differential effects of antipsychotics on expression of antioxidant enzymes and membrane lipid...

Differential effects of antipsychotics on expression of antioxidantenzymes and membrane lipid peroxidation in rat brain

Vinay Parikha,b, Mohammad M. Khana,b, Sahebarao P. Mahadika,b,*aPsychiatry and Health Behavior, Medical College of Georgia, Augusta, Georgia, USA

bMedical Research Service (242), Veterans Affairs Medical Center, 5B-103, 1 Freedom Way, Augusta, GA-30904, USA

Received 4 April 2002; received in revised form 27 June 2002; accepted 1 July 2002

Abstract

Typical and atypical antipsychotics significantly differ in their neurotransmitter receptor affinity profiles, and their efficacy andside effects in schizophrenic patients. Typical antipsychotics have been found to increase the oxidative (i.e. free radical-mediated)

cellular injury in rats. Since schizophrenia also involves oxidative injury, the understanding of differential effects of these anti-psychotics on expression of antioxidant enzymes and oxidative injury may be very critical. The effect of chronic exposure of halo-peridol (HAL), a typical antipsychotic, was compared to effects of risperidone (RIS) or clozapine (CLZ) or olanzapine (OLZ),atypical antipsychotics on antioxidant defense enzymes and lipid peroxidation in the rat brain. The levels of antioxidant enzymes

and hydroxyalkenals (HAEs) were measured in rat brain cytosol and fatty acids were measured in brain cell membranes. ChronicHAL treatment for both 45 and 90 days significantly decreased manganese-superoxide dismutase (MnSOD), copper-zinc super-oxide dismutase (CuZnSOD) and catalase (CAT) activity with parallel marked increase in (HAEs), a marker of lipid peroxidation

in rat brain. The levels of enzymatic activity very well correlated with the levels of enzyme proteins indicating that the changes wereprobably in the expression of net protein. However, RIS, CLZ and OLZ treatments did not produce any alterations in the levels ofantioxidant enzymes and HAEs, both after 45 and 90 days. There were no alterations in the levels of saturated as well as poly-

unsaturated fatty acids in brain membranes. These findings indicate that chronic administration of HAL, but none of the studiedatypicals induce oxidative stress by persistent changes in the levels of antioxidant enzymes and cause membrane lipid peroxidation.# 2002 Elsevier Science Ltd. All rights reserved.

Keywords: Oxidative stress; Atypical antipsychotics; Antioxidant enzymes; Lipid peroxidation; Brain; Fatty acids

1. Introduction

Neuroleptics often referred to as antipsychotics havebeen used to treat schizophrenia for over 40 years.Haloperidol (HAL), a commonly prescribed typicalantipsychotic has been found to induce/exacerbate theextrapyramidal side effects and often leads to persis-tent tardive dyskinesia (TD), and it has also beenfound to have a limited efficacy in treating the corenegative symptoms and cognitive deficits in schizo-phrenia. Recently, use of atypical antipsychotics suchas clozapine (CLZ), risperidone (RISP), olanzapine(OLZ) etc. has demonstrated that it is possible to haveantipsychotic efficacy with minimum neurological

adverse effects and even improve the cognitive perfor-mance in schizophrenic patients (Sharma and Mockler,1998).The knowledge of biochemical mechanisms under-

lying the pathophysiology of side effects may providemeans to improve the use of these medications. One ofthe hypotheses has implicated the role of neuroleptictreatment associated increased reactive oxygen species(ROS) in the pathogenesis of extrapyramidal sideeffects, particularly TD (Cadet and Lohr, 1989; Lohr etal., 1988, 1990). Production of ROS is a ubiquitousevent during cellular aerobic metabolism. ROS cancause cellular damage such as peroxidation of mem-brane lipids, oxidation of proteins and damage to DNA,if not removed by antioxidant defenses. Under normalcircumstances, enzymatic antioxidant defense, super-oxide dismutase (SOD), catalase (CAT) and glutathioneperoxidase (GPx) can protect against oxidative injury.

0022-3956/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.

PI I : S0022-3956(02 )00048-1

Journal of Psychiatric Research 37 (2003) 43–51

www.elsevier.com/locate/jpsychires

* Corresponding author. Tel.: +1-706-733-0188x 2490; fax: +1-

706-823-3977.

E-mail address: [email protected] (S.P. Mahadik).

http://www.elsevier.com/locate/jpsychires/a4.3d

mailto:[email protected]

SOD dismutates superoxide radicals to form hydrogenperoxide, which inturn is decomposed to water andoxygen by CAT and GPx thereby preventing the for-mation of hydroxyl radicals.It has also been suggested that some antipsychotics,

primarily typical antipsychotics may have pro-oxidanteffects, i.e. increase the oxidative stress and oxidativecell injury, preferentially in the brain (Halliwell, 1992;Jeding et al., 1995; Mahadik and Mukherjee, 1996;Reddy and Yao, 1996). HAL administration to rats hasbeen reported to increase the brain lipid peroxidationand decrease the levels of reduced glutathione (Shiva-kumar and Ravindranath, 1993). Moreover, neurotoxiceffect of HAL has been suggested to be a consequenceof increased lipid peroxidation in cerebral cortex (Sawasand Gilbert, 1985). Treatment of animals with typicalneuroleptics such as HAL and fluphenazine has beenfound to increase oxidative stress by altering the levelsof antioxidant enzymes, and cause oxidative injury inthe brain (Murthy et al., 1989; Cadet and Perumal,1990). Chlorpromazine metabolites have also been sug-gested to generate H2O2 by autooxidation (Heikkila etal., 1975). The ability of neuroleptics to interact both in-vivo and in-vitro with iron, that plays a critical role inROS-mediated cell injury, is well established.There is substantial evidence that increased oxidative

stress and oxidative cell injury may also be a part of thepathogenesis of schizophrenia (Mahadik et al., 1999;Yao et al., 2001). This is based on several findings ofabnormal activities of critical antioxidant enzymes andother indices of lipid peroxidation in plasma, red bloodcells, and cerebrospinal fluid in chronic patients pri-marily treated with typical antipsychotics (reviewed byMahadik and Mukherjee, 1996; Reddy and Yao, 1996;Mahadik et al., 1999) or treated with CLZ (Liday et al.,1995), and in never medicated first-episode psychoticpatients (McCreadie et al., 1995; Mahadik et al., 1998).Such abnormalities have been also associated with TD(Cadet and Cohr, 1989; Peet et al., 1993) and negativesymptoms (Buckman et al., 1990). This has led to thesuggestion that oxidative cell injury may be a con-tributing factor in the etiopathophysiology of schizo-phrenia, which may be exacerbated by the treatmentwith antipsychotics with pro-oxidant properties (Maha-dik and Mukherjee, 1996), and atypicals may not inducethe oxidative stress and peroxidative cell injury.It is indicated that lipid peroxidation may be one of

the major mechanisms of the consistently observedreduction of EPUFAs in schizophrenia (Mahadik et al.,1999). There is a substantial evidence for altered mem-brane phospholipids (PL) primarily due to abnormalEPUFAs metabolism in schizophrenia (Horrobin et al.,1994; Horrobin, 1999; Yao et al., 1998; Yao, 1999).EPUFAs are critical for brain development as well asfunction (Simopoulos, 1991; Wainwright, 1992) that areaffected in schizophrenia.

Based on these findings, we have hypothesized thattypical antipsychotics such as HAL but not the atypicalantipsychotics such as RIS, CLZ or OLZ will induce theoxidative stress and lead to oxidative cell injury thatmay induce/exacerbate the side effects associated withchronic treatment. Since HAL is the most frequentlyused typical antipsychotic and most of the clinical stud-ies have used this drug to compare the clinical effectswith atypical antipsychotics, we used it as prototypeagent for typical antipsychotic in our study. We investi-gated systematically this hypothesis by studying anti-oxidant enzymes (indices of oxidative stress) activitywith their net protein content and membrane lipid per-oxidation (index of oxidative injury) in rat brain afterchronic treatment with antipsychotics. The findingssupport the hypothesis, and may have implications forthe improved means of treatment and management ofschizophrenia.

2. Materials and methods

2.1. Materials

HAL was purchased from Sigma (St. Louis, MO,USA). Janssen Pharmaceutica (Trenton, NJ) providedRIS and CLZ and Eli Lilly (Indianapolis, IN) providedthe OLZ. Rabbit anti-rat manganese-superoxide dis-mutase (MnSOD) IgG fraction was purchased fromResearch Diagnostics Inc. (Flanders, NJ, USA). Rabbitanti-rat copper-zinc superoxide dismutase (CuZnSOD),anti-rat CAT polyclonal antibody, peroxidase con-jugated anti-CuZnSOD and anti-CAT IgG fraction andtetramethylbenzidine (TMB) were from Rockland Inc.(Gilbertsville, PA, USA). All other chemicals were fromSigma (St. Louis, MO, USA).

2.2. Animals and drug treatment

Male albino Wistar rats (225–250 g) were obtainedfrom Harlan Sprague–Dawley, Inc (Indianapolis, IN)and housed in a temperature controlled room (25 �C)with a 12-h light/dark cycle. Upon arrival, each animalwas provided with tap water and food (Purina RatChow1) ad libitum for 1 week. Thereafter tap waterwas replaced with the solutions described below. Allprocedures employed during this study were reviewedand approved by the Medical College of Georgia Com-mittee on Animal Use for Research and the VeteransAffairs Medical Center Subcommittee on Animal Useand were in accordance with the Guide for the Care andUse of Laboratory Animals as adopted and promul-gated by US National Institute of Health. All drugswere prepared daily and administered in solutions,which replaced drinking water. This method was pre-ferred over multiple IM injections to maintain more

44 V. Parikh et al. / Journal of Psychiatric Research 37 (2003) 43–51

constant drug levels and to reduce stress and neuro-muscular damage. HAL, RIS, CLZ and OLZ were eachdissolved in a 0.1 M solution of acetic acid and subse-quently diluted (1:100) with tap water to administer thefinal daily dose of drug. The amount of drug intake wasmeasured everyday and adjustments were madedepending upon the fluid consumed and weight of ani-mal. Rats (N=16 per drug group) were exposed to aHAL (2 mg/kg/day), RIS (2.5 mg/kg/day), CLZ (20 mg/kg/day) and OLZ (10 mg/kg/day) for 45 and 90 daysrespectively. The HAL dose was chosen based on pre-vious studies (Mahadik et al., 1988), where it was foundto result in plasma levels comparable to levels in human.This dose is also comparable to doses of atypical anti-psychotics used in rats. The RIS, CLZ and OLZ doseswere selected based on the published reports (Skarsfeldt,1996; Didriksen, 1995; Bymaster et al., 1996). Tap watercontaining 0.1 M acetic acid (100�1) was used for con-trol group to assure that unanticipated effect of thevehicle was not present. After 45 days, eight animalsfrom each group were given a 4-day washout and thensacrificed. The remaining eight animals in each druggroup were continued on the same drug dosage for anadditional 45 days (for a total of 90 days of continuousdrug exposure) and then given a 4-day washout andsacrificed. Four-day washout was considered to be ade-quate to reduce brain levels based on no acute effectswere observed in behavioral tests. The brains of sacri-ficed animals were removed and kept frozen at �70 �Cfor further biochemical analysis.

2.3. Tissue sample preparation

The frozen whole brain tissues were homogenized inice cold PBS (10 mM Na-phosphate, 30 mM NaCl, pH7.4) containing 2 mM EDTA, 4 mM EGTA, 5 mMBHT, 1 mg/ml aprotinin and 20 mg/ml trypsin inhibitorusing polytron homogenizer (Brinkmann Instruments,NY, USA). The homogenate was centrifuged at 15000 gfor 30 min. The supernatant representing the cytosolicfraction was removed and used for protein, lipid per-oxidation and antioxidant enzyme analysis. The residuerepresenting the brain membrane fraction was sus-pended in PBS and used for fatty acid analyses. Proteincontent of each sample was determined by using TotalProtein assay kit (Sigma Diagnostics Inc., St Louis,MO, USA).

2.4. Analyses of antioxidant enzymes

The contents of antioxidant enzymes were determinedby both measuring the enzymatic activity and proteinexpression. Protein contents of MnSOD, CuZnSOD,and CAT were measured in the brain cytosol to rule outany effect produced on enzymatic activities due toknown interaction of free antipsychotics with metallic

cofactors, in addition to use of 4 days of washout toallow substantial reduction of free drug in the brain.Protein levels of GPx were not studied due to unavail-ability of immunoassay kit as well as suitable antibodiesfor cellular GPx.

2.4.1. SOD ActivityTotal SOD activity was determined in the cytosol by

commercially available BIOXYTECH SOD-525 kit(OXIS International, Inc, Portland, OR, USA). Theprotocol was followed according to the direction ofmanufacturer. In a separate reaction, specific MnSODactivity was measured by the addition of 1 mmolsodium cyanide to the sample to inhibit CuZnSOD.SOD activity was calculated as units per mg of protein.We calculated CuZnSOD activity by subtracting thevalue using cyanide from the total SOD value.

2.4.2. CAT and GPx ActivityCAT and GPx activity was determined in the brain

cytosol by commercially available BIOXYTECH Cata-lase-520 and BIOXYTECH GPx-340 kit (OXIS Inter-national, Inc, Portland, OR, USA). The protocol wasfollowed according to the direction of manufacturer.Enzyme activity is expressed as Units/mg protein.

2.4.3. MnSOD ProteinIndirect ELISA was used to determine MnSOD pro-

tein content in rat brain. Aliquots of each of the tissuesample were diluted with coating buffer (0.1 M bicar-bonate buffer, pH 9.6) to achieve a final protein con-centration of 10 mg/ml. Microplates were coatedovernight at 4 �C with 100 ml of sample. Wells werewashed with 300 ml of washing buffer (PBS containing0.1% Tween 20). Unoccupied binding sites on theimmunoplates were blocked by 1-h incubation with 300ml 1% BSA in coating buffer. After washing, wells wereincubated with 100 ml of 2 mg/ml rabbit anti-ratMnSOD polyclonal antibody for 2 h at room tempera-ture. Wells were washed and incubated with 1:2000diluted secondary antibody, HRP-conjugated goat anti-rabbit for 1 h at room temperature. TMB (100 ml)containing 0.0003% H2O2 was added in each well afterextensive washing and plates were kept in dark at roomtemperature for 30 min. The reaction was stopped byaddition of 50 ml of 5 M H2SO4 and the absorbancewas measured at 450 nm with Multiskan microplatereader (Flow Labs Inc., McLean, Virginia, USA).Absorbance values for control groups were normalizedto 100% and data is expressed as% change in absor-bance in various treatment groups. For further con-firmation of results samples showing highest variationin absorbance values among treatment groups weresubjected to western blot analyses for immunodetectionof MnSOD protein. E. Coli MnSOD was used as stan-dard to detect the band.

V. Parikh et al. / Journal of Psychiatric Research 37 (2003) 43–51 45

2.4.4. CuZnSOD and CAT proteinA double antibody sandwich ELISA was developed to

quantitate rat cellular CuZnSOD and CAT. All themeasurements were performed in duplicate. Ninety-sixwell immunoplates were coated overnight at 4 �C with0.1 ml of 1 mg/ml rabbit anti-CuZnSOD polyclonalantibody or rabbit anti-CAT polyclonal antibody pre-pared in coating buffer. After blocking with 1% BSAwells were incubated with 100 ml of rat brain cytosolappropriately diluted in PBS containing 0.1% BSA for2 h at room temperature. After another washing, wellswere incubated with 1:10000 diluted horseradish perox-idase (HRP)—conjugated anti-rat CuZnSOD or anti-ratCAT secondary antibody for 1 h at room temperature.HRP substrate TMB 100 ml containing 0.0003% H2O2

was added in each well after extensive washing andplates were kept in dark at room temperature for 30min. The reaction was stopped by addition of 50 ml of5M H2SO4 and the absorbance was measured. Standardcurve was obtained by fitting the absorbance to differentconcentration of bovine CuZnSOD or CAT by logisticcurve fit. Rat brain CuZnSOD or CAT levels were nor-malized to total protein content and expressed as ng/mgprotein equivalent to bovine CuZnSOD or CAT.

2.5. Fatty acids in rat brain membrane

Total fatty acids were extracted from brain membranesuspension and analyzed by gas chromatography per-formed on a gas chromatograph (Hewlett-Packard,model 5890 SERIES-II, Atlanta, GA, USA) using acapillary column of 30 m�0.32 mm�0.20 mm dimen-sions (Supelco, Inc., Bellefonte, PA, USA) as describedearlier (Mahadik et al., 1996).

2.6. Determination of HAEs: indicator of lipidperoxidation

Measurement of 4-hydroxyalkenals (HAEs) was usedas an indicator of lipid peroxidation. Lipid peroxidesderived from EPUFAs are unstable and decompose toform HAE. HAEs were estimated in rat brain cytosolby commercially available BIOXYTECH HAE-586 kit(OXIS International, Inc, Portland, OR, USA).

2.7. Statistical analysis

The data is expressed as mean�SE. One-way analysisof variance (ANOVA) was applied to compare the dif-ference between control and treatment groups. One-wayANOVA was used as there was only one independentvariable and the data was subjected to post-hoc com-parisons. Kruskal–Wallis one-way ANOVA on rankswas applied when the normality test on data failed.When significant differences were observed, pairwisemultiple comparison by Student–Newman–Keuls test

applied. All statistical tests were performed with thestatistical program SigmaStat.

3. Results

3.1. Effect of antipsychotics on antioxidant enzymeactivities in rat brain

Significant reduction in MnSOD activity wasobserved in HAL treated rats after 45 days (6.7�0.6 vscontrol 10.3�0.7 U/mg protein, P<0.001) as well as 90days (7.2�0.7 vs control 10.4�0.4 U/mg protein,P<0.01) of chronic treatment (Fig. 1). HAL treatmentcaused similar reduction also in CuZnSOD activity (45days, 2.6�0.4 vs control 6.2�0.1 U/mg protein,P<0.001; 90 days 3.5�0.4 vs control 6.7�0.7 U/mgprotein, P<0.01) in rat brain (Fig. 2). No statisticallysignificant differences were observed in SOD enzymeactivities in RIS, CLZ and OLZ treated rats as com-pared to controls (Figs. 1 and 2).CAT activity was also significantly lower after 45 days

of HAL treatment (1.8�0.3 vs control 4.0�0.4 U/mgprotein, P<0.001). However, after 90 days of HALtreatment, the levels of catalase were not significantlydifferent from control (Fig. 3). RIS, CLZ and OLZtreatment did not produce any change in CAT activityafter 45 or 90 days of treatment (Fig. 3). No alterationswere also observed in GPx activity after 45 and 90 daysof antipsychotic treatment (Fig. 4).

3.2. Effect of antipsychotics on content of antioxidantenzyme protein in rat brain

Analyses of antioxidant enzyme protein levelsrevealed similar results as for the respective enzyme

Fig. 1. Effect of chronic (45 and 90 days) treatment with anti-

psychotics on MnSOD activity (units/mg cytosol protein) in rat brain.

Bar graphs show means�SEM (n=6 in each group CON, vehicle

controls group, HAL, haloperidol treated group; RIS, risperidone

treated group; CLZ, clozapine treated group and OLZ, olanzapine

treated group. ** P<0.01 vs control; *** P<0.001 vs control.


activity. As shown in Fig. 5, MnSOD protein levels werealso significantly reduced in HAL treated rats after both45 days (67�4%, P<0.05) and 90 days (74�5%,P<0.05) of treatment compared to controls. No chan-ges were observed in RIS, CLZ and OLZ treated rats ascompared to control (Fig. 5). Western blot analysisconfirmed the specificity of the polyclonal rabbit anti-MnSOD antibody for rat MnSOD that recognized a 25kD band. A representative western blot of MnSOD inbrain samples of various drug groups showing relativedifferences is shown in Fig. 5. CuZnSOD protein levels(Fig. 6) were markedly declined after HAL treatmentfor 45 days (8.2�1.3 Vs control 15.3�1.6 ng/mg pro-tein, P<0.05) and 90 days (8.1�1 Vs control 15.8�2.5ng/mg protein, P<0.05). No such reduction in CuZn-SOD protein was observed after RIS, CLZ and OLZtreatment (Fig. 6). Similar to CAT enzymatic activity,CAT protein content was also found to be significantlyreduced in HAL treated rats (2.6�0.35 vs control5.6�0.9 ng/mg protein, P<0.05) after 45-day treatment

(Fig. 7). The mean value of CAT protein was slightlylower (4.4�0.7 vs control 5.6�0.8 ng/mg protein) in 90days treated HAL rats. CAT protein was also more orless similar to controls after both 45 and 90 days oftreatment with RIS, CLZ and OLZ (Fig. 7).


psychotics on CuZnSOD activity (units/mg cytosol protein) in rat

brain. Bar graphs show means�SEM (n=6 in each group CON,

vehicle controls group, HAL, haloperidol treated group; RIS, risper-

idone treated group; CLZ, clozapine treated group and OLZ, olanz-

apine treated group. ** P<0.01 vs control; *** P<0.001 vs control.


psychotics on CAT activity (units/mg cytosol protein) in rat brain. Bar

graphs show means�SEM (n=8 in each group CON, vehicle controls

group, HAL, haloperidol treated group; RIS, risperidone treated

group; CLZ, clozapine treated group and OLZ, olanzapine treated

group. *** P<0.001 vs control.


psychotics on GPx activity (units/mg cytosol protein) in rat brain. Bar

graphs show means�SEM (n=6 in each group CON, vehicle controls




psychotics on protein content of MnSOD (percent/mg cytosol protein)

in rat brain (top). Bar graphs show means�SEM (n=8 in each group)

from ELISA of all brains. Protein levels in the treated groups are

expressed relative to protein levels in control, which were set to 100%.

CON, vehicle controls group, HAL, haloperidol treated group; RIS,

risperidone treated group; CLZ, clozapine treated group and OLZ,

olanzapine treated group. * P<0.05 vs control. Representative wes-

tern blot analysis (described in text) of MnSOD in rat brain after 45

day chronic treatment with antipsychotics (bottom). Total protein (20

mg) was loaded in SDS-polyacrylamide gel. Proteins were transferred

to PVDF membrane by electroblotting and MnSOD was detected as a

25 kD band by enhanced chemiluminescence using rabbit polyclonal

antibody specific to rat MnSOD. STD, standard MnSOD protein

from E. Coli, CON, vehicle controls group, HAL, haloperidol treated

group; RIS, risperidone treated group; CLZ, clozapine treated group

and OLZ, olanzapine treated group.


3.3. Effect of antipsychotics on fatty acid compositionand lipid peroxidation in rat brain

Both 45 as well as 90 days of treatment with eithertypical or atypical antipsychotics did not produce sig-nificant reductions in the saturated as well as unsatu-rated (a trend P=0.07 for linoleic acid, C18:2) fattyacids in brain membranes (Table 1). However, there wasa marked increase in HAEs (0.66�0.08 nmol/mg pro-tein, P<0.001) in HAL treated rats as compared tocorresponding control (0.21�0.021nmol/mg protein)after 45 days’ treatment (Fig. 8). Similarly significantincrease in HAEs was observed in HAL treated rats(0.57�0.05 vs controls 0.20�0.03 nmol/mg protein,P<0.001) also after 90 days treatment. RIS, CLZ andOLZ did not produce any alteration in HAEs levels

both after 45 and 90 days of treatment and the levelswere similar to controls (Fig. 8). However, slight butinsignificant increase in HAE levels was observed after90 days of RIS treatment (0.32�0.03 nmol/mg protein,P=0.11; Fig. 8).

4. Discussion

In the present study, chronic treatment for both 45 aswell as 90 days with typical antipsychotic HAL in ratsproduced severe oxidative stress, as manifested bymarked changes in antioxidant defense enzymes andassociated oxidative injury as indicated by increasedlipid peroxidation in the brain. Our findings are inagreement with earlier studies, which reported changes


psychotics on protein contents of CAT (ng/mg cytosol protein) in rat

brain. Bar graphs show means�SEM (n=8 in each group) from

ELISA of all brains. CON, vehicle controls group, HAL, haloperidol

treated group; RIS, risperidone treated group; CLZ, clozapine treated

group and OLZ, olanzapine treated group. * P<0.05 vs control.


psychotics on protein contents of CuZnSOD (ng/mg cytosol protein)

in rat brain. Bar graphs show means�SEM (n=8 in each group) from

ELISA of all brains. CON, vehicle controls group, HAL, haloperidol

treated group; RIS, risperidone treated group; CLZ, clozapine treated

group and OLZ, olanzapine treated group. * P<0.05 vs control.

Table 1

Fatty acid composition in rat brain membranes after treatment with antipsychoticsa

Fatty Acid (FA) Drug treatment

CON HAL RIS CLZ OLZ

45 Days 90 Days 45 Days 90 Days 45 Days 90 Days 45 Days 90 Days 45 Days 90 Days

Saturated FA 42.13 41.88 42.20 42.81 41.07 42.05 40.78 41.21 40.88 42.05

�0.69 �0.69 �0.50 �0.11 �0.21 �0.47 �0.35 �0.54 �0.45 �0.36

Monounsaturated FA 26.18 26.45 25.06 25.55 26.39 25.55 26.4 26.28 25.03 25.51

�0.44 �0.44 �0.64 �0.25 �0.39 �0.48 �0.41 �0.56 �0.62 �0.41

Polyunsaturated FA

C18:2o6 (LA) 0.53 0.57 0.54 0.31* 0.59 0.58 0.47 0.51 0.53 0.45

�0.09 �0.13 �0.10 �0.10 �0.09 �0.06 �0.09 �0.09 �0.08 �0.11

C20:4 o6 (AA) 11.03 10.83 11.43 11.15 10.99 11.18 11.38 10.73 11.66 10.86

�0.34 �0.33 �0.36 �0.22 �0.13 �0.24 �0.20 �0.18 �0.32 �0.21

C22:5 o6 (DPA) 4.23 4.06 4.38 4.18 4.27 4.61 4.08 4.36 4.7 3.67

�0.30 �0.41 �0.17 �0.19 �0.46 �0.19 �0.19 �0.18 �0.09 �0.27

C22:6 o6 (DHA) 14.54 14.35 14.91 14.46 14.19 14.60 14.06 14.11 15.70 14.65

�0.45 �0.26 �0.62 �0.25 �0.62 �0.37 �0.39 �0.25 �0.63 �0.23

a P=0.07 vs control. Values are Mean�SEM, percent of the total fatty acids (n=8 in each group). CON, vehicle control group, HAL, halo-

peridol treated group; RIS, risperidone treated group; CLZ, clozapine treated group and OLZ, olanzapine treated group.


in levels of some antioxidant enzyme activity afterchronic treatment with typical antipsychotics (Shivaku-mar and Ravindranath, 1993; Cadet and Perumal,1990). However, for the first time, our study shows thatchronic treatment up to 90 days with atypical anti-psychotics; either RIS or CLZ or OLZ did not changethe level of critical antioxidant enzymes and the lipidperoxidation. This indicates that these medications donot induce the oxidative stress and the associated ser-ious membrane injury, membrane lipid peroxidationproducts in the brain.The cellular oxidative stress increases when the levels

of ROS exceed the levels of antioxidants particularly thelevels of critical antioxidant enzymes (CuZnSOD,MnSOD, GPx and CAT), since their synergistic actioncontrols the oxidative stress and thereby oxidativeinjury (Halliwell, 1992; Mahadik and Mukherjee, 1996).The data on both the enzymatic activities as well as theprotein content of MnSOD, CuZnSOD and CAT fromour carefully designed study, clearly demonstrates thatthe HAL treatment probably affects the genetic expres-sion of these enzymes and differences observed are notrelated to its interaction with metallic cofactors. CATactivity and its protein content was markedly lowerafter 45 days’ HAL treatment, however this differencewas not significant after 90 days’ treatment, and GPxactivity remained unaffected throughout suggesting thatHAL treatment may have differential action on theexpression of these enzymes. These findings are con-sistent with those of Cadet and Perumal (1990), whoreported alteration of SOD and CAT activity but notGPx in rat brain after chronic fluphenazine (anothertypical antipsychotic) treatment. However, for the firsttime, none of the atypical antipsychotics studied here,RIS or CLZ or OLZ were found to cause oxidativestress or oxidative injury.These data suggest that altered levels of one or more

antioxidant enzymes reported for the last over 30 years

in schizophrenic patients, may be partly a result oftreatment with typical antipsychotics. It is also impor-tant to indicate that we have reported altered levels ofthese antioxidant enzymes in never-medicated first-epi-sode psychotic patients (Mukherjee et al., 1996) andothers and we have also reported in chronic patientstreated with typical antipsychotic (Reddy et al., 1991;Yao et al., 1998). The differential effects of these anti-psychotic effects on antioxidant enzymes in rat brain,may be very informative to plan proper studies on theperipheral indices of oxidative stress and injury inpatients. We have recently reported that atypical anti-psychotic treatment of never-medicated first-episodepsychotic patients with risperidone or olanzapinereduced the plasma lipid peroxides and increased themembrane essential fatty acids (Evans et al., 2001). Wehave also reported that compared to never-medicatedfirst-episode psychotic patients, chronic patients treatedwith atypicals have lower levels of plasma lipid per-oxides and higher membrane fatty acids (Khan et al.,2002; Arvindakshan et al., 2002). These differences werealso significantly correlated with the psychopathology.These findings indicate that treatment with atypicalsmay have ‘‘antioxidant’’ effects in patients. This needsfurther investigated within subject studies.Haloperidol may increase oxidative stress primarily

by altered antioxidant defense. Furthermore, it mayexacerbate the oxidative stress by increased oxidativedopamine metabolism following dopamine receptorblockade (Cadet and Lohr, 1989; Mahadik andMukherjee, 1996) as well as by its oxidative bio-transformation to produce a potentially neurotoxicpyridinium metabolite (HAL+) in rat brain (Sub-ramanyam et al., 1990). There is substantial evidencethat this haloperidol’s effect may have pathophysiolo-gical consequences. We have earlier reported thathaloperidol treatment decreased level of cholinergicactivity in several regions of brain (Mahadik et al.,1988) that was prevented by co-treatment with a gly-colipid with anti-oxidant effects (Mahadik andMukherjee, 1995). Now, we have found that treatmentwith haloperidol but none of the atypicals testedreduced the cholinergic activity with concomitantimpaired cognitive performance in rats (Mahadik et al.,2001a). Although a role for free radicals has been sug-gested in clozapine-induced agranulocytosis (Fisher,1991), most of the atypicals have not been investigatedfor a broad range of actions similar to HAL. It isinteresting that high dose risperidone has clinical effectssimilar to typicals, but it did show significant differ-ences compared to haloperidol. It is possible thatlonger periods of treatment may show similarity, sincea trend towards increased levels of lipid peroxides andreduced fatty acids were observed, and earlier we havealso found reduced cholinergic activity after 90 days oftreatment.


psychotics on HAEs (nM/mg cytosol protein) in rat brain. Bar graphs

show means�SEM (n=8 in each group). CON, vehicle controls



group. *** P<0.001 vs control.


Lipid peroxidation is one of the well-established indexof cellular peroxidative membrane injury associatedwith increased oxidative stress, which is generally asso-ciated with the reduced membrane fatty acids (Mahadiket al., 1999). Haloperidol did increase the lipid peroxides(HAEs) but did not significantly reduce the membranefatty acids. It is possible that increased lipid peroxi-dation is compensated by increased incorporation offatty acids in normal rats, since these rats are providedwith sufficient fatty acids in their food. It is also possiblethat the fatty acid changes vary among brain regions,i.e. reduced in one region and increased in other, sincehaloperidol has differential effects on brain regions.Moreover, there was a trend towards the reduction inlevels of linoleic acid (18:2) after 90 days of haloperidoltreatment. HAEs react readily with nucleophilic groups,leading to protein and nucleic acid adducts (Esterbaueret al., 1991) that are known to be cytotoxic, mutagenicand may further alter signal transduction and geneexpression (Burcham, 1998; Keller and Mattson, 1998),which are implicated in schizophrenia (Mahadik et al.,1999). Since antipsychotics are going to be the drugs ofchoice for the treatment of psychotic disorders, theunderstanding of the effects of their action on oxidativestress and oxidative cellular injury may be very impor-tant. However it is important to point out that suchpathology can be effectively prevented by dietary sup-plementation with antioxidants and essential fatty acids(Mahadik and Gowda, 1996; Mahadik and Scheffer,1996; Reddy and Yao, 1996; Mahadik and Evans, 1997;Mahadik et al., 2001b).In summary, the findings of this study indicate some

of the mechanisms underlying the oxidative stress andresultant peroxidative cell injury associated with thechronic administration of HAL. In contrast, neitherRIS, CLZ nor OLZ cause these effects. This maystrongly support the proposed contribution of oxidativeinjury in exacerbating the extrapyramidal and someother serious side effects in patients treated with HALbut not treated with atypical antipsychotics such as RIS,CLZ and OLZ. It is further suggested that there may bethe advantage of dietary supplementation of anti-oxidants and EPUFAs to augment the oxidative injuryand improve outcome by treatment with atypical anti-psychotics.

Acknowledgements

This work was supported partly by a grant fromNIH/NCCAM (R01 AT00147) and partly by financialsupports from Janssen Pharmaceutica Research Foun-dation and Eli Lilly and Company. The authors wouldlike to thank Joe Carothers, Justin White and KavyaSebastian for their excellent technical assistance.

References

Arvindakshan M, Sitasawad S, Debsikdar V, Ghate M, Horrobin D,

Bennett C, Ranjekar PK, Mahadik SP. Membrane essential poly-

unsaturated fatty acids (EPUFA) and schizophrenia outcome:

EPUFA and lipid peroxide levels in never-medicated and medicated

schizophrenics. Biological Psychiatry 2002, (in press).

Buckman TD, Kling AS, Sutphin MS, Steinberg A, Eiduson S. Plate-

let glutathione peroxidase and monoamine oxidase activity in

schizophrenics with CT scan abnormalities: relation to psychosocial

variables. Psychiatry Research 1990;31:1–14.

Burcham PC. Genotoxic lipid peroxidation products: their DNA

damaging properties and role in formation of endogenous DNA

adducts. Mutagenesis 1998;13:287–305.

Bymaster FP, Hemrick-Luecke SK, Perry KW, Fuller RW. Neuro-

chemical evidence for antagonism by olanzapine of dopamine, ser-

otonin, a1-adrenergic and muscarinic receptors in vivo in rats.

Psychopharmacology (Berl) 1996;124:87–94.

Cadet JL, Perumal AS. Chronic treatment with prolixine causes oxi-

dative stress in rat brain. Biological Psychiatry 1990;28:738–40.

Cadet JL, Lohr JB. Possible involvement of free radicals in neurolep-

tic-induced movement disorders. Evidence from treatment of tardive

dyskinesia with vitamin E. Annals of the New York Academy of

Sciences 1989;570:176–85.

Didriksen M. Effects of antipsychotics on cognitive behavior in rats

using the delayed non-match to position paradigm. European Jour-

nal of Pharmacology 1995;28:241–50.

Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of

4-hydroxyalkenal, malondialdehyde and related aldehydes. Free

Radical Biology and Medicine 1991;11:81–128.

Evans DR, Mahadik SP, Akin J, Evans W, Jain D. Cell membrane

essential fatty acid status in drug-naı̈ve first-episode psychotic

patients and its relation to outcome. Schizophrenia Research

(Abstr) 2001;49:83.

Fisher AB. Possible role of free radical formation in clozapine (clo-

zaril)-induced agranulocytosis. Molecular Pharmacology 1991;40:

846–53.

Halliwell B. Reactive oxygen species and the central nervous system.

Journal of Neurochemistry 1992;59:1609–23.

Heikkila RE, Cohen G, Manian AA. Reactivity of various phe-

nothiazine derivatives with oxygen and oxygen radicals. Biochem-

ical Pharmacology 1975;24:363–8.

Horrobin DF, Glen I, Vaddadi K. The membrane hypothesis of schi-

zophrenia. Schizophrenia Research 1994;13:195–208.

Horrobin DF. The phospholipid concept of psychiatric disorders and

its relationship to the neurodevelopmental concept of schizophrenia.

In: Peet M, Glen I, Horrobin DF, editors. Phospholipid Spectrum

Disorders in Psychiatry. Lancashire: Marius Press; 1999. p. 3–20.

Jeding I, Evans PJ, Akanmu D, Dexter D, Spencer JD, Aruoma OI,

Jenner P, Halliwell B. Characterization of the potential antioxidant

and pro-oxidant actions of some neuroleptic drugs. Biochemical

Pharmacology 1995;49:359–65.

Keller JN, Mattson MP. Roles of lipid peroxidation in modulation of

cellular signaling pathways, cell dysfunction and death in the ner-

vous system. Reviews in the Neuroscience 1998;9:105–16.

Khan MM, Evans DR, Gunna V, Scheffer RE, Parikh VV, Mahadik

SP. Reduced erythrocyte membrane essential fatty acids and

increased lipid peroxides in schizophrenia at the never-mediated

first-episode of psychosis and after years of treatment with anti-

psychotics. Schizophrenia Research 2002, (in press).

Liday LA, Pippinger CE, Howard AA, Lieberman JA. Free radical

scavenging enzyme activity and related trace metals in clozapine-

induced agranulocytosis. A pilot study. Journal of Psychopharma-

cology 1995;15:353–60.

Lohr JB, Cadet JL, Lohr MA, Larson L, Wasli E, Wade L, Hylton R,

Vidoni C, Jeste DV, Wyatt RJ. Vitamine E in the treatment of


tardive dyskinesia: the possible involvement of free radical mechan-

isms. Schizophrenia Bulletin 1988;14:291–6.

Lohr JB, Underhill S, Moir S, Jeste DV. Increased indices of free

radical activity in the cerebrospinal fluid of patients with tardive

dyskinesia. Biological Psychiatry 1990;28:535–9.

Mahadik SP, Evans DR. Essential fatty acids in the treatment of

schizophrenia. Drugs of Today 1997;33:1–13.

Mahadik SP, Evans D, Lal H. Oxidative stress and role of antioxidant

and o-3 essential fatty acid supplementation in schizophrenia. Pro-

gress Neuro-Psychopharmacology and Biological Psychiatry 2001b;

25:463–93.

Mahadik SP, Evans DR, Terry A, Hill WD. Neuroprotective action of

atypical antipsychotics in schizophrenia: impaired cognitive perfor-

mance and underlying mechanism of action. Schizophrenia

Research (Abstr) 2001a;49:94.

Mahadik SP, Gowda S. Antioxidants in the treatment of schizo-

phrenia. Drugs of Today 1996;32:553–65.

Mahadik SP, Heljo L, Korenovsky A, Karpiak SE. Haloperidol alters

rat cholinergic system: enzymatic and morphological analysis. Bio-

logical Psychiatry 1988;24:199–217.

Mahadik SP, Mukherjee S. Monosialoganglioside co treatment pre-

vents haloperidol treatment-associated loss of cholinergic enzymes

in rat brain. Biological Psychiatry 1995;38:246–54.

Mahadik SP, Mukherjee S. Free Radical pathology and antioxidant

defense in schizophrenia: a review. Schizophrenia Research 1996;19:

1–18.

Mahadik SP, Mukherjee S, Correnti EE, Scheffer R, Mahadik J. Ele-

vated plasma lipid peroxides at the onset of non-affective psychosis.

Biological Psychiatry 1998;43:674–9.

Mahadik SP, Mukherjee S, Horrobin D, Jenkins K, Correnti EE,

Scheffer RE. Plasmamembrane phospholipid fatty acid composition of

cultured skin fibroblasts from schizophrenic patients: comparison with

bipolar and normal controls. Psychiatry Research 1996;63:133–42.

Mahadik SP, Scheffer RE. Oxidative injury and potential use of anti-

oxidants in schizophrenia. Prostaglandins Leukotrienes Essential

Fatty Acids 1996;55:45–54.

Mahadik SP, Sitasawad V, Mulchandani M. Membrane peroxidation

and the neuropathology of Schizophrenia. In: Peet M, Glen I, Hor-

robin DF, editors. Phospholipid Spectrum Disorders in Psychiatry.

Lancashire: Marius Press; 1999. p. 99–111.

McCreadie RG, Macdonald E, Wiles D, Campbell G, Pateson JR. The

Nithsdale Schizophrenia Surveys XIV: plasma lipid peroxide and

serum vitamin E levels in patients with and without tardive dyskinesia

and in normal subjects. British Journal of Psychiatry 1995;167:1–8.

Mukherjee S, Mahadik SP, Scheefer R, Correnti EE, Kelkar H.

Impaired antioxidant defense at the onset of psychosis. Schizo-

phrenia Research 1996;19:19–26.

Murthy JN, Laev H, Karpiak S, Mahadik SP. Enzymes of oxyradical

metabolism after haloperidol treatment of rat. Society of Neu-

roscience (Abstr) 1989;15:139.

Peet M, Laugharne J, Rangarajan N, Reynold GP. Tardive dyskine-

sia, lipid peroxidation, and sustained amelioration with vitamin E

treatment. International Journal of Clinical Psychopharmacology

1993;8:151–3.

Reddy R, Mahadik SP, Mukherjee S, Murthy JN. Enzymes of the

Antioxidant Defense System in Chronic Schizophrenic Patients.

Biological Psychiatry 1991;30:409–12.

Reddy R, Yao JK. Free radical pathology in schizophrenia: a review.

Prostaglandins Leukotrienes and Essential Fatty Acids 1996;55:33–

43.

Sawas AH, Gilbert JC. Lipid peroxidation as a possible mechanism

for the neurotoxic and nephrotoxic effects of a combination of

lithium carbonate and haloperidol. Archives Internationales Phar-

macodynamie et de Therapie 1985;276:301–12.

Sharma T, Mockler D. The cognitive efficacy of atypical anti-

psychotics in schizophrenia. Journal of Clinical Psychopharmacol-

ogy 1998;18(Suppl 1):12S–19S.

Shivakumar BR, Ravindranath B. Oxidative stress and thiol modifi-

cation induced by chronic administration of haloperidol. Journal of

Pharmacology Experimental Therapeutics 1993;265:1137–41.

Simopoulos AP. Omega-3 fatty acids in health and disease, and in

growth and development. American Journal of Clinical Nutrition

1991;54:438–63.

Skarsfeldt T. Differential effect of antipsychotics on place navigation

of rats in the Morris water maze. A comparative study between

novel and reference antipsychotics. Psychopharmacology (Berl)

1996;124:126–33.

Subramanyam B, Rollema H, Woolf T, Castagnoli N. Identification of

a potentially neurotoxic pyridinium metabolite of haloperidol in

rats. Biochemical Biophysical Research Communications 1990;166:

238–44.

Wainwright PE. Do essential fatty acids play a role in brain and

behavioral development? Neuroscience and Biobehavioral Reviews

1992;16:193–205.

Yao JK. Red blood cell and platelet fatty acid metabolism in schizo-

phrenia. In: Peet M, Glen I, Horrobin DF, editors. Phospholipid

Spectrum Disorders in Psychiatry. Lancashire: Marius Press; 1999.

p. 57–71.

Yao JK, Reddy R, McElhinny LG, vanKammen DP. Effects of halo-

peridol on antioxidant defense system enzymes in schizophrenia.

Journal of Psychiatric Research 1998;32:385–91.

Yao JK, Reddy RD, vanKammen DP. Oxidative damage and Schi-

zophrenia: an overview of its evidence and therapeutic implications.

CNS Drugs 2001;15:287–310.


differential effects of antipsychotics on expression of antioxidant enzymes and membrane lipid...

Documents