Tiron protects against UVB-induced senescence-like characteristics in human dermal fibroblasts by the inhibition of superoxide anion production and glutathione depletion

ORIGINAL RESEARCH

Tiron protects against UVB-induced senescence-likecharacteristics in human dermal fibroblasts by the

inhibition of superoxide anion production andglutathione depletionajd_912 1..9

Yong Fang,* Xiao-Hui Hu,* Zhi-Gang Jia, Mang-Hua Xu, Zhu-Ying Guo and Feng-Hou Gao

NO.3 People’s Hospital affiliated to Shanghai Jiao-Tong University School of Medicine (SJTU-SM),Shanghai, China

ABSTRACT

Background/Objectives: Free radicals and reac-tive oxygen species (ROS), which are generated byUV irradiation, may induce an irreversible growtharrest similar to senescence. Tiron, 4,5-dihydroxy-1,3-benzene disulfonic acid, is a widely used antioxi-dant to rescue ROS-evoked cell death. The aim of thearticle was to explore the effects of tiron on skin pho-toaging and associated mechanisms.Methods: The effects of tiron on cell proliferationwere determined using 3-(4,5-dimethylthiazol-2-Yl)-2,5-diphenyltetrazolium bromide. Senescent cellswere determined by morphology and senescence-associated b-galactosidase activity analysis. Intra-cellular hydrogen peroxide, superoxide anion andglutathione concentration were analysed by a fluores-cent probe. The concomitant changes of proteinexpression were analysed with Western blot.Results: Human dermal fibroblasts were induced topremature senescence by sub-cytotoxic doses ofirradiated UVB. Strong senescence-associated b-galactosidase activity and increased intracellularsuperoxide anion were observed in human dermalfibroblasts irradiated by UVB. Tiron blocks UVB-induced glutathione depletion and increase of sup-eroxide anion and protects against UVB-inducedsenescence-like characteristics in human dermalfibroblasts. Compared with normal fibroblasts,UVB-irradiated human dermal fibroblasts showed a

higher ratio of active (hypophosphorylated) to inactive(phosphorylated) forms of Rb and p38, upregulation ofp53 or p16 and c-Myc and insulin-like growth factor 1(IGF-1) downregulation. After treatment with tiron,p53, p16 c-Myc and IGF-1 as well as phosphorylationRb and p38 could partially recover.Conclusion: These results indicate that tiron pro-tects against UVB-induced senescence-like character-istics in human dermal fibroblasts via the inhibition ofproduction of superoxide anion and glutathione deple-tion, and modulation of related senescence proteins.

Key words: fibroblast, reactive oxygen species,senescence, Tiron, ultraviolet irradiation.

Correspondence: Dr Feng-Hou Gao, No. 3 People’s Hospital, affili-ated to Shanghai Jiao-Tong University School of Medicine, Shanghai201900, China. Email: [email protected]

Yong Fang, PhD. Xiao-Hui Hu, M.S. Zhi-Gang Jia, M.S. Mang-HuaXu, BSc. Zhu-Ying Guo, M.S. Feng-Hou Gao, PhD.

*These authors are equal contributors to this work.Submitted 14 October 2011; accepted 22 March 2012.

Abbreviations:

CMFDA 5-chloromethylfluorescein diacetateDCF 2,7-dichlorofluoresceinDHE dihydroethidiumDMEM Dulbecco’s modified Eagle’s mediumDMSO dimethyl sulfoxideFBS fetal bovine serumGAPDH anti-glyceraldehyde 3-phosphate dehydrogenaseH2DCFDA 2′,7′dichlorodihydrofluorescein diacetateHDF human dermal fibroblastsHRP horseradish peroxidaseIGF Insulin-like growth factorMAPK mitogen-activated protein kinaseMTT 3-(4,5-dimethylthiazol-2-Yl)-2,5-diphenyltetrazolium

bromidePBS phosphate buffered salinePVDF polyvinylidene fluorideROS reactive oxygen speciesRT-PCR reverse transcriptase polymerase chain reactionSASP senescence-associated secretory phenotypeSDS-PAGE sodium dodecyl sulfate polyacrylamide gel

electrophoresisSOD superoxide dismutaseXOX xanthine oxidase

bs_bs_banner

Australasian Journal of Dermatology (2012) ••, ••–•• doi: 10.1111/j.1440-0960.2012.00912.x

© 2012 The AuthorsAustralasian Journal of Dermatology © 2012 The Australasian College of Dermatologists

INTRODUCTION

The effects of sunlight have fascinated researchers fordecades because nearly every living thing on earth is likelyto be exposed to sunlight and its UV fraction. In addition todetrimental long-term effects such as immunosuppressionand skin cancer, premature aging of the skin (photoaging)is a well-documented consequence of exposure to UVA andUVB.1 Although UVB is mostly absorbed in the epidermisand predominantly affects epidermal cells, about 10–30% ofUVB can penetrate the epidermis, reach the upper dermisand harm fibroblasts and the extracellular matrix as well.2,3

Chronic exposure to high-intensity UVB causes cumulativedamage, and is regarded as one of the most important envi-ronmental hazards to human skin. The effects of light havebeen attributed to UVB irradiated-induced reactive oxygenspecies (ROS) formation.4–7 Therefore, the use of antioxi-dants is an effective approach to prevent symptoms relatedto photo-induced aging of the skin.

Several phenolic compounds have been shown to be usefulantioxidants by virtue of donating a hydrogen atom to peroxysuperoxide, or alkoxy radicals, thus forming a less toxichydroperoxide and a less reactive phenoxyl radical.8–10 Tiron,4,5-dihydroxy-1,3-benzene disulfonic acid, which is watersoluble and cell permeable and is an effective antioxidantcapable of scavenging a variety of radicals.9,11 Tiron can alsoprevent the ozone-mediated inactivation of superoxide dis-mutase and catalase as well as the metal-catalysed peroxi-dation of DNA or lipids.9,12 Because tiron is a potentiallyuseful agent protecting against the lethal effects of oxidativestress and offers protection from it by chelating redox-activetransition metal ions, it has been widely used in a num-ber of studies on oxidative stress-related physiologicalresearch.9,13,14 Although studies have described the protec-tive effect of tiron on oxidative stress, however, little hasbeen said about its effects on photo-induced aging of the skin.

Here we report on the supplementation of tiron, a super-oxide anion scavenger, that inhibits UVB-induced cellularsenescence-like characteristics of human dermal fibro-blasts and the possible mechanisms involved in the activity.

MATERIALS AND METHODS

Reagents

Tiron, 3-(4,5-dimethylthiazol-2-Yl)-2,5-diphenyltetrazoliumbromide (MTT), dimethyl sulfoxide (DMSO) and anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH) wereobtained from Sigma (St Louis, MO, USA). High glucose-containing Dulbecco’s modified Eagle’s medium (DMEM),fetal bovine serum (FBS) and phosphate buffered saline(PBS) were obtained from Gibco-BRL (Gaithersburg, MD,USA). Antibodies of p16, p53, and pi-Rb, c-Myc, and horse-radish peroxidase HRP-conjugated immunoglobulin (Ig)Gwere obtained from Santa Cruz Biochemicals (CA, USA).

Isolation and cultivation of human dermalfibroblasts

Human dermal fibroblasts were aseptically isolated from acircumcised neonatal foreskin. The epidermis and dermis

were separated by incubation in 0.9 units/mL dispase in amedium for 16 h at 4°C. After the epidermis and dermiswere mechanically separated, the dermis was minced andattached to the surface of a tissue culture flask and fed withDMEM containing 10% FBS for 1–2 weeks. The dermalfibroblasts spreading as a radial outgrowth from theattached pieces of dermis were harvested and sub-culturedin DMEM with 10% FBS.

Ultraviolet irradiation

For UVB-irradiated cells, we exposed the human dermalfibroblasts to a UVB cross-linker (6 ¥ 8 W, 312 nm, ModelCL-508 M, Vilber Lourmat, Paris, France). In brief, serum-starved confluent human dermal fibroblasts were rinsedtwice with PBS and all irradiations of the cells were per-formed under a thin layer of PBS. Immediately after irra-diation, a fresh, serum-free medium was added to thecells. Responses were measured after an incubation periodfor the times indicated. Mock-irradiated controls followedthe same schedule of medium changes without beingirradiated.

Determination of cell viability

The effect of tiron or UVB on the cell viability of humandermal fibroblast was determined using an MTT assay.Briefly, cells at 2 ¥ l04 cells/well were treated with variousconcentrations of tiron or UVB. After incubation for 72 h, thecells were washed twice with PBS, then MTT (0.5 mg/mLPBS) was added to each well and incubated at 37°C for30 min. The formazan crystals formed were dissolved byadding DMSO (100 mL/well) and optical density was read at490 nm using a microplate reader (Model 3550, BIO-RAD,Richmond, CA, USA) for cell viability. The reduction inviability of tiron or UVB-treated HDF was expressed as apercentage of control cells, which were considered to be100% viable.

Detection of intracellular hydrogen peroxide andsuperoxide anion concentration

Two oxidation-sensitive fluorescent probe dye,2’,7’dichlorodihydrofluorescein-diacetate (H2DCFDA) (Invi-trogen Molecular Probes, Eugene, OR, USA) and dihydroet-hidium (DHE) were used to detected the intracellularhydrogen peroxide or superoxide anion concentration,respectively. H2DCFDA was deacetylated intracellularlyby nonspecific esterase, which was further oxidised bycellular peroxides to the fluorescent compound 2,7-dichlorofluorescein (DCF). Dihydroethidium (DHE) (Invit-rogen Molecular Probes) is a fluorogenic probe that is highlyselective for superoxide anion radical detection. DHE iscell permeable and reacts with superoxide anion to formethidium, which in turn intercalates in the deoxyribonucleicacid, thereby exhibiting red fluorescence. Briefly, cells weretreated with UVB in the presence or absence of ROS scaven-gers for varying lengths of time. After having been washed

2 Y Fang et al.


with PBS the cells were incubated with 20 mmol/L H2DCFDAor 5 mmol/L DHE at 37°C for 30 min according to themanufacturer’s instructions. The fluorescence signals weredetected by FACStar flow cytometer (Beckman Coulter, Brea,USA). For each sample, 5000 or 10 000 events were collected.Hydrogen peroxide and superoxide anion production wereexpressed as mean fluorescence intensity.15,16

Detection of intracellular glutathione

Cellular glutathione levels were analysed using 5-chloromethylfluorescein diacetate (CMFDA, MolecularProbes). Cytoplasmic esterases convert non-fluorescentCMFDA to fluorescent 5-chloromethylfluorescein, whichcan then react with the glutathione. CMFDA is a usefulmembrane-permeable dye for determining levels of intrac-ellular glutathione. Briefly, cells were treated with UVB inthe presence or absence of ROS scavengers for the timesindicated. After having been washed with PBS the cells wereincubated with 5 mmol/L CMFDA at 37°C for 30 min accord-ing to the manufacturer’s instructions. Chloromethylfluo-rescein (CMF) fluorescence was detected by the FACStarflow cytometer (Beckman Coulter).17,18 For each sample5000 or 10 000 events were collected.

Senescence-associated b-galactosidase assays

At 48 h after the last stress, the cells were trypsinised andseeded in squared 35-mm culture dishes (Falcon, Becton-Dickinson UK) containing 2 mL of basal medium eagle(BME) and 1% FCS at a density of 700 cells/cm2. At 24 h after,senescence-associated b-galactosidase activity was deter-mined as described by Dimri et al.19 The population ofsenescence-associated b-galactosidase positive cells wasdetermined by counting 400 cells per dish. The proportionsof cells positive for senescence-associated b-galactosidaseactivity are given as a percentage of the total number of cellscounted in each dish. Triplicate experiments were alwaysperformed. The results are expressed as means � SD.

Western blot analysis

The human dermal fibroblasts (2 ¥ 106 cells) were pretreatedwith tiron for 2 h and then irradiated with UVB (40 mJ/cm2)and the cells were cultured with tiron for 72 h. The cells werelysed with the same volume of 2 ¥ SDS Reagent (PierceBiotechnology, Rockford, IL, USA) on ice. In the lysates theprotein concentration was determined using the bicincho-ninic acid (BCA) method. Samples were separated by sodiumdodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) with 10% acrylamide running and 5% acrylamidestacking gels and then transferred to hybond-polyvinylidenefluoride (PVDF) membranes using a Western blot apparatus.The PVDF membranes were blotted with 1 mg/mL of primaryantibodies for p53, p16, pi-Rb, c-Myc, piP38, P38 and GAPDH.HRP-conjugated IgG was used as a secondary antibody. Theprotein expression levels were then determined by analysingthe signals captured on the PVDF membrane using an imageanalyser (Las-1000, Fuji Film, Tokyo, Japan).

RNA extraction and reverse transcriptase reversetranscription-polymerase chain reaction

The total RNA was extracted from the cells using RNAisoTMplus (TaKaRa, Shiga, Japan). The concentration and purityof the RNA were determined by absorbance at 260/280 nm.cDNA was synthesised from 1 mg total RNA using a TaKaRaa reverse transcriptase polymerase chain reaction (RT-PCR)kit (TaKaRa). Total RNA was extracted with TRIzol reagent(Invitrogen) 72 h after the stress, according to the manufac-turer’s instructions as previously described. Total RNA wasreverse transcribed (10 mL of total volume) at 42°C for30 min, 99°C for 5 min and 4°C for 5 min. The total volumeof the PCR reaction was 20 mL. The PCR conditions were onecycle of 94°C at 2 min, 35 cycles of 94°C at 30 s, 55°C at 30 s,72°C at 1 min. The primers used for GAPDH were forward:5′-TGG ACT CTG GAA TCC ATT CTG-3′, reverse: 5′-AAAATC CCT GTT CCC ACT CA-3′, and for insulin-like growthfactor 1(IGF)-1 they were forward: 5′-ACC TAC AGT GAAGAT GCA CAC CA-3′; reverse: 5′-AGC GAG CTG ACT TGGCAG GCT TGA-3′. The PCR products were then separated by1% agarose gel electrophoresis and photographed.

Superoxide dismutase (SOD) activity

Upon reaching confluence, the cell cultures in the 96-wellplates were exposed to varying concentrations of tiron orUVB at various time points. The cells were washed twicewith PBS and then detached with a rubber policeman insample buffer suspensions and sonicated before analysingthe total SOD activity. The SOD activity was determinedusing a colorimetric SOD assay kit (Cayman Chemicals AnnArbor, MI, USA). This SOD kit utilises the water-soluble2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium, monosodium salt, which produces a highlywater soluble formazan dye upon reduction with superox-ide anions. The rate of reduction with oxygen is linearlyrelated to xanthine oxidase (XOX) activity, and it is inhibitedby SOD. This assay system has the advantage of determin-ing SOD activity over a wide range. Absorbance was read at450 nm using the SpectraMax M5E microplate reader(Molecular Devices LLC, Sunnyvale, USA). SOD activity wasexpressed as percentage of XOX activity inhibition.

Statistical analysis

All experiments were repeated a minimum of three times.Error bars indicate standard errors of mean. MicrosoftExcel or Instat software (GraphPad Prism4, San Diego, CA,USA) was used to analyse the data. A Student’s t-test orone-way analysis of variance (ANOVA) with post hoc analysisusing Tukey’s multiple comparison test was used for para-metric data. Statistical significance was defined as P < 0.05.

RESULTS

Inhibitory effect of tiron or UVB irradiatedhuman diploid fibroblasts

The cytotoxic effect of tiron on human diploid fibroblastscells was determined with varying concentrations of tiron

Tiron blocks UVB-induced HDF senescence 3


(72 h) by MTT assay. As is shown in Figure 1a, no inhibitoryeffect of tiron on HDF cell viability was observed when theconcentration of tiron was lower than 500 mmol/L. We alsoexamined the effect of UVB exposure on the cell viability ofhuman diploid fibroblasts. There was a decrease in theviability of human diploid fibroblasts with increasing dosesof UVB irradiation (10–100 mJ/cm2) (Fig. 1b), and UVB atdoses of 20 and 100 mJ/cm2 significantly inhibited cell viabil-ity. Our results indicate that tiron shows little inhibitoryeffect on the cell viability of human diploid fibroblasts, whileUVB irradiation induces the decrease of human diploid fibro-blasts cell viability in a dose-dependent manner.

Tiron or UVB irradiation brought aboutmorphological changes in humandiploid fibroblasts

Cells were examined for the morphological featuresof both senescence and apoptosis. Most of the cells treated

with 20 and 40 mJ/cm2 UVB (Fig. 2a) appeared enlarged,flattened and less spindle shaped than the controlcells. Stereological analyses of these cells revealedan increased adherent surface and a bigger cell andnuclear volume than in the control cells. In addition,a small proportion of the coils in this culture wore foundto contract and detach from the dish. This populationof condensed detached cells greatly increased in aUVB concentration-dependent manner up to the death ofall the cells (60 and 80 mJ/cm2). These morphologicalchanges increased in a UVB concentration-dependentmanner, with the most pronounced phenotype beingobserved at 40 mJ/cm2 (Fig. 2a). Therefore the sub-cytotoxic dose of UVB used throughout this study was40 mJ/cm2. Although high-dose tiron (1000 mmol/L) inhib-ited the viability of human diploid fibroblast, significantnumbers of both enlarged adherent and condenseddetached cells were not observed in the tiron-treated cul-tures (Fig. 2b).

Figure 1 Inhibitory effect of tiron orUVB irradiated on human dermal fibro-blasts viability. (a) Human diploid fibro-blasts cells were treated with 0, 10, 50,100, 200, 500 and 1000 mmol/L tiron for72 h. The effects of tiron on cell prolif-eration were determined by using MTT.(b) Exposure to UVB irradiation was per-formed at 0, 10, 20, 40, 60, 80 and100 mJ/cm2. UVB-irradiated cells werecultured for 72 h. Cell viability wasobserved with MTT. *P < 0.05 comparedwith control. The data are presented asmeans � SD (N = 5) and all experimentswere done in triplicate.

A B

120

100

80

60

40

20

0

120

100

80

60

40

20

0

Cel

l Via

bili

ty (

%)

0 10 50 100 200 500 1000 0 10 20 40 60 80 100

Tiron μmol/L UVB mJ/cm2

Figure 2 Effect of tiron or UVB irradia-tion on human dermal fibroblasts mor-phology. (a) Human dermal fibroblastswere treated with 0, 10, 20, 40, 6 and80 mJ/cm2 UVB. UVB-irradiated cellswere cultured for 72 h. (b) Humandiploid fibroblasts cells were treatedwith 0, 10, 50, 100, 200 and 500 mmol/Ltiron for 72 h. The morphology of controlcells and UVB or tiron-treated cells wereexamined at 72 h by phase-contrastmicroscopy (¥100). The experimentswere done in triplicate.

Tiron μmol/L

UVB mJ/cm2

0

A

B

10 20 40 60 80

0 10 50 100 200 500

4 Y Fang et al.


UVB increases the superoxide anion the level andsenescence-associated b-galactosidase activity inhuman diploid fibroblasts

First, we assessed the production of intracellular hydrogenperoxide in UVB-treated human diploid fibroblasts cellswith H2DCFDA fluorescence dye. As is shown in Figure 3a,UVB did not increase the intracellular hydrogen peroxidelevels in a time-dependent manner. Next, we assessed thechange of intracellular superoxide anion in UVB-treatedhuman diploid fibroblasts with DHE staining. Red fluores-cence derived from DHE, reflecting superoxide anion accu-mulation, increased significantly in UVB-treated humandiploid fibroblast cells (Fig. 3b). UVB treatment at 40 mJ/cm2 for 2 h increased intracellular superoxide anion levels.The maximum level of superoxide anion was reached at

about 180 min in the human diploid fibroblasts with 40 mJ/cm2 UVB and was about thrice higher than that of controlcells.

As the important component of antioxidant system, cellu-lar glutathione has been shown to be crucial for the regu-lation of cell proliferation, cell cycle progression, apoptosisand senescence. We therefore analysed the changes of glu-tathione level in UVB-treated human diploid fibroblastsusing a CMFDA fluorescence probe. UVB at 40 mJ/cm2 sig-nificantly decreased the level of intracellular glutathionecontent at 2 h (Fig. 3c), indicating the depletion of the intra-cellular glutathione content of human diploid fibroblasts.

One of the morphological changes that was frequentlyobserved in the UVB-treated cells is the flattening of theadherent cells with increased granularity, which is a typicalmorphological change associated with cellular senescence

Mea

n D

CF

Flu

ores

cenc

e (M

FT

)M

ean

CM

F F

luor

esce

nce

(MF

T)

Mea

n D

HE

Flu

ores

cenc

e (M

FT

)

20

15

10

5

0

1.4

1.2

1

0.8

0.6

0.4

0.2

00 10 30 60 120 180 minutes

40 mJ/cm2 UVB

0 10 30 60 120 180 minutes

40 mJ/cm2 UVB

0 10

24Con 48 72 hours

30 60 120 180 minutes

40 mJ/cm2 UVB

40 mJ/cm2 UVB

*

*

*

140

120

100

80

60

40

20

0

A B

C D

Figure 3 Effects of UVB on intracellular hydrogen peroxide, superoxide anion, glutathione production and senescence-associatedb-galactosidase activity in human diploid fibroblasts. Exponentially growing cells were treated with 40 mJ/cm2 UVB for varying lengths oftime. (a) The intracellular hydrogen peroxide level was determined by FACStar flow cytometer. The graph shows the levels of mean2,7-dichlorofluorescein fluorescence for the designated concentrations of UVB and times. (b) Intracellular superoxide anion level wasdetermined by a FACStar flow cytometer. The graph shows the levels of mean DHE fluorescence for the designated concentrations of UVBand times. (c) Intracellular glutathione level was determined by a FACStar flow cytometer. The graph shows the levels of mean CMFfluorescence for the indicated concentrations of UVB and times. *P < 0.05 compared with the control group. (d) Senescent cells weredetermined by senescence-associated b-galactosidase activity analysis (¥200). The experiments were done in triplicate.



(Fig. 2a). By examining the senescence-associated expres-sion of b-galactosidase activity we confirmed that cellularsenescence was indeed induced in those flattened humandiploid fibroblasts treated with UVB (Fig. 3d).

Tiron blocks ROS production, glutathionedepletion and senescence in UVB-treated humandermal fibroblasts

To further determine the role of superoxide anion produc-tion and glutathione depletion in UVB-induced senescence,the human dermal fibroblasts were treated with UVB in thepresence or absence of tiron for varying lengths of time. Asexpected, the co-treatment of tiron with UVB inhibited UVB-induced superoxide anion increase and glutathione decline(Fig. 4a,b). We therefore examined whether tiron could

prevent UVB-induced cell senescence in human dermalfibroblasts. As is shown in Figure 4c,d, co-treatment of tironwith UVB protected the cells from UVB-induced senescencein human dermal fibroblasts.

Tiron abrogates UVB-induced changes of pi-Rb,p53, p16, c-Myc, pi-P38 and IGF-1 in humandermal fibroblasts

UVB-induced human dermal fibroblasts senescence isrelated to the upregulation of p16 and p53 downregulationof pi-Rb and c-Myc. As tiron was able to inhibit UVB-inducedsenescence, we examined the effect of tiron on the expres-sion of these proteins. As shown in Figure 5a, UVB treat-ment caused changes in p53, p16, p-Rb and c-Myc. UVBtreatment at 40 mJ/cm2 mainly induced senescence in

20

16

12

8

4

0

Mea

n D

HE

Flu

ores

cenc

e (M

FT

)

Mea

n C

MF

Flu

ores

cenc

e (M

FT

)S

A-β

-Gal

(%

)

*

*

*

UVB

Tiron

UVB

Tiron

_ + +

_ _ +

_ + +

_ _ +

UVB

Tiron

_ + +

_ _ +

40 mJ/cm2 UVB

Control_

Trion

140

120

100

80

60

40

20

0

100

80

60

40

20

0

A B

C D

Figure 4 Effects of tiron on intracellular superoxide anion, glutathione production and senescence in UVB-treated human dermal fibro-blasts. Human dermal fibroblasts pretreated with 200 mmol/L tiron for 2 h were exposed to UVB irradiation of 40 mJ/cm2 at 312 nm. Theintracellular superoxide anion (a) and glutathione (b) were determined by a FACStar flow cytometer. The graphs show the levels of meanDHE (a) CMF, and (b) fluorescence. The cells were incubated with or without tiron at concentrations of 200 mmol/L tiron for 72 h. (c)Senescent cells were determined by a senescence-associated b-galactosidase activity analysis (¥100). (d) The graphs show the percentageof senescent-positive cells. *P < 0.05 compared with the cells treated with UVB alone. The experiments were done in triplicate.

6 Y Fang et al.


human dermal fibroblasts, as reflected by the induction ofp16, a hallmark of both replicative and accelerated senes-cence, and the co-treatment of tiron inhibited the inductionof p16 (Fig. 5a). C-Myc inactivation and p53 and Rb signalpathway activation have both been reported to be involvedin senescence. UVB treatment resulted in c-Myc inactiva-tion and p53 and Rb activation, while tiron co-treatmentrescued the expression of c-Myc, p53 and Rb parallelto senescence inhibition. p38 mitogen-activated proteinkinase (MAPK) is a novel response-independent regulatorof the senescence-associated secretory phenotype. Asshown in Figure 5b, UVB treatment activated p38, but tironinhibited the activation of p38 UVB-induced. IGF-1 issecreted by a variety of cell types, including human fibro-blasts. UVB down-regulated IGF-1 mRNA levels, whileco-treatment with tiron could rescued the mRNA levels ofIGF-1 (Fig. 5c).

Tiron abolishes the inhibition of SOD activityinduced-UVB in fibroblasts

Treatment of fibroblasts with 40 mJ/cm2 UVB resulted in adecline of cellular SOD activity at 60, 120, 180 min. Althoughthe SOD activity was not significantly increased in fibro-blasts treated with 0, 10, 50, 100, 200, 500 mmol/L tiron for2 h, tiron was able to significantly restore the SOD activity offibroblasts treated by 40mJ/cm2 UVB (Fig. 6).

DISCUSSION

There is some evidence to indicate that tiron exerts effectsrelated to its role as a scavenger of superoxide anion.20–22

However, the relationships between tiron and UVB-inducedhuman dermal fibroblasts senescence were unclear. Theresults we obtained demonstrate that tiron protects againstUVB-induced senescence-like characteristics in humandermal fibroblast by the inhibition of production of super-oxide anion and glutathione depletion.

In 2005, stress-induced premature senescence of humandiploid fibroblasts was induced by a series of subcytotoxicexposures to UVB,23 now we focused on the involvement ofROS such as hydrogen peroxide and superoxide anion andglutathione in UVB-induced human diploid fibroblasts. Ourdata suggest that UVB could not increase the intracellularhydrogen peroxide levels in time dependent manner,whereas treatment of UVB could increase intracellularsuperoxide anion levels. Glutathione, as the main nonpro-tein antioxidant in the cell, could clear away the superoxideanion free radical and provide electrons for enzymes suchas glutathione peroxidase, which reduces hydrogen perox-ide to H2O.17,24,25 Likewise, the results indicated the depletionof intracellular glutathione content by UVB in humandiploid fibroblasts. Most of the cells treated with UVBappeared enlarged, flattened, which is a typical morpho-logical change associated with cellular senescence. We con-firmed that cellular senescence was indeed induced inthose flattened human diploid fibroblasts by b-galactosidaseactivity.

Thus, we extrapolated that superoxide anion and glu-tathione depletion play a major role in UVB-induced senes-cence in human diploid fibroblasts. In our study, based onthis in vitro model system, we evaluated the protectiveeffect of tiron against photoaging in human diploid fibro-blasts and its mechanism of action. Tiron, a scavenger ofsuperoxide anion, could significantly reduce the level ofsuperoxide anion and glutathione depletion in humandiploid fibroblasts treated with 40 mJ/cm2 UVB. Accord-ingly, the scavengers could reduce senescence in UVB-treated cells. These results suggest that the changes ofintracellular ROS, especially superoxide anion and glu-tathione depletion by UVB were directly related to senes-cence in human diploid fibroblasts and the anti-aging effectof tiron in premature human diploid fibroblasts induced byUVB exposure.

It has been well established that cellular aging is accom-panied by increased oxidative stress and altered expression

Figure 5 Effects of tiron on UVB-induced senescence related proteins andmRNA levels in human diploid fibro-blasts. Human dermal fibroblasts pre-treated with 200 mmol/L tiron for 2 hwere exposed to UVB irradiation of40 mJ/cm2 at 312 nm. The cells wereincubated with or without 200 mmol/Ltiron for 72 h. Aliquots of 40 mg of proteinextracts were resolved by 10% sodiumdodecyl sulfate polyacrylamide gel elec-trophoresis gel, transferred onto thepolyvinylidene fluoride membrane, andimmunoblotted with the antibodies indi-cated (a) and (b). A semi-quantitativereverse transcriptase polymerase chainreaction was used to examine thechanges of mRNA of insulin-like growthfactor-1 (c).



of aging-related genes.26 However, the exact mechanisms ofUVB on skin aging and related signalling pathways are stilla matter of debate. Previous studies by this and other labo-ratories have suggested that down regulation of c-Myc, p-Rband expression of p53 may be involved in the acceleratedsenescence response to stress.27,28 Western blotting indi-cated that the c-Myc, p-Rb and p53 involved in the acceler-ated senescence response to UVB in human diploidfibroblasts. This was supported by the observation that thec-Myc, p-Rb and p53 in tiron-treated cells were stronglyrecoveried by tiron. Cells with high levels of p16 at senes-cence failed to proliferate, p16 over-expression, which is ahallmark of both replicative and accelerated senescence.29

One interesting finding was that the expression of p16was significantly upregulated by UVB treatment, whereasco-treatment of tiron inhibited the induction of p16. It iswell known that the methylation of p16 promoter repressthe transcription.30 How UVB and tiron regulate the expres-

sion of p16 deserved further investigation. There are2 phases of senescence-the morphological and thesenescence-associated secretory phenotype (SASP). p38MAPK is one of the important regulatory molecules on theSASP. The Western blot results confirmed that the p38 sig-nalling pathway was involved in UVB-induced fibroblastsaging process, Tiron itself could not activate the p38, but itinhibited the activation of p38 UVB induced in fibroblasts.The study found that the IGF-1 is highly expressed innormal but not in aging fibroblasts. RT-PCR results showedthat IGF-1 was hardly affected by tiron that it was able torestore the UVB-induced downregulation of IGF-1. Thesedata suggest that tiron antagonistic UVB induced fibroblastsenescence.

Because SOD is an extremely efficient enzyme that cataly-ses the neutralisation of superoxide nearly as fast as the twocan diffuse together spontaneously in solution, the treatmentof UVB could increase intracellular superoxide anion levels.

100

90

80

70

60

50

40

30

20

10

0

100

90

80

70

60

50

40

30

20

10

0

80

70

60

50

40

30

20

10

0

SO

D a

ctiv

ity

(% o

f X

O in

hibi

tion

)S

OD

act

ivit

y (%

of

XO

inhi

biti

on)

SO

D a

ctiv

ity

(% o

f X

O in

hibi

tion

)

0 10 50 100 200 500 0 10 30 60 120 180 minutesμmol/L

40 mJ/cm2 UVBTiron

*

*

*

*

UVB

Tiron

_ _ +

_ +

+

_ +

A

C

B

Figure 6 Tiron and UVB affect superoxide dismutase (SOD) activity in fibroblasts. (a) Human diploid fibroblasts cells were treated with 0,10, 50, 100, 200, and 500 mmol/L tiron for 2 h. (b) Human diploid fibroblasts cultured in 96-well plates were exposed to UVB irradiation of40 mJ/cm2 at 312 nm. The SOD activity was determined at the time specified. (c) Human dermal fibroblasts pretreated with 200 mmol/L tironfor 2 h were exposed to 40 mJ/cm2 UVB. The cells were incubated with or without 200 mmol/L tiron for 2 h. The SOD activity was determined.SOD activity was expressed as the percentage of xanthine oxidase activity inhibition. The experiments were done in triplicate.

8 Y Fang et al.


We hypothesised that UVB may reduce the activityof SOD in fibroblasts, resulting in intracellular superoxideanions increased, and thus induce fibroblast aging. In fact,after the treatment of fibroblasts with UVB, SOD activity infibroblasts was decreased significantly. The experimentalresults indicated that tiron did not significantly increase theactivity of SOD in the fibroblasts but it could restore SODactivity in fibroblasts UVB-induced. These data suggest thattiron may antagonise the UVB-induced senescence by stabi-lising the activity of SOD. As for, how the tiron stableSOD activity in fibroblasts UVB-induced is worth furtherexploration.

In summar, tiron could abrogate the increase of superox-ide anion and glutathione depletion and also block UVB-induced senescence. Our results supported the hypothesisthat tiron protects skin fibroblasts against senescence bypreventing intracellular oxidative stress.

REFERENCES

1. Brenneisen P, Sies H, Scharffetter-Kochanek K. Ultraviolet-Birradiation and matrix metalloproteinases: from induction viasignaling to initial events. Ann. N. Y. Acad. Sci. 2002; 973: 31–43.

2. Wang YN, Wu W, Chen HC et al. Genistein protects againstUVB-induced senescence-like characteristics in human der-mal fibroblast by p66Shc down-regulation. J. Dermatol. Sci.2010; 58: 19–27.

3. Anna B, Blazej Z, Jacqueline G et al. Mechanism of UV-relatedcarcinogenesis and its contribution to nevi/melanoma. ExpertRev. Dermatol. 2007; 2: 451–69.

4. Buechner N, Schroeder P, Jakob S et al. Changes of MMP-1 andcollagen type Ialpha1 by UVA, UVB and IRA are differentiallyregulated by Trx-1. Exp. Gerontol. 2008; 43: 633–7.

5. Lee BC, Lee SY, Lee HJ et al. Anti-oxidative and photo-protective effects of coumarins isolated from Fraxinus chinen-sis. Arch. Pharm. Res. 2007; 30: 1293–301.

6. Catalano A, Rodilossi S, Caprari P et al. 5-lipoxygenase regu-lates senescence-like growth arrest by promoting ROS-dependent p53 activation. EMBO J. 2005; 24: 170–9.

7. Macip S, Igarashi M, Fang L et al. Inhibition of p21-mediatedROS accumulation can rescue p21-induced senescence. EMBOJ. 2002; 21: 2180–8.

8. Krishna CM, Liebmann JE, Kaufman D et al. The cate-cholic metal sequestering agent 1,2-dihydroxybenzene-3,5-disulfonate confers protection against oxidative cell damage.Arch. Biochem. Biophys. 1992; 294: 98–106.

9. Kim JS, Cho EW, Chung HW et al. Effects of tiron, 4,5-dihydroxy-1,3-benzene disulfonic acid, on human promyeloticHL-60 leukemia cell differentiation and death. Toxicology2006; 223: 36–45.

10. Middleton E Jr, Kandaswami C, Theoharides TC. The effects ofplant flavonoids on mammalian cells: implications for inflam-mation, heart disease, and cancer. Pharmacol. Rev. 2000; 52:673–751.

11. Shibayama-Imazu T, Sonoda I, Sakairi S et al. Production ofsuperoxide and dissipation of mitochondrial transmembranepotential by vitamin K2 trigger apoptosis in human ovariancancer TYK-nu cells. Apoptosis 2006; 11: 1535–43.

12. Nirala SK, Bhadauria M, Upadhyay AK et al. Reversal of effectsof intra peritoneally administered beryllium nitrate by tironand CaNa3DTPA alone or in combination with a-tocopherol.Indian J. Exp. Biol. 2009; 47: 955–63.

13. Ito Y, Pagano PJ, Tornheim K et al. Oxidative stress increasesglyceraldehyde-3-phosphate dehydrogenase mRNA levels inisolated rabbit aorta. Am. J. Physiol. 1996; 270: H81–7.

14. Kobayashi M, Sasaki K, Enomoto M et al. Highly sensitivedetermination of transient generation of biophotons duringhypersensitive response to cucumber mosaic virus in cowpea.J. Exp. Bot. 2007; 58: 465–72.

15. Furuichi T, Liu W, Shi H et al. Generation of hydrogen perox-ide during brief oxygen-glucose deprivation induces precon-ditioning neuronal protection in primary cultured neurons. J.Neurosci. Res. 2005; 79: 816–24.

16. Fink B, Laude K, McCann L et al. Detection of intracellularsuperoxide formation in endothelial cells and intact tissuesusing dihydroethidium and an HPLC-based assay. Am. J.Physiol. Cell Physiol. 2004; 287: C895–902.

17. Park WH, Han YW, Kim SH et al. An ROS generator, antimycinA, inhibits the growth of HeLa cells via apoptosis. J. Cell.Biochem. 2007; 102: 98–109.

18. Han YH, Moon HJ, You BR et al. The anti-apoptotic effects ofcaspase inhibitors on propyl gallate-treated HeLa cells in rela-tion to reactive oxygen species and glutathione levels. Arch.Toxicol. 2009; 83: 825–33.

19. Dimri GP, Lee X, Basile G et al. A biomarker that identifiessenescent human cells in culture and in aging skin in vivo.Proc. Natl Acad. Sci. U.S.A. 1995; 92: 9363–7.

20. Bridges JP, Gilbert JS, Colson D et al. Oxidative stress con-tributes to soluble Fms-like tyrosine kinase-1 induced vasculardysfunction in pregnant rats. Am. J. Hypertens. 2009; 22:564–8.

21. Fan L, Sawbridge D, George V et al. Chronic cocaine-inducedcardiac oxidative stress and mitogen-activated protein kinaseactivation: the role of Nox2 oxidase. J. Pharmacol. Exp. Ther.2009; 328: 99–106.

22. Tsou PS, Addanki V, Fung HL. Dissociation between superox-ide accumulation and nitroglycerin-induced tolerance. J. Phar-macol. Exp. Ther. 2008; 327: 97–104.

23. Debacq-Chainiaux F, Borlon C, Pascal T et al. Repeated expo-sure of human skin fibroblasts to UVB at subcytotoxic leveltriggers premature senescence through the TGF-b1 signalingpathway. J. Cell Sci. 2005; 118: 743–58.

24. Han YH, Kim SH, Kim SZ et al. Apoptosis in arsenic trioxide-treated Calu-6 lung cells is correlated with the depletion ofGSH levels rather than the changes of ROS levels. J. Cell.Biochem. 2008; 104: 862–78.

25. Han YH, Moon HJ, You BR et al. The effect of MAPK inhibitorson arsenic trioxide-treated Calu-6 lung cells in relation to celldeath, ROS and GSH levels. Anticancer Res. 2009; 29: 3837–44.

26. Weinreb O, Drigues N, Sagi Y et al. The application of proteom-ics and genomics to the study of age-related neurodegenera-tion and neuroprotection. Antioxid. Redox Signal. 2007; 9:169–79.

27. Sashida G, Liu Y, Elf S et al. ELF4/MEF activates MDM2expression and blocks oncogene-induced p16 activation topromote transformation. Mol. Cell. Biol. 2009; 29: 3687–99.

28. Shamma A, Takegami Y, Miki T et al. Rb regulates DNAdamage response and cellular senescence through E2F-dependent suppression of N-ras isoprenylation. Cancer Cell2009; 15: 255–69.

29. Weebadda WK, Jackson TJ, Lin AW. Expression of p16INK4Avariants in senescent human fibroblasts independent ofprotein phosphorylation. J. Cell. Biochem. 2005; 94: 1135–47.

30. Kondo E, Gu Z, Horii A et al. The thymine DNA glycosylaseMBD4 represses transcription and is associated with methy-lated p16(INK4a) and hMLH1 genes. Mol. Cell. Biol. 2005; 25:4388–96.



Documents

Tiron protects against UVB-induced senescence-like characteristics in human dermal fibroblasts by the inhibition of superoxide anion production and glutathione depletion