JOURNAL OF MEDICINAL FOOD

J Med Food 13 (5) 2010, 1111–1115

# Mary Ann Liebert, Inc. and Korean Society of Food Science and Nutrition
DOI: 10.1089=jmf.2009.0255

Genoprotective Effects of the Green Tea-Derived

Polyphenol=Epicatechin Gallate in C6 Astroglial Cells
Renata T. Abib,1 André Quincozes-Santos,1 Caroline Zanotto,1 Fares Zeidán-Chuliá,2
Paula S. Lunardi,1 Carlos-Alberto Gonçalves,1 and Carmem Gottfried1
1
Neuroglial Plasticity Laboratory, Department of Biochemistry, Postgraduate Program in Biochemistry, Institute of Basic Health
Sciences, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil; and 2Medical Biochemistry
and Developmental Biology, Institute of Biomedicine, University of Helsinki, Helsinki, Finland

ABSTRACT In vitro and in vivo studies have recently reported significant chemopreventive effects of green tea-derived
polyphenols in different diseases. However, it remains unclear how such effects could be triggered. In order to elucidate the
effects of epicatechin gallate (ECG) in C6 cells, both by itself and against H2O2-induced genotoxicity, measurements of DNA
strand breaks and chromosome loss were performed. DNA damage was measured by comet and micronucleus assays. The
present study shows for the first time how ECG, the major green tea-derived polyphenol, is able to exert dose-dependent
genoprotective effects in an H2O2-induced toxicity model of C6 astroglial cells. We demonstrate that doses of ECG in a range
from 0.1 to 1 mM were able to completely prevent H2O2-induced genotoxicity in vitro. In contrast, considerably higher
concentrations of ECG (10 mM) were able to reverse previous positive effects in a dose- and time-dependent manner. The
same results were confirmed by both comet (F3,9 ¼ 336,148; P < .001) and micronucleus (F3,9 ¼ 23,228; P < .001) methods.
Together, our data show ECG as a dose-dependent genoprotective compound in C6 astroglial cells. This indicates that small
doses of polyphenols included in our diet could have beneficial effects on neural cells, contributing to prevention of oxidative
stress-associated brain pathologies. In addition, our data highlight the importance of strictly modulating doses and=or con-
sumption of antioxidant-fortified foods or additional supplements containing such beneficial molecules.

KEY WORDS:  antioxidant  astrocyte  DNA damage  epicatechin gallate  oxidative stress

INTRODUCTION hydroxyl groups, and other possible chemical substitutions

in their organic structure.6 The relative antioxidant activity
S ince early civilization, people have been using plants
as medicine, and their consumption has significantly
increased nowadays.1
of green tea-derived catechins is (-)-epigallocatechin
3-gallate ¼ ECG > (-)-epigallocatechin > (-)-epicatechin.7
Some studies have demonstrated that excessive production
Green tea, from Camellia sinensis, is obtained as a result
of free radicals and reactive oxygen species, such as hy-
of minimal oxidation during the production process. It is
drogen peroxide (H2O2), is indeed harmful, promotes aging,
widely consumed throughout the world and has received a
and can strongly influence the development of many dis-
great deal of attention since tea polyphenols were shown to
eases such as cancer8 as well as cardiovascular and neuro-
be strong antioxidants.2
degenerative disorders.9–11 Tea polyphenols are retained in
Green tea-derived polyphenols, such as catechins, may be
the brain and may exert neuroprotective effects whenever
responsible for reactive oxygen species quenching.3 The
their consumption is controlled.12 In fact, prolonged green
chemopreventive effects of green tea have been attributed to
tea ingestion is able to significantly protect hippocampal
biochemical activities induced by its polyphenolic constit-
proteins and lipids against oxidative damage.13
uents such as (-)-epicatechin gallate (ECG), (-)-epicatechin,
Previous works have demonstrated that intracellular re-
(-)-epigallocatechin 3-gallate, and (-)-epigallocatechin.4,5
active oxygen species and reactive nitrogen species such as
The capability of flavonoids to act as an antioxidant de-
superoxide anion, hydroxyl radicals, H2O2, lipid peroxyl
pends on their molecular structure, the position of their
radicals, nitric oxide, and peroxynitrite can lead to de-
struction of different cellular components, including lipids,
proteins, and DNA.6,14 However, cells can still survive
Manuscript received 12 November 2009. Revision accepted 28 January 2010.
thanks to antioxidant-dependent repair mechanisms.15
Address correspondence to: Renata T. Abib, M.Sc., Neuroglial Plasticity Laboratory, Characteristic properties of polyphenols such as penetration
Department of Biochemistry, Institute of Basic Health Sciences, Federal University of Rio
Grande do Sul, Rua Ramiro Barcelos, 2600–anexo–Bairro Santana, 90035–003 Porto
in brain tissue as well as their antioxidant and iron-chelating
Alegre, RS, Brazil, E-mail: [email protected] capabilities may make such compounds putative natural

1111
1112 ABIB ET AL.

drugs for the treatment of oxidative-related neurodegener- maldehyde in phosphate-buffered saline, stained with
ative diseases.16 0.2 mg=mL 40 ,60 -diamidino-2-phenylindole for 1 hour, and
Glial cells, particularly astrocytes, are known to exten- visualized under a fluorescent microscope (Nikon [Tokyo,
sively interact with neuronal elements in the brain, influ- Japan] inverted microscope using a TE-FM epi-fluorescence
encing their activity and exerting a prominent role in both accessory). Apoptotic cells were morphologically identified
protection and repair of nervous tissue after damage.17 The by nuclear shrinkage and chromatin condensation and=or
C6 cell line was originally derived from rat tumors in- fragmentation.
duced by N-nitrosomethylurea,18 and it is widely used as an
astrocyte-like cell line.19–25 Cytokinesis-block micronucleus assay
The main aims of our study were to elucidate the effects
of ECG in C6 cells, both by itself and against H2O2-induced Micronuclei are DNA-containing structures that result
genotoxicity by measuring DNA strand breaks and chro- from chromosomal loss during mitosis. They represent a
mosome loss. subgroup of all chromosomal aberrations. This makes the
micronucleus frequency test a widely accepted method for
investigating both in vitro and in vivo genotoxicity in human
EXPERIMENTAL PROCEDURES biomonitoring studies.28 The cytokinesis-block micronu-
Materials cleus technique was performed as previously described29,30
with minor modifications. After treatment, cells were incu-
Ethidium bromide, material for cell culture, and ECG bated with 2 mg=mL cytochalasin B for 24 hours, fixed with
were purchased from Sigma (St. Louis, MO, USA). chilled methanol=glacial acetic acid (3:1 vol=vol) for 5
40 ,60 -Diamidino-2-phenylindole was from Calbiochem (La minutes, and stained with Giemsa. Each data point repre-
Jolla, CA, USA). Dulbecco’s modified Eagle’s medium was sents the mean of eight independent experiments. In each
purchased from Gibco BRL (Carlsbad, CA, USA), and fetal experiment, 1,000 binucleated cells=sample were analyzed.
bovine serum was purchased from Cultilab (Campinas, SP,
Brazil). All other chemicals were obtained from regular Comet assay
commercial suppliers.
After different treatments as described above, C6 cells
Cell culture were detached by incubating in the presence of trypsin=
EDTA (0.05%). During trypsinization, cells were carefully
C6 astroglial cells were cultured as previously de- manipulated to avoid mechanical stress. Comet assay (single-
scribed.25 In our preparations, more than 95% of the cells cell gel electrophoresis) was performed as previously de-
exhibited positive immunoreactivity to glial fibrillary acidic scribed.27 In brief, slides were prepared by mixing 30 mL of
protein. Late passages of cells (100 passages, minimum) C6 cell suspension with 70 mL of low-melting -point agarose
were seeded in flasks and cultured in Dulbecco’s modified (0.75%). Following electrophoresis, slides were incubated
Eagle’s medium (pH 7.4) supplemented with 5% fetal bo- with 5 mg=mL ethidium bromide and left in the dark for 20
vine serum, 2.5 mg=mL amphotericin B (FungizoneÒ, minutes to stain the DNA. Images of 100 randomly selected
Bristol-Myers Squibb, Princeton, NJ, USA), and 100 U=L nuclei (50 nuclei from two replicated slides) were analyzed
gentamicin. Exponentially growing cells were incubated for for each treatment. Nuclei were scored visually for comet tail
1, 6, 12, and 24 hours at 378C in an atmosphere of 5% size based on an arbitrary scale of 0–4, i.e., ranging from no
CO2=95% air in Dulbecco’s modified Eagle’s medium (pH damage to extensive damage of DNA. Therefore, the damage
7.4) without serum in the absence or presence of ECG (0.1, index scale could range from 0 (all nuclei without tail, 100
1, or 10 mM). The concentrations of ECG used in these ex- cells0) to 400 (all nuclei with maximally elongated tails,
periments were obtained from previous determinations.26 100 cells4). Slides were viewed on a Nikon inverted mi-
croscope using a TE-FM epifluorescence accessory, and im-
H2O2 treatment ages were transferred to a computer with a digital camera
In order to investigate the genoprotective effects of ECG (Sound Vision Inc., Wayland, MA, USA).
against H2O2-induced oxidative stress, cells were pre-
incubated with different concentrations of ECG (0.1, 1, and Statistical analysis
10 mM) for 1 hour at 378C in an atmosphere of 5% CO2=95%
To verify the dose- and time course-dependent effect of
air in Dulbecco’s modified Eagle’s medium (pH 7.4) with-
ECG, we used one-way analysis of variance for repeated
out serum. After this time, the medium was maintained, and
measures, followed by a post hoc analysis (Tukey’s test).
1 mM H2O2 was added.27 Cells were incubated in the same
ECG effectsdifferent doses of H2O2 were analyzed sta-
conditions for an additional 30 minutes.
tistically by two-way analysis of variance followed by a post
hoc analysis (Tukey’s test). Data are mean  SEM values.
Nuclear morphology assay
Values with P < .05 were considered to be significant. All
C6 cells were cultured on circular glass coverslips and analyses were carried out in a PC-compatible computer
treated with or without 1 and 10 mM ECG for 1, 6, 12, or 24 using the Statistical Package for Social Sciences (SPSS)
hours. Cells were fixed for 20 minutes with 4% parafor- software (SPSS Inc., Chicago, IL, USA).
 EPICATECHIN GALLATE IN ASTROGLIAL CELLS 1113

Table 1. Micronucleus Frequency in C6 Cells After 1, 6, 12, or 24 hours. As shown in Figure 1, ECG induced
Incubation with Epicatechin Gallate DNA damage (F3,64 ¼ 1946,747; P < .001). This effect was
Micronucleus frequency among
concentration and time dependent (F3,9 ¼ 336,148;
1,000 binucleated cells at ECG (mM) P < .001). Only 2% of cells exposed to 10 mM ECG after 24
hours presented nuclear fragmentation by 40 ,60 -diamidino-2-
Time (hours) 0 0.1 1 10 phenylindole staining assay (data not shown).
a b b
1 1.0  0.09 1.3  0.11 1.4  0.12 1.6  0.12c
6 1.0  0.07a 1.3  0.09b 1.3  0.11b 1.7  0.12c Genoprotective effects of ECG against H2O2-induced
12 1.4  0.11b 2.0  0.15d 2.0  0.19d 2.3  0.17e DNA damage in C6 cells
24 1.6  0.13c 2.3  0.14e 2.4  0.17e 2.7  0.19f
To investigate the effect of ECG on DNA damage in-
Cells were incubated with epicatechin gallate (ECG) (0.1, 1.0, and 10 mM) duced by H2O2, cells were preincubated with different
for 1, 6, 12, or 24 hours. Data are mean  SEM values for micronucleus concentrations of ECG (0.1, 1, and 10 mM) for 1 hour. After
frequency among 1,000 binucleated cells, from eight independent experiments
this time, 1 mM H2O2 was added and maintained for 30
performed in duplicate. To verify the main effect of time course and different
doses of ECG, repeated-measures one-way analysis of variance was used, minutes (Fig. 2A). Such an assay was performed at the
followed by post hoc analysis with Tukey’s test. 1-hour time point taking into account that at this time ECG
abcdef
Values that do not share a common letter differ significantly at by itself had minimal influence on DNA integrity (Fig. 1).
P < .001. ECG-induced genoprotective effects after treatment are
shown in Figure 2B. The index of DNA damage observed
when cells were incubated in the presence of H2O2 in the
RESULTS absence of ECG was 42.5  3.9. In this context, ECG at 0.1
Effect of ECG on chromosome loss in C6 cells and 1.0 mM (20.0  1.3 and 22.0  1.5, respectively)
(F ¼ 47,529; P < .001) was able to significantly prevent the
Micronucleus frequency (Table 1) was increased by H2O2-induced genotoxicity in vitro.
ECG, compared to their respective control values
(F3,48 ¼ 262,381; P < .001). The analysis also indicate a
significant interaction between concentration and time
(F3,9 ¼ 23,228; P < .001).

Effect of ECG on nuclear morphology

and DNA strand breaks in C6 cells
To assess the direct effect of ECG on DNA integrity, cells
were incubated in the presence of 0.1, 1, and 10 mM ECG for

FIG. 1. In vitro time- and dose-dependent effects of ECG on DNA

damage. C6 astroglial cells were exposed to different concentrations FIG. 2. Protective effect of ECG against H2O2-induced DNA
of ECG (0.1, 1, and 10 mM) for different time intervals (1–24 hours). damage in C6 astroglial cells. Cells were preincubated for 1 hour in
The extent of damage to DNA was determined by the comet assay. the presence of ECG (0.1, 1, and 10 mM) before exposure to H2O2.
The index of DNA damage was calculated as described in Materials Culture medium was maintained, and then 0.1 mM H2O2 was added.
and Methods. Data are mean  SEM values of 16 experimental de- Cells were incubated in the same conditions for 30 minutes longer.
terminations performed in duplicates. To verify the dose- and time The extent of damage to DNA was determined by the comet assay.
course-dependent effect of ECG, repeated-measures analysis of var- The index of DNA damage was calculated as described in Materials
iance was performed, followed by post hoc analysis with Tukey’s and Methods. Data are mean  SEM values. Differences were sta-
test. abcdefColumns that do not share a common letter differ signifi- tistically analyzed by two-way analysis of variance followed by
cantly at P < .001. Tukey’s test. *Significant differences from control values (P < .001).
1114 ABIB ET AL.

DISCUSSION suggesting that, at these concentrations, ECG plays a ben-

eficial effect probably due to its antioxidant properties. The
It is very well documented that several redox-active reaction between ECG and hydroxyl radicals is particularly
compounds could have a dualistic effect, either beneficial important for preventing oxidative injury because hydroxyl
or toxic, depending on the concentration used.25,27,31,32 radicals have been shown to be highly responsible for sev-
The ability of polyphenols to scavenge reactive oxygen eral pathogeneses related to a wide range of diseases.
species depends on their chemical structures.33 In addition, Despite the fact that our in vitro data demonstrated a sig-
antioxidant=pro-oxidant activity of different redox-active nificant astroglial genoprotection upon ECG treatment, some
compounds, such as polyphenols, largely depends on the limitations should be under consideration regarding the dif-
levels consumed within the diet and may potentially cause ficulty of correlating in vitro experiments with in vivo treat-
DNA damage,34 probably triggered by direct binding of ments. Plasma ECG concentrations in rats range from 0.2 to
polyphenol to DNA.35 135 nM after 7.3 mg=kg ECG administration via the tail
As previously reported by our group,26 10 mM ECG in- vein.37 The pharmacokinetics of catechins in rat brain fetuses
duces morphological alterations in the C6 astroglial cell line after pregnant ingestion of green tea extract could result in a
(process-bearing cells) with a small increase in propidium ECG range from 10 to 80 pmol=g.38 Moreover, under in vitro
iodide incorporation after 24 hours of incubation (up to 4%). experiments, drug concentrations need to be usually higher
In the present work, we observed that, at least up to 24 than chronic in vivo treatments in order to mimic long-term
hours, ECG did not induce any indicative alteration in nu- physiological cell activities and their functions.27
clear morphology related to apoptotic process (data not
shown). DNA strand breaks in individual cells result from CONCLUSIONS
events such as direct scission of the DNA backbone.
In the present in vitro experimental model (1 hour in the Our results clearly demonstrated that ECG can affect
presence of ECG, before exposure to H2O2 for 30 minutes), DNA in our cell model, but, most importantly, it is able to
ECG was not cytotoxic to C6 cells. However, it is important protect against H2O2-induced DNA damage, showing a
to mention that index values below 30 are obtained from less statistically significant dose-dependent genoprotective ef-
prominent DNA strand breaks. In such context, the first 6 fect. This indicates that small doses of polyphenols included
hours of ECG incubation resulted in mild DNA damage. in our daily diet could play genoprotective effects in neural
Thereafter, ECG clearly induced a time- and dose-dependent cells and, therefore, could ameliorate oxidative stress-
genotoxicity. associated brain pathologies. Nevertheless, the dualistic
The degree of chromosome loss, analyzed by the micro- genotoxic effect exerted by high doses of ECG suggests that
nucleus frequency test (Table 1), reflects the capacity of the caution is recommended when large quantities of antioxi-
cells to resist oxidative stress and repair single-strand dants and=or supplements in foods are consumed.
breaks.34 As ECG by itself induced DNA damage at higher
doses, we decided to investigate if this genotoxic effect ACKNOWLEDGMENTS
could impair DNA repair. Actually, micronuclei frequency We would like to thank Dr. Richard Rodnight for critical
values from ECG exposure were significantly higher than reading and helpful discussion of the manuscript. This work
control values. The highest dose of ECG induced an increase was supported by the Conselho Nacional de Desenvolvi-
(about 60%) in all incubation time points (Fig. 1). It indi- mento Cientı́fico e Tecnológico, Coordenação de Aperfei-
cates that ECG genotoxicity at higher doses affects the re- çoamento de Pessoal de Nı́vel Superior, FINEP=Rede IBN
pair system, probably via a pro-oxidant effect.32 grant 01.06.0842-00, and the National Institute of Science
In order to investigate possible genoprotective effects of and Technology for Excitotoxicity and Neuroprotection.
ECG against DNA damage induced by oxidant conditions,
we used an experimental procedure previously established AUTHOR DISCLOSURE STATEMENT
in our group.27 In this model, cells were preincubated with
antioxidant for 1 hour before H2O2-induced insult. H2O2 is The authors declare that no competing financial interests
particularly attractive as an oxidant model because its cel- exist.
lular actions and fate are well understood. It readily crosses
the cellular membrane and gives rise to highly reactive REFERENCES
hydroxyl radicals, which have the ability to react with dif- 1. Silva CJ, Bastos JK, Takahashi CS: Evaluation of the genotoxic
ferent macromolecules, including DNA, proteins, and lipids, and cytotoxic effects of crude extracts of Cordia ecalyculata and
and to ultimately damage a cell.36 Echinodorus grandiflorus. J Ethnopharmacol 2010;127:445–450.
Following the observation that the 1-hour time point of 2. Khan N, Mukhtar H: Tea polyphenols for health promotion. Life
ECG exposure did not induce genotoxic effects by itself in Sci 2007;81:519–533.
C6 cells, we decided to keep the same time point for ECG 3. Al-Bloushi S, Safer AM, Afzal M, Mousa SA: Green tea mod-
preincubation, just before the H2O2 pulse at 30 minutes, ulates reserpine toxicity in animal models. J Toxicol Sci
which is known to be sufficient and enough to induce gen- 2009;34:77–87.
otoxic properties.27 We found that H2O2-induced geno- 4. Henning SM, Aronson W, Niu Y, et al.: Tea polyphenols and
toxicity was significantly prevented by 0.1 and 1 mM ECG, theaflavins are present in prostate tissue of humans and mice
 EPICATECHIN GALLATE IN ASTROGLIAL CELLS 1115

after green and black tea consumption. J Nutr 2006;136:1839– lular pH and tyrosine phosphorylation. Brain Res 2002;946:
1843. 12–23.
5. Tipoe GL, Leung TM, Hung MW, Fung ML: Green tea poly- 22. Cechin SR, Dunkley PR, Rodnight R: Signal transduction
phenols as an anti-oxidant and anti-inflammatory agent for car- mechanisms involved in the proliferation of C6 glioma cells
diovascular protection. Cardiovasc Hematol Disord Drug induced by lysophosphatidic acid. Neurochem Res 2005;30:603–
Targets 2007;7:135–144. 611.
6. Sun AY, Wang Q, Simonyi A, Sun GY: Botanical phenolics and 23. Chen TJ, Jeng JY, Lin CW, Wu CY, Chen YC: Quercetin inhi-
brain health. Neuromol Med 2008;10:259–274. bition of ROS-dependent and -independent apoptosis in rat gli-
7. Haque AM, Hashimoto M, Katakura M, et al.: Long-term ad- oma C6 cells. Toxicology 2006;223:113–126.
ministration of green tea catechins improves spatial cognition 24. Funchal C, Dos Santos AQ, Jacques-Silva MC, et al.: Branched-
learning ability in rats. J Nutr 2006;136:1043–1047. chain alpha-keto acids accumulating in maple syrup urine disease
8. Ravindranath MH, Ramasamy V, Moon S, Ruiz C, Muthu- induce reorganization of phosphorylated GFAP in C6-glioma
gounder S: Differential growth suppression of human melanoma cells. Metab Brain Dis 2005;20:205–217.
cells by tea (Camellia sinensis) epicatechins (ECG, EGC and 25. dos Santos AQ, Nardin P, Funchal C, et al.: Resveratrol increases
EGCG). Evid Based Complement Alternat Med 2009;6:523–530. glutamate uptake and glutamine synthetase activity in C6 glioma
9. Nomizu K, Hashida K, Makino R, Ohara S: Antioxidants from cells. Arch Biochem Biophys 2006;453:161–167.
steamed used tea leaves and their reaction behavior. Biosci 26. Abib RT, Quincozes-Santos A, Nardin P, et al.: Epicatechin
Biotechnol Biochem 2008;72:1682–1689. gallate increases glutamate uptake and S100B secretion in C6
10. Nakamura T, Lipton SA: Molecular mechanisms of nitrosative cell lineage. Mol Cell Biochem 2008;310:153–158.
stress-mediated protein misfolding in neurodegenerative dis- 27. Quincozes-Santos A, Andreazza AC, Nardin P, et al.: Resveratrol
eases. Cell Mol Life Sci 2007;64:1609–1620. attenuates oxidative-induced DNA damage in C6 glioma cells.
11. Uehara T, Nakamura T, Yao D, et al.: S-nitrosylated protein- Neurotoxicology 2007;28:886–891.
disulphide isomerase links protein misfolding to neurodegen- 28. Fink K, Brink A, Vienken J, Heidland A, Stopper H: Homo-
eration. Nature 2006;441:513–517. cysteine exerts genotoxic and antioxidative effects in vitro.
12. Mandel S, Amit T, Reznichenko L, Weinreb O, Youdim MB: Toxicol In Vitro 2007;21:1402–1408.
Green tea catechins as brain-permeable, natural iron chelators- 29. Fenech M: Cytokinesis-block micronucleus cytome assay. Nat
antioxidants for the treatment of neurodegenerative disorders. Protoc 2007;2:1084–1104.
Mol Nutr Food Res 2006;50:229–234. 30. Reyes S, Herrera LA, Ostrosky P, Sotelo J: Quinacrine enhances
13. Assuncao M, Santos-Marques MJ, Carvalho F, Lukoyanov NV, carmustine therapy of experimental rat glioma. Neurosurgery
Andrade JP: Chronic green tea consumption prevents age-related 2001;49:969–973.
changes in rat hippocampal formation. Neurobiol Aging 2009 31. de la Lastra CA, Villegas I: Resveratrol as an antioxidant and
May 1 [Epub ahead of print]. pro-oxidant agent: mechanisms and clinical implications. Bio-
14. Hwang ES, Kim GH: Biomarkers for oxidative stress status of chem Soc Trans 2007;35:1156–1160.
DNA, lipids, and proteins in vitro and in vivo cancer research. 32. Halliwell B: Dietary polyphenols: good, bad, or indifferent for
Toxicology 2007;229:1–10. your health? Cardiovasc Res 2007;73:341–347.
15. Gorbunova V, Seluanov A, Mao Z, Hine C: Changes in DNA 33. Yamamoto T, Lewis J, Wataha J, et al.: Roles of catalase
repair during aging. Nucleic Acids Res 2007;35:7466–7474. and hydrogen peroxide in green tea polyphenol-induced
16. Levites Y, Weinreb O, Maor G, Youdim MB, Mandel S: Green chemopreventive effects. J Pharmacol Exp Ther 2004;308:317–
tea polyphenol (-)-epigallocatechin-3-gallate prevents N-methyl- 323.
4-phenyl-1,2,3,6-tetrahydropyridine-induced dopaminergic neu- 34. Noel S, Kasinathan M, Rath SK: Evaluation of apigenin using in
rodegeneration. J Neurochem 2001;78:1073–1082. vitro cytochalasin blocked micronucleus assay. Toxicol In Vitro
17. Suzumura A, Takeuchi H, Zhang G, Kuno R, Mizuno T: Roles of 2006;20:1168–1172.
glia-derived cytokines on neuronal degeneration and regenera- 35. Kuzuhara T, Tanabe A, Sei Y, et al.: Synergistic effects of
tion. Ann N Y Acad Sci 2006;1088:219–229. multiple treatments, and both DNA and RNA direct bindings on,
18. Benda P, Lightbody J, Sato G, Levine L, Sweet W: Differentiated green tea catechins. Mol Carcinog 2007;46:640–645.
rat glial cell strain in tissue culture. Science 1968;161:370–371. 36. Halliwell B, Clement MV, Long LH: Hydrogen peroxide in the
19. Mangoura D, Sakellaridis N, Jones J, Vernadakis A: Early and human body. FEBS Lett 2000;486:10–13.
late passage C-6 glial cell growth: similarities with primary glial 37. Fu T, Liang J, Han G, Lv L, Li N: Simultaneous determination of
cells in culture. Neurochem Res 1989;14:941–947. the major active components of tea polyphenols in rat plasma by
20. Feng Z, Zhang JT: Protective effect of melatonin on beta- a simple and specific HPLC assay. J Chromatogr B Analyt
amyloid-induced apoptosis in rat astroglioma C6 cells and its Technol Biomed Life Sci 2008;875:363–367.
mechanism. Free Radic Biol Med 2004;37:1790–1801. 38. Chu KO, Wang CC, Chu CY, et al.: Pharmacokinetic studies of
21. Cechin SR, Gottfried C, Prestes CC, et al.: Astrocyte stellation in green tea catechins in maternal plasma and fetuses in rats. J
saline media lacking bicarbonate: possible relation to intracel- Pharm Sci 2006;95:1372–1381.