cDNA gene expression profile of rat hippocampus after ... - Springer Link

DOI 10.1007/s00702-003-0077-8 J Neural Transm (2003) 110: 1413–1436

cDNA gene expression profile of rat hippocampus after chronic treatment with antidepressant drugs N. Drigues1 , T. Poltyrev2, C. Bejar2, M. Weinstock2 , and M. B. H. Youdim1 1 Eve Topf and National Parkinson Foundation Centers of Excellence for Neurodegenerative Diseases Research and Department of Pharmacology, Technion – Faculty of Medicine, Haifa, and 2 Department of Pharmacology, Hebrew University Faculty of Medicine, Jerusalem, Israel

Accepted September 24, 2003

Summary. Background. Chronic antidepressant treatment causes alterations in several hippocampal genes, which participate in neuronal plasticity. However the full picture of their mechanism of action is not known. The advent of genomics enables to identify a broader mechanism of action and identify novel targets for antidepressant development. Methods. The present study examined the cDNA microarray gene expression profile in the hippocampus induced by chronic antidepressant treatment, in rats exposed to the forced swim test. Animals were treated for 2 weeks with moclobemide, clorgyline and amitriptyline. Results. The three antidepressants significantly reduced immobility in the forced swim test and initiated significant homologous changes in gene expressions. These include up regulation of cAMP response element binding protein and down regulation of corticotrophin releasing hormone. Other gene changes noted were those related to neuropeptides, neurogenesis and synaptogenesis, including synaptophysin and neogenin. Some 89 genes were changed by at least 2 drugs, out of which 53 were changed oppositely by forced swim test. Confirmation of gene changes, have come from real time RT-PCR. Conclusions. A significant number and homology in gene expression were observed with the three antidepressants. Many of the genes were associated with neurogenesis and synaptogenesis, including synaptophysin and neogenin. Keywords: Depressive disorder, antidepressant, gene expression profiling, oligonucleotide array sequence analysis, hippocampus, rat. Introduction The current concept on the mechanism of action of antidepressant drugs (ADDs) results from two events, an immediate pharmacological action and a

1414

N. Drigues et al.

long term one on post-receptor signalling pathways. The latter includes changes in neural gene expression (Duman et al., 1997; Hyman and Nestler, 1996; Post, 1997). Chronic antidepressant treatments, ADDs and electro-convulsive therapy (ECT), increase the levels of cAMP response element binding protein (CREB) mRNA, protein, and function in the hippocampus (Butterweck et al., 2001; Dowlatshahi et al., 1998; Duman and Vaidya, 1998; Jeon et al., 1997; Nibuya et al., 1996). The time course for these changes is consistent with that for the therapeutic action of antidepressant treatments. Brain derived neurotrophic factor (BDNF), a known CREB target, is increased following chronic antidepressant treatment as well as ECT (Nibuya et al., 1995, 1996). A variety of endocrine changes, particularly thyroid and adrenal disturbances, have also been identified in patients with mood disorders. Studies consistently describe corticotrophin-releasing hormone (CRH) and cortisol hyper secretion during major depression episodes (Gold et al., 1995; Wong et al., 2000; Young et al., 1994). This has been attributed to an inadequate feedback of glucocorticoids at multiple levels of the hypothalamic-pituitary-adrenal axis (Holsboer, 2000; Nemeroff et al., 1984). There is evidence of reduced neurotrophic element in the pathophysiology of depression. Stress and corticosterone administration cause a dramatic reduction in BDNF expression in hippocampus of rats (Nitta et al., 1997; Smith et al., 1995). Thus, the largest group of animal models involves exposure to a stressful event. Models of acute stress events include the learned helplessness and the forced swimming tests (FST). Recent studies have shown that decreased neurogenesis occurs in response to both acute and chronic stress (Gould et al., 2000). Chronic administration of ADDs or electro-convulsive shock (ECS) in rats, increase proliferation and survival of new neurons in the hippocampus (Jacobs et al., 2000; Madsen et al., 2000; Malberg et al., 2000). Molecular and biological experiments have demonstrated that ADDs have a significant affect on the cAMP cascade and in all probability a homologous pathway exists for all classes of antidepressants. However, sporadic observations, using regular biochemical and molecular assays, can only suggest limited possible changes. Sequencing of the human genome has fostered an unprecedented development of new high-throughput technologies that can simultaneously and globally examine expression of thousands of genes at once. cDNA microarray technique has successfully been applied to cancer pathology and therapy (DeRisi et al., 1996; Hedenfalk et al., 2001). Also, it has been used in brain research, including brain tumours, demyelination diseases and schizophrenia (Mirnics et al., 2000; Sallinen et al., 2000; Whitney et al., 1999). More recently we have successfully employed cDNA microarray gene expression to determine what genes initiate neurodegeneration in animal models of Parkinson’s disease, how these relate to gene expression in Parkisonian brains and how neuroprotective drugs induce neuroprotection by altering cell survival= death genes (Mandel et al., 2003). This study determines gene expression changes elicited by FST model of depressive illness and chronic treatment with different classes of ADDs in rat hippocampus. We show that ADDs change the expression of genes, many relate to making of new neurons and new synaptic connections.

cDNA gene expression and antidepressants

1415

Methods and materials Animal model Animals. All procedures are in accordance with the NIH Guide for the Care and Use of Laboratory Animals and are approved by the Technion Animal Ethics Committee, Haifa, Israel. Experiments were conducted in male Sprague Dawley rats (Harlan Laboratories Limited, Israel) weighing 200–250 g. These rats were housed under standard conditions and fed ad libitum. Forced Swimming Test. Forced swimming test was conducted according to Porsolt et al. (Porsolt et al., 1978), with slight modifications (Weinstock et al., 2002). On treatment day 13, each rat was placed for 15 min in a transparent cylinder (40 cm high, 18 cm in diameter) containing water 15 cm deep maintained at 25 C. After an initially frenzied attempt to escape, animals assumed an immobile posture. The duration of immobility was timed during the last 5 min of a test. On sacrifice day, rats received a 5 min swim test. In the 5 min test, the duration of immobility was recorded. Responsiveness of rats was assessed 60 min after drug administration. Animal treatment. Rats were randomly assigned to treatment groups and received by oral gavage doses of vehicle (water), moclobemide (20 mg=kg=day), amitriptyline (10 mg=kg=day) and clorgyline (0.5 mg=kg=day) (drugs were kindly provided by Teva Pharmaceutical industries, Ltd.) for 14 days. Following treatment, animals were sacrificed by decapitation, and hippocampi were collected into RNALater (Ambion, Inc.) for RNA isolation.

Total RNA isolation Total RNA was isolated using TriReagent (Sigma-Aldrich, Inc.) according to the manufacturer’s recommendations. In order to assess yield and purity, total RNA O.D. was measured at 260 nm and 280 nm. RNA integrity was checked by electrophoresis in a 1.2% agarose=formaldehyde gel. Testing for DNA contamination was done by PCR with primers for the well-characterized gene, b-actin (primer sequences are presented in Table 1).

mRNA expression pattern using cDNA microarray To enable reproduction and verification of this array based gene expression monitoring experiment, the methodology used is reported according to the guidelines described in MIAME (minimum information about a microarray experiment) (Brazma et al., 2001). The MIAME report is presents in Appendix 1.

Real-time RT-PCR Experiment was repeated with 6 or 8 animals in each group. Total RNA was pooled from RNA of 2 animals and was reversed transcribed to cDNA using MMLV reverse transcriptase enzyme (Promega Corporation, Inc.). The cDNA was amplified by DNA polymerase enzyme using specific primers (Invitrogen Corporation, Inc.) as shown in Table 1. PCR reaction conditions were optimized for each mRNA evaluated, these include cDNA, MgCl and primers concentrations and annealing temperature. Real time PCR using SYBR Green I was performed in duplicates for each RNA sample using the LightCycler (Roche Diagnostics Corporation, Inc.) instrument. We utilized the ready-to use FastStart DNA Master SYBR Green I PCR mix kits (Roche Diagnostics Corporation, Inc.) according to the manufacturer’s protocol. Serial dilutions of cDNA were used as a standard curve for the relative quantitation of mRNA in each sample. A melting curve was constructed at the end of the assay to evaluate the specificity of the reaction.

Statistics Statistic techniques include one-way analysis of variance (ANOVA), followed by the Tukey test, two-ways student t-test and Pearson correlation that were performed, using the Analyse-it Software (Analyse-it Software, Ltd.). P values of less than 0.05 were considered significant.

Name

Beta-actin 18S CREB Neogenin Synaptophysin CRH

GeneBank #

NM_031144 X00640 U38938 U68726 X06655 M54987

ACCAACTGGGACGATATGGAGAAGA GTAACCCGTTGAACCCCATT GCCCACCAGCTACAAAGTGT AGAATGCAAACGCAACA TGGTATCCTACCGCATTC ACTGATGGAGATTATCGGG

Forward

TACGACCAGAGGCATACAGGGACAA CCATCCAATCGGTAGTAGCG ACTTGTGAGCAGCACATTGG TCGTTCTCGGTTTTCCT ACTCACCTCATAGCTCC CTGGGTGACTTCCATCTG

Reverse

Table 1. Primer sequences used in real-time PCR

– 152 594 201 379 351

Amplicon (bp)

1416 N. Drigues et al.


1417

Results Forced swimming test Chronic treatment with amitriptyline, moclobemide and clorgyline for two weeks with dosage indicated in the Methods, significantly decreased duration of immobility in the rat FST (Fig. 1) similar to our recent publication on this subject (Weinstock et al., 2002). No significant difference was noted between the three drugs. mRNA expression pattern RNA was isolated from each rat treated with an antidepressant and was pooled for each antidepressant treated group. RNA was reversed transcribed to construct cDNA radioactive probes that were subsequently hybridized to Clontech’s Rat 1.2 cDNA microarray. Genes were regarded as expressed if their adjusted intensity (intensity above background) was more than 5 standard deviations of the background noise. Significant changes were considered when the ratio of expression was more than two standard deviations of the ratio distribution (between 1.6 and 2.1, dependent on the variations between arrays). When a gene was significantly expressed in one group and significantly not expressed in the other it was considered as changed. The number of genes changed by each treatment and the calculated percentages are presented in Table 2. Moclobemide treatment changed 124 genes, clorgyline changed 123 and amitriptyline changed 128 genes. The percentage of genes changed by each antidepressant drug demonstrates the homology between different ADDs (see Table 2). Each drug changed the expression of about 125 genes out of 1185 on the cDNA array, equal to roughly 10.5% of total genes on the array. However, 32% of the genes changed by moclobemide were also changed by clorgyline; 45% of the genes changed by moclobemide were also changed by amitriptyline;

Fig. 1. Effect of chronic antidepressant drug treatment on time (sec) spent in immobility in the forced swimming test. ANOVA followed by the Tukey test: : p< 0.05 : p < 0.01

1185 124 123 128 40 57 53 32

All array moclob clorg amit moclob þ clorg moclob þ amit clorg þ amit moclob þ clorg þ amit 10.5%

moclob 10.4%

clorg 10.8%

amit

Moclob moclobemide; Clorg clorgyline; Amit amitriptyline

All array

No. of genes 3.4% 32.3% 32.5%

moclob þ clorg

44.5%

4.8% 46.0%

moclob þ amit

43.1% 41.4%

4.5%

clorg þ amit

2.7% 25.8% 26.0% 25.0% 80.0% 56.1% 60.4%

moclob þ clorg þ amit

Table 2. Number of genes changed in each array and the percentages calculated each time from a different set of genes on the left column

1418 N. Drigues et al.


1419

Table 3. A partial list of genes that where changed by at least 2 of the 3 antidepressant drugs GenBank #

Adj. Intens

Difference OR Log2(ratio)

Control(þs)

Control(s)

Protein=gene

Moclob

Clorg

Amit

Neurotransmitters L31771 5388 U20907 5682 M61099 1482

633 543 564

25387 163 3782

24341 24230 73

23976 24509 2477

alpha-1D adrenergic R 5-HT R 4 mGluR 1

Neuropeptides Z11504 5448 M54987 8585 U67863 6203

1054 1291 1634

1675 21.15 1766

23695 20.87 23912

22471 27079 24981

NPY R type 1 CRH Melanocortin R 4

cAMP cascade M55075 2931 L26986 1 U38938 2602

1508 3098 5354

1589 5801 5293

805 4592 807

1699 6286 8133

AC type III AC type VIII CREB

Kinase- serine=threonin L24907 1704

2545

3830

4353

3535

CamK I

Cell cycle regulation X70871 6432 L11007 4950 D83792 5030 L24388 5045 D10863 5618 D25233 1689 M61219 1984

24050 610 1603 25044 0.95 3102 895

25450 1177 23661 22983 1.39 7246 590

25925 22382 24388 24233 1478 1343 2075

26091 21968 22157 1667 1.33 5304 2752

G2=M-specific cyclin G CDK 4 p27Kip1 GTA protein kinase Id-2; DNA-binding protein inhibitor Retinoblastoma Prohibitin

DNA replication X98490 4926 Y00047 2182

22918 197

23393 4497

23086 5503

24751 6197

Replication protein A PCNA

Synapse formation and proteins X06655 2645 1804 X52772 2488 969 U26402 2226 116 Y13381 2819 2491 Y13380 2712 2458 U68726 2515 1169

5532 4704 2127 1273 5138 3164

2542 138 556 166 4875 754

6659 2916 3107 1529 3223 3822

Metabolism- carbohydrates M17701 6241

3579

Other M20636 U97143

3268 5238

4898 231

1.56 7062 4309

1.05 3505 5668

1.42 5720 230

Synaptophysin, p38 Synaptotagamin I Synaptotagmin V Amphiphysin Amphiphysin II Neogenin GAPDH PLC beta 1 GDNF R alpha 2

Control(þs), animals treated with water and subjected to the forced swimming test; Control(s): animals treated with water but not subjected to the forced swimming test; Moclob moclobemide; Clorg clorgyline; Amit amitriptyline. Up regulated genes are in a bold font; Down regulated genes are in a smaller, bold and italic font. Adj. Intensity adjusted intensity (spot intensity minus background intensity); Difference adjusted intensity in treated array minus its adjusted intensity in the control(þs) array; Log2(ratio) (grey background), the log2 of the ratio between the adjusted intensity in treated array and its adjusted intensity in the control(þs) array

1420

N. Drigues et al.

and 42% of the genes changed by clorgyline were also changed by amitriptyline (Table 2). Moreover, from 40 genes that were changed by moclobemide and clorgyline 78% were also changed by amitriptyline. Consistent with our hypothesis we decided to concentrate only on genes differentially expressed by at least 2 drugs. Out of 1185 genes on the cDNA membrane 58 genes were up regulated and 31 genes were down regulated by at least 2 of the 3 ADDs used. 18 genes were up- and 14 were down regulated by the 3 drugs. A partial list of the genes that were changed by the antidepressant treatments is shown in Table 3. For the full list of changed genes see Appendix 2. Gene changes included acetylcholine receptors, adrenergic receptors, glutamate and GABA transmission, neuropeptide receptors, calcium cascade, cyclic AMP cascade, potassium channels, small G proteins, tyrosine kinase cascade, cell cycle regulators, synaptogenesis and synaptic proteins. There were also significant other gene changes. Of the genes altered by the ADDs the FST model of depression changed 53 in the opposite direction (Appendix 2) strengthening the belief that stress has a major role in the pathogenesis of depressive illness. The results with the gene expressions were corroborated in a second experiment. Groups of 6 or 8 animals were treated with the three ADDs for two weeks. RNA was extracted and was pooled only from two animals, resulting in 3 or 4 replicates in each group. The expression of four genes was analyzed using real-time RT-PCR. Expression of CREB, CRH, synaptophysin and neogenin was measured and normalized to the expression of the housekeeping gene ribosomal 18S RNA. The three ADDs up regulated CREB significantly (Fig. 2).

Fig. 2. Gene expression changes of cAMP response element binding protein, neogenin and synaptophysin by chronic treatment with antidepressant drugs. Control(þs): animals treated Control(s): animals treated with with water and subjected to the forced swimming test; water but not subjected to the forced swimming test; Moclobemide; Clorgyline; Amitriptyline. ANOVA followed by the Tukey test: : p < 0.05 : p< 0.01 : p < 0.001


1421

Fig. 3. Correlation between the expression of synaptophysin and of neogenin in control animals and after different antidepressant drugs treatments. Pearson correlation: r ¼ 0.95, p< 0.0001

Neogenin and synaptophysin expressions were increased by the ADDs (Fig. 2) and direct correlation between them was observed (Fig. 3). By contrast, moclobemide, clorgyline and amitriptyline significantly decreased CRH expression (Fig. 4).

Fig. 4. Gene expression changes of corticotrophin releasing hormone by chronic treatment Control(þs): animals treated with water and subjected to the with antidepressant drugs. Control(s): animals treated with water but not subjected to the forced swimming test; forced swimming test; Moclobemide; Clorgyline; Amitriptyline. ANOVA followed by the Tukey test: : p< 0.05 : p < 0.01

1422

N. Drigues et al.

Discussion The mechanisms by which antidepressants modalities, monoamine oxidase inhibitors, amine uptake inhibitors and ECT produce their effect is not fully known or established. Several weeks of treatment are required until their action and beneficial effect is observed. Thus, it has been inferred on several occasions that this period is associated with both presynaptic and postsynaptic adaptive changes (Hyman and Nestler, 1996). The most prominent features of ADDs’ action are up regulation of CREB and down regulation of CRH (Butterweck et al., 2001; Dowlatshahi et al., 1998; Gold et al., 1995; Nibuya et al., 1996), which in depressed patients have been reported to be altered in an opposite direction respectively. These changes are complimented by increase in BDNF expression and neurogenesis (Madsen et al., 2000; Malberg et al., 2000; Nibuya et al., 1995). Thus, it is apparent that during prolong ADDs treatment there may be a need for serial cascades of biochemical events before their antidepressant activity be expressed. The advent of genomics, namely cDNA microarray gene expression and proteomics, may contribute to elucidate this complex cascade. In this paper we describe for the first time cDNA microarray gene expression for three antidepressants and FST model of depressive illness. Previous studies used whole brain tissue to obtain a cDNA gene expression profiles (Rausch et al., 2002) or were unable to confirm CREB and CRH changes (Landgrebe et al., 2002). A recent study utilized cDNA microarray to study desipramine effect on hippocampal cells in culture (Chen et al., 2003) and desipramine and tranylcypromine were shown to increase GAP-43 expression. The present use of different classes of ADDs provides a dissecting tool, so far not examined by others, that enabled us to reduce the ‘‘noise’’ in the gene array data as described in the Method section. Although using only a single hybridization per condition, we used several ADDs and focused on genes that were changed by at least 2 drugs in order to minimize the false-positive results. By taking genes that were changed by the three drugs we would have increased our false-negative results. For example CREB was found to be increased only after moclobemide and amitriptyline treatment but not after clorgyline treatment (Table 3). However, our PCR results show that CREB is up regulated by the three ADDs (Fig. 2). Our mRNA expression results also show significant homology in gene expressions between different ADDs, which is greater when it is calculated from a narrower set of genes (as denoted in Table 2). This strengthens the belief that our results are not a consequence of random changes, but relate to drug’s therapeutic action. The results of the present study do not represent all the changes that may occur after ADDs treatment in rats’ hippocampus, since, the microarray we used consisted of 1,200 genes, which is 4% of the rat genome. However, it gives us a picture of changes occurring following drug treatment, and enables us to explore new molecular cascades that might take place in the therapeutic action of ADDs. Out of 89 found to be expressed differentially after ADDs treatments we have so far confirmed 4 and other genes are presently being analyzed. Nevertheless, our microarray methodology confirmed the two most prominent features of ADDs action on protein expression, up regulation of CREB and down regulation of CRH (Dowlatshahi et al., 1998; Nibuya et al., 1996; Stout et al., 2002; Thome et al., 2000). Thus, our microarray results plus


1423

the up regulation of synaptophysin and neogenin as shown by RT-PCR, ascertain the veracity of the gene changes found. Neurotransmitters It was found that moclobemide and amitriptyline increase the expression of glutamate receptor subunit 1 (GluR1). Previous report shows that chronic treatment with desipramine and paroxetine increase GluR1 protein levels in rats’ hippocampus (Martinez-Turrillas et al., 2002). Neuropeptides Higher CRH levels are found in the cerebro-spinal fluid of depressed patients as compared to healthy subjects (Galard et al., 2002; Gold et al., 1995). CRH mRNA in the paraventricular nucleus is reduced following antidepressant administration (Gold et al., 1995; Stout et al., 2002), as confirmed in this study. NPY reduces immobility in the FST while pre-treatment with NPY-1 antagonists significantly block the anti-immobility effects. These results suggest that NPY may have antidepressant-like activity in the FST, which is mediated by NPY-1 receptor subtype (Redrobe et al., 2002). Consistent with these results, NPY-1 was reduced following chronic treatment with ADDs (Table 3). It is possible that both NPY and ADDs have a common pathway of action, which, when activated inhibits NPY-1 expression by a negative feedback mechanism (Fig. 5). MCL0129, a selective melanocortin 4 receptor antagonist, produces an

Fig. 5. Signaling pathways altered after treatment with antidepressant drugs: (a) cAMP cascade; (b) calcium mediated cascade; (c) cAMP response element binding protein dependent genes. Genes in red were up regulated and genes in blue were down regulated. Red arrows indicate activation, blue arrows indicate inhibition and green arrows indicate increase in expression. NPY1R neuropeptide Y 1 receptor; AC adenylate cyclase; PKA protein kinase A; PLC phospholipase C; CamK calcium=calmodulin dependent kinase; CREB cAMP response element binding protein; PCNA proliferating cell nuclear antigen; BDNF brain derived neurotrophic factor; NOS nitric oxide synthase

1424

N. Drigues et al.

antidepressant-like effect both in the FST and the learned helplessness test (Chaki et al., 2003). Our results indicate that melanocrtin-4 receptor expression is reduced following chronic ADDs treatment in rats’ hippocampus. cAMP cascade Adenylate cyclase (AC) activity is reported to be lower in persons with major depression (Abou-Saleh et al., 1994; Menninger and Tabakoff, 1997; Newman et al., 1992; Wang et al., 1974). We found that AC III and VIII expression is up regulated following chronic ADDs treatment. The present ADDs induced up regulation of CREB mRNA in rat hippocampus, confirming previous reports (Gold et al., 1995; Nibuya et al., 1996). Dowlatshahi et al. have found lower CREB concentrations in hippocampi of depressed patients not on ADDs and higher CREB in the temporal cortex of depressed patients treated with ADDs than in untreated patients (Dowlatshahi et al., 1998). Several CREB-dependent genes were also up regulated, probably secondary to CREB up regulation (Fig. 5). Among these proteins are proliferating cell nuclear antigen (PCNA), neuronal nitric oxide synthase, retinoblastoma and huntingtin (Mayr and Montminy, 2001). Calcium cascade Calcium=calmodulin-dependant kinase (CamK) I expression was increased following ADDs treatment, supporting previous reports for involvement of CamK in the action of ADDs (Popoli et al., 2001) (Fig. 5). The finding that ADDs reduces phospholipase C b activity and mRNA in rats’ hippocampus (Dwivedi et al., 2002) contrasts with our finding that its expression is increased (Table 3). Proliferation and differentiation related genes A substantial role for cell-cycle control and neurogenesis in the mechanism of action of ADDs has been identified (Table 3). ADDs increase cell proliferation and neurogenesis in adult brain, while stress reduces them (Duman et al., 1999, 2001). Thus, chronic treatment with different classes of ADDs increase labeling of dividing cells in adult rat hippocampus (Lee et al., 2001; Malberg et al., 2000; Manev et al., 2001). Furthermore neurogenesis in dentate gyrus of adult rats is increased in response to ECS (Madsen et al., 2000). Our results indicate that ADDs regulate the cell cycle mechanism of neural progenitors. By inhibiting cell cycle progression and proliferation progenitors may differentiate to newly formed neurons. Progression through the cell cycle is initiated by enzymes composed of catalytic (cyclin-dependent kinase, CDK) and regulatory (cyclin) subunits. CDK4 and Cyclin G, two proteins important for cell cycle progression (Ekholm and Reed, 2000), were down regulated by ADDs treatment. Galactosyltransferase associated (GTA) protein kinase, a cdc2-related protein kinase, favours cell cycle progression (Kerr et al., 1994). GTA kinase expression was decreased following ADDs treatment, strengthening drugs’ role in inducing differentiation. Retinoblastoma and prohibitin, which act as tumour suppressor genes by decreasing proliferation (Wang et al., 1999), were up regulated.


1425

Some gene changes observed are associated with increase in cell proliferation capacity. Thus p27Kip1, a CDK inhibitor present in contact-inhibited cells and during neural differentiation (Ekholm and Reed, 2000), was found to be down regulated following ADDs treatment. The involvement of p27Kip1 in neural commitment is suggested by experiments showing that its over-expression in cells leads to neural cell cycle growth arrest, while its depletion causes increased proliferative capacity (Doetsch et al., 2002; Legrier et al., 2001). Id-2, which is increased following ADDs treatment, inhibits transcription of differentiation related genes (Hara et al., 1994; Iavarone et al., 1994) and is increased in several tumours (Langlands et al., 2000; Maruyama et al., 1999). Two genes associated with DNA replication were also changed (Table 3). Replication protein A (RPA) is part of the cell DNA replication mechanism (Wold, 1997), while PCNA is a co-factor for DNA polymerase (Travali et al., 1989) and is known as an index for cell proliferation (Belvindrah et al., 2002). RPA was down regulated, emphasizing ADDs role in differentiation while PCNA was increased, pointing for a proliferative role. Synaptogenesis and plasticity One of the most prominent set of genes differentially expressed after treatment with ADDs relates to synapse formation and axonal outgrowth. Synaptotagmin I, synaptotagmin V and synaptophysin, all synaptic proteins (Mochida, 2000) were up regulated following ADDs treatment. Synaptophysin up regulation was confirmed using real-time PCR. There is no current knowledge on the effect of ADDs on synaptic protein transcription; however, stress exposure has been shown to reduce the expression of synaptophysin in the hippocampus (Thome et al., 2001). Moreover, 5-HT reuptake blockers cause an increase in phosphorylation of synaptotagmin (Popoli et al., 1997). Amphiphysin is located at the nerve terminal and is important for endocytosis of synaptic vesicles (Wigge and McMahon, 1998) and for neurite outgrowth in cultured hippocampal cells (Mundigl et al., 1998). Amphiphysin I and II were up regulated following chronic treatment with ADDs. Neogenin, a receptor for the netrin family of proteins, is known to mediate attraction of growing axons (Livesey, 1999). Volenec et al. have shown up regulation of synaptophysin and deleted in colorectal cancer (DCC) mRNA, a different netrin receptor, following chronic treatment with selective serotonin reuptake inhibitors (Volenec et al., 2002). The PCR confirmation used a small number of replicates, 3 or 4. Thus, the variation between samples prevented the changes being significant. However a trend toward up regulation can be seen in the three ADDs. When studying expression of genes using real-time PCR a high correlation between expression of neogenin and synaptophysin was found (r ¼ 0.95, p