Cell Biology International 32 (2008) 1380e1387www.elsevier.com/locate/cellbi

Impaired activation of caspase cascade during cell death inducedby newly synthesized singlet oxygen generator,1-buthylnaphthalene-4-propionate endoperoxide

Kaoru Otsu a,*, Kazuaki Sato b, Michihiko Sato c, Hideyu Ono d, Yoshihiro Ohba b,Yohtaro Katagata e

a Environmental Preservation Center, Yamagata University, 2-2-2 Iida-nishi, Yamagata 990-9585, Japanb Department of Chemical Engineering, Faculty of Engineering, Yamagata University, 4-3-16 Jonan, Yonezawa 992-8510, Japan

c Central Laboratory for Research and Education, Yamagata University School of Medicine, 2-2-2 Iida-nishi, Yamagata 990-9585, Japand Department of Biochemistry, Yamagata University School of Medicine, 2-2-2 Iida-nishi, Yamagata 990-9585, Japan

e Department of Biochemistry and Biotechnology, Faculty of Agriculture and Life Science, Hirosaki University, Hirosaki 036-8561, Japan

Received 3 September 2007; revised 28 April 2008; accepted 12 August 2008

Abstract

Endoperoxides of naphthalene derivatives generate singlet oxygen under physiological conditions. Here we have synthesized a new endo-peroxide of a naphthalene derivative, 1-buthylnaphthalene-4-propionate endoperoxide (BNPE), and studied its cytotoxic properties on HepG2and HaCaT cells. BNPE induced cell death at much lower concentration than 1-methylnaphthalene-4-propionate endoperoxide (MNPE) andnaphthalene dipropionate endoperoxide (NDPE). A positive correlation exists between the amount of endoperoxide incorporated into cells andits cytotoxic ability. The cytotoxic effect of BNPE was attenuated by a-tocopherol but not by sodium azide. In contrast, the effects of MNPE andNDPE were attenuated by both a-tocopherol and sodium azide. The caspase cascade in cells treated with endoperoxide was impaired. Caspaseactivity in a soluble protein fraction were inhibited similarly by the above three endoperoxides. These results suggest an abortive apoptoticpathway due to the suppression of caspase activation is a general feature of cell death induced by singlet oxygen.� 2008 International Federation for Cell Biology. Published by Elsevier Ltd. All rights reserved.

Keywords: Singlet oxygen; Endoperoxide; Naphthalene derivative; Caspase cascade; Apoptosis

1. Introduction

Singlet oxygen is one of the reactive oxygen speciesresponsible for skin damage induced by long-wave ultraviolet(UVA) radiation (Krutmann, 2000; Klotz et al., 2001; Tyrrell,2004) and for the cytotoxic effect in photodynamic therapyagainst tumors (Pass, 1993; Moan and Peng, 2003; Tries-scheijn et al., 2006). Several methods for generating singletoxygen in cells have been developed and utilized, and one ofthe most widely used methods so far involves treating cellswith permeable photosensitizers followed by visible light

* Corresponding author. Tel.: þ81 23 628 5142; fax: þ81 23 628 5143.

E-mail address: yootsu@med.id.yamagata-u.ac.jp (K. Otsu).

1065-6995/$ - see front matter � 2008 International Federation for Cell Biology.

doi:10.1016/j.cellbi.2008.08.008

radiation (e.g. Muller-Breitkreutz et al., 1995; Zhuang et al.,1998). This method is easy to perform and does not requirespecial equipment. Rose Bengal and Methylene Blue are themost frequently used photosensitizers. Their limitation,however, is the generation of free radicals in addition to singletoxygen (Halliwell and Gutteridge, 1999). To avoid thisproblem, endoperoxides of naphthalene derivatives weresynthesized and used for over 20 years (Saito et al., 1981;Nieuwint et al., 1985; Pierlot et al., 2000). Endoperoxidesgenerate ‘‘pure’’ singlet oxygen with no radical formationthrough thermal decomposition in physiological conditions.

The mechanism of cell death induced by singlet oxygen hasbeen studied using various singlet oxygen generating systems.When photosensitizers were used for generating singlet

Published by Elsevier Ltd. All rights reserved.

1381K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

oxygen, most reports stated that an apoptotic cell death wasinduced. For example, HeLa cells treated with a benzopor-phyrin derivative monoacid ring A (Granville et al., 1998),HL-60 cells with Rose Bengal (Zhuang et al., 1998), thyroidfollicular carcinoma-derived cell line XTC.UC1 with Methy-lene Blue (Allia et al., 2003) and human colon carcinoma cellline 320DM with hexaminolevulinate (Shahzidi et al., 2006)all produced apoptotic cell death after the photodynamictreatments. UVA irradiation e another well known method forgenerating singlet oxygen inside cells e has been used onhuman T helper cells (Morita et al., 1997), human T and Blymphocytes (Godar, 1999), and human keratinocyte cell lineHaCaT cells (Morley et al., 2006) and were reported to induceapoptosis. In contrast, we recently found cell death induced by1-methylnaphthalene-4-propionate endoperoxide (MNPE)does not induce a typical apoptosis, cells showing incompletenuclear fragmentation, little DNA ladder formation and weakactivation of the caspase cascade (Otsu et al., 2005).

The above phenomenon induced by MNPE seems to be dueto singlet oxygen via inhibition of caspase activity (Otsu et al.,2005). Several reports describe the modification of enzymeactivities by singlet oxygen (Vinnikova et al., 1992; Tatsuzawaet al., 1998; Diaz et al., 2005). In addition to these, Glaseret al. (1988) suggested that a selective loss of histidine residueby singlet oxygen leads to inhibition of mitochondrial F1-ATPase activity. von Montfort et al. (2006) found that Cys215within the active site of the protein tyrosine phosphatase-1Bwas the only cysteine residue oxidized by singlet oxygen.Nagaoka et al. (2005), on the other hand, suggested that thecysteine residue in an active center of the enzyme was notoxidized by singlet oxygen. More work is clearly required toexplain the action of singlet oxygen on such enzymes.

Since the mechanism of cell death and the mode of caspaseinactivation induced by singlet oxygen remain obscure asabove we synthesized a new naphthalene endoperoxidederivative to help elucidate the action of singlet oxygen oncells.

2. Materials and methods

2.1. Materials

MNPE and naphthalene dipropionate endoperoxide(NDPE) were synthesized as described previously (Aubryet al., 1989). 1-Buthylnaphthalene-4-propionate endoperoxide(BNPE) was synthesized using a similar method as MNPEsynthesis (Scheme 1). The structure and purity of the endo-peroxide was determined by NMR. The half-lives of MNPE,NDPE, and BNPE were approx. 25, 27 and 20 min at 37 �C,respectively. The generation of singlet oxygen from MNPEand NDPE was detected by ESR using 4,40-bis-1-p-carbox-yphenyl- 3-methyl-5-hydroxyl]pyrazole (DRD156) as a sensi-tive singlet oxygen-detecting probe, which specifically reactswith singlet oxygen among the reactive oxygen species (Liuet al., 2001). Quantitative generation of singlet oxygen fromMNPE was reported by Saito et al. (1983) and Nakano et al.(1998).

2.2. Cell culture

The HaCaT cells, a spontaneously immortalized aneuploidhuman keratinocyte cell line, were a generous gift from DrFusenig (Boukamp et al., 1988). HaCaT cells and HepG2cells, a human hepatocellular carcinoma cell line, weremaintained in Dulbecco’s modified minimal essential medium(Sigma) containing 100 units/ml penicillin and 100 mg/mlstreptomycin supplemented with 10% fetal bovine serum(Invitrogen). These cells were grown at 37 �C in a humidifiedatmosphere containing 5% CO2 in air.

2.3. HPLC analysis of endoperoxides in cells

The delivery of endoperoxides to the cells was assessedessentially as before (Otsu et al., 2005). After incubation with2 mM BNPE, MNPE, or NDPE for 30 min, HepG2 cells(5 � 106) or HaCaT cells (4 � 106) were washed with phos-phate buffered saline (PBS) and scraped. Cells suspended in100 ml of PBS were disrupted for 1 min by a Bioruptor(Cosmo Bio, Tokyo, Japan) at 200 W with cooling in ice-coldwater. The sample was centrifuged at 100,000 � g for 60 min,and the resulting supernatant (PBS extract) was saved. Theprecipitate was suspended in 300 ml of ethanol by sonicationand centrifuged at 100,000 � g for 60 min. The supernatantwas saved, dried under a nitrogen stream, and dissolved inPBS (ethanol extract). Two extracts were subjected separatelyto HPLC (LC-2000; Jasco, Tokyo, Japan) on a reversed-phaseC18 column (Jasco, Tokyo, Japan). The decompositionproduct, 1-methylnaphthalene-4-propionate (MNP), naphtha-lene dipropionate (NDP), or 1-buthylnaphthalene-4-propionate(BNP) was eluted at 40 �C at 1 ml/min. The mobile phase wasmethanol/50 mM ammonium acetate, pH 7.0, with 20%methanol for 1 min, increased to 100% methanol in 8 min, andwas kept in 100% methanol for 4 min. Absorbance of theeluent was monitored at 230 nm. Amounts of the decomposedcompounds that had been incorporated into cells were stan-dardized based on the recovery of a known amount of MNP,NDP, or BNP added to the supernatant prepared fromuntreated control cells.

2.4. Evaluation of the viability of cells by lactatedehydrogenase (LDH) activity

LDH activity was measured to assess the viability of cellsafter endoperoxide treatments (Otsu et al., 2005). At 24 h afterthe treatments, portions of the medium were collected formeasuring LDH activity. The cells were disrupted by briefsonication in PBS containing 0.1% Tween 20. Cellular extractsfree of debris were prepared by centrifugation at 15,000 � gfor 10 min. The reaction mixture contained 2 ml of the sample,0.1 mM NADH, 1 mM sodium pyruvate, and 50 mM sodiumphosphate buffer, pH 7.2 in 200 ml. The initial rates of NADHdecrease were measured by monitoring the absorbance at340 nm using a microplate reader SUNRISE Remote(TECAN). The viability of the cells was calculated as thepercentage of LDH activity recovered in cellular extract

BrS

C4H9 C4H9

CH2Br

C4H9

CH2CH(COOC2H5)2C4H9

CH2CH(COOH)2

C4H9

CH2CH2COOH

C4H9

CH2CH2COOH

OO

a b c d

e f g

Scheme 1. Reagents and conditions: (a) 2-bromothiophene, Mg, Ni(dppp)Cl2, RT, 1 h; (b) RanneyeNickel, 50 �C, 2 h; (c) paraformaldehyde, HBr, AcOH, 100 �C,

1 h; (d) dried EtOH, malonic ester, 5 h; (e) EtOH, 5% NaOH, 65 �C, 2 h; (f) 150 �C, vacuum sublimation; (g) NaOH, Na2MoO2, pH 10e11 buffer, H2O2, 1.5 h.

1382 K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

against the total (cellular extract plus medium) recoveredactivity.

2.5. Caspase activity assay

A

20

40

60

80

100

Via

bilit

y (

)

BNPEBNPMNPEMNPNDPENDP

Caspase activity was determined essentially as described byStennicke and Salvesen (1997). Acetyl-Leu-Glu-His-Asp-4-methyl-coumaryl-7-amide and acetyl-Asp-Glu-Val-Asp-4-ethyl-coumaryl-7-amide (Wako, Osaka, Japan) were used forthe measurement of caspase-9 and caspase-3 activity, respec-tively. The reaction mixture contained 25 mM piperazine-1,40-bis(2-ethanesulphonic acid) (PIPES)eKOH, pH 7.2, 5 mMMgCl2, 10 mM dithiothreitol, 0.1% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulphonate (CHAPS), 10%sucrose and 0.05 mM substrate in 200 ml. The initial rates ofenzymatic hydrolysis were measured at 37 �C by the release ofmethyl-coumaryl-7-amide from the substrate as the emissionat 460 nm upon excitation at 380 nm using a SpectraMaxGEMINI EM Multiplate Spectrofluorometer (MolecularDevices Corporation Japan, Tokyo, Japan) equipped witha thermostated plate reader.

00.1 1 10 100 1000

2.6. Detection of fragmented DNA by agarosegel electrophoresis

Concentration (mM)

B

80

100BNPEBNPMNPEMNPNDPE

DNA was isolated from HepG2 and HaCaT cells (1 � 106

cells each) according to Ishizawa et al. (1991). Isolated DNAsamples were electrophoresed on a 2% agarose gel in TriseborateeEDTA buffer. DNA was visualized on a UV illumi-nator after staining with 0.5 mg/ml ethidium bromide.

60

lity

()

NDP

2.7. Statistics

0

20

40

0.1 1 10 100 1000

Concentration (mM)

Via

bi

Data are presented as the means � SD in triplicate assays.Student’s t-test was used to compare the significance ofdifferences between data. Values of P < 0.05 were consideredsignificant.

3. Results

Fig. 1. Comparison of cytotoxic effect of endoperoxide on HepG2 and HaCaT

3.1. Cytotoxic effect of endoperoxide cells. HepG2 cells (A) and HaCaT cells (B) were exposed to each endoper-

oxide (BNPE, MNPE, or NDPE) or its precursor compound (BNP, MNP, or

NDP) and viability of the cells was assessed by measuring LDH activity. The

means � SD of triplicate assays are shown.

To compare the cytotoxic effect of BNPE, MNPE andNDPE, LDH release in the medium was measured after

treating cells with each compound. HepG2 and HaCaT cellswere adopted since the former cell line was used in ourprevious study describing the mechanism of cell death inducedby MNPE (Otsu et al., 2005) and the latter was a spontane-ously transformed human epithelial cell line from adult skinwhich was suitable for elucidating the physiological role ofsinglet oxygen. BNPE was the most effective among the threeendoperoxides. The concentration of BNPE that lead to a 50%reduction in cellular viability (IC50) was 0.4 mM for HepG2cells and 1.0 mM for HaCaT cells (Fig. 1). MNPE was

Fig. 2. Morphological change of cell nuclei after treating with endoperoxides or etoposide. HepG2 and HaCaT cells were treated with each endoperoxide for an

initial 2 h, changed to fresh medium, and incubated for an additional 22 h. For etoposide treatment (Etop), 0.4 mM or 0.3 mM etoposide for HepG2 or HaCaT cells,

respectively, was present at all times. After the treatments, cells were trypsinized, washed in PBS, and suspended in PBS. About 5 � 105 cells in 10 ml were stained

with 0.03 mM DAPI, and photographs were taken with a digital camera under a fluorescent microscope (Olympus BX 50, Tokyo, Japan).

1383K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

Table 1

Incorporation of endoperoxide into HepG2 and HaCaT cellsa

Cells Endoperoxides PBS extract

(nmol/106 cells per 30 min)

Ethanol extract

(nmol/106 cells per 30 min)

Total

(nmol/106 cells per 30 min) (% in ethanol extract)

HepG2 BNPE 1.194 � 0.080 2.117 � 0.109 3.310 � 0.140 (64.0)

MNPE 0.092 � 0.007 0.028 � 0.002 0.120 � 0.007 (23.6)

NDPE 0.022 � 0.005 NDb 0.022 � 0.005 (0)

HaCaT BNPE 0.179 � 0.017 0.395 � 0.059 0.574 � 0.076 (68.8)

MNPE 0.041 � 0.006 0.008 � 0.002 0.049 � 0.008 (16.9)

NDPE 0.013 � 0.001 NDb 0.013 � 0.001 (0)

a HepG2 cells and HaCaT cells were incubated at 37 �C with 2 mM each of BNPE, MNPE, or NDPE for 30 min. The cells were washed with PBS, then extracted

sequentially with PBS and ethanol as described in Section 2. The amount of endoperoxide in the extracts was determined by HPLC. Data are presented as

means � SD for triplicate samples.b ND, not detected (less than 0.001 nmol/106 cells per 30 min).

A

20

40

60

80

100V

iabi

lity

()

noneNaN3VE

*

***

***

**

***

1384 K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

moderately effective and NDPE was the least effective. HepG2cells were approximately twice as sensitive to the respectiveendoperoxides as HaCaT cells. The precursor compoundsthemselves exhibited cytotoxicity, which were about twicethose of endoperoxides, except those of MNP on HepG2 cells.The effective concentration of MNPE to HepG2 cells was onesixth that of MNP.

The morphology of nuclei of HepG2 and HaCaT cellstreated with endoperoxides was examined under fluorescentmicroscopy. The morphological change of the nuclei aftertreating cells with each reagent was almost the same in theabove two cell types (Fig. 2). After treatment with etoposide,both cells exhibited fragmented nuclei representing cell deathfrom typical apoptosis. The nuclei after being treated witheach endoperoxide showed similar morphology to one another,thus most nuclei were condensed and some were slightlyfragmented. These results indicate that cell death induced byendoperoxides was not typical for apoptosis.

0CTR BNPE MNPE NDPE

100

3.2. BNPE is more easily delivered into cells thanMNPE and NDPE

noneNaN3VE

**

**

*****

***

***

B

0

20

40

60

80

CTR BNPE MNPE NDPE

Via

bilit

y (

)

Fig. 3. Antioxidants attenuate the cytotoxic effect of endoperoxides. HepG2

cells (A) and HaCaT cells (B) were preincubated with 5 mM sodium azide

(NaN3) for 30 min or 10 mM a-tocopherol (VE) for 16 h, then treated with

each endoperoxide for 2 h at 37 �C in a humidified atmosphere containing air

with 5% CO2. One day after the treatment, viability of the cells was assessed

by measuring LDH activity. CTR, control. The means � SD of triplicate

assays are shown. *P < 0.05 versus control; **P < 0.01 versus control;

***P < 0.001 versus control.

The amount of endoperoxide incorporated into cells duringthe treatment was assessed in order to estimate the level ofsinglet oxygen generation inside the cells. Although theabsolute values of each endoperoxide incorporated into HepG2and HaCaT cells were different from each other, BNPE wasthe most easily incorporated endoperoxide for both cells(Table 1). The amount of BNPE incorporated into respectivecells was 3.3 � 0.14 nmol/106 cells per 30 min, and this wasabout 28 times that of MNPE in HepG2 cells and0.57 � 0.08 nmol/106 cells per 30 min, and was about 12times that in HaCaT cells. Incorporation of NDPE into cellswas much less than that of MNPE. In addition, more than 64%of BNPE was found in the ethanol extract, which representsthe compound incorporated in the membrane compartment,and made a good contrast with more than 76% of MNPE andall of the NDPE which were recovered in the PBS extract,representing the compound incorporated in the cytosol frac-tion. The amount of BNPE, MNPE or NDPE recovered fromHepG2 cells was 5.8-, 2.4- or 1.7-fold that recovered fromHaCaT cells. This shows that HepG2 cells are larger than

HaCaT cells. The amounts of each endoperoxide incorporatedinto the above two cells differed according to the hydropho-bicity of the compounds. This may indicate that the cellsurface of HepG2 cells can more easily be accessed by

0

1

2

3

4

5

E P E P E P

CTR Etop BNP MNP NDP

Rel

ativ

e A

ctiv

ity

*

**

B

0

1

2

3

4

5

6

7

8

9

10

E P E P E P

CTR Etop BNP MNP NDP

Rel

ativ

e A

ctiv

ity

******

****** *

**

Caspase-9Caspase-3

Caspase-9Caspase-3

A

Fig. 4. Effect of BNPE, MNPE, or NDPE on caspase activation during cell

death. HepG2 cells (A) and HaCaT cells (B) were treated with each endo-

peroxide (E) or its precursor compound (P) for an initial 2 h, changed to fresh

medium, and incubated for an additional 15 h. For etoposide treatment (Etop),

0.4 or 0.3 mM etoposide for HepG2 cells or HaCaT cells, respectively, was

present at all times. A cellular lysate was prepared at the end of the incubation,

and activities of caspase-9 and -3 were assayed using specific substrates. CTR,

control. The means � SD of triplicate assays are shown. *P < 0.05 versus

precursor compound; **P < 0.01 versus precursor compound; ***P < 0.001

versus precursor compound.

A

0

20

40

60

80

100

120

0.1 1.0 10.0

Concentration (mM)

Rel

ativ

e ac

tivi

ty (

)

B

0

20

40

60

80

100

120

0.1 1.0 10.0

Concentration (mM)

Rel

ativ

e ac

tivi

ty (

)

BNPEBNPMNPEMNPNDPENDP

BNPEBNPMNPEMNPNDPENDP

Fig. 5. Inhibition of caspase activity in cell extract by endoperoxides. HepG2

cells were treated with 0.4 mM etoposide for 17 h then extracted with caspase

extraction buffer. The extract containing activated caspases was incubated with

various concentrations of each endoperoxide or its precursor compound at

37 �C for 2 h. Caspase-9 (A) and caspase-3 (B) activities were assayed with

their specific substrates. Relative activities to that of the original extract were

calculated. The means � SD of triplicate assays are shown.

1385K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

hydrophobic compounds than that of HaCaT cells and/or themembrane system, cell organelle, inside HepG2 cells is moredeveloped than that in HaCaT cells. It is obvious that there isa positive correlation between cytotoxicity and the amount ofeach endoperoxide incorporated in cells.

3.3. Effect of singlet oxygen inhibitors on the cytotoxicityof endoperoxides

Investigating the mechanisms of cell death induced bysinglet oxygen in more detail, the effects of singlet oxygeninhibitors on the cytotoxicity of endoperoxide were investi-gated. Two inhibitors, sodium azide (Sigma) and a-tocopherol(Sigma), were selected for this experiment since the former isa hydrophilic and the latter is a hydrophobic compound. Whena-tocopherol was added to the culture media, the viability ofcells after treatment with BNPE increased from 40 to 90% forHepG2 cells and from 38 to 60% for HaCaT cells (Fig. 3).Sodium azide, on the other hand, did not attenuate the

cytotoxic effect of BNPE. In contrast to these results, thecytotoxic effect of MNPE and NDPE was attenuated by boththe inhibitors although there was a tendency for a-tocopheroland sodium azide to be more effective for HepG2 cells andHaCaT cells, respectively.

3.4. Effect of endoperoxide on caspase cascade

Caspase cascade activation in endoperoxide treated cellswas examined to find the mode of cell death. When HepG2cells were treated with endoperoxide neither caspase-9 norcaspase-3 was activated; all the caspase activities in endo-peroxide treated cells were lower than those in cells treatedwith the respective precursor compound, except that of thecaspase-3 activity in MNPE treated cells (Fig. 4A). However,there was a slight but significant level of caspase activation inHaCaT cells after MNPE or NDPE treatment (Fig. 4B). Cas-pase activity in HaCaT cells treated with BNPE were alsohigher than in cells treated with BNP, although the differenceswere not significant. The reason for the opposite effect oncaspase activation in HepG2 compared with HaCaT cells after

M CTR Etop BNP MNP NDP

E P E P E P

Fig. 6. Effect of endoperoxide on DNA ladder formation. HaCaT cells were

incubated with 1.0 mM BNPE/BNP, 12 mM MNPE/MNP, or 45 mM NDPE/

NDP for an initial 2 h, changed to fresh media, and incubated for an additional

15 h. For etoposide treatment, 0.3 mM etoposide (Etop) was present all the

time. DNA was extracted, separated on 2% agarose gel, and stained with

ethidium bromide as described in Section 2. M, 200 bp DNA ladder; CTR,

control. Typical data from three separate experiments are shown.

1386 K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

endoperoxide treatment is not clear, but it should be empha-sized that the magnitude of caspase activation in HaCaT cellsduring typical apoptosis induced by etoposide was muchhigher than in HepG2 cells (Fig. 3).

Otsu et al. (2005) reported that MNPE inhibited the caspaseactivity in a soluble protein fraction extracted from etoposidetreated cells. Inhibition was dose-dependent, with a half-maximal inhibition at w1 mM MNPE. We explored whetherBNPE and NDPE had a similar inhibitory effect on caspaseactivity. A soluble protein fraction containing high caspaseactivity was prepared from etoposide treated HepG2 cells: itdisplayed, along with BNPE or NDPE, a dose-dependentsuppression of caspase-9 and -3 (Fig. 5). The doseeresponsecurves of BNPE and NDPE were essentially the same as thoseof MNPE with a half-maximal inhibition at 1 mM for HepG2cells and 2 mM for HaCaT cells. The data suggest that none ofthe three endoperoxides has a specific affinity for caspases,and singlet oxygen attacks some site(s) away from theenzymes. The precursor compounds had only a slight effect oncaspase activity even at the highest concentrations.

DNA ladder formation is one of the representativephenomena evoked by caspase cascade activation, and is themost common criterion of apoptotic cell death. DNA wasextracted from HaCaT cells treated with endoperoxides. DNAladder formation was much less in cells treated with endo-peroxide than in cells treated with etoposide (Fig. 6), indi-cating that both BNPE and NDPE also induce cell deaththrough an abortive apoptotic pathway, as previously reportedwith MNPE (Otsu et al., 2005).

4. Discussion

Although naphthalene endoperoxide derivatives have beenused in research during the past two decades, only a few reportsdescribe their delivery to cells. Only Klotz et al. (1999) reportedquantitative data on the incorporation of endoperoxides intocells, finding that lipophilic N,N0-di(2,3-dihydroxypropyl)-1,4-naphthalene dipropionate endoperoxide was easily incorporatedinto cells, but NDPE was barely detectable. The data presentedin Table 1 are consistent with those of Klotz et al. (1999). Thus,hydrophobic BNPE was easily incorporated into cells, buthydrophilic NDPE was not. In addition to these, the sequentialextraction procedure with PBS and ethanol that was adoptedgave intimate information about the cellular compartmentwhere each endoperoxide was incorporated. The data show thatendoperoxide had a characteristic pattern in its delivery intocellular compartments, and suggests that the sites and amountsof singlet oxygen generation after the three endoperoxidetreatments were different from one another. Furthermore, sinceBNPE might generate singlet oxygen inside the cellularmembrane, and the site where singlet oxygen initiates thecutaneous signaling events after UVA radiation is the cellmembrane or some molecules closely associated with it (Klotzet al., 2001; Wenk et al., 2004), BNPE should be the mostsuitable naphthalene endoperoxide derivative for mimicking thephysiological effects of singlet oxygen during UVA radiation.

The cytotoxic effect of BNPE was attenuated by a-tocoph-erol but not by sodium azide, and the cytotoxic effects of MNPEand NDPE were attenuated by both inhibitors (Fig. 3). BNPE ismoderately hydrophobic and two-thirds may be incorporatedinto the membrane compartment of cells (Table 1). a-Tocoph-erol is also hydrophobic, and is incorporated into the membranecompartment of cells (Otsu et al., 2005). Therefore, a-tocoph-erol could attenuate the cytotoxic effect of BNPE because thesetwo compounds are supposed to exist in close proximity to oneanother. Since sodium azide is hydrophilic, it can attenuate theeffect of the above two hydrophilic endoperoxides. The fact thata-tocopherol attenuated the cytotoxic effect of hydrophilicendoperoxides suggests that the component(s) inside or theclose proximity of cellular membrane, e.g. cytochrome c (Sutoet al., 2005), are possible primary targets of singlet oxygen andmitochondria could be heavily involved in cell death.

Cell death induced by three naphthalene endoperoxidederivatives was not associated with typical caspase cascadeactivation, as with etoposide treatment (Fig. 4). This isinconsistent with the previous reports that cell death inducedby singlet oxygen is accompanied by caspase activation(Granville et al., 1998; Valencia and Moran, 2004; Ding et al.,2004; Wieder et al., 2006). All these studies had utilizedphotosensitizer as singlet oxygen generators. Free radicals, inaddition to singlet oxygen, are generated when a photo-sensitizing reaction is used (Halliwell and Gutteridge, 1999),and many reports describe cell death induced by free radicalsas being accompanied by caspase activation (e.g. Ryter et al.,2007). Therefore, caspase cascade activation after the photo-sensitizer reaction may not be the result of singlet oxygen, butof free radicals.

1387K. Otsu et al. / Cell Biology International 32 (2008) 1380e1387

Collectively, the amount and compartment of singletoxygen generation in cells treated with the three endoperox-ides seem different from each other. Nevertheless, theyexhibited a similar inhibitory effect on caspase cascade acti-vation during cell death. Thus an abortive apoptotic pathwaydue to the suppression of caspase activation could be a generalfeature of cell death induced by singlet oxygen.

Acknowledgments

We thank Ms. Masako Seki for maintenance of laboratoryequipment and secretarial services.

References

Allia E, Cassoni P, Marrocco T, Volante M, Bussolati B, Wong M, et al.

Oxyphilic and non-oxyphilic thyroid carcinoma cell lines differ in

expressing apoptosis-related genes. J Endocrinol Invest 2003;26:660e7.

Aubry JM, Cazin B, Duprat F. Chemical sources of singlet oxygen 3. Perox-

idation of water-soluble singlet oxygen carriers with the hydrogen per-

oxideemolybdate system. J Org Chem 1989;54:726e8.

Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J, Markham A,

Fusenig NE. Normal keratinization in a spontaneously immortalized

aneuploid human keratinocyte cell line. J Cell Biol 1988;106:761e71.

Diaz A, Munoz-Clares RA, Rangel P, Valdes VJ, Hansberg W. Functional and

structural analysis of catalase oxidized by singlet oxygen. Biochimie 2005;

87:205e14.

Ding X, Xu Q, Liu F, Zhou P, Gu Y, Zeng J, et al. Hematoporphyrin mono-

methyl ether photodynamic damage on HeLa cells by means of reactive

oxygen species production and cytosolic free calcium concentration

elevation. Cancer Lett 2004;216:43e54.

Glaser E, Cadenas E, Andell S, Ernster L. Inhibition of the mitochondrial F1-

ATPase by Rose Bengal mediated photooxidation. Interaction of the Fe2þ

chelate of bathophenanthroline with the sensitizer. Acta Chem Scand B

1988;42:175e82.

Godar DE. UVA1 radiation triggers two different final apoptotic pathways. J

Invest Dermatol 1999;112:3e12.

Granville DJ, Carthy CM, Jiang H, Shore GC, McManus BM, Hunt DW. Rapid

cytochrome c release, activation of caspases 3, 6, 7 and 8 followed by

Bap31 cleavage in HeLa cells treated with photodynamic therapy. FEBS

Lett 1998;437:5e10.

Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine. 3rd ed.

Oxford: Clarendon Press; 1999. chapter 2.5.3.

Ishizawa M, Kobayashi Y, Miyamura T, Matsuura S. Simple procedure of

DNA isolation from human serum. Nucleic Acids Res 1991;19:5792.

Klotz LO, Pellieux C, Briviba K, Pierlot C, Aubry JM, Sies H. Mitogen-

activated protein kinase (p38-, JNK-, ERK-) activation pattern induced by

extracellular and intracellular singlet oxygen and UVA. Eur J Biochem

1999;260:917e22.

Klotz LO, Holbrook NJ, Sies H. UVA and singlet oxygen as inducers of

cutaneous signaling events. Curr Probl Dermatol 2001;29:95e113.

Krutmann J. Ultraviolet A radiation-induced biological effects in human skin:

relevance for photoaging and photodermatosis. J Dermatol Sci 2000;

23(Suppl 1):S22e6.

Liu W, Ogata T, Sato K, Ohba Y, Sakurai K, Igarashi T. Syntheses of water-

soluble endoperoxides as a singlet oxygen source. ITE Lett Batter New

Technol Med 2001;2:98e101.

Moan J, Peng Q. An outline of the hundred-year history of PDT. Anticancer

Res 2003;23:3591e600.

Morita A, Werfel T, Stege H, Ahrens C, Karmann K, Grewe M, et al. Evidence that

singlet oxygen-induced human T helper cell apoptosis is the basic mechanism

of ultraviolet-A radiation phototherapy. J Exp Med 1997;186:1763e8.

Morley N, Rapp A, Dittmar H, Salter L, Gould D, Greulich KO, et al. UVA-

induced apoptosis studied by the new apo/necro-Comet-assay which

distinguishes viable, apoptotic and necrotic cells. Mutagenesis 2006;21:

105e14.

Muller-Breitkreutz K, Mohr H, Briviba K, Sies H. Inactivation of viruses by

chemically and photochemically generated singlet molecular oxygen. J

Photochem Photobiol B 1995;30:63e70.

Nagaoka Y, Otsu K, Okada F, Sato K, Ohba Y, Kotani N, et al. Specific

inactivation of cysteine protease-type cathepsin by singlet oxygen gener-

ated from naphthalene endoperoxides. Biochem Biophys Res Commun

2005;331:215e23.

Nakano M, Kambayashi Y, Tatsuzawa H, Komiyama T, Fujimori K. Useful

1O2 (1delta g) generator, 3-(40-methyl-10-naphthyl)-propionic acid, 10,40-endoperoxide (NEPO), for dioxygenation of squalence (a skin surface

lipid) in an organic solvent and bacterial killing in aqueous medium. FEBS

Lett 1998;432:9e12.

Nieuwint AW, Aubry JM, Arwert F, Kortbeek H, Herzberg S, Joenje H.

Inability of chemically generated singlet oxygen to break the DNA

backbone. Free Radic Res Commun 1985;1:1e9.

Otsu K, Sato K, Ikeda Y, Imai H, Nakagawa Y, Ohba Y, et al. An abortive

apoptotic pathway induced by singlet oxygen is due to the suppression of

caspase activation. Biochem J 2005;389:197e206.

Pass HI. Photodynamic therapy in oncology: mechanisms and clinical use. J

Natl Cancer Inst 1993;85:443e56.

Pierlot C, Aubry JM, Briviba K, Sies H, Di Mascio P. Naphthalene endoper-

oxides as generators of singlet oxygen in biological media. Methods

Enzymol 2000;319:3e20.

Ryter SW, Kim HP, Hoetzel A, Park JW, Nakahira K, Wang X, et al. Mechanisms

of cell death in oxidative stress. Antioxid Redox Signal 2007;9:49e89.

Saito I, Matsuura T, Inoue K. Formation of superoxide ion from singlet oxygen

on the use of a water-soluble singlet oxygen source. J Am Chem Soc 1981;

103:188e90.

Saito I, Matsuura T, Inoue K. Formation of superoxide ion via one-electron

transfer from electron donors to singlet oxygen. J Am Chem Soc 1983;105:

3200e6.

Shahzidi S, Stokke T, Soltani H, Nesland JM, Peng Q. Induction of apoptosis by

hexaminolevulinate-mediated photodynamic therapy in human colon carci-

noma cell line 320DM. J Environ Pathol Toxicol Oncol 2006;25:159e71.

Stennicke HR, Salvesen GS. Biochemical characteristics of caspases-3, -6, -7,

and -8. J Biol Chem 1997;272:25719e23.

Suto D, Sato K, Ohba Y, Yoshimura T, Fujii J. Suppression of the pro-

apoptotic function of cytochrome c by singlet oxygen via a haem redox

state-independent mechanism. Biochem J 2005;392:399e406.

Tatsuzawa H, Maruyama T, Misawa N, Fujimori K, Hori K, Sano Y, et al.

Inactivation of bacterial respiratory chain enzymes by singlet oxygen.

FEBS Lett 1998;439:329e33.

Triesscheijn M, Baas P, Schellens JH, Stewart FA. Photodynamic therapy in

oncology. Oncologist 2006;11:1034e44.

Tyrrell RM. Solar ultraviolet A radiation: an oxidizing skin carcinogen that

activates heme oxygenase-1. Antioxid Redox Signal 2004;6:835e40.

Valencia A, Moran J. Reactive oxygen species induce different cell death

mechanisms in cultured neurons. Free Radic Biol Med 2004;36:1112e25.

Vinnikova AK, Kukreja RC, Hess ML. Singlet oxygen-induced inhibition of

cardiac sarcolemmal NaþKþ-ATPase. J Mol Cell Cardiol 1992;24:465e70.

von Montfort C, Sharov VS, Metzger S, Schoneich C, Sies H, Klotz LO.

Singlet oxygen inactivates protein tyrosine phosphatase-1B by oxidation of

the active site cysteine. Biol Chem 2006;387:1399e404.

Wenk J, Schuller J, Hinrichs C, Syrovets T, Azoitei N, Podda M, et al.

Overexpression of phospholipid-hydroperoxide glutathione peroxidase in

human dermal fibroblasts abrogates UVA irradiation-induced expression of

interstitial collagenase/matrix metalloproteinase-1 by suppression of

phosphatidylcholine hydroperoxide-mediated NFkappaB activation and

interleukin-6 release. J Biol Chem 2004;279:45634e42.

Wieder ME, Hone DC, Cook MJ, Handsley MM, Gavrilovic J, Russell DA.

Intracellular photodynamic therapy with photosensitizer-nanoparticle

conjugates: cancer therapy using a ’Trojan horse’. Photochem Photobiol

Sci 2006;5:727e34.

Zhuang S, Lynch MC, Kochevar IE. Activation of protein kinase C is required

for protection of cells against apoptosis induced by singlet oxygen. FEBS

Lett 1998;437:158e62.