• adenovirus;
  • dopaminergic neurons;
  • gene therapy;
  • oxidative stress;
  • Parkinson's disease


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Several mechanisms are thought to be involved in the progressive decline in neurons of the substantia nigra pars compacta (SNpc) that leads to Parkinson's disease (PD). Neurotoxin 6-hydroxydopamine (6-OHDA), which induces parkinsonian symptoms in experimental animals, is thought to be formed endogenously in patients with PD through dopamine (DA) oxidation and may cause dopaminergic cell death via a free radical mechanism. We therefore investigated protection against 6-OHDA by inhibiting oxidative stress using a gene transfer strategy. We overexpressed the antioxidative Cu/Zn-superoxide dismutase (SOD1) enzyme in primary culture dopaminergic cells by infection with an adenovirus carrying the human SOD1 gene (Ad-hSOD1). Survival of the dopaminergic cells exposed to 6-OHDA was 50% higher among the SOD1-producing cells than the cells infected with control adenoviruses. In contrast, no significant increased survival of (6-OHDA)-treated dopaminergic cells was observed when they were infected with an adenovirus expressing the H2O2-scavenging glutathione peroxidase (GPx) enzyme. These results underline the major contribution of superoxide in the dopaminergic cell death process induced by 6-OHDA in primary cultures. Overall, this study demonstrates that the survival of the dopaminergic neurons can be highly increased by the adenoviral gene transfer of SOD1. An antioxidant gene transfer strategy using viral vectors expressing SOD1 is therefore potentially beneficial for protecting dopaminergic neurons in PD.

Abbreviations used

adenovirus encoding β-galactosidase


recombinant adenovirus encoding SOD1


recombinant adenovirus encoding SOD1


brain-derived neurotrophic factor








Dulbecco's modified Eagle's medium


glutathione peroxidase


glutathione reductase




Hank's balanced salt solution


hydrogen peroxide


Mn(III) tetrakis(4-benzoic acid)porphyrin chloride


multiplicity of infection


nitroblue tetrazolium




phosphate-buffered saline


Parkinson's disease




Rous sarcoma virus


substantia nigra pars compacta


copper-zinc superoxide dismutase




tyrosine hydroxylase immunopositive



Parkinson's disease (PD) is a neurological disorder characterized by a progressive and massive degeneration of dopaminergic cells in the substantia nigra pars compacta (SNpc). The resulting dopamine (DA) depletion in the striatum leads to the deterioration of motor function which manifests itself by tremor, bradykinesia, and rigidity. There is evidence that oxidative stress contributes to the nigral dopaminergic neuronal degeneration. This stress results from both increased generation of free radicals and impairment of the cell defence system for scavenging them (Dunnett and Björklund 1999). First, the nonenzymatic auto-oxidation of DA or its oxidation by monoamine oxidase-B would generate cytotoxic reactive oxygen species like hydrogen peroxide (H2O2) and superoxide (Olanow 1993). Activation of N-methyl-d-aspartate (NMDA) receptors by glutamate released from the neocortical or subthalamic inputs to the SN could further increase the generation of toxic free radicals. This process could be enhanced by the high levels of iron associated with low levels of ferritin in the SN of PD patients (Dexter et al. 1989, 1990) yielding cytotoxic hydroxyl radicals through the Fenton reaction (Youdim et al. 1993). In addition, the SN of PD patients have anormally low levels of reduced glutathione (GSH) (Perry et al. 1981) and of antioxidative enzyme activities (Ambani et al. 1975; Kish et al. 1985; Youdim et al. 1993), as well as abnormally high levels of lipid peroxides (Dexter et al. 1989).

The 6-hydroxydopamine (6-OHDA)-induced lesion of the nigrostriatal dopaminergic system is commonly used as a model for PD. The neurotoxin is taken up by the dopaminergic neurons leading to the generation of neurotoxic reactive oxygen species resulting from DA auto-oxidation (Kumar et al. 1995). 6-OHDA has been detected in the urine of parkinsonian patients (Andrew et al. 1993), and it has been suggested to act endogenously in PD (Glinka et al. 1997). Injections of the neurotoxin into the rat striatum induce a reduction in the levels of GSH, GSH peroxidase (GPx), and SOD1, together with an increase in lipid peroxidation (Perumal et al. 1989; Kumar et al. 1995), a situation that mimics some aspects of the oxidative stress associated with PD. Antioxidants such as N-acetylcysteine, Mn(III) tetrakis(4-benzoic acid)porphyrin chloride (MnTBAP), or the C3 carboxyfullerene derivative, protect against 6-OHDA-mediated toxicity in dopaminergic neurons in vitro (Choi et al. 1999; Lotharius et al. 1999). Moreover, the protective potential of the SOD1 antioxidative enzyme against (6-OHDA)-induced toxicity has been demonstrated both in vitro using human SY5Y neuroblastoma cells (Tiffany-Castiglioni et al. 1982) and in vivo with SOD1 transgenic mice. In this study, the mice were protected against the neurotoxic effects of intracerebroventricular injection of the toxin (Asanuma et al. 1998). These studies all implicate free radicals in the (6-OHDA)-induced neurotoxicity.

One way to increase the antioxidant potential of dopaminergic cells both in vitro and in vivo is the transfer of genes encoding antioxidative enzymes into specific cell compartments, such that there is long-term overexpression of the corresponding proteins. This ‘antioxidant’ strategy of neuroprotection based on gene transfer technology should help increase the resistance of dopaminergic neurons to the oxidative stress generated during PD. We previously described the efficacy of this methodology for increasing survival of neuronal cells in which death was induced by exposure to glutamate or to β-amyloid in vitro (Barkats et al. 1996, 2000), and to the stress mediated by the intracerebral grafting procedure (Barkats et al. 1997). To analyse the protective potential of the DA neuronal gene transfer of SOD1 against (6-OHDA)-induced neurotoxicity, rat embryonic mesencephalic neurons were transduced with an adenovirus vector encoding the human SOD1 enzyme (Ad-SOD1) prior to exposure to 6-OHDA. Active human SOD1 was expressed in the (Ad-SOD1)-infected dopaminergic neurons. These neurons were significantly more resistant to 6-OHDA than either non-infected or control adenovirus-infected neurons. In contrast, the adenovirally mediated overproduction of GPx in dopaminergic neurons did not protect them against the toxic effects of 6-OHDA.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Recombinant adenoviruses

The human CuZnSOD (Ad-SOD1) and the bovine GPx genes (Ad-GPx) were cloned in an adenoviral backbone under the control of the Rous sarcoma virus (RSV) promoter (Barkats et al. 1996, 2000). One control adenovirus, encoding βgalactosidase, expressed the LacZ reporter gene under the control of the same promoter (Stratford-Perricaudet et al. 1992). The second control adenovirus (empty adenovirus) contained the cytomegalovirus (CMV) promoter without any transgene (Vector Developments, Gencell, Aventis, France). All viral stocks were prepared as previously described (Stratford-Perricaudet et al. 1992).

Primary cultures of ventral mesencephalon

We established serum-free ventral mesencephalon cultures enriched for dopaminergic neurons from embryonic day 14 Sprague–Dawley rat midbrain. Tissues were rinsed in phosphate-buffered saline (PBS) containing 0.6% glucose and were incubated at 37°C in a Hank's balanced salt solution (HBSS) containing 0.1% trypsin and 0.05% DNase. After 30 min of incubation, the tissues were mechanically dissociated in Dulbecco's modified Eagle's medium (DMEM) containing 4.5 g/L glucose and 0.58 g/L d-glutamine. This medium was supplemented with 100 µg/mL transferrin, 25 µg/mL insulin, 10 µg/mL putrescin, 5 ng/mL sodium selenite and 6.3 ng/mL progesterone (all from Sigma, St Quentin Fallavier, France). Viable cells were counted using trypan blue cell exclusion and plated at a density of 500 000 cells/well on 24-well polyornithine-coated culture plates. Cells were grown for 5 days, alone or in the presence of adenovirus, in a humidified incubator at 37°C in 5% CO2/90% air atmosphere.

Adenoviral infections

To determine the adenoviral transduction efficacy of embryonic dopaminergic cells, three culture wells were infected with 25 pfu of Ad-βgal, and the number of tyrosine hydroxylase immunopositive (TH+) cells expressing β-galactosidase was assessed 48 h after infection (six fields were counted per well).

To investigate for neuroprotection against 6-OHDA, nine culture wells were infected with Ad-βgal (five at MOI 25, four at MOI 50), 11 with Ad-SOD (seven at MOI 25, four at MOI 50), five with Ad-GPx (all at MOI 25), and eight with empty vectors (four at MOI 25, four at MOI 50). Eight control wells were uninfected.

Twenty-four hours after plating, the cell medium was removed, and replaced with DMEM containing appropriate dilutions of each adenovirus. After 45 min of incubation, the adenoviral solutions were replaced by fresh culture medium. Non-infected cells were incubated in parallel in virus-free DMEM.

6-OHDA toxicity

Two days after adenoviral infection, 50 µd 6-OHDA (Sigma) or PBS (for controls) was added to cultures for 2 h. All wells but two per group were treated with the 6-OHDA toxin. The viability of dopaminergic cells was assessed using TH immunohistochemistry and counting of the TH+ cells. All the TH+ cells were counted in each well. The percent viability was calculated as the ratio of TH+ cells in 6-OHDA-treated cultures to that in non-treated cultures.


TH immunocytochemistry

Cells were fixed with 4% paraformaldehyde (PFA) in 0.1 d PBS, and incubated in 0.1 d PBS containing 10% goat serum and 0.1% Triton X-100 (blocking solution) for 1 h. Cells were then incubated overnight in the blocking solution containing a rabbit polyclonal anti-TH antibody (Jacques Boy 1 : 2000). The anti-TH antibody was detected using goat anti-rabbit Ig and avidin-biotin peroxidase complex (ABC kit, Vector) with 3,3′-diaminobenzidine (DAB) as the chromogenic substrate.

SOD immunocytochemistry

Cells were fixed with 4% PFA in 0.1 d PBS, and incubated in 0.1 d PBS containing 10% swine serum and 0.1% Triton X-100 (blocking solution) for 1 h. Cells were then incubated overnight in the blocking solution containing swine polyclonal anti-human SOD1 antibody (Valbiotech, Abcys, Paris, France, 1 : 500). The anti-SOD1 antibody was detected using swine anti-sheep/goat Ig (Amersham, Orsay, France) and avidin–biotin peroxidase complex (ABC kit, Vector) with 3,3′-DAB as the chromogenic substrate.

Double immunofluorescence TH/SOD1

Cultures were fixed with 4% PFA in PBS for 10 min. The cells were washed with 0.1 d PBS, and then incubated in a blocking solution (10% horse serum, 10% rabbit serum, and 0.1% Triton in 0.1 d PBS) for 1 h. They were then incubated overnight with sheep/goat polyclonal anti-human SOD1 antibody (Valbiotech, 1 : 500) and murine monoclonal anti-TH antibody (Boeringher, 1 : 300) diluted in the blocking solution. The anti-TH antibody was detected using horse anti-mouse Ig (ABC kit, Vector) and streptavidin–biotin phycoerythrin complex (1 : 150, Amersham). For detection of anti-human SOD1 antibody, cells were incubated with rabbit FITC-conjugated anti-sheep Ig (Biosys, Biovalley Marne La Vallee, France, 1 : 300).

Double staining X-gal/TH

Cells were fixed in PBS containing 4% PFA, and then incubated for 2 h at 37°C in a PBS solution containing 0.4 mg/mL of 5-bromo-4-chloro-3-indoyl-β-d-galactosidase (X-gal) substrate (Appligene, Illkirch, France) with 4 md potassium ferricyanide (Sigma), 4 md potassium ferrocyanide (Merck, VWR International, Strasbourg, France) and 4 md MgCl2 (Merck).

The cultures were washed with 0.1 d PBS, and processed for TH immunocytochemistry as described above.

SOD enzymatic activity (NBT assay)

SOD activity was determined by gel electrophoresis followed by nitroblue tetrazolium (NBT) staining. Forty-eight hours after infection, Nonidet P-40 extracts were prepared from adenovirus-treated or non-treated cultures. Cell extracts were loaded on a 15% non-denaturing polyacrylamide gel, and electrophoresis was performed at 100 V. SOD activity was revealed by soaking the gel in distilled water containing 0.3 md NBT and 0.26 md riboflavin (20 min, room temperature) followed incubation in 90 md tetramethylethylenediamine (TEMED) (for 20 min at room temperature).

Statistical analysis

The intergroup differences between the survival rates of TH+ cells were compared using a two-factor analysis of variance (dnova): virus × mutiplicity of infection (MOI).


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

To analyse the neuroprotective potential of dopaminergic SOD1 gene transfer against (6-OHDA)-induced oxidative stress, we determined the survival rate of primary culture dopaminergic neurons infected with recombinant adenoviruses expressing antioxidative enzymes and exposed to neurotoxic concentrations of 6-OHDA.

Expression of functional enzymes in mesencephalic primary cultures

To investigate the adenoviral infection rate of embryonic dopaminergic neurons, we infected mesencephalic primary culture cells with the control adenovirus encoding Escherichia coli LacZ (Ad-βgal) at MOI of 25. Both dopaminergic and non-dopaminergic were efficiently transduced. Co-labelling experiments using combined X-gal cytochemistry and TH immunocytochemistry showed that 57.5 ± 5.75 of the TH+ neurons were transduced (Fig. 1).


Figure 1. Efficacy of adenoviral transduction in rat embryonic dopaminergic cultures. Combined X-gal cytochemistry and TH immunocytochemistry in mesencephalic cultures infected with 25 MOI of Ad-βgal. Expression of β-galactosidase was detected both in dopaminergic (TH+) and non-dopaminergic cells. 57.5 ± 5.7 TH+ neurons were efficiently transduced with Ad-βgal. Arrows, TH+ neurons expressing β-galactosidase (Xgal/TH double-labelled cells); arrowheads, non-transduced TH neurons. Scale bar = 30 µm.

Download figure to PowerPoint

To determine if Ad-SOD1 was able to direct expression of a functional human SOD1 protein in the cells, we infected the cultures with 25 MOI of vector and investigated the level of SOD1 enzymatic activity 2 days later in adenoviral-infected and non-infected cultures either exposed or not exposed to 6-OHDA. Protein extracts from the mesencephalic cells were run on gels, and SOD activity was revealed by NBT staining. This assay, which is based on the determination of the level of superoxide quenching in the gel, allows a semiquantitative analysis of SOD enzymatic activity. As endogenous rat SOD and human exogenous SOD could not be discriminated by their different mobilities in the gel, total SOD activity was quantified in the (Ad-SOD1)-infected cells and compared to that of the control (Ad-βgal)-infected and uninfected cells (Fig. 2a). Cell infection with 25 MOI of Ad-SOD1 nearly doubled the SOD enzymatic activity of cultures. SOD activities in non-infected cells and in cells infected with the control adenovirus were the same. The increased SOD activity in (Ad-SOD1)-infected cultures was thus due to the production of the recombinant human SOD1 protein in the transduced neurons. In addition, the 6-OHDA treatment had no significant effect on the level of SOD activity in any of the conditions tested (Fig. 2b).


Figure 2. SOD enzymatic activity. Gel electrophoresis of protein extracts from primary culture rat mecencephalic cells non-infected or infected with Ad-SOD1 or Ad-βgal (MOI 25). SOD enzymatic activity was determined using the NBT assay in (a) non-treated cells and in (b) 6-OHDA treated cells. The intensity of the band corresponding to SOD activity was nearly double that in cells infected with Ad-SOD1.

Download figure to PowerPoint

We then tested whether the recombinant human SOD1 protein was produced in the dopaminergic neurons (about 5% of the cells). Immunostaining specific for the human SOD1 protein was detected in the soma and neuronal processes of mesencephalic cells infected with a series of concentrations of Ad-SOD1 (Figs 3a–d). We detected no marked difference in the transduction efficacy between 25 and 75 MOI, but a slight viral toxicity was apparent at the latter dose. Co-labelling experiments using combined immunocytochemistry for human SOD1 and rat TH showed that the human protein was present in neurons either expressing or not expressing TH. Fig. 3(e–h) illustrate some TH+ neurons expressing the human SOD1 protein.


Figure 3. Human SOD1 expression in (Ad-SOD1)-infected mesencephalic cells. (a–d) Immunocytochemical detection of the human SOD1 protein in primary culture rat mesencephalic cells (a) non-infected, (b) infected with 25 MOI of Ad-SOD1, (c) infected with 50 MOI of Ad-SOD1, and (d) infected with 75 MOI of Ad-SOD1. Scale bar = 100 µm (e–h): Co-labelling experiments showing TH+ dopaminergic neurons expressing the human SOD1 (hSOD1) protein. (e–g) phycoerythrin, red, TH. (f–h) fluorescein, green, human SOD1. White arrows: hSOD1/TH double-labelled dopaminergic cells. Scale bar = 30 µm.

Download figure to PowerPoint

Neuroprotective effect of SOD1 overexpression

Exposure of mesencephalic cells to 50 µdnova 6-OHDA resulted in the death of about 65% of the TH-positive neurons. To investigate the protective effect of SOD1 overexpression on dopaminergic neurons, cultures were treated with Ad-SOD1 (MOI 25 and 50) two days before exposure to 6-OHDA. The viability of (Ad-SOD1)-infected cells was compared to that of non-infected cells, and of cells infected with Ad-βgal or empty control adenoviruses (Fig. 4). There was no significant difference between the viability of either (Ad-βgal)- or empty vector-infected cells, and therefore values were pooled into a single control virus group (mock). The two-factor analysis of variance (dnova) revealed a significant global effect of the adenoviral treatment (p = 0.0012) and a lack of effect of the viral MOI (p = 0.63). Since there was no significant ‘virus–MOI’ interaction (p = 0.42), comparisons were done between the virus-infected and non-infected groups, independently of the dose that was used (Fig. 4a). Cell transduction with the Ad-SOD1 vector resulted in a significant increase in cell viability compared to mock-infection (54% of increase, p = 0,0003). Cell viability of (Ad-SOD1)-treated cells was also significantly higher than that of uninfected cells (27.4% of increase, p < 0.038) (results are illustrated in Fig. 5).


Figure 4. Ad-SOD1 infection increased resistance to 6-OHDA cytotoxicity. Cytotoxic response of primary culture dopaminergic cells to 6-OHDA. About 65% of TH+ cells were killed by a 24-h exposure to 50 µdnova 6-OHDA. (a) Comparison of the dopaminergic cell viability between Ad-SOD1-infected cells (Ad-SOD1) and cells either uninfected or infected with control adenoviruses (mock) using a two-factor analysis of variance (dnova): ‘viral dose’ (25, 50) × ‘viral treatment’ (Ad-SOD1, Ad-βgal, mock, uninfected). The viability of TH+ cells was significantly increased in Ad-SOD1-infected cultures. (*p < 0.05, ****p < 0.0001, two-factor analysis of variance, post-hoc Fisher's test). (b) Analysis of cell viability in mesencephalic cultures infected with 25 MOI of Ad-GPx. Results shows no apparent neuroprotective effect of Ad-GPx infection, while Ad-SOD1 infected cells were significantly more resistant to 6-OHDA than mock-infected cells (***p < 0.001, one-factor analysis of variance, post-hoc Fisher's test). Cell viability was assayed by counting all TH+ cells per culture well, and results are expressed as percent viability, corresponding to the number of 6-OHDA-treated cells relative to untreated cells.

Download figure to PowerPoint


Figure 5. Resistance of Ad-SOD1-transduced dopaminergic cells to 6-OHDA. Dopaminergic cells were evidenced using TH-immunostaining. Cells received (a, c and e) no. 6-OHDA treatment (b, d and f) 50 µdnova 6-OHDA (24 h exposure). (a and b) Control uninfected mesencephalic TH+ cells. (c and d) Mesencephalic cells infected with 25 MOI of Ad-βgal (double-staining Xgal/TH). (e and f) Mesencephalic cells infected with 25 MOI of Ad-SOD1. More than half of the dopaminergic cells have degenerated in uninfected (b) and Ad-βgal infected (d) cells. Note the neuroprotective effect of SOD1 overexpression in Ad-SOD1 infected cultures (f). Scale bar = 100 µm.

Download figure to PowerPoint

The viability of Ad-GPx infected cells (MOI 25) was then compared to that of uninfected and mock-infected viruses. No significant increase in resistance to 6-OHDA was found when cells overexpressed GPx (one factor dnova, p > 0.2) (Fig. 4b). In contrast, cultures infected with 25 MOI of Ad-SOD1 were significantly more resistant to 6-OHDA than those infected with the control viruses (mock-infected). We did not further test the effect of infection with Ad-GPx at 50 MOI since no marked increase in the transduction efficacy of adenoviral vectors was found at this concentration (see also Barkats et al. 2000). The functionality of this adenoviral vector was previously evidenced using RT-PCR and enzymatic activity assays in 293 cells (Barkats et al. 2000).


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

6-OHDA, which is commonly used to induce PD in experimental animals, is thought to cause dopaminergic cell death via a free radical mechanism. Using primary cultures of rat ventral mesencephalon, we examined the effect of SOD1 overexpression on the survival of dopaminergic neurons exposed to 6-OHDA. To overexpress the antioxidative enzyme in neurons, we used a gene delivery method based on recombinant adenoviruses. These viral vectors are among the most efficient for transducing postmitotic cells like neurons 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).

Co-labelling experiments combining Xgal and TH stainings in (Ad-βgal)-infected cultures revealed the transgene expression within both dopaminergic and non-dopaminergic cells. The percentage of dopaminergic neurons that were transduced after exposure to 25 MOI of Ad-βgal went beyond 50%. Production of the human SOD1 protein was evidenced in the mesencephalic cultures using a semiquantitative enzymatic assay and immunocytochemistry specific for the recombinant protein. Although we did not determine the exact number of the (Ad-SOD1)-transduced TH cells, it might be comparable to that of colabelled Xgal/TH cells (approximately 50%) since similar promoters (RSV) were used in both Ad-βgal and Ad-SOD1 constructions. The semiquantitative NBT assay showed that the intracellular SOD enzymatic activity in extracts of mesencephalic cells infected with 25 MOI Ad-SOD1 was nearly double that in uninfected cell extracts. This result suggests that cell infection with Ad-SOD1 led to a more than two-fold increase in cytoplasmic antioxidative activity (since only approximately 50% of dopaminergic neurons could be transduced with 25 MOI of adenoviral vectors).

Infection of mesencephalic cells with Ad-SOD1 prevented the death of a significant number of dopaminergic neurons exposed to 6-OHDA. The percent survival of TH+ neurons in cultures infected with Ad-SOD1 was significantly increased compared to controls treated with Adβgal or empty vectors (approximately two-fold). The difference with non-infected cells was less pronounced probably due to a residual toxicity of the adenoviral vector. The SOD1 enzyme is not secreted, and is exclusively synthesized in the cytoplasmic cell compartment. This suggests that only the (Ad-SOD1)-infected dopaminergic cells could be protected from (6-OHDA)-induced neurotoxicity. As only half the TH+ cells were transduced by adenoviral vectors, our results suggest that a large majority of the (Ad-SOD1)-transduced dopaminergic neurons were protected from 6-OHDA.

Potential trophic effects mediated by glia were insignificant in our experiments since as less than 1% of the cells in culture (under serum-free conditions) were glial cells (Frodl et al. 1994). A non-specific protective effect of treatment with adenoviral vectors, conceivably induced by a stress adaptative response of the cells, can also be excluded since no protective reaction against 6-OHDA toxicity was detected with control adenoviruses (empty vectors or vectors expressing β-galactosidase). Therefore, our results strongly suggest that the intraneuronal overexpression of SOD1 can inhibit dopaminergic cell death in (Ad-SOD1)-infected cultures.

The mechanisms underlying prevention of dopaminergic cell death by SOD1 overexpression are related to the sensitivity of dopaminergic neurons to (6-OHDA)-mediated oxidative stress. Earlier in vivo studies show that 6-OHDA injected into the striatum of rodents is transported into dopaminergic neurons where it is oxidized into reactive oxygen species (Kumar et al. 1995). Superoxide generated by 6-OHDA (Heikkila and Cohen 1973) can further cause dopaminergic cell death by interacting with nitric oxide to form the highly reactive peroxynitrite radical, or it can react with iron or copper to generate hydroxyl radicals. SOD, which detoxifies free radicals, may have a protective effect on dopaminergic neurons by scavenging (6-OHDA)-generated superoxide. Indeed, when overexpressed in transgenic animals, the superoxide-scavenging SOD1 protein was reported to the prevent the death of dopaminergic neurons induced by intracerebroventricular injection of 6-OHDA (Asanuma et al. 1998).

It has also been suggested that the neurotoxin 6-OHDA exerts its toxic effects on dopaminergic cells by production of H2O2 (Perumal et al. 1989). However, we did not find any significant effect of the overexpression of the H2O2-scavenging enzyme GPx in cultures infected with Ad-GPx before exposure to 6-OHDA. The lack of neuroprotective effect of GPx was unexpected because a partial protection of dopaminergic neurons against (6-OHDA)-induced neurotoxicity has been described in transgenic mice overexpressing GPx (Bensadoun et al. 1998). The discrepancy may result from a decreased availability of reduced glutathione (GSH) in our primary culture model. Indeed, the level of oxidized glutathione (GSSG) increases substantially after treatment with 6-OHDA (Spina et al. 1992). The concomitant decrease of GSH (Perumal et al. 1989) and/or glutathione reductase (GR) may therefore have limited the neuroprotective action of GPx in our experiments. This idea was also discussed by Bensadoun et al. (1998). The question of whether detoxification of H2O2 may contribute to reducing (6-OHDA)-mediated neurotoxicity could therefore be investigated by cotransducing cells with viral vectors encoding both GPx and GR, or GPx and the brain-derived neurotrophic factor (BDNF) which is known to induce the increase of GR activity (Spina et al. 1992).

In a previous study using ex vivo gene transfer technology, we showed that overexpression of SOD1 in dopaminergic neurons prior intrastriatal grafts in hemiparkinsonian rats increased the functional recovery of the animals (Barkats et al. 1997). This study suggested that free radicals were important agents in the dopaminergic cell death induced by the grafting procedure. Here, we further demonstrate the great neuroprotective potential of recombinant SOD1 adenoviral vectors for dopaminergic neurons in culture exposed to the 6-OHDA neurotoxin. Since it is clear that free radical homeostasis is extremely important for the survival of dopaminergic neurons, antioxidant gene transfer based on viral vector methodologies similar to those used in our study allows direct in vivo application of neuroprotective strategies in experimental models of neurodegenerative diseases. Scavenging enzymes, like SOD1, delivered directly to degenerating dopaminergic neurons by means of viral vectors, have enormous potential as neuroprotective agents in these experimental diseases. These antioxidant-based gene transfer strategies may reveal themselves more efficient than those using growth factors such as the glial cell line-derived neurotrophic factor (GDNF) or the BDNF. Indeed, a carboxyfullerene derivative (the C3 potent antioxidant) has recently been reported to outperform the highly protective effect of GDNF in the 6-OHDA model (Lotharius et al. 1999). Moreover, both GDNF and BDNF were recently suggested to act on dopaminergic neuron survival by activating the antioxidative enzyme systems (Spina et al. 1992; Chao and Lee 1999).

In conclusion, our study demonstrates the potential of an adenoviral vector encoding the human SOD1 to prevent (6-OHDA)-induced dopaminergic cell death by increasing the antioxidant potential of neurons. As 6-OHDA could act as an endogenous neurotoxin in PD, and because there is evidence for the contribution of oxidative stress in this disease, these results may be of significance for gene therapy of PD.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Sue Orsoni for expertise in English. This work was supported by the European Commission Biotech Program (QLRT-1999–02189), the Centre National de la Recherche Scientifique, Aventis, Institut pour la Recherche sur la Moelle Epinière, and the Association Française Retinis Pigmentosa. M. Barkats was supported by the Association France Alzheimer.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. 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, 224228.
  • Ambani L. M., Van Woert M. H. and Murphy S. (1975) Brain peroxidase and catalase in Parkinson disease. Arch. Neurol. 32, 114118.
  • Andrew R., Watson D. G., Best S. A., Midgley J. M., Wenlong H. and Petty R. K. (1993) The determination of hydroxydopamines and other trace amines in the urine of parkinsonian patients and normal controls. Neurochem. Res. 18, 11751177.
  • Asanuma M., Hirata H. and Cadet J. L. (1998) Attenuation of 6-hydroxydopamine-induced dopaminergic nigrostriatal lesions in superoxide dismutase transgenic mice. Neuroscience 85, 907917.DOI: 10.1016/S0306-4522(97)00665-9
  • 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, 229234.
  • 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 neurones against glutamate toxicity. Neuroreport 7, 497501.
  • 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, 703713.
  • Barkats M., Millecamps S., Abrioux P., Geoffroy M. C. and Mallet J. (2000) Overexpression of glutathione peroxidase increases the resistance of neuronal cells to Abeta-mediated neurotoxicity. J. Neurochem. 75, 14381446.DOI: 10.1046/j.1471-4159.2000.0751438.x
  • Bensadoun J. C., Mirochnitchenko O., Inouye M., Aebischer P. and Zurn A. D. (1998) Attenuation of 6-OHDA-induced neurotoxicity in glutathione peroxidase transgenic mice. Eur. J. Neurosci. 10, 32313236.
  • Chao C. C. and Lee E. H. (1999) Neuroprotective mechanism of glial cell line-derived neurotrophic factor on dopamine neurons: role of antioxidation. Neuropharmacology 38, 913916.
  • Choi W. S., Yoon S. Y., Oh T. H., Choi E. J., O'Malley K. L. and Oh Y. J. (1999) Two distinct mechanisms are involved in 6-hydroxydopamine- and MPP+-induced dopaminergic neuronal cell death: role of caspases, ROS, and JNK. J. Neurosci. Res. 57, 8694.
  • 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, 219223.
  • Dexter D. T., Wells F. R., Lees A. J., Agid F., Agid Y., Jenner P. and Marsden C. D. (1989) Increased nigral iron content and alterations in other metal ions occurring in brain in Parkinson's disease. J. Neurochem. 52, 18301836.
  • Dexter D. T., Carayon A., Vidailhet M., Ruberg M., Agid F., Agid Y., Lees A. J., Wells F. R., Jenner P. and Marsden C. D. (1990) Decreased ferritin levels in brain in Parkinson's disease. J. Neurochem. 55, 1620.
  • Dunnett S. B. and Björklund A. (1999) Prospects for new restorative and neuroprotective treatments in Parkinson's disease. Nature 399, A32A39.
  • Frodl E. M., Nakao N. and Brundin P. (1994) Lazaroids improve the survival of cultured rat embryonic mesencephalic neurones. Neuroreport 5, 23932396.
  • Glinka Y., Gassen M. and Youdim M. B. (1997) Mechanism of 6-hydroxydopamine neurotoxicity. J. Neural Transm. Suppl. 50, 5566.
  • Heikkila R. E. and Cohen G. (1973) 6-Hydroxydopamine: evidence for superoxide radical as an oxidative intermediate. Science 181, 456457.
  • Kish S. J., Morito C. and Hornykiewicz O. (1985) Glutathione peroxidase activity in Parkinson's disease brain. Neurosci. Lett. 58, 343346.
  • Kumar R., Agarwal A. K. and Seth P. K. (1995) Free radical-generated neurotoxicity of 6-hydroxydopamine. J. Neurochem. 64, 17031707.
  • Le Gal La Salle G., Robert J.-J., Berrard S., Ridoux V., Stratford-Perricaudet L. D., Perricaudet M. and Mallet J. (1993) An adenovirus vector for gene transfer into neurons and glia in the brain. Science 259, 988990.
  • Lotharius J., Dugan L. L. and O'Malley K. L. (1999) Distinct mechanisms underlie neurotoxin-mediated cell death in cultured dopaminergic neurons. J. Neurosci. 19, 12841293.
  • Olanow C. W. (1993) A radical hypothesis for neurodegeneration. Trends Neurosci. 16, 439444.
  • Perry T. L., Kish S. J., Hansen S., Wright J. M., Wall R. A., Dunn W. L. and Bellward G. D. (1981) Elevation of brain GABA content by chronic low-dosage administration of hydrazine, a metabolite of isoniazid. J. Neurochem. 37, 3239.
  • Perumal A. S., Tordzro W. K., Katz M., Jackson-Lewis V., Cooper T. B., Fahn S. and Cadet J. L. (1989) Regional effects of 6-hydroxydopamine (6-OHDA) on free radical scavengers in rat brain. Brain Res. 504, 139141.
  • Spina M. B., Squinto S. P., Miller J., Lindsay R. M. and Hyman C. (1992) Brain-derived neurotrophic factor protects dopamine neurons against 6-hydroxydopamine and N-methyl-4-phenylpyridinium ion toxicity: involvement of the glutathione system. J. Neurochem. 59, 99106.
  • 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, 626630.
  • Tiffany-Castiglioni E., Saneto R. P., Proctor P. H. and Perez-Polo J. R. (1982) Participation of active oxygen species in 6-hydroxydopamine toxicity to a human neuroblastoma cell line. Biochem. Pharmacol. 31, 181188.
  • Youdim M. B., Ben-Shachar D. and Riederer P. (1993) The possible role of iron in the etiopathology of Parkinson's disease. Mov. Disord. 8, 112.