Overexpression of Glutathione Peroxidase ... - Wiley Online Library

2 downloads 0 Views 304KB Size Report
AЯ peptide (Borchelt et al., 1996; Duff et al., 1996;. Citron et al., 1997). Recently .... grown in a humidified incubator at 37°C in 5% CO2/90% air atmosphere. ..... Fidani L., Giuffra L., Haynes A., Irving N., James L., et al. (1991). Segregation of a ...
Journal of Neurochemistry Lippincott Williams & Wilkins, Inc., Philadelphia © 2000 International Society for Neurochemistry

Overexpression of Glutathione Peroxidase Increases the Resistance of Neuronal Cells to A␤-Mediated Neurotoxicity Martine Barkats, Ste´phanie Millecamps, Pascale Abrioux, Marie-Claude Geoffroy, and Jacques Mallet Laboratoire de Ge´ne´tique Mole´culaire de la Neurotransmission et des Processus Neurode´ge´ne´ratifs, CNRS UMR C9923, Hoˆpital de la Pitie´ Salpeˆtrie`re, Paris, France

Abstract: Senile plaques are neuropathological manifestations in Alzheimer’s disease (AD) and are composed mainly of extracellular deposits of amyloid ␤-peptide (A␤). Various data suggest that the accumulation of A␤ may contribute to neuronal degeneration and that A␤ neurotoxicity could be mediated by oxygen free radicals. Removal of free radicals by antioxidant scavengers or enzymes was found to protect neuronal cells in culture from A␤ toxicity. However, the nature of the free radicals involved is still unclear. In this study, we investigated whether the neuronal overexpression of glutathione peroxidase (GPx), the major hydrogen peroxide (H2O2)-degrading enzyme in neurons, could increase their survival in a cellular model of A␤-induced neurotoxicity. We infected pheochromocytoma (PC12) cells and rat embryonic cultured cortical neurons with an adenoviral vector encoding GPx (Ad-GPx) prior to exposure to toxic concentrations of A␤(25–35) or (1– 40). Both PC12 and cortical Ad-GPx-infected cells were significantly more resistant to A␤-induced injury. These data strengthen the hypothesis of a role of H2O2 in the mechanism of A␤ toxicity and highlight the potential of Ad-GPx to reduce A␤-induced damage to neurons. These findings may have applications in gene therapy for AD. Key Words: Gene therapy—Gene transfer—Adenovirus—␤-Amyloid—Alzheimer’s disease —Oxidative stress. J. Neurochem. 75, 1438 –1446 (2000).

observed in the brain of AD patients (Quon et al., 1991; Games et al., 1995; Hsiao et al., 1995; LaFerla et al., 1995; Moran et al., 1995; Moechars et al., 1996). Familial forms of early-onset AD have also been linked to mutations in the presenilin genes (Levy-Lahad et al., 1995; Sherrington et al., 1995). These mutations result in increased production of the 42-amino acid form of the A␤ peptide (Borchelt et al., 1996; Duff et al., 1996; Citron et al., 1997). Recently, immunization with A␤ has been found to attenuate the AD-like pathology in a transgenic mouse overexpressing a mutant human APP (Schenk et al., 1999), confirming the central role of A␤ in AD neuropathology. Finally, A␤ has been found to be directly toxic to neuronal cells, both in vitro (Yankner et al., 1990) and in vivo (Geula et al., 1998). However, the mechanism of A␤ toxicity and the cellular pathways involved in neuronal degeneration are not yet fully understood. The involvement of oxygen free radicals in A␤-induced neuronal death has been suggested in a large number of studies (Behl, 1999). First, there is evidence that in aqueous solution, the peptide is fragmented and generates free radicals (Hensley et al., 1994). Reciprocally, A␤ aggregation and amyloidogenicity are catalyzed in vitro by oxidation (Dyrks et al., 1992). Furthermore, a number of antioxidant scavengers decreased A␤ toxicity in culture (Manelli and Puttfarcken, 1995; Bruce et al., 1996; Cafe et al., 1996; Zhou

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder leading to dementia in the elderly. It is characterized by anatomopathological changes in brain including neuronal death, senile plaques, and neurofibrillary tangles (Tomlinson et al., 1970). The amyloid ␤-peptide (A␤) is the main constituent of senile plaques in the brain of AD patients. There is evidence that A␤ is involved in AD pathogenesis (Selkoe, 1991; Yankner, 1996): Mutations in the A␤-peptide precursor protein (APP) have been detected in cases of early-onset familial AD (Chartier-Harlin et al., 1991; Goate et al., 1991; Murrell et al., 1991), and transgenic mice overexpressing APP mutations or cytoplasmic A␤ develop neuropathological and/or behavioral abnormalities resembling those

Received February 28, 2000; revised manuscript received May 3, 2000; accepted May 16, 2000. Address correspondence and reprint requests to Dr. J. Mallet at UMR CNRS C9923, LGN, Bat CERVI, Hoˆpital de la Pitie´ Salpeˆtrie`re, 83 Bd de l’hoˆpital, 75013 Paris, France. E-mail: [email protected] Abbreviations used: A␤, amyloid ␤-peptide; AD, Alzheimer’s disease; Ad-␤gal, adenoviral vector encoding ␤-galactosidase; Ad-GPx, adenoviral vector encoding glutathione peroxidase; APP, amyloid ␤-peptide precursor protein; DMEM, Dulbecco’s modified Eagle’s medium; FDA, fluorescein diacetate; ␤-gal, ␤-galactosidase; X-gal, 5-bromo-4-chloro-3-indolyl-␤-D-galactoside; GPx, glutathione peroxidase; MOI, multiplicity of infection; MTT, 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide; PBS, phosphate-buffered saline; PI, propidium iodide; RSV, Rous sarcoma virus; SOD, superoxide dismutase.

1438

GPx ADENOVIRUS AGAINST A␤ NEUROTOXICITY et al., 1996). Hydrogen peroxide (H2O2) was initially proposed as a mediator of A␤ neurotoxicity (Behl et al., 1994), and transfected B12 cells overexpressing the H2O2-degrading enzymes catalase and glutathione peroxidase (GPx) were more resistant to A␤ injury (Sagara et al., 1996). In addition, cell clones selected for their resistance to A␤ displayed high intracellular levels of these antioxidative enzymes (Sagara et al., 1996). The involvement of free radicals in A␤-induced neurotoxicity was more recently demonstrated in PC6 pheochromocytoma cells overexpressing the presenilin-1 mutation. These cells expressing presenilin-1 mutations exhibited increased superoxide production, nitrotyrosine formation, and lipid peroxidation following exposure to A␤ (Guo et al., 1997, 1999). In the present study, we used a gene transfer strategy to increase the antioxidant potential of neuronal cells prior to exposure to neurotoxic fragments of A␤. More precisely, we evaluated whether intracellular overexpression of GPx could increase resistance of PC12 pheochromocytoma cells and rat embryonic cortical neurons to A␤ toxicity. As adenovirus is an efficient vector for transferring genes into postmitotic nervous cells both in vitro and in vivo (Akli et al., 1993; Bajocchi et al., 1993; Davidson et al., 1993; Le Gal La Salle et al., 1993), increased intracellular levels of GPx were obtained using this vector. In this study, we found that transduction of PC12 and cortical neurons in culture with a recombinant adenovirus vector expressing GPx (Ad-GPx) allowed the overproduction of the enzyme and rendered the cells more resistant to A␤ toxicity by increasing intracellular levels of peroxidase activity. Adenovirus-mediated transduction of neuronal cells with antioxidative genes is thus an attractive strategy for improving their antioxidant defense mechanism and evaluating the implication of oxidative stress in cellular or animal models of A␤-induced injury. MATERIALS AND METHODS Chemicals RNAzol-B was obtained from Biotecx Laboratories. RNA preparations were reverse-transcribed using a commercial kit from Promega. GPx activity was determined using a commercial kit from Randox Laboratories. The Bio-Rad assay for protein determination was from Bio-Rad Laboratories. Serum for cell culture was from Roche Diagnostics. Culture medium was from Life Technologies. The 5-bromo-4-chloro-3-indolyl␤-D-galactoside (X-gal) chromogen was from Euromedex. Hormones for supplementation of the cortical cell culture medium, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), fluorescein diacetate (FDA), propidium iodide (PI), and other chemicals were from Sigma.

Adenoviral constructs The bovine GPx cDNA (generously given by Dr. P. Amstad, ISREC, Switzerland) and the SV40 polyadenylation sequence [poly(A)] were first inserted into the pBluescript SK ⫹/⫺ plasmid. The 1.1-kbp SacI/KpnI fragment containing the bovine GPx cDNA and the SV40 poly(A) was excised, rendered

1439

blunt-ended, and inserted into the EcoRV site of a shuttle vector downstream from the Rous sarcoma virus (RSV) promoter. The resulting plasmid (pGPx) contained additional adenoviral sequences including the inverted terminal repeat, the encapsidation sequence, and the protein IX adenoviral sequence allowing homologous recombination. To generate the recombinant adenovirus, transformed human kidney 293 cells were co-transfected with pGPx and the large ClaI fragment of a plasmid containing the complete adenovirus genome (pAd-5) using the calcium phosphate–DNA precipitation method. The recombinant GPx adenovirus was isolated from a single plaque using restriction enzyme analysis and PCR. Viral stock was prepared after expansion in the 293 cell line (from human kidney) and purified by double CsCl gradient ultracentrifugation. Virus titers were determined by measuring optical density at 260 nm. A similar strategy was used to obtain control vectors: the recombinant adenovirus expressing ␤-galactosidase (␤-gal) under the control of the RSV promoter (Ad-␤gal) (Stratford-Perricaudet et al., 1992) and the empty adenovirus (with a cytomegalovirus promoter, without transgene; a gift from Dr. Jean-Franc¸ois Dedieu, Aventis, France).

RT-PCR analysis of specific adenoviral GPx (AdGPx) transcripts

Total RNA was isolated using RNAzol-B from 1 ⫻ 106 uninfected 293 cells and cells infected with a multiplicity of infection (MOI) of 100 and 200 of Ad-GPx. The RNA preparations were reverse-transcribed, and RT-PCR was performed using primers specific for the Ad-GPx mRNA transcripts. The sense oligonucleotide was chosen in a part of the bovine GPx cDNA displaying a low similarity with the rat cDNA sequence (5⬘-TGCTCTGGATTCGGAAACGGATACC-3⬘); the antisense nucleotide was chosen in a pBluescript SK polylinker sequence between the bovine GPx cDNA and the SV40 poly(A) (3⬘-GCTTGATATGCAATTCCTGCAGCCC-5⬘). One half of the RNA preparation was used in a control reaction from which the reverse transcriptase was omitted.

GPx enzymatic activity GPx activity was measured using a method based on that of Paglia and Valentine (1967). GPx catalyses the oxidation of glutathione by cumene hydroperoxide. In the presence of glutathione reductase and NADPH, the oxidized glutathione is converted to the reduced form with a concomitant oxidation of NADPH. The resulting decrease in absorbance at 340 nm can be measured in a spectrophotometer (Beckman). Cells were homogenized in cold 0.05 M potassium phosphate buffer (pH 7.4) containing 1 mM EGTA and centrifuged at 13,000 g for 10 min. The supernatants were added to the incubation mixture consisting of 4 mM glutathione, 0.28 mM NADPH, and 0.5 U/L glutathione reductase in 0.05 M phosphate buffer (pH 7.2). The reaction was initiated by the addition of 0.18 mM cumene hydroperoxide. One unit of GPx activity was defined as the amount required to oxidize 1 ␮mol of NADPH/min, based on the NADPH molar absorptivity of 6.22 ⫻ 10⫺6. GPx activity was expressed relatively to the amount of proteins (in milligrams) in cell extracts determined by the Bio-Rad assay.

Cell line and primary culture Rat pheochromocytoma (PC12) cells were cultivated in RPMI 1640 medium containing 10% horse serum and 5% fetal calf serum. Primary culture cortical cells were derived from embryonic day 17 Sprague–Dawley rats. Cortical tissue was dissected and rinsed in phosphate-buffered saline (PBS) containing 0.6% glucose. The PBS– glucose solution was then

J. Neurochem., Vol. 75, No. 4, 2000

1440

M. BARKATS ET AL.

removed and replaced with pyruvate-free Dulbecco’s modified Eagle’s medium (DMEM) containing 4.5 g/L glucose and 0.58 g/L L-glutamine. After mechanical dissociation of the tissue in DMEM, the cell solution was centrifuged for 10 min at 800 g. The supernatant was removed, and the cell pellet was resuspended in a serum-free culture medium consisting of DMEM supplemented with 100 ␮g/ml transferrin, 25 ␮g/ml insulin, 10 ␮g/ml putrescin, 5 ng/ml sodium selenite, and 6.3 ng/ml progesterone (all from Sigma). Viable cells were counted using trypan blue cell exclusion, plated at a density of 100,000 cells/cm2 surface area on polyornithine-plated cell dishes, and grown in a humidified incubator at 37°C in 5% CO2/90% air atmosphere.

Adenoviral cell infection The day of plating or after 1 day in culture (for PC12 or cortical cells, respectively), the cell medium was removed. The serum-free viral solution was then added to the cells, and incubation was continued for 45 min at 37°C. A new medium was then added to the cultures.

X-gal histochemistry Cells were fixed for 10 min in 4% paraformaldehyde in PBS. ␤-Gal activity was detected after incubation of the cells for 2 h at 37°C in an X-gal solution consisting of 4 mM potassium ferricyanide, 4 mM potassium ferrocyanide, 4 mM MgCl2, and 0.4 mg/ml chromogen X-gal in PBS.

A␤ toxicity

The peptides A␤(25–35) and A␤(1– 40) were dissolved as 1 mM stock in PBS and added in the culture medium to final concentrations of 1, 10, or 100 ␮M. Controls were treated with PBS. To assess cell viability after A␤ exposure, two assays were utilized: the modified MTT assay (Hansen et al., 1989) and a cell count method after intravital staining of the culture using FDA and PI (Didier et al., 1990). For the MTT assay, 30 ␮l of a 4 mg/ml MTT stock in PBS was added to each well containing 300 ␮l of culture medium, and incubation was continued for 4 h. A solution of propanolol-2 containing 8% 1 M HCl was added to the cultures, and cell dishes were agitated for 10 min until complete dissolution of the formazan blue crystals. Absorption values were immediately determined at 540 nm. For intravital staining, cells were washed with Locke’s solution (154 mM NaCl, 5.6 mM KCl, 1 mM MgCl2, 2.3 mM CaCl2, 5.6 mM D-glucose, and 8.6 mM HEPES, pH 7.4) and incubated for 5 min at 37°C in the Locke’s solution containing 15 ␮g/ml FDA and 15 ␮g/ml PI. The medium was then replaced with fresh Locke’s solution, and cultures were immediately examined under a fluorescence microscope at 488 nm (FDA) and 514 nm (PI). Viable and injured cells were counted in three representative fields per well. The percentage of viable cells was scored as the FDA/(PI ⫹ FDA) ratio.

RESULTS A␤ toxicity We determined the optimal conditions for obtaining a significant A␤ toxicity in cultured rat cortical neurons. Cells were exposed for 1–3 days to 10 ␮M A␤(25–35), the biologically active fragment of A␤ (Yankner et al., 1990), 4 days after plating. This A␤ concentration has been reported to have a significant toxic effect on neuronal survival (Behl et al., 1994; Shearman et al., 1995; Prehn et al., 1996). In these conditions, A␤(25–35) caused a significant reduction in cell viability, as reJ. Neurochem., Vol. 75, No. 4, 2000

FIG. 1. Cytotoxic response of embryonic cortical rat cells to A␤. A: Cells were exposed to 10 ␮M A␤(25–35) for 24, 48, or 72 h. Cell viability was assayed using the MTT cell metabolism assay. B: Cells were exposed to 10 ␮M A␤(25–35) and 10 ␮M of the reverse peptide A␤(35–25) for 48 h. Data are means ⫾ SEM for triplicate determinations, and differences between groups were statistically analyzed using the Student’s t test (**p ⬍ 0.01, ***p ⬍ 0.001). OD units, optical density units.

vealed by the inhibition of the intracellular reduction of MTT to insoluble product. A time-dependent decrease in cell viability was observed after exposure to A␤ following increasing exposure time, and a 2-day exposure period to 10 ␮M A␤(25–35) was sufficient enough to kill ⬎50% of primary cortical cells (Fig. 1A). In contrast, incubation with the reverse peptide A␤(35–25) was not neurotoxic (Fig. 1B). It was demonstrated that exposure of cells to 10 ␮M A␤(1– 40) fragment also induced degeneration by a mechanism similar to that of A␤(25– 35) (Yankner et al., 1990; Pike et al., 1993). Similar conditions of A␤ toxicity induced a significant degeneration of PC12 cells (data not shown). Construction of Ad-GPx and production of GPx mRNA and protein in Ad-GPx-infected cells To investigate whether overexpression of GPx protected cells from A␤-induced neurotoxicity, we constructed a recombinant adenovirus allowing transfer of the gene encoding GPx into neuronal cells. This adenoviral vector was derived from the type 5 adenovirus and deleted in the early 1 and early 3 sequences; the bovine

GPx ADENOVIRUS AGAINST A␤ NEUROTOXICITY

FIG. 2. Schematic map of the replication-defective recombinant adenovirus containing the bovine GPx cDNA under the transcriptional control of the long terminal repeat of the Rous sarcoma virus (LTR-RSV) promoter. ITR, inverted terminal repeat; ⌿, encapsidation sequence; SV40 polyA, polyadenylation sequence of the simian virus; PIX, protein IX (allowing the homologous recombination).

GPx cDNA was inserted downstream from the RSV promoter (Fig. 2). To determine if the recombinant vector was able to direct the intracellular production of a functional GPx protein, we first infected 293 cells with 100 and 200 MOI of Ad-GPx. The synthesis of exogenous GPx mRNA in Ad-GPx-infected cells was investigated 48 h after infection by RT of total RNA followed by PCR using primers specific for the GPx transgene. A single band of the expected size was detected only in cells that were infected with Ad-GPx, at both 100 and 200 MOI, the band intensity being greater at the higher MOI (Fig. 3A). No signal was detected in the control reactions in which the reverse transcriptase was omitted or in which RNA cell extracts were replaced by water, so the GPx band did not correspond to DNA contaminating the RNA preparation. The production of a functional GPx protein within Ad-GPx-infected cells was tested by measuring the peroxidase activity in 293 cells infected with 100 MOI of Ad-GPx and Ad-␤gal vectors and in noninfected cells. Cells infected with Ad-GPx showed a 3- and a 1.7-fold higher GPx activity than Ad-␤galinfected and noninfected cells, respectively (Fig. 3B). Differences between the Ad-GPx group and the others were statistically significant. Adenoviral infectability and cytotoxicity in cortical cultures To test whether PC12 and rat embryonic cortical cells transduced with the adenovirus encoding GPx were protected against A␤ toxicity, we determined the optimal conditions for infecting these cells with recombinant replication-defective adenoviruses. A series of concentrations of Ad-␤gal (0 –200 MOI) were added to the cells 1 day after plating. Three days later, ␤-gal-expressing cortical cells were counted after incubation with the chromogenic substrate X-gal. Numerous cortical cells were intensely stained, even at the lowest MOI (Fig. 4A). The percentage of transduced cells increased with the concentration of the viral solution from 25 to 200 MOI, with a maximum of 66% ␤-gal-expressing cells at 200 MOI (Fig. 4B). Although viral toxicity was apparent for the MOI of 200, this concentration was chosen as a compromise between infection rate and toxicity. For PC12 cells, no significant viral toxicity was apparent for this MOI, as revealed by the MTT cell metabolism assay, which measures the level of MTT reduction as an index of viability (data not shown).

1441

More resistance of Ad-GPx-infected cells to A␤ toxicity Infection of PC12 cells and cultured rat cortical neurons with optimal concentrations of the recombinant adenovirus encoding GPx was used to produce overexpression of the antioxidative protein in the cells (Table 1). PC12 cells infected with 200 MOI of Ad-GPx displayed a significantly higher GPx activity (about fivefold) than noninfected cells. Similarly, infection of cultured rat cortical neurons with 200 MOI of Ad-GPx led to a significant increase of GPx activity (threefold) compared with control noninfected cells. No increased GPx activity was detected in cells infected with a control empty vector. We tested whether the cells transduced with Ad-GPx were more resistant to A␤ toxicity. On day 1 (24 h after

FIG. 3. Production of GPx mRNA and protein in Ad-GPx-infected cells. A: RT-PCR analysis of bovine GPx mRNAs in 293 cells using primers specific for the adenoviral transcripts. The production of adenoviral GPx transcripts was detected in cells infected with 100 and 200 MOI of Ad-GPx (lanes 3 and 5, respectively). No signal was obtained in noninfected cells (lane 1) either in control reactions in which the reverse transcriptase was omitted (lanes 2, 4, 6 for noninfected, 100 MOI or 200 MOI Ad-GPx-infected cells) or in reactions in which RNA cell extracts were replaced by water (lane 7). B: GPx enzymatic activity in 293 cells either not infected or infected with Ad-GPx or Ad-␤gal. GPx activity was assayed according to Paglia and Valentine (1967). Activity units are given in micromoles of NADPH oxidized per minute and per milligram of protein. GPx activity was increased in Ad-GPx-infected cells compared with noninfected (⫻1.7) or Ad-␤gal-infected (⫻3) cells. Data are means ⫾ SEM for triplicate determinations. Differences between groups were analyzed using the Student’s t test (*p ⬍ 0.05, **p ⬍ 0.01).

J. Neurochem., Vol. 75, No. 4, 2000

1442

M. BARKATS ET AL.

FIG. 4. Adenoviral infectability and cytotoxicity in rat cortical primary cultures. A: Adenoviral infectability and cytotoxicity in embryonic cortical cells infected with increasing concentrations of Ad-␤gal. Cell viability was assessed by visual counting after X-gal cytochemistry. Results are expressed as percentage of transduced cells (X-gal-positive) relative to the total number of cells. Data are means ⫾ SEM for five separate wells. B: X-gal staining of embryonic cortical cells infected with 200 MOI of Ad-␤gal for 48 h.

plating), cultured cortical neurons and PC12 cells were exposed to Ad-GPx, to a control adenovirus (Ad-␤gal or empty vector), or to vehicle, 2 days prior to exposure to 10 ␮M A␤ for 2 days. A␤ toxicity was first assessed in Ad-GPx-infected PC12 cells using the MTT assay. The survival rate of Ad-GPx-infected PC12 cells was significantly increased compared with control cells infected with a control vector or noninfected (Fig. 5). In these cells, treatment with 200 MOI of Ad-GPx produced a significant increase in cell viability up to 122 and 125% of the Ad-␤gal (Fig. 5A) and empty virus (Fig. 5B) control groups, respectively. The neurotoxic effect of A␤ was further evaluated in cultured cortical neurons infected with Ad-GPx. In these experiments, cell viability was assessed by visual counting after intravital staining with FDA and PI fluorochromes. The number of viable cells was quantified by J. Neurochem., Vol. 75, No. 4, 2000

assessment of FDA deesterification, which produces a green–yellow fluorescence only in living cells. Neuronal injury facilitates neuronal PI uptake, and its interaction with DNA produces red fluorescence in dying cells. As observed in PC12 cells, infection of cultured cortical neurons with Ad-GPx increased their resistance to A␤ neurotoxicity (Fig. 6). Ad-GPx-infected cortical neurons exposed to 10 ␮M A␤(25–35) showed an increase in viability to 128% of control cultures infected with Ad-␤gal. DISCUSSION AD is a progressive and fatal neurodegenerative disease, for which no efficient treatment is available. Accumulating evidence implicates A␤ in the pathogenesis of AD (Selkoe, 1991; Yankner, 1996). As oxidative stress has been suggested to be involved in A␤-mediated neu-

GPx ADENOVIRUS AGAINST A␤ NEUROTOXICITY

1443

TABLE 1. GPx activity in PC12 and cultured cortical neurons infected with Ad-GPx Infection

PC12 Cortical neurons

Noninfected

Ad-GPx

Increase

257 ⫾ 88 579 ⫾ 114

1,336 ⫾ 257a 1,722 ⫾ 350a

5⫻ 3⫻

Cells were infected 24 h after plating at a MOI of 200. GPx activity was assayed according to Paglia and Valentine (1967). Activity units are given in micromoles of NADPH oxidized per minute and per milligram of protein. Data are means ⫾ SEM for triplicate determinations. a p ⬍ 0.01, differences between groups analyzed with the Student’s t test.

rodegeneration (Hensley et al., 1996), the transfer of genes encoding antioxidative proteins into neuronal cells could be valuable both for elucidating the role of free radicals in A␤ toxicity and for possible AD gene therapy.

FIG. 5. Cytotoxic response of Ad-GPx-infected PC12 cells to A␤. PC12 cells were infected the day of plating at a MOI of 200. After 4 days, cells were exposed to 10 ␮M A␤(1– 40) for 48 h. Cell viability was assayed using MTT, and results are expressed as percent MTT reduction relative to untreated cells. The effect of Ad-GPx was compared with that of a control Ad-␤gal (A) and control empty adenovirus (B). Data are means ⫾ SEM of six values (from two independent experiments). *p ⬍ 0.05, ***p ⬍ 0.001, ****p ⬍ 0.0001 (Student’s t test).

FIG. 6. Cytotoxic response to A␤ of Ad-GPx-infected rat cultured cortical neurons. Cells were infected 24 h after plating at a MOI of 200. Forty-eight hours after infection, cells were exposed to 10 ␮M A␤(25–35) for 48 h. Cell viability was assessed using the FDA/PI cell-counting assay, and results are expressed as percent viability relative to untreated cells. Data are means ⫾ SEM of nine values (from three independent experiments). **p ⫽ 0.01, different from control (Ad-␤gal-infected, A␤-treated) cells (Student’s t test).

In this study, we constructed an “antioxidative” adenoviral vector directing overproduction of GPx mRNA and protein in transduced cells. Both PC12 and cultured cortical neurons infected with Ad-GPx showed an increased resistance to A␤ toxicity, although the protection was not complete. After infection with Ad-␤gal, ⬃66% of the cortical cells were efficiently transduced, as revealed by counting the number of X-gal-positive cells in the culture (Fig. 4A). In the experiments of neuroprotection, only Ad-GPx-transduced cells (i.e., 66% of the cells) contributed to the observed 28% increase in the cortical culture viability (Fig. 6). In these cultures, A␤ toxicity was 31%, suggesting that the majority of the Ad-GPx-transduced cells were protected from A␤ toxicity. Similar results of neuroprotection were reported in a study in which B12 cells co-transfected with catalase and GPx cDNAs were not completely resistant to A␤ (Sagara et al., 1996). In this study, it is suggested that a further increase in antioxidative enzyme activity would result in complete neuroprotection of cortical cells. A possible protective effect of subtoxic levels of treatment with the adenoviral vectors could result from a stress response increasing resistance of cells to A␤ peptide toxicity. However, the fact that control viruses (adenovirus expressing ␤-gal or empty adenovirus) did not trigger any protective reaction against A␤ toxicity strongly argues against the occurrence of an adaptive protective response to viral infection. Early findings suggested that the cytotoxic action of A␤ on neurons resulted from an increase in H2O2 accumulation and lipid peroxidation (Behl et al., 1994). Exogenous addition of the antioxidant enzyme catalase and J. Neurochem., Vol. 75, No. 4, 2000

1444

M. BARKATS ET AL.

of a large number of antioxidants was found to protect cells from A␤ toxicity. In addition, cell clones that were selected for resistance to A␤ displayed highly elevated mRNA and protein levels of catalase and GPx (Sagara et al., 1996). However, other data either do not support a direct oxidative mechanism of A␤ toxicity (Zhang et al., 1996) or suggest that although A␤ can increase oxidative stress in neurons, free radical-induced degeneration is not an essential component of A␤-induced cell death (Pike et al., 1997). In one of these studies, addition of catalase was demonstrated to effectively protect neurons in culture against A␤-mediated toxicity, but by a direct interaction of catalase with A␤ itself (resulting from effects on A␤ conformation and aggregation) rather than by a mechanism involving the capacity of the enzyme to scavenge H2O2 (Lockhart et al., 1994; Zhang et al., 1996). In cells subjected to toxic A␤ concentrations, the observed neuroprotective potential of the H2O2-degrading enzyme GPx suggests that H2O2 is a mediator of A␤ neurotoxicity. In our model, the gene encoding GPx was introduced into nervous cells using an adenoviral vector that directed the intracellular production of the protein. As GPx is not secretable, its direct interaction with A␤ can be excluded. As H2O2 is freely diffusible through the cell membrane, the A␤-induced extracellular H2O2 concentration would effectively be lowered by intracellular GPx after H2O2 diffusion across the cell membrane. Our data therefore provide additional evidence for the involvement of reactive oxygen species in A␤-mediated toxicity. Various experimental factors have also been suggested to contribute to the controversial results in the literature concerning the effect of antioxidants on A␤ toxicity. One of these variables is the length of time the cultures are maintained in vitro before experimentation (Pike et al., 1997). Antioxidants may be effective only in cells maintained in culture for long periods (⬎7 days), which may increase their basal level of oxidative stress above that of those cultured for shorter periods. Such resistance of mature cultures to A␤-induced toxicity has been observed, for example, in organotypic hippocampal cultures (Bruce et al., 1996). In our study, A␤ toxicity and GPx-mediated neuroprotection were reported in PC12 and cultured cortical cells that were exposed to A␤ only 4 days after plating. The observed protective effect of GPx is therefore not consequent to a time-dependent accumulation of oxidative stress in the culture or to neuronal differentiation factors in addition to increased oxidative stress, as previously suggested (Pike et al., 1997). The A␤-mediated increase in the H2O2 concentration may reflect the overproduction of the superoxide radical that is subsequently reduced to H2O2 by superoxide dismutase (SOD). Superoxide has not been clearly demonstrated to have a direct role in A␤ neurotoxicity in neuronal cultures. A␤ treatment did not increase the production of superoxide anions in cultured rat hippocampal neurons, as monitored with the fluorescent probe hydroethidine (Prehn et al., 1996). In this study, J. Neurochem., Vol. 75, No. 4, 2000

further overexpression of the human copper–zinc superoxide dismutase (SOD1) using a recombinant adenoviral vector failed to protect the cells against A␤ insult. In contrast, other studies report a reduction of apoptotic features and oxidative stress in PC6 cells transfected with an expression vector containing the human MnSOD cDNA prior to exposure to A␤, suggesting a primary role for superoxide in A␤ injury (Keller et al., 1998). However, levels of GPx activity were significantly increased in PC6 cells overexpressing MnSOD, probably by a compensatory mechanism for the increased H2O2 production. The increased removal of H2O2 consequent to the increased GPx activity, rather than a direct action of MnSOD, may therefore account for the resistance of PC6 –MnSOD cells to A␤ insult. Our results, showing a large protection of cortical cultured neurons infected with Ad-GPx, do not support superoxide radicals having a crucial role in the neuronal cell death induced by A␤ insult. Rather, they suggest a role of H2O2 in this neurotoxicity. In this study, we describe, for the first time, a gene transfer system based on the use of recombinant viral vectors expressing GPx that is able to render neuronal cells in primary culture more resistant to A␤-induced toxicity by increasing their peroxidase activity. The development of these recombinant adenoviral vectors not only confirms previous information on the mechanism of A␤ neurotoxicity but may have applications in gene therapy for AD and for other neurodegenerative diseases in which oxidative stress may be involved, such as Parkinson’s disease (Jenner, 1998) and amyotrophic lateral sclerosis (Rosen et al., 1993). Because of the global nature of AD, the development of viral vectors encoding secreted forms of antioxidative enzymes is of considerable interest for the treatment of this disease. In addition to its neurotoxic role, A␤ has been reported to induce vascular dysfunction, which may be a contributing factor to AD (Thomas et al., 1996). Recently, an important role of free radicals was demonstrated in the cerebral endothelial dysfunction observed in mice overexpressing APP (Iadecola et al., 1999). As adenoviruses are efficient vectors for in vivo gene transfer, “antioxidant” adenoviruses encoding SOD (Jordan et al., 1995; Barkats et al., 1996, 1997) or GPx may offer new opportunities for exploring the pathophysiology of AD in these transgenic animal models and elucidate the role of oxidative stress in this neurodegenerative disease. Acknowledgment: We are grateful to Dr. P. Amstad for providing us with the GPx cDNA and to Dr. J. F. Dedieu for providing the control empty adenovirus. We also thank J. J. Robert for contribution to the preparation of Ad-GPx, C. Serguera for his critical reading of the manuscript, and G. Pitiot and P. Ravassard for helpful advice. M. Barkats was supported by fellowships from the 4th EEC Biotech program and the Foundation France Alzheimer. S. Millecamps was supported by the Ministe`re de l’Education Nationale, de l’Enseignement Supe´rieur et de la Recherche, and the Association Franc¸aise contre les Myopathies. This work was supported by the EEC Biotech program, the Centre National de la Recherche Scien-

GPx ADENOVIRUS AGAINST A␤ NEUROTOXICITY tifique, Aventis, and the Association Franc¸aise Retinis Pigmentosa.

REFERENCES Akli S., Caillaud C., Vigne E., Stratford-Perricaudet L. D., Poenaru L., Perricaudet M., Kahn A., and Peschanski M. R. (1993) Transfer of a foreign gene into the brain using adenovirus vectors. Nat. Genet. 3, 224 –228. Bajocchi G., Feldman S. H., Crystal R. G., and Mastrangeli A. (1993) Direct in vivo gene transfer to ependymal cells in the central nervous system using recombinant adenovirus vectors. Nat. Genet. 3, 229 –234. Barkats M., Bemelmans A. P., Geoffroy M. C., Robert J. J., Loquet I., Horellou P., Revah F., and Mallet J. (1996) An adenovirus encoding CuZnSOD protects cultured striatal neurons against glutamate toxicity. Neuroreport 7, 497–501. Barkats M., Nakao N., Grasbon-Frodl E. M., Bilang-Bleuel A., Revah F., Mallet J., and Brundin P. (1997) Intrastriatal grafts of embryonic mesencephalic rat neurons genetically modified using an adenovirus encoding human Cu/Zn superoxide dismutase. Neuroscience 78, 703–713. Behl C. (1999) Alzheimer’s disease and oxidative stress: implications for novel therapeutic approaches. Prog. Neurobiol. 57, 301–323. Behl C., Davis J. B., Lesley R., and Schubert D. (1994) Hydrogen peroxide mediates amyloid beta protein toxicity. Cell 77, 817– 827. Borchelt D. R., Davis J., Fischer M., Lee M. K., Slunt H. H., Ratovitsky T., Regard J., Copeland N. G., Jenkins N. A., Sisodia S. S., and Price D. L. (1996) Familial Alzheimer’s disease-linked presenilin 1 variants elevate A beta 1– 42/1– 40 ratio in vitro and in vivo. Neuron 17, 1005–1013. Bruce A. J., Malfroy B., and Baudry M. (1996) Beta-amyloid toxicity in organotypic hippocampal cultures: protection by EUK-8, a synthetic catalytic free radical scavenger. Proc. Natl. Acad. Sci. USA 93, 2312–2316. Cafe C., Torri C., Bertorelli L., Angeretti N., Lucca E., Forloni G., and Marzatico F. (1996) Oxidative stress after acute and chronic application of beta-amyloid fragment 25–35 in cortical cultures. Neurosci. Lett. 203, 61– 65. Chartier-Harlin M. C., Crawford F., Houlden H., Warren A., Hughes D., Fidani L., Goate A., Rossor M., Roques P., Hardy J., et al. (1991) Early-onset Alzheimer’s disease caused by mutations at codon 717 of the beta-amyloid precursor protein gene. Nature 353, 844 – 846. Citron M., Westaway D., Xia W., Carlson G., Diehl T., Levesque G., Johnson-Wood K., Lee M., Seubert P., Davis A., Kholodenko D., Motter R., Sherrington R., Perry B., Yao H., Strome R., Lieberburg I., Rommens J., Kim S., Schenk D., Fraser P., St. George Hyslop P., and Selkoe D. J. (1997) Mutant presenilins of Alzheimer’s disease increase production of 42-residue amyloid betaprotein in both transfected cells and transgenic mice. Nat. Med. 3, 67–72. Davidson B. L., Allen E. D., Kozarsky K. F., Wilson J. M., and Roessler B. J. (1993) A model system for in vivo gene transfer into the central nervous system using an adenoviral vector. Nat. Genet. 3, 219 –223. Didier M., Heaulme M., Soubrie P., Bockaert J., and Pin J. P. (1990) Rapid, sensitive, and simple method for quantification of both neurotoxic and neurotrophic effects of NMDA on cultured cerebellar granule cells. J. Neurosci. Res. 27, 25–35. Duff K., Eckman C., Zehr C., Yu X., Prada C. M., Perez-Tur J., Hutton M., Buee L., Harigaya Y., Yager D., Morgan D., Gordon M. N., Holcomb L., Refolo L., Zenk B., Hardy J., and Younkin S. (1996) Increased amyloid-beta42(43) in brains of mice expressing mutant presenilin 1. Nature 383, 710 –713. Dyrks T., Dyrks E., Hartmann T., Masters C., and Beyreuther K. (1992) Amyloidogenicity of beta A4 and beta A4-bearing amyloid protein precursor fragments by metal-catalyzed oxidation. J. Biol. Chem. 267, 18210 –18217. Games D., Adams D., Alessandrini R., Barbour R., Berthelette P., Blackwell C., Carr T., Clemens J., Donaldson T., Gillespie F.,

1445

et al. (1995) Alzheimer-type neuropathology in transgenic mice overexpressing V717F beta-amyloid precursor protein. Nature 373, 523–527. Geula C., Wu C. K., Saroff D., Lorenzo A., Yuan M., and Yankner B. A. (1998) Aging renders the brain vulnerable to amyloid beta-protein neurotoxicity. Nat. Med. 4, 827– 831. Goate A., Chartier-Harlin M. C., Mullan M., Brown J., Crawford F., Fidani L., Giuffra L., Haynes A., Irving N., James L., et al. (1991) Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer’s disease. Nature 349, 704 – 706. Guo Q., Sopher B. L., Furukawa K., Pham D. G., Robinson N., Martin G. M., and Mattson M. P. (1997) Alzheimer’s presenilin mutation sensitizes neural cells to apoptosis induced by trophic factor withdrawal and amyloid beta-peptide: involvement of calcium and oxyradicals. J. Neurosci. 17, 4212– 4222. Guo Q., Fu W., Holtsberg F. W., Steiner S. M., and Mattson M. P. (1999) Superoxide mediates the cell-death-enhancing action of presenilin-1 mutations. J. Neurosci. Res. 56, 457– 470. Hansen M. B., Nielsen S. E., and Berg K. (1989) Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. Immunol. Methods 119, 203–210. Hensley K., Carney J. M., Mattson M. P., Aksenova M., Harris M., Wu J. F., Floyd R. A., and Butterfield D. A. (1994) A model for beta-amyloid aggregation and neurotoxicity based on free radical generation by the peptide: relevance to Alzheimer disease. Proc. Natl. Acad. Sci. USA 91, 3270 –3274. Hensley K., Butterfield D. A., Hall N., Cole P., Subramaniam R., Mark R., Mattson M. P., Markesbery W. R., Harris M. E., Aksenov M., et al. (1996) Reactive oxygen species as causal agents in the neurotoxicity of the Alzheimer’s disease-associated amyloid beta peptide. Ann. N.Y. Acad. Sci. 786, 120 –134. Hsiao K. K., Borchelt D. R., Olson K., Johannsdottir R., Kitt C., Yunis W., Xu S., Eckman C., Younkin S., Price D., et al. (1995) Age-related CNS disorder and early death in transgenic FVB/N mice overexpressing Alzheimer amyloid precursor proteins. Neuron 15, 1203–1218. Iadecola C., Zhang F., Niwa K., Eckman C., Turner S. K., Fischer E., Younkin S., Borchelt D. R., Hsiao K. K., and Carlson G. A. (1999) SOD1 rescues cerebral endothelial dysfunction in mice overexpressing amyloid precursor protein. Nat. Neurosci. 2, 157–161. Jenner P. (1998) Oxidative mechanisms in nigral cell death in Parkinson’s disease. Mov. Disord. 13 (Suppl. 1), 24 –34. Jordan J., Ghadge G. D., Prehn J. H., Toth P. T., Roos R. P., and Miller R. J. (1995) Expression of human copper/zinc-superoxide dismutase inhibits the death of rat sympathetic neurons caused by withdrawal of nerve growth factor. Mol. Pharmacol. 47, 1095– 1100. Keller J. N., Guo Q., Holtsberg F. W., Bruce-Keller A. J., and Mattson M. P. (1998) Increased sensitivity to mitochondrial toxin-induced apoptosis in neural cells expressing mutant presenilin-1 is linked to perturbed calcium homeostasis and enhanced oxyradical production. J. Neurosci. 18, 4439 – 4450. LaFerla F. M., Tinkle B. T., Bieberich C. J., Haudenschild C. C., and Jay G. (1995) The Alzheimer’s A beta peptide induces neurodegeneration and apoptotic cell death in transgenic mice. Nat. Genet. 9, 21–30. Le Gal La Salle G., Robert J. J., Berrard S., Ridoux V., StratfordPerricaudet L. D., Perricaudet M., and Mallet J. (1993) An adenovirus vector for gene transfer into neurons and glia in the brain. Science 259, 988 –990. Levy-Lahad E., Wasco W., Poorkaj P., Romano D. M., Oshima J., Pettingell W. H., Yu C. E., Jondro P. D., Schmidt S. D., Wang K., et al. (1995) Candidate gene for the chromosome 1 familial Alzheimer’s disease locus. Science 269, 973–977. Lockhart B. P., Benicourt C., Junien J. L., and Privat A. (1994) Inhibitors of free radical formation fail to attenuate direct betaamyloid 25–35 peptide-mediated neurotoxicity in rat hippocampal cultures. J. Neurosci. Res. 39, 494 –505. Manelli A. M., and Puttfarcken P. S. (1995) Beta-amyloid-induced toxicity in rat hippocampal cells: in vitro evidence for the involvement of free radicals. Brain Res. Bull. 38, 569 –576.

J. Neurochem., Vol. 75, No. 4, 2000

1446

M. BARKATS ET AL.

Moechars D., Lorent K., De Strooper B., Dewachter I., and Van Leuven F. (1996) Expression in brain of amyloid precursor protein mutated in the alpha-secretase site causes disturbed behavior, neuronal degeneration and premature death in transgenic mice. EMBO J. 15, 1265–1274. Moran P. M., Higgins L. S., Cordell B., and Moser P. C. (1995) Age-related learning deficits in transgenic mice expressing the 751-amino acid isoform of human beta-amyloid precursor protein. Proc. Natl. Acad. Sci. USA 92, 5341–5345. Murrell J., Farlow M., Ghetti B., and Benson M. D. (1991) A mutation in the amyloid precursor protein associated with hereditary Alzheimer’s disease. Science 254, 97–99. Paglia D. E., and Valentine W. N. (1967) Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J. Lab. Clin. Med. 70, 158 –169. Pike C. J., Burdick D., Walencewicz A. J., Glabe C. G., and Cotman C. W. (1993) Neurodegeneration induced by beta-amyloid peptides in vitro: the role of peptide assembly state. J. Neurosci. 13, 1676 –1687. Pike C. J., Ramezan-Arab N., and Cotman C. W. (1997) ␤-Amyloid neurotoxicity in vitro: evidence of oxidative stress but not protection by antioxidants. J. Neurochem. 69, 1601–1611. Prehn J. H., Bindokas V. P., Jordan J., Galindo M. F., Ghadge G. D., Roos R. P., Boise L. H., Thompson C. B., Krajewski S., Reed J. C., and Miller R. J. (1996) Protective effect of transforming growth factor-beta 1 on beta-amyloid neurotoxicity in rat hippocampal neurons. Mol. Pharmacol. 49, 319 –328. Quon D., Wang Y., Catalano R., Scardina J. M., Murakami K., and Cordell B. (1991) Formation of beta-amyloid protein deposits in brains of transgenic mice. Nature 352, 239 –241. Rosen D. R., Siddique T., Patterson D., Figlewicz D. A., Sapp P., Hentati A., Donaldson D., Goto J., O’Regan J. P., Deng H. X., et al. (1993) Mutations in Cu/Zn superoxide dismutase gene are associated with familial amyotrophic lateral sclerosis. Nature 362, 59 – 62. Sagara Y., Dargusch R., Klier F. G., Schubert D., and Behl C. (1996) Increased antioxidant enzyme activity in amyloid beta proteinresistant cells. J. Neurosci. 16, 497–505. Schenk D., Barbour R., Dunn W., Gordon G., Grajeda H., Guido T., Hu K., Huang J., Johnson-Wood K., Khan K., Kholodenko D., Lee

J. Neurochem., Vol. 75, No. 4, 2000

M., Liao Z., Lieberburg I., Motter R., Mutter L., Soriano F., Shopp G., Vasquez N., Vandevert C., Walker S., Wogulis M., Yednock T., Games D., and Seubert P. (1999) Immunization with amyloidbeta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 400, 173–177. Selkoe D. J. (1991) The molecular pathology of Alzheimer’s disease. Neuron 6, 487– 498. Shearman M. S., Hawtin S. R., and Tailor V. J. (1995) The intracellular component of cellular 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction is specifically inhibited by ␤-amyloid peptides. J. Neurochem. 65, 218 –227. Sherrington R., Rogaev E. I., Liang Y., Rogaeva E. A., Levesque G., Ikeda M., Chi H., Lin C., Li G., Holman K., et al. (1995) Cloning of a gene bearing missense mutations in early-onset familial Alzheimer’s disease. Nature 375, 754 –760. Stratford-Perricaudet L. D., Makeh I., Perricaudet M., and Briand P. (1992) Widespread long-term gene transfer to mouse skeletal muscles and heart. J. Clin. Invest. 90, 626 – 630. Thomas T., Thomas G., McLendon C., Sutton T., and Mullan M. (1996) ␤-Amyloid-mediated vasoactivity and vascular endothelial damage. Nature 38, 168 –171. Tomlinson B. E., Blessed G., and Roth M. (1970) Observation on the brains of demented old people. J. Neurol. Sci. 11, 205–242. Yankner B. A. (1996) Mechanisms of neuronal degeneration in Alzheimer’s disease. Neuron 16, 921–932. Yankner B. A., Duffy L. K., and Kirschner D. A. (1990) Neurotrophic and neurotoxic effects of amyloid beta protein: reversal by tachykinin neuropeptides. Science 250, 279 –282. Zhang Z., Rydel R. E., Drzewiecki G. J., Fuson K., Wright S., Wogulis M., Audia J. E., May P. C., and Hyslop P. A. (1996) Amyloid ␤-mediated oxidative and metabolic stress in rat cortical neurons: no direct evidence for a role for H2O2 generation. J. Neurochem. 67, 1595–1606. Zhou Y., Gopalakrishnan V., and Richardson J. S. (1996) Actions of neurotoxic ␤-amyloid on calcium homeostasis and viability of PC12 cells are blocked by antioxidants but not by calcium channel antagonists. J. Neurochem. 67, 1419 –1425.