Oxidative stress increases internal calcium stores and reduces a key mitochondrial enzyme

Biochimica et Biophysica Acta 1586 (2001) 177^189 www.bba-direct.com

Oxidative stress increases internal calcium stores and reduces a key mitochondrial enzyme Gary E. Gibson *, Hui Zhang, Hui Xu, Larry C.H. Park, Thomas M. Jeitner Cornell University Medical College at Burke Medical Research Institute, 785 Mamaroneck Avenue, White Plains, NY 10605, USA Received 20 March 2001; received in revised form 13 September 2001; accepted 4 October 2001

Abstract Fibroblasts from patients with genetic and non-genetic forms of Alzheimer’s disease (AD) show many abnormalities including increased bombesin-releasable calcium stores (BRCS), diminished activities of the mitochondrial K-ketoglutarate dehydrogenase complex (KGDHC), and an altered ability to handle oxidative stress. The link between genetic mutations (and the unknown primary event in non-genetic forms) and these other cellular abnormalities is unknown. To determine whether oxidative stress could be a convergence point that produces the other AD-related changes, these experiments tested in fibroblasts the effects of H2 O2 , in the presence or absence of select antioxidants, on BRCS and KGDHC. H2 O2 concentrations that elevated carboxy-dichlorofluorescein (c-H2 DCF)-detectable ROS increased BRCS and decreased KGDHC activity. These changes are in the same direction as those in fibroblasts from AD patients. Acute treatments with the antioxidants Trolox, or DMSO decreased c-H2 DCF-detectable ROS by about 90%, but exaggerated the H2 O2 -induced increases in BRCS by about 4-fold and did not alter the reduction in KGDHC. Chronic pretreatments with Trolox more than doubled the BRCS, tripled KGDHC activities, and reduced the effects of H2 O2 . Pretreatment with DMSO or N-acetyl cysteine diminished the BRCS and either had no effect, or exaggerated the H2 O2 -induced changes in these variables. The results demonstrate that BRCS and KGDHC are more sensitive to H2 O2 derived species than c-H2 DCF, and that oxidized derivatives of the antioxidants exaggerate the actions of H2 O2 . The findings support the hypothesis that select abnormalities in oxidative processes are a critical part of a cascade that leads to the cellular abnormalities in cells from AD patients. ß 2001 Elsevier Science B.V. All rights reserved. Keywords: Alzheimer’s disease; Calcium; Reactive oxygen species; Fibroblast; K-Ketoglutarate dehydrogenase complex; Antioxidant

1. Introduction

Abbreviations: AD, Alzheimer’s disease; BRCS, bombesin-releasable calcium stores; c-H2 DCF-AM, 6-carboxy-dichlorodihydro£uorescein diacetate acetoxymethyl ester; c-H2 DCF, hydrolyzed form; c-DCF, oxidized form; DMSO, dimethylsulfoxide; KGDHC, K-ketoglutarate dehydrogenase complex; NAC, N-acetyl cysteine; RFU, relative £uorescence units; ROS, reactive oxygen species; Tx, Trolox * Corresponding author. Fax: +1-914-597-2757. E-mail address: [email protected] (G.E. Gibson).

Overwhelming evidence indicates that oxidative stress occurs in brains of patients with Alzheimer’s disease (AD) [1^4]. Although reports suggest that oxidative damage is the earliest event in AD [5], studies of autopsy brain cannot distinguish if the markers of oxidative stress result from, or cause the neurodegeneration. Oxidative stress also occurs in brains of mice bearing genetic mutations that reproduce some phenotypic features of AD [6,7]. Oxi-

0925-4439 / 01 / $ ^ see front matter ß 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 5 - 4 4 3 9 ( 0 1 ) 0 0 0 9 1 - 6

BBADIS 62076 10-4-02

178

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

dative stress precedes plaque formation in mice with mutations that lead to AD-like plaques, [8]. Previous studies in humans indicate that metabolic diseases also cause plaques [9] and tangles [10]. Oxidative stress in cells promotes production of amyloidogenic forms of amyloid precursor protein [11,12] and tangle formation [13,14]. Amyloid-L-peptide impairs metabolism [15], causes oxidative stress [1], and can directly produce H2 O2 through metal ion reduction [16]. Antioxidants protect against amyloid-L-peptide toxicity [1]. Cells from AD patients [17], cells bearing mutations that cause AD [18] and cybrids [19] from AD patients have abnormalities in the ability to handle oxidative stress. Calcium uptake by mitochondria isolated from ¢broblasts of AD patients is sensitive to selective oxidative stressors and to antioxidants [20]. The persistence of abnormalities in the ability to handle oxidative stress in cultured ¢broblasts from AD patients suggests that this is an inherent property of AD cells [17,20,21]. Together these ¢ndings suggest that oxidative stress is a critical part of a cascade of events leading to AD. Cellular signaling systems and mitochondrial processes are abnormal in cultured ¢broblasts from patients with AD and in transfected cells bearing ADcausing mutations. Fibroblasts from AD patients have abnormalities in amyloid-L-peptide production, calcium dynamics, energy metabolism, L-adrenergicinduced formation of cyclic AMP, G protein interactions, and the phosphatidyl inositide cascade (see reviews [22,23]). An exaggeration of the bombesinreleasable calcium store (BRCS), or IP3 -releasable calcium pool, is one of the most replicable ¢ndings [24], and is related to reported changes in capacitative calcium entry [5]. BRCS has been examined in ¢broblasts from multiple individuals with AD including non-genetic forms of AD, and in those bearing PS1 [25] or APP mutations [26], and appears to be diagnostic in some populations [27]. BRCS is also altered in brain and ¢broblasts from PS-1 transgenic mice [5]. The mitochondrial enzyme K-ketoglutarate dehydrogenase complex (KGDHC) is reduced in brains from AD patients [28,29], and ¢broblasts from familial [30,31] and non-familial AD patients [30,31]. Furthermore, KGDHC activity in ¢broblasts bearing AD-causing presenilin-1 mutations is more sensitive to oxidative stress [32]. Thus, the occurrence

of multiple AD-related changes in ¢broblasts is well documented. Fibroblasts can be used to examine the interactions of these abnormalities and to determine the cascade of events leading to multiple changes. The goal of the current studies was to determine whether oxidative stress in ¢broblasts could lead to the changes in calcium and KGDHC that occur in cells from patients with AD. A ¢broblast line from a young individual was used because the interpretation would be complicated if other unknown factors (e.g., age-related changes in membranes) had already compromised the cell line. The current experiments lay the foundation for future studies that will be done on cells from multiple AD patients in which antioxidants can be tested in cells with the same genetic background of the patients. Oxidative stress was induced in cells with a general oxidant, H2 O2 , which produces numerous reactive oxygen species (ROS). Speci¢c antioxidants were tested with and without H2 O2 to better de¢ne the nature of the ROS that leads to the changes. The antioxidants that were used included Trolox (Tx, a water soluble analogue of vitamin E which is a general antioxidant that acts on membranes and in the cytosol), N-acetyl cysteine (NAC, a precursor of glutathione) and dimethylsulfoxide (DMSO, a hydroxyl radical scavenger) [33]. Comparisons of the ability of the antioxidants to alter KGDHC or BRCS, to diminish H2 O2 -induced ROS and to reverse H2 O2 -induced changes in KGDHC and BRCS were used to test the role of particular ROS.

2. Materials and methods 2.1. Reagents Fura-2 AM and 6-carboxy-dichlorodihydro£uorescein diacetate acetoxymethyl ester (c-H2 DCF-AM) were purchased from Molecular Probes (Eugene, OR, USA). The £uorescent dyes were made into stocks in dry DMSO and stored under nitrogen at 380‡C. The ¢nal DMSO concentration while loading the dyes into the cells was 0.2%. Tx was from Calbiochem (San Diego, CA, USA). Bombesin, DMSO and all other chemicals were from Sigma Chemical (St. Louis, MO, USA).

BBADIS 62076 10-4-02

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

2.2. Fibroblasts A ¢broblast line from a young control donor GM8399 was purchased from Coriell Cell Repository (Camden, NJ, USA) and maintained by our established procedures [34]. Viability was determined with Erythrosin-B [34]. Cells were grown in Dulbecco’s modi¢ed Eagle media with 1 g/l glucose (Gibco, Grand Island, NY, USA) and supplemented with 10% fetal bovine serum (Sigma). Cells were sub-cultured weekly by trypsinizing and reseeding at 104 / cm2 . For imaging experiments, ¢broblasts were seeded at 500/cm2 on MatTek glass bottom Petri dishes (Ashland, MA, USA) 5^6 days before the experiments. 2.3. Fluorescence imaging of intracellular calcium ([Ca2þ ]i ) [25,26] Fibroblasts were loaded with 2 WM-Fura-2 AM in balanced salt solution (BSS; in mM: NaCl 140, KCl 5, MgCl2 1.5, glucose 5, HEPES 10, CaCl2 2.5; pH 7.4) for 1 h at room temperature (23‡C) in the dark. Cells were then rinsed and kept in calcium free BSS bu¡er for measurement of [Ca2þ ]i on the microscope stage. The calcium free BSS was the same as BSS except that added CaCl2 was reduced to 0.1 mM and EGTA was added to 1 mM. The calculated free [Ca2þ ] in the bu¡er was 7 nM (MaxChelator v 6.63 by Chris Patton, Stanford University, Hopkins Marine Station, Paci¢c Grove, CA 93950-3094, USA). Measurements of [Ca2þ ]i were started exactly 3 min after the cells were changed into calcium free BSS. In all the treatment groups, basal [Ca2þ ]i was measured for 1 min before any manipulations. Measurements were made with an ImageMaster system from PTI (Photon Technology International, Lawrenceville, NJ, USA). Excitation wavelengths of 350 and 378 nm (band passes 2^4 nm) were selected with two monochrometers and interchanged with chopper mirror installed in the light source box. An XF04 dichroic cube (Omega Optical, Brattleboro, VT, USA) with no excitation ¢lter, dichroic 400 nm, and emission ¢lter 510 nm, band pass 40 nm, was installed in an Olympus IMT-2 microscope with a 10UDAPO UV340 objective lens. Pairs of £uorescent images at excitation 350/378 nm were taken with a Hamamatu C2400 SIT camera at 5 s intervals.

179

Each image was the average of 32 images taken within 2 s. Standard images of fura-2 solutions with minimum (0) and maximum ( s 0.1 mM) [Ca2þ ] were taken during each day’s experiment for calculation of intracellular free calcium within the PTI program according to the method of Grynkiewicz et al. [35]. 2.4. Bombesin-releasable [Ca2þ ]i pool Fibroblasts in calcium free BSS bu¡er have basal [Ca2þ ]i of about 66 nM. This value did not change signi¢cantly at the 4 min time point of the measurement (i.e., after 7 min of incubation in calcium free bu¡er). Bombesin (1 WM) increased [Ca2þ ]i in about half of the cells by over 100% of the value before bombesin addition. The bombesin-induced change of [Ca2þ ]i was expressed as the integration of [Ca2þ ]i over the 3 min interval of 240 to 420 s (Fig. 3A). 2.5. Measurement of ROS in ¢broblasts by £uorescence imaging Fibroblasts were loaded with 10 WM 6-c-H2 DCFAM (Molecular Probes, Eugene, OR, USA) in BSS at room temperature for 60 min. During this time the AM group is cleaved to leave the c-H2 DCF trapped in the cells. Cells were then rinsed in BSS and transferred onto the microscope stage for measurements. The £uorescence of the oxidized form of c-H2 DCF (c-DCF) was measured with an excitation wavelength of 488 nm (band pass 3 nm), emission at 535 nm (band pass 45 nm) with the PTI system using a 20U UV Fluor objective. Composite images (i.e., the average of 64 images taken within 2 s) were taken every 10 s. An image of BSS bu¡er in a Petri dish with same quality as those used for the experiments was taken as background for each day’s experiment. Another image of 2 ml of 100 nM c-DCF in BSS in the same dish was also taken. This concentration of c-DCF provided optimal images to correct for background, uniformity and to calibrate intensity for all the images taken on that day. All images were divided by this 100 nm c-DCF image after background subtraction. The ratioed images were then multiplied by 100. The derived values therefore represent the £uorescence of equal concentration of c-DCF in nM.

BBADIS 62076 10-4-02

180

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

2.6. Measurement of ROS in ¢broblasts with a £uorescent plate reader

amounts of water were added to the controls. The antioxidants and bombesin were all dissolved in BSS.

Fibroblasts were grown on 96 well cell culture plates to con£uence in 7 days. The cells were rinsed once with BSS bu¡er and incubated with 10 WM cH2 DCF AM in BSS bu¡er for 60 min at room temperature to load the dye. The cells were then rinsed and incubated with 1 mM H2 O2 in BSS with various treatments at 37‡C in a 5% CO2 incubator. Cells were rinsed once at the end of treatments and 100 Wl of BSS was added into each well. c-DCF £uorescence of 485/538 nm (Ex/Em) was measured in a Molecular Devices (Sunnyvale, CA, USA) GEMINI £uorescence plate reader. 100 Wl of 500 nM c-DCF was measured along with each time’s measurement as a control. The day-to-day variation of the £uorescence of the c-DCF solution was less than 10%. The c-DCF signal can be converted to concentration by the relation that one relative £uorescence unit (RFU) equals 0.07 pmol.

2.9. Data analysis and statistics All imaging experiments were repeated in three dishes each day on at least 2 separate days. Typically, about 30^50 cells per image ¢eld were measured for calcium studies and 10^20 cells per image ¢eld for ROS measurements. Cell values ([Ca2þ ]i in nM, or £uorescence equivalent to nM c-DCF, respectively) were averaged on per cell basis to derive mean and standard error of mean. ANOVA and Student’s t-tests were performed with with Excel. Other measurements were run in triplicates and repeated at least one more time on a separate day.

3. Results 3.1. ROS generation in ¢broblasts and the e¡ects of antioxidants

2.7. Measurements of KGDHC activity Fibroblasts were sub-cultured on 6 well culture plates to reach con£uence in 7 days. After the indicated treatments, cells were collected by scraping in a lysate bu¡er [26] and sheared by passing through a 27 G needle ¢ve times. KGDHC activities of the cell lysate were measured by the conversion rate of NAD to NADH with a 96 well £uorescence plate reader (SpectraMax GEMINI, from Molecular Devices, Sunnyvale, CA, USA). Brie£y, the emission £uorescence at 460 nM was measured after excitation at 340 nm at 30‡C in a reaction mix (pH 7.8) containing in mM: MgCl2 1,CaCl2 1, EDTA 0.5, TPP 0.3, DTT 1, L-NAD 1, Coenzyme A 0.163, K-ketoglutarate 1, Tris^HCl 50, and 0.1% Triton X-100. The £uorescence was converted into NADH with an NADH standard on the day of experiment. An aliquot of the cell lysate was saved for measurement of protein concentration with a Bio-Rad 500-006 kit based on the Bradford dye-binding procedure that utilizes the color change of Coomassie brilliant blue G-250 [36]. 2.8. Dilution of drugs H2 O2 was diluted with H2 0. Thus, equally small

The metabolism of ROS was measured by monitoring the oxidation of c-H2 DCF. Cells were loaded with c-H2 DCF AM. Once inside the cell the AM group is hydrolyzed to give c-H2 DCF. The cH2 DCF serves as an oxidizable substrate for cellular peroxidases and the oxidized product, c-DCF, is highly £uorescent. H2 O2 produced a time- and concentration-dependent increase in c-DCF-detectable ROS formation in ¢broblasts, whether the response was monitored with a microscope (Fig. 1) or with 96 well plates. On the microscope stage, concentrations of H2 O2 ranging from 88 WM to 88 mM produced a robust increase in ROS over the whole 3 min interval. When the analysis was done on the plate reader, 1 mM H2 O2 for 2 h increased the c-DCF signal from 20.7 þ 0.4 to 68.1 þ 6.3 RFU/well (P 6 0.001). Subsequent studies were monitored with the plate reader. 3.2. Acute and chronic e¡ects of antioxidants on H2 O2 -induced ROS generation To assess acute e¡ects of antioxidants, several concentrations of Tx, NAC and DMSO were incubated with the ¢broblasts for 1 h before measurements, and

BBADIS 62076 10-4-02

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

181

were pretreated with antioxidants for 5 days and the e¡ects on H2 O2 -induced ROS formation was determined in the absence of antioxidants (Fig. 2, bottom panel). Tx (200 WM) increased the H2 O2 -

Fig. 1. H2 O2 increases c-DCF-detectable ROS generation in ¢broblasts. c-DCF was monitored in ¢broblasts by £uorescence microscope as described in Section 2. The top panel shows the treatment paradigm. Images were taken every 10 s. H2 O2 (10 Wl or 10 Wl H2 O for controls) at variable concentrations was added 1 min after the start of measurements and monitoring continued for 3 min. Values on the vertical axis are £uorescence equivalent to the concentration of c-DCF. The bottom panel presents the slopes of the c-DCF £uorescence increase over the 3 min interval after the addition of H2 O2 . Each bar represents one tracing in the top panel. Each time point or bar is mean þ S.E.M. from measurements of about 50 cells in ¢ve or six dishes measured on 2 separate days. ***Denotes values are signi¢cantly di¡erent (P 6 0.001) from controls.

the e¡ects on H2 O2 -induced £uorescence were monitored in the presence of the antioxidants (Top panel of Fig. 2). Tx reduced the H2 O2 -induced ROS in a concentration-dependent fashion. The inhibition at the maximal Tx concentration (2 mM) was 84%. NAC at the highest concentration (10 mM) diminished ROS by about 90%. DMSO (0.5%) reduced the H2 O2 -induced response by about 60%, and higher DMSO concentrations produced only slightly greater e¡ects. The above values were obtained using the plate reader, and similar values were obtained with the microscope. To assess chronic e¡ects of antioxidants, cells

Fig. 2. Antioxidants selectively alter H2 O2 -induced ROS generation. c-DCF £uorescence was determined with a plate reader as described in Section 2. The top panel shows the e¡ects of a 1 h pretreatment with the antioxidants with subsequent testing in the presence of the antioxidants. Antioxidants were dissolved directly in BSS bu¡er. The drugs were incubated with the cells for 1 h while the c-H2 DCF-AM was being loaded. Then the cells were changed into 1 mM H2 O2 in BSS or BSS (as a control) and incubated for another 2 h in the presence of the antioxidants. For the chronic pretreatment in the bottom panel, cells were incubated with the antioxidants in culture media for 5 days with subsequent testing in the absence of antioxidants. The media were changed to BSS and the cells were loaded with c-H2 DCF for 1 h in the absence of antioxidants. Cells were then treated with H2 O2 for 2 h. Data are the percentage change induced by H2 O2 over each group’s own control. The 10 mM NAC chronic treatment group is not shown because it slowed growth or caused cell death. Each bar is the mean þ S.E.M. of six measurements on 2 separate days. Asterisks show signi¢cant di¡erence from controls: *P 6 0.05, **P 6 0.01, ***P 6 0.001.

BBADIS 62076 10-4-02

182

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

induced elevation in ROS, while 2 mM Tx reduced the H2 O2 e¡ect on ROS by more than 50%. Pretreatment with 1 mM NAC did not a¡ect H2 O2 induced ROS. Many cells treated with 10 mM NAC did not adhere to the dish so those data are not included. Chronic treatment with DMSO was more e¡ective than the acute treatment and reduced the H2 O2 -induced ROS more than 80%. Thus, Tx and NAC are more e¡ective at diminishing H2 O2 induced ROS with acute, than chronic treatment.

The reverse is true for DMSO. Thus, chronic treatment with antioxidants provided a unique pattern of protection against H2 O2 -induced ROS as determined with c-DCF. 3.3. H2 O2 and cellular calcium dynamics H2 O2 -induced changes in basal [Ca2þ ]i and in BRCS. Bombesin releases calcium from internal calcium stores to increase [Ca2þ ]i . The concentrations of H2 O2 that increased c-DCF-sensitive ROS (see Fig. 1) proportionally elevated basal cytosolic free calcium ([Ca2þ ]i ) and BRCS (Fig. 3). Basal [Ca2þ ]i was minimally, but signi¢cantly, a¡ected by 88 and 880 WM. Higher concentrations of H2 O2 (8.8 mM) produced larger changes in [Ca2þ ]i . H2 O2 at very high concentration (88 mM) increased the basal calcium nearly 5-fold. H2 O2 concentrations as low as 88 WM increased the BRCS by more than 100% over control (Fig. 3; bottom right panel), and 8.8 mM H2 O2 nearly tripled BRCS. At the highest H2 O2 concentration the bombesin-induced change was similar to that in controls. Thus, except at the highest concentrations, H2 O2 proportional increased c-DCF-detectable ROS and BRCS. 3.4. E¡ect of antioxidants on ¢broblasts calcium dynamics (Fig. 4)

Fig. 3. H2 O2 alters the [Ca2þ ]i and BRCS. The top panel shows the experimental paradigm. Fura-2AM was loaded in BSS immediately before the measurements (see Section 2 for details). After loading, the media were changed to calcium free BSS, and exactly 3 min later the calcium measurements were initiated. The [Ca2þ ]i was measured for a total of 7 min at 5 s intervals in each cell dish. The arrow indicates the additions of 10 Wl of either H2 O (control) or H2 O2 . Bombesin was added 3 min later. The bottom panel shows the e¡ect of H2 O2 . The bottom left panel is the e¡ect of 3 min H2 O2 on basal cytosolic free calcium (i.e., the [Ca2þ ]i measurements at 4 min time point of the tracing.) The bottom right panel shows the e¡ects of H2 O2 on BRCS. Values are the integration of the [Ca2þ ]i peak after bombesin addition. The integration started at the 270 s time point to avoid the arti¢cial peak caused by drug addition. Data are mean þ S.E.M. Each time point represents the measurement of 150^250 cells in at least six dishes measured on 2 separate days.

Before assessing the interaction of antioxidants and H2 O2 with calcium dynamics, the e¡ects of the antioxidants on cellular calcium pools were determined. The e¡ects of the antioxidants were studied in the presence of the antioxidant after a 1 h pretreatment (acute), or in the absence of the antioxidant after a 5-day pretreatment (chronic). Basal calcium was measured for 1 min and then H2 O was added (as a control for the experiments in the following paragraph) and bombesin was added 3 min later. Acute treatment of ¢broblasts with antioxidants had minimal e¡ects on BRCS (Fig. 4). However, chronic treatment of ¢broblasts had a striking e¡ect on BRCS and the direction of the e¡ects varied with the antioxidant. All concentrations of Tx elevated the BRCS, and the highest concentration (2 mM) more than doubled the BRCS. On the other hand, NAC and DMSO both diminished the BRCS (Fig. 4).

BBADIS 62076 10-4-02

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

183

induced changes in BSCS, the drugs were added 1 h before measuring calcium and were present during the measurements. One h pretreatment with 20 WM or 200 WM Tx exaggerated the increase in BRCS due to 88 WM H2 O2 . Indeed, the higher concentration (200 WM) nearly quadrupled the BRCS over the control values. Neither the highest Tx concentration (2 mM), nor any concentration of NAC, altered the H2 O2 -induced change. However, DMSO exaggerated the H2 O2 -induced change in BSCS by approximately 4-fold. Thus, antioxidant treatments that diminish cDCF-sensitive ROS exaggerate the e¡ects of H2 O2 on BRCS.

Fig. 4. Antioxidants and BRCS. Calcium measurements followed the paradigm in Fig. 3. For acute antioxidant treatments, drugs were present during the 1 h fura-2AM loading period in BSS and during the subsequent [Ca2þ ]i measurements in calcium free BSS. H2 O (10 Wl) was added 1 min after the initiation of the calcium measures (as a control for studies in Fig. 5). For chronic studies, antioxidants were incubated with the cells in culture media for 5 days. Drugs were absent during the fura2AM loading in BSS and subsequent [Ca2þ ]i measurements were in calcium free BSS. H2 O (as a control for Fig. 5) was added after 1 min measurement of basal calcium. The bars show the integrated [Ca2þ ]i peak (nM min) of the bombesin response as shown in the bottom right of Fig. 3. A control group was included with each drug because di¡erent drugs were tested on separate days. Note that NAC at 10 mM for 5 days or 100 mM for 1 h were not included in the graph, because they slowed the cell growth or killed the cells when cells were seeded at a light density for imaging. Data are means þ S.E.M. Each column represents the measurement of 250^350 cells in at least six dishes measured on two separate days. Asterisks indicate di¡erence from the control group: *P 6 0.05, **P 6 0.01, ***P 6 0.001.

3.5. Antioxidants and the H2 O2 -induced changes in bombesin-sensitive calcium stores (Fig. 5) To test the acute e¡ects of antioxidants on H2 O2 -

Fig. 5. Antioxidants selectively alter the H2 O2 -induced increase in BSCS. The experimental paradigm for both the acute and chronic treatment was identical to that in Fig. 4, except that 10 Wl of H2 O2 (88 WM ¢nal) was added in place of H2 O. However, the presentation is di¡erent. Each bar indicates the percent increase in BSCS induced by H2 O2 over each group’s H2 O control (in Fig. 4). Data were calculated on per cell basis and are means þ S.E.M. Each column represents the measurement of 250^350 cells in at least six dishes measured on 2 separate days. Asterisks indicate di¡erence from the control group: *P 6 0.05, **P 6 0.01, ***P 6 0.001.

BBADIS 62076 10-4-02

184

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

ately. H2 O2 reduced KGDHC activity in ¢broblasts in a concentration-dependent fashion at both 3 min and 2 h. 3.7. E¡ect of antioxidants on KGDHC and H2 O2 induced reductions in KGDHC (Fig. 7) Short-term treatment (1 h) with the antioxidants that diminish H2 O2 -induced ROS selectively a¡ected KGDHC activities. In the absence of H2 O2 , the antioxidants had almost no e¡ect on KGDHC. SurprisFig. 6. E¡ects of H2 O2 on KGDHC activities in ¢broblasts. Either 1 or 10 mM H2 O2 was incubated with cells in BSS at 37‡C for either 3 min or 2 h. The cells were then scraped and KGDHC was measured as described in Section 2. KGDHC activity was calculated as mU/mg protein. The average control value was 3.82 þ 0.1 mU/mg protein. Asterisks show signi¢cant di¡erences (P 6 0.001) from the control for same time interval. Each value is the mean þ S.E.M. of seven measurements done on 2 separate days.

To test the chronic e¡ects of antioxidants, they were added 5 days before testing, and the drugs were removed before the calcium measurements. Only the highest concentration of Tx (2 mM) diminished the H2 O2 exaggeration in the BRCS under these conditions (Fig. 5, bottom panel). Although Tx diminished the H2 O2 -induced change in BRCS, the level of BRCS was still higher than control because of the very large induction of BRCS by 2 mM TX alone (see Fig. 4). On the other hand, long-term treatment with 1 mM NAC exaggerated the H2 O2 response. Five-day treatments with NAC at higher concentrations (10 or 100 mM) caused cells to detach so those results are not presented. DMSO had no e¡ect on the H2 O2 -induced changes in calcium (Fig. 5, bottom panel). Thus, although chronic treatment with Tx (2 mM) and DMSO (2%) diminished H2 O2 -induced, DCF-sensitive ROS (see Fig. 2), Tx reversed the resulting calcium de¢cit, but DMSO did not. 3.6. H2 O2 inactivates KGDHC activities in ¢broblasts (Fig. 6) Fibroblasts were incubated with 1 or 10 mM H2 O2 for 3 min or 2 h. The cells were then rinsed, lysed and their KGDHC activity was determined immedi-

Fig. 7. E¡ects of antioxidants on KGDHC in ¢broblasts and their e¡ects on H2 O2 -induced reduction in KGDHC. Fibroblasts were incubated with antioxidants in BSS for 1 h (acute) or in culture media for 5 days (chronic) at 37‡C. Cells were then treated with H2 O2 for 2 h and KGDHC was measured as in Fig. 6. Each column is the mean þ S.E.M. of measurements made on at least 2 days in triplicates. Asterisks indicate di¡erences between the BSS control group and the antioxidanttreated: *P 6 0.05, **P 6 0.01, ***P 6 0.001. þ Indicates di¡erences between H2 O2 control group and the antioxidants and H2 O2 treated group: þ P 6 0.05, þþ P 6 0.01, þþþ P 6 0.001.

BBADIS 62076 10-4-02

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

ingly, the H2 O2 -induced reductions of KGDHC activities were only minimally altered by concentrations of Tx that signi¢cantly diminished H2 O2 -induced ROS. Furthermore, concentrations of NAC that reduced ROS exaggerated the H2 O2 -induced reduction in KGDHC. DMSO did not alter KGDHC in the presence or absence of H2 O2 . Chronic treatment with the antioxidants produced a very striking response on KGDHC. All concentrations of Tx increased KGDHC activities in the absence of H2 O2 . The elevation was maximal at 2 mM and the e¡ect was nearly 3-fold. Following H2 O2 treatment, all Tx-treated cells had higher KGDHC activities than the H2 O2 alone. The intermediate concentration of Tx diminished the H2 O2 -induced reduction in KGDHC activities by 50%, but the activity was still higher than the non-H2 O2 -treated control, because of the very large induction of KGDHC by 2 mM Tx. On the other hand, NAC exaggerated H2 O2 -induced de¢cits on KGDHC activities and DMSO was ine¡ective. 3.8. H2 O2 and cell death The e¡ects of H2 O2 on KGDHC and BRCS were not related to cell death. Cells were treated with either 1 or 10 mM H2 O2 for 2 h. The ability of the cells to exclude ethidium bromide was used to determine cell death. No cell death was observed at 1 mM H2 O2 , which approximates the concentration used in most of our experiments. Even 10 mM H2 O2 only produced minimal cell death (Fig. 8).

185

Fig. 8. Fibroblast viability following treatment with H2 O2 . Cells were treated with the indicated concentration of H2 O2 in BSS for 2 h at 37‡C. Viability was determined with ethidium bromide [34]. At least 100 cells were counted in each group. Values are means þ S.E.M.

sults are consistent with the following three part hypothesis: (i) The diminution of the c-DCF signal by the antioxidants suggests that they are better substrates for cellular peroxidases than c-DCF. (ii) The reaction of these antioxidants with H2 O2 -derived species converted them to speci¢c oxidants. (iii) The resulting oxidants select between the two cell targets KGDHC and BRCS. These suggestions are developed in the following paragraphs. The interaction of cellular peroxidases with NAC, Tx, DMSO and c-DCF provides a plausible explanation for the ability of the antioxidants to diminish the c-DCF signal and to subsequently alter cell processes. In cells, H2 O2 is catabolized by either catalase or peroxidases that reduce H2 O2 at the expense of an oxidizable substrate (RH; c-H2 CDF, Tx, NAC, DMSO), as shown below:

4. Discussion

2RH ! 2R þ 2Hþ þ 2e3

The experiments demonstrate that BRCS and KGDHC are cellular targets of H2 O2 -induced ROS, but the mechanism(s) for the H2 O2 -induced change is unknown. The current studies used antioxidants in the presence or absence of H2 O2 to demonstrate that di¡erent H2 O2 -derived oxidative species act on the two cellular targets, BRCS and KGDHC. As expected, antioxidants diminished the H2 O2 -induced formation of c-DCF-sensitive ROS. Unexpectedly, the reduction in c-DCF-sensitive ROS did not predict the bene¢cial e¡ects on the cellular targets following either chronic or acute treatments. The re-

H2 O2 þ 2e3 þ 2Hþ ! H2 O Since the identity of the peroxidase is unknown, subsequent discussion will refer to enzyme as cH2 DCF-oxidizing peroxidase. NAC, Tx [37^44], DMSO [45,46] and c-H2 DCF [47^49] are oxidizable substrates for the cellular c-H2 DCF-oxidizing peroxidase. Thus, the diminution of the c-DCF signal by acute exposure to NAC, Tx and DMSO suggests that these antioxidants are better substrates for the cellular c-H2 DCF-oxidizing peroxidase than c-H2 DCF. As indicated by the equation, the diminished c-

BBADIS 62076 10-4-02

186

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

DCF signal by each antioxidant indicates the production of an oxidant. The oxidation of Tx and NAC by peroxidases generates phenoxyl and thiyl radicals, respectively [38^44,50,51]. The thiyl and phenoxyl radicals generated from NAC and Tx can then react with oxygen to produce a spectrum of radical species [48]. DMSO can also generate a variety of radical species [52,53] and von Sonntag [54] has described several prooxidant e¡ects of DMSO in irradiation studies. The utilization of NAC, Tx and DMSO by a c-H2 DCF-oxidizing peroxidase could also account for the ability of these compounds, which markedly di¡er on the basis of structure and reactivity with H2 O2 , to produce similar e¡ects on the oxidation of c-H2 DCF. The response of BRCS to H2 O2 with and without the antioxidants demonstrates that these calcium stores are sensitive to oxidative stress. The response of the BRCS to H2 O2 and the antioxidants suggests some regulator of BRCS is an oxidizable substrate for the peroxidase that oxidizes c-H2 DCF. Moreover, the data suggest that the oxidation of this regulator stimulates BRCS. Thus, the addition of H2 O2 oxidizes c-H2 DCF and the regulator of BRCS. The inability of Tx, DMSO or NAC to protect this ‘process’ from H2 O2 suggests that the BRCS regulator is a better substrate for the c-H2 DCF-oxidizing peroxidase than any of these antioxidants. Thus, the substrate preference for this peroxidase would be BRCS regulator s Tx, NAC, DMSO s c-H2 DCF. The exaggeration of the H2 O2 -induced response by Tx and DMSO suggests that the reaction of Tx or DMSO with peroxidase produces products that further oxidize the regulator of BRCS. The relative abilities of these antioxidants and cH2 DCF to act as substrates for the cellular cH2 DCF-oxidizing peroxidase could explain the effects of H2 O2 and the antioxidants on KGDHC. Although the antioxidants diminished the H2 O2 -induced changes in c-H2 DCF, as described above, they did not diminish the H2 O2 -induced de¢cit in KGDHC activities. This suggests that the oxidantsensitive component of KGDHC is a better substrate for the c-H2 DCF-oxidizing peroxidase than any of the antioxidants. A likely candidate is the lipoamide component of the KGDHC reaction. High concentrations of NAC exaggerated the H2 O2 -induced deficits in KGDHC, whereas DMSO and Tx were in-

e¡ective. This suggests that other actions of NAC contribute to its e¡ects on KGDHC. One possibility is that the stimulation of cellular GSH production by NAC. Svensson has shown that GSH stimulates the oxidation of the thiol substrates [37]. Thus, the effects of NAC on KGDHC may be due the combined actions of NAC and NAC-derived GSH. The e¡ect of NAC, DMSO and Tx is a di¡erent than the e¡ect that these antioxidants have on BRCS, which were altered by DMSO and Tx. The chronic pretreatment with these drugs and the subsequent interaction with H2 O2 suggests that the oxidant species produced from these antioxidants selectively a¡ect the cells’ processes. Chronic Tx treatment increased BRCS whereas the NAC and DMSO reduced the BRCS. The results suggest that the regulator of BRCS that was sensitive to the Tx-derived oxidant species was already maximally oxidized following chronic Tx, so that the BRCS was no longer increased by H2 O2 in combination with Tx. Indeed, it was diminished. This suggests that the same oxidant species was involved in the chronic treatment and the acute treatment in the presence of H2 O2 . On the other hand, chronic incubation with NAC produced a species that reacted with H2 O2 to produce a species that oxidized (i.e., enhanced) the process leading to BSCR process. DMSO was ine¡ective. These responses are just reverse to those occurring with acute antioxidant treatment in the presence of H2 O2 , which suggests that di¡erent oxidants are involved. Thus, each of the antioxidants produced selective changes in calcium depending upon the length of the incubation and the presence or absence of H2 O2 . Just as in the acute experiments, the e¡ects of antioxidant pretreatment on c-DCF £uorescence did not predict the response of the cellular markers. Chronic treatment with the antioxidants had a selective e¡ect on KGDHC that di¡ered from the effect on BRCS. Chronic treatment with Tx generated a species that increased KGDHC activities 3-fold, but neither NAC nor DMSO a¡ected KGDHC. KGDHC activities respond like an antioxidant enzyme. In the short term, it is oxidized, but low-grade oxidative stress over time leads to its induction. Once induced it is still sensitive to H2 O2 , but a large reserve is now present so that in spite of the reduction, the remaining activity is still higher than the nonstressed control. The KGDHC activities are mea-

BBADIS 62076 10-4-02

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

sured in the presence of the reducing agent dithiothreitol, so the Tx-induced increase is not just an e¡ect on the oxidation state of KGDHC. Interactions of oxidative stress with internal calcium stores, especially in the endoplasmic reticulum, may be a common link in a variety of neurodegenerative processes, but the mechanism may vary between disorders (see review in [55,56]). For example, BRCS are altered in ¢broblasts from AD patients, as well as in ¢broblasts and neurons from transgenic mice bearing gene mutations that lead to AD [5]. In these cells the alteration is due to an increase in the size of the endoplasmic reticulum calcium store [5]. In some tissues, oxidative stress including H2 O2 inactivates the endoplasmic reticulum Ca-ATPase leading to impaired calcium re-uptake into the endoplasmic reticulum [57,58]. H2 O2 had only small effects on resting [Ca2þ ]i, whereas the increase in the calcium response following bombesin was large, which suggests that release from the internal stores were selectively increased. Reductions in KGDHC have been implicated in AD and other neurodegenerative disorders [59], and KGDHC is particularly sensitive to oxidative stress [32]. The current studies show these e¡ects can be speci¢c for various ROS. Our previous studies show that peroxynitrite and NO both inactivate the isolated enzyme and the intracellular enzyme [60]. Others have shown that hydroxynonenal inactivates KGDHC in mitochondria [61], and that H2 O2 inactivates KGDHC in isolated nerve endings [62]. The current data suggest that the low-grade oxidative stress induces KGDHC activity, while higher levels of oxidative stress inactivate it. The current studies expand those studies by showing that it occurs in yet another cell type, and by demonstrating the relation of the changes in KGDHC to those in alterations of calcium pools. In addition, there studies reveal a selective action with speci¢c antioxidants. H2 O2 -induced oxidative stress produces changes in cellular signaling and metabolism that resemble the abnormalities that occur in ¢broblasts from patients with AD. However, bene¢cial e¡ects cannot be predicted from changes in c-DCF, since antioxidant treatments are complex and antioxidants can exaggerate rather than diminish the H2 O2 -induced abnormalities.

187

Acknowledgements This work was supported by the Burke Medical Research Institute and the following Grants: NIA RO-1 (AG11921); NIA PP (AG14930); NIA RO-1 (AG19589).

References [1] K. Hensley, D.A. Butter¢eld, N. Hall, P. Cole, R. Subramaniam, R. Mark, M.P. Mattson, W.R. Markesbery, M.E. Harris, M. Aksenov et al., Reactive oxygen species as causal agents in the neuro toxicity of the Alzheimer’s disease-associated amyloid beta peptide, Ann. N. Y. Acad. Sci. 786 (1996) 120^134. [2] M.A. Smith, L.M. Sayre, V.E. Anderson, P.L. Harris, M.F. Beal, N. Kowall, G. Perry, Cytochemical demonstration of oxidative damage in Alzheimer disease by immuno chemical enhancement of the carbonyl reaction with 2,4- dinitrophenylhydrazine, J. Histochem. Cytochem. 46 (1998) 731^735. [3] M.A. Smith, C.A. Rottkamp, A. Nunomura, A.K. Raina, G. Perry, Oxidative stress in Alzheimer’s disease, Biochim. Biophys. Acta 1502 (2000) 139^144. [4] W.R. Markesbery, J.M. Carney, Oxidative alterations in Alzheimer’s disease, Brain Pathol. 9 (1999) 133^146. [5] M.A. Leissring, Y. Akbari, C.M. Fanger, M.D. Cahalan, M.P. Mattson, F.M. LaFerla, Capacitative calcium entry de¢cits and elevated luminal calcium content in mutant presenilin-1 knockin mice, J. Cell Biol. 149 (2000) 793^798. [6] A.J. Bruce, B. Malfroy, M. Baudry, beta-Amyloid toxicity in organotypic hippocampal cultures: protection by EUK-8 a synthetic catalytic free radical scavenger, Proc. Natl. Acad. Sci. USA 93 (1996) 2312^2316. [7] A. Nunomura, G. Perry, G. Aliev, K. Hirai, A. Takeda, E.K. Balraj, P.K. Jones, H. Ghanbari, T. Wataya, S. Shimohama, S. Chiba, C.S. Atwood, R.B. Petersen, M.A. Smith, Oxidative damage is the earliest event in Alzheimer disease, J. Neuropathol. Exp. Neurol. 60 (2001) 759^767. [8] D. Pratico, K. Uryu, S. Leight, J.Q. Trojanoswki, V.M. Lee, Increased lipid peroxidation precedes amyloid plaque formation in an animal model of Alzheimer amyloidosis, J. Neurosci. 21 (2001) 4183^4187. [9] M. Kaido, H. Fujimura, F. Soga, K. Toyooka, H. Yoshikawa, T. Nishimura, T. Higashi, K. Inui, H. Imanishi, S. Yorifuji, T. Yanagihara, Alzheimer-type pathology in a patient with mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS), Acta Neuropathol. Berl. 92 (1996) 312^318. [10] K.M. Cullen, G.M. Halliday, Neuro¢brillary tangles in chronic alcoholics, Neuropathol. Appl. Neurobiol. 21 (1995) 312^318. [11] H. Misonou, M. Morishima-Kawashima, Y. Ihara, Oxidative stress induces intracellular accumulation of amyloid

BBADIS 62076 10-4-02

188

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189 beta-protein (Abeta) in human neuroblastoma cells, Biochemistry 39 (2000) 6951^6959. D. Gabuzda, J. Busciglio, L.B. Chen, P. Matsudaira, B.A. Yankner, Inhibition of energy metabolism alters the processing of amyloid precursor protein and induces a potentially amyloidogenic derivative, J. Biol. Chem. 269 (1994) 13623^ 13628. J.P. Blass, A.C. Baker, L.W. Ko, R.S. Black, Induction of Alzheimer antigens by an uncoupler of oxidative phosphorylation, Arch. Neurol. 47 (1990) 864^869. B. Cheng, M.P. Mattson, Glucose deprivation elicits neuro¢brillary tangle-like antigenic changes in hippocampal neurons: prevention by NGF and bFGF, Exp. Neurol. 117 (1992) 114^123. M.S. Shearman, C.I. Ragan, L.L. Iversen, Inhibition of PC12 cell redox activity is a speci¢c early indicator of the mechanism of beta-amyloid-mediated cell death, Proc. Natl. Acad. Sci. USA 91 (1994) 1470^1474. X. Huang, C.S. Atwood, M.A. Hartshorn, G. Multhaup, L.E. Goldstein, R.C. Scarpa, M.P. Cuajungco, D.N. Gray, J. Lim, R.D. Moir, R.E. Tanzi, A.I. Bush, The A beta peptide of Alzheimer’s disease directly produces hydrogen peroxide through metal ion reduction, Biochemistry 38 (1999) 7609^7616. G.E. Gibson, H. Zhang, K.R. Sheu, L.C. Park, Di¡erential alterations in antioxidant capacity in cells from Alzheimer patients, Biochim. Biophys. Acta 1502 (2000) 319^329. J.N. Keller, Q. Guo, F.W. Holtsberg, A.J. Bruce-Keller, M.P. Mattson, Increased sensitivity to mitochondrial toxininduced apoptosis in neural cells expressing mutant presenilin-1 is linked to perturbed calcium homeostasis and enhanced oxyradical production, J. Neurosci. 18 (1998) 4439^ 4450. J.P. Sheehan, R.H. Swerdlow, S.W. Miller, R.E. Davis, J.K. Parks, W.D. Parker, J.B. Tuttle, Calcium homeostasis and reactive oxygen species production in cells transformed by mitochondria from individuals with sporadic Alzheimer’s disease, J. Neurosci. 17 (1997) 15. U. Kumar, D.M. Dunlop, J.S. Richardson, Mitochondria from Alzheimer’s ¢broblasts show decreased uptake of calcium and increased sensitivity to free radicals, Life Sci. 54 (1994) 1855^1860. G.E. Gibson, H. Zhang. E¡ects of Ginkgo biloba (EGb761) on metabolism of reactive oxygen species in ¢broblasts from Alzheimer’s Disease patients and controls, in: Y. Christen (Ed.), Advances in Ginkgo biloba Extract Research Vol. 8. Ginkgo biloba Extract (EGB 761) as a Neuroprotective agent: from basic studies to clinical trials, IPSEN Foundation, Salal editeur, Marseille, 2000. H.M. Huang, R. Martins, S. Gandy, R. Etcheberrigaray, E. Ito, D.L. Alkon, J.P. Blass, G.E. Gibson. The use of cultured ¢broblasts in elucidating the pathophysiology and diagnosis of Alzheimer’s disease, in: J.F. Disterhoft, W.H. Gispen, J. Traber, Z. Khachaturian (Eds.), Calcium Hypothesis of Aging and Dementia. Ann. N. Y. Acad. Sci. 747 (1994) 225^244.

[23] L. Gasparini, M. Racchi, G. Binetti, M. Trabucchi, S.B. Solerte, D. Alkon, R. Etcheberrigaray, G. Gibson, J. Blass, R. Paoletti, S. Govoni, Peripheral markers in testing pathophysiological hypotheses and diagnosing Alzheimer’s disease, FASEB J. 12 (1998) 17^34. [24] R. Etcheberrigaray, N. Hirashima, L. Nee, J. Prince, S. Govoni, M. Racchi, R.E. Tanzi, D.L. Alkon, Calcium responses in ¢broblasts from a symptomatic members of Alzheimer’s disease families, Neurobiol. Dis. 5 (1998) 37^ 45. [25] E. Ito, K. Oka, R. Etcheberrigaray, T.J. Nelson, D.L. McPhie, B. Tofel-Grehl, G.E. Gibson, D.L. Alkon, Internal Ca2þ mobilization is altered in ¢broblasts from patients with Alzheimer disease, Proc. Natl. Acad. Sci. USA 91 (1994) 534^538. [26] G.E. Gibson, M. Vestling, H. Zhang, S. Szolosi, D. Alkon, L. Lannfelt, S. Gandy, R.F. Cowburn, Abnormalities in Alzheimer’s disease ¢broblasts bearing the APP670/671 mutation, Neurobiol. Aging 18 (1997) 573^580. [27] N. Hirashima, R. Etcheberrigaray, S. Bergamaschi, M. Racchi, F. Battaini, G. Binetti, S. Govoni, D.L. Alkon, Calcium responses in human ¢broblasts: a diagnostic molecular pro¢le for Alzheimer’s disease, Neurobiol. Aging 17 (1996) 549^ 555. [28] G.E. Gibson, V. Haroutunian, H. Zhang, L.C. Park, Q. Shi, M. Lesser, R.C. Mohs, R.K. Sheu, J.P. Blass, Mitochondrial damage in Alzheimer’s disease varies with apolipoprotein E genotype, Ann. Neurol. 48 (2000) 297^303. [29] F. Mastrogiacoma, J.G. Lindsay, L. Bettendor¡, J. Rice, S.J. Kish, Brain protein and alpha-ketoglutarate dehydrogenase complex activity in Alzheimer’s disease, Ann. Neurol. 39 (1996) 592^598. [30] K.F. Sheu, A.J. Cooper, K. Koike, M. Koike, J.G. Lindsay, J.P. Blass, Abnormality of the alpha-ketoglutarate dehydrogenase complex in ¢broblasts from familial Alzheimer’s disease, Ann. Neurol. 35 (1994) 312^318. [31] J.P. Blass, K.F. Sheu, S. Piacentini, S. Sorbi, Inherent abnormalities in oxidative metabolism in Alzheimer’s disease: interaction with vascular abnormalities, Ann. N. Y. Acad. Sci. 826 (1997) 382^385. [32] G.E. Gibson, L.C. Park, H. Zhang, S. Sorbi, N.Y. Calingasan, Oxidative stress and a key metabolic enzyme in Alzheimer brains, cultured cells, and an animal model of chronic oxidative de¢cits, Ann. N. Y. Acad. Sci. 893 (1999) 79^94. [33] B. Halliwell, J.M.C. Gutteridge. Free Radicals in Biology and Medicine, Clarendon Press, Oxford. [34] G.E. Gibson, B. Tofel-Grehl, K. Sche¡old, V.J. Cristofalo, J.P. Blass, A reproducible procedure for primary culture and subsequent maintenance of multiple Lines of human skin ¢broblasts, AGE 21 (1998) 7^14. [35] G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2þ indicators with greatly improved £uorescence properties, J. Biol. Chem. 260 (1985) 3440^3450. [36] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the prin-

BBADIS 62076 10-4-02

G.E. Gibson et al. / Biochimica et Biophysica Acta 1586 (2001) 177^189

[37] [38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

ciple of protein^dye binding, Anal. Biochem. 72 (1976) 248^ 254. B.E. Svensson, Thiols as myeloperoxidase-oxidase substrates, Biochem. J. 253 (1988) 441^449. M. Nakamura, One-electron oxidation of Trolox C (a vitamin E analogue) by peroxidases, J. Biochem. (Tokyo) 108 (1990) 245^249. M. Nakamura, One-electron oxidation of Trolox C and vitamin E by peroxidases, J. Biochem. (Tokyo) 110 (1991) 595^597. F.J. Romero, I. Ordonez, A. Arduini, E. Cadenas, The reactivity of thiols and disul¢des with di¡erent redox states of myoglobin. Redox and addition reactions and formation of thiyl radical intermediates, J. Biol. Chem. 267 (1992) 1680^ 1688. A. Puppo, M.J. Davies, The reactivity of thiol compounds with di¡erent redox states of ghaemoglobin evidence for competing reduction and addition pathways, Biochim. Biophys. Acta 1246 (1995) 74^81. U. Burner, C. Obinger, Transient-state and steady-state kinetics of the oxidation of aliphatic and aromatic thiols by horseradish peroxidase, FEBS Lett. 411 (1997) 269^274. E. Nunez Delicado, A. Sanchez Ferrer, F. Garcia Carmona, A kinetic study of the one-electron oxidation of Trolox C by the hydroperoxidase activity of lipoxygenase, Biochim. Biophys. Acta 1335 (1997) 127^134. Q. Guo, L. Packer, Ascorbate-dependent recycling of the vitamin E homologue Trolox by dihydrolipoate and glutathione in murine skin homogenates, Free Radic. Biol. Med. 29 (2000) 368^374. D.E. Feierman, A.I. Cederbaum, The e¡ect of EDTA and iron on the oxidation of hydroxyl radical scavenging agents and ethanol by rat liver microsomes, Biochem. Biophys. Res. Commun. 116 (1983) 765^770. J. Geigert, S.K. DeWitt, S.L. Neidleman, G. Lee, D.J. Dalietos, M. Moreland, DMSO is a substrate for chloroperoxidase, Biochem. Biophys. Res. Commun. 116 (1983) 82^85. G. Rothe, G. Valet, Flow cytometric analysis of respiratory burst activity in phagocytes with hydroethidine and 2P,7Pdichloro£uorescin, J. Leukoc. Biol. 47 (1990) 440^448. C. Schoneich, U. Dillinger, F. von Bruchhausen, K.D. Asmus, Oxidation of polyunsaturated fatty acids and lipids through thiyl and sulfonyl radicals reaction kinetics, and in£uence of oxygen and structure of thiyl radicals, Arch. Biochem. Biophys. 292 (1992) 456^467. C. Rota, Y.C. Fann, R.P. Mason, Phenoxyl free radical formation during the oxidation of the £uorescent dye 2P,7P-

[50]

[51]

[52] [53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

189

dichloro£uorescein by horseradish peroxidase. Possible consequences for oxidative stress measurements, J. Biol. Chem. 274 (1999) 28161^28168. B.E. Svensson, A. Graslund, G. Strom, P. Moldeus, Thiols as peroxidase substrates, Free Radic. Biol. Med. 14 (1993) 167^175. J.A. Royall, H. Ischiropoulos, Evaluation of 2P,7P-dichloro£uorescin and dihydrorhodamine 123 as £uorescent probes for intracellular H2 O2 in cultured endothelial cells, Arch. Biochem. Biophys. 302 (1993) 348^355. K. Stolze, R.P. Mason, Spin trapping artifacts in DMSO, Biochem. Biophys. Res. Commun. 143 (1987) 941^946. R.C. Scaduto Jr., Oxidation of DMSO and methanesul¢nic acid by the hydroxyl radical, Free Radic. Biol. Med. 18 (1995) 271^277. C. von Sonntag, New aspects in the free-radical chemistry of pyrimidine nucleobases, Free Radic. Res. Commun. 2 (1987) 217^224. A.K. Grover, S.E. Samson, C.M. Misquitta, Sarco(endo)plasmic reticulum Ca2þ pump isoform SERCA3 is more resistant than SERCA2b to peroxide, Am. J. Physiol. 273 (1997) C420^425. W. Paschen, J. Doutheil, Disturbances of the functioning of endoplasmic reticulum: a key mechanism underlying neuronal cell injury?, J. Cereb. Blood Flow Metab. 19 (1999) 1^18. E.V. Menshikova, V.B. Ritov, A.A. Shvedova, N. Elsayed, M.H. Karol, V.E. Kagan, Pulmonary microsomes contain a Ca2þ -transport system sensitive to oxidative stress, Biochim. Biophys. Acta 1228 (1995) 165^174. M.P. Mattson, F.M. LaFerla, S.L. Chan, M.A. Leissring, P.N. Shepel, J.D. Geiger, Calcium signaling in the ER: its role in neuronal plasticity and neurodegenerative disorders, Trends Neurosci. 23 (2000) 222^229. G.E. Gibson, L.C. Park, K.F. Sheu, J.P. Blass, N.Y. Calingasan, The alpha-ketoglutarate dehydrogenase complex in neurodegeneration, Neurochem. Int. 36 (2000) 97^112. L.C. Park, H. Zhang, K.F. Sheu, N.Y. Calingasan, B.S. Kristal, J.G. Lindsay, G.E. Gibson, Metabolic impairment induces oxidative stress, compromises in£ammatory responses and inactivates a key mitochondrial enzyme in microglia, J. Neurochem. 72 (1999) 1948^1958. K.M. Humphries, Y. Yoo, L.I. Szweda, Inhibition of NADH-linked mitochondrial respiration by 4-hydroxy-2nonenal, Biochemistry 37 (1998) 552^557. L. Tretter, V. Adam-Vizi, Inhibition of Kreb’s cycle enzymes by hydrogen peroxide: A key role, J. Neurosci. 20 (2000) 8972^8979.

BBADIS 62076 10-4-02