National Center for Toxicological Research, U.S. Food and Drug

UVA photoirradiation ofbenzo[a]pyrene metabolites:induction of cytotoxicity, reactiveoxygen species, and lipid peroxidation

Qingsu Xia1, Hsiu-Mei Chiang1,2, Jun-Jie Yin3,Shoujun Cheng1, Lining Cai4, Hongtao Yu5andPeter P Fu1

Administration, Jefferson, AR, USA

Department of Cosmecutics, China Medical University,

Taichung, Taiwan, Republic of China

Center for Food Safety and Applied Nutrition, U.S. Food and

Drug Administration, College Park, MD, USA

Biotranex LLC, Monmouth Junction, NJ, USA

Department of Chemistry and Biochemistry, Jackson State

University, Jackson, MS, USA

Corresponding author:

Peter P Fu, National Center for Toxicological Research, U.S. Food

and Drug Administration, Jefferson, AR 72079, USA.

Email:

Abstract

Benzo[a]pyrene (BaP) is a prototype for studying carcinogenesis of polycyclic aromatic hydrocarbons (PAHs).We have long been interested in studying the phototoxicity of PAHs. In this study, we determined thatmetabolism of BaP by human skin HaCaTkeratinocytes resulted in six identified phase I metabolites, forexample, BaP trans-7,8-dihydrodiol (BaP t-7,8-diol), BaP t-4,5-diol, BaP t-9,10-diol, 3-hydroxybenzo[a]pyrene(3-OH-BaP), BaP (7,10/8,9)tetrol, and BaP (7/8,9,10)tetrol. The photocytotoxicity of BaP, 3-OH-BaP, BaPt-7,8-diol, BaP trans-7,8-diol-anti-9,10-epoxide (BPDE), and BaP (7,10/8,9)tetrol in the HaCaTkeratinocyteswas examined. When irradiated with 1.0 J UVA light, these compounds when tested at doses of 0.1, 0.2, and0.5 mM, all induced photocytotoxicity in a dose-dependent manner. When photoirradiation was conducted inthe presence of a lipid (methyl linoleate), BaP metabolites, BPDE, and three related PAHs, pyrene, 7,8,9,10-tetrahydro-BaP trans-7,8-diol, and 7,8,9,10-tetrahydro-BaP trans-9,10-diol, all induced lipid peroxidation. Theformation of lipid peroxides by BaP t-7,8-diol was inhibited by NaN 3 and enhanced by deuterated methanol,which suggests that singlet oxygen may be involved in the generation of lipid peroxides. The formation of lipidhydroperoxides was partially inhibited by superoxide dismutase (SOD). Electron spin resonance spin trappingexperiments indicated that both singlet oxygen and superoxide radical anion were generated from UVA photo-irradiation of BPDE in a light dose responding manner.

Keywords

Polycyclic aromatic hydrocarbons (PAHs), Benzo[a]pyrene, metabolite, lipid peroxidation, reactive oxygenspecies (ROS)

Introduction

Polycyclic aromatic hydrocarbons (PAHs)1are wide-spread genotoxic environmental pollutants. PAHsrequire metabolic activation in order to exert theirgenotoxic and carcinogenic activities (Harvey, 1997;NTP, 1998). There are three metabolic activationpathways of PAHs leading to tumorigenicity, namely,

metabolism into bay-region diolepoxides, formationof radical cation intermediates, and formation ofquinones (Harvey, 1997). Among these mechanisms,metabolism into bay-region diolepoxides has been

most extensively studied and has been determined toplay the most important role in vivo (Harvey, 1991,1997). Benzo[a]pyrene (BaP) is a prototype for stud-

ies of PAH-induced carcinogenesis. The metabolic

activation of BaP leading to bay-region diolepoxides

for tumor formation is shown in Figure 1.

Since PAHs are phototoxic (Yu, 2002), it is

important and significant to determine human health

risks posed by PAHs concomitantly exposed to

sunlight. It is known that human skin can metabolize

PAHs (Mukhtar, 1995). It is also important to

determine whether or not PAH metabolites are

phototoxic.

We have long been interested in studying the

UVA-induced phototoxicity and lipid peroxidation

of PAHs and derivatives, including parent PAHs

(Chiang et al., 2010; Yu et al., 2005, 2006), methy-

lated PAHs (Xia et al., 2006a), ethylated PAHs (Chen

et al., 2007), and halogenated PAHs (Herreno-Saenz

et al., 2006). We found that UVA photoirradiation

of PAHs generated reactive oxygen species (ROS)

that mediated the lipid peroxide formation (Xia

et al., 2006b). As a continuation of our studies, in this

present investigation we studied the metabolism of

BaP by human skin HaCaTkeratinocytes, identified

the metabolites, performed UVA photoirradiation of

BaP and its metabolites in the presence of methyl

linoleate to determine lipid peroxidation formation,

and employed electron spin resonance (ESR) spin

trapping methodology to detect ROS formation. We

found that the tested compounds generated lipid

peroxidation mediated by ROS.

Methods and materials

Materials

BaP, pyrene, methyl linoleate, sodium azide (NaN3 ),

superoxide dismutase (SOD), berberine and 2,2,6,6-

tetramethyl-piperidine (TEMP) were purchased from

Sigma-Aldrich (St. Louis, MO).BaP (7,10/8,9)-tetrol

(BaPtetrol-Ia) and BaP (7/8,9,10)-tetrol (BaPtetrol-

Ib) were purchased from the NCI Chemical Carcino-

gen Reference Standard Repositories at the Midwest

Research Institute (Kansas City, KS). 3-Hydroxy-

BaP (3-OH-BaP), BaP trans-4,5-dihydrodiol (BaP

t-4,5-diol), BaP trans-7,8-dihydrodiol (BaP t-7,8-

diol), BaP trans-9,10-dihydrodiol (BaP t-9,10-diol),

BaP trans-7,8-diol anti-9,10-epoxide (BPDE),

7,8,9,10-tetrahydro-BaP trans-7,8-diol (H 4 BaP 7,8-

diol), and 7,8,9,10-tetrahydro-BaP trans-9,10-diol

(H 4 BaP 9,10-diol) were synthesized in our laboratory

according to previously described methods (Chen

et al., 1997; Fu et al., 1978, 1982; Harvey and Fu,

1978). The purity of all the PAHs and metabolites was

analyzed by high-performance liquid chromatography

(HPLC) and found to be >99% pure. Dulbecco’s mod-

ified Eagle’s medium (DMEM), phosphate-buffered

saline (PBS, pH 7.4), fetal bovine serum (FBS), tryp-

sin, penicillin, and streptomycin were purchased from

Invitrogen Inc. (Grand Island, NY). The CellTiter

One Solution Cell Proliferation Assay (MTS) kit

was purchased from Promega Co. (Madison, WI). All

other reagents used were at least of analytical grade.

The nitrone spin trap, 5-tert-butoxycarbonyl 5-methyl-

1-pyrroline N-oxide (BMPO), was purchased from

Applied Bioanalytical Labs (Sarasota, FL).

Human skin HaCaTkeratinocytes, a transformed

human epidermal (HaCaT) cell line, were obtained

from Dr. Norbert Fusenig of the German Cancer

Research Centre (Heidelberg, Germany).

Light sources

The UVA light box was custom made using 4 UVA

lamps (National Biologics, Twinsburg, OH) (Cherng

et al., 2005). The spectral irradiance of the light box

was determined using an Optronics OL754 Spectrora-

diometer (Optronics Laboratories, Orlando, FL), and

the light dose was routinely measured using a Solar

Light PMA-2110 UVA detector (Solar Light Inc.,

Philadelphia, PA). The maximum emission of the

UVA light box was determined to be between 340 and

355 nm. The light intensities at wavelengths below

320 nm (UVB light) and above 400 nm (visible light)

are approximately two orders of magnitude lower

than the maximum in the 340–355 nm spectral region

(Chiang et al., 2010).

For the study of BaP and metabolites in keratino-

cytes concomitantly exposed to UVA light, UVA

light of 1, 2, or 4 J/cm

(at the light dose 9 mW/

, 1 J equal to 2 min 10sec of light) exposure was

used. For the UVA light-induced lipid peroxidation

byBaP and metabolites, the UVA-irradiation dose

was from 7 to 21 J/cm

, which were obtained by

approximately 23 to 69 min exposures at the dose rate

of 5 mW/cm

. UVA of 10 J/cm

equates to about 2 h

of exposure at the noon time of sunny days during

summer around the world, based upon observations

of UVA intensity of 2.1 mW/cm

in Okayama, Japan,

in September (Arimoto-Kobayashi et al., 2000), 3.6

mW/cm

in Jackson, MS, USA, in August (Yu

et al., 2001), 5.4 mW/cm

in Paris, France, in July

(Jeanmougin and Civatte, 1987), and 6.6 mW/cm

in Coimbatore, India, in July (Balasaraswathy et al.,

2002).

Instrumentation

A Waters HPLC system consisting of a Model 600

controller, a Model 996 photodiode array detector,

and pump was used for the separation and purification

of the methyl linoleatehydroperoxide products. Con-

ventional ESR spectra were obtained with a Bruker

EMX spectrometer (Bruker Instruments Inc, MA).

The scan width was 100 G for both BMPO and TEMPexperiments. ESR signals were recorded with 15 mW

incident microwave and 100 kHz field modulation

of 1.25 (for TEMP) and 1 G (for BMPO-OOH).

The scan ranges were 80 (for TEMP) and 100 G (for

BMPO-OOH), respectively. All measurements were

performed at room temperature.

Metabolism of BaP in human HaCaT

keratinocytes

The HaCaT cells, with a density of 8 10

permL,

were cultured in two 15-cm petri dishes each contain-

ing 25 mL DMEM supplemented with 10% (v/v) FBS

and antibiotics (100 U/mL penicillin and 100 mg/mL

streptomycin) at 37

C in a humidified atmosphere

containing 5% CO 2 . After incubation for 20 h, the

DMEM was removed, a solution of 25 mL of FBS free

DMEM and 50 mL of 10 mMBaP in DMSO was

added to the cells in each dish, and the resulting mix-

tures were incubated at 37

C for 48 h. The cells and

medium in each dish were transferred into a 25-mL

centrifuge tube with scraping followed by the addition

of 2.5 mL acetone and stirring for 5 min to quench the

enzymatic reaction and break the cell membranes.

Upon centrifugation at 3,000 ppm for 10 min, the

resulting supernatants were filtered, combined, and

extracted twice with 150 mL ethyl acetate. The ethyl

acetate layer was collected and the solvent removed

under reduced pressure by rotary evaporator, and the

resulting residue was subjected for liquid chromato-

graphy–mass spectroscopy (LC/MS) analysis of the

metabolites formed.

LC/MS analysis

Liquid chromatography. The liquid handling system

consisted of a Shimadzu Prominence HPLC system,

including a CBM-20A system controller, two LC-

20AD pumps, an SIL-20AC HT autosampler, an

SPD-20A UV/VIS detector (Shimadzu Scientific

Instrument, Columbia, MD), and an automated

switching valve (TPMV, Rheodyne, Cotati, CA). The

switching valve was used to divert the column efflu-

ent to either waste or the MS instrument. The

Shimadzu Prominence HPLC system was used for

sample injection and separation. Each sample was

loaded onto a reverse phase column (Ace 3 C18,

4.6 mm 150 mm, 3 mm, MAC-MOD Analytical,

Chadds Ford, PA) eluded with a water/acetonitrile

gradient at 0.6 mL/min, and the sample components

were eluted into the mass spectrometer. The mobilephases were 10 mM ammonium acetate with 0.1%

formic acid in water and acetonitrile containing

0.1% formic acid. The initial gradient consisted of

5% acetonitrile for 2 min followed by a linear gradient

up to 95% acetonitrile over 36 min, after holding 95%

acetonitrile for 4 min, and the initial conditions were

reset in 1 min. The analytical column was equilibrated

with the starting mobile phase for 10 min.

Mass spectrometry. An AB Sciex 4000 QTrap LC/MS/

MS system (AB Sciex, Foster City, CA), equipped

with a Turbo V ™ ion source with a desolvation tem-

perature setting at 450

C. Nitrogen was used as cur-

tain gas, nebulizer gas, heater gas, and collision gas.

The samples were acquired in positive electrospray

ionization (ESI) mode and the scan types were the

enhanced MS (EMS) (scan from 100 to 800 Da in

0.8031s) and enhanced product ion (EPI). The meta-

bolites were identified by comparing to their HPLC

retention times and product ion fragmentation with

the authentic standards under the same LC/MS

conditions.

Photocytotoxicity assays in HaCaTkeratinocytes

Human skin HaCaTkeratinocyte cells were cultured

in DMEM supplemented with 10% (v/v) FBS and

antibiotics (100 U/mL penicillin and 100 mg/mL

streptomycin) at 37

C in a humidified atmosphere

containing 5% CO 2 . One day before the experiment,

cells were plated into 96-well plates at a density of

1 10

cells per well in 100 mL culture medium and

incubated for overnight. Afterward, culture medium

in the plates was replaced by 50 mL/well phenol red

and serum free DMEM containing control or different

concentrations (0.1, 0.2, 0.5, and 1 mM) of test chem-

ical and incubated for 2 h. The cells were exposed to

UVA light of 1, 2, or 4 J/cm

. After light irradiation,

the chemical containing medium was aspirated, 100

mL/well DMEM with 2% FBS was added and the

incubation was continued for 24 h.

The effects of tested chemicals on the cell viability

were evaluated by 3-(4,5-dimethylthiazole-2-yl)-5-

(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tet

razo-lium (MTS) cell viability assay using the CellTi-

ter 961 One Solution Cell Proliferation Assay kit

according to the manufacture’s instruction. Briefly,

20 mL per well MTS solution was added to each well

and incubated at 37

C for 3 h, then optical density

(OD) was read on a microplate reader (Bio Tec, USA AQ 1 )

at 490 nm. The results were expressed as percentageviability compared with the untreated controls.

Berberine (10 mM) was used as a positive control.

UVA light-induced lipid peroxidation by BaP, BaP

metabolites, and related compounds

Experiments were conducted with a solution of

100 mM methyl linoleate and 1.0 mM substrate in

methanol. The tested compounds include BaP,

3-OH-BaP, BaP t-4,5-diol, BaP t-7,8-diol), BaP

t-9,10-diol, BPDE, BaPtetrolIa, pyrene, H 4 BaP 7,8-

diol, and H 4 BaP 9,10-diol. Samples were placed in a

UV-transparent cuvette and irradiated with 0, 7, and

21 J/cm

of UVA light. After irradiation, the methyl

linoleatehydroperoxide products were separated by

HPLC using a Prodigy 5 mm ODS column (4.6

250 mm, Phenomenex, Torrance, CA) eluted isocrati-

cally with 10% water in methanol (v/v) at 1 mL/min.

The levels of lipid peroxidation were determined by

HPLC and quantified by monitoring the elution of

HPLC peak areas at 235 nm (Cherng et al., 2005;

Tokita and Morita, 2000 AQ 2 ) followed by conversion to

the concentration based on the molar extinction coef-

ficient (at 235 nm) reported by Gibian and Vanden-

berg (1987).

UVA light-induced lipid peroxidation by BaP

t-7,8-diol in the presence of a free radical

scavenger or enhancer

The experiments were carried out as described above,

with the exception that parallel experiments were

conducted in the presence NaN 3 or SOD. The concen-

tration of SOD was 200 U/mL and NaN 3 was 20 mM.

It has been established that the lifetime of singlet

oxygen is longer in deuterated solvents, such as deu-

terium oxide (D 2 O) or deuterated methanol (CH 3 OD),

than in protic solvent (Foote, 1991 AQ 3 ). The effect on the

levels of lipid peroxide formation induced by UVA

photoirradiation of BaP t-7,8-diol with methanol

(CH 3 OH) and CH 3 OD was analyzed.

Detection of ROS using ESR spectroscopy spin

trapping

The UVA light photoirradiation of BPDE was

performed in situ. Two 50 mL quartz capillaries were

used. The UVA light was provided by a Schoeffel

500 W Xenon lamp coupled with a Schoeffel grating

monochromator. All experiments were performed in

duplicate. The data were obtained with error of less

than 10%.

Detection of singlet oxygen. Singlet oxygen generated

from the photoirradiation of BPDE with UVA light

was detected by the ERS spin trapping detection

method using spin trap TEMP (Rinalducci et al.,

2004). A solution of BPDE, 0.2 mM in 95% acetoni-

trile in water, and 5 mM spin trap TEMP was photo-

irradiated with UVA light at wavelength 344 nm. The

ESR spectra were recorded at room temperature at 0,

5, 10, 20, and 30 min, respectively.

Detection of superoxide anion radicals. The commonly

used specific spin trap BMPO is employed for detection

of superoxide anion radicals (Xia et al., 2006b; Zhao

et al., 2001). UVA photoirradiation of 0.2 mM BPDE

and 25 mM spin trap BMPO in 20% acetonitrile in water

was carried out at wavelength 344 nm, and the ESR

spectra were recorded at room temperature after 0, 5,

10, 20, and 30 min of irradiation, respectively.

Statistical analysis

The experiments were repeated twice or three times

independently. All data were analyzed using t test, with

P < 0.05 considered to be a significant difference.

Results

Metabolism of BaP in human skin HaCaT

keratinocytes

After BaP was incubated with human skin HaCaT

keratinocytes, the cells and medium were centrifuged

and the resulting supernatants were collected and

extracted by ethyl acetate. The ethyl acetate layer was

collected, the solvent was removed under reduced

pressure, and the resulting residue was analyzed by

LC/MS. The reversed phase LC/MS profiles of the

incubation mixture obtained from metabolism of BaP

by human skin HaCaTkeratinocytes and the identi-

fied organic extractable metabolites are shown in

Figure 2. Based on comparison of their HPLC reten-

tion time, UV-visible spectrum, and mass spectral

pattern with those of synthetic standards, the metabo-

lites contained in chromatographic peaks eluted at

18.1, 19.3, 23.9, 25.3, 25.7, and 33.3 min were iden-

tified as BaP tetrol Ia, BaP tetrol Ib, BaP t-9,10-

diol, BaP t-4,5-diol, BaP t-7,8-diol, and 3-OH-BaP,

respectively (Figure 2). The chromatographic UV

peak eluted at 40.1 min is the recovered BaP (not

shown in Figure 2). BaPtetrol-Ia and BaPtetrol-Ib are

hydrolysis products of BaP anti-7,8-diol anti-9,10-

epoxide with the 9,10-epoxy ring opened at the sameside and the opposite of the 7-hyroxyl group, respec-

tively. Based on their peak areas shown in the top

panel of Figure 2, BaPtetrol-Ia was formed at a quan-

tity larger than BaPtetrol-Ib, which is consistent with

the results obtained from liver microsomalmetabo-

lism of BaP (Yang et al., 1978). The HPLC elution

order of BaP t-9,10-diol, BaP t-4,5-diol, and BaP t-

7,8-diol is also consistent with that reported from liver

microsomal metabolism of BaP (Yang et al., 1978).

Photocytotoxicity of BaP and its metabolites in

human skin HaCaTkeratinocytes

Induction of photocytotoxicity of BaP and its four meta-

bolites, 3-OH-BaP, BaP t-7,8-diol, BPDE, and BaP

tetrol-Ia in human skin HaCaT keratinocytes was

assessed. In the absence of light irradiation, all the tested

chemicals with concentrations at 0.1, 0.2, 0.5, or 1 mM

did not induce cytotoxicity (Table 1). Thus, for photocy-

totoxicity determination, these concentrations (0.1, 0.2,

0.5, and 1 mM) were used. The cells were treated with

substrates in serum free DMEM for 24 h, cell viabilities

were examined by the MTS assay, which is based on the

reduction of MTS by cellular metabolism to a soluble

formazan product measured by its absorbance at

490 nm, and the quantity of its formation is directly pro-

portional to the number of live cells.

The compounds at doses of 0.1, 0.2, and 0.5 mM

concomitantly irradiated with 1.0 J UVA light all

inducedphotocytotoxicity in a light dose-dependent

and substrate dose-dependent manner; although BaP

t-7,8-diol, BPDE, and BaPtetrol-Ia at low dose (0.1

mM) did not exert photocytotoxicity (Table 1). It is

worthy to note that BaPtetrol-Ia exhibited the lowest

photocytotoxicity. Even when BaPtetrol-Ia at 1.0 mM

concomitantly exposed to 1 or 2 J UVA light, no

photocytotoxicity occurred. When exposed to 4.0 J,

BaPtetrol-Ia at 4.0 mM resulted in 20% cell death

(data not shown).

The overall results indicate that 3-OH-BaP is most

photocytotoxic and BaPtetrol-Ia is least photocyto-

toxic in human skin HaCaTkeratinocytes (Table 1).

The results summarized in Table 1 also indicate that

berberine, the positive control, possesses higher

photocytotoxicity than all the tested compounds.

UVA light-induced lipid peroxidation by BaP, BaP

metabolites, and related compounds