Baicalein inhibition of oxidative-stress-induced apoptosis via modulation of ERKs activation and induction of HO-1 gene expression in rat glioma cells C6 .

Baicalein inhibition of oxidative-stress-induced apoptosis

via modulation of ERKs activation and induction of

HO-1 gene expression in rat glioma cells C6

Yen-Chou Chen

a,

⁎

, Jyh-Ming Chow

b

, Cheng-Wei Lin

a

, Chin-Yen Wu

a

, Shing-Chuan Shen

c,d

a

Graduate Institute of Pharmacognosy, School of Pharmacy, Taipei Medical University, Taipei, Taiwan

b

Section of Hematology-Oncology, Department of Internal, Medicine, Taipei Municipal Wan-Fang Hospital, Taipei Medical University, Taipei, Taiwan

c

Department of Dermatology, School of Medicine, Taipei Medical University, Taipei, Taiwan

d

Department of Dermatology, Taipei Municipal Wan-Fang Hospital, Taipei, Taiwan Received 13 January 2006; revised 9 May 2006; accepted 14 May 2006

Available online 19 May 2006

Abstract

In the present study, we examined the protective mechanism of baicalein (BE) and its glycoside, baicalin (BI), on hydrogen-peroxide (H

2

O

2

)-induced cell death in rat glioma C6 cells. Results of the MTT assay, LDH release assay, and morphological observation showed that H

2

O

2

addition

reduced the viability of C6 cells, and this was prevented by the addition of BE but not BI. Incubation of C6 cells with BE significantly decreased

the intracellular peroxide level induced by H

2

O

2

according to flow cytometric analysis using DCHF-DA as a fluorescent substrate. Suppression of

H

2

O

2

-induced apoptotic events including DNA ladders, hypodiploid cells, and activation of caspases 3, 8, and, 9 by BE but not BI was identified

in C6 cells. The cytotoxicity and phosphorylation of ERK proteins induced by H

2

O

2

were blocked by the ERK inhibitor PD98059. Catalase

addition prevented H

2

O

2

-induced ROS production, ERKs protein phosphorylation, and cell death, and BE dose-dependently inhibited H

2

O

2

-induced ERK protein phosphorylation in C6 cells. These data suggest that ROS-scavenging activity is involved in BE prevention of H

2

O

2

-induced

cell death via blocking ERKs activation. Additionally, BE but not BI induced heat shock protein 32 (HSP32; HO-1) protein expression in both

time- and dose-dependent manners, but not heme oxygenase 2 (HO-2), heat shock protein 70 (HSP70), or heat shock protein 90 (HSP90) protein

expression. In the absence of H

2

O

2

, BE induces ERKs protein phosphorylation, and HO-1 protein expression induced by BE was blocked by the

addition of cycloheximide, actinomycin D, and the ERK inhibitor PD98059. The addition of the HO inhibitor ZnPP inhibited the protective effect

of BE against H

2

O

2

-induced cytotoxicity in C6 cells according to the MTT assay and apoptotic morphology under microscopic observation,

accompanied by blocking the ROS-scavenging activity of BE in C6 cells. However, BE treatment was unable to protect C6 cells from

C2-ceramide-induced cell death. These data indicate that BE possesses abilities to inhibit ROS-mediated cytotoxic effects through modulation of

ERKs activation and induction of HO-1 protein expression. The role of HO-1 in ROS-scavenging activity of BE is proposed.

© 2006 Elsevier Inc. All rights reserved.

Keywords: Baicalein; Hydrogen peroxide; Apoptosis; ERKs; HO-1

Introduction

Baicalein (BE) is one of the major flavonoids in Scutellaria

baicalensis, which has long been extensively used in Chinese

herbal medicine. Several biological effects of BE such as

anti-viral, anti-inflammation, anti-hepatotoxicity, and anti-tumor

properties have been reported (

Ahn et al., 2001; Huang et al.,

2005; Hwang et al., 2005

). Most activities of BE are attributed to

its antioxidant and prooxidant capacities.

Wang et al. (2004)

indicated that BE induced apoptosis in tumor cells via the

Abbreviations: BE, baicalein; BI, baicalin; LPS, lipopolysaccharide; TPA,

12-O-tetradecanoylphorbol 13-acetate; H2O2, hydrogen peroxide; HO-1, heme

oxygenase 1; HSP70, heat shock protein 70; HSP90, heat shock protein 90; HO-2, heme oxygenase 2; LDH, lactate dehydrogenase; ZnPP, zinc protoporphyrin; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; NBT, nitroblue tetrazolium; BCIP, 5-bromo-4-chloro-3-indolyl phosphate; ERKs, extracellular regulated kinases; JNKs, c-Jun N-terminal kinases; ROS, reactive oxygen species.

⁎ Corresponding author. Fax: +1 886 2 23787139. E-mail address:[email protected](Y.-C. Chen).

0041-008X/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2006.05.008

production and accumulation of intracellular reactive oxygen

species (ROS). There is additional evidence showing that BE

exhibits protective effects against ischemia-induced apoptosis in

cardiomyocytes via reducing hydrogen peroxide production

(

Shao et al., 1999, 2002

). Increasing evidence indicates that BE

possesses the ability to protect against cellular damage induced

Fig. 1. BE but not BI protects glioma C6 cells from H2O2-induced cytotoxicity. (A) Chemical structures of baicalein (BE) and baicalin (BI). BI is a BE glycoside and

contains a glucoside moiety at C7 of BE. (B) C6 cells were treated with BE or BI (100μM) for 30 min followed by H2O2(400μM) incubation for a further 12 h. The

morphological changes were detected under microscopic observation via Giemsa staining. (C) As described in (B), C6 cells were treated with BE or BI (50 and 100μM) for 30 min followed by H2O2(400μM) incubation for a further 12 h. The viability of C6 cells under different treatments was examined by an MTT assay as

described in Materials and methods. (D) Under conditions described in (C), the amount of LDH released from cells after different treatments was detected, and data are expressed as the percentage of cytotoxicity as described in Materials and methods. C or CON, control group. Data are expressed as the mean ± SE. **pb 0.01 indicates a significant difference from the H2O2-treated group.

by ROS and chemotoxic agents.

Shieh et al. (2000)

indicated

that BE and BI possess potent eliminative activities on free

radical production via suppressing xanthine oxidase activity. In

the presence of lipopolysaccharide (LPS) treatment, BE prevents

the degeneration of dopaminergic neurons induced by LPS, and

the suppression of nitric oxide (NO)-induced cytotoxicity by BE

in microglia cells via blocking NF-

κB activation was identified

(

Chen et al., 2004a,b; Li et al., 2005

). Although several

biological activities of BE have been reported, the mechanism

by which BE prevents oxidative-stress-induced cell death in

glioma cells is still unclear.

Structural modification of flavonoids occurs extensively in

plants, and glycosylation, one type of structural modification,

commonly appears in the metabolism of flavonoids (

Hollman et

al., 1999; Rivera et al., 2004

). Several previous studies indicated

that glycosylation of flavonoids increases their hydrophilicity and

induces resistance to enzyme oxidation in plants (

Birt et al., 2001;

Chen et al., 2001; Regev-Shoshani et al., 2003

). Evidence related

to the effects of glycoside addition on the biological activities of

flavonoids is still lacking. Our previous study demonstrated that

aglycon flavonoids such as quercetin showed more significant

NO inhibitory activity and apoptosis-inducing activities than its

glycosides, rutin and quercitrin, in RAW264.7 macrophages and

human leukemia HL-60 cells, respectively (

Chen et al., 2001;

Shen et al., 2003

). Additionally, the aglycons, hesperitin and

naringenin, but not their respective glycosides, hesperidin and

naringin, inhibit NO production in macrophages via inducing

heme oxygenase 1 (HO-1) protein expression, as well as inducing

apoptosis in leukemia cells (

Chen et al., 2003; Lin et al., 2005

).

These data indicate that glycosides may play as a negative moiety

in the biological actions of flavonoids. However, the effect of

glycoside in BE prevention of oxidative-stress-induced cell death

in glioma cells has yet to be investigated.

It has been reported that ROS participate in causing such

human diseases as diabetes, tumors, atherosclerosis, stroke, and

neurodegenerative diseases such as Parkinson's disease, stroke,

amyotrophic lateral sclerosis, and Alzheimer's disease (

Crack

and Taylor, 2005; Henze et al., 2005; Laufs et al., 2005; Park

et al., 2005

). Glial cells play a major role in maintaining the

functions of neurons in the brain, and several studies have

indicated that ROS accumulation in glial cells leads to cell death

and dysfunction of the surrounding neurons (

de Bernardo et al.,

2004; Juravleva et al., 2005; Qian et al., 2005

). Therefore, agents

with the ability to protect glial cells from ROS-dependent cell

death may possess the potential to treat neurodegenerative

diseases. Heme oxygenases (HOs) are enzymes which catalyze

heme to bilirubin, biliverdin, carbon monoxide (CO), and

ferrous iron (Fe

2+

). At least three HOs including HO-1, HO-2,

and HO-3 have been identified. HO-1 is inducible in response to

several stimuli such as heme, heavy metals, LPS, and

inflammatory cytokines. Previous studies showed that induction

of HO-1 gene expression protects cells from cell death (

Chen

et al., 2000, 2002; Choi et al., 2004; Petrache et al., 2000

), and

upregulation of the HO-1 protein by the HO-1 inducers, hemin

and cadmium, induces resistance to apoptotic stimuli in human

gastric cancer cells (

Liu et al., 2004

). Overexpression of the

HO-1 protein protects vascular smooth muscle cells from

angio-tensin-II-induced damages (

Morita et al., 2005

). These data

support the protective function of the HO-1 protein; however,

the role of the HO-1 protein in the biological action of BE is still

undefined.

In the present study, we investigated the mechanism of BE

and its glycoside, baicalin, on oxidative-stress (H

2

O

2

)-induced

apoptosis in rat glioma C6 cells, and the roles of the HO-1

protein, ERK protein, and ROS-scavenging activity in the

pre-ventive mechanism of BE were investigated.

Materials and methods

Cell culture. Rat glioma C6 cells from ATCC (American Type Culture Collection; Rockville, MD) were incubated in RPMI-1640 medium supplemented with 2 mM glutamine, antibiotics (100 U/ml of penicillin A and 100 U/ml of streptomycin), and 10% heat-inactivated fetal bovine serum and maintained at 37 °C in a humidified incubator containing 5% CO2.

Chemicals. The structurally related flavonoids including baicalein and baicalin were obtained from Sigma Chemical (St. Louis, MO). (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide) (MTT), hydrogen peroxide (H2O2), zinc

protophyrin (ZnPP), actinomycin D, cycloheximide, 2 ′,7′-dichlorodihydrofluor-escein-diacetate (DCHF-DA), and propidium iodine (PI) were also obtained from Sigma. The antibodies of anti-HO-1, anti-α-tubulin, pERKs, HO-2, anti-iNOS, and anti-HSP90 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). PD98059, Ac-DEVD-pNA, Ac-IETD-pNA, and Ac-LEHD-pNA were obtained from Calbiochem (La Jolla, CA).

Cell viability. MTT was used as an indicator of cell viability as determined by its mitochondrial-dependent reduction to formazone. Cells were plated at a density of 4 × 105cells/well in 24-well plates for 12 h followed by treatment with different concentrations of BE for a further 12 h. Cells were washed with PBS three times, and MTT (50 mg/ml) was added to the medium for 4 h. Then, the supernatant was removed, and the formazone crystals were dissolved using 0.04 N HCl in isopropanol. The absorbance was read at 600 nm with an ELISA analyzer (Dyna-tech MR-7000; Dyna(Dyna-tech Laboratories).

LDH release assay. The percentage of LDH release was expressed as the proportion of LDH released into the medium compared to the total amount of LDH present in cells treated with 2% Triton X-100. The activity was monitored as the oxidation of NADH at 530 nm by an LDH assay kit (Roche Applied Science). The cytotoxicity (%) was determined by the equation [(OD530 of the treated group − OD530 of the control group)/(OD530 of the Triton X-100-treated group− OD530 of the control group)] × 100%.

Caspase activity assay. After different treatments, glioma C6 cells were collected and washed three times with PBS and resuspended in 50 mM Tris– HCl (pH 7.4), 1 mM EDTA, and 10 mM ethyleneglycoltetraacetic acid (EGTA). Cell lysates were clarified by centrifugation at 20,000 × g for 3 min, and clear lysates containing 100μg of protein were incubated with 100 μM enzyme-specific colorimetric substrates including Ac-DEVD-pNA for caspase 3/CPP32, Ac-IETD-pNA for caspase 8, and Ac-LEHD-pNA for caspase 9 at 37 °C for 1 h. Alternative activity of the indicated caspases was described as the cleavage of colorimetric substrate by measuring the absorbance at 405 nm.

Western blotting. Total cellular extracts were prepared, and an equal amount of proteins from each group was separated on 8%–12% SDS-polyacrylamide minigels followed by transfer to immobilon polyvinylidenedifluoride membranes (Millipore, Bedford, MA). Membranes were incubated with 1% bovine serum albumin and then incubated with specific antibodies overnight at 4 °C. Protein expression was detected by staining with nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-3-indolyl phosphate (BCIP) (Sigma).

DNA gel electrophoresis. Cells under different treatments were collected, washed twice with PBS, lysed in 80μl of lysis buffer (50 mM Tris (pH 8.0), 10 mM ethylenediaminetetraacetic acid (EDTA), 0.5% sodium sarkosinate, and 1 mg/ml proteinase K) for 3 h at 56 °C, and then treated with 0.5 mg/ml RNase A for an additional 1 h at 56 °C. DNA was extracted with phenol/chloroform/

isoamyl alcohol (25/24/1, v/v) before loading. Samples were mixed with loading buffer (50 mM Tris, 10 mM EDTA, 1% (w/w) low-melting point agarose, and 0.025% (w/w) bromophenol blue) and loaded onto a pre-solidified 2% agarose gel containing 0.1 mg/ml ethidium bromide. The agarose gels were run at 50 V for 90 min in TBE buffer, after which they were observed and photographed under UV light.

Determination of ROS production by flow cytometry analysis and fluorescent microscopic observation. The production of ROS was monitored by flow cytometry using DCHF-DA. This dye is a stable compound that readily diffuses into cells and is hydrolyzed by intracellular esterase to yield DCHF, which is trapped within cells. Hydrogen peroxide or low-molecular-weight hydroperoxides produced by cells oxidize DCHF to the highly fluorescent compound, 2 ′,7′-dichlorofluorescein (DCF). Thus, the fluorescence intensity is proportional to the amount of peroxide produced by cells. In the present study, cells were treated with BE for 30 min followed by H2O2addition for 1 h. Then, compound-treated cells

were washed twice with PBS to remove the extracellular compounds, and DCHF-DA (100μM) green fluorescence was added, excited using an argon laser, and detected using a 525-nm (FL1-H) band-pass filter by flow cytometric analysis. As the same part of experiments, the green fluorescence in cells was observed by fluorescent microscope.

Statistical analysis. All experiments were performed in triplicate. ANOVA was used to determine differences among groups. If a significant difference was found, t test was used to determine the location of the difference between indicated two groups; values of pb 0.05 and p b 0.01 were considered statistically significant.

Results

BE but not BI protects glioma C6 cells from H

2

O

2

-induced cell

death

The structures of BE and BI are depicted in

Fig. 1

A; BI

contains a glycoside at C7 of BE (

Fig. 1

A). In order to

investigate if BE or BI exhibits the ability to protect glioma

cells from ROS-induced cell death, C6 cells were treated with

H

2

O

2

(400

μM) in the presence or absence of BE or BI

incubation, and the viability of cells was examined by

morphological observation and MTT and LDH release assays.

Results of the morphological observation via Giemsa staining

indicated that H

2

O

2

induced the appearance of condensed cells

(dark ones) in C6 cells, which was prevented by the addition of

BE but not BI (100

μM) (

Fig. 1

B). Data from the MTT and

LDH release assays showed that a decrease in the viability of

C6 cells was detected in H

2

O

2

-treated C6 cells and that this

was blocked by adding BE but not BI (

Figs. 1

C, D). These

data revealed that BE but not BI protected C6 cells from H

2

O

2

-induced cytotoxicity.

Fig. 2. BE inhibits H2O2-induced apoptotic events in C6 cells. (A) BE prevention of DNA fragmentation induced by H2O2in C6 cells. C6 cells were treated with BE or BI

(50 or 100μM) for 30 min followed by H2O2treatment for a further 12 h. The integrity of the DNA in cells under different treatments was determined via agarose

electrophoresis. (B) BE inhibition of H2O2-induced hypodiploid cells (sub-G1) in C6 cells. As described in (A), the ratio of hypodiploid cells was measured by flow

cytometric analysis. (C–E) BE inhibition of H2O2-induced caspase 3, 8, and 9 enzyme activity in glioma C6 cells. Activities of caspase 3 (C), caspase 8 (D), and caspase 9 (E)

enzymes stimulated by H2O2in the presence or absence of BE or BI (100μM) were detected using specific peptidyl substrates as described in Materials and methods. (F) BE

reversed H2O2-induced decreases in the expressions of pro-caspases 3, 8, and 9 and pro-PARP protein. As described previously, the expression of the pro-form of caspases 3,

8, and 9 and the PARP protein was detected by Western blotting. The expression ofα-tubulin protein was used as an internal control. Each value is presented as the mean ± SE. **pb 0.01 indicates a significant difference from the control group; and##pb 0.01 indicates a significant difference between indicated groups.

BE inhibits H2O2-induced apoptotic events in rat glioma C6

cells

We further investigated the effect of BE and BI on H

2

O

2

-induced apoptotic characteristics including DNA ladders,

hypo-diploid cells (sub-G1), and caspase activation in rat glioma cells

C6. Results in

Fig. 2

A show that H

2

O

2

induced a loss in the

integrity of DNA in accordance with the appearance of

frag-mented DNA in C6 cells, and this was prevented by the addition

of BE but not BI to C6 cells (

Fig. 2

A). Data of the flow cytometric

analysis showed that the ratio of hypodiploid cells induced by

H

2

O

2

was attenuated by the addition of BE but not BI (

Fig. 2

B).

An increase in the activation of the indicated caspases, including

caspases 3, 8, and 9, was detected in H

2

O

2

-treated C6 cells using

specific colorimetric peptidyl substrates (Ac-DEVD-pNA for

caspase 3, Ac-IETD-pNA for caspase 8, and Ac-LEHD-pNA for

caspase 9), and those effects were significantly attenuated by BE

but not BI (

Figs. 2

B–D). Data from Western blotting showed that

H

2

O

2

treatment induced decreases in the levels of pro-caspases 3,

8, and 9 and caspase 3 substrate PARP protein in C6 cells, and

these were blocked by BE but not BI (

Fig. 2

E). A similar level of

α-tubulin protein expression in each lane was used as an internal

control. This suggests that BE prevents H

2

O

2

-induced cell death

via blocking cells from undergoing apoptosis.

Activation of ERKs protein via phosphorylation induction is

involved in H2O2-induced C6 cell death which was blocked by

BE

Since activation of ERKs by H

2

O

2

has extensively been

shown, we investigated the role of ERK activation in the

pre-ventive mechanism of BE. Results in

Fig. 3

A show that the

Fig. 3. Involvement of ERK activation in H2O2-induced cell death. (A) H2O2-induced ERK protein phosphorylation in a time-dependent manner. C6 cells were treated

with H2O2(400μM) for different times (10, 20, 40, 60, and 80 min), and the expressions of phosphorylated ERK (p-ERK) and total ERK (t-ERK) proteins were

detected by Western blotting using specific antibodies. (B) (Lower panel) PD98059, a specific ERK activity inhibitor, suppressed H2O2-induced cytotoxicity in C6

cells according to MTT assay. Cells were treated with different doses of PD98059 (2.5, 5, 10, and 20μM) for 30 min followed by H2O2treatment for a further 12 h. The

viability of cells was measured by the MTT assay. (Upper panel) PD98059 dose-dependently attenuated H2O2-induced phosphorylation of ERK protein. Cells were

treated with different doses of PD98059 (2.5, 5, 10, and 20μM) for 30 min followed by H2O2treatment for a further 20 min. The expressions of phosphorylated ERK

(p-ERK) and total ERK (t-ERK) proteins were detected by Western blotting using specific antibodies. (C) Catalase addition inhibited H2O2-induced ERK protein

phosphorylation in accordance with suppression of the H2O2-induced cytotoxic effects in C6 cells. (Upper panel) C6 cells were treated with catalase (25, 50, and

100 U/ml) for 30 min followed by H2O2(400μM) treatment for 20 min (ERK protein detection) and 12 h (MTT assay). The expressions of p-ERK and t-ERK proteins

(upper panel) and the viability of C6 cells (lower panel) under different treatments were examined by Western blotting and the MTT assay, respectively. (D) BE dose-dependently inhibited ERK activation induced by H2O2. Cells were treated with different doses of BE (25, 50, and 100μM) for 30 min followed by H2O2treatment for

a further 20 min. The expressions of p-ERK and t-ERK proteins were detected by Western blotting using specific antibodies. Each value is presented as the mean ± SE. **pb 0.01 indicates a significant difference from the control group; and##pb 0.01 indicates a significant difference from H2O2-treated groups.

expression of phosphorylated ERKs proteins was induced in

H

2

O

2

-treated C6 cells, and the time course of phosphorylated

ERKs proteins was appeared in 10 to 60 min after H

2

O

2

treat-ment. Incubation of C6 cells with the chemical ERK inhibitor,

PD98059, dose-dependently protected C6 cells from H

2

O

2

-induced cytotoxicity according to the MTT assay (

Fig. 3

B; lower

panel), accompanied by suppression of H

2

O

2

-induced ERKs

protein phosphorylation (

Fig. 3

B; upper panel). Catalase addition

dose-dependently prevented H

2

O

2

-induced cytotoxicity according to

the MTT assay by blocking ERKs protein phosphorylation in C6

cells (

Fig. 3

C). Furthermore, results in

Fig. 3

D show that BE

dose-dependently inhibited H

2

O

2

-induced ERK protein phosphorylation

in C6 cells. These data indicated that blocking ERK protein

phosphorylation induced by H

2

O

2

is involved in the action of BE in

rat glioma C6 cells.

ROS-scavenging activity participates in BE protection against

H2O2-induced cell death in rat glioma C6 cells

It is important to examine if BE protects C6 cells from H

2

O

2

-induced apoptosis via reducing intracellular ROS production.

DCHF-DA has been used to examine changes in intracellular

peroxide levels through flow cytometric analysis. Results in

Fig. 4

A indicate that H

2

O

2

(400

μM) treatment significantly

induced intracellular peroxide production, and the percentage of

ROS-overexpressed C6 cells (M1) is 90.4 ± 2.1%. A decrease

Fig. 4. Free radical-scavenging activities of BE and BI by the DCHF-DA assay. (A) C6 cells were treated with different doses of BE and BI (25, 50, and 100μM) or catalase (50 U/ml) for 30 min followed by adding H2O2(400μM) for a further 1 h. The level of intracellular peroxide was detected by the DCHF-DA assay using flow cytometric

analysis. The percentage of cells under different treatments in M1 was detected and expressed as the mean ± SE from three independent experiments. *pb 0.05, **p b 0.01 indicate a significant difference between indicated groups, as analyzed by Student's t test. (B) As described in (A), cells were treated with or without BE (100μM), BI (100μM), or catalase (CAT; 50 U/ml) for 30 min followed by H2O2(400μM) treatment. The fluorescence in cells was detected by fluorescent microscopic observation.

in intracellular fluorescent intensity was detected in C6 cells

without DCHF-DA addition as a negative control to verify the

specificity of the reaction (N-CON). Incubation of C6 cells with

different doses (25, 50, and 100

μM) of BE significantly

re-duced H

2

O

2

(400

μM)-induced intracellular peroxide

produc-tion, and the percentages of M1 in 25, 50, and 100

μM of

BE-treated cells are 34.5 ± 1.8%, 36.5 ± 3.2%, and 32.1 ± 0.9%,

respectively. BI performs significant, but less effective than BE,

inhibitory effect on H

2

O

2

-induced ROS production, and the

percentages of M1 in 25, 50, and 100

μM of BI-treated cells are

62.7 ± 4.6%, 66.8 ± 3.6%, and 56.9 ± 2.5%, respectively.

Catalase (CAT) reduction of H

2

O

2

-induced intracellular

perox-ide production to the control level was used as a positive

control. As the same part of experiment, data of fluorescent

microscopic observation indicated that an increase in

intracel-lular fluorescent intensity was detected in H

2

O

2

-treated C6 cells

and was blocked by BE and catalase. These data suggest the

ROS-scavenging activity of BE in C6 cells.

BE but not BI induces HO-1 protein expression in rat glioma

C6 cells

Induction of HSPs has been shown to protect cells from

oxidative-stress-induced cell death, therefore we investigated

the effect of BE on the expressions of HO-1, HO-2, HSP70, and

HSP90 in C6 cells. As illustrated in

Fig. 5

A, BE but not BI

dose-dependently induced HO-1, but not HO-2, HSP70, or

HSP90, protein expression in C6 cells. In the same part of the

experiment, BE induced HO-1 protein expression in a

time-dependent manner (

Fig. 5

B). Both with and without FBS, BE

expressed similar inductive effects on HO-1 protein expression

in C6 cells (

Fig. 5

C). An increase in HO-1 protein expression

by the traditional HO-1 inducer hemin was used as a positive

control here.

Induction of ERKs, but not p38 or JNKs, protein phosphorylation

is involved in BE induction of HO-1 protein expression

Data from

Fig. 6

A show that the induction of ERKs, but not

p38 or JNKs, protein phosphorylation was detected in

BE-treated glioma C6 cells by Western blotting using specific

antibodies. No change in the expression of total ERKs, p38, and

Fig. 5. BE but not BI induced HO-1 protein expression in C6 cells. (A)

Dose-dependent induction of HO-1 protein by BE (but not BI) in C6 cells. Cells were treated with BE or BI (50, 100, and 200μM) for 12 h, and the expressions of HO-1, HO-2, HSP70, and HSP90 protein were detected by Western blotting. (B) BE and hemin (HEM) time-dependently induced HO-1 protein expression in C6 cells. Cells were treated with BE (100μM) or hemin (HEM, 10 μM) for different times (2, 4, 8, and 12 h), and the expressions of HO-1, HO-2, and HSP90 protein were detected. (C) A similar inductive pattern of HO-1 protein stimulated by BE (50, 100, and 200μM) was detected in the presence (FBS) or absence (SF) of 10% FBS. Cells were treated with different doses of BE with or without FBS, and the expressions of HO-1, HO-2, and HSP90 proteins were detected by Western blotting.

Fig. 6. Induction of phosphorylated ERK (but not p38 or JNK) proteins is involved in BE-induced HO-1 protein expression. (A) BE induced ERK but not p38 or JNK protein phosphorylation in C6 cells. C6 cells were treated with BE or BI (100μM) for different times (20, 40, and 60 min), and the expressions of phosphorylated and total ERK, p38, and JNK proteins were detected by Western blotting using specific antibodies. (B) PD98059 dose-dependently inhibited BE-induced HO-1 protein expression in C6 cells. Cells were treated with PD98059 (5, 10, and 20μM) for 30 min followed by the addition of BE (100 μM) for a further 12 h. The expressions of HO-1 and HO-2 proteins were detected by Western blotting using specific antibodies. (C) Addition of the translational inhibitor cycloheximide (CHX; 1 μg/ml) or the transcriptional inhibitor actinomycin D (Act D; 1μg/ml) reduced HO-1 protein expression induced by BE (100μM). C6 cells were treated with CHX or Act D for 30 min followed by BE treatment for a further 12 h (for detecting HO-1, HO-2, and HSP90 proteins) or 20 min (for detecting phosphorylated and total ERK proteins).

JNKs protein was detected to verify similar amount of proteins

loaded in each lane. Adding PD98059 blocked HO-1 (but not

HO-2) protein expression induced by BE (

Fig. 6

B).

Incuba-tion of C6 cells with the translaIncuba-tional inhibitor cycloheximide

(CHX) and the transcriptional inhibitor actinomycin D (Act D)

significantly inhibited HO-1, but not HO-2, HSP70, or HSP90,

protein expression in BE-treated C6 cells. However, CHX and

Act D were unable to block ERK protein phosphorylation

induced by BE (

Fig. 6

C). These data suggest that HO-1 protein

expression induced by BE occurs via de novo protein synthesis

and that it is located downstream of ERK activation in C6 cells.

The HO-1 activity inhibitor zinc protoporphyrin suppresses the

inhibitory effect of BE on H2O2-induced cell death

Zinc protoporphyrin (ZnPP) has been used as an HO inhibitor.

In order to delineate if HO-1 induction participates in BE protection

of C6 cells against H

2

O

2

-induced cytotoxicity, cells were treated

with BE (100

μM) for 8 h to induce HO-1 protein expression

followed by adding different doses (0.5, 1, 2

μM) of ZnPP for

30 min and incubated with H

2

O

2

for a further 12 h. Results of the

MTT assay showed that the protective effect of BE on H

2

O

2

-induced cytotoxicity was dose-dependently reversed by the

addi-tion of ZnPP (

Fig. 7

A). Data of the morphological observations

indicated that condensed cells reappeared in ZnPP-treated cells in

the presence of BE and H

2

O

2

treatments (

Fig. 7

B). Analysis of

intracellular peroxide production shows that ZnPP suppresses the

ROS-scavenging activity of BE, and the percentages of M1 in

control, H

2

O

2

, ZnPP, H

2

O

2

plus BE, H

2

O

2

plus ZnPP, and H

2

O

2

plus BE plus ZnPP-treated C6 cells are 6.2 ± 0.9%, 92.5 ± 2.7%,

17.2 ± 1.8%, 22.8 ± 3.1%, 75.5 ± 4.3%, and 41.6 ± 1.5%,

respectively (

Fig. 7

C). It suggests that induction of HO-1 gene

expression participates in BE against H

2

O

2

-induced cell death and

ROS production in C6 cells.

Fig. 7. The chemical HO-1 enzyme inhibitor ZnPP reversed the protective effect and ROS-scavenging activity of BE against H2O2-induced cell death. (A) C6 cells

were treated with BE (100μM) for 8 h followed by incubating with different doses of ZnPP (0.5, 1, and 2 μM) for 30 min and addition of H2O2(400μM) into the cells

for a further 12 h. The viability of cells was detected by the MTT assay. (B) As described in (A), morphological changes in the condition with or without BE (100μM) or ZnPP (2μM) followed by H2O2stimulation for 12 h were observed microscopically using Giemsa staining. (C) Alternations in intracellular peroxide production

were detected by DCHF-DA assay. C6 cells were treated with BE (100μM) for 8 h followed by incubating with or without ZnPP (2 μM) for 30 min. H2O2(400μM)

BE expresses no protective effect against C2-ceramide-induced

cell death in rat glioma C6 cells

We examined if BE possesses an inhibitory effect on

intracellular ROS-induced cell death in C6 cells. C2-ceramide

has been shown to induce apoptosis in several cells via inducing

intracellular ROS production. Therefore, we investigated the

effect of BE on the C2-ceramide-induced cytotoxic effect in C6

cells. As illustrated in

Fig. 8

A, the addition of BE and BI

produced no protective effect on the C2-ceramide-induced

cytotoxic effect in C6 cells according to the MTT assay. Loss of

DNA integrity was detected in C2-ceramide-treated C6 cells;

however, neither BE nor BI inhibited C2-ceramide-induced

DNA fragmentation (

Fig. 8

B). In addition, condensed and

rounded cells appeared in C2-ceramide-treated C6 cells under

microscopic observation, and their formation was not prevented

by BE or BI treatment (

Fig. 8

C).

Discussion

Several studies have indicated that the protective effects of

flavonoids are derived from their free radical-scavenging

acti-vities, whereas the antioxidant properties are insufficient to

explain the protective mechanism of flavonoids (

Horvathova

et al., 2003; Singh and Chopra, 2004

). BE have been

demon-strated to prevent ROS-induced cell damage by acting as a free

radical scavenger due to its high trolox equivalent antioxidant

capacity (TEAC) and DPPH free radical scavenging activity

(

Pietta, 2000; Ishige et al., 2001

). In the present study, BE

inhibited apoptosis induced by H

2

O

2

with a reduction in

intra-cellular peroxide levels. This suggests that the antioxidative

activity may participate in BE protection of C6 cells against

H

2

O

2

-induced cytotoxic events. C2-ceramide has been shown

to induce apoptosis via accumulating intracellular ROS in

mi-tochondria (

Kannan et al., 2004

), therefore C2-ceramide was

Fig. 8. Neither BE nor BI suppressed C2-ceramide-induced cell death in glioma C6 cells. (A) Cells were treated with BE or BI (50 or 100μM) for 30 min followed by the addition of C2-ceramide (20μM) for a further 24 h. The viability of C6 cells under different treatments was detected by the MTT assay. (B) As described in (A), the DNA integrity in cells under different treatments was analyzed by agarose electrophoresis. (C) The morphological changes induced by C2-ceramide were not altered by the addition of BE or BI (100μM). As described in (A), the morphology of cells was observed microscopically via Giemsa staining.

applied to examine the role of BE in intracellular ROS-mediated

cell death in the present study. Data of the present study

indi-cated no protective effect of BE against C2-ceramide-induced

cell death in C6 cells. Differential protective effects of BE

against H

2

O

2

and intracellular ROS (C2-ceramide)-mediated

cytotoxicity were observed. The reason for the ineffectiveness

of BE against C2-ceramide-induced cell death is still unclear.

Several studies reported that ceramide-mediated cell death is

Ca

2+

-dependent (

Pinton et al., 2001; Townley et al., 2005; Wu

et al., 2005

).

Maher and Hanneken (2005)

delineated that BE

protected against oxidative stress induced by GSH depletion,

tBOOH, and H

2

O

2

toxicity, but not Ca

2+

influx. Induction of

intracellular ROS and cytosolic Ca

2+

and long-lasting lose of

Ca

2+

in mitochondria were detected in ceramide-treated cells,

and long-lasting lose of calcium in mitochondria, but not others,

led to cell death (

Darios et al., 2003

). These data supported an

important role of Ca

2+

on ceramide-induced apoptosis. It

sug-gests that no inhibitory effect of BE on C2-ceramide-induced

cell death may in part attribute to its ineffectiveness on Ca

2+

-induced cytotoxicity in cells.

Flavonoids have been shown to regulate the activity of

protein kinases such as mitogen activated protein kinases

(MAPKs), tyrosine kinases, and protein kinase C (

Shen et al.,

2004; Ko et al., 2005

).

Nakahata et al. (2003)

indicated that BE

inhibited histamine- and A23187-induced ERK activation in C6

cells. In the present study, activation of ERKs was detected in

H

2

O

2

-treated C6 cells, and blocking ERK activation by

PD98059 or catalase significantly attenuated H

2

O

2

-induced

cell death. Suppression of H

2

O

2

-induced ERK protein

phos-phorylation by the addition of BE was identified in glioma C6

cells. This suggests that the protective effect of BE against H

2

O

2

-induced apoptosis is via suppression of ERK activation.

However, BE induced HO-1 protein expression via stimulation

of ERK protein phosphorylation in the absence of H

2

O

2

treatment. Therefore, a contradictory effect of ERK activation

was observed in BE-treated C6 cells in conditions with and

without H

2

O

2

. This suggests that BE possesses the ability to

induce intracellular kinase cascades such as ERKs to activate

cellular protective genes such as HO-1. In response to oxidative

stress, BE may directly or indirectly scavenge ROS production

and reduce oxidative-stress-induced kinase cascades such as

ERKs via its antioxidant activity. A double action of flavonoids

on the activation and inhibition of ERK activity, depending on

the extracellular oxidative condition, is proposed.

HO-1 protein has been shown to protect cells from ischemia/

reperfusion-induced damage (

Szabo et al., 2004

). Our previous

study indicated that induction of HO-1 protein attenuated

LPS-induced inducible nitric oxide synthase (iNOS) protein

expres-sion and NO production (

Lin et al., 2003

).

Chow et al. (2005)

indicated that HO-1 induction participates in quercetin protection

against cell death in RAW264.7 macrophages.

Chen et al. (2005)

demonstrated that resveratrol induction of HO-1 via activation of

Nrf2-ARE signaling participated in augmenting cellular

antiox-idant defense capacity to protect PC12 cells from

oxidative-stress-induced cell death. However, the role of the HO-1 protein in the

protective effect of BE against H

2

O

2

-induced cell death is still

undefined. An increase in HO-1 protein expression was detected

in BE (but not BI)-treated C6 cells. The chemical HO enzyme

inhibitor ZnPP suppressed the protective effect of BE against

H

2

O

2

-induced cell death. We found that ZnPP addition was able

to reverse the ROS-scavenging activity of BE. This suggests that

induction of HO-1 protein expression by BE may in part

participate in BE protection against H

2

O

2

-induced cell death.

Although several biological activities of BE have been

re-ported, data of the present study provide additional evidence to

suggest that BE possesses beneficial effects which protect against

H

2

O

2

-induced cell death, and modulation of ERKs activation and

induction of HO-1 protein expression via reduction of ROS

production were demonstrated. The potential for using BE in

experimental and practical application is highlighted for further

investigation.

Acknowledgments

This study was supported by the National Science Council of

Taiwan (NSC93-2321-B-038-009 and

NSC94-2320-B-038-049), Center of Excellence for Clinical Trial and Research in

Neurology Specialty, and Topnotch Stroke Research Center

Grant, Ministry of Education.

References

Ahn, H.C., Lee, S.Y., Kim, J.W., Son, W.S., Shin, C.G., Lee, B.J., 2001. Binding aspects of baicalein to HIV-1 integrase. Mol. Cells 12, 127–130. Birt, D.F., Hendrich, S., Wang, W., 2001. Dietary agents in cancer prevention:

flavonoids and isoflavonoids. Pharmacol. Ther. 90, 157–177.

Chen, K., Gunter, K., Maines, M.D., 2000. Neurons overexpressing heme oxygenase-1 resist oxidative stress-mediated cell death. J. Neurochem. 75, 304–313.

Chen, Y.C., Shen, S.C., Lee, W.R., Hou, W.C., Yang, L.L., Lee, T.J., 2001. Inhibition of nitric oxide synthase inhibitors and lipopolysaccharide induced inducible NOS and cyclooxygenase-2 gene expressions by rutin, quercetin, and quercetin pentaacetate in RAW 264.7 macrophages. J. Cell. Biochem. 82, 537–548.

Chen, Y.C., Shen, S.C., Lee, W.R., Lin, H.Y., Ko, C.H., Lee, T.J., 2002. Nitric oxide and prostaglandin E2 participate in lipopolysaccharide/interferon-gamma-induced heme oxygenase 1 and prevent RAW264.7 macrophages from UV-irradiation-induced cell death. J. Cell. Biochem. 86, 331–339. Chen, Y.C., Shen, S.C., Lin, H.Y., 2003. Rutinoside at C7 attenuates the

apoptosis-inducing activity of flavonoids. Biochem. Pharmacol. 66, 1139–1150. Chen, C.J., Raung, S.L., Liao, S.L., Chen, S.Y., 2004a. Inhibition of inducible

nitric oxide synthase expression by baicalein in endotoxin/cytokine-stimu-lated microglia. Biochem. Pharmacol. 67, 957–965.

Chen, T.J., Shen, S.C., Lin, H.Y., Chien, L.L., Chen, Y.C., 2004b. Lipopolysac-charide enhancement of 12-o-tetradecanoylphorbol 13-acetate-mediated transformation in rat glioma C6, accompanied by induction of inducible nitric oxide synthase. Toxicol. Lett. 147, 1–13.

Chen, C.Y., Jang, J.H., Li, M.H., Surh, Y.J., 2005. Resveratrol upregulates heme oxygenase-1 expression via activation of NF-E2-related factor 2 in PC12 cells. Biochem. Biophys. Res. Commun. 331, 993–1000.

Choi, B.M., Pae, H.O., Jeong, Y.R., Oh, G.S., Jun, C.D., Kim, B.R., Kim, Y.M., Chung, H.T., 2004. Overexpression of heme oxygenase (HO)-1 renders Jurkat T cells resistant to fas-mediated apoptosis: involvement of iron released by HO-1. Free Radic. Biol. Med. 36, 858–871.

Chow, J.M., Shen, S.C., Huan, S.K., Lin, H.Y., Chen, Y.C., 2005. Quercetin, but not rutin and quercitrin, prevention of H2O2-induced apoptosis via

anti-oxidant activity and heme oxygenase 1 gene expression in macrophages. Biochem. Pharmacol. 69, 1839–1851.

Crack, P.J., Taylor, J.M., 2005. Reactive oxygen species and the modulation of stroke. Free Radic. Biol. Med. 38, 1433–1444.

Darios, F., Lambeng, N., Troadec, J.D., Michel, P.P., Ruberg, M., 2003. Ceramide increases mitochondrial free calcium levels via caspase 8 and Bid: role in initiation of cell death. J. Neurochem. 84, 643–654.

de Bernardo, S., Canals, S., Casarejos, M.J., Solano, R.M., Menendez, J., Mena, M.A., 2004. Role of extracellular signal-regulated protein kinase in neuronal cell death induced by glutathione depletion in neuron/glia mesencephalic cultures. J. Neurochem. 91, 667–682.

Henze, C., Earl, C., Sautter, J., Schmidt, N., Themann, C., Hartmann, A., Oertel, W.H., 2005. Reactive oxidative and nitrogen species in the nigrostriatal system following striatal 6-hydroxydopamine lesion in rats. Brain Res. 1052, 97–104.

Hollman, P.C., Bijsman, M.N., van Gameren, Y., Cnossen, E.P., de Vries, J.H., Katan, M.B., 1999. The sugar moiety is a major determinant of the absorption of dietary flavonoid glycosides in man. Free Radic. Res. 31, 569–573. Horvathova, K., Novotny, L., Vachalkova, A., 2003. The free radical scavenging

activity of four flavonoids determined by the comet assay. Neoplasma 50, 291–295.

Huang, W.H., Lee, A.R., Chien, P.Y., Chou, T.C., 2005. Synthesis of baicalein derivatives as potential anti-aggregatory and anti-inflammatory agents. J. Pharm. Pharmacol. 57, 219–225.

Hwang, J.M., Tseng, T.H., Tsai, Y.Y., Lee, H.J., Chou, F.P., Wang, C.J., Chu, C.Y., 2005. Protective effects of baicalein on tert-butyl hydroperoxide-induced hepatic toxicity in rat hepatocytes. J. Biomed. Sci. 12, 389–397.

Ishige, K., Schubert, D., Sagara, Y., 2001. Flavonoids protect neuronal cells from oxidative stress by three distinct mechanisms. Free Radic. Biol. Med. 30, 433–446. Juravleva, E., Barbakadze, T., Mikeladze, D., Kekelidze, T., 2005. Creatine enhances survival of glutamate-treated neuronal/glial cells, modulates Ras/ NF-kappaB signaling, and increases the generation of reactive oxygen species. J. Neurosci. Res. 79, 224–230.

Kannan, R., Jin, M., Gamulescu, M.A., Hinton, D.R., 2004. Ceramide-induced apoptosis: role of catalase and hepatocyte growth factor. Free Radic. Biol. Med. 37, 166–175.

Ko, C.H., Shen, S.C., Lee, T.J., Chen, Y.C., 2005. Myricetin inhibits matrix metalloproteinase 2 protein expression and enzyme activity in colorectal carcinoma cells. Mol. Cancer Ther. 4, 281–290.

Laufs, U., Wassmann, S., Czech, T., Munzel, T., Eisenhauer, M., Bohm, M., Nickenig, G., 2005. Physical inactivity increases oxidative stress, endothelial dysfunction, and atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 25, 809–814. Li, F.Q., Wang, T., Pei, Z., Liu, B., Hong, J.S., 2005. Inhibition of microglial activation by the herbal flavonoid baicalein attenuates inflammation-media-ted degeneration of dopaminergic neurons. J. Neural Transm. 112, 331–347. Lin, H.Y., Juan, S.H., Shen, S.C., Hsu, F.L., Chen, Y.C., 2003. Inhibition of lipopolysaccharide-induced nitric oxide production by flavonoids in RAW264.7 macrophages involves heme oxygenase-1. Biochem. Pharmacol. 66, 1821–1832. Lin, H.Y., Shen, S.C., Chen, Y.C., 2005. Anti-inflammatory effect of heme oxygenase 1: glycosylation and nitric oxide inhibition in macrophages. J. Cell. Physiol. 202, 579–590.

Liu, Z.M., Chen, G.G., Ng, E.K., Leung, W.K., Sung, J.J., Chung, S.C., 2004. Upregulation of heme oxygenase-1 and p21 confers resistance to apoptosis in human gastric cancer cells. Oncogene 23, 503–513.

Maher, P., Hanneken, A., 2005. Flavonoids protect retinal ganglion cells from oxidative stress-induced death. Invest. Ophthalmol. Visual Sci. 46, 4796–4803.

Morita, T., Imai, T., Sugiyama, T., Katayama, S., Yoshino, G., 2005. Heme oxygenase-1 in vascular smooth muscle cells counteracts cardiovascular damage induced by angiotensin II. Curr. Neurovasc. Res. 2, 113–120.

Nakahata, N., Tsuchiya, C., Nakatani, K., Ohizumi, Y., Ohkubo, S., 2003. Baicalein inhibits Raf-1-mediated phosphorylation of MEK-1 in C6 rat glioma cells. Eur. J. Pharmacol. 461, 1–7.

Park, L., Anrather, J., Zhou, P., Frys, K., Pitstick, R., Younkin, S., Carlson, G.A., Iadecola, C., 2005. NADPH-oxidase-derived reactive oxygen species med-iate the cerebrovascular dysfunction induced by the amyloid beta peptide. J. Neurosci. 25, 1769–1777.

Petrache, I., Otterbein, L.E., Alam, J., Wiegand, G.W., Choi, A.M., 2000. Heme oxygenase-1 inhibits TNF-alpha-induced apoptosis in cultured fibroblasts. Am. J. Physiol., Lung Cell. Mol. Physiol. 278, L312–L319.

Pietta, P.G., 2000. Flavonoids as antioxidants. J. Nat. Prod. 63, 1035–1042. Pinton, P., Ferrari, D., Rapizzi, E., Di Virgilio, F., Pozzan, T., Rizzuto, R., 2001.

The Ca2+_{concentration of the endoplasmic reticulum is a key determinant of}

ceramide-induced apoptosis: significance for the molecular mechanism of Bcl-2 action. EMBO J. 20, 2690–2701.

Qian, Y., Zheng, Y., Ramos, K.S., Tiffany-Castiglioni, E., 2005. GRP78 compart-mentalized redistribution in Pb-treated glia: role of GRP78 in lead-induced oxidative stress. Neurotoxicology 26, 267–275.

Regev-Shoshani, G., Shoseyov, O., Bilkis, I., Kerem, Z., 2003. Glycosylation of resveratrol protects it from enzymic oxidation. Biochem. J. 374, 157–163. Rivera, F., Urbanavicius, J., Gervaz, E., Morquio, A., Dajas, F., 2004. Some aspects of the in vivo neuroprotective capacity of flavonoids: bioavailability and structure–activity relationship. Neurotoxicol. Res. 6, 543–553. Shao, Z.H., Li, C.Q., Vanden Hoek, T.L., Becker, L.B., Schumacker, P.T., Wu,

J.A., Attele, A.S., Yuan, C.S., 1999. Extract from Scutellaria baicalensis Georgi attenuates oxidant stress in cardiomyocytes. J. Mol. Cell. Cardiol. 31, 1885–1895.

Shao, Z.H., Vanden Hoek, T.L., Qin, Y., Becker, L.B., Schumacker, P.T., Li, C.Q., Dey, L., Barth, E., Halpern, H., Rosen, G.M., Yuan, C.S., 2002. Baicalein attenuates oxidant stress in cardiomyocytes. Am. J. Physiol. 282, H999–H1006.

Shen, S.C., Chen, Y.C., Hsu, F.L., Lee, W.R., 2003. Differential apoptosis-inducing effect of quercetin and its glycosides in human promyeloleukemic HL-60 cells by alternative activation of the caspase 3 cascade. J. Cell. Biochem. 89, 1044–1055.

Shen, S.C., Ko, C.H., Hsu, K.C., Chen, Y.C., 2004. 3-OH flavone inhibition of epidermal growth factor-induced proliferation through blocking prostaglan-din E2 production. Int. J. Cancer 108, 502–510.

Shieh, D.E., Liu, L.T., Lin, C.C., 2000. Antioxidant and free radical scavenging effects of baicalein, baicalin and wogonin. Anticancer Res. 20, 2861–2865. Singh, D., Chopra, K., 2004. The effect of naringin, a bioflavonoid on ischemia–reperfusion induced renal injury in rats. Pharmacol. Res. 50, 187–193.

Szabo, M.E., Gallyas, E., Bak, I., Rakotovao, A., Boucher, F., de Leiris, J., Nagy, N., Varga, E., Tosaki, A., 2004. Heme oxygenase-1-related carbon monoxide and flavonoids in ischemic/reperfused rat retina. Invest. Ophthalmol. Visual Sci. 45, 3727–3732.

Townley, H.E., McDonald, K., Jenkins, G.I., Knight, M.R., Leaver, C.J., 2005. Ceramides induce programmed cell death in Arabidopsis cells in a calcium-dependent manner. Biol. Chem. 386, 161–166.

Wang, J., Yu, Y., Hashimoto, F., Sakata, Y., Fujii, M., Hou, D.X., 2004. Baicalein induces apoptosis through ROS-mediated mitochondrial dysfunction pathway in HL-60 cells. Int. J. Mol. Med. 14, 627–632.

Wu, Z., Tandon, R., Ziembicki, J., Nagano, J., Hujer, K.M., Miller, R.T., Huang, C., 2005. Role of ceramide in Ca2+-sensing receptor-induced apoptosis. J. Lipid Res. 46, 1396–1404.