Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO...

7
]CANCER RESEARCH53, 4505-4510, October 1, 1993] Singlet Oxygen: A Primary Effector in the Ultraviolet A/Near-Visible Light Induction of the Human Heme Oxygenase Gene 1 Sharmila Basu-Modak and Rex M. Tyrrell 2 Swiss Institute for Experimental Cancer Research, Physical Carcinogenesis Unit, CH-1066 Epalinges/Lausanne, Switzerland ABSTRACT Both singlet oxygen and the hydroxyl radical are generated in mam- malian ccHs by UVA (~3Z0---~3{50 rim) and possibly near-visible (~{50--420 nm) radiation. We have modulated the cellular levels of these two reactive oxygen species in order to compare their involvement in the induction of the human heme0xygenase (I40) geneby broadspectrum UVA/near- visible light 0LIVA/NVL). Irradiation in deuterium oxide (in which singlet oxygen has a longer half-life) enhances the broad spectrum UVA/NVL induction of this gene. Sodium azide and L-histidine which are scavengers of both singlet oxygen and the hydroxyl radical reduce the fluence-depen- dent accumulation of HO mRNA, while compounds which are only hy- droxyl radical scavengers, namely, mannitol and dimethyl sulfoxide do not. Rose Bengal, a known generator of singlet oxygen, also increases the HO mRNA levels, and this induction is enhanced in deuterium oxide. We conclude that the observed effects of deuterium oxide and singlet oxygen scavengers on HO mRNA levels are not due to a nonspecific effect on transcription but that singlet oxygen is a primary effector in the UVA/NVL induction of the human heme oxygenase gene. INTRODUCTION Heine oxygenase (EC 1.14.99.3), the rate-limiting enzyme in the heme degradation pathway, is ubiquitous in higher eukaryotes. Gen- erally, the enzyme is induced in vivo and in vitro by diverse factors such as heine compounds, therapeutic agents, heat shock, hormones, bacterial toxins, sulfhydryl reagents, metal ions, etc. (1). In human skin fibroblasts, HO 3 is induced by UVA (320-380 nm), a component of sunlight clearly established as a carcinogen in animals (2). The human HO gene is also induced by the oxidizing agent hydrogen peroxide (3, 4) but not by heat shock (3, 5, 6). In rat liver, induction of HO enzyme activity by either heroin, cadmium chloride or bromo- benzene als0 caused an increase in the translatable H0 mRNA (7). This was the first indication that the induction of no gene expression was probably by transcriptional activation. Nuclear run-on transcrip- tion assays in human skin fibroblasts (8) showed that the gene is induced by a transient increase in the transcription rate of HO mRNA, reaching a maximum within 1 h after treatment and then declining steadily to basal levels by 6 h. The initial increase in the rate of HO mRNA transcription is also correlated with an accumulation of HO mRNA to a maximum between 2 and 4 h, followed by a decrease to basal level by 8 h. Thus, the HO gene induction by UVA involves transcriptional activation, but the signal transduction events are not yet well understood. In a previous report (9), it was shown that the presence of iron chelators, o-phenanthroline and desferrioxamine, lowered the H202 induction of both HO mRNA and protein by 50-80%. A similar effect of these iron chelators was observed on the induction of the HO Received 11/4/92; accepted 7/28/93. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by grants from the Swiss League against Cancer and the Swiss National Science Foundation (31.30880.91). 2 To whom reprint requests should be addressed, at Ch. des Boveresses 155, Ch-1066 Epalinges, Switzerland. 3 The abbreviations used are: HO, Heine oxygenase 1; D20, deuterium oxide; DMSO, dimethyl sulfoxide; PBS, phosphate-buffered saline; GAPDH, glyceraldehyde phosphate dehydrogenase; NVL, near-visible light; .OH, hydroxyl radical; 1Oa, singlet oxygen; RB, Rose Bengal. protein by near-monochromatic UVA (365 nm) radiation. From this study, it was concluded that the .OH generated by the iron catalyzed Fenton reaction was involved in the induction of the HO gene. How- ever, iron is also involved in other reactions, such as decomposition of lipid hydroperoxides and autooxidation of a number of biomolecules. Thus, in addition to the .OH, iron-catalyzed reactions could generate a number of other active oxygen species such as the lipid peroxyl radical and singlet oxygen (10). Furthermore, chemical modulation of UVA inactivation of human fibroblasts (11) indicated that 10 2 gen- eration plays a major role in the cytotoxicity of UVA. This prompted us to address the question of the involvement of -OH and 102 in HO gene induction by UVA in more detail. A common approach to detect 102 involvement in reactions is the use of scavengers. A disadvantage of this approach is that some of the commonly used scavengers of ~O2 are also efficient scavengers of the 9 OH as seen by the in vitro rate constants shown in Table 1 (12, 13). Another approach is to replace water by D20 in which 102 has a longer half-life (14). A combination of the two approaches as used in this study will provide complementary information. It has been shown that UVA irradiation of macromolecules in vitro causes the generation of H202 and superoxide anion (15, 16). Fur- thermore, it has been proposed that the .OH could be generated by the iron-catalyzed reduction of H202 by superoxide anion (17), and this was demonstrated in vitro (18). Studies of prokaryotic and eukaryotic cells indicate that the .OH could be generated in vivo by UVA irra- diation (15, 16). On the other hand, a number of reactions capable of generating 102 under conditions relevant in vivo have been proposed (19). It has also been suggested that 102 is involved in the UVA- induced lipid peroxidation of liposomal membranes in vitro (20, 21), and lipid peroxidation by UVA has been demonstrated recently in cultured human fibr0blasts (22). Sinceboth ,OH and 10a are gener- ated in vivo by EVA irradiation, we have modulated the cellular levels of these active oxygen species by various chemical agents. We find that 102 is a primary effector in the UVA/NVL induction of HO mRNA. MATERIALS AND METHODS Cell Culture. The normal human fibroblast cell line FEK4 (23) was cul- tured at 37~ in Earle's modified minimal essential medium supplemented with penicillin, streptomycin, glutamine, sodium bicarbonate, and 15% fetal calf serum. Cells were passaged at weekly intervals by trypsinization and used for experiments between passages 8 and 17. Chemical Treatments and UVA Irradiation. All test agents (100 mM sodium azide, 10 mr~ L-histidine, 0.1 Mmannitol, and 4% DMSO) were dis- solved in PBS. D20 buffer solution was prepared by dissolving 1 PBS tablet (Oxoid, Basingstoke, England) in 100 ml of 100% D20. Cells were seeded (5 105/10-cm dish) in 10 ml of medium 3 days prior to the experiment. Prior to treatment with the test agent and irradiation, the culture medium was removed and kept aside. The cell monolayer was rinsed with PBS, 5 ml of the test agent in PBS (supplemented with Ca2+/Mg 2+, 0.01% each) was added to each dish and incubated at 37~ for 15 min. Control dishes were incubated for the same time in 5 ml of PBS containing Ca2§ 2§ After pretreatment with test agents, cells were irradiated with different fluences (0-1 MJ/m 2) of UVA using a broad spectrum (330-450 nm, with maximum energy between 360 and 420 nm) UVASUN 3000 lamp (Mutzhas, Munich, Germany) at a distance which did not cause heating during irradiation. Although this lamp is consid- 4505 Research. on February 6, 2021. © 1993 American Association for Cancer cancerres.aacrjournals.org Downloaded from

Transcript of Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO...

Page 1: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

]CANCER RESEARCH 53, 4505-4510, October 1, 1993]

Singlet Oxygen: A Pr imary Effector in the Ultraviolet A/Near-Vis ible Light

Induct ion of the H u m a n H e m e Oxygenase Gene 1

Sharmila Basu-Modak and Rex M. Tyrrell 2

Swiss Institute for Experimental Cancer Research, Physical Carcinogenesis Unit, CH-1066 Epalinges/Lausanne, Switzerland

ABSTRACT

Both singlet oxygen and the hydroxyl radical are generated in mam- malian ccHs by UVA (~3Z0---~3{50 rim) and possibly near-visible (~{50--420 nm) radiation. We have modulated the cellular levels of these two reactive oxygen species in order to compare their involvement in the induction of

the human heme 0xygenase (I40) gene by broad spectrum UVA/near- visible light 0LIVA/NVL). Irradiation in deuterium oxide (in which singlet oxygen has a longer half-life) enhances the broad spectrum U V A / N V L

induction of this gene. Sodium azide and L-histidine which are scavengers of both singlet oxygen and the hydroxyl radical reduce the fluence-depen- dent accumulation of HO mRNA, while compounds which are only hy- droxyl radical scavengers, namely, mannitol and dimethyl sulfoxide do not. Rose Bengal, a known generator of singlet oxygen, also increases the HO mRNA levels, and this induction is enhanced in deuterium oxide. We conclude that the observed effects of deuterium oxide and singlet oxygen scavengers on HO mRNA levels are not due to a nonspecific effect on transcription but that singlet oxygen is a primary effector in the UVA/NVL induction of the human heme oxygenase gene.

INTRODUCTION

Heine oxygenase (EC 1.14.99.3), the rate-limiting enzyme in the heme degradation pathway, is ubiquitous in higher eukaryotes. Gen- erally, the enzyme is induced in vivo and in vitro by diverse factors such as heine compounds, therapeutic agents, heat shock, hormones, bacterial toxins, sulfhydryl reagents, metal ions, etc. (1). In human skin fibroblasts, HO 3 is induced by UVA (320-380 nm), a component of sunlight clearly established as a carcinogen in animals (2). The human HO gene is also induced by the oxidizing agent hydrogen peroxide (3, 4) but not by heat shock (3, 5, 6). In rat liver, induction of HO enzyme activity by either heroin, cadmium chloride or bromo-

benzene als0 caused an increase in the translatable H0 mRNA (7). T h i s w a s the f i rs t i n d i c a t i o n tha t the i n d u c t i o n o f n o g e n e e x p r e s s i o n

was probably by transcriptional activation. Nuclear run-on transcrip- tion assays in human skin fibroblasts (8) showed that the gene is induced by a transient increase in the transcription rate of HO mRNA, reaching a maximum within 1 h after treatment and then declining steadily to basal levels by 6 h. The initial increase in the rate of HO mRNA transcription is also correlated with an accumulation of HO mRNA to a maximum between 2 and 4 h, followed by a decrease to basal level by 8 h. Thus, the HO gene induction by UVA involves transcriptional activation, but the signal transduction events are not yet well understood.

In a previous report (9), it was shown that the presence of iron chelators, o-phenanthroline and desferrioxamine, lowered the H202 induction of both HO mRNA and protein by 50-80%. A similar effect of these iron chelators was observed on the induction of the HO

Received 11/4/92; accepted 7/28/93. The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by grants from the Swiss League against Cancer and the Swiss National Science Foundation (31.30880.91).

2 To whom reprint requests should be addressed, at Ch. des Boveresses 155, Ch-1066 Epalinges, Switzerland.

3 The abbreviations used are: HO, Heine oxygenase 1; D20, deuterium oxide; DMSO, dimethyl sulfoxide; PBS, phosphate-buffered saline; GAPDH, glyceraldehyde phosphate dehydrogenase; NVL, near-visible light; .OH, hydroxyl radical; 1Oa, singlet oxygen; RB, Rose Bengal.

protein by near-monochromatic UVA (365 nm) radiation. From this

study, it was concluded that the .OH generated by the iron catalyzed Fenton reaction was involved in the induction of the HO gene. How- ever, iron is also involved in other reactions, such as decomposition of lipid hydroperoxides and autooxidation of a number of biomolecules. Thus, in addition to the .OH, iron-catalyzed reactions could generate

a number of other active oxygen species such as the lipid peroxyl radical and singlet oxygen (10). Furthermore, chemical modulation of UVA inactivation of human fibroblasts (11) indicated that 10 2 gen- eration plays a major role in the cytotoxicity of UVA. This prompted us to address the question of the involvement of -OH and 102 in HO

gene induction by UVA in more detail. A common approach to detect 102 involvement in reactions is the

use of scavengers. A disadvantage of this approach is that some of the commonly used scavengers of ~O2 are also efficient scavengers of the �9 OH as seen by the in vitro rate constants shown in Table 1 (12, 13). Another approach is to replace water by D20 in which 102 has a longer half-life (14). A combination of the two approaches as used in this study will provide complementary information.

It has been shown that UVA irradiation of macromolecules in vitro

causes the generation of H202 and superoxide anion (15, 16). Fur- thermore, it has been proposed that the .OH could be generated by the iron-catalyzed reduction of H202 by superoxide anion (17), and this was demonstrated in vitro (18). Studies of prokaryotic and eukaryotic cells indicate that the .OH could be generated in vivo by UVA irra- diation (15, 16). On the other hand, a number of reactions capable of generating 102 under conditions relevant in vivo have been proposed (19). It has also been suggested that 102 is involved in the UVA- induced lipid peroxidation of liposomal membranes in vitro (20, 21), and lipid peroxidation by UVA has been demonstrated recently in

cultured human fibr0blasts (22). Since both ,OH and 10a are gener- ated in v ivo b y E V A i r r ad i a t ion , w e h a v e m o d u l a t e d the c e l l u l a r l e v e l s

of these active oxygen species by various chemical agents. We find that 102 is a primary effector in the UVA/NVL induction of HO mRNA.

MATERIALS AND M E T H O D S

Cell Culture. The normal human fibroblast cell line FEK4 (23) was cul- tured at 37~ in Earle's modified minimal essential medium supplemented with penicillin, streptomycin, glutamine, sodium bicarbonate, and 15% fetal calf serum. Cells were passaged at weekly intervals by trypsinization and used for experiments between passages 8 and 17.

Chemical Treatments and UVA Irradiation. All test agents (100 mM sodium azide, 10 mr~ L-histidine, 0.1 M mannitol, and 4% DMSO) were dis- solved in PBS. D20 buffer solution was prepared by dissolving 1 PBS tablet (Oxoid, Basingstoke, England) in 100 ml of 100% D20. Cells were seeded (5 • 105/10-cm dish) in 10 ml of medium 3 days prior to the experiment. Prior to treatment with the test agent and irradiation, the culture medium was removed and kept aside. The cell monolayer was rinsed with PBS, 5 ml of the test agent in PBS (supplemented with Ca2+/Mg 2+, 0.01% each) was added to each dish and incubated at 37~ for 15 min. Control dishes were incubated for the same time in 5 ml of PBS containing Ca2§ 2§ After pretreatment with test agents, cells were irradiated with different fluences (0-1 MJ/m 2) of UVA using a broad spectrum (330-450 nm, with maximum energy between 360 and 420 nm) UVASUN 3000 lamp (Mutzhas, Munich, Germany) at a distance which did not cause heating during irradiation. Although this lamp is consid-

4505

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 2: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE

Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical (12, 13)

Rate constant (M-is -1)

Test agent 10 2 .OH

Azide 2.2 x 10 9 1.1 x 101~

(pn 9.0) L-Histidine 3.2 • 10 7 5 x 10 9

(pn 7.1) (pn 6-7) DMSO 3.4 >( 10 4 7.1 x 10 9

(pH 7.0) Mannitol <103 2.7 X 109

(pH 7,0)

ered primarily as a UVA-emitting lamp, we cannot exclude the possibility that the near-visible (380-420 nm) component contributes to the effects observed; hence, the term UVA/NVL is used for the radiation emitted by this lamp. After irradiation, the PBS containing test agent was aspirated, cells were washed with fresh PBS, the original medium was added back, and the cells were incubated for 3 h at 37~ prior to the extraction of total RNA.

RNA Isolation and Analysis. Total cellular RNA was isolated by the acid guanidinium thiocyanate-phenol-chloroform extraction method (24), electro- phoresed (15 ~g/well) in a 2.2 M formaldehyde-l.3% agarose gel (25) and transferred onto a Gene Screen nylon membrane (NEN Research Products) by capillary blotting (conditions proposed by the manufacturer were used for RNA transfer and hybridization). Baked blots were hybridized to the larger (1 kilobase) EcoRI fragment of a fun-length HO cDNA clone [clone 2/10; (4)] which was 32p labeled by the random-priming DNA-labeling method (26). After autoradiography, the HO probe was stripped off by boiling in 0.1• standard sodium citrate (15 mM sodium chloride, 1.5 mM sodium citrate) and 0.1% sodium dodecyl sulfate (twice for 20 min each time), and the blots were reprobed with the 32p-labeled PstI fragment (1300 base pairs) of rat GAPDH cDNA. The GAPDH-RNA signal was used as an internal control for the loading error between samples since the constitutive levels of this RNA are unaffected by the test agents used.

For autoradiography, films were preflashed, and blots were exposed at -70~ Radioactive signals were quantified by densitometry on a Elscript 400 (Hirschmann) densitometer. The HO mRNA signals (corrected for the loading error with GAPDH) was expressed as a relative (fold) increase above basal level and plotted as a function of fluence.

The ratio of the corrected HO mRNA signal area of treated sample and the corresponding control was also calculated for each sample, and mean values of 3-5 experiments for each test agent were plotted as a function of fiuence. These plots represent the general pattern of the fluence-dependent effect of the test agents.

RNA Synthesis in Cells. Cells were cultured in 12-well tissue culture dishes (104 cells/2 ml/well) for 24 h and labeled with [14C]thymidine (10 nCi/wen of 57 mCi/mmol specific activity) for another 24 h. After the cells were treated with test agent in PBS (1 ml/well) and UVA/NVL as described above, 1.5 ml of the original culture medium supplemented with 3 p.Ci of [3H]uridine (specific activity, 46 Ci/mmol) was added back to each well and incubated for 3 h at 37~ Cells were lysed in 100 p.1 of lysis buffer [2% sodium dodecyl sulfate-0.1 mg/ml bovine serum albumin-20 mM 4-(2-hy- droxyethyl)-l-piperazineethanesulfonic acid, pH 7.3] and spotted onto GF/C discs prewetted with 1 N HC1 (27). Discs were washed with 100% ethanol and air dried, and radioactivity was determined in a liquid scintillation counter. The ratio of 3I"I/14C cpm was used as a measure of RNA synthesis.

Modulation of HO mRNA Levels by a Photosensitizer. A stock solution of Rose Bengal (1 mM), a known generator of 10~, was prepared in DMSO. Cells were seeded in 10-cm dishes and prepared for chemical treatment as described above. For chemical treatment, 5 ml of PBS (supplemented with Ca2+/Mg 2+) containing 1 /xM RB was added to each dish, and cells were incubated in the dark for 30 min at 37~ Control cells were incubated simi- larly in Ca2+/Mg2+-PBS without the photosensitizer. After the pretreatment, cells were irradiated with a range of fluences (0-1500 J/m 2) of broad spectrum (400-700 nm) visible light by placing the culture dishes on a fluorescent light box. After irradiation, the PBS-containing RB was aspirated, the monolayer was washed with fresh PBS, and the original medium was added back to incubate for 3 h at 37~ prior to RNA extraction.

R E S U L T S

M o d u l a t i o n o f U V A / N V L I n d u c t i o n o f H e m e O x y g e n a s e m R N A

by D e u t e r i u m Oxide . UVA/NVL irradiation of human skin fibro-

blasts induces accumulat ion of HO m R N A (Fig. 1, A and B). Since

both "OH and 102 are generated in UVA-irradiated cells, we studied

the f luence-dependent accumulat ion o f HO m R N A in cells irradiated

with UVA/NVL in the presence of D20, a solvent in which 10 2 is

known (14) to have a longer half-life. Cells were incubated for 15 min

in buffer prepared with 100k D20 prior to irradiation with UVA/NVL f luences ran~in~ f rom 0--0.5 MJ/m =. We find (Fi~. 1~ A and B) that in

control cells irradiated with UVA/NVL the HO m R N A signal intensity

increases steadily to 17-fold above basal level at the m a x i m u m fluence

used. While t reatment with D 2 0 alone does not change the basal level

of HO m R N A significantly, the f luence-dependent accumulat ion of

this message is enhanced (Fig. 1, A and B). In D20- t rea ted cells, the

HO m R N A levels first increase to a m a x i m u m of 24-fold at 0.2

MJ/m 2, and at higher f luences these levels decrease rapidly, reaching

basal levels at 0.5 MJ/m: . Ratios of the HO m R N A signal intensities

in DeO-treated and control cells (Fig. 1C) show the general pattern of

the f luence-dependent DzO effect. We find that the m a x i m u m en-

hancement of the HO m R N A levels is at 0.1 MJ/m 2. At higher flu-

ences, the effect declines rapidly, and at 0.5 MJ/m 2 the HO m R N A

levels in DzO-treated cells are only 10% of the control levels.

A

MJm .2 D20

HO

0 0.1 0.2 0.3 0.5 " - 4- " ' - 4- " " 4- " ' - 4- " " 4. "

~ B 30

~ e m Control ---O-.--- D20

.~e ~0-

0 " C s

E o o ~ o~

~ 0 �9 ~ �9 0.0 0.1 0.2 0.3 0.4 0.5 0.6

FIuence ( MJm "2 ) Fig. 1. Effect of 020 on the HO mRNA accumulation induced by UVA/near-visible

light. Human skin fibroblasts were irradiated with increasing fluences of UVA/NVL in the presence (+) or absence (-) of 100% DzO. A, Northern blot of total cellular RNA was first probed for HO mRNA and then for GAPDH mRNA. B, HO and GAPDH mRNA signals were quantified by densitometry. The GAPDH signal was used as an internal loading control. The HO mRNA signal, corrected with GAPDH for difference in loading between samples, was expressed as a relative increase above basal level and plotted as a function of UVA/NVL fluence. C, ratio of the HO mRNA signals (corrected for loading error) in D20-treated and control cells was calculated for each sample. A mean of 3-5 experiments was plotted as a function of UVA/NVL fluence. Point (bar), mean (-+SD).

4506

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 3: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

SINGLET OXYGEN INDUCTION OF THE HUMAN HO I GENE

These results indicate that it is primarily the 10 2 generated by UVA/NVL irradiation which is involved in the induction of the HO

gene. Effect of Sodium Azide and L-Histidine. To further test the in-

volvement of 102, we studied the effect of sodium azide (28) and L-histidine (29) which are known to scavenge 102 almost as effec- tively as the .OH (Table 1). Monolayer cultures were preincubated with 100 mM sodium azide or 10 mM L-histidine for 15 min and then irradiated with a similar range of UVA/NVL fluences in the presence of the scavenger. The fluence-dependent increase in HO mRNA levels in cells treated with these agents is shown in Fig. 2, A, B, D, and E. These scavengers alone have little effect on the basal level of HO mRNA. In cells treated with sodium azide, the fluence-dependent HO message induction lags initially and reaches corresponding control levels only at 0.4 MJ/m 2 (Fig. 2B). L-Histidine also suppresses the fluence-dependent accumulation of HO mRNA (Fig. 2E). The ratios of HO mRNA signal intensity between cells treated with 102 scav- enger and control cells (Fig. 2, C and F) indicate that in the presence of sodium azide the HO mRNA levels are reduced to 40% of the corresponding control at 0.2 MJ/m 2 (Fig. 2C). In histidine-treated cells (Fig. 2F), the HO mRNA levels decrease to 50% of the controls at 0.3 MJ/m 2. At higher fluences, the HO mRNA levels in cells treated with either scavenger approach that in the corresponding controls.

While these results support those obtained in D20, the possible involvement of the .OH as a primary effector needed further experi- mental consideration since the 10 z scavengers mentioned above are also excellent scavengers of the .OH as predicted by their in vitro rate constants (Table 1).

Hydroxyl Radical Scavengers. In order to eliminate the possibil- ity that the observed effect of sodium azide or L-histidine on HO

induction was really due to scavenging of the -OH radical, we used DMSO (30) and mannitol which scavenge the .OH far more specifi- cally because of their low rate constants for reaction with 102 (Table 1). We find that these agents themselves have little effect on the basal level of HO mRNA, and the fluence-dependent increase of this mes- sage in cells treated with either 4% DMSO or 0.1 M mannitol is similar to that observed in the corresponding control cells (Fig. 3, A, B, D, and E). Plots of fluence-dependent change in ratios (Fig. 3, C and F) indicate that the HO mRNA levels in the .OH scavenger-treated cells are either almost equal to the corresponding controls (in mannitol) or slightly higher (in DMSO) than the levels in the corresponding un- treated cells.

Thus, scavengers of the .OH do not decrease the fluence-dependent accumulation of HO mRNA, and at least in cells treated with DMSO cause a slight enhancement of the HO mRNA levels.

Effect of Test Agents on Total RNA Synthesis. In order to deter- mine whether the modulation of HO mRNA levels by the various test agents is due to a general effect on transcription, we measured the total [3H]uridine incorporation during 3 h in cells which had been pre- treated with D20, sodium azide, or DMSO and then irradiated with increasing fluences of UVA/NVL in the presence of these agents. In control cells, the RNA synthesis is inhibited with increasing fluences and reaches almost zero at 1 MJ/m 2 (Fig. 4A). In cells treated with D20, there is further suppression of transcriptional activity at fluences above 0.25 MJ/m 2. In contrast, sodium azide seems to protect against the toxic effects of UVA on transcription, as seen by the higher incorporation of [3H]uridine in cells irradiated in the presence of azide (Fig. 4B). Finally, the .OH scavenger DMSO does not have a signifi- cant effect on the fluence-dependent decrease in transcription (Fig. 4C).

Fig. 2. Effect of sodium azide (NaN3) and L-histidine on UVA/NVL-induced HO mRNA accumulation. Northern blot analysis of HO and GAPDH mRNA in the presence (+) or absence ( - ) of either 100 mM sodium azide (A) or 10 mM L-histidine (D). B and E, HO mRNA signals quantified and corrected for loading error and expressed as a relative increase over basal levels. C and F, ratios of the HO signal intensity in scavenger-treated and control cells plotted as a function of UVA/NVL fluence. Point (bar), mean (•

4507

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 4: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE

Fig. 3. UVA/NVL-induced HO mRNA accumulation in the presence of hydroxyl radical scavengers. Northern blot analysis of HO and GAPDH mRNA in the presence (+) and absence (-) of 4% DMSO (A) and 0.1 M mannitol (D). B and E, relative increase in HO mRNA accumulation quantified by densitometry of autoradiographs. C and F, ratios of the signal intensities of HO mRNA (corrected for loading error with GAPDH) in hydroxyl radical scavenger-treated and control cells. Point (bar), mean (_+SD).

These results indicate that the enhancement of HO mRNA accu- mulation in D20 and suppression of this accumulation in sodium azide are not due to a general effect of these agents on transcription.

Modulation of HO mRNA Levels by Rose Bengal. It is known (31) that the photosensitizer RB generates 102 when illuminated with visible light. Therefore, we studied the effect of this photosensitizer on HO mRNA accumulation. We find that irradiation with a range of fluences of visible light alone does not induce HO (Fig. 5, A and C). In cells treated with 1 /XM RB, the HO mRNA levels increase to 29-fold above basal levels at the maximum fluence of visible light used. In the presence of D20, the RB/visible light-induced increase in HO mRNA levels is enhanced; the HO message levels increase to 82-fold above basal level at 1 kJ/m 2 (Fig. 5, B and C). At higher fluences, the HO mRNA levels decline rapidly to basal levels. Thus, the enhancement of HO mRNA accumulation by RB/visible light in the presence of D20 is similar to that observed in the UVA/NVL- induced accumulation of this message.

These results are entirely consistent with the conclusion that it is the 102 which is a primary effector in the UVA/NVL induction of the human heme oxygenase gene.

DISCUSSION

In the present study, we have compared the relative involvement of 102 and .OH in the UVA/NVL induction of the HO gene by chemi- cally modulating the cellular concentration of these active oxygen species, and we find that 102 is a primary effector in this induction. An accompanying observation is that 102 also appears to be the active intermediate involved in the UVA/NVL fluence-dependent inhibition of overall transcription. Since it was not possible to measure effective intracellular concentrations of the test agents used, we used the opti-

mum concentrations as determined in a previous study (11) on radical involvement in UVA inactivation of human fibroblasts.

The UVA/NVL fluence-dependent increase in HO mRNA levels is initially enhanced in DEO (Fig. 1). Although the enhancement of lifetime of 102 in vitro by D20 is by a factor of 10-15, this is not reflected in the enhancement of HO mRNA levels in vivo presumably due to the interaction of this active oxygen intermediate with biomol- ecules. Furthermore, deuteration of biological antioxidants could con- tribute to lowering the threshold of UVA/NVL fluence required to induce HO gene in the presence of DzO. At fluences higher than 0.2 MJ/m z, the HO mRNA levels decrease rapidly (Fig. 1, A and B) probably due to a further suppression of the UVA/NVL fluence- dependent decrease in total RNA synthesis in the presence of DzO (Fig. 4A). We have observed in untreated cells (data not shown) that at fluences above 0.5 MJ/m 2, the HO mRNA levels also decline, indicating that severe oxidant stress shuts off transcription of this gene, probably because of a general inhibitory effect on total RNA synthesis in cells. In the presence of sodium azide and L-histidine, the fluence-dependent induction of HO mRNA is significantly (50--60%) lower than the controls, and this is not due to an inhibitory effect on transcription (Figs. 2 and 4B). Thus, in the fluence range in which D20 has no effect on total RNA synthesis, it enhances the UVA/NVL induction of HO mRNA. On the other hand, sodium azide protects to some extent against the the toxic effects of UVA/NVL on total RNA synthesis in cells but decreases the UVA/NVL fluence-dependent HO mRNA accumulation up to fluences of 0.3 MJ/m 2. At higher fluences, these agents are no longer effective, probably because of the genera- tion of radicals by the scavengers themselves. Since sodium azide and L-histidine scavenge both the .OH and 1 0 2 with equal efficiency as predicted by their in vitro rate constants (Table 1), our results could be

4508

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 5: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

S I N G L E T O X Y G E N I N D U C T I O N O F T H E H U M A N H O 1 G E N E

9.0 i T -----On Control

3.06"0 ~ m D20

A " t ' ~ . ' ~ 0.0 c 9.0 T T ~ Control

E �9 NaN 3 t- i- 6.0 O -

e 3.0 .E "o 'r.. B

0.0 . . . . . 9.O 1 ~ Control

6.0 ~

3.0

c 0.0 . . . . . . . . . . .

0 0.2 0.4 0.6 0.8 1.0

Fluence ( MJm -2)

Fig. 4. Cells were labeled with [14C]thymidine for 24 h, followed by incubation in 100% D20, 100 mM sodium azide, or 4% DMSO for 15 min at 37~ first and then irradiation with a range (0-1 MJ/m 2) of UVA/NVL fluences prior to incubation with [3H]uridine for 3 h to measure total RNA synthesis. Because the [14C]thymidine cpm (used here as an indicator of cell number) was relatively consistent in all samples, the ratio of 3H/14C cpm was used as a measure of total RNA synthesis. Ratios of [3H]uridine/ [14C]thymidine cpm are plotted as a function of UVA/NVL fluence for cells treated with D20 (A), sodium azide (B), and DMSO (C). Point (bar), mean (•

kJm -2

Rose Bengal

HO

GAPDH

0 f ~

0.5 1.0 1.5 II 71 I

- + + - +

.o [, GAPDH I--~

A

- I m - [ __,____..,...] B

Control B e ~ Rose Bengal - - - ~ D20- control m A ~ D20. Rose Bengal

~ 80 *

~ 60

~ ~176 1 ._c Z

I~ 40 ._>

a:._c

| �9 i �9 | , |

0 500 1000 1500 Fluence (Jm "2)

C

Fig. 5. Modulation of HO mRNA levels by Rose Bengal and visible light. HO mRNA levels were induced in cells by Rose Bengal and visible light. A, Northern blots of total cellular RNA from cells treated with increasing fluences of visible light in the presence (+) and absence (-) of 1 p.M RB probed for HO and GAPDH. B, Northern blot analysis of total RNA from cells induced with RB/visible light in the presence of D20. C, autoradiographs quantified by densitometry and the relative increases in HO mRNA levels expressed as a function of fluence.

due to scavenging of either one or both of these species. However, DMSO and mannitol which preferentially scavenge the .OH (Table 1) have little or no effect on either the fluence-dependent increase in HO mRNA levels or total RNA synthesis in cells (Figs. 3 and 4C). The possibility exists that the -OH scavengers do not reach the target molecules within the cell. However, for UVA/NVL induction of the HO gene, the cell membrane is a more likely site for triggering a cascade of signal transduction events since UVB and UVC radiations (3, 32), which cause more DNA damage than UVA radiation, do not induce this gene. Any .OH generated by UVA/NVL at or close to cell membranes would be expected to be easily accessible to scavengers. Use of RB (which is known to generate 102) to modulate HO mRNA levels supports the idea of 10 2 involvement. Interpretation of results obtained using RB is complicated by the fact that it can generate radicals (type I reactions) in addition to 102 (type II reactions). To avoid this problem, we have also induced HO with RB/visible light in the presence of D20 in which type I reactions are presumably un- affected (33). The D20 enhancement of the increase in HO mRNA levels by RB/visible light confirms the effectiveness of 102 in the induction of the HO gene. Furthermore, in the presence of L-histidine, which has little reactivity toward the superoxide anion (28), there is a decrease in the HO mRNA levels induced by RB/visible light (data not shown). Taken together, these results are consistent with the conclu- sion that it is the 102 which is the primary effector in the UVA/NVL induction of this gene in human skin fibroblasts.

An indication that a crucial early step in HO gene induction could be a membrane event is that porphyrins, which constitute an important

class of endogenous photodynamic sensitizers, cause oxidative stress by generating 102 (34) and are present in cell membranes. Moreover, it has been demonstrated in Propionibacterium acnes (35) that the maximum sensitivity to blue light (380-440 nm) was in the region of 415 nm, which corresponded to the absorption maxima of porphyrins in cells. A porphyrin involvement has also been suggested (36) in near UV (300-400 nm) inactivation events in Escherichia coll. It has been shown in bacteria that RB accumulates in the cell membrane (37) and that the photodynamic inactivation of bacteria by this dye is similar to that caused by use of chemically pure 102 on bacterial cells (38). Thus, the RB-mediated photodynamic inactivation seems to be a membrane event in bacteria, probably because of the generation of 102. If we assume that RB also accumulates in the cell membrane of mammalian cells, the HO induction in the presence of this dye and visible light could be interpreted to be a membrane event. There is ample evidence that UVA can cause lipid peroxidation of cell mem- branes in human skin fibroblasts (22) and, based on experiments with a wide variety of scavengers, this process is believed to be mediated by 10 2 (20, 21).

Thus, we propose that 102 may be involved as a crucial interme- diate at an early step in the signal transduction events that lead to the UVA/NVL induction of the human heme oxygenase gene.

A C K N O W L E D G M E N T S

We thank Drs. Glenn Vile and Sohan Modak for useful discussions during the course of this work.

4509

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 6: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

SINGLET OXYGEN INDUCTION OF THE HUMAN ttO 1 GENE

R E F E R E N C E S

1. Maines, M. D. Heine oxygenase: function, multiplicity, regulatory mechanisms, and clinical applications. FASEB J., 2: 2557-2568, 1988.

2. Sterenborg, H. J. C. M., and van der Leun, J. C. Tumorigenesis by long wavelength UVA. In: E Urbach (ed.), Biological Responses to Ultraviolet A Radiation, pp. 295-300. Kansas City, KS: Valdenmar Publishing Company, 1992.

3. Keyse, S. M., and Tyrrell, R. M. Both near ultraviolet radiation and the oxidizing agent hydrogen peroxide induce a 32-kDa stress protein in normal human skin fibroblasts. J. Biol. Chem., 262: 14821-14825, 1987.

4. Keyse, S. M., and Tyrrell, R. M. Heme oxygenase is the major 32-kDa stress protein induced in human skin fibroblasts by UVA radiation, hydrogen peroxide and sodium arsenite. Proc. Natl. Acad. Sci. USA, 86: 99-103, 1989.

5. Taketani, S., Kohno, H., Yoshinaga, T., and Tokunaga, R. The human 32-kDa stress protein induced by exposure to arsenite and cadmium ions is heine oxygenase. FEBS Lett., 245: 173-176, 1989.

6. Yoshida, T., Biro, P., Cohen, T., Muller, R. M., and Shibahara, S. Human heme oxygenase cDNA and induction of its mRNA by hemin. Eur. J. Biochem., 171: 457-461, 1988.

7. Ishizawa, S., Yoshida, T., and Klkuchi, G. Induction of heme oxygenase in rat liver. J. Biol. Chem., 258: 4220--4225, 1983.

8. Keyse, S. M., Applegate, L. A., Tromvoukis, Y., and TyrreU, R. M. Oxidant stress leads to transcriptional activation of the human heme oxygenase gene in cultured skin fibroblasts. Mol. Cell. Biol,, 10: 4967-4969, 1990.

9. Keyse, S. M., and Tyrrell, R. M. Induction of the heine oxygenase gene in human skin fibroblasts by hydrogen peroxide and UVA (365 nm) radiation: evidence for the involvement of the hydroxyl radical. Carcinogenesis (lond.), 11: 787-791, 1989.

10. Koga, S., Nakano, M., and Uehara, K. Mechanism for the generation of superoxide anion and singlet oxygen during heme compound-catalysed linoleic acid hydroper- oxide decomposition. Arch. Biochem. Biophys., 289: 223-229, 1991.

11. Tyrrell, R. M., and Pidoux, M. Singlet oxygen involvement in the inactivation of cultured human fibroblasts by UVA (334 nm, 365 rim) and near visible (405 nm) radiations. Photochem. Photobiol., 49: 407-412, 1989.

12. Farhataziz, and Ross, A. B. Selected specific rates of reactions of transients from water in aqueous solution. III. Hydroxyl radical and perhydroxyl radical and their radical ions. Natl. Stand. Ref. Data Ser. (U. S. Natl. Bur. Stand.), 59: 1-113, 1977.

13. Wilkinson, E, and Brummer, J. Rate constants for the decay and reactions of the lowest electronically excited singlet state of molecular oxygen in solution. J. Phys. Chem. Ref. Data, 10: 809-999, 1981.

14. Lindig, B. A., Rodgers, M. A. J., and Schaap, A. P. Determination of the lifetime of singlet oxygen in D20 using 9,10-Anthracenedipropionic acid, a water-soluble probe. J. Am. Chem. Soc., 102: 5590-5593, 1980.

15. Tyrrell, R. M. UVA (320-380 nm) radiation as an oxidative stress. In: H. Sies (ed.), Oxidative Stress: Oxidants and Antioxidants, pp. 57-83. London: Academic Press, 1991.

16. Tyrrell, R. M., Keyse, S. M., and Moraes, E. C. Cellular defence against UVA (320-380 nm) and UVB (290-320 nm) radiations. In: E. Riklis (ed.), Photobiology, pp. 861-871. New York: Plenum Press, 1991.

17. McCord, J. M., and Day, E. D. Superoxide-dependent production of hydroxyl radical catalysed by iron-EDTA complex. FEBS Lett., 86: 139-142, 1978.

18. Beauchamp, C., and Fridovich, I. A mechanism for the production of ethylene from methional. The generation of the hydroxyl radical by xanthine oxidase. J. Biol. Chem., 245: 4641--4646, 1970.

19. Badwey, J. A., and Karnovsky, M. L. Active oxygen species and the functions of phagocytic leukocytes. Annu. Rev. Biochem., 49: 695-726, 1980.

20. Bose, B., Agarwal, S., and Chatterjee, S. N. UV-A induced lipid peroxidation in liposomal membrane. Radiat. Environ. Biophys., 28: 59-65, 1989.

21. Bose, B., Agarwal, S., and Chatterjee, S. N. Membrane lipid peroxidation by UV-A: mechanism and implications. Biotechnol. Appl. Biochem., 12: 557-561, 1990.

22. Morli6re, P., Moysan, A., Santus, R., Huppe, G., Mazi6re, J-C., and Dubertret, L. UVA-induced lipid peroxidation in cultured human fibroblasts. Biochim. Biophys. Acta, 1084: 261-268, 1991.

23. Tyrrell, R. M., and Pidoux, M. Quantitative differences in host cell reactivation of ultraviolet-damaged virus in human skin fibroblasts and epidermal keratinocytes cultured from the same foreskin biopsy. Cancer Res., 46: 2665-2669, 1986.

24. Chomczynski, P., and Sacchi, N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162: 156-- 159, 1987.

25. Davis, L. G., Dibner, M. D., and Battey, J. E Basic Methods in Molecular Biology, p. 143. New York: Elsevier, 1986.

26. Feinberg, A. P., and Vogelstein, B. A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem., 137: 266-267, 1984.

27. Lechner, J. E, McClendon, I. A., LaVeck, M. A., Shamsuddin, A. M., and Harris, C. H. Differential control by platelet factors of squamous differentiation in normal and malignant human bronchial epithelial cells. Cancer Res., 43: 5915-5921, 1983.

28. Singh, A. Chemical and biochemical aspects of superoxide radicals and related species of activated oxygen. Can. J. Physiol. Pharmacol., 60: 1330-1345, 1982.

29. Dahl, T. A., Midden, W. R., and Hartman, P. E. Some prevalent biomolecules as defences against oxygen damage. Photochem. Photobiol., 47: 357-362, 1988.

30. Klein, S. M., Cohen, G., and Cederbaum, A. I. Production of formaldehyde during metabolism of dimethyl sulfoxide by hydroxyl radical generating systems. Biochem- istry, 20: 6006-6012, 1981.

31. Midden, W. R., and Wang, S. Y. Singlet oxygen generation for solution kinetics: clean and simple. J. Am. Chem. Soc., 105: 4129-4135, 1983.

32. Applegate, L. A., Luscher, P., and Tyrrell, R. M. Induction of heine oxygenase: a general response to oxidant stress in cultured mammalian cells. Cancer Res., 51: 974-978, 1991.

33. Halliwell, B., and Gutteridge, J. M. C. Free Radicals in Biology and Medicine. Oxford, UK: Clarendon Press, 1989.

34. Tyrrell, R. M., and Keyse, S. M. The interaction of UVA radiation with cultured cells. J. Photochem. Photobiol., 4: 349-361, 1990.

35. Kjeldstad, B., and Johnsson, A. An action spectrum for blue and near ultraviolet inactivation of Propionibacterium acnes; with emphasis on a possible porphyrin photosensitization. Photochem. Photobiol., 43: 67-70, 1986.

36. Tuveson, R. W., and Sammartano, L. J. Sensitivity of HemA mutant Escherichia coli ceils to inactivation by near-UV light depends on the level of supplementation with d=aminolevulinic acid. Photochem. Photobiol., 43: 621--626, 1986.

37. Dahl, T. A., Valdes-Aguilera, O., Midden, W. R., and Neckers, D. C. Partition of Rose Bengal anion from aqueous medium into a lipophilic environment in the cell envelope of Salmonella typhimurium: implications for cell-type targeting in photodynamic therapy. J. Photochem. Photobiol. B, 4: 171-184, 1989.

38. Dahl, T. A., Midden, W. R., and Neckers, D. C. Comparison of photodynamic action by rose bengal in Gram-positive and Gram-negative bacteria. Photochem. Photobiol., 48: 607-612, 1988.

4510

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 7: Singlet Oxygen: A Primary Effector in the Ultraviolet A ...SINGLET OXYGEN INDUCTION OF THE HUMAN HO 1 GENE Table 1 Rate constants for reaction of singlet oxygen and the hydroxyl radical

1993;53:4505-4510. Cancer Res   Sharmila Basu-Modak and Rex M. Tyrrell  Gene

OxygenaseA/Near-Visible Light Induction of the Human Heme Singlet Oxygen: A Primary Effector in the Ultraviolet

  Updated version

  http://cancerres.aacrjournals.org/content/53/19/4505

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/53/19/4505To request permission to re-use all or part of this article, use this link

Research. on February 6, 2021. © 1993 American Association for Cancercancerres.aacrjournals.org Downloaded from