Proteomic Investigation of the Hippocampus in Prenatally Stressed Mice Implicates Changes in Membrane Trafficking, Cytoskeletal, and Metabolic Function

Original Paper

Dev Neurosci 2014;36:432–442 Received: December 4, 2013 DOI: 10.1159/000365327 Accepted after revision: June 17, 2014 Published online: August 14, 2014

a d d d Melanie Föcking Rianne Opstelten Jos Prickaerts Harry W.M. Steinbusch b d, e a, c Michael J. Dunn Daniël L.A. van den Hove David R. Cotter a b Department of Psychiatry, Royal College of Surgeons in Ireland, Proteome Research Centre, UCD Conway c Institute of Biomolecular and Biomedical Research, School of Medicine and Medical Sciences, and Department of d Psychiatry, Beaumont Hospital, Dublin , Ireland; Department of Neuroscience, Faculty of Health Medicine and e Life Sciences, Maastricht University, Maastricht , The Netherlands; Institute of Molecular Psychiatry, Laboratory of Translational Neuroscience, Department of Psychiatry, Psychosomatics and Psychotherapy, University of Würzburg, Würzburg , Germany

Key Words blotting. In comparison to controls, 47 protein spots (38 in- Proteomics · 2-Dimensional difference in gel dividual proteins) were found to be differentially expressed electrophoresis · Hippocampus · Prenatal maternal stress · in the hippocampus of prenatally stressed mice. Functional Trafficking grouping of these proteins revealed that prenatal stress influenced the expression of proteins involved in brain development, cytoskeletal composition, stress response, and en- Abstract ergy metabolism. Western blotting was utilized to validate Prenatal stress influences the development of the fetal brain the changes in calretinin, hippocalcin, profilin-1 and the sig- and so contributes to the risk of the development of psychi- nal-transducing adaptor molecule STAM1. Septin-5 could atric disorders in later life. The hippocampus is particularly not be validated via Western blotting due to methodological sensitive to prenatal stress, and robust abnormalities have issues. Closer investigation of the validated proteins also been described in the hippocampus in schizophrenia and pointed to an interesting role for membrane trafficking def- depression. The aim of this study was to determine whether icits mediated by prenatal stress. Our findings demonstrate prenatal stress is associated with distinct patterns of differ- that prenatal stress leads to altered hippocampal protein ex- ential protein expression in the hippocampus using a vali- pression, implicating numerous molecular pathways that dated mouse model. We therefore performed a comparative may provide new targets for psychotropic drug develop- proteomic study assessing female hippocampal samples ment. © 2014 S. Karger AG, Basel from 8 prenatally stressed mice and 8 control mice. Differen- tial protein expression was assessed using 2-dimensional difference in gel electrophoresis and subsequent mass spectrometry. The observed changes in a selected group of dif- Daniël L.A. van den Hove and David R. Cotter are senior/final authors ferentially expressed proteins were confirmed by Western of this work.

© 2014 S. Karger AG, Basel Melanie Föcking 0378–5866/14/0365–0432$39.50/0 Department of Psychiatry, Royal College of Surgeons in Ireland Education and Research Centre, Beaumont Hospital E-Mail [email protected] Dublin 9 (Ireland) www.karger.com/dne E-Mail mfocking @ rcsi.ie Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM Introduction Based on previous proteomic studies on psychiatric disorders in postmortem brains [31–33] , we expected that met- Even though the onset of schizophrenia and affective abolic and cytoskeletal pathways and processes would be disorders takes place in early adulthood, the underlying dysregulated in the PS mouse model. In order to direct cause may occur much earlier, possibly already in utero. future research, we applied a hypothesis-generating ap- Developmental lesions may be the result of intrinsic (epi) proach in this study, allowing for an extensive discussion genetic changes, extrinsic stressors, or, most likely, both on possibly interesting pathways that are affected by PS. [1] . Stress exposure of the pregnant mother has been shown to induce alterations in the fetal environment [2, 3] and may predispose the fetus to adult psychopathology. M e t h o d s In humans, the types of prenatal stressors are widely variable, ranging from malnutrition to infection and Animal Samples and PS Tissue samples used for proteomic analyses were obtained emotional trauma [4] . Likewise, the long-term behavior- from 8 female PS mice that underwent behavioral testing for de- al and psychiatric outcomes of prenatal stress (PS) vary pressive-like symptoms in a previous study (for details, see online greatly and include attention, motor, and cognitive defi- suppl. material; for all online suppl. material, see www.karger. cits as well as psychiatric disorders [2, 5] . Emotional ma- com/doi/10.1159/000365327 and [23]). Ethical approval for this ternal stress, in particular when experienced during early study was granted by the Animal Ethics Board of the University of Maastricht. to mid-gestation, has been shown to increase the risk of the development of depression [6, 7] and schizophrenia Separation of Proteins by 2-Dimensional Gel Electrophoresis [8–10] in adult offspring. However, studies differ in terms and Image Analysis of which gender is more vulnerable and with regard to the Samples were processed and separated by 2-dimensional differ- sensitive time window; some studies have even failed to ence in gel electrophoresis (2D-DIGE) as described previously by our group [34–37] (for details, see online suppl. material). After electro- show any effects [11] . phoresis, scanning of the gels with CyDye-labeled proteins was per- Prenatal restraint stress is a well-established animal formed on a Typhoon TM 9410 image scanner (GE Healthcare, UK). model that is widely validated in terms of behavior [5, 12– Prescans were performed to adjust the photomultiplier tube to obtain 15]. In the restraint stress procedure, pregnant dams are images with a maximum intensity of 60,000–80,000 units. Images TM restrained in their last week of pregnancy for 45 min in 3 were cropped using ImageQuant software (GE Healthcare), and protein quantification across experimental groups was carried out daily sessions [16] . The model is particularly congruent using Progenesis SameSpots® (Nonlinear Dynamics, UK). with the human situation in terms of the increased anxi- ety-related [17, 18] and depressive-like behavior [19, 20] Statistical Analysis observed in the PS offspring and also shows some overlap For the 2D-DIGE data, the Progenesis SameSpots software with deficits that are typically seen in schizophrenia [21] . (Nonlinear Dynamics) was used to match and average individual runs for each group, which generated a list of significantly differ- Furthermore, similar to human pregnancy, stress hor- entially expressed protein spots (ANOVA, p ≤ 0.05) between the mones from the pregnant mouse can reach the fetal blood PS and control groups. circulation, thereby affecting brain development and in- creasing stress sensitivity [22]. The effects of PS seem to Protein Identification by Mass Spectrometry depend on the genetic background of the animal and are Statistically significant protein spots were selected for protein identification by mass spectrometry based on protein spot intensity. predominantly sex specific [23–25]. In a previous behav- Protein spots were extracted from preparative gels, destained, and ioral study by our group, only the female PS mouse off- digested with trypsin as previously described [38] . Mass spectrom- spring showed a depressive-like phenotype [23] . etry was performed on an Agilent 6520 Accurate-Mass Q-TOF with The aim of this study was to define distinct patterns of HPLC-Chip Cube and a 1200 series HPLC system attached to pro- differential protein expression in the hippocampus of mice vide nanoflow separation (for details, see online suppl. material). exposed to PS. We selected the hippocampus because it is of major importance in the pathophysiology of both depression [26] and schizophrenia [27] . Furthermore, this R e s u l t s brain structure has been shown to be affected by early life stress in humans [28, 29] . As the female mice displayed the 2D-DIGE Analysis most pronounced behavioral phenotype in our previous On the master gel image, a total of 1,418 protein spots study [23] as well as in a recent microarray study [30] , their were detected and matched across all gel images and sta- hippocampal tissue was the focus of the present study. tistically analyzed using ANOVA. In total, 47 protein

Proteomic Investigation of the Dev Neurosci 2014;36:432–442 433 Hippocampus DOI: 10.1159/000365327 Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM 4 pH 7 70 Color version available online MW (kDa)

Fig. 1. Representative 2D-DIGE (pH 4–7) indicating the protein spots found to be significantly altered in PS mice and subse- quently identified with mass spectrometry. Protein identification numbers relate to the proteins listed in table 1. MW = Molecular 6 weight. spots were found to be differentially expressed in the hip- In brief, the following proteins were assessed using pocampal tissue of the PS group compared to the control Western blotting. These proteins were selected on the ba- animals. The identified protein spots are indicated on a sis of antibody availability, fold changes, the strength of representative 2D-DIGE gel in figure 1. statistical differences, and the degree to which the proteins represented implicated functional pathways. Functional Clustering of Differentially Expressed Calretinin. The results of 2D-DIGE showed an in- Proteins creased expression in the PS group (+1.21-fold; p < 0.01). The proteins shown to be significantly altered by 2D- An increased expression in PS mice compared to controls DIGE and identified by mass spectrometry were grouped was confirmed (p < 0.05). according to biological function by searching the UniProt Hippocalcin. The results of 2D-DIGE showed an in- database (http://www.uniprot.org). A list of these find- creased expression in the PS group (+1.25-fold; p < 0.05). ings, including the functional ontology of the proteins, Western blotting of individual samples from the PS group can be found in table 1 . The proteins could be subdivided and controls confirmed upregulation in the PS group (p < into 7 functional clusters (figure 2). The highest number 0.05). of altered proteins (n = 14) was matched to developmen- Profilin1. The results of 2D-DIGE showed an increased tal pathways and the cellular stress response (n = 9). En- expression in the PS group (+1.45-fold; p < 0.05). An in- ergy metabolism (n = 8) and protein modification path- creased expression in PS mice compared to controls was ways (n = 6) were also affected by PS. Of note, proteins confirmed (p < 0.05). with a role in cellular trafficking (n = 4), calcium buffer- Signal-Transducing Adaptor Molecule. Results from ing (n = 2), and myelination processes (n = 2) were found 2D-DIGE showed a reduced expression in the PS group to be differentially expressed in the PS group. (–1.31-fold; p < 0.05). A reduced expression in PS mice compared to controls was confirmed (p < 0.05). Validation of Proteomic Findings Septin-5. 2D-DIGE showed a downregulation of 22.6% Extensive validation to confirm the findings obtained (p < 0.01) in the PS offspring. Immunoblot data could not using 2D-DIGE was conducted via Western blotting. See confirm differential expression in septin-5 (SEPT5). figure 3 for the details of this validation work. Since SEPT5 has been reported to have 5 isoforms in hu-

434 Dev Neurosci 2014;36:432–442 Föcking/Opstelten/Prickaerts/Steinbusch/ DOI: 10.1159/000365327 Dunn/van den Hove/Cotter Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM 0.023 0.048 0.048 0.049 0.046 0.05 0.033 0.003 0.007 0.049 0.014 0.006 0.005 0.044 0.04 0.016 0.0005 0.04 0.014 0.007 0.044 0.046 0.035 0.033 0.049 0.023 0.0005 0.043 0.029 0.004 0.035 0.036 0.032 0.029 0.008 0.015 0.029 0.030 1.12 1.04 1.12 1.06 1.03 1.34 1.10 1.21 1.25 1.14 1.20 1.20 1.11 1.07 1.41 1.56 1.25 1.20 1.10 1.60 1.14 1.04 1.45 –1.22 –1.12 –1.13 –1.31 –1.23 –1.20 –1.03 –1.13 –1.1.20 –1.06 –1.06 –1.09 –1.38 –1.09 –1.22 Two isoforms differentially expressed. a ENSMUSG00000061983 ENSMUSG00000025967 ENSMUSG00000002102 ENSMUSG00000068749 ENSMUSG00000028967 ENSMUSG00000044573 ENSMUSG00000026718 ENSMUSG00000009394 ENSMUSG00000072214 ENSMUSG00000021546 ENSMUSG00000032261 ENSMUSG00000033953 ENSMUSG00000003657 ENSMUSG00000028785 ENSMUSG00000022055 ENSMUSG00000026202 ENSMUSG00000058672 ENSMUSG00000062825 ENSMUSG00000034336 Gene ID Fold change p value ENSMUSG00000020766 ENSMUSG00000041607 ENSMUSG00000024359 ENSMUSG00000015656 ENSMUSG00000020048 ENSMUSG00000028964 ENSMUSG00000027248 ENSMUSG00000005161 ENSMUSG00000024953 ENSMUSG00000032279 ENSMUSG00000027332 ENSMUSG00000035561 ENSMUSG00000025393 ENSMUSG00000004267 ENSMUSG00000022048 ENSMUSG00000024501 ENSMUSG00000020849 ENSMUSG00000051391 ENSMUSG00000018293 s. ANOVA, p ≤ 0.05. RPS12 EEF1B2 PSMC3 PSMA5 ERRFI1 ACP1 STAM1 SYN2 SEPT5 HNRNPK SH3BGRL2 PPP3R1 CALB2 HPCA NEFL TUBA4A TUBB2A ACTG1 INA Gene name GALK1 MBP HSPA9 HSPA8 HSP90B1 PARK7 PDIA3 PRDX2 PRDX5 IDH3A IVD ALDH1B1 ATP5B ENO2 DPYSL2 DPYSL3 YWHAE YWHAG PFN1 P63323 O70251 O88685 Q9Z2U1 Q99JZ7 Q9D358 P70297 Q64332 Q9Z2Q6 P61979 Q8BG73 Q63810 Q08331 P84075 P08551 P68368 Q7TMM9 P63260 P46660 accession No. Q9R0N0 P04370 P38647 P63017 P08113 Q99LX0 P27773 Q61171 P99029 Q9D6R2 Q9JHI5 Q9CZS1 P56480 P17183 O08553 Q62188 P62259 P61982 P62962 b

a a

a 26S protease regulatory subunit 6A Proteasome subunit alpha type-5 ERB-B receptor feedback inhibitor Low-molecular-weight phosphotyrosine protein phosphatase Elongation factor 1-beta Isocitrate dehydrogenase subunit alpha, mitochondrial Signal-transducing adapter molecule 1 Synapsin-2 Septin-5 Dihydropyrimidase-related protein 2 SH3 domain-binding glutamic acid-rich-like protein 2 Protein DJ1 Protein disulfide isomearse A3 Peroxiredoxin-2 Peroxirdoxin-5, mitochondrial Calcineurin subunit B type Calretinin Neuron-specific calcium-binding protein hippocalcin Gamma-enolase Tubulin alpha-4 chain Tubulin beta-2A chain Actin, cytoplasmic 2 Alpha-internexin Protein name UniProt Dihydropyrimidase-related protein 3 14-3-3 protein epsilon Myelin basic protein Endoplasmin Isovaleryl CoA dehydrogenase, mitochondrial Aldehyde dehydrogenase X, mitochondrial ATP synthase beta, mitochondrial 14-3-3 protein gamma Profilin-1 Heat shock cognate protein, 71 kDa Catabolism Phosphatase Heat shock Stress 70 protein, mitochondrial Biosynthesis 40 S ribosomal protein S12 Mitochondrial ATP production Redox homeostatsis Cytoskeletal components Neurofilament light polypeptide Specific function Summary of the differentially expressed protein spots Identified by mass spectrometry together with information on the fold change, functional ontology, and grouping of protein Three isoforms differentially expressed. Stress response Protein modifications Energy metabolism Endo-/ exocytosis CNS development/ cytoskeletal Other Heterogeneous nuclear riboprotein K Calcium sensoring Table 1. Functional ontology b Myelination Galaktokinase

Proteomic Investigation of the Dev Neurosci 2014;36:432–442 435 Hippocampus DOI: 10.1159/000365327 Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM 30

10 Proteins regulated by PS (%) regulated Proteins 0 Other cytoskeletal Fig. 2. Chart displaying the categories into Myelination

which significantly altered protein spots response Stress could be grouped. Clustering was done Calcium sensors CNS development/ Energy metabolism Energy Protein modifications Protein manually using the UniProt database. Membrane trafficking CNS = Central nervous system.

110 * 100 90 CALB2 31 kDa 80 70 ERK2 42 kDa Control (%) Control 60 CCPPPCPPC P 50 Color version available online a Control PS

120 110 * 100 PFN1 15 kDa 90 80 ERK2 42 kDa 70 Control (%) Control Fig. 3. Validation of differentially ex- 60 CPPCPPPCP P pressed proteins. * p ≤ 0.05, ** p < 0.01. 50 Protein expression changes were assessed b Control PS by Western blot analysis of whole hippocampal lysates from 8 animals per group. 110 Results were averaged over 3 consecutive 100 runs for each protein of interest. The re- ** 90 STAM1 78 kDa sults in the bar graphs are presented as per- 80 centages of the control group. Error bars 70 ERK2 42 kDa indicate the standard error of the mean. (%) Control 60 Next to each bar graph is a representative 50 CCCCCPPCC P image of an immunoblot stained with a c Control PS specific antibody. Samples were loaded onto SDS-PAGE gels in a randomized or- 120 der. Immunoblots were incubated with an- 110 * tibodies specific to calretinin at ∼ 31 kDa 100 HPCA 22 kDa (a ), PFN1 at ∼ 15 kDa (b ), STAM1 at ∼ 78 90 80 kDa (c ), and HPCA at ∼ 22 kDa (d ). All ERK2 42 kDa 70 Control (%) Control membranes were counterstained with an 60 CCCCPPPPC C antibody against ERK2 at ∼ 43 kDa to con- 50 trol for loading differences. CALB2 = Cal- d Control PS retinin.

436 Dev Neurosci 2014;36:432–442 Föcking/Opstelten/Prickaerts/Steinbusch/ DOI: 10.1159/000365327 Dunn/van den Hove/Cotter Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM 506 sonable to speculate that the effect of downregulation of 840 9 one of these isoforms (as identified by 2D-DIGE) was masked by different changes in the others in immunob- lotting, and to the best of our knowledge no isoform-specific antibodies are available for SEPT5.

1,228 938 Discussion

Fig. 4. 2-D Western blot of SEPT5 in whole hippocampal lysate. While prenatal maternal stress has known effects on Tissue was pooled from all samples, separated in 2 dimensions (pH and molecular weight), and blotted onto a membrane. The protein behavior and cognition, the mechanism by which these spot identified as being differentially expressed in the primary effects occur is not fully understood. Our results demon- analysis (No. 9) is listed, in addition to the others identified by 2-D strate that female mice that have been stressed in utero blot (No. 506, 840, 938, and 1,228). See table 3 for fold differences show various protein expression differences in the hip- between the PS and control groups for these spots (corresponding pocampus in comparison to controls. In total, 47 protein numbers are used). spots – representing 38 individual proteins – were found to be differentially expressed in whole hippocampal lysates of the PS group. Functional clustering of these ex- Table 2. Five isoforms of SEPT5 (1 confirmed by mass spectrom- pression changes indicated alterations in cytoskeletal etry and 4 additionally confirmed by 2-D Western blot) proteins, metabolism, membrane trafficking, and my- Master spot No. Fold change ANOVA elination pathways in the brains of PS mice.

20 –1.226 0.007 Comparisons to Differentially Expressed Proteins 506 +1.080 0.083 from the Literature 840 +1.040 0.461 938 –1.032 0.563 To complement the results gathered in our hypothe- 1,228 +1.012 0.804 sis-neutral approach, we compared our results to those of previous proteomic animal studies investigating the ef- The altered expression and significance level of each of these fects of PS on the hippocampus. We found that 13 out of protein spots in prenatal stress in comparison to controls are listed. Protein spot 20 was revealed by primary analysis (p = 0.007) and our 38 proteins had been shown to be differentially ex- spots 506, 840, 938, and 1,228 could be traced back to the original pressed in these other animal studies. Furthermore, we 2D-DIGE data after 2-D Western blot analysis. None of the latter also compared our results to those of postmortem studies spots gained significance; the majority of the spots were found to in humans and observed that 7 proteins found to be dys- be slightly upregulated, and one was minimally downregulated. regulated by PS in this study had previously been shown This probably explains why the validation of SEPT5 in our sample failed when using 1-D Western blotting, as there is no marker for to be dysregulated in major psychiatric disorders (for de- specific isoforms of SEPT5. tails, see table 3 ). Comparison to Other Proteomic Studies. A recent proteomics study [39] looking at the effects of PS on the rat hippocampus identified 26 differentially expressed pro- mans, which could result from differences in posttransla- teins. While there was an overlap in only 3 individual pro- tional modifications, possibly including phosphoryla- teins between that rat study and our own study (i.e. dihy- tions [36] , we performed a 2-D Western blot (for details, dropyramidase-related protein-2, synapsin-2, and see online suppl. material) where the membrane was α-isocitrate dehydrogenase), there was a significant over- stained with the same SEPT5-specific antibody that had lap in terms of pathways and processes as both studies been used for the 1-D Western blot. This confirmed the implicated cytoskeletal proteins, energy metabolism, syn- presence of 5 SEPT5 isoforms, i.e. 5 spots for SEPT5, aptic transduction, and synaptic vesicles. within the mouse hippocampus ( fig. 4 ). Consequentially, Ten of the 47 proteins identified in our study were also we went back to the primary analysis and reexamined the found to be differentially expressed in a recent proteomic data for the 4 newly identified spots and indeed found 1 study in human hippocampal tissue from schizophrenia further spot to be minimally downregulated, whereas the and bipolar disorder patients [33]; these included chang- other 3 were moderately upregulated ( table 2 ). It is rea- es in 6 cytoskeletal proteins (ACTG1, DPYSL2, DPYSL3,

Proteomic Investigation of the Dev Neurosci 2014;36:432–442 437 Hippocampus DOI: 10.1159/000365327 Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM Table 3. Comparison to differentially expressed proteins in the literature

Animal studies Human studies current study PS mouse microarray data PS rat study SCZ BPD biomarker for SCZ? EG × E

CNS development/cytoskeletal proteins ACTG1 ↑↑ DPYSL2 ↓↓↓X DPYSL3 ↓↓↓ INAa ↓↓ NEFL ↓↓↓X PFN1 ↑b ↓ TUBB2A ↓↑ ↑ X YWHAE ↓↓↑ ↓↓X Membrane trafficking STAM1 ↓b SYN2 ↑↓ ↓ HPCA ↑b Protein modifications PSMA5 ↑↑ Stress response HSPA8 ↓↓↓ HSPA9 ↓↓a PARK7 ↑↓ Energy metabolism ATP5B ↑↓↑↓↑ ENO2 ↑↓X IDH3A ↑↓ IVD ↑↓ Calcium sensors CALB2 ↑b Myelination MBP ↑↑↑ X

All proteins found to be significantly altered in this study were chitectural changes were mentioned in at least 1 of these publica- compared to significant findings in a human study that evaluated tions. Two proteins involved in membrane trafficking could be protein expression in the hippocampus of schizophrenia and bipo- retrieved in at least 1 other study, as could 1 of the proteins in- lar patients [31] and to mouse microarray data of prenatally volved in protein modifications, 3 proteins of the cellular response stressed PS mice (E), some of which were genetically manipulated to stress, 4 proteins involved in energy metabolism, 1 myelination to make them more vulnerable to PS (G) [24]. We also compared protein, and 1 calcium sensor. Proteins successfully validated in our findings to those of a recent proteomics study investigating the this study were highlighted in bold. ↑ = Upregulated; ↓ = down- rat hippocampus [37]. Furthermore, a recently published article regulated; SCZ = schizophrenia; BPD = bipolar disorder; E = en- evaluated proteomic studies in schizophrenia brain tissue, listing vironmental factor (here: PS); G × E = here: 5-Htt +/– × PS; ? = some of the proteins found in this study as potential biomarkers suggests proteins as biomarkers [29]. a p < 0.01; b As validated by for schizophrenia [29]. Eight of the 10 proteins involved in cytoar- Western blot.

INA, NEFL, and YWHAE), 2 stress response proteins tential biomarkers for schizophrenia in 2 recent reviews (HSPA8 and PARK7), and 2 proteins that are involved in of proteomic findings [31, 40] . energy metabolism (ATP5B and ENO2) (for further de- Gene Expression in the Brain Following PS. We also tails, see table 3 ). Some of these proteins, i.e. NEFL, compared our results to a recent microarray experiment DPYSL2, PRDX2, and PARK7, were also suggested as po- using a gene- environment interaction (G × E), i.e. a defi-

438 Dev Neurosci 2014;36:432–442 Föcking/Opstelten/Prickaerts/Steinbusch/ DOI: 10.1159/000365327 Dunn/van den Hove/Cotter Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM cient serotonin transporter (5-Htt) × PS paradigm, based interaction with ubiquitinated cargo proteins on the ear- on the same PS mouse model [25]. Five genes that were ly endosome. STAM proteins thus participate in the sort- affected by PS (E) at the mRNA level were also observed ing of cargo proteins for trafficking to the lysosome [51] . as protein changes in the same direction in our study Cytoskeletal Protein Changes. Our results showed ex- (TUBB2a, YHWAE, PSMA5, HSPA9, and MBP). Three pression changes in components of the cytoskeleton, such genes showed an opposite PS effect (E) on gene expres- as NEFL, tubulin subunits, and INA. Furthermore, we ob- sion compared to the protein expression changes found served downregulation of DPYSL2 and 3 and upregulation in this study [profilin1 (PFN1), SYN2, and IVD]. Of note, of PFN1, which may interfere with remodeling of the cyto- MBP was identified as being significantly altered in both skeleton during development, a requirement for axon guid- the E and the G × E condition, as well as at the protein ance [52, 53] . 14-3-3 protein epsilon, found to be down- level, in our study. A detailed summary is given in table 3 . regulated by PS, plays an essential role in neuronal migration by interacting with NUDEL and DISC1 [54, 55]. Of Particular Pathways and Processes Found to Be note, single-nucleotide polymorphisms in 14-3-3 proteins Affected after PS (YWHAE and YWHAG) are associated with severe neuro- Membrane Trafficking. Roughly 10% of the proteins af- logical diseases, and therefore alterations may involve non- fected by PS were proteins involved in membrane traffick- specific mechanisms of neuronal pathology. In keeping ing. SEPT5, which seems to be subjected to isoform-spe- with our findings, cytoskeletal changes in the brain in af- cific changes after PS, as verified by our 2-D Western blot, fective disorders and schizophrenia are well recognized in is an important element in vesicle trafficking, particularly many brain areas [32, 34, 35, 56, 57] , including the hippo- in clathrin-mediated endocytosis (CME) [36, 41, 42] . Hip- campus [58, 59]. Together, these findings point to an effect pocalcin (HPCA) was upregulated in the hippocampus of PS on neuronal migration and axon guidance. following PS, as confirmed by Western blot. HPCA is a Metabolic Changes and Stress Response. Energy me- neuronal Ca 2+ sensor thought to influence cognitive pro- tabolism was implicated by roughly 13% of the proteins cesses via the regulation of long-term potentiation and de- found to be differentially expressed following PS. Four pression. Long-term depression requires the internaliza- of these, i.e. IDH3A, IVD, ALDH1B1, and ATP5B, are tion of AMPA and NMDA receptors, both of which are mitochondrial proteins. ENO2, for which 2 isoforms dependent on CME. HPCA binds directly to the clathrin were altered by PS, is involved in glucose metabolism adaptor complex AP2 [43] and is required for the proper and such changes following PS are in keeping with the formation of the complex, after which CME can be initi- impaired glucose metabolism in schizophrenia [60, 61] . ated [44, 45] . Interestingly, the female PS mice that were Furthermore, we found proteins involved in the stress analyzed in this study showed significant memory impair- response, i.e. HSPA9, HSPA8, and HSP90B1, as well as ment during behavioral testing [23] . These results were the redox homeostasis proteins PARK7, PDIA3, and comparable to those of a study in HPCA-deficient mice PRDX2, to be differentially expressed following PS. that displayed a decline in spatial and associative memory These changes likely reflect sensitivity to mitochondrial [46]. Dysregulation of HPCA could thus contribute to the stress following PS, findings which are in keeping with cognitive deficits associated with affective disorders by the alterations observed in affective disorder and schizo- disturbing CME and thereby synaptic plasticity. phrenia [31, 62] . PFN1 was upregulated following PS, which could be Calcium-Sensing Proteins. Calretinin has a function confirmed by Western blot. It belongs to a family of small similar to that of HPCA and was also found to be signifi- proteins that act as actin polymerization promoters in cantly upregulated in the PS group of this study. In schizo- vivo [47, 48] and form complexes with regulators of en- phrenia, the focus on the role of calcium-binding proteins docytosis, synaptic vesicle recycling, and actin assembly. like calretinin or calbindin [63] has shifted towards a dys- Profilins colocalize with dynamin-1 and synapsin in axo- function in cell signaling in the GABAergic interneurons nal and dendritic processes [49] and have roles shaping of the hippocampus. This is understood to impair the the synaptic structure [50] . synchronized firing of neurons in the hippocampal-pre- The signal-transducing adaptor molecule (STAM) ex- frontal cortex circuit by destabilizing GABAergic inhibi- pression was reduced following PS, which we observed tory signaling [64]. Our findings emphasize the crucial both using 2D-DIGE and using Western blot. STAM1 is role of Ca 2+-binding proteins such as HPCA and cal- suggested to function as an adaptor molecule involved in retinin in the maintenance of cellular homeostasis and the downstream signaling of cytokine receptors through demonstrate how this process is an outcome of PS.

Proteomic Investigation of the Dev Neurosci 2014;36:432–442 439 Hippocampus DOI: 10.1159/000365327 Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM Myelin Changes. MBP expression was upregulated fol- Concluding Remarks lowing PS at both the protein level (the current study) and the mRNA level [25] . These findings are not consistent with In summary, this investigation is the first proteomic the myelin gene expression reductions typically observed in study of the hippocampus in PS mice and provides evi- schizophrenia and, to a lesser extent, in affective disorder dence that the mouse model reproduces hippocampal [65–67] and may be speculated to represent a compensatory deficits relevant to the pathology of schizophrenia and reaction to myelin dysfunction such as that observed in ma- affective disorders. Our results implicate that PS affects jor psychiatric disorders [68, 69]. Interestingly, MBP was cytoskeletal and synaptic function, energy metabolism, also shown to be differentially methylated in a hippocampal myelination, and membrane trafficking. These changes DNA methylation screening of PS mice (in preparation). are consistent with the changes observed in schizophrenia and affective disorders [31, 32] and so our findings Methodological Considerations strengthen previous evidence that PS in the mouse is a A primary limitation of this study is exposure of the valid animal model for affective disorders and schizo- mice to extensive behavioral testing prior to their sacrifice. phrenia [25] . Furthermore, our work contributes to the Although the behavioral task exposure was identical for all focus on new proteins such as calretinin, HPCA, SEPT5, animals, this repeated stress exposure may have influenced PFN1, STAM1, and 14-3-3 (YWHAE) and processes such our results, possibly masking some of the effects that were as calcium buffering that may be a part of the pathophys- induced purely by exposing the mice to PS and thus per- iology of affective disorders. The results of this study en- manently imprinting on hippocampal protein expression hance our knowledge of the nature and extent of protein patterns in a condition-dependent manner. Secondly, the changes following PS and provide valuable information lack of an overlap between our study and the study pub- about the molecular mechanisms that are involved in the lished by Mairesse et al. [39] might be explained by the fact severe illnesses that manifest themselves in human PS off- that the PS paradigm in rats shows a distinct sex specific- spring. ity, i.e. males are more affected by PS exposure whereas in mice females are more affected [23]. Thirdly, our 2D- DIGE results are rather small in term of fold change. How- Acknowledgements ever, spot volume normalization, within a gel and across We thank Caitriona Scaife, BSc, for technical assistance with gels (using the internal standard), adds to the sensitivity of the 2D-DIGE image analysis. Access to and use of the mass spec- the fluorescent dyes and their wide linear dynamic range. trometry instrumentation of the Conway Institute is gratefully ac- Of the 5 proteins that we attempted to validate using West- knowledged; Giuliano Elia, PhD, Cathy Rooney, BSc, and Kasper ern blotting, we successfully confirmed 4. The lack of con- Pedersen, BSc, provided technical assistance with mass spectrom- firmation of SEPT5 may be due to isoform-specific chang- etry. This work was supported by the Irish Health Research Board, es to which the antibody used in our Western blot experi- Science Foundation Ireland, The Brain and Behavior Research ments was insensitive. However, we cannot entirely Foundation (formerly NARSAD), and the Stanley Medical Re- exclude that the SEPT5 finding might be a false positive. search Institute.

References

1 Bale TL, Baram TZ, Brown AS, Goldstein JM, 4 Markham JA, Koenig JI: Prenatal stress: role lowing prenatal exposure to severe life events: Insel TR, McCarthy MM, Nemeroff CB, in psychotic and depressive diseases. Psycho- a Danish population-based cohort study. J

Reyes TM, Simerly RB, Susser ES, Nestler EJ: pharmacology (Berl) 2011; 214: 89–106. Psychiatr Res 2011; 45: 879–885. Early life programming and neurodevelop- 5 Beydoun H, Saftlas AF: Physical and mental 8 Khashan AS, Abel KM, McNamee R, Peders-

mental disorders. Biol Psychiatry 2010; 68: health outcomes of prenatal maternal stress in en MG, Webb RT, Baker PN, Kenny LC, 314–319. human and animal studies: a review of recent Mortensen PB: Higher risk of offspring

2 Weinstock M: The potential influence of ma- evidence. Paediatr Perinat Epidemiol 2008; schizophrenia following antenatal maternal

ternal stress hormones on development and 22: 438–466. exposure to severe adverse life events. Arch

mental health of the offspring. Brain Behav 6 Watson JB, Mednick SA, Huttunen M, Wang Gen Psychiatry 2008; 65: 146–152.

Immun 2005; 19: 296–308. X: Prenatal teratogens and the development 9 Malaspina D, Corcoran C, Kleinhaus KR, 3 O’Donnell K, O’Connor TG, Glover V: of adult mental illness. Dev Psychopathol Perrin MC, Fennig S, Nahon D, Friedlander

P renatal stress and neurodevelopment of 1999; 11: 457–466. Y, Harlap S: Acute maternal stress in preg- the child: focus on the HPA axis and role of 7 Khashan AS, McNamee R, Henriksen TB, nancy and schizophrenia in offspring: a co-

the placenta. Dev Neurosci 2009; 31: 285– Pedersen MG, Kenny LC, Abel KM, hort prospective study. BMC Psychiatry

292. Mortensen PB: Risk of affective disorders fol- 2008; 8: 71.

440 Dev Neurosci 2014;36:432–442 Föcking/Opstelten/Prickaerts/Steinbusch/ DOI: 10.1159/000365327 Dunn/van den Hove/Cotter Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM 10 van Os J, Selten JP: Prenatal exposure to ma- 22 Weinstock M: The long-term behavioural solateral prefrontal cortex in schizophrenia ternal stress and subsequent schizophrenia: consequences of prenatal stress. Neurosci and bipolar disorder reveals alterations in

the May 1940 invasion of the Netherlands. Br Biobehav Rev 2008; 32: 1073–1086. LAMP, STXBP1 and BASP1 protein expres-

J Psychiatry 1998; 172: 324–326. 23 Behan AT, van den Hove DL, Mueller L, sion. Mol Psychiatry 2009; 14: 601–613. 11 Selten JP, van der Graaf Y, van Duursen R, Jetten MJ, Steinbusch HW, Cotter DR, Prick- 35 English JA, Dicker P, Focking M, Dunn MJ, Gispen-de Wied CC, Kahn RS: Psychotic ill- aerts J: Evidence of female-specific glial defi- Cotter DR: 2-D DIGE analysis implicates cy- ness after prenatal exposure to the 1953 Dutch cits in the hippocampus in a mouse model of toskeletal abnormalities in psychiatric dis-

flood disaster. Schizophr Res 1999; 35: 243– prenatal stress. Eur Neuropsychopharmacol ease. Proteomics 2009; 9: 3368–3382.

245. 2011; 21: 71–79. 36 Pennington K, Beasley CL, Dicker P, Fagan A, 12 Burton CL, Chatterjee D, Chatterjee- 24 Weinstock M: Gender differences in the ef- English J, Pariante CM, Wait R, Dunn MJ, Chakraborty M, Lovic V, Grella SL, Steiner M, fects of prenatal stress on brain development Cotter DR: Prominent synaptic and metabol-

Fleming AS: Prenatal restraint stress and and behaviour. Neurochem Res 2007; 32: ic abnormalities revealed by proteomic analy- motherless rearing disrupts expression of 1730–1740. sis of the dorsolateral prefrontal cortex in plasticity markers and stress-induced corti- 25 Van den Hove D, Jakob SB, Schraut KG, Ke- schizophrenia and bipolar disorder. Mol Psy-

costerone release in adult female Sprague- nis G, Schmitt AG, Kneitz S, Scholz CJ, Wie- chiatry 2008; 13: 1102–1117.

Dawley rats. Brain Res 2007; 1158: 28–38. scholleck V, Ortega G, Prickaerts J, Stein- 37 Focking M, Boersema PJ, O’Donoghue N, Lu- 13 Jutapakdeegul N, Afadlal S, Polaboon N, busch H, Lesch KP: Differential effects of pre- bec G, Pennington SR, Cotter DR, Dunn MJ: Phansuwan-Pujito P, Govitrapong P: Repeat- natal stress in 5-Htt deficient mice: towards 2-D DIGE as a quantitative tool for investigat- ed restraint stress and corticosterone injec- molecular mechanisms of gene × environ- ing the HUPO Brain Proteome Project mouse

tions during late pregnancy alter GAP-43 ex- ment interactions. PLoS One 2011; 6:e22715. series. Proteomics 2006; 6: 4914–4931. pression in the hippocampus and prefrontal 26 Campbell S, Macqueen G: The role of the hip- 38 Gharahdaghi F, Weinberg CR, Meagher DA,

cortex of rat pups. Int J Dev Neurosci 2010; 28: pocampus in the pathophysiology of major Imai BS, Mische SM: Mass spectrometric

83–90. depression. J Psychiatry Neurosci 2004; 29: identification of proteins from silver-stained 14 Afadlal S, Polaboon N, Surakul P, Govit- 417–426. polyacrylamide gel: a method for the removal rapong P, Jutapakdeegul N: Prenatal stress al- 27 Harrison PJ: The hippocampus in schizo- of silver ions to enhance sensitivity. Electro-

ters presynaptic marker proteins in the hip- phrenia: a review of the neuropathological phoresis 1999; 20: 601–605.

pocampus of rat pups. Neurosci Lett 2010; evidence and its pathophysiological implica- 39 Mairesse J, Vercoutter-Edouart AS, Marrocco

470: 24–27. tions. Psychopharmacology (Berl) 2004; 174: J, Zuena AR, Giovine A, Nicoletti F, Michal- 15 Zuena AR, Mairesse J, Casolini P, Cinque C, 151–162. ski JC, Maccari S, Morley-Fletcher S: Pro- Alema GS, Morley-Fletcher S, Chiodi V, Sp- 28 Vythilingam M, Heim C, Newport J, Miller teomic characterization in the hippocampus

agnoli LG, Gradini R, Catalani A, Nicoletti F, AH, Anderson E, Bronen R, Brummer M, of prenatally stressed rats. J Proteomics 2012;

Maccari S: Prenatal restraint stress generates Staib L, Vermetten E, Charney DS, Nemeroff 75: 1764–1770. two distinct behavioral and neurochemical CB, Bremner JD: Childhood trauma associ- 40 Martins-De-Souza D, Dias-Neto E, Schmitt profiles in male and female rats. PLoS One ated with smaller hippocampal volume in A, Falkai P, Gormanns P, Maccarrone G,

2008; 3:e2170. women with major depression. Am J Psychia- Turck C, Gattaz W: Proteome analysis of

16 Maccari S, Morley-Fletcher S: Effects of pre- try 2002; 159: 2072–2080. schizophrenia brain tissue. World J Biol Psy-

natal restraint stress on the hypothalamus- 29 Heim C, Binder EB: Current research trends chiatry 2010; 11: 110–120. pituitary-adrenal axis and related behavioural in early life stress and depression: review of 41 Peterson EA, Petty EM: Conquering the com- and neurobiological alterations. Psychoneu- human studies on sensitive periods, gene-en- plex world of human septins: implications for

roendocrinology 2007; 32(suppl 1):S10–S15. vironment interactions, and epigenetics. Exp health and disease. Clin Genet 2010; 77: 511–

17 Van den Hove DL, Blanco CE, Aendekerk B, Neurol 2012; 233: 102–111. 524. Desbonnet L, Bruschettini M, Steinbusch HP, 30 Mychasiuk R, Gibb R, Kolb B: Prenatal stress 42 Hall PA, Russell SE: The pathobiology of the

Prickaerts J, Steinbusch HW: Prenatal reproduces sexually dimorphic and regionally septin gene family. J Pathol 2004; 204: 489– straint stress and long-term affective conse- specific changes in gene expression in hippo- 505.

quences. Dev Neurosci 2005; 27: 313–320. campus and frontal cortex of developing rat 43 McMahon HT, Mills IG: Cop and clathrin-

18 Barros VG, Rodriguez P, Martijena ID, Perez offspring. Dev Neurosci 2011; 33: 531–538. coated vesicle budding: different pathways, A, Molina VA, Antonelli MC: Prenatal stress 31 English JA, Pennington K, Dunn MJ, Cotter common approaches. Curr Opin Cell Biol

and early adoption effects on benzodiazepine DR: The neuroproteomics of schizophrenia. 2004; 16: 379–391.

receptors and anxiogenic behavior in the Biol Psychiatry 2011; 69: 163–172. 44 Palmer CL, Lim W, Hastie PG, Toward M,

adult rat brain. Synapse 2006; 60: 609–618. 32 Martins-de-Souza D, Gattaz WF, Schmitt A, Korolchuk VI, Burbidge SA, Banting G, 19 Morley-Fletcher S, Mairesse J, Soumier A, Maccarrone G, Hunyadi-Gulyas E, Eberlin Collingridge GL, Isaac JT, Henley JM: Hip- Banasr M, Fagioli F, Gabriel C, Mocaer E, MN, Souza GH, Marangoni S, Novello JC, pocalcin functions as a calcium sensor in hip-

Daszuta A, McEwen B, Nicoletti F, Maccari S: Turck CW, Dias-Neto E: Proteomic analysis pocampal LTD. Neuron 2005; 47: 487–494. Chronic agomelatine treatment corrects be- of dorsolateral prefrontal cortex indicates the 45 Jo J, Son GH, Winters BL, Kim MJ, Whitcomb havioral, cellular, and biochemical abnormal- involvement of cytoskeleton, oligodendro- DJ, Dickinson BA, Lee YB, Futai K, Amici M, ities induced by prenatal stress in rats. Psy- cyte, energy metabolism and new potential Sheng M, Collingridge GL, Cho K: Muscarin-

chopharmacology (Berl) 2011; 217: 301–313. markers in schizophrenia. J Psychiatr Res ic receptors induce LTD of NMDAR EPSCs

20 Morley-Fletcher S, Darnaudery M, Koehl M, 2009; 43: 978–986. via a mechanism involving hippocalcin, AP2

Casolini P, Van Reeth O, Maccari S: Prenatal 33 Föcking M, Dicker P, English J, Schubert K, and PSD-95. Nat Neurosci 2010; 13: 1216– stress in rats predicts immobility behavior in Dunn M, Cotter D: Common proteomic 1224. the forced swim test: effects of a chronic treat- changes in the hippocampus in schizophrenia 46 Kobayashi M, Masaki T, Hori K, Masuo Y,

ment with tianeptine. Brain Res 2003; 989: and bipolar disorder and particular evidence Miyamoto M, Tsubokawa H, Noguchi H, No- 246–251. for involvement of cornu ammonis regions 2 mura M, Takamatsu K: Hippocalcin-deficient

21 Bethus I, Lemaire V, Lhomme M, Goodall G: and 3. Arch Gen Psychiatry 2011; 68: 477–488. mice display a defect in camp response ele- Does prenatal stress affect latent inhibition? It 34 Behan AT, Byrne C, Dunn MJ, Cagney G, ment-binding protein activation associated

depends on the gender. Behav Brain Res 2005; Cotter DR: Proteomic analysis of membrane with impaired spatial and associative memo-

158: 331–338. microdomain-associated proteins in the dor- ry. Neuroscience 2005; 133: 471–484.

Proteomic Investigation of the Dev Neurosci 2014;36:432–442 441 Hippocampus DOI: 10.1159/000365327 Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM 47 Pantaloni D, Carlier MF: How profilin pro- 55 Taya S, Shinoda T, Tsuboi D, Asaki J, Nagai 63 Cotter D, Landau S, Beasley C, Stevenson R, motes actin filament assembly in the presence K, Hikita T, Kuroda S, Kuroda K, Shimizu M, Chana G, MacMillan L, Everall I: The density

of thymosin beta 4. Cell 1993; 75: 1007–1014. Hirotsune S, Iwamatsu A, Kaibuchi K: DISC1 and spatial distribution of GABAergic neu- 48 Qualmann B, Kessels MM: Endocytosis and regulates the transport of the NUDEL/ rons, labelled using calcium binding proteins,

the cytoskeleton. Int Rev Cytol 2002; 220: 93– LIS1/14-3-3 epsilon complex through kine- in the anterior cingulate cortex in major de-

144. sin-1. J Neurosci 2007; 27: 15–26. pressive disorder, bipolar disorder, and schizo-

49 Witke W, Podtelejnikov AV, Di Nardo A, 56 Beasley CL, Pennington K, Behan A, Wait R, phrenia. Biol Psychiatry 2002; 51: 377–386. Sutherland JD, Gurniak CB, Dotti C, Mann Dunn MJ, Cotter D: Proteomic analysis of the 64 Eyles DW, McGrath JJ, Reynolds GP: Neuro- M: In mouse brain profilin I and profilin II anterior cingulate cortex in the major psychi- nal calcium-binding proteins and schizophre-

associate with regulators of the endocytic atric disorders: evidence for disease-associat- nia. Schizophr Res 2002; 57: 27–34.

pathway and actin assembly. EMBO J 1998; ed changes. Proteomics 2006; 6: 3414–3425. 65 Martins-de-Souza D, Maccarrone G, Wo-

17: 967–976. 57 Clark D, Dedova I, Cordwell S, Matsumoto I: brock T, Zerr I, Gormanns P, Reckow S, 50 Neuhoff H, Sassoe-Pognetto M, Panzanelli P, A proteome analysis of the anterior cingulate Falkai P, Schmitt A, Turck CW: Proteome Maas C, Witke W, Kneussel M: The actin- cortex gray matter in schizophrenia. Mol Psy- analysis of the thalamus and cerebrospinal

binding protein profilin I is localized at syn- chiatry 2006; 11: 459–470, 423. fluid reveals glycolysis dysfunction and po- aptic sites in an activity-regulated manner. 58 Zaidel D, Esiri M, Harrison P: The hippocam- tential biomarkers candidates for schizophre-

Eur J Neurosci 2005; 21: 15–25. pus in schizophrenia: lateralized increase in nia. J Psychiatr Res 2010; 44: 1176–1189. 51 Mizuno E, Kawahata K, Kato M, Kitamura N, neuronal density and altered cytoarchitectur- 66 Matthews PR, Eastwood SL, Harrison PJ: Re-

Komada M: STAM proteins bind ubiquitinat- al asymmetry. Psychol Med 1997; 27: 703–713. duced myelin basic protein and actin-related ed proteins on the early endosome via the 59 Nakatani N, Ohnishi T, Iwamoto K, Wata- gene expression in visual cortex in schizo-

VHS domain and ubiquitin-interacting mo- nabe A, Iwayama Y, Yamashita S, Ishitsuka Y, phrenia. PLoS One 2012; 7:e38211.

tif. Mol Biol Cell 2003; 14: 3675–3689. Moriyama K, Nakajima M, Tatebayashi Y, 67 Davis KL, Stewart DG, Friedman JI, Buchs- 52 Kitamura K, Takayama M, Hamajima N, Na- Akiyama H, Higuchi T, Kato T, Yoshikawa T: baum M, Harvey PD, Hof PR, Buxbaum J, Ha- kanishi M, Sasaki M, Endo Y, Takemoto T, Expression analysis of actin-related genes as routunian V: White matter changes in schizo- Kimura H, Iwaki M, Nonaka M: Character- an underlying mechanism for mood disor- phrenia: evidence for myelin-related dysfunc-

ization of the human dihydropyrimidinase- ders. Biochem Biophys Res Commun 2007; tion. Arch Gen Psychiatry 2003; 60: 443–456.

related protein 2 (DRP-2) gene. DNA Res 352: 780–786. 68 Hof PR, Haroutunian V, Friedrich VL Jr,

1999; 6: 291–297. 60 Ryan MC, Collins P, Thakore JH: Impaired Byne W, Buitron C, Perl DP, Davis KL: Loss 53 Krishnan K, Moens P: Structure and function fasting glucose tolerance in first-episode, and altered spatial distribution of oligoden-

of profilins. Biophys Rev 2009; 1: 71–81. drug-naive patients with schizophrenia. Am J drocytes in the superior frontal gyrus in

54 Toyo-oka K, Shionoya A, Gambello MJ, Car- Psychiatry 2003; 160: 284–289. schizophrenia. Biol Psychiatry 2003; 53: 1075– doso C, Leventer R, Ward HL, Ayala R, Tsai 61 van Nimwegen LJ, Storosum JG, Blumer RM, 1085. LH, Dobyns W, Ledbetter D, Hirotsune S, Allick G, Venema HW, de Haan L, Becker H, 69 Roussos P, Katsel P, Davis KL, Bitsios P, Giak- Wynshaw-Boris A: 14-3-3 epsilon is impor- van Amelsvoort T, Ackermans MT, Fliers E, oumaki SG, Jogia J, Rozsnyai K, Collier D, Fr- tant for neuronal migration by binding to Serlie MJ, Sauerwein HP: Hepatic insulin re- angou S, Siever LJ, Haroutunian V: Molecular NUDEL: a molecular explanation for Miller- sistance in antipsychotic naive schizophrenic and genetic evidence for abnormalities in the

Dieker syndrome. Nat Genet 2003; 34: 274– patients: stable isotope studies of glucose me- nodes of Ranvier in schizophrenia. Arch Gen

285. tabolism. J Clin Endocrinol Metab 2008; 93: Psychiatry 2012; 69: 7–15. 572–577. 62 Ryan MM, Lockstone HE, Huffaker SJ, Way- land MT, Webster MJ, Bahn S: Gene expression analysis of bipolar disorder reveals downregulation of the ubiquitin cycle and alterations in synaptic genes. Mol Psychiatry

2006; 11: 965–978.

442 Dev Neurosci 2014;36:432–442 Föcking/Opstelten/Prickaerts/Steinbusch/ DOI: 10.1159/000365327 Dunn/van den Hove/Cotter Downloaded by: Univ. of California San Diego 198.143.33.65 - 8/31/2015 3:41:02 PM