001; Stephens et al. 2007). As revealed in Fig. 2, only treatment with α-NG (Fig. 2a, c, e) and not with α-NM (Fig. 2b, d, f) leads to a significant concentrationdependent decrease in the length of neurite-like processes in NGF-stimulated PC12 cells. As revealed by phase-contrast image, control PC12 cells treated with 10 ng/ml NGF for 4 days, but in the absence of either α-NG (Fig. 2a) or α-NM (Fig. 2b), revealed a normal formation of neurite-like processes. In contrast, a 4-day treatment with 10 ng/ml NGF together with either 1 μg/ml (Fig. 2c) or 10 μg/ml α-NG (Fig. 2e) leads to a distinct decrease in the length of neurite-like processes as compared to control cells, which turned out to be statistically significant (Fig. 2g). In contrast, no such effect could be observed after a 4-day treatment of PC12 cells with 10 ng/ ml NGF together with α-NM at concentrations of either 1 μg/ ml (Fig. 2d, h) or 10 μg/ml (Fig. 2f, h). A further question with regard to a putative relevance for schizophrenia pathogenesis of the observed effects of α-NG treatment on the formation of neurite-like processes in NGFstimulated PC12 cells is whether these effects can be reversed

356

J Mol Neurosci (2014) 52:353–365

Fig. 1 Effect of α-NG treatment on neurite length and numbers in NGF-stimulated PC12 cells. a Phase-contrast image of PC12 cells before treatment with NGF. b Phase-contrast image of PC12 cells before treatment with NGF together with α-NG. c Phasecontrast image of PC12 cells after 4 days of treatment with 10 ng/ml NGF. d Phase-contrast image of PC12 cells of PC12 cells after 4 days of treatment with 10 ng/ml NGF together with 10 μg/ml αNG. e Diagram of the statistical evaluation of neurite length in a series of experiments as shown in a–d, demonstrating a significant decrease in neurite length by αNG treatment in NGF-stimulated PC12 cells. f Diagram of the statistical evaluation of neurite numbers per cell in a series of experiments as shown in a–d, demonstrating a significant decrease in numbers of neurites per cell by α-NG in NGFstimulated PC12 cells. The large arrowheads shown in the figures depict the origin of a representative neurite at the cell body, whereas the double arrows label the corresponding neurite tip

by the parallel application of neuroleptic drugs. Surprisingly, as revealed in Fig. 3, this was indeed the case, since the α-NGdependent reduction in the outgrowth of neurite-like processes in NGF-stimulated PC12 cells could be almost completely compensated by a parallel application of either of the antipsychotic drugs HAL (0.1 μmol/l), RIS (1 μmol/l), CLZ (0.1 μmol/l), or OLA (10 μmol/l). As revealed by phasecontrast image, PC12 cells treated for 4 days with 10 ng/ml NGF revealed a normal formation of neurite-like processes (Fig. 3a) that was distinctly reduced upon parallel application of 10 μg/ml α-NG (Fig. 3b). If in addition to NGF and α-NG, 0.1 μmol/l haloperidol was added to the culture medium, length of neurite-like processes again became almost normal (Fig. 3c). A similar effect could be observed also for 1 μmol/l risperidone (Fig. 3d), 0.1 μmol/l clozapine (Fig. 3e), and 10 μmol/l olanzapine (Fig. 3f), and all the three drugs were also able to overcome the impairing effect of α-NG treatment on neurite outgrowth in PC12 cells. As revealed on the

diagram of the statistical evaluation of a series of such experiments (Fig. 3g), these effects were highly significant. Another important question with regard to the observed effects of α-NG treatment on growth of neurite-like processes in NGF-treated PC12 cells is which intracellular regulatory factors are potentially involved in this process. To clarify this, expression and phosphorylation of the transcription factors FoxO3a and Stat3, which have been previously demonstrated to be involved in the regulation of neurite outgrowth of PC12 cells (Wen et al. 2011; Zhou and Too 2011), were investigated under control conditions either without NGF or in the presence of 10 ng/ml NGF or with an additional treatment with 10 μg/ml α-NG (Fig. 4). As revealed by Western blot analysis (Fig. 4a), expression of FoxO3a was distinctly induced by NGF treatment, but levels of FoxO3a immunoreactivity were not further changed by additional treatment with α-NG. In contrast to this, an antibody directed to the phosphorylated form of FoxO3a revealed that the phosphorylation of FoxO3a

J Mol Neurosci (2014) 52:353–365

357

Fig. 2 Effect of treatment with different concentrations of either α-NG or α-NM on neurite outgrowth in NGF-stimulated PC12 cells. a Phase-contrast image of PC12 cells before treatment with α-NG. b Phasecontrast image of PC12 cells before treatment with α-NM. c Phase-contrast image of PC12 cells after 4 days of treatment with 10 ng/ml NGF together with 1 μg/ml α-NG. d Phase-contrast image of PC12 cells after 4 days of treatment with 10 ng/ml NGF together with 1 μg/ml α-NM. e Phase-contrast image of PC12 cells after 4 days of treatment with 10 ng/ml NGF together with 10 μg/ml α-NG. f Phase-contrast image of PC12 cells after 4 days of treatment with 10 ng/ml NGF together with 1 μg/ml α-NM. g Diagram of the statistical evaluation of a series of experiments as shown in a–f, demonstrating a significant concentration-dependent decrease in neurite length by α-NG treatment in NGF-stimulated PC12 cells, whereas treatment with α-NM elicited no such effect. The large arrowheads shown in the figures depict the origin of a representative neurite at the cell body, whereas the double arrows label the corresponding neurite tip

was also distinctly increased in cultures treated with NGF, as compared to untreated cultures and this phosphorylation was distinctly diminished, when the cells were additionally treated with α-NG (Fig. 4b). Similar results were obtained for Stat3, the expression of which was also distinctly induced by NGF treatment but not further changed by additional application of α-NG (Fig. 4c). In contrast to this, Stat3 protein phosphorylation

(Fig. 4d) was increased in cultures treated with NGF and distinctly diminished when the cells were additionally treated with α-NG. As a loading control, blots as shown in Fig. 4(a–d) were stripped and reincubated with an antibody directed to the housekeeping protein β-actin (Fig. 4e). Densitometric evaluation and statistical analysis of the changes in FoxO3a phosphoprotein immunoreactivity revealed the observed changes in FoxO3a

358

J Mol Neurosci (2014) 52:353–365

Fig. 3 Effect of different neuroleptic drugs on the decrease in neurite length in NGFstimulated PC12 cells caused by α-NG. a Phase-contrast image of PC12 cells treated for 4 days with 10 ng/ml NGF. b Phase-contrast image of PC12 cells treated for 4 days with 10 ng/ml NGF together with 10 μg/ml α-NG. c Phase-contrast image of PC12 cells treated for 4 days with 10 ng/ml NGF, 10 μg/ml α-NG, and 0.1 μmol/l haloperidol. d Phase-contrast image of PC12 cells treated for 4 days with 10 ng/ml NGF, 10 μg/ml α-NG, and 1 μmol/l risperidone. e Phase-contrast image of PC12 cells treated for 4 days with 10 ng/ml NGF, 10 μg/ml α-NG, and 0.1 μmol/l clozapine. f Phase-contrast image of PC12 cells treated for 4 days with 10 ng/ml NGF, 10 μg/ml α-NG, and 10 μmol/l olanzapine. g Diagram of the statistical evaluation of a series of experiments as shown in a–f, demonstrating a significant recovery of the decrease in neurite length caused by α-NG treatment when the cells were treated in parallel by neuroleptic drugs. The large arrowheads shown in the figures depict the origin of a representative neurite at the cell body, whereas the double arrows label the corresponding neurite tip

protein phosphorylation to be highly significant (Fig. 4f). Likewise, also the changes in phosphorylation of the transcription factor Stat3 in PC12 cells, treated with α-NG and NGF, were highly significant (Fig. 4g). As revealed in Fig. 5, the effect of α-NG on phosphorylation of FoxO3a and Stat3 could also be observed in individual cells as revealed by immunofluorescent staining. Thus, in control

cells without NGF treatment, only weak immunostaining for the phosphorylated forms of both FoxO3a (Fig. 5a) and Stat3 (Fig. 5b) could be observed. Staining intensity was then distinctly increased, when the cells were treated for 4 days with NGF alone (Fig. 5c, d), whereas in cells treated for 4 days with both NGF and α-NG, staining intensity for both antibodies was again distinctly reduced.

J Mol Neurosci (2014) 52:353–365

359

Fig. 4 Effect of a 4-day α-NG treatment of NGF-stimulated PC12 cells, on phosphorylation of the neurite outgrowth-related transcription factors FoxO3a and Stat3, as revealed on the protein level by Western blot analysis. a Western blot detection of the transcription factor FoxO3a in total cellular protein of control (CON)-treated, NGF-treated, or α-NG plus NGF-treated cultures of PC12 cells. b Western blot detection of the phosphorylated form of the transcription factor FoxO3a in total cellular protein of CON-treated, NGF-treated, or α-NG plus NGF-treated cultures of PC12 cells. c Western blot detection of the transcription factor Stat3 in total cellular protein of CON-treated, NGF-treated, or α-NG plus NGFtreated cultures of PC12 cells. d Western blot detection of the phosphorylated form of the transcription factor Stat3 in total cellular protein of CON-treated, NGF-treated, or α-NG plus NGF-treated cultures of PC12

cells. e Western blot detection of the cytoskeletal housekeeping protein β-actin in total cellular protein of CON-treated, NGF-treated, or α-NG plus NGF-treated cultures of PC12 cells as a loading control. f Diagram of the statistical evaluation of the densitometric evaluation of a series of experiments as shown in b, demonstrating a significant decrease in phosphorylation of the transcription factor FoxO3a in PC12 cells treated with α-NG together with NGF, as compared to cultures treated with NGF alone. g Diagram of the statistical evaluation of the densitometric evaluation of a series of experiments as shown in d, demonstrating a significant decrease in phosphorylation of the transcription factor Stat3 in PC12 cells treated with α-NG together with NGF, as compared to cultures treated with NGF alone

Findings of the previous section, together with the already described rescue effect of haloperidol, risperidone, clozapine, and olanzapine on α-NG-dependent defects in neurite outgrowth raised the question: Whether also the changes in

protein phosphorylation of FoxO3a and Stat3 caused by αNG could be reversed by the application of neuroleptic drugs? To clarify this, α-NG was applied to PC12 cells during NGFdependent neurite outgrowth either alone or in the presence of

360

J Mol Neurosci (2014) 52:353–365

Fig. 5 Effect of a 4-day α-NG treatment of NGF-stimulated PC12 cells, on phosphorylation of the neurite outgrowth-related transcription factors FoxO3a and Stat3, as revealed on the cellular level by immunofluorescent detection. a Fluorescence image of the immunoreactivity for the phosphorylated form of the transcription factor FoxO3a in nondifferentiated PC12 cells. b Fluorescence image of the immunoreactivity for the phosphorylated form of the transcription factor Stat3 in nondifferentiated PC12 cells. c Fluorescence image of the immunoreactivity for the phosphorylated form of the transcription factor FoxO3a in PC12 cells treated for 4 days with NGF alone. d Fluorescence image of the immunoreactivity for the phosphorylated form of the transcription factor Stat3 in PC12 cells treated for 4 days with NGF alone. e Fluorescence image of the immunoreactivity for the phosphorylated form of the transcription factor FoxO3a in PC12 cells treated for 4 days with NGF together with α-NG. f Fluorescence image of the immunoreactivity for the phosphorylated form of the transcription factor Stat3 in PC12 cells treated for 4 days with NGF in combination with α-NG

the neuroleptic drugs haloperidol, risperidone, clozapine, and olanzapine, and phosphorylation of both FoxO3a and Stat3 was then analyzed by Western blot analysis. As revealed in Fig. 6, in NGF-treated PC12 cells (Fig. 6a), phosphorylation of FoxO3a was again reduced by application of α-NG; however, application of neither haloperidol nor clozapine was able to restore phosphorylation rates of this protein. Similarly, for Stat3 (Fig. 6b), no recovery of phosphorylation upon application of either haloperidol or clozapine occurred. Detection of the housekeeping protein β-actin confirmed equal protein loading on each lane (Fig. 6c). As the diagrams for the statistical evaluation of a series of experiments as shown in Fig. 6(a, b) show that there is no recovery of α-NG-dependent effects on protein phosphorylation of FoxO3a (Fig. 6d) and Stat3 (Fig. 6e), this could be obtained by parallel application of the neuroleptic drugs haloperidol and clozapine. Likewise, risperidone and olanzapine failed to restore phosphorylation of both FoxO3a (Fig. 6f) and Stat3 (Fig. 6g), as revealed by

Western blot analysis. Again, equal protein loading was confirmed by the detection of the housekeeping protein β-actin (Fig. 6h), and the diagrams of the statistical evaluation of a series of experiments as shown in Fig. 6(f, g) reveal that there is no recovery of α-NG-dependent effects on protein phosphorylation of FoxO3a (Fig. 6i) and Stat3 (Fig. 6k) by parallel application of the neuroleptic drugs risperidone and olanzapine.

Discussion Brain-reactive autoantibodies are a common feature of several neurological disorders, such as Sydenham’s chorea, pediatric autoimmune neuropsychiatric disorder associated with streptococcal infection (PANDAS), and NMDA receptor encephalitis (Dale and Brilot 2012), which, in the case of PANDAS, are linked to a previous history of streptococcal infections

J Mol Neurosci (2014) 52:353–365

(Swedo et al. 1998). Also for schizophrenia, bacterial infections seem to play a role for disease etiology, since among others, first trimester prenatal infections with NG have been shown to be able to increase the risk for the offspring of infected mothers to suffer from psychosis in later life (Babulas et al. 2006; Sørensen et al. 2009). Thus, bacterial infections might be one of multiple factors leading to the neurodevelopmental alterations in brain architecture found in this disease (Murray et al. 1992). NG is a common pathological agent during acute and subacute genital tract infections in both males and females (Edwards and Butler 2011), and NG-specific antibodies can be frequently detected in the blood serum of these patients (Hoffman et al. 1979). This suggests that unspecific crossreactivity of these antibodies with neuronal antigens could be causal for perturbed brain development upon early fetal exposure during pregnancy. Such a molecular mimicry like mode of action has already been demonstrated for several autoimmune pathologies (Moran et al. 1996), and according to this hypothesis, our results demonstrate that antibodies directed to NG are able to impair the length and numbers of neurites in PC12 cells in vitro. These in vitro findings could be of importance in order to explain a decreased neuropil formation in the prefrontal region and other cortical regions of the schizophrenic brain (Selemon et al. 1995; Black et al. 2004). Schizophrenia is known to be associated with enlarged ventricles and decreased cortical volume (Gur et al. 1994; Van Horn and McManus 1992; Woods et al. 2005; Wright et al. 2000) as well as with increased neuronal density probably due to impaired neurite formation (Benes et al. 1986, 1991; Selemon and Goldman-Rakic 1999; Selemon et al. 1995, 1998, 2003). According to this, at the cellular level, shorter basilar dendrites, smaller dendritic trees, and fewer collaterals have been shown in cortical neurons of schizophrenic patients (Kalus et al. 2000, 2002; Black et al. 2004; Broadbelt et al. 2002). This suggested that the findings of the present study provided new cues to the understanding of schizophrenia pathology. Such conclusions have to be drawn with extreme caution, since PC12 cells are only a pheochromocytoma cell line which by accident can be differentiated into a cholinergic-like neuronal phenotype (Sofroniew et al. 2001). They are indeed able to form neurite-like processes; however, these processes are not fully polarized, since they express both axonal and dendritic markers in the same process. Nevertheless, synapse-like structures are established between neurite-like processes in PC12 cells and functional synaptic transmission can also be detected (Jeon et al. 2010b). Due to these properties, one should always keep in mind that PC12 cells are an in vitro system and, by far, not sufficient to cover all the complex changes in brain development, leading in the end to schizophrenic psychosis in vivo. Due to the blood–brain barrier, the central nervous system is an immune-privileged space, and of which, antibodies and

361

immune cells are normally excluded (Muldoon et al. 2013). Therefore, if antibodies specific to NG should indeed be able to impair neurite outgrowth in vivo, the question remains: How they could pass the blood–brain barrier in order to reach their targets in the developing and/or mature brain? At the moment, one can only speculate about this; however, it has already been shown that blood–brain barrier permeability is increased at least in a subset of schizophrenic patients and that this can be associated with increased levels of IgG molecules in the cerebrospinal fluid (Müller and Ackenheil 1995; Vasic et al. 2012). An important question with regard to α-NG-dependent actions on neurite outgrowth in PC12 cells is whether these interactions occur in an antibody-specific fashion and not simply due to nonspecific interactions of immunoglobulins, in general. We clarified this by applying an antiserum directed to another member of the Neisseriaceae family of infectious bacteria, NM. It turned out that in contrast to α-NG, NMspecific antibodies were not able to impair neurite outgrowth in PC12 cells, thus confirming the specificity of the effects of α-NG on neurite outgrowth in PC12 cells. An important hypothesis for the pathogenesis of schizophrenia is the so-called dopamine hypothesis (Meltzer and Stahl 1976), which is based on the fact that neuroleptic drugs, the only effective treatment for psychotic symptomatology, are all known to interact with aminergic neurotransmitter receptors, i.e., type 2 dopamine receptors. However, the underlying mechanisms have still not yet been fully clarified (Nord and Farde 2011). With regard to this, the present study demonstrates that the α-NG-dependent impairment of neurite outgrowth in PC12 cells can be compensated by parallel treatment with neuroleptic drugs like haloperidol, risperidone, clozapine, and olanzapine. These findings confirm earlier studies on effects of atypical neuroleptic drugs on neurite outgrowth in PC12 cells (Lu and Dwyer 2005; Ishima et al. 2012). In contrast to the study of Lu and Dwyer (2005), we found here that also haloperidol was able to revert the α-NGdependent impairment of neurite outgrowth in PC12 cells. This discrepancy might be due to the lower concentrations of this drug applied here, whereas the higher concentration applied by Lu and Dwyer (2005) was found to be cytotoxic in our hands. The small therapeutic window of haloperidol is well-known from several clinical studies (Ulrich et al. 1998; Darby et al. 1995). NGF-dependent neurite outgrowth in PC12 cells is known to trigger changes in phosphorylation of a variety of transcription factors including FoxO3a and Stat3. According to this, results of the present study reveal that effects of α-NGspecific antibodies on neurite outgrowth in PC12 cells correlate indeed to changes in the phosphorylation state of FoxO3a and Stat3 (Wang et al. 2013; Ng et al. 2006), both of which are known to act neuritogenically. A role of Stat3 for neurite outgrowth in PC12 cells has been already stated earlier (Wu

362

and Bradshaw 1996), where especially mitochondrially located phosphorylated forms of this protein seem to be involved in NGF-dependent growth of neural processes (Zhou and Too 2011). Together with an earlier finding that the phosphorylation of Stat3 is involved in the NGF receptor signaling cascade (Ng et al. 2006), this suggests that the effects of α-NG on neurite outgrowth in PC12 cells could be probably due to direct interference with the NGF signaling cascade. Also for FoxO3a, a role in NGF-dependent neurite outgrowth of PC12 cells has already been previously reported (Wang et al. 2013) where FoxO3a has been shown to negatively regulate neurite formation in these cells. This effect could not be confirmed by our experiments, which are, however, in accordance to an earlier report (Wen et al. 2011) which demonstrates NGFdependent phosphorylation of FoxO3a at serine residue 253 during neurite outgrowth in PC12 cells. This discrepancy

J Mol Neurosci (2014) 52:353–365

might be due to different antibodies and/or cell clones used in our study as compared to the study of Wang et al. (2013). Despite the obvious correlation of the α-NG-dependent impairment of neurite outgrowth with phosphorylation of Stat3 and FoxO3a in vitro, the compensatory effect of neuroleptic drugs on this impairment revealed no such correlation. In this case, probably, other signaling pathways triggered by NGF or other growth factors might be the point of interference. Thus, also FGF-2 (Jeon et al. 2010a), cAMP (Chen et al. 2010), IGF-1 (Pugazhenthi et al. 1999), and EGF (Kasai et al. 2005) have already been demonstrated to act positively or negatively on neurite outgrowth in PC12 cells and are therefore likely candidates for such an alternative mechanism. Also the PI3-kinase/AKT pathway has already been demonstrated to be modulated by neuroleptic drugs in vitro (Lu and Dwyer 2005) and, therefore, could be involved as well.

J Mol Neurosci (2014) 52:353–365

ƒFig. 6

Effect of different neuroleptic drugs, on the α-NG-dependent decreases in phosphorylation of the neurite outgrowth-related transcription factors FoxO3a and Stat3, as revealed by Western blot analysis. a Western blot detection of the phosphorylated form of the transcription factor FoxO3a in total cellular protein of PC12 cells treated with NGF, or NGF and α-NG, either alone or in combination with haloperidol (HAL) or clozapine (CLZ). b Western blot detection of the phosphorylated form of the transcription factor Stat3 in total cellular protein of PC12 cells treated with NGF or NGF and α-NG, either alone or in combination with either HAL or CLZ. c Western blot detection of the cytoskeletal housekeeping protein β-actin in total cellular protein of PC12 cells treated with NGF or NGF and α-NG, either alone or in combination with either HAL or CLZ, as a loading control. d Diagram of the statistical evaluation of the densitometric evaluation of a series of experiments as shown in a, revealing a significant decrease in phosphorylation of the transcription factor FoxO3a in PC12 cells treated with NGF together with α-NG, as compared to control cells treated with NGF only. In contrast, no significant increase in phosphorylation of this protein could be detected with regard to cells treated with NGF together with α-NG, when the cells were treated additionally with HAL or CLZ. e Diagram of the statistical evaluation of the densitometric evaluation of a series of experiments as shown in (b), revealing a significant decrease in phosphorylation of the transcription factor Stat3 in PC12 cells treated with NGF together with αNG, as compared to control cells treated with NGF only. In contrast, no significant increase in phosphorylation of this protein could be detected with regard to cells treated with NGF together with α-NG, when the cells were treated additionally with HAL or CLZ. f Western blot detection of the phosphorylated form of the transcription factor FoxO3a in total cellular protein of PC12 cells treated with NGF or NGF and α-NG, either alone or in combination with risperidone (RIS ) or olanzapine (OLA). g Western blot detection of the phosphorylated form of the transcription factor Stat3 in total cellular protein of PC12 cells treated with NGF or NGF and α-NG, either alone or in combination with either RIS or OLA. h Western blot detection of the cytoskeletal housekeeping protein β-actin in total cellular protein of PC12 cells treated with NGF or NGF and α-NG, either alone or in combination with either RIS or OLA, as a loading control. i Diagram of the statistical evaluation of the densitometric evaluation of a series of experiments as shown in f , revealing a significant decrease in phosphorylation of the transcription factor FoxO3a in PC12 cells treated with NGF together with α-NG, as compared to control cells treated with NGF only. In contrast, no significant increase in phosphorylation of this protein could be detected with regard to cells treated with NGF together with α-NG, when the cells were treated additionally with RIS or OLA. k Diagram of the statistical evaluation of the densitometric evaluation of a series of experiments as shown in g, revealing a significant decrease in phosphorylation of the transcription factor Stat3 in PC12 cells treated with NGF together with αNG, as compared to control cells treated with NGF only. In contrast, no significant increase in phosphorylation of this protein could be detected with regard to cells treated with NGF together with α-NG, when the cells were treated additionally with RIS or OLA

This leads finally to the question: How NG-specific autoantibodies might act on neurite outgrowth in vitro and in vivo? One could speculate that either a secreted factor such as NGF or another member of the neurotrophin family of growth factors or a membrane-bound receptor such as one of the members of the Trk family of neurotrophin receptors would be most likely to be involved in such a mechanism. Altered levels of NGF and other neurotrophins in the serum and cerebrospinal fluid of schizophrenic patients have already been reported (Martinotti et al. 2012) as well as autoantibodies

363

directed to NGF in the serum of both mothers of early onset schizophrenic patients (Bashina et al. 1997) as well as in the patients themselves (Zakharenko et al. 1999). However, it remains a challenging task for an upcoming project to identify possible neuronal interaction partners for α-NG on the molecular level. In conclusion, the present in vitro study demonstrates for the first time that antibodies directed to NG-specific bacterial surface proteins are able to cross-react with neuronal antigens, thereby leading to impaired neurite outgrowth. They demonstrate also that this impairment can be overcome by a parallel treatment of the cells with neuroleptic drugs. In addition, they show that the effects of bacteria-specific antibodies may be mediated by the phosphorylation of the neuritogenic transcription factors FoxO3a and Stat3. Thus, they outline a putative mechanism by which bacterially elicited autoantibodies could interfere with brain development and add new evidence to an effect of neuroleptic drugs on neuropil formation. Together, these findings provide a new in vitro mechanism with putative importance for the understanding of the brain pathology underlying schizophrenic psychoses. However, the exact in vivo relevance of this mechanism remains still to be clarified in the future. Acknowledgments I would like to thank the Medical Faculty of the University of Göttingen (UMG) for their persistent and reliable support of my work. Conflict of Interest interests

The author declares no competing financial

References Babulas V, Factor-Litvak P, Goetz R, Schaefer CA, Brown AS (2006) Prenatal exposure to maternal genital and reproductive infections and adult schizophrenia. Am J Psychiatry 163:927–929 Bashina VM, Kozlova IA, Kliushnik TP, Simashkova NV, Danilovskaia EV, Gorbachevskaia NL, Turkova IL, Iakupova LP, Grachev VV (1997) An elevation in the level of auto antibodies to nerve-growth factor in the blood serum of schizophrenic children. Zh Nevrol Psikhiatr Im S S Korsakova 97:47–51 Benes FM, Davidson J, Bird ED (1986) Quantitative cytoarchitectural studies of the cerebral cortex of schizophrenics. Arch Gen Psychiatry 43:31–35 Benes FM, McSparren J, Bird ED, SanGiovanni JP, Vincent SL (1991) Deficits in small interneurons in prefrontal and cingulate cortices of schizophrenic and schizoaffective patients. Arch Gen Psychiatry 48: 996–1001 Black JE, Kodish IM, Grossman AW, Klintsova AY, Orlovskaya D, Vostrikov V, Uranova N, Greenough WT (2004) Pathology of layer V pyramidal neurons in the prefrontal cortex of patients with schizophrenia. Am J Psychiatry 161:742–744 Broadbelt K, Byne W, Jones LB (2002) Evidence for a decrease in basilar dendrites of pyramidal cells in schizophrenic medial prefrontal cortex. Schizophr Res 58:75–81 Chen MC, Lin H, Hsu FN, Huang PH, Lee GS, Wang PS (2010) Involvement of cAMP in nerve growth factor-triggered p35/Cdk5

364 activation and differentiation in PC12-cells. Am J Physiol Cell Physiol 299:C516–C527 Dale RC, Brilot F (2012) Autoimmune basal ganglia disorders. J Child Neurol 27:1470–1481 Darby JK, Pasta DJ, Dabiri L, Clark L, Mosbacher D (1995) Haloperidol dose and blood level variability: toxicity and interindividual and intraindividual variability in the nonresponder patient in the clinical practice setting. J Clin Psychopharmacol 15:334–340 Edwards JL, Butler EK (2011) The pathobiology of Neisseria gonorrhoeae lower female genital tract infection. Front Microbiol 2(102):1–12 Greene LA (1978) Nerve growth factor prevents the death and stimulates the neuronal differentiation of clonal PC12 pheochromocytoma cells in serum-free medium. J Cell Biol 78:747–755 Gur RE, Mozley PD, Shtasel DL, Cannon TD, Gallacher F, Turetsky B, Grossman R, Gur RC (1994) Clinical subtypes of schizophrenia: differences in brain and CSF volume. Am J Psychiatry 151:343–350 Harrison PJ (1999) The neuropathology of schizophrenia. A critical review of the data and their interpretation. Brain 122:593–624 Henkel AW, Bieger SC (1994) Quantification of proteins dissolved in an electrophoresis sample buffer. Anal Biochem 223:329–331 Hoffman TA, Damus AJ, Sands L (1979) Evaluation of a gonococcal serologic test. Am J Clin Pathol 71:184–189 Ishima T, Iyo M, Hashimoto K (2012) Neurite outgrowth mediated by the heat shock protein Hsp90α: a novel target for the antipsychotic drug aripiprazole. Transl Psychiatry 2(e170):1–12 Jeon CY, Kim HJ, Morii H, Mori N, Settleman J, Lee JY, Kim J, Kim SC, Park JB (2010a) Neurite outgrowth from PC12-cells by basic fibroblast growth factor (bFGF) is mediated by RhoA inactivation through p190RhoGAP and ARAP3. J Cell Physiol 224:786–794 Jeon CY, Jin JK, Koh YH, Chun W, Choi IG, Kown HJ, Kim YS, Park JB (2010b) Neurites from PC12 cells are connected to each other by synapse-like structures. Synapse 64:765–772 Kalus P, Muller TJ, Zuschratter W, Senitz D (2000) The dendritic architecture of prefrontal pyramidal neurons in schizophrenic patients. Neuroreport 11:3621–3625 Kalus P, Bondzio J, Federspiel A, Muller TJ, Zuschratter W (2002) Celltype specific alterations of cortical interneurons in schizophrenic patients. Neuroreport 13:713–717 Kaplan DR, Stephens RM (1994) Neurotrophin signal transduction by the Trk receptor. J Neurobiol 25:1404–1417 Kasai A, Shima T, Okada M (2005) Role of Src family tyrosine kinases in the down-regulation of epidermal growth factor signaling in PC12cells. Genes Cells 10:1175–1187 Khandaker GM, Zimbron J, Lewis G, Jones PB (2013) Prenatal maternal infection, neurodevelopment and adult schizophrenia: a systematic review of population-based studies. Psychol Med 43:239–257 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 Levi A, Biocca S, Cattaneo A, Calissano P (1988) The mode of action of nerve growth factor in PC12-cells. Mol Neurobiol 2:201–226 Lu XH, Dwyer DS (2005) Second-generation antipsychotic drugs, olanzapine, quetiapine, and clozapine enhance neurite outgrowth in PC12-cells via PI3K/AKT, ERK, and pertussis toxin-sensitive pathways. J Mol Neurosci 27:43–64 Martinotti G, Di Iorio G, Marini S, Ricci V, De Berardis D, Di Giannantonio M (2012) Nerve growth factor and brain-derived neurotrophic factor concentrations in schizophrenia: a review. J Biol Regul Homeost Agents 26:347–356 Meltzer HY, Stahl SM (1976) The dopamine hypothesis of schizophrenia: a review. Schizophr Bull 2:19–76 Moran AP, Prendergast MM, Appelmelk BJ (1996) Molecular mimicry of host structures by bacterial lipopolysaccharides and its contribution to disease. FEMS Immunol Med Microbiol 16:105–115 Muldoon LL, Alvarez JI, Begley DJ, Boado RJ, Del Zoppo GJ, Doolittle ND, Engelhardt B, Hallenbeck JM, Lonser RR, Ohlfest JR, Prat A,

J Mol Neurosci (2014) 52:353–365 Scarpa M, Smeyne RJ, Drewes LR, Neuwelt EA (2013) Immunologic privilege in the central nervous system and the blood–brain barrier. J Cereb Blood Flow Metab 33:13–21 Müller N, Ackenheil M (1995) Immunoglobulin and albumin content of cerebrospinal fluid in schizophrenic patients: relationship to negative symptomatology. Schizophr Res 14:223–228 Murray RM, O’Callaghan E, Castle DJ, Lewis SW (1992) A neurodevelopmental approach to the classification of schizophrenia. Schizophr Bull 18:319–332 Ng YP, Cheung ZH, Ip NY (2006) STAT3 as a downstream mediator of Trk signaling and functions. J Biol Chem 281:15636–15644 Nord M, Farde L (2011) Antipsychotic occupancy of dopamine receptors in schizophrenia. CNS Neurosci Ther 17:97–103 Pugazhenthi S, Boras T, O’Connor D, Meintzer MK, Heidenreich KA, Reusch JE (1999) Insulin-like growth factor I-mediated activation of the transcription factor cAMP response element-binding protein in PC12-cells. Involvement of p38 mitogen-activated protein kinasemediated pathway. J Biol Chem 274:2829–2837 Rosenstein NE, Perkins BA, Stephens DS, Popovic T, Hughes JM (2001) Meningococcal disease. N Engl J Med 344:1378–1388 Selemon LD, Goldman-Rakic PS (1999) The reduced neuropil hypothesis: a circuit based model of schizophrenia. Biol Psychiatry 45:17– 25 Selemon LD, Rajkowska G, Goldman-Rakic PS (1995) Abnormally high neuronal density in the schizophrenic cortex. A morphometric analysis of prefrontal area 9 and occipital area 17. Arch Gen Psychiatry 52:805–818 Selemon LD, Rajkowska G, Goldman-Rakic PS (1998) Elevated neuronal density in prefrontal area 46 in brains from schizophrenic patients: application of a three dimensional, stereologic counting method. J Comp Neurol 392:402–412 Selemon LD, Mrzljak J, Kleinman JE, Herman MM, Goldman-Rakic PS (2003) Regional specificity in the neuropathologic substrates of schizophrenia: a morphometric analysis of Broca’s area 44 and area 9. Arch Gen Psychiatry 60:69–77 Sofroniew MV, Howe CL, Mobley WC (2001) Nerve growth factor signaling, neuroprotection, and neural repair. Annu Rev Neurosci 24:1217–1281 Sørensen HJ, Mortensen EL, Reinisch JM, Mednick SA (2009) Association between prenatal exposure to bacterial infection and risk of schizophrenia. Schizophr Bull 35:631–637 Stephens DS, Greenwood B, Brandtzaeg P (2007) Epidemic meningitis, meningococcaemia, and Neisseria meningitidis. Lancet 369:2196– 2210 Swedo SE, Leonard HL, Garvey M, Mittleman B, Allen AJ, Perlmutter S, Lougee L, Dow S, Zamkoff J, Dubbert BK (1998) Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases. Am J Psychiatry 155:264–271 Tandon R, Nasrallah HA, Keshavan MS (2009) Schizophrenia, “just the facts” 4. Clinical features and conceptualization. Schizophr Res 110: 1–23 Ulrich S, Neuhof S, Braun V, Meyer FP (1998) Therapeutic window of serum haloperidol concentration in acute schizophrenia and schizoaffective disorder. Pharmacopsychiatry 31:163–169 Van Horn JD, McManus IC (1992) Ventricular enlargement in schizophrenia. A meta-analysis of studies of the ventricle:brain ratio (VBR). Br J Psychiatry 160:687–697 Vasic N, Connemann BJ, Wolf RC, Tumani H, Brettschneider J (2012) Cerebrospinal fluid biomarker candidates of schizophrenia: where do we stand? Eur Arch Psychiatry Clin Neurosci 262:375–391 Wang H, Duan X, Ren Y, Liu Y, Huang M, Liu P, Wang R, Gao G, Zhou L, Feng Z, Zheng W (2013) FoxO3a negatively regulates nerve growth factor-induced neuronal differentiation through inhibiting the expression of neurochondrin in PC12-cells. Mol Neurobiol 47: 24–36

J Mol Neurosci (2014) 52:353–365 Wen Q, Duan X, Liao R, Little P, Gao G, Jiang H, Lalit S, Quirion R, Zheng W (2011) Characterization of intracellular translocation of Forkhead transcription factor O (FoxO) members induced by NGF in PC12-cells. Neurosci Lett 498:31–36 Woods BT, Ward KE, Johnson EH (2005) Meta-analysis of the time-course of brain volume reduction in schizophrenia: implications for pathogenesis and early treatment. Schizophr Res 73:221–228 Wright IC, Rabe-Hesketh S, Woodruff PW, David AS, Murray RM, Bullmore ET (2000) Meta-analysis of regional brain volumes in schizophrenia. Am J Psychiatry 157:16–25

365 Wu YY, Bradshaw RA (1996) Induction of neurite outgrowth by interleukin-6 is accompanied by activation of Stat3 signaling pathway in a variant PC12 cell (E2) line. J Biol Chem 271: 13023–13032 Zakharenko OM, Kliushnik TP, Kozlova IA, Kozlovskaia GV, Danilovskaia EV, Bashina VM, Simashkova VM, Goriunova AV, Kalinina MA (1999) Nerve growth factor auto-antibodies in the sera of mothers of schizophrenic children and children from high risk group. Zh Nevrol Psikhiatr Im S S Korsakova 99:44–46 Zhou L, Too HP (2011) Mitochondrial localized STAT3 is involved in NGF induced neurite outgrowth. Plops One 6(e21680):1–13