Green tea polyphenol (–)-epigallocatechin-3-gallate prevents N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-induced dopaminergic neurodegeneration


  • Yona Levites,

    1. *Eve Topf and USA National Parkinson Foundation, Centers of Excellence for Neurodegenerative Diseases Research, and †Department of Cell Biology, Technion-Faculty of Medicine, Haifa, Israel
    Search for more papers by this author
  • Orly Weinreb,

    1. *Eve Topf and USA National Parkinson Foundation, Centers of Excellence for Neurodegenerative Diseases Research, and †Department of Cell Biology, Technion-Faculty of Medicine, Haifa, Israel
    Search for more papers by this author
  • Gila Maor,

    1. *Eve Topf and USA National Parkinson Foundation, Centers of Excellence for Neurodegenerative Diseases Research, and †Department of Cell Biology, Technion-Faculty of Medicine, Haifa, Israel
    Search for more papers by this author
  • Moussa B. H. Youdim,

    1. *Eve Topf and USA National Parkinson Foundation, Centers of Excellence for Neurodegenerative Diseases Research, and †Department of Cell Biology, Technion-Faculty of Medicine, Haifa, Israel
    Search for more papers by this author
  • Silvia Mandel

    1. *Eve Topf and USA National Parkinson Foundation, Centers of Excellence for Neurodegenerative Diseases Research, and †Department of Cell Biology, Technion-Faculty of Medicine, Haifa, Israel
    Search for more papers by this author

Address correspondence and reprint requests to Professor M. B. H. Youdim, Department of Pharmacology, Technion- Faculty of Medicine, P.O.B. 9697, 31096 Haifa, Israel. E-mail:


In the present study we demonstrate neuroprotective property of green tea extract and (–)-epigallocatechin-3-gallate in N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mice model of Parkinson's disease. N-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine neurotoxin caused dopamine neuron loss in substantia nigra concomitant with a depletion in striatal dopamine and tyrosine hydroxylase protein levels. Pretreatment of mice with either green tea extract (0.5 and 1 mg/kg) or (–)-epigallocatechin-3-gallate (2 and 10 mg/kg) prevented these effects. In addition, the neurotoxin caused an elevation in striatal antioxidant enzymes superoxide dismutase (240%) and catalase (165%) activities, both effects being prevented by (–)-epigallocatechin-3-gallate. (–)-Epigallocatechin-3-gallate itself also increased the activities of both enzymes in the brain. The neuroprotective effects are not likely to be caused by inhibition of N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine conversion to its active metabolite 1-methyl-4-phenylpyridinium by monoamine oxidase-B, as both green tea and (–)-epigallocatechin-3-gallate are very poor inhibitors of this enzyme in vitro (770 µg/mL and 660 µM, respectively). Brain penetrating property of polyphenols, as well as their antioxidant and iron-chelating properties may make such compounds an important class of drugs to be developed for treatment of neurodegenerative diseases where oxidative stress has been implicated.

Abbreviations used

black tea




3,4-dihydroxyphenylacetic acid




green tea


homovanillic acid






oxidative stress


Parkinson's disease


reactive oxygen species


substantia nigra pars compacta


tyrosine hydroxylase


tumor necrosis factor-α


superoxide dismutase




3′-{1-[(phenylamino)-carbonyl]-3,4-tetrazolium}-bis(4-methoxy-6-nitro)benzenesulfonic acid hydrate.

The presence of ongoing oxidative stress (OS) and inflammatory processes occurring selectively in the substantia nigra pars compacta (SNPC) of parkinsonian brains and in animal models of Parkinson's disease (PD; Youdim et al. 1993; Gerlach et al. 1994; Jenner and Olanow 1996) has recently been substantiated by cDNA microarray gene expression (Mandel et al. 2000; Grünblatt et al. 2001a). This includes alterations in the activities of the antioxidant enzymes superoxide dismutase (SOD) and catalase, a decrease in the levels of reduced glutathione (GSH) (Riederer et al. 1989; Di Monte and Chan Piu Sandy 1992; Grünblatt et al. 2001c), proliferation of reactive microglia with highly significant increase in cytokines tumor necrosis factor-α (TNF-α), interleukin (IL)-1 and IL-2 (Mogi et al. 1994a,b), along with an increase in the levels of iron within the reactive microglia and the melanin containing dopamine (DA) neurons (Riederer et al. 1989; Iacopino and Christakos 1990; Youdim et al. 1993; Jellinger 1999).

The approaches to neuroprotection in PD reflect the current concepts of the etiology of the disease. Antioxidant strategies, and iron-chelating/anti-inflammatory strategies have been at the focus of attention (Slikker et al. 1999). Among antioxidants and iron chelators of potential therapy, tea and tea polyphenols have attracted increasing interest because of their well reported biological effects both in vivo and in vitro, including radical scavenging (Salah et al. 1995; Nanjo et al. 1996), iron chelating (Grinberg et al. 1997; Guo et al. 1996), anti-inflammatory (Haqqi et al. 1999; Pan et al. 2000), anticarcinogenic (Wang et al. 1994; Lin et al. 1999) and anti-angiogenic (Cao and Cao 1999) actions. Tea and tea polyphenols have been shown to increase the plasma total antioxidant capacity. Ingestion of either black tea (BT) or green tea (GT) protects plasma low-density lipoprotein oxidation in humans (Serafini et al. 1996) and in rats fed with GT extract (Anderson et al. 1998). In addition, liver and kidney slices from EGCG or GT extract fed rats, showed reduction of lipid peroxidation (Sano et al. 1995).

Recent reports have revealed that polyphenol flavonoids may be neuroprotective in neuronal primary cell cultures: the Ginkgo biloba extract, known to be enriched with flavonoids, has been shown to protect hippocampal neurons from nitric oxide or beta-amyloid derived peptides-induced neurotoxicity (Bastianetto et al. 2000a,b). Moreover, the polyphenol flavanol epicatechin was shown to attenuate neurotoxicity induced by oxidized low-density lipoprotein in mouse-derived striatal neurons (Schroeter et al. 2000). Tea extracts and EGCG attenuated the neurotoxic action of 6-hydroxydopamine in rat PC12 and human neuroblastoma SH-SY5Y cells (Levites et al. 2001).

The consumption of flavonoid-rich blueberries or strawberries has been shown to reverse the age-related cognitive and motor behavioral deficits in rats (Joseph et al. 1999). The protective effect of tea polyphenols has also been described in a mice model of cerebral ischemia, where intravenous injection of tea catechins (Matsuoka et al. 1995) or intraperitoneal injection of (–)-epigallocatechin-3-gallate (EGCG; Lee et al. 2000) immediately after ischemia improved the memory impairment and reduced hippocampal neuronal damage, respectively. EGCG has also been reported to access the brain: mice fed with radioactively labeled EGCG, have shown a substantial incorporation into their brains similar to the levels found in lung, kidney, heart, liver spleen and pancreas (Suganuma et al. 1998).

The pharmacological actions of GT extract and its polyphenols, together with their ability to penetrate the brain, fulfilled the requirements for a potential neuroprotective action. We report here for the first time the neuroprotective effects of GT extract and its isolated EGCG polyphenol on striatal DA depletion and neuronal loss in substantia nigra of mice induced by N-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), a potent neurotoxin known to specifically induce dopaminergic neurodegeneration via OS (for review see Grünblatt et al. 2000).

Experimental procedures

Green tea extract

Dried GT and BT extracts (from Indonesia, ART. 85007 and 90177, respectively, Plantextrakt GmbH & Co. KG, Vestenbergsgreuth, Germany) were obtained from Wissotzky company (Tel Aviv, Israel). The same dried extract powders were used for every preparation of the tea solutions in each experiment, thus minimizing the possible variations in the concentration of the active components, that occur between extract preparations. Each 0.12 g dried material derives from 100 mL original liquid extract and is the recommended dissolution to prepare a regular cup of tea. Normally, 3–7 kg of GT raw material have been used for producing 1 kg of powdered extracts. The total polyphenol concentration in GT extract, as specified by the manufacturers, ranged from 18–20% and the catechin content (analyzed by HPLC) from 12–17%. We also determined the total polyphenol content in the extract spectrophotometrically with phosphomolybdic phosphotungstic acid reagent (Narr Ben et al. 1996) and quercetin as a standard. The polyphenol concentration was 20% of the powdered extract.

Animal treatments and brain catecholamine determination by HPLC

All procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Technion Animal Ethics committee, Haifa, Israel. All the efforts were made to minimize animal suffering and to use only the number of animals necessary to produce reliable scientific data. Male C57/BL mice (20–25 g) received two doses of GT extract (0.5 mg/kg, i.p.) at 6 h intervals on the first day. Sixteen hours later they were injected with MPTP (Sigma Chemical Co., St Louis, MO, USA; 24 mg/kg, i.p.) for four consecutive days. Each MPTP injection was preceded by one of GT. Control mice received only saline or GT. For EGCG studies, mice were orally administered with EGCG alone (2 and 10 mg/kg/day) for 10 days and the following 4 days the animals received a combination of EGCG and MPTP (24 mg/kg/day). The animals were killed by decapitation 3 days after the last injection and their striata and hippocampus from one hemisphere were immediately homogenized in 0.1 mol/L perchloric acid for HPLC (high performance liquid chromatography) analysis (Ben-Shachar and Youdim 1990) of DA, 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA). Striata from the second halves of mice brains were used for preparation of homogenates for enzyme activity and western-blot analysis (Grünblatt et al. 1999).

Preparation of tissue homogenates

Striata and liver were homogenized in 0.3 mol/L sucrose/0.05 mol/L potassium phosphate buffer pH 7.4, containing a mix of protease inhibitors (completeTM, Boheringer, Mannheim, Germany). Homogenates were centrifuged at 12 000 g for 10 min at 4°C and supernatant was taken for further analysis.

Western blotting analysis for tyrosine hydroxylase

Analysis of tyrosine hydroxylase (TH; EC levels by western blotting were conducted essentially as previously described (Grünblatt et al. 2001a).

Tyrosine hydroxylase activity

TH activity in the striatum was determined by the method of Kato et al. (1981) with the following modifications: 190 µL of striatal homogenate (1 mg protein/mL) were incubated as described. After the incubation, the reaction was stopped with 4.5 mL of ethanol solution containing 300 pmol of 3,4-dihydroxybenzylamine as an internal standard and 100 µL of 0.2 mol/L EDTA (ethylenediaminotetraacetic acid), and then centrifuged as described. The pH (8–8.5) of the supernatant was adjusted, and was passed through the 100 mg alumina column (J & W Scientific, Folsom, CA, USA). The column was washed with 1.5 mL of 0.5 mol/L Tris-buffer (pH 8.6) twice, 1.5 mL water, and 100 µL of 0.5 mol/L HCl. DOPA and 3,4-dihydroxybenzylamine were eluted with 300 µL of 0.5 mol/L HCl and quantitated by HPLC (same specifications as for catecholamines).

Superoxide dismutase activity (SOD)

SOD (EC activity was assessed based on the method of Ukeda et al. (1997). Twenty microliters of striatal (40 µg/mL protein) or liver (8 µg/mL protein) homogenate, were incubated in 50 mmol/L carbonate buffer (pH 10.2) containing xanthine (3 mmol), EDTA (3 mmol/L) and 3′-{1-[(phenylamino)-carbonyl]-3,4-tetrazolium}-bis(4-methoxy-6-nitro)benzenesulfonic acid hydrate (XTT; 0.75 mmol/L) at a final volume of 250 µL in a 96 microtiter plate. The reaction was started by the addition of 25 µL xanthine oxidase (36 miliunits/mL). The plate was immediately introduced into an ELISA reader instrument for 3 min and the data was analyzed by basic Magellan Wizard Software V2.22 (TECAN Austria GmbH, Salzburg, Austria). The slope of the initial reaction (without homogenate) was adjusted to 0.02 OD470/min. All reagents are from Sigma.

Catalase activity

Twenty microliters of striatal or liver homogenates were used for enzymatic analysis of catalase (EC as described (Aebi 1974).

Immunohistochemistry for TH

Paraffin sections were reacted for 2 h at room temperature with a specific antibody against mouse TH (Sigma) (Mori et al. 1988). Detection was done by the appropriate biotinylated second antibody with a streptavidin-peroxidase conjugate and S-(2-aminoethyl)-l-cysteine (AEC) as substrate. Counterstaining was done with hematoxylin.

Monoamine oxidase activity (MAO)

MAO (EC activity were estimated in striatal homogenates using [14C]phenylethylamine [PEA; 10–20 µmol/L (NEN, Zaventum, Belgium)] for MAO-B (Youdim 1975). Striata from naive C57-BL mice were homogenized in 20 volumes of 0.32 mol/L sucrose. Samples were preincubated with increasing concentrations of GT, EGCG or l-deprenyl for 30 min at 37°C before addition of the substrates.


One-way analysis of variance followed by Tukey's test, or Student's t-test, were performed using the scientific statistic software GraphPad InstatTM version 2.04 (GraphPad Software Inc., San Diego, CA, USA). p-values of less than 0.05 were considered significant.


Effect of green tea extract on MPTP-induced mouse striatal dopamine and TH depletion

In order to establish possible neuroprotective properties of GT in vivo, the MPTP model of PD was employed. Mice were pretreated with GT extract (0.5, 1 and 5 mg/kg corresponding to 0.3, 0.6 and 3 µmol total polyphenols/kg/day, i.p.) before the injection of MPTP (24 mg/kg, i.p.). MPTP caused a marked reduction in striatal DA levels after 4 days treatment (Table 1). However, GT extract conferred significant protection against MPTP-induced DA decrease with either 0.5 mg/kg or 1 mg/kg (from 45 ± 4% to 73 ± 5% and 66 ± 5% of control, respectively). However, doses of 5 and 10 mg/kg were not effective. GT extract (0.5 and 1 mg/kg) also diminished the increased striatal DA turnover (TO) induced by MPTP, as measured by the DOPAC + HVA/DA ratio, while GT alone had no significant effect (Table 1). In contrast, the higher dose of GT (5 mg/kg) elevated DA turnover to values higher than those observed with MPTP (Table 1). No significant changes were detected in the neurotransmitters norepinephrine and serotonin levels in hippocampus (data not shown). It is interesting to note that GT itself at 0.5 mg/kg (0.3 µmol total polyphenols/kg) caused an elevation in DA levels (130 ± 9% of control). These findings correlated with the changes observed in striatal TH protein content (Fig. 1a): MPTP treatment reduced TH levels to 62 ± 7.0% of control, whereas pretreatment with 0.5 mg/kg GT prevented this decrease (110 ± 9.6% of control). Similar results were obtained with 1 mg/kg (from 53 ± 5.7 to 79 ± 6.1% of control, respectively; data not shown). In addition, TH activity was significantly decreased in MPTP treated mice striata (65 ± 4.3%) whereas pretreatment with 0.5 mg/kg GT completely prevented this effect (101 ± 6.3% of control, Fig. 1b). GT extract alone did not affect TH activity.

Table 1.   Striatal DA content and turnover (DOPAC + HVA/DA) following GT treatment
 GT (% of control)
 0.5 mg/kg1 mg/kg5 mg/kg
Group treatmentDATODATODATO
  1. C57-BL mice were injected with GT (0.5–5 mg/kg/day, containing 0.3–3 µmol total polyphenols/kg, i.p.) followed by a dose of MPTP (24 mg/kg per day) for 4 days. Respective controls received saline or GT only. Striatal DA and metabolites were measured by HPLC, and the DA-turnover was calculated (DOPAC + HVA/DA). Absolute DA, DOPAC and HVA values in control, untreated mice are: 74.32 ± 1.73, 3.77 ± 0.40 and 1.07 ± 0.10 pmol/mg tissue, respectively. The results represent the mean ± SEM (n = 6 mice). One-way anova: ap < 0.001 versus control; bp < 0.01 versus MPTP. Student's t-test: cp < 0.05 versus MPTP.

Control100 ± 9100 + 18100 ± 9100 + 18100 ± 9100 + 13
Green tea130 ± 999 ± 1393 ± 9117 ± 2278 ± 871 ± 8
MPTP (24 mg/kg/day)45 ± 4a179 ± 1845 ± 4a179 ± 1828 ± 4a136 ± 13
Green tea + MPTP73 ± 5b112 ± 20c66 ± 5130 ± 2035 ± 8165 ± 6
Figure 1.

 Effect of GT on TH content and activity in MPTP-treated mice. The striata from the second halves of the mice brains treated with GT (0.5 mg/kg/day) from Table 1, were homogenized as described in Materials and methods. (a) Twenty-five micrograms of protein were electrophoresed on 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Triplicate bands per group represent a pool of two animals per each lane. Duplicate bands per group represent a pool of three animals in each lane. Bands were detected with mouse monoclonal anti-TH antibody and visualized by chemiluminescence. Densitometric analysis of the bands, given in arbitrary units, is represented graphically. The values are the mean ± SEM. (b) Striatal TH activity was measured according to the method described by Kato et al. (1981). The results represent the mean ± SEM of triplicates from pooled striata (n = 6). Student's t-test, *p < 0.05 versus MPTP. Absolute TH activity values in control, untreated mice are: 130 ± 3.9 nmol/min/mg protein.

Immunohistochemistry for TH positive neurons in SN

In order to investigate whether the neuroprotection exerted by GT extract, as indicated by striatal DA and TH levels, is a consequence of neuronal survival or a compensatory event, immunohistochemical analysis for TH positive neurons in SN was performed. MPTP treatment caused a marked 40% loss of TH positive neurons in SN (dopaminergic cell count was reduced from 159.0 ± 11.15 to 96.0 ± 7.47; Fig. 2b and Table 2), whereas GT extract (0.5 mg/kg) pretreatment prevented this, resulting in a 15% reduction, not significantly different from control values (Figs 2a and c and Table 2).

Figure 2.

 Immunohistochemical analysis for TH-positive neurons in substantia nigra. Animals were treated as described in Table 1. Transversed parallel paraffin sections were reacted with mouse anti-TH antibody followed by biotinylated second antibody, streptavidin-peroxidase conjugated, and AEC substrate. TH positive neurons were observed specifically in SN of (a) control (b) MPTP and (c) GT plus MPTP-treated mice and counted accordingly (Table 2).

Table 2.   Determination of dopaminergic cell count in substantia nigra of saline, MPTP- and green tea-treated mice
ControlMPTP (24 mg/kg)MPTP (24 mg/kg) + GT (0.5 mg/kg)
  1. Mice were treated with MPTP and GT as described in Fig. 2. TH positive neurons were counted. Values represent the mean ± SEM (n = 6). Student's t-test: p < 0.01 versus control; *p < 0.02 versus MPTP.

159.0 ± 11.1596.0 ± 7.47133.8 ± 13.70*

EGCG prevents striatal DA and TH reduction induced by MPTP

The biological activities of GT extract have been attributed to its polyphenolic content. We investigated the effect of the major polyphenol constituent of green tea, EGCG, on MPTP-induced neurodegeneration. For this purpose, EGCG (2 and 10 mg/kg) was given orally, alone for 10 days to achieve a relatively chronic consumption of the polyphenol and then MPTP was administered in combination with EGCG for another 4 days. EGCG (2 and 10 mg/kg), similarly to GT extract, considerably prevented the decrease in striatal DA induced by MPTP (from 41 ± 7.8% to 79 ± 7.6% and from 33 ± 5.4% to 70 ± 10% of control, respectively; Fig. 3). EGCG, as observed with GT, also restored the increased striatal DA turnover induced by MPTP, while EGCG alone had no significant effect (Table 3).

Figure 3.

 Effect of EGCG on striatal dopamine content in MPTP-treated mice. C57-BL mice received EGCG (2 and 10 mg/kg, orally) for 10 days and the following four days the animals received a combination of EGCG and MPTP (24 mg/kg/day, i.p.). Controls received saline or EGCG only. Striatal DA was measured by HPLC. The results represent the mean ± SEM (n = 6 mice). One-way anova; *p < 0.05 versus MPTP.

Table 3.   Striatal dopamine turnover (DOPAC + HVA/DA) following EGCG treatment
 EGCG (% of control)
Group treatment2 mg/kg10 mg/kg
  1. C57-BL mice were injected with EGCG (orally, 2 and 10 mg/kg per day) for 14 days. At the last 4 days a dose of EGCG was followed by a dose of MPTP (24 mg/kg per day). Respective controls received saline or EGCG only. Striatal DA and metabolites were measured by HPLC, and the DA-turnover was calculated (DOPAC + HVA/DA). The results represent the mean ± SEM (n = 6). Student's t-test: *p < 0.05 versus MPTP.

Control100 ± 18100 ± 6
EGCG81 ± 12120 ± 11
MPTP (24 mg/kg/day)154 ± 15175 ± 23
EGCG + MPTP97 ± 14*115 ± 21

The decreased levels of TH protein caused by MPTP were restored by pretreatment with both 2 and 10 mg/kg/day of EGCG (from 46 ± 14% to 97 ± 13% and from 23 ± 0.5% to 82 ± 2.3% of control, respectively; Fig. 4a). Similarly, the reduction in TH activity induced by MPTP (55 ± 2.6% and 61 ± 1.2%, respectively) was restored to control values by pretreatment with 2 mg/kg (109 ± 4.8% of control) whereas 10 mg/kg was less effective (84 ± 3.9% of control; Fig. 4b). EGCG alone (2 mg/kg), caused a marked reduction in the levels of TH protein (68 ± 7.3% of control) while its activity was up-regulated at 2 and 10 mg/kg (166 ± 11.5% and 145 ± 7.2% of control, respectively).

Figure 4.

 Effect of EGCG on TH content and activity in MPTP-treated mice. The striata from the second halves of the mice brains from Fig. 3 were homogenized as described in Materials and methods. (a) and (b) as described in Fig. 1. Student's t-test, *p < 0.05 versus MPTP; **p < 0.005 versus MPTP; †p < 0.05 versus control.

MPTP induction of antioxidant enzymes SOD and catalase is prevented by EGCG

The activities of SOD and catalase enzymes were assessed in striatum and liver of mice treated with either EGCG, MPTP or a combination of the two. MPTP was found to increase both enzyme activities in striatum (SOD, 240 ± 26.9%, catalase, 165 ± 16.9%), while pretreatment with EGCG prevented this effect (Fig. 5), except for the higher EGCG dose (10 mg/kg) which did not prevent the increase in SOD activity induced by MPTP. EGCG itself elevated the activities of striatal SOD and catalase by 223 ± 25.4 and 202 ± 32.1, respectively, at a concentration of 2 mg/kg. At 10 mg/kg only SOD was increased. Liver SOD activity was not affected by EGCG at the two concentrations tested, while liver catalase activity was increased (150 ± 1.4%) only with 10 mg/kg (Fig. 5).

Figure 5.

 Effect of EGCG on catalase and SOD activities in MPTP-treated mice. Catalase and SOD activities were determined in (a) striatal homogenates from Fig. 4 and in (b) liver homogenates from same mice. Absolute catalase and SOD activity values in control, untreated mice are: 3.17 ± 0.15 µmol H2O2/min/mg protein and 0.39 ± 0.01 units/mg protein, respectively, in striatum and 230 ± 11.6 µmol H2O2/min/mg protein and 12.4 ± 0.24 units/mg protein, in liver, respectively. The results represent the mean ± SEM (n = 6) of a representative experiment. Student's t-test: †p < 0.05 versus control; ††p < 0.001 versus control; *p < 0.05 versus MPTP.

MAO-B activity

The activity of the MPTP metabolizing enzyme, MAO-B, was examined in vitro using mouse brain homogenates, with increasing concentrations of GT, EGCG or deprenyl with phenylethylamine as substrate. GT extract inhibited MAO-B at a relatively high IC50 (713 µg/mL). Similarly, EGCG displayed an IC50 of 662 µm. In contrast, the irreversible MAO-B inhibitor, l-deprenyl, caused a 50% inhibition at a concentration of 6 nm(Table 4).

Table 4.   Inhibitory potency of GT, EGCG and deprenyl on MAO-B activity
  1. Brain homogenates were preincubated with increasing concentrations of GT, EGCG and deprenyl. MAO-B activity was measured and IC50 values were calculated. The values represent the mean of 2–3 experiments.

Green tea713 µg/mL
EGCG662 µm (303 µg/mL)
Deprenyl6 nm (2 ng/mL)


This study is the first to show neuroprotective activity of green tea extract and its individual polyphenol, EGCG, in an MPTP mouse model of PD. The current hypothesis concerning the pathogenesis of idiopathic PD points to selective OS that expresses itself with compatible biochemical changes in SNPC. Therefore, drugs that exhibit free radical-scavenging and iron-chelating properties, may serve as potential candidates for the treatment of PD. One possible mechanism underlying the effectiveness of GT and EGCG against MPTP neurotoxicity, as shown in the present study, may involve its catechol-like structure, since it is known that catechol-containing compounds are potent radical scavengers and chelators of ferric ion (Guo et al. 1996; van Acker et al. 1996; Grinberg et al. 1997; Gassen and Youdim 1997; Gassen and Youdim 1999). Indeed, we have previously shown that the catechol derivatives, R- and S-apomorphine, are potent iron chelators and radical scavengers in brain mitochondrial fraction (Gassen et al. 1996). They also inhibit MAO-B reversibly, with similar IC50 concentration (300 µm) (Grünblatt et al. 2001a) to that reported for EGCG. In addition, they protect PC12 cells against toxicity induced by 6-OHDA and H2O2 (Gassen et al. 1998) and, in vivo, R-apomorphine protects SN dopamine neurons against MPTP-induced cell loss (Grünblatt et al. 1999, 2001a). EGCG, the main constituent of GT, was shown to be easily absorbed from the digestive tract and was widely distributed into various organs, including the brain, which had a similar concentration to that of the liver (Suganuma et al. 1998). At this stage we cannot state whether the conjugated or the free form of EGCG (or both) is the one that penetrates the brain. In future studies we will concentrate on determination of conjugated polyphenols in brain.

In the present study we used the mouse MPTP model of PD to study potential protective activities of GT extract and EGCG. MPTP neurotoxin produces a pattern of neurodegeneration and a neuropathology that is similar to that seen in parkinsonian brains (Heikkila et al. 1984; Langston 1996; Grünblatt et al. 2001a). Administration of GT extract (0.5 and 1 mg/kg, equivalent to 0.3 and 0.6 µmol total polyphenols/kg/day) protected against MPTP-induced decrease in striatal DA content, as manifested by the almost normalized levels of DA. DOPAC and HVA metabolites reduction was also significantly prevented by GT (data not shown). Higher GT doses of 5 and 10 mg/kg were not effective in prevention of DA depletion by MPTP. These data suggest that there is a concentration-dependent window of action, beyond which the effect of GT becomes detrimental to the neurons. In this context GT acts like true antioxidants such as vitamin C (Halliwell 1996) and R-apomorphine (Gassen et al. 1998), being protective at low concentrations, while at high concentrations become pro-oxidants. Immunohistochemistry for dopaminergic TH positive neurons revealed the truly neuroprotective ability of GT extract, since it not only prevented the decrease in TH protein content induced by MPTP, but also the dopaminergic neuron death in substantia nigra (Fig. 2 and Table 2). The full protection of striatal TH content and activity by GT extract, suggest that even lower concentrations may prevent MPTP neurotoxicity. The protective effect of tea polyphenols was described in a mouse model of cerebral ischemia, where intravenous injection of tea catechins or intraperitoneal injection of EGCG (25–50 mg/kg) immediately after ischemia improved the memory impairment (Matsuoka et al. 1995) and reduced hippocampal neuronal damage (Lee et al. 2000), respectively. We have found that EGCG also prevented the MPTP-induced depletion of DA, as well as that of TH protein and TH activity, where the lower dose, like GT, was more effective in preventing loss of TH activity. It is of interest to note the negative effect of EGCG on TH protein levels when administered alone, especially with the lower 2 mg/kg concentration. This effect occurred in parallel with increased TH activity. This result may be an adaptive response to the treatment, since EGCG prevented the decline in DA, TH protein and TH activity and the increase in SOD and catalase activities induced by MPTP. If EGCG was toxic, then MPTP neurotoxicity should have been exacerbated, however, this did not occur and would suggest a protective property of EGCG on brain neuronal cells. It reduced the levels of TH production by the neurons while increasing the activity of the enzyme. In this way DA neurons may function at a lower rate, while maintaining normal catecholamine levels.

GT extract and EGCG inhibit striatal MAO-B at high concentrations [IC50s = 713 µg/mL and 662 µm (303 µg/mL), respectively]. These concentrations are far higher than those for selective irreversible MAO-B inhibition by l-deprenyl, and those expected to be found in circulating blood after the administration of GT extract or EGCG. Furthermore, like S- and R-apomorphine (Grünblatt et al. 2001a), they did not increase the level of DA, 5-HT and NE in non-MPTP treated mice striatum, as would be expected if MAO was inhibited. Thus, MAO inactivation is not involved in neuroprotection exerted by these compounds. This finding also indicates that neither the tea extract nor EGCG are likely to act by interfering with the conversion of MPTP to its active metabolite MPP+, a reaction mediated by MAO-B.

MPTP increased both SOD and catalase activities in striatum, a finding that supports previous studies (Thiffault et al. 1995; Cassarino et al. 1997). This increase may reflect an adaptive response to due to a leakage of superoxide anion resulting from mitochondrial respiration impairment, as suggested by Thiffault et al. (1995). Alternatively, increased enzyme activity due to MPTP-induced gliosis could not be ruled out. However, pretreatment of mice with EGCG completely prevented catalase induction at the two concentrations assessed, and SOD levels were partially reduced by pretreatment with 2 mg/kg, while they were not affected by the higher dose. Interestingly, oral EGCG alone did also elevated the activity of SOD and catalase in striatum. This is the first report on the positive effects of EGCG in vivo on SOD and catalase activities in brain. A recent study reported that an antioxidant cocktail drink, β-catechin, consisting of green tea and sunflower seed extract, as well as ascorbic acid, caused a slight, not significant increase in striatal and midbrain SOD activity, after administration of 1 month (Komatsu and Hiramatsu 2000). Our findings suggest that EGCG neuroprotective activity may also involve regulation of antioxidant enzymes. Indeed, EGCG was shown to induce the expression of a reporter gene containing the antioxidant regulatory element present in the promoter of phase II antioxidant enzymes (Chen et al. 2000). GT extract given at relatively high concentrations for 4 weeks, decreased hepatic catalase activity in rats, while a lower concentration did not affect it (Bu-Abbas et al. 1998). However, in an other study a GT polyphenol fraction (0.2% w/v) given to mice for 30 days, was reported to increase the activities of the antioxidant enzymes glutathione peroxidase, catalase, and quinone reductase in small bowel, liver, and lungs (Khan et al. 1992); brain was not examined. In our experience, EGCG did not affect the activities of SOD and catalase in liver, except at the higher 10 mg/kg concentration. These discrepancies may reflect different routes of administration, concentration, animals and kind of drug used (individual polyphenol, tea polyphenolic fraction or tea extract). The observation that a dose of 2 mg/kg EGCG has differential effects on antioxidant enzymes in brain compared with liver tissue, may indicate the selectivity of EGCG on the antioxidant capacity in brain.

Both tea extract and EGCG exert neuroprotection in MPTP model of PD and 6-hydroxydopamine induced rat PC12 and human blastoma SH-SY5Y cell death in culture (Levites 2001). The lower GT and EGCG concentrations tested appear to best protect brain DA neurons. Although the exact specific cell targets of polyphenol action are still unrevealed, we assume that the antioxidant and iron chelating properties of polyphenols previously reported, may contribute to the neuroprotection induced by EGCG against MPTP. This neurotoxin has been shown to give rise to highly elevated contents of iron in the SNPC, which is thought to participate in OS via generation of hydroxyl radical by the Fenton reaction, and iron chelator desferal prevents its neurotoxicity.

It is now apparent that the future approach to neuroprotection must consider either the use of neuroprotective drug combinations (Slikker et al. 1999) or individual drugs having multipharmacological activities (Weinstock and Youdim 2001). This hypothesis is derived from clinical observations that a single drug therapy has not provided significant neuroprotection in PD, ischemia and other neurodegenerative diseases (Slikker et al. 1999). Tea extracts may be ideal because they contain variety of compounds with different known potent pharmacological actions. As a consequence the neuroprotective activities of individual and the combination of the purified major GT polyphenols may provide new insights into their major neuroprotective-related pathways. This is presently being investigated.


We thank Dr Bianca Fuhrman for her help with total polyphenol determination assay. We also thank Professor Karl Skorecki for kindly revising this manuscript.