Expression of Brain-Derived NeurotrophicFactor mRNA in Rat Hippocampus AfterTreatment With Antipsychotic Drugs

Ou Bai, Jennifer Chlan-Fourney, Rudy Bowen, David Keegan, and Xin-Min Li*Neuropsychiatry Research Unit, Department of Psychiatry, University of Saskatchewan, Saskatoon,Saskatchewan, Canada

Typical and atypical antipsychotic drugs, though botheffective, act on different neurotransmitter receptors andare dissimilar in some clinical effects and side effects.The typical antipsychotic drug haloperidol has beenshown to cause a decrease in the expression of brain-derived neurotrophic factor (BDNF), which plays an im-portant role in neuronal cell survival, differentiation, andneuronal connectivity. However, it is still unknownwhether atypical antipsychotic drugs similarly regulateBDNF expression. We examined the effects of chronic(28 days) administration of typical and atypical antipsy-chotic drugs on BDNF mRNA expression in the rat hip-pocampus using in situ hybridization. Quantitative anal-ysis revealed that the typical antipsychotic drughaloperidol (1 mg/kg) down-regulated BDNF mRNA ex-pression in both CA1 (P � 0.05) and dentate gyrus (P �0.01) regions compared with vehicle control. In contrast,the atypical antipsychotic agents clozapine (10 mg/kg)and olanzapine (2.7 mg/kg) up-regulated BDNF mRNAexpression in CA1, CA3, and dentate gyrus regions of therat hippocampus compared with their respective con-trols (P � 0.01). These findings demonstrate that thetypical and atypical antipsychotic drugs differentially reg-ulate BDNF mRNA expression in rat hippocampus.© 2002 Wiley-Liss, Inc.

Key words: haloperidol; clozapine; olanzapine; BDNF;hippocampus

The antipsychotic drugs, which are effective in treat-ing several psychiatric illnesses, such as schizophrenia, aredivided into two classes. Atypical antipsychotic drugs, suchas clozapine and olanzapine, block both dopamine andserotonin receptors, improve both the positive and thenegative symptoms of schizophrenia, and have a lowerpropensity to induce extrapyramidal side effects. Typicalantipsychotic drugs, such as haloperidol, block dopaminereceptors and improve positive symptoms of schizophreniabut have a high propensity to cause motor side effects(Meltzer, 1995). Clinical evidence demonstrates that earlyand prolonged intervention with these drugs will improvethe long-term therapeutic outcome (Wyatt and Henter,1998) and that the clinical improvements are not observeduntil 2–3 weeks following antipsychotic drug administra-

tion (Kinon and Liebermann, 1996). These findings indi-cate that the clinical effects are not solely due to theblockade of neurotransmitters. Recent studies showed thatthe typical antipsychotic agent haloperidol increased nervegrowth factor (NGF) expression in the hippocampus, piri-form cortex, striatum, and nucleus accumbens of mice(Ozaki, 2000); however, other studies demonstrated thathaloperidol significantly decreased BDNF protein expres-sion in the frontal cortex, occipital cortex, and hippocam-pus of rats (Angelucci et al., 2000; Lipska et al., 2001).

Brain-derived neurotrophic factor (BDNF) is widelyexpressed in the CNS; its highest levels are found in thehippocampus and cortex (Hofer et al., 1990), whereBDNF supports neuronal cell survival, differentiation,connectivity, and morphology (Ghosh et al., 1994;Thoenen, 1995), particularly of dopaminergic neurons(Hyman et al., 1991). Functional and anatomical evidenceimplicates hippocampal involvement in schizophrenia(Falkai and Bogerts, 1986; Weinberger, 1999) and BDNFin hippocampal neuron function. Because haloperidoldown-regulates BDNF expression, it is possible thatBDNF is also involved in the action of atypical antipsy-chotic drugs. The aim of this study was to measure BDNFmRNA expression in the hippocampus of rats that hadbeen treated with the atypical antipsychotic drugs cloza-pine or olanzapine and the typical antipsychotic drughaloperidol using in situ hybridization.

MATERIALS AND METHODS

Animals

Male Wistar rats (200–350 g; Charles River Canada,Montreal, Quebec, Canada) were individually housed under a12 hr light-dark cycle at a temperature of 19–20°C, with free

Contract grant sponsor: Health Services Utilization and Research Com-mission; Contract grant sponsor: Canadian Psychiatric Research Founda-tion and the Schizophrenia Society of Saskatchewan.

*Correspondence to: Dr. Xin-Min Li, Neuropsychiatry Research Unit,Department of Psychiatry, University of Saskatchewan, A114 MedicalResearch Building 103 Wiggins Road, Saskatoon, Saskatchewan S7N 5E4,Canada. E-mail: xin-min.li@usask.ca

Received 30 April 2002; Revised 22 July 2002; Accepted 23 July 2002

Journal of Neuroscience Research 71:127–131 (2003)

© 2002 Wiley-Liss, Inc.

access to both food and water. All procedures involving animalswere performed in accordance with the Canadian Council onAnimal Care Guidelines and approved by the University ofSaskatchewan’s Committee on Animal Care and Supply. After1 week of acclimatization, rats were randomly divided into fourgroups (n � 3) and were given daily drug injections (i.p.) for 28days. Group 1 received vehicle control (0.4% acetic acid in 0.9%saline); group 2 received haloperidol (1 mg/kg); group 3 re-ceived clozapine (10 mg/kg); group 4 received olanzapine(2.7 mg/kg). The rats were sacrificed 18 hr after the last injec-tion. The doses for haloperidol and clozapine were chosen onthe basis of 5-HT2 and D2 ex vivo receptor occupancies in therat brain as determined by Schotte et al. (1993), and the dose forolanzapine was calculated based on a study by Beasley et al.(1997). The treatments did not induce changes in weight.

In Situ Hybridization

Brains to be used for in situ hybridization were removedand immediately frozen in isopentane at –45°C and stored at –70°C. Twelve micrometer coronal sections through the hip-pocampus were thaw mounted on slides doubly coated withpoly-L-lysine and stored at –20°C. The prehybridization wascarried out within 72 hr. Mounted sections were fixed with 4%paraformaldehyde, acetylated in acetic anhydride, dehydratedthrough an increasing ethanol gradient, delipidated in chloro-form for 10 min, rehydrated to 95% ethanol, air dried, andstored at –20°C until the actual hybridization was performed.

BDNF sense and antisense RNA probes were labeled with35S-CTP, and 2 � 106 c.p.m. were applied to each slide. Slideswere then incubated at 54°C overnight. On the following day,slides were rinsed in 2� SSC buffer (sodium chloride andsodium citrate) four times for 5 min each and treated with10 mg/ml RNase for 30 min, followed by a rinse in RNasebuffer for 30 min at room temperature. Subsequently, the sec-tions were washed in 2� SSC at room temperature for 4 min,2� SSC at 50°C for 30 min, and 0.2� SSC at 55°C for 30 min.Finally, sections were dried in ascending ethanol concentrationsand allowed to air dry for 1 hr before exposure to photographicemulsion (Li et al., 1998).

Photographic Emulsion

Kodak NTB2 photographic emulsion was diluted 1:1with nanopure water and warmed to 42°C. The slides weredipped for 3 sec and left to dry for at least 2 hr and were thenplaced in dark boxes with dessicant at 4°C for 2–6 weeks. Theslides were developed for 2 min in Kodak D-19 developer, fixedfor 30 sec in Kodak fixer, rinsed in water, and air dried. Thesections were lightly counterstained with cresyl violet, dehy-drated, cleared with xylene, and mounted for analysis.

Quantitative Analysis of BDNF mRNA

The emulsion-dipped slides were quantitatively analyzedwith image-analysis software (Image-Pro Plus version 4.0). Thearea covered with silver grains was measured using an Olympus(BH-2) microscope with darkfield illumination, and cell numberwas counted with lightfield illumination. The background, anarea of negative expression, was subtracted. Silver grain den-sity � area of (positive expression – negative expression)/cellnumber. Three fields from each section and three sections from

each animal were counted. Average percentage area values wereexpressed as percentage of vehicle control values.

Data Analysis

The data are presented as mean � SD. The groups werecompared using one-way ANOVA, followed by the post hocScheffe’s test. Significance is reported as follows: *P � 0.05,**P � 0.01.

RESULTSBrain sections from the four treatment groups hy-

bridized with 35S-labeled BDNF RNA probes are shownin Figure 1. Low-magnification (2.5� lightfield) photo-graphs (top row) show strong expression of BDNFmRNA on hippocampal sections from rats treated withclozapine or olanzapine. High-magnification (40� dark-field) photographs (three lower rows) show BDNFmRNA expression with silver grain density in the CA1,CA3, and dentate gyrus regions of the hippocampus.Lower silver grain densities are seen in the CA1, CA3, anddentate gyrus regions of hippocampus in rats treated withhaloperidol and the highest density on sections from clo-zapine or olanzapine groups. The decreased expression ofBDNF mRNA is indicated by a lower density of silvergrains, and the increased expression of BDNF is shown bya higher density.

Quantitative analysis of BDNF mRNA expressionwas performed on three regions of the hippocampus.Analysis by one-way ANOVA showed significant effectson BDNF expression in CA1 [F(3,8) � 67.7], CA3[F(3,8) � 33.7], and dentate gyrus regions [F(3,8) � 98.6]after drug treatments. Post hoc analysis revealed signifi-cance in all three treatment groups (Fig. 2). Haloperidoldecreased BDNF mRNA expression in both CA1 (P �0.05) and dentate gyrus (P � 0.01) regions compared withvehicle control. In contrast, clozapine and olanzapine sig-nificantly increased gene expression in all three regions ofthe hippocampus compared with control (P � 0.01).

DISCUSSIONThe present study demonstrates that chronic admin-

istration of the typical antipsychotic drug haloperidol re-sults in a reduction of BDNF mRNA expression in theCA1 and dentate gyrus regions of rat hippocampus,whereas administration of clozapine or olanzapine resultsin significant up-regulation in CA1, CA3, and dentategyrus regions. There are at least three possible explanationsfor the discrepant effects of atypical and typical antipsy-chotics on the expression of BDNF mRNA.

First, it has been reported that exogenous adminis-tration of the indirect dopamine agonist levodopa in-creases the expression of BDNF mRNA, an action that ispartly blocked by coadministration of haloperidol (Oka-zawa et al., 1992). This links dopamine receptor stimula-tion to BDNF synthesis and binding of the receptors withhaloperidol to down-regulation of BDNF expression. Theatypical antipsychotic drugs clozapine and olanzapine havethe propensity to dissociate rapidly from D2 receptors and

128 Bai et al.

to allow normal dopamine neurotransmission (Kapur andRemington, 2001).

Second, studies have demonstrated that the5-HT2A/2C receptor agonist DOI (4-iodo-2,5-dimethoxy-phenylisopropylamine) significantly decreases BDNFmRNA expression in the hippocampus, and this effect isblocked by pretreatment with the selective 5-HT2A re-ceptor antagonist ketanserin (Vaidya et al., 1997). Theatypical antipsychotic agents clozapine and olanzapine arealso 5-HT2A receptor antagonists (Kapur and Remington,2001), which may result in up-regulation of BDNF ex-pression.

Third, the regulation of BDNF expression and re-lease from hippocampal neurons is increased by non-N-methyl-D-aspartate (NMDA)-type glutamate receptor ac-tivation (Zafra et al., 1990; Wetmore et al., 1994). It has

been reported that haloperidol increases NMDAR1 pro-tein and mRNA levels in the striatum after chronic treat-ment. Clozapine does not significantly affect striatalNMDAR1 levels but induces an increase in GluR2 pro-tein levels in the frontal parietal cortex, nucleus accum-bens, and hippocampus (Fitzgerald et al., 1995) andmRNA expression in rat cortex and striatum (Healy andMeador-Woodruff, 1997). Studies have demonstrated thatthe presence of GluR2 subunits in functional non-NMDAreceptors, such as AMPA receptors, increases BDNFmRNA expression and decreases calcium influx (Hollmanet al., 1991; Burnashev et al., 1992), whereas NMDA,which effectively enhances calcium influx, does not affectBDNF mRNA levels (Zafra et al., 1990). It is possible thatclozapine increases BDNF mRNA levels subsequent toup-regulation of the subunit of non-NMDA glutamate

Fig. 1. Animals were given daily injections of vehicle, haloperidol,clozapine, or olanzapine for 28 days and were sacrificed 18 hr after thelast injection. Brain sections were hybridized with 35S-labeled BDNFRNA probes and dipped into photographic emulsion. Low-magnification (2.5�) photomicrographs showed BDNF mRNA ex-

pression in the hippocampus (top row). High-magnification (40�)photomicrographs show detailed changes silver grain density in theCA1, CA3, and dentate gyrus (DG) regions compared with the control;many cells show positive expression of BDNF mRNA after clozapineor olanzapine treatments.

Expression of BDNF After Antipsychotics 129

receptors. Because olanzapine is similar in structure andpharmacology to clozapine (Fulton and Gao, 1997), thisdrug may also have the same effect as clozapine on BDNFregulation.

MK-801, a glutamatergic antagonist, can causepsychosis-like effects in humans (Javitt and Znkin, 1991)and reduce the basal levels of BDNF mRNA in in vitro aswell as in vivo models of the rat hippocampus (Castren etal., 1993). Furthermore, BDNF mRNA was attenuated bypretreatment with the atypical antipsychotic drug cloza-pine (Linden et al., 2000). It is also reported that there isa significant decrease of BDNF in the hippocampus ofschizophrenic patients (Durany et al., 2001) and thatBDNF expression in the hippocampus of schizophrenicpatients is increased following antipsychotic drug treat-ment (Takahashi et al., 2000). Our data demonstrate thatchronic administration of the atypical antipsychotic drugsclozapine and olazapine up-regulates BDNF mRNA lev-els, and it has been reported that the typical antipsychoticdrug haloperidol increases NGF expression (Ozaki, 2000).Taken together, the up-regulation of neurotrophins in-duced by antipsychotics may account, in part, for thelong-term therapeutic effects of these drugs, and neuro-trophins may be involved in the action of antipsychotics.

The present finding is consistent with reports on theeffects of the typical antipsychotic drug haloperidol onBDNF expression (Angelucci et al., 2000; Lipska et al.,2001). Down-regulation of BDNF could affect neuronalsurvival and neuronal connectivity and, therefore, may be

related to its extrapyramidal side effects, which are due todopamine D2 receptor blockade (Coffin et al., 1989).

In summary, our study demonstrates that the anti-psychotic drugs clozapine and olanzapine increase, buthaloperidol decreases, BDNF mRNA expression in rathippocampus. These findings indicate that the differentregulation of BDNF expression might represent a possiblemechanism for the typical and atypical clinical profile.

ACKNOWLEDGMENTSWe thank Dr. Sergey Fedoroff and Dr. Darrell

Mousseau for their critical and helpful comments andGabriel Stegeman for her excellent technical assistance.This study was supported by Health Services Utilizationand Research Commission Postdoctoral Fellowship7-01807 to O.B.

REFERENCESAngelucci F, Mathe AA, Aloe L. 2000. Brain-derived neurotrophic factor

and tyrosine kinase receptor Trk B in rat brain are significantly alteredafter haloperidol and risperidone administration. J Neurosci Res 60:783–794.

Beasley CM, Tollefson GD, Tran PV. 1997. Effects of olanzapine: anoverview of pivotal clinical trials. J Clin Psychiatr 58:7–12.

Burnashev N, Monyer H, Seeburg PH, Sakmann B. 1992. Divalent ionpermeability of AMPA receptor channels is dominated by the edited formof a single subunit. Neuron 8:189–198.

Castren E, Berzaghi MP, Lindholm D, Thoenen H. 1993. Differentialeffects of MK-801 on brain-derived neurotrophic factor mRNA levels indifferent regions of rat brain. Exp Neurol 122:244–252.

Coffin VL, Latranyi MB, Chipkin RE 1989. Acute extrapyramidal syn-drome in cebus monkeys: development mediated by dopamine D2 butnot D1 receptors. J Pharmacol Exp Ther 249:769–774.

Durany N, Michel T, Zochling R, Boissl KW, Cruz-Sanchez FF, RiedererP, Thome J. 2001. Brain-derived neurotrophic factor and neurotrophin-3in schizophrenic psychoses. Schizophr Res 52:79–86.

Falkai P, Bogerts B. 1986. Cell loss in the hippocampus of schizophrenia.Eur Arch Psychiatr Neurol Sci 236:154–161.

Fitzgerald LW, Deutch AY, Gasic G, Heinemann SF, Nestler EJ. 1995.Regulation of cortical and subcortical glutamate receptor subunit expres-sion by antipsychotic drugs. J Neurosci 15:2453–2461.

Fulton B, Goa KL. 1997. Olanzapine. a review of its pharmacologicalproperties and therapeutic efficacy in the management of schizophreniaand related psychoses. Drugs 53:281–298.

Ghosh A, Carnahan J, Greenberg ME. 1994. Requirement for BDNF inactivity-dependent survival of cortical neurons. Science 263:1618–1623.

Healy DJ, Meador-Woodruff JH. 1997. Clozapine and haloperidol differ-entially affect AMPA and kainate receptor subunit mRNA levels in ratcortex and striatum. Brain Res Mol Brain Res 47:331–338.

Hofer M, Pagliusi SR, Hohn A, Leibrock J, Barde YA. 1990. Regionaldistribution of brain-derived neurotrophic factor mRNA in the adultmouse brain. EMBO J 9:2459–2464.

Hollmann M, Hartley M, Heinemann S. 1991. C2� permeability of KA-AMPA-gated glutamate receptor channels depends on subunit composi-tion. Science 252:851–853.

Hyman C, Hofer M, Barde YA, Juhasz M, Yancopoulos GD, Squinto SP,Lindsay RM. 1991. BDNF is a neurotrophic factor for dopaminergicneurons of the substantia nigra. Nature 350:230–232.

Javitt DC, Znkin SK. 1991. Recent advances in the phencyclidine model ofschizophrenia. Am J Psychiatr 148:1301–1308.

Kapur S, Remington G. 2001. Atypical antipsychotics: new directions andnew challenges in the treatment of schizophrenia. Annu Rev Med 52:503–517.

Fig. 2. Animals were given various doses of drugs daily for 28 days andthen were sacrificed 18 hr after the last injection. Quantitative analysisof BDNF mRNA expression was performed on three regions ofhippocampus. The area covered with silver grains was measured in thedarkfield, and cell number was counted in the lightfield. Silver graindensity � area of (positive expression – background)/cell number.Three fields from each section and three sections from each animalwere counted. Average percentage area values are presented as percent-age of vehicle control values. Data represents mean � SD. Significancewas determined by using one-way ANOVA, followed by the post hocScheffe’s test. *P � 0.05, **P � 0.01.

130 Bai et al.

Kinon BJ, Lieberman JA. 1996. Mechanisms of action of atypical antipsy-chotic drugs: a critical analysis. Psychopharmacology 124:2–34.

Li XM, Juorio AV, Qi J, Boulton AA. 1998. L-deprenyl induces aromaticL-amino acid decarboxylase (AADC) mRNA in the rat substantia nigraland ventral tegmentum. Mol Chem Neuropathol 35:149–155.

Linden AM, Vaisanen J, Lakso M, Nawa H, Wong G, Castren E. 2000.Expression of neurotrophins BDNF and NT-3, and their receptors in ratbrain after administration of antipsychotic and psychotrophic agents. JMol Neurosci 14:27–37.

Lipska BK, Khaing ZZ, Weickert CS, Weinberger DR. 2001. BDNFmRNA expression in rat hippocampus and prefrontal cortex: effects ofneonatal ventral hippocampal damage and antipsychotic drugs. Eur J Neu-rosci 14:135–144.

Meltzer HY. 1995. Atypical antipsychotic drugs. In: Psychopharmacology:the fourth generation of progress. New York: Raven Press. p 1277–1286.

Okazawa H, Murata M, Watanabe M, Kamei M, Kanazawa I. 1992.Dopaminergic stimulation up-regulates the in vivo expression of brain-derived neurotrophic factor (BDNF) in the striatum. FEBS Lett 313:138–142.

Ozaki T. 2000. Comparative effects of dopamine D(1) and D(2) receptorantagonists on nerve growth factor protein induction. Eur J Pharmacol402:39–44.

Schotte A, Janssen PF, Megens AA, Leysen JE. 1993. Occupancy of centralneurotransmitter receptors by risperidone, clozapine and haloperidol,

measured ex vivo by quantitative autoradiography. Brain Res 631:191–202.

Takahashi M, Shirakawa O, Toyooka K, Kitamura N, Hashimoto T,Maeda K, Koizumi S, Wakabayashi K, Takahashi H, Someya T, Nawa H.2000. Abnormal expression of brain-derived neurotrophic factor and itsreceptor in the corticolimbic system of schizophrenic patients. Mol Psy-chiatr 5:293–300.

Thoenen H. 1995. Neurotrophins and neuronal plasticity. Science 270:593–598.

Vaidya VA, Marek GJ, Aghajanian GK, Duman RS. 1997. 5-HT2A

receptor-mediated regulation of brain-derived neurotrophic factormRNA in the hippocampus and the neocortex. J Neurosci 17:2785–2795.

Weinberger DR. 1999. Cell biology of the hippocampal formation inschizophrenia. Biol Psychiatr 45:395–402.

Wetmore C, Olson L, Bean AJ. 1994. Regulation of brain-derived neu-rotrophic factor (BDNF) expression and release from hippocampal neu-rons is mediated by non-NMDA type glutamate receptors. J Neurosci14:1688–1700.

Wyatt RT, Henter ID. 1998. The effects of early and sustained interventionon the long-term morbidity of schizophrenia. J Psychiatr Res 32:169–177.

Zafra F, Hengerer B, Leibrock J, Thoenen H, Lindholm D. 1990. Activitydependent regulation of BDNF and NGF mRNA in the rat hippocampusis mediated by non-NMDA glutamate receptors. EMBO J 9:3545–3550.

Expression of BDNF After Antipsychotics 131