Original Contribution

INTERLEUKIN-6 PROTECTS PC12 CELLS FROM 4-HYDROXYNONENAL-INDUCED CYTOTOXICITY BY INCREASING INTRACELLULAR

GLUTATHIONE LEVELS

AKIRA NAKAJIMA , KIYOFUMI YAMADA , LI-BO ZOU, YIJIN YAN, MAKOTO MIZUNO, and TOSHITAKA NABESHIMA

Department of Neuropsychopharmacology and Hospital Pharmacy, Nagoya University Graduate School of Medicine,Nagoya, Japan

(Received 23 October 2001;Revised 7 March 2002;Accepted 27 March 2002)

Abstract—Oxidative stress plays an important role in neuronal cell death associated with many different neurodegen-erative conditions, and it is reported that 4-hydroxynonenal (HNE), an aldehydic product of membrane lipid peroxi-dation, is a key mediator of neuronal cell death induced by oxidative stress. Previously, we have demonstrated thatinterleukin-6 (IL-6) protects PC12 cells from serum deprivation and 6-hydroxydopamine-induced toxicity. Therefore, inthe present study, we examined the effects of interleukins on HNE toxicity in PC12 cells. Exposure of PC12 cells toHNE resulted in a decrease in levels of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction,which was due to necrotic and apoptotic cell death. Addition of IL-6 24 h before HNE treatment provided aconcentration-dependent protection against HNE toxicity, whereas neither IL-1� nor IL-2 had any effect. Addition ofglutathione (GSH)-ethyl ester, but not superoxide dismutase or catalase, before HNE treatment to the culture mediumprotected PC12 cells from HNE toxicity. We found that IL-6 increases intracellular GSH levels and the activity of�-glutamylcysteine synthetase (�-GCS) in PC12 cells. Buthionine sulfoximine (BSO), an inhibitor of�-GCS, reversedthe protective effect of IL-6 against HNE toxicity. These results suggest that IL-6 protects PC12 cells from HNE-induced cytotoxicity by increasing intracellular levels of GSH. © 2002 Elsevier Science Inc.

Keywords—IL-6, 4-Hydroxynonenal, Glutathione, Alzheimer’s disease, Parkinson’s disease, Oxidative stress, Freeradicals

INTRODUCTION

Oxidative stress may contribute to neuronal dysfunctionand death in a variety of disorders, including Alzhei-mer’s disease (AD) and Parkinson’s disease (PD) [1–3].Reactive oxygen species (ROS) are generated in severalmetabolic pathways, a major source being mitochondri-ally derived superoxide anion radical, which gives rise tohydrogen peroxide. Hydrogen peroxide is further con-verted to hydroxyl radical, which induces membranelipid peroxidation [4]. Systems that detoxify ROS in-clude the enzymes superoxide dismutase (SOD), cata-lase, and glutathione peroxidase, and the thiol tripeptideglutathione (GSH). 4-Hydroxynonenal (HNE), an alde-

hydic product of lipid peroxidation, is cytotoxic to non-neuronal and neuronal cells when cells are exposed tovarious oxidative insults [5,6]. HNE forms covalentcross-links with proteins via Michael addition to lysine,cysteine, and histidine residues [7,8]. It is reported thatHNE levels are significantly elevated in the ventricularfluid of Alzheimer’s patients [9] and that HNE-modifiedproteins are significantly elevated in nigral neurons ofParkinson’s patients [10]. HNE is normally detoxified byconjugation with GSH [11–13]. Decreased glutathionelevels are observed in AD and PD [14–16].

In addition to neurotrophic factors such as NGF, re-cent evidence suggests that cytokines play an importantrole in neuronal development, maturation, survival, andregeneration in the nervous system. For instance, inter-leukin-1� (IL-1�), which is an astroglial growth factor[17], increases neuronal survival in dissociated spinalcord cultures [18]. IL-2 enhances the viability of primarycultured cortical neurons [19]. IL-6 has been shown to

Address correspondence to: Toshitaka Nabeshima, Ph.D., Depart-ment of Neuropsychopharmacology and Hospital Pharmacy, NagoyaUniversity Graduate School of Medicine, Showa-ku, Nagoya 466-8560, Japan; Tel:�81 (52) 744-2674; Fax:�81 (52) 744-2682;E-Mail: tnabeshi@med.nagoya-u.ac.jp.

Free Radical Biology & Medicine, Vol. 32, No. 12, pp. 1324–1332, 2002Copyright © 2002 Elsevier Science Inc.Printed in the USA. All rights reserved

0891-5849/02/$–see front matter

PII S0891-5849(02)00845-6

1324

support the survival of cultured rat cholinergic and cat-echolaminergic neurons [20,21], and to protect rat hip-pocampal neurons against glutamate-induced neurotox-icity [22]. We recently demonstrated that IL-6 protectsPC12 cells from serum deprivation-, calcium ionophoreA23187-, and 6-hydroxydopamine-induced toxicity [23,24].

In this study, we examined the effects of IL-1�, IL-2,and IL-6 on HNE toxicity in PC12 cells. We show thatIL-6 protects PC12 cells from HNE-induced neurotoxic-ity and that the protective effect of IL-6 is associatedwith an increase in intracellular GSH levels.

MATERIALS AND METHODS

Materials

Recombinant human IL-6 was a gift from DainipponPharmaceutical Co. Ltd. (Tokyo, Japan), recombinanthuman IL-1� was a gift from Otsuka Pharmaceutical Co.Ltd. (Tokyo, Japan), and recombinant human IL-2 wasdonated by Takeda Chemical Industry (Osaka, Japan).Anti-human-IL-6 antibody was purchased from R & DSystem Inc. (Minneapolis, MN, USA). Nerve growthfactor (NGF) (2.5s, from mouse submaxillary glands),GSH-ethyl ester, buthionine sulfoximine (BSO), cata-lase, and SOD were obtained from Sigma Chemical Co.(St. Louis, MO, USA). 4-HNE was from Cayman chem-ical (Ann Arbor, MI, USA).

Cell culture

PC12 cells were maintained at 37°C in a humidifiedatmosphere containing 5% CO2. Cells were grown inDulbecco’s modified Eagle’s medium (DMEM) with10% (v/v) fetal bovine serum, 5% (v/v) horse serum, 100units/ml penicillin, and 100 �g/ml streptomycin. PC12cells that were cultured in the presence of NGF (50ng/ml) for at least 7 d to differentiate into neuron-likecells were used for all experiments. All the experimentsdescribed below were performed in the presence of NGF(50 ng/ml).

We examined the effects of various interleukins, in-cluding IL-1�, IL-2, and IL-6, on the number of viablecells after incubation with HNE. PC12 cells were platedat a density of 1 � 104 per well in a 96 well collagen-coated culture plate. All interleukins were dissolved inphosphate-buffered saline (PBS) containing 0.1% (w/v)bovine serum albumin and then added to the PC12 cellculture. Under standard experimental conditions, PC12cells were incubated with test agents for 24 h beforeHNE was added, and the cells were cultured for anadditional 24 h in the medium containing test agents andHNE. Twenty-four hours after the addition of HNE, the

effects of interleukins were judged according to the num-ber of viable cells as assessed by the MTT assay [25].The conversion of the dye MTT to formazan dye crystalsin cells has been shown to be related to mitochondrialredox state [26]. Ten microliters of MTT solution (5mg/ml) was added to the culture, and the culture mediumwas incubated for 4 h at 37°C. Cells were then lysed in100 �l of 0.04 M HCl in isopropanol, and absorbancewas quantified using a plate reader.

Morphological assessment of necrotic andapoptotic cells

After HNE treatment, attached and floating cells wereharvested, stained with trypan blue, and counted fortrypan blue-excluding (viable) and trypan blue-stainedcells (necrotic) [27]. For assessment of apoptotic cells,cells were washed three times with PBS and fixed with4% paraformaldehyde for 30 min. After staining with 1�g/ml 4, 6-diamidino-2-phenylindole (DAPI) (SigmaChemical Co.) for 30 min, nuclei were examined byfluorescence microscopy and apoptotic cells scored [27].At least 300 cells in four fields were counted. Necroticand apoptotic cells were expressed as percentage of totalcells.

GSH measurement

Intracellular GSH levels were measured by the enzy-matic recycling procedure of Tietze [28]. Cells (1 � 106)were washed with PBS, and lysed with 10% trichloro-acetic acid at 4°C. The supernatant was obtained forassay after centrifugation (12,000 � g for 15 min at4°C). Determination of GSH levels was carried out fol-lowing removal of the protein precipitant from the su-pernatant solutions by extraction with ether. Residualtraces of ether were removed by vigorous shaking undernitrogen gas. GSH levels were measured by the rate ofcolorimetric change of 625 �M 5,5�-dithiobis(nitroben-zoic acid) at 412 nm in the presence of 0.6 units ofglutathione reductase and 0.25 mM NADPH. GSH levelswere adjusted to sample protein content determined withProtein Assay Rapid Kit Wako (Wako Chemical, Osaka,Japan).

�-Glutamylcysteine synthetase (�-GCS) activity

The activity of �-GCS was measured according to themethod of Seeling and Meister [29], which is based onthe rate of ADP formation. This method, which wasdesigned to measure the activity of the pure enzyme, wasmodified to accurately measure the �-GCS activity ofcrude cell extracts by introducing a short preincubation

1325IL-6 protects PC12 cells

step before initiation of the �-GCS reaction [30]. Cellextracts were prepared by lysing cells (suspended in 100mM Tris-HCl, pH 8.2 containing 20 mM MgCl2, 50 mMKCl, and 2 mM ethylenediaminetetraacetate [EDTA]) bysonication on ice. The lysate was centrifuged (12,000 �g for 15 min at 4°C) and supernatant was used for �-GCSactivity assay. Supernatant was added to prewarmedassay medium containing 100 mM Tris-HCl buffer, 150mM KCl, 5 mM Na2ATP, 2 mM posphoenolpyruvate(PEP), 10 mM sodium L-glutamate, 20 mM MgCl2, 2mM Na2EDTA, 0.2 mM NADH, 17 �g of pyruvatekinase, and 17 �g of lactate dehydrogenase and prein-cubated for 2 min at 37°C. Immediately after preincuba-tion, the �-GCS reaction was initiated at 37°C by addi-tion of L-�-aminobutyrate to a final concentration of 10mM to the test medium, while an equal volume of waterwas added to the control assay medium. The change inabsorbance at 340 nm was monitored for 5 min and the rateof NADH oxidation determined. The enzyme activity wasexpressed as 1 �mol of substrate changed per min.

RT-PCR

PC12 cells were incubated with IL-6 (10 ng/ml) for24 h, then total RNA was extracted from PC12 cells bya method previously described [31,32]. The levels of�-GCS mRNA in PC12 cells were determined by RT-PCR. The mRNA for GAPDH was used as an internalcontrol, to be coamplified with �-GCS mRNA. TotalRNA (1 �g) was converted into cDNA using oligo (dT)12-18 primer (Life Technologies, Tokyo, Japan) andMoloney murine leukemia virus reverse transcriptase(Life Technologies) in a total volume of 25 �l (RT-reaction mixture). PCR was performed using one twenty-fifth of the RT-reaction mixture, 0.4 �M of each (for-ward and reverse) primer, and Ready To Go PCR Beads(Amersham Pharmacia Biotech, Tokyo, Japan) in a totalreaction volume of 25 �l. The primers used were asfollows: �-GCS: 5�-CCTTCTGGCACAGCACGTTG-3�(forward) and 5�-TAAGACGGCATCTCGCTCCT-3�(reverse) [33] GAPDH: 5�-AGGTCGGTGTCAACG-GATTT-3� (forward) and 5�-CAGCATCAAAGGTG-GAGGAA-3� (reverse) [34]. In a preliminary experi-ment, the number of PCR cycles and denaturationtemperature were tested to ascertain a linear workingrange for all PCR products. The experimental amplifica-tion protocol consisted of a first round at 94°C for 5 minand then 27 cycles of denaturation for 1 min at 94°C,annealing for 1 min at 57°C, and extension for 1 min at72°C on a programmable thermal cycler (PCR ThermalCycler; Takara, Shiga, Japan). The PCR products werevisualized by ethidium bromide staining under UV lightafter electrophoresis on a 1.5% agarose gel.

Statistical analysis

The data are given as the mean � SEM. Significancetesting was performed using one-way analysis of vari-ance (ANOVA) followed by Bonferroni’s multiple com-parison test.

RESULTS

IL-6 protects PC12 cells from HNE-inducedcytotoxicity

PC12 cells were exposed to 50 �M HNE for 24 h andthe viability was quantified by MTT assay. Exposure ofPC12 cells to HNE resulted in a decrease in MTT reduc-tion to � 55% of basal levels (Table 1). To investigatethe effects of interleukins, PC12 cells were preincubatedwith interleukins for 24 h before HNE treatment. Table 1shows that treatment with IL-6 (0.1–50 ng/ml) resulted ina dose-dependent protection against HNE-induced cyto-toxicity, whereas IL-1� and IL-2 had no effects in thesame dose range. The protective effect of IL-6 wasnegated when IL-6 was preadsorbed with anti-IL-6 anti-body. Anti-IL-6 antibody alone did not affect HNE-induced cytotoxicity (Fig. 1). These results suggest thatIL-6 specifically protects PC12 cells from HNE-inducedcytotoxicity.

It has been reported that comparable concentrations ofHNE induce both necrosis and apoptosis in human colo-rectal carcinoma cells [27]. During necrosis, the plasmamembrane transport function fails, resulting in cells thatcannot exclude trypan blue or other dyes [35]. Recog-nizable features of apoptotic cells include chromatincondensation, nuclear disintegration, and fragmented nu-clei. Treatment of HNE (50–100 �M) in PC12 cells led

Table 1. Effects of IL-1�, IL-2, and IL-6 on HNE-inducedCytotoxicity in PC 12 Cells

Treatment Conc. (ng/ml) MTT reduction (% of control)

Vehicle 100 � 3.6HNE 50 �M 54.0 � 3.6IL-1� 0.1 49.7 � 7.3

1 60.9 � 2.310 53.8 � 9.750 59.9 � 9.7

IL-2 0.1 59.0 � 1.21 59.9 � 2.5

10 54.3 � 1.550 51.6 � 9.2

IL-6 0.1 63.9 � 8.21 70.9 � 9.1

10 76.5 � 9.4*50 84.1 � 8.0*

PC12 cells were preincubated with interleukins for 24 h, and thenexposed to 50 �M HNE. Cytotoxicity was quantified by MTT assay24 h after HNE treatment. Values are the mean � SEM (n � 4).

* p � .05 vs. HNE alone.

1326 A. NAKAJIMA et al.

to a dose-dependent increase in the number of trypanblue-positive cells, and a moderate (10–20%) increase inapoptotic cells 24 h after HNE treatment (Fig. 2A).Pretreatment of IL-6 10 ng/ml significantly decreased thenumber of trypan blue-positive cells (Fig. 2B) and apo-ptotic cells induced by HNE 50 �M (Fig. 2C). Theseresults suggest that IL-6 protected PC12 cells from HNE-induced both necrosis and apoptosis.

GSH protects PC12 cells from HNE-inducedcytotoxicity

GSH is the most abundant intracellular thiol and actsas a major cellular antioxidant [36]. Previous studiesshowed that GSH can conjugate with, and thereby de-toxify, HNE in nonneuronal and neuronal cells [6,11].Consistent with previous studies, pretreatment of PC12cells with a membrane-permeant form of GSH (GSH-ethyl ester, 1 mM) 2 h before HNE treatment afforded asignificant protection against HNE-induced cytotoxicity(HNE alone: 32.5 � 12.1% MTT reduction, HNE �GSH-ethyl ester: 89.1 � 13.4%, p � .05) (Fig. 3).However, the addition of GSH-ethyl ester to the culturemedium 24 h before HNE treatment had no effect. Thisappears to reflect that GSH-ethyl ester was metabolizedduring the pretreatment period. It has been reported thatthe addition of superoxide dismutase (SOD) and catalase,which catalyze the dismutation of O2

� to H2O2 and thebreak down of H2O2, respectively, has protective effectsagainst 6-hydroxydopamine-induced cytotoxicity inPC12 cells [24,37]. In this study, in contrast to GSH-

ethyl ester, pretreatment of PC12 cells with SOD (1–100units/ml) and catalase (1–100 units/ml) had no effectagainst HNE-induced cytotoxicity (data not shown), con-sistent with HNE acting as a downstream effector of lipidperoxidation-induced toxicity.

IL-6 increases intracellular levels of GSH

It has been reported that cytokines play a role in GSHsynthesis. For example, treatment with tumor necrosisfactor-� (TNF-�) and IL-1� resulted in a marked in-crease in intracellular levels of GSH in mouse endothe-

Fig. 1. Effects of anti-IL-6 antibody on the protective effect of IL-6against HNE-induced cytotoxicity in PC12 cells. PC12 cells werepreincubated with vehicle, IL-6, and anti-IL-6 antibody for 24 h, andexposed to 50 �M HNE. Cytotoxicity was quantified by MTT assay24 h after HNE treatment. Values are the mean � SEM (n � 3). *p �.05 vs. HNE alone. #p � .05 vs. HNE � IL-6.

Fig. 2. Effects of HNE on the numbers of trypan blue-positive andapoptotic cells (A), and effects of IL-6 on HNE-induced cell death (B)(C). (A) PC12 cells were treated with the indicated concentrations ofHNE for 24 h. Cells were stained with trypan blue or DAPI forassessment of necrotic and apoptotic cell phenotypes, respectively. (B)(C) PC12 cells were incubated with IL-6 at the concentration of 10ng/ml for 24 h, and then exposed to 50 �M HNE. Cells were stainedwith trypan blue (B) or DAPI (C) 24 h after HNE treatment. Values arethe mean � SEM (n � 4). *p � .05, **p � .01 vs. HNE alone.

1327IL-6 protects PC12 cells

lial cells [38]. To investigate the effects of interleukinson intracellular levels of GSH in PC12 cells, we mea-sured intracellular GSH levels in PC12 cells 24 h aftertreatment with interleukins. Figure 4A shows that treat-ment with IL-6 at 1-50 ng/ml resulted in a markedincrease in GSH levels, whereas IL-1� and IL-2 had noeffects (Fig. 4B and C). Since exogenous GSH protectedPC12 cells against HNE-induced toxicity (Fig. 3), it isplausible that the protective effect of IL-6 may be asso-ciated with an increase in intracellular GSH levels.

IL-6 increases �-GCS activity and mRNA levels

To investigate the cellular mechanism for the IL-6-induced increase in GSH levels, we measured the activityof �-GCS, the rate limiting enzyme in GSH synthesis,24 h after treatment with IL-6 in PC12 cells. Figure 5Ashows that treatment with IL-6 at 10 ng/ml for 24 hresulted in a marked increase in the activity of �-GCS.Furthermore, we measured the effect of IL-6 on theexpression of �-GCS mRNA in PC 12 cells by RT-PCR.Treatment with IL-6 at a concentration of 10 ng/ml for24 h caused an increase in the �-GCS mRNA level by20% (Fig. 5B and C). These results suggest that IL-6increases GSH levels in PC12 cells due to an increase in�-GCS activity in part caused by the elevation of itstranscriptional levels.

BSO inhibits the protective effect of IL-6 againstHNE-induced cytotoxicity

To test whether the increase in intracellular GSHlevels caused by IL-6 is necessary for its survival-pro-moting effects, we treated PC12 cells with IL-6 and

buthionine-sulfoximine (BSO) for 24 h. BSO inhibits�-GCS activity and by this means causes the depletion ofintracellular GSH [39]. Figure 6A shows that treatmentwith IL-6 alone resulted in a significant increase in GSH,

Fig. 3. Effects of GSH-ethyl ester on HNE-induced cytotoxicity. PC12cells were preincubated with vehicle or GSH-ethyl ester for 2 or 24 h,and exposed to 50 �M HNE for 24 h. Cytotoxicity was quantified byMTT assay 24 h after HNE treatment. Values are the mean � SEM(n � 3). *p � .05 vs. HNE alone.

Fig. 4. Effects of IL-6 (A), IL-1� (B), and IL-2 (C) on intracellularGSH levels in PC12 cells. PC12 cells were incubated with vehicle orthe indicated concentrations of interleukins for 24 h, then intracellularGSH levels were measured. Values are the mean � SEM (n � 3–5).*p � .05 vs. vehicle.

1328 A. NAKAJIMA et al.

and that the IL-6-induced increase in intracellular levelsof GSH was effectively inhibited by 0.2 mM BSO.Figure 6B shows that IL-6 protects PC12 cells fromHNE-induced cytotoxicity (HNE alone: 46.5 � 5.9%MTT reduction, HNE � IL-6: 80.0 � 7.0%, p � .01),and that the protective effect of IL-6 is significantlyinhibited by 0.2 mM BSO (HNE � IL-6 � BSO: 40.4 �7.8% MTT reduction, p � .01). Furthermore, it is indi-cated that GSH depletion caused by BSO potentiated

HNE-induced cytotoxicity (HNE alone: 46.5 � 5.9%MTT reduction, HNE � BSO: 24.4 � 6.2%, p � .05).

DISCUSSION

It has been reported that IL-6 has protective effectsagainst neuronal toxicity [20–24]. The present findingsdemonstrate that IL-6 protects PC12 cells from HNE-

Fig. 5. Effects of IL-6 on �-GCS activity (A) and expression of �-GCSmRNA (B) (C) in PC12 cells. PC12 cells were incubated with vehicleor IL-6 at the concentration of 10 ng/ml for 24 h, then �-GCS activityand expression of �-GCS mRNA were evaluated. (B) A representativepicture of RT-PCR product. Values are the mean � SEM (n � 3). *p �.05 vs. vehicle.

Fig. 6. Effects of BSO on intracellular GSH levels (A) and on theprotective effects of IL-6 against HNE-induced cytotoxicity (B). (A) PC12cells were incubated for 24 h with either 10 ng/ml IL-6, or 10 ng/ml IL-6 �0.2 mM BSO, 0.2 mM BSO. Values are the mean � SEM (n � 4). *p �.05 vs. control. #p � .05 vs. IL-6. (B) PC12 cells were incubated for 24 hwith either 50 �M HNE, 50 �M HNE � 10 ng/ml IL-6, 50 �M HNE �10 ng/ml IL-6 � 0.2 mM BSO, 50 �M HNE � 0.2 mM BSO, or 0.2 mMBSO. Values are the mean � SEM (n � 5). *p � .05, **p � .01 vs. HNEalone. ##p � .01 vs. HNE � IL-6.

1329IL-6 protects PC12 cells

induced cytotoxicity, by inducing �-GCS activity and asubsequent increase in intracellular levels of GSH.

Mitochondrial alterations, including decreases in trans-membrane potential and decreased levels of MTT reduc-tion, occur after oxidative insults. Kruman et al. [6] andShearman et al. [26] have found that �-amyloid protein(A�), a major component of senile plaques in AD [40],rapidly reduces the extent of MTT reduction in PC12 cells.HNE also impairs mitochondrial function in synaptosomes[41], and in PC12 cells [6]. GSH, an inhibitor of HNE-induced cytotoxicity, attenuates oxidative stress-induceddisruption of mitochondrial transmembrane potential inlymphocytes, suggesting that HNE contributes to oxidativestress-induced mitochondrial dysfunction [42]. The firstmajor finding in this study is that IL-6, at a dose range of10–50 ng/ml, protects PC12 cells from the HNE-induceddecrease in MTT reduction (Table 1). Furthermore, sinceIL-6 decreased the number of trypan blue-positive cells andapoptotic cells induced by HNE 50 �M (Fig. 2B and C),IL-6 may have protective effects against HNE-induced bothnecrosis and apoptosis.

GSH, which has been shown previously to bind anddetoxify HNE [5,6,11], protected PC12 cells from HNE-induced cytotoxicity (Fig. 3). Pieri et al. [42] reported thatGSH attenuated oxidative stress-induced disruption of mi-tochondrial transmembrane potential in lymphocytes. It isreported that glutathione depletion in PC12 resulted inselective inhibition of mitochondrial complex I activity andthe subsequent mitochondrial dysfunction [43]. We foundthat IL-6 significantly increased intracellular GSH levels(Fig. 4A), and that the neuroprotective effect of IL-6 wasnegated when GSH synthesis was inhibited by BSO (Fig.6B). Collectively, our findings suggest that an increase inintracellular GSH levels is necessary for the protectiveeffect of IL-6 against HNE-induced cytotoxicity.

It has been reported that cytokines and neurotrophicfactors play a role in GSH synthesis. For instance, NGFincreases GSH concentrations in PC12 cells after 24 h oftreatment by stimulating �-GCS activity and rescues cellsfrom oxidative attack [44]. Hack et al. [45] have reportedthat in vivo treatment of IL-6 in C57BL/6 mice induceswithin 4 h a marked increase in �-GCS activity in erythro-cytes. This �-GCS activity is still increased 17 h aftertreatment with IL-6, and the intracellular GSH levels areincreased to levels higher than pretreatment levels. Wefound in the present study that IL-6 stimulated the expres-sion of �-GCS mRNA and increased the activity of thisenzyme, which led to increased GSH levels in PC12 cells24 h after IL-6 treatment (Fig. 5). It is considered that theactivity of �-GCS is regulated by several mechanisms. Forexample, Urata et al. [46] have reported that TNF-�- orIL-1�-induced increase in �-GCS activity in endothelialcells is associated with an increase in its mRNA expression.In contrast, Pan and Perez-Polo [44] have reported that

NGF does not increase the activity of �-GCS at the tran-scription level, rather it extends the half-life of �-GCSmRNA. In this study, IL-6 stimulated the expression of�-GCS mRNA, but its effect was limited by 20%. There-fore, it is considered that IL-6 may increase the activity of�-GCS at the transcription and posttranscription levels.Further studies are needed to clarify the mechanism thatIL-6 increases the activity of �-GCS.

Although Kalman et al. [47] have described a positiveassociation between IL-6 serum levels and severity ofAD, a significant reduction of IL-6 levels in the cerebro-spinal fluid of AD patients has been reported [48]. Mulleret al. [49] have reported that IL-6 levels in cerebrospinalfluid inversely correlate to severity of PD, especiallybradykinesia more than rigidity but not tremor. Severalstudies indicate an inverse linear relation between theUPDRS bradykinesia score and the nigral dopaminergiccell count [50]. Considering the neuroprotective effectsof IL-6 in this study, these observations lead to thehypothesis that the upregulation of IL-6 probably takesplace at the beginning or in an early phase of the degen-erative process in order to protect or even regeneratelesioned neurons in the mild form of PD. Further studiesare needed to clarify the role of increasing IL-6 inneurodegenerative disease.

Acknowledgements — This study was supported in part by Grants-in-Aid for Science Research (number 12210075), a COE Grant, andSpecial Coordination Funds for the Promotion of Science and Tech-nology, Target-Oriented Brain Science Research Program from theMinistry of Education, Culture, Sports, Science, and Technology ofJapan, and by an SRF Grant for Biochemical Research.

REFERENCES

[1] Jesberger, J. A.; Richardson, J. S. Oxygen free radicals and braindysfunction. Int. J. Neurosci. 57:1–17; 1991.

[2] Mattson, M. P.; Furukawa, K.; Bruce, A. J.; Mark, R. J.; Blanc,E. M. Calcium homeostasis and free radical metabolism as con-vergence points in the pathophysiology of dementia. In: Wasco,W.; Tanzi, R.E., eds. Molecular mechanisms of dementia. To-towa, NJ: Humana; 1996:103–143.

[3] Simonian, N. A.; Coyle, J. T. Oxidative stress in neurodegenera-tive disease. Annu. Rev. Pharmacol. Toxicol. 36:83–106; 1996.

[4] Evans, P. Free radicals in brain metabolism and pathology. Br.Med. Bull. 49:577–587; 1993.

[5] Esterbauer, H.; Schaur, R. J.; Zollner, H. Chemistry and biochem-istry of 4-hydroxynonenal, malonaldehyde, and related aldehyde.Free. Radic. Biol. Med. 11:81–128; 1991.

[6] Kruman, I.; Bruce, A. J.; Bredesen, D. E.; Waeg, G.; Mattson,M. P. Evidence that 4-hydroxynonenal mediates oxidative stress-induced neuronal apoptosis. J. Neurosci. 17:5089–5100; 1997.

[7] Uchida, K.; Stadtman, E. R. Covalent attachment of 4-hy-droxynonenal to glyceraldehyde-3-phosphate dehydrogenase.J. Biol. Chem. 268:6388–6393; 1993.

[8] Uchida, K.; Toyokuni, S.; Nishikawa, K.; Kawakishi, S.; Oda, H.;Hiai, H.; Stadman, E. R. Michael addition-type 4-hydroxy-2-nonenal adducts in modified low-density lipoproteins: markers foratherosclerosis. Biochemistry 33:12487–12494; 1994.

[9] Lovell, M. A.; Ehmann, W. D.; Mattson, M. P.; Markesbery,W. R. Elevated 4-hydroxynonenal in ventricular fluid in Alzhei-mer’s disease. Neurobiol. Aging 18:457–461; 1997.

1330 A. NAKAJIMA et al.

[10] Yoritaka, A.; Hattori, N.; Uchida, K.; Tanaka, M.; Stadtman, E.;Mizuno, Y. Immunohistochemical detection of 4-hydroxynonenalprotein adducts in Parkinson disease. Proc. Natl. Acad. Sci. USA93:2696–2701; 1996.

[11] Spitz, D. R.; Malcom, R. R.; Roberts, R. J. Cytotoxicity andmetabolism of 4-hydroxy-2-nonenal and 2-nonenal in H2O2-re-sistant cells lines. Do aldehyde by-products of lipid peroxidationcontribute to oxidative stress? Biochem. J. 267:453–459; 1990.

[12] Grune, T.; Siem, W. G.; Zollner, H.; Esterbauer, H. Metabolismof 4-hydroxynonenal, a cytotoxic lipid peroxidation product, inEhrlich mouse ascites cells at different proliferation stages. Can-cer. Res. 54:5231–5235; 1994.

[13] Ullrich, O.; Grune, T.; Henke, W.; Esterbauer, H.; Siems, W. G.Identification of metabolic pathways of the lipid peroxidationproduct 4-hydroxynonenal by mitochondria isolated from rat kid-ney cortex. FEBS Lett. 352:84–86; 1994.

[14] Cecchi, C.; Latorraca, S.; Sorbi, S.; Iantomasi, T.; Favilli, F.;Vincenzini, M. T.; Liguri, G. Glutathione level is altered inlymphoblasts from patients with familial Alzheimer’s disease.Neurosci. Lett. 275:152–154; 1999.

[15] Sian, J.; Dexter, D. T.; Lees, A. J.; Daniel, S.; Agid, Y.; Javoy,A. F.; Jenner, P.; Marsden, C. D. Alterations in glutathione levelsin Parkinson’s disease and other neurodegenerative disorders af-fecting basal ganglia. Ann. Neurol. 36:348–355; 1994.

[16] Perry, T. L.; Godin, D. V.; Hansen, S. Parkinson’s disease: adisorder due to nigral glutathione deficiency? Neurosci. Lett.33:305–310; 1982.

[17] Giulian, D.; Young, D. G.; Woodward, J.; Brown, D. C.; Lach-man, L. B. Interleukin-1 is an astroglial growth factor in thedeveloping brain. J. Neurosci. 8:709–714; 1988.

[18] Brenneman, D. E.; Schultzberg, M.; Barfai, T.; Gozes, I. Cytokineregulation of neuronal survival. J. Neurochem. 58:454–460; 1992.

[19] Shimoji, M.; Imai, Y.; Nahajima, K.; Mizushima, S.; Uemura, A.;Kohsaka, S. IL-2 enhances the viability of primary cultured ratneocortical neurons. Neurosci. Lett. 151:170–173; 1993.

[20] Hama, T.; Miyamoto, M.; Tsukui, H.; Nishio, C.; Hatanaka, H.Interleukin-6 as a neurotrophic factor for promoting the survivalof cultured basal forebrain cholinergic neurons from postnatalrats. Neurosci. Lett. 104:340–344; 1989.

[21] Hama, T.; Kushima, Y.; Miyamoto, M.; Kubota, M.; Takei, N.;Hatanaka, H. Interleukin-6 improves the survival of mesence-phalic catecholaminergic and septal cholinergic neurons frompostnatal, two-week-old rats in cultures. Neuroscience 40:445–452; 1991.

[22] Yamada, M.; Hatanaka, H. Interleukin-6 protects cultured rathippocampal neurons against glutamate-induced cell death. BrainRes. 643:173–180; 1994.

[23] Umegaki, H.; Yamada, K.; Naito, M.; Kameyama, T.; Iguchi, A.;Nabeshima, T. Protective effect of interleukin-6 against the deathof PC12 cells caused by serum deprivation or by the addition ofa calcium ionophore. Biochem. Pharmacol. 52:911–916; 1996.

[24] Yamada, K.; Umegaki, H.; Maezawa, I.; Iguchi, A.; Kameyama,T.; Nabeshima, T. Possible involvement of catalase in the protec-tive effect of interleukin-6 against 6-hydroxydopamine toxicity inPC12 cells. Brain Res. Bull. 43:573–577; 1997.

[25] Mossman, T. Rapid colorimetric assay for cellular growth andsurvival: application to proliferation and cytotoxicity assays.J. Immunol. Methods 65:55–63; 1983.

[26] Shearman, M.; Hawtin, S.; Taylor, V. The intracellular compo-nent of cellular 3-(4:5-dimethylthiazol-2-yl)-2:5-diphenyltetrazo-lium bromide (MTT) reduction is specifically inhibited by �-amy-loid peptides. J. Neurochem. 65:218–227; 1995.

[27] Chuan, J.; Ventkataraman, A.; Jennifer, A. P.; Lawrence, J. M.4-Hydroxynonenal induces apoptosis via caspase-3 activation andcytochrome c release. Chem. Res. Toxicol. 14:1090–1096; 2001.

[28] Tietze, F. Enzymatic method for quantative determination ofnanogram amounts of total and oxidized glutathione: applicationsto mammalian blood and other tissues. Anal. Biochem. 27:502–522; 1969.

[29] Seeling, G. F.; Meister, A. �-Glutamylcystyeine synthetase.J. Biol. Chem. 259:3534–3538; 1984.

[30] Ray, S.; Misso, N. L. A.; Lenzo, J. C.; Robinson, C.; Thompson,P. J. Gamma-glutamylcysteine synthetase activity in human lungepithelial (A549) cells: factors influencing its measurement. FreeRadic. Biol. Med. 27:1346–1356; 1999.

[31] Mizuno, M.; Yamada, K.; Olariu, A.; Nawa, H.; Nabeshima, T.Involvement of brain-derived neurotrophic factor in spatial mem-ory formation and maintenance in a radial maze arm test in rats.J. Neurosci. 20:7116–7121; 2000.

[32] Chamberlain, L. H.; Burgoyne, R. D. Identification of a novel cysteinstring protein variant and expression of cystein string proteins innonneuronal cells. J. Biol. Chem. 271:7320–7323; 1996.

[33] Mousatassim, S. E.; Gruerin, P.; Menezo, Y. Mammalian oviductand protection against free oxygen radicals: expression of genesencoding antioxidant enzymes in human and mouse. Eur. J.Obstet. Gynecol. Reprod. Biol. 89:1–6; 2000.

[34] Siripurkpong, P.; Harnyuttanakorn, P.; Chindaduangratana, C.;Kotchabhakdi, N.; Wichyanuwat, P.; Casalotti, S. O. Dexameth-asone, but not stress, induce measurable change of mitochondrialbenzodiazepine receptor mRNA levels in rats. Eur. J. Pharmacol.331:227–235; 1997.

[35] Darzynkiewicz, Z.; Juan, G.; Li, X.; Gorczyca, W.; Murakami, T.;Traganos, F. Cytometry in cell necrobiology: analysis of apopto-sis and accidental cell death (necrosis). Cytometry 27:1–20; 1997.

[36] Meister, A.; Anderson, M. Glutathione. Annu. Rev. Biochem.52:711–760; 1983.

[37] Kim, D. H.; Jang, Y. Y.; Han, E. S.; Lee, C. S. Protective effectof harmaline and harmalol against dopamine- and 6-hydroxydo-pamine-induced oxidative damage of brain mitochondria and syn-aptosomes, and viability loss of Pc12 cells. Eur. J. Neurosci.13:1861–1872; 2001.

[38] Urata, Y.; Yamamoto, H.; Goto, S.; Tsushima, H.; Akazawa, S.;Yamashita, S.; Nagataki, S.; Kondo, T. Long exposure to highglucose concentration impairs the responsive expression of �-glu-tamylcysteine synthetase by interleukin-1� and tumor necrosisfactor-� in mouse endothelial cells. J. Biol. Chem. 271:15146–15152; 1996.

[39] Griffith, O. W.; Anderson, M. E.; Meister, A. Inhibitor of gluta-thione biosynthesis by prothionine sulfoximine (S-n-propyl ho-mocysteine sulfoximine), a selective inhibitor of �-glutamylcys-teine synthetase. J. Biol. Chem. 254:1205–1210; 1979.

[40] Yamada, K.; Nabeshima, T. Animal models of Alzheimer’s dis-ease and evaluation of anti-dementia drugs. Pharmacol. Ther.88:93–113; 2000.

[41] Keller, J. N.; Mark, R. J.; Bruce, A. J.; Blanc, E. M.; Rothstein,J. D.; Uchida, K.; Mattson, M. P. 4-Hydroxynonenal, an aldehy-dic product of membrane lipid peroxidation, impairs glutamatetransport and mitochondrial function in synaptosomes. Neuro-science 80:685–696; 1997.

[42] Pieri, C.; Recchioni, R.; Marcheselli, F.; Moroni, F.; Marra, M.;Benatti, C. The impairment of mitochondrial membrane potentialand mass in proliferating lymphocytes from vitamine E-deficientanimals is recovered by glutathione. Cell. Mol. Biol. (Noisy-le-grand) 41:755–762; 1995.

[43] Jha, N.; Jurmas, O.; Lalli, G.; Liu, Y.; Pettus, E. H.; Greenamyre,J. T.; Liu, R. M.; Forman, H. J.; Andersen, J. K. Glutathionedepletion in PC12 results in selective inhibition of mitochondrialcomplex I activity. J. Biol. Chem. 275:26096–26101; 2000.

[44] Pan, Z.; Perez-Polo, R. Role of nerve growth factor in oxidanthomeostasis: glutathione metabolism. J. Neurochem. 61:1713–1721; 1993.

[45] Hack, V.; Gross, A.; Kinscherf, R.; Bockstette, M.; Fiers, W.;Berke, G.; Droge, W. Abnormal glutathione and sulfate levelsafter interleukin 6 treatment and in tumor-induced cachexia.FASEB. J. 10:1219–1226; 1996.

[46] Urata, Y.; Yamamoto, H.; Goto, S.; Tsushima, H.; Akazawa, S.;Yamashita, S.; Nagataki, S.; Kondo, T. Long exposure to highglucose concentration impairs the responsive expression of �-glu-tamylcysteine synthetase by interleukin-1� and tumor necrosisfactor-� in mouse endothelial cells. J. Biol. Chem. 271:15146–15152; 1996.

[47] Kalman, J.; Juhasz, A.; Laird, G. Serum interleukin-6 levels

1331IL-6 protects PC12 cells

correlate with the severity of dementia in Down syndrome and inAlzheimer’s disease. Acta Neurol. Scand. 96:236–240; 1997.

[48] Yamada, K.; Kondo, K.; Umegaki, H.; Yamada, K.; Iguchi, A.;Fukatsu, T.; Nakashima, N.; Nishiwaki, H.; Shimad, Y.; Sugita,Y.; Yamamoto, T.; Hasegawa, T.; Nabeshima, T. Decreased in-terleukin-6 level in the cerebrospinal fluid of patients with Alzhe-imer-type dementia. Neurosci. Lett. 186:219–221; 1995.

[49] Muller, T.; Blum-Degan, D.; Przuntek, H.; Kuhn, W. Interleu-kin-6 levels in cerebrospinal fluid inversely correlate to severityof Parkinson’s disease. Acta Neurol. Scand. 98:142–144; 1998.

[50] Lee, C. S.; Schulzer, M.; Mak, E.; Hammerstad, J. P.; Calne, S.;Calne, D. B. Patterns of asymmetry do not change over the courseof idiopathic parkinsonism: implications for pathogenesis. Neu-rology 45:435–439; 1995.

ABBREVIATIONS

AD—Alzheimer’s diseaseBSO—buthionine sulfoximine

DAPI—4, 6-diamidino-2-phenylindole�-GCS—�-glutamylcysteine synthetaseGSH—glutathioneHNE—4-hydroxynonenalIL-1�—interleukin-1�IL-2—interleukin-2IL-6—interleukin-6MTT—3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-

zolium bromideNGF—nerve growth factorPBS—phosphate-buffered salinePD—Parkinson’s diseaseROS—reactive oxygen speciesSOD—superoxide dismutaseTNF-�—tumor necrosis factor-�

1332 A. NAKAJIMA et al.