Document aBGedwqo5Xp75yZJ2ywEEa0xN

I- -J "r 1433, Environmental and Molecular Mutagenesis 21:339-348 (1993) \ Benzene Metabolite, 1,2,4=Benzenetriol, Induces Micronuclei and Oxidative DNA Damage in Human lymphocytes and HL60 Cells Luoping Zhang, Moire 1. Robertson, Prema Kolachana, Allan J. Davison, and Martyn T. Smith Department of Biomedical and Environmental Health Sciences, School of Public Health, University o f California, Berkeley (M.L.R., P.K., M.T.S.), and Bioenergetics Research Lab, Faculty of Applied Science, Simon Fraser University, Burnaby, B.C., Canada (L.Z., A.J.D.) The triphenolic metobolite of benzene, 1,2,4- respectively. A linear dase-related increase in benzenetriol (BT), is readily oxidized to its corre- total MN, mainly in K+-MN, was observed in sponding quinone via a semiquinone radical. both HL60 cells and lymphocytes. Addition of During this process, active oxygen species are copper ions (Cu2+) potentiated the effect of BT formed that may damage DNA and other cellu- on MN induction threefold in HL60 cells and allar macromolecules. The ability of BT to induce tered the pattern of MNformation from predommicronuclei (MN) and oxidative DNA damage inantly K + to K-. BT also increased the level of has been investigated in both humon lympho- 8-hydroxy-2'-deoxyguonosine (8-OH-dG), a cytes and HL60 cells. An antikinetochore anti- marker of active oxygen-induced D N A damage. body based micronucleus assay was used to dis- Cu2+ again enhanced this effect. Thus, ET hos tinguish MN containing kinetochores ond the potential to cause both numerical and struc- potentiolly entire chromosomes (kinetochore- tural chromosomal changes in human cells. Fur- positive, K+) from those contoining acentric ther, it may cause point mutations indirectly by chromosome fragments (kinetochore-negative, generating oxygen radicals. BT may therefore K-). BT increased the frequency of MN forma- ploy an important role in benzene-induced leution twofold in lymphocytes and eightfold in kemia. o 1993 WiIey-Liss, tnc. HL60 cells with the MN being 62% and 82% K+, ~, Key words: 1,2,4-benzenetrioI, micronuclei, oxygen radicals, 8-hydroxy-2'-deoxyguanorine, human leukemia INTRODUCTION Benzene is an important industrial chemical and an ubiquitous environmental pollutant. It causes myelotoxicity and myelocytic leukemia in humans [Goldstein, 1977; IARC, 19881 and multiple forms of cancer in rodents [Huff et al., 19891. Benzene itself is unlikely to be the ultimate carcinogen. It is metabolized in the liver to its primary metabolite phenol (PH), which is further hydroxylated to catechol (CAT), hydroquinone (HQ), and 1,2,4-benzenetriol (BT) [Inoue et al., 1989a, b]. truns,truns-Muconic acid, a ringopened benzene metabolite, is also formed via a different pathway [Inoue et al., 1989~1E. ach of these metabolites and intermediary species may play a role in benzene-induced carcinogenesis. Even though BT is a relatively minor benzene metabolite, its chemical properties suggest a high biological potency. BT is more unstable in molecular oxygen than the other metabolites, has a strong ability to bind transition metal ions, and rapidly autoxidizes to its corresponding quinone via semiquinone radical intermediates [Greenlee et al., 1981; Bandy et al., 19901. During the autoxidation of BT, 0 1993 Wiley-Liss, Inc. oxygen is converted to active oxygen-derived species, including superoxide anion radical, hydrogen peroxide, hydroxyl radical, and perhaps singlet oxygen [Greenlee et al., 1981; Pellack and Blumer, 1986; Bandy et al., 1990; Zhang and Davison, 19901. These reactive oxygen species can damage DNA and other cellular macromolecules [Halliwell and Aruoma, 1991; Stadtman and Oliver, 19911. BT has been reported to induce DNA strand breaks, which are elevated by Cu2+ and Fe3+ but inhibited by superoxide dismutase, catalase and hydroxyl radical scavengers [Lewis et al., 1988; Kawanishi et al., 19891. BT also induces sister chromatid exchanges (SCEs) in human lymphocytes [Erexson et al., 19851 and gene mutations (6-thioguanine resistance) in V79 cells [Glatt et al., 19891. Further, BT suppresses mitogenic responses and interferes with microtubule assembly in Received June 23. 1992; revised and accepted January 5 , 1993 Address reprint requests to M.T. Smith, Department of Biomedical and Environmental Health Sciences, School of Public Health, University of California. Berkeley. CA 94720. .e- l 340 Zhang et al. lymphocytes presumably via formation of quinones and ac- production is catalyzed by Cu2+, are involved in K--MN tive oxygen species [Irons and Neptun, 1980; Irons et al., formation. To further investigate their role, we used an 1981; Irons, 19851. These data suggest that BT may produce assay for the hydroxylated DNA base, 8-hydroxy-2'-deox- both chromosomal or DNA breakage (clastogenicity) via yguanosine (8-OH-dG) and show that BT increases levels of oxidative mechanisms and chromosomal lag (aneuploidy) this modified base in HL60 cells, an effect also potentiated by disrupting microtubules. by Cuz+. These data indicate the involvement of active Here, we have investigated the ability of BT to produce oxygen species in BT-induced DNA damage and implicate numerical and structural types of chromosomal changes us- the potential importance of BT-mediated genetic damage in ing a modified micronucleus assay [Fenech and Morley, benzene-induced human leukemia. 1985; Eastmond and Tucker, 19891. Using cells that have been blocked from cytokinesis by cytochalasin B, the pres- MATERIALS A N D METHODS ence or absence of micronuclei (MN) can easily be scored in cells having completed exactly one nuclear division. Use of Cell Culture an antikinetochore antibody allows differentiation of MN Human lymphocytes were isolated from peripheral blood containing entire chromosomes from those containing chro- of a healthy adult female using Ficoll-Paque (Pharmacia, mosome fragments. Kinetochore-positive micronuclei (K+- Piscataway, NJ) density gradients and cultured as described MN) have a high probability of containing entire chromo- previously [Yager et al., 1990; Robertson et al., 19911. somes, whereas kinetochore-negative micronuclei (K-- Briefly, 1 X IO6 lymphocytes were seeded into 2 ml of . .. .I.. MN) contain only chromosomal fragments [Vig and culture medium consisting of RPMI 1640 supplemented Swearngin, 1986; Degrassi and Tanzarella, 1988; Gudi et with 10% fetal bovine serum (Hyclone, Logan, UT), 2 mM al., 19901. We have successfully used this technique previ- L-glutamine, 100 unitslml penicillin, 100 p.g/ml streptomy- ously to show that benzene's primary phenolic metabolites cin (Gibco, Grand Island, NY), and 1.5% phytohemaggluti- PH, CAT and HQ are each capable of producing both nin (PHA) (HA15, Burroughs-Wellcome, Greenville, NC). K+-MN and K--MN in human lymphocytes [Yager et al., The lymphocytes were grown in a humidified incubator with 19901. We have also discovered that HQ and CAT act in 5% CO, at 37C for 72 hr. HL60 cells, a human myeloid cell concert to produce a synergistic induction of primarily line originally derived from the peripheral blood leukocytes K+-MN in human lymphocytes [Robertson et al., 19911. of a patient with acute myeloid leukemia [Gallagher et al., Whereas lymphocytes have been used in this and previous 19791, were obtained from American Type Culture Collec- studies, the human HL60 myeloid leukemia cell line is a tion (Rockville, MD). HL60 cells were also cultured in particularly suitable model to characterize the effects of RPMI 1640 supplemented with 10% fetal bovine serum and benzene's metabolites and to study the mechanisms in- 50 pg/ml gentamicin sulfate (UCSF, San Francisco, CA) at volved [Meyer et al., 19911. HL60 cells are representative 37C in a 5% CO, moist atmosphere and passaged twice of early lineage myelocytic bone marrow cells and contain weekly to maintain a density between 1 X IO5 to 1 X lo6 high levels of myeloperoxidase (MPO), which facilitates the cells/ml. The cells in the plateau phase of growth were then bioactivation of many benzene metabolites, including BT seeded in 15 ml sterile centrifuge tubes and diluted with [Gallagher et al., 1979; Smith et al., 1989; Subrahmanyam fresh complete medium to a density of 0.5 X lo6cells/ml (2 __I et al., 19921. In this report, we have extended our studies to ml per each culture). .......'?..,.. . . *...-I. BT and shown that BT induces both K+-MN and K--MN in human lymphocytes and HL60 cells. In combination with Treatment Conditions copper (Cu2'), the pattern of MN formation was altered from primarily K+ to K- accompanied by an increase in 1,2,4-benzenetriol (99%) (Aldrich, Milwaukee, WI) was total MN. This suggests that active oxygen species, whose dissolved in phosphate-buffered saline (PBS, Cu2+ and Mg2+ free, pH 7.4) immediately prior to treatment, and a solution of copper (11) chloride (Aldrich, Milwaukee, WI) was made fresh in distilled-deionized water. At 24 hr after Abbreviations culture initiation, lymphocytes were treated with BT in com- BN BT dG KK+ MN MPO 8-OH-dG binucleated cells 1,2,4-benzenetriol deoxyguanosine kinetochore-negative kinetochore-positive micronuclei myeloperoxidase 8-hydroxy-2'-deoxyguanosine plete medium for 24 hr, and HL60 cells were treated in PBS with BT only or in combination with Cu2+ for 1 hr, then washed and resuspended in complete medium. All treatments were performed in duplicate for each dose and repeated at least three times in HL60 cells. Cytochalasin B (Sigma, S t . Louis, MO) was added at 44 hr for lymphocytes and at 28 hr for HL60 cells to block the cells in cytokinesis (3 p.g/ml, final concentration). Both Genotoxicity of Benzenetriol in lymphocytes and HL60 Cells 341 TABLE I. Viability and Replicative Index (R. I.) of Lymphocytes and HL60 Cells Treated With 1,2,4-Benzenetriol (BT) and Copper Ions (Cuz')* Cell treatment [BT](pM) Lymphocytes" BT/Medium Viability (%) R.I. BTIPBS Viability (%) HL60 Cellsh (BT] + 5 pM Cu2+ R.I. Viability (%) R.I. 0 I 95 91 2.05 94 1.85 - I .56 - 94 - I .-52 5 - - 91 I .62 91 I .57 10 86 1.88 91 1.61 95 I .48 20 - - 90 I .60 92 I S O 25 89 1.86 - - - - 50 81 1.93 84 I .50 86' 1.50' 100 65 I .47 15' N/Ad - - +*Cell viability (70)= (number of counted living cells1400) x 100%; R.I. = [(70mononuclear cells) (2 X % binuclear cells) + (3 X % tri- or > trinuclear cells)]/100. "Lymphocytes were treated with BT in complete medium. bHL60 cells were treated with BT in PBS. 'Results were the average from only two experiments. 'R.1. was unscoreable because of the toxicity of 100 p M BT to HL60 cells. .I. ...... .. . treated lymphocytes and HL60 cells were harvested onto cein filter. Both the total number of MN, including K+-MN glass slides at 72 hr and 48 hr, respectively, using a cytocen- and K--MN, and the number of cells containing at least one trifuge (Cytospin-11, Shandon, Sewickley, PA) at 600 rpm micronucleus were recorded. for 10 min. Staining Procedures Detailed procedures for performing the micronucleus assay with the antikinetochore antibody have been described previously [Eastmond and Tucker, 1989; Yager et al., 19901. Briefly, following fixation of cells in methanol for 15 min, slides were dried completely in air, rinsed in PBS containing 0.1% Tween 20 (Fisher, Fair Lawn, NJ) for 5 min, and stained with an antikinetochore antibody (Chemicon, Los Angeles, CA) diluted 1:l with 0.2% Tween 20 in PBS. The slides were then coverslipped and incubated for 1 hr, washed twice in 0.1% Tween 20 in PBS and incubated with fluoresceinated rabbit or goat antihuman IgG (Chemicon, Los Angeles, CA) diluted 1:120 with PBS containing 0.5% Tween 20 for another hour. Again the slides were washed twice and then stained with the DNA-specific stain 4,6-diamino-2-phenylindole(DAPI) (Sigma, St. Lou is, MO) prepared in an antifade solution [Johnson and Nogueira Araujo, 19811. Scoring The stained slides were randomized and coded prior to scoring; 1,000binucleated cells per dose (500 per duplicate) were then scored for the presence of MN using a Nikon microscope equipped with epifluorescent illumination, a l 0 0 X oil immersion lens and the appropriate filters for fluo- rescein and DAPI [Robertson et al., 19911. The presence or absence of a kinetochore spot in each micronucleus was determined by switching from the DAPI filter to the fluores- Cell Viability and Cell Division Kinetics Cell viabilities were determined at 72 hr (lymphocytes) or 48 hr (HL60 cells) using trypan blue (0.16%). Replicative index (R. I . ) , a measure of cell division kinetics, was calculated by scoring 400 cells per dose (200 per duplicate) (see Table I legend). Extraction, Purification, and Enzymatic Hydrolysis of DNA from HL60 Cells DNA was isolated from HL60 cells according to a phenol extraction procedure [Gupta, 19841. High purity doubledistilled phenol (International Biotechnologies, New Haven, CT) was used for extraction to avoid additional oxidative DNA damage by peroxide or quinone contaminants in phenol. Following the DNA isolation, 200-400 p g of DNA from HL60 cells was resuspended in 200 kl of 20 mM sodium acetate (pH 4.8) and digested to nucleotides with 20 p g of nuclease PI (Sigma, St. Louis, MO) at 70C for 15 min. Then, 20 p1 of 1 M Tris-HC1 (pH 7.4) was added to the nucleotide mixture to adjust the pH, and the mixture was subsequently treated with 1.3 units of E . coli alkaline phosphatase (Sigma) at 37C for 60 min. Synthesis of 8-OH-dG Standard Standards of 8-OH-dG were synthesized by Udenfriend's hydroxylating system [Kasai and Nishimura, 19841. Deoxyguanosine (dG) was hydroxylated at the C-8 position by sequentially adding 25 p1 of 0.1 M dG, 14 p1 of 1 M 342 Zhang et al. -ascorbic acid, 6.5 p1of I M ethylene diamine tetra acetic -cn 50 acid (EDTA), and 13 p1of 0.13 M FeSO, to 0.942 ml of 0. I 3 M sodium phosphate buffer (pH 6.8). The mixture was incubated for 15 min at 37C in the dark with vigorous shaking. Following incubation, aliquots of the reaction mixture were analyzed by high pressure liquid chromatography Ual cm. !!u 40 3 C mI Kinelochon,+ (HPLC). Fractions containing 8-OH-dG were collected from HPLC eluates, lyophilized and stored at 4C. Determination of 8-OH-dG in DNA The hydrolyzed DNA solution was filtered using an "ultrafree" millipore filtration system (10,000 dalton cutoff). The amount of 8-OH-dG in DNA was measured by HPLC equipped with electrochemical detection using an ESA Cou- 0 2 30 L -.-nQ) 20 a C =.$lo c 0 A A lochem detector (model 6010). The applied potential was 0.05 V at the first detector and 0.35 V at the second detec- o.'E'n.t.,., .f . . tor. Levels of 8-OH-dG were expressed relative to the con.. , tent of DNA detected by absorbance at 260 nm. Standard solutions of 8-OH-dG (1-5 pmol) and dG (0.5 mM) were simultaneously analyzed to calibrate levels of 8-OH-dG and DNA, respectively. Data were digitized by a Nelson 760 (Cupertino, CA) analytical interface and processed by Nel- son analytical series 4400 data acquisition software on a Hewlett-Packard 98 16 computer. Results are expressed in pmol8-OH-dG/p.g DNA. Statistical Analysis R. 1. ( I .47) was also observed at this concentration (Table I). BT is therefore only a moderate inducer of MN in human lymphocytes. The statistical comparisons of multiple means were as- Effect of BT on Micronucleus Induction in sessed by analysis of variance (ANOVA). Individual means HL60 Cells were compared using a one-tailed Fisher exact test. The minimum level of significance chosen wasp < 0.05. Slopes To avoid potential confounding effects of antioxidants of dose response and Pearson's correlation coefficient (r- and other substances present in complete medium, HL60 value) were determined, where appropriate, by regression cells were treated with BT for 1 hr in PBS rather than in analysis. medium. A dose-dependent increase (p < 0.01) in the total ..'.. :...v:.:.Ii.,:I number of MN was observed in HL60 cells treated with BT between 5 and 50 p M (Fig. 2). The proportion of K+-MN ._ ~ RESULTS was consistently increased (p < 0.001)over the BT concen- Effect of BT on Micronucleus Induction in Human lymphocytes tration range tested. The minimum concentration of BT tested (5 p M ) induced a twofold increase in MN with 73% being K + . The maximum effective concentration of BT used Figure 1 shows the induction of MN in human lympho- (50 p M ) induced an eightfold increase in MN with 82% cytes treated with a range of BT concentrations in a com- being K+. I n contrast, 68% of the total MN were K--MN in plete medium. BT increased the total number of MN per untreated HL60 cells. At 100 p M BT, MN frequency and 1000binucleated cells in a dose-dependent manner (Fig. I ) . the R. I . were unscorable and cell viability was decreased to The mechanism of MN formation was examined using an 15% (Table 1). Regression analysis showed that the dose- antikinetochore antibody to distinguish between K+-MN related increase in both total MN and K+-MN was linear and K--MN. More K+-MN were formed in BT treated (correlation coefficient r > 0.99). The slopes of the lymphocytes than K--MN at each concentration (Fig. I ) . dose response curves were MN: 0.947 > K+-MN: Although BT produced a minimal response between I O and 0.825 >> K--MN: 0.122. The increase in K--MN was not 50 pM,at 100 p M (the maximum concentration tested) it statistically significant. Consequently, the increase in total caused a 2.3-fold increase (p < 0.01) of total MN above number of MN is almost entirely due to an increase in background, with 62% being K + . However, notable cyto- K+-MN. BT therefore induces predominantly K+-MN in i toxicity as indicated by trypan blue dye exclusion (65%) and HL60 cells. ' 80 -60 50 - 30 - Genotoxicity of Benzenetriol in Lymphocytes and HL60 Cells 343 Similarly, BT ( 5 pM) with varying concentrations of CuZ+(1-10 pM) induced primarily K--MN in HL60 cells, with at least twice as many K--MN as K+-MN (Fig. 3B). At the maximum response, the simultaneous administration of 5 p M BT and 2 p M Cuz+ produced a threefold increase ( p < 0.001) in the total number of MN with 85% being K--MN in comparison with 5 p M BT alone, which produced only 27% K--MN. Although the combination of BT ( 5 pM) with 5 or 10 p M Cu2+ again decreased the total number of MN, these concentrations were essentially nontoxic as measured by cell viability and R . I. (data not shown). Cuz+ (1-10 p.M), by itself, had no statistically significant effect on the induction of MN (data not shown). In summary, co-incubation with Cu2+ increases the total number of MN induced by BT and alters the pattern of MN formation from Kf-MN to K--MN. 0 10 20 30 40 50 60 Effect of BT on 8-OH-dG Levels in DNA of Concentrationof 1,2,4-Benzenetriol (pM) HL60 Cells Fig. 2. Effect of 1,2.4-benzenetriol on micronucleus induction in HL60 cells. The data represent the mean t S E of at least three different experiments. 8-OH-dG is a DNA-hydroxyl radical or DNA-singlet oxygen adduct [Kasai and Nishimura, 1984; Floyd et al., 1986, 1989; Devasagayam et al.. 19911. It has become a useful and sensitive marker of oxidative DNA damage [Floyd et a ] . , 19861. To test BT's ability to cause oxidative DNA damage, 8-OH-dG formation in the DNA of HL60 Effect of Cu2+ on the Induction of Micronuclei in HL60 Cells by BT cells treated with different concentrations of BT ( 5 , 10, and 50 p.M) in PBS was measured (Fig. 4A). BT, at all concen- Since transition metal ions (Cu" and Fe3+) stimulate BT-induced DNA strand breaks and accelerate the autoxidation of BT [Lewis et al., 1988; Kawanishi et al., 1989; Zhang and Davison, 19901, the effect of Cuz+ on the level and type of MN induced by BT was investigated. Cu'+ was chosen because it is more efficient than Fe3+ [Kawanishi et al., 1989; Zhang and Davison, 19901. When HL60 cells were incubated with varying concentrations of BT (0-50 pM) while maintaining Cu'+ constant at 5 pM, the total number of MN induced reached a maximum at 10 p M BT (Fig. 3A). When compared with 10 p M BT alone (Fig. 2 ) . the presence of 5 p M Cu2+ increased the incidence of total MN twofold (p = 0.05) and the formation of K--MN five- fold (p < 0.001) but decreased the proportion of K+-MN from 75% to 30%. Treatment with 5 pM Cu2+by itself did trations tested, increased the 8-OH-dG content in DNA to 0.10-0.30 pmol/pg DNA above the background level of 0.06 pmol/pg DNA. Treatments with 5 and 10 pM BT for 30 min caused a twofold and fivefold increase over the control, respectively. These values returned to steady-state levels by 60 min, suggesting an efficient DNA repair mechanism. Cells treated with 50 p M BT took twice as long (60 min) to show a fourfold increase in the 8-OH-dG level, which returned back to the original level after further incubation for 24 hr at 37" (data not shown). Though the reason for this is unclear. it may be due to excess BT andlor its semiquinone acting as an antioxidant and reacting with active oxygen species, thereby delaying the rise in 8-OH-dG. In PBS controls, the 8-OH-dG level remained almost constant around 0.06 pmol/pg DNA (Fig. 4A). not significantly effect MN or K--MN formation. Thus the Effect of Cu2+ on BT-Induced 8-OH-dG Formation addition of Cu2+ not only potentiated the MN formation but also altered the nature of BT-induced MN from K + to K-. Because simultaneous treatment with 5 p M BT and 2 pM In contrast to the linear dose response seen in Figure 2, the Cuzt induced the maximum response in K--MN formation total number of MN decreased at concentrations above 20 (Fig. 3B), the same combination was chosen to examine the pM BT in the presence of Cu" (Fig. 3A). This may be formation of 8-OH-dG in HL60 cells (Fig. 4B). As men- attributed to the combined toxicity of BT and Cu2+. How- tioned earlier, 5 pM BT alone increases the 8-OH-dG level ever, cell viability and the R. I. were not measurably reduced twofold after 30 min. Levels of 8-OH-dG were almost un- I at these doses (Table I). A possible explanation for this is changed when cells were treated with 2 p M Cu2+ alone for that the measurements of cytotoxicity may not estimate the true 15,30, and 60 min. However, at 15 and 30 min, the combi- toxicity because they were assessed 24 hr after the treatment nation of 5 p M BT and 2 FM Cu2+ increased 8-OH-dG and damaged cells might have been lost to analysis. levels by 0.04 pmol/p.g DNA above the 5 p M BT treatment 344 Zhang et al. -fn 50 J' 3 5 Kinetochore+ E Kinetochore- T I0 T 0 0 7- L al -.Q- al 0 3 C; f .c 0 naLl 5 z Control 0 5 10 20 50 Concentrationof 1,2,4-BenzenetrioI (pM) [Cu2+]=5pM -fn 3 "fl -al0 -mCI al 40 .0Ea- m Kinetochore - 0 0 0 30 T- L -.na-l al 0 20 a .E0e- z c 10 0 naLl E =I 2 Control 0 1 2 5 Concentrationof Cu2+ (FM) 10 [BT]=SpM Fig. 3. Effect of Cu" on I ,2,4-benzenetriol-induced micronuclei in HL60 cells. A: HL60 cells treated with different concentrations of I,2.4-benzenetriol in combination with 5 pM Cu'+. B: HL60 cells treated with different concentrations of CuZt in combination with 5 pM I .2,4-benzenetriol. Control indicates cells treated with neither BT nor CuZf. Each bar represents the mean 2 SE of at least three different experiments. I .. A 0.351 . Genotoxicity of Benzenetriol in lymphocytes and H160 Cells -+ control 5pMBT o.2010.24 I 6 IOph4BT `AM BT 0.161 345 0 1 5 30 45 60 75 Time (min) 0.00 ! 0 15 30 45 60 Time (min) Fig. 4. Effect of I .2,4-benzenetrioI on the 8-OH-dG level in the DNA of HL60 cells. A: Dose-response for I ,2,4-benzenetriol-induced8-OH-dG formation. Cells were treated with 5 p M BT. IO pM BT, and 50 pM BT in PBS for 0, 15, 30, and 60 min. Control cells were incubated in PBS and sampled at the same time points. B: Effect of I ,2,4-benzenetriol and Cu" in combination on 8-OH-dG formation in DNA of HL60 cells. Cells were treated with 5 FM BT, 2 pM Cu". and 5 pM BT + 2 pM Cu" in PBS for 0. 15. 30, and 60 min. 75 (Fig. 4B). Thus the presence of Cu2+ potentiated BT-in- gen (IO2). These reactive species are known to damage duced oxidative DNA damage. Moreover, the level of DNA and other cellular macromolecules [Halliwell and 8-OH-dG failed to return to steady-state by 60 min. This Aruoma, 1991; Stadtman and Oliver, 19911, The products may be explained by Cu2+ catalyzing the continuous pro- of BT oxidation, mainly, 2-hydroxy- 1,4-benzoquinone duction of hydroxyl radicals, singlet oxygen, and other ac- (2-OH-BQ) and its corresponding semiquinone radicals tive oxygen species that damage DNA. The presence of (s-Q'), could also produce genetic damage by, e.g., adduc- Cu2+ would therefore be expected to enhance BT-induced tion to DNA or by disrupting the mitotic spindle leading to chromosome breakage, which appears to be the case since chromosome loss. Different pathways for the induction of Cu2+ also potentiates BT-induced K--MN formation. genetic damage by BT and its oxidative products in cellular systems are proposed in Figure 5. DISCUSSION Here, we have shown that BT increases the incidence of MN in human lymphocytes (Fig. 1). MN arise when repli- Our laboratory continues to investigate the role of differ- cating cell populations are subjected to chromosomal break- ent benzene metabolites in benzene-induced human leuke- age by clastogens or to chromosome lag by mitotic spindle ... mia. Each metabolite tested so far, including PH, CAT, dysfunction. In the antikinetochore antibody modification of HQ, and 1,4-benzoquinone, induces genetic damage in hu- the MN assay [Fenech and Morley, 1985; Eastmond and man lymphocytes as measured by their ability to increase the Tucker, 19891, K+-MN represent the misincorporation of incidence of micronuclei (MN) [Yager et ai., 1990; Robert- whole chromosomes into the daughter nuclei or aneuploidy son et al., 19911. In this report, we have extended our and K--MN indicate the formation of acentric chromosome investigations to BT, another benzene metabolite, which is fragments or clastogenicity [Degrassi and Tanzarella, 1988; relatively unknown. BT has been detected in the urine of Vig and Swearngin, 1986; Eastmond and Tucker, 1989; workers exposed to benzene [Inoue et a ] . , 1989b1. Even Gudi et al., 19901. Using this assay, we have shown that though it is a minor metabolite in quantitative terms (Rusch BT can increase both chromosomal lagging and chromo- et al., 19771, BT may be important due to its instability in somal breakage. These events are significant genetically air and its strong ability to bind metals [Greenlee et al., since they represent a high probability of a net loss of ge- 1981; Pellack and Blumer, 1986; Bandy et al., 1990; Zhang netic material. and Davison, 19901. Thus it readily autoxidizes and acti- Relatively few studies have been conducted on the geno- vates molecule oxygen to a number of reactive species in- toxicity of BT and data regarding BT's potency relative to cluding superoxide anion (0;), hydrogen peroxide (H,O,), other benzene metabolites are mixed. Our results show that and hydroxyl radical (`OH) as well as perhaps singlet oxy- in lymphocytes BT behaves similarly to HQ, with the dose 346 Zhang et al. (---)+?DNA-quinone Adduct? DNA Macromolecules quinone \ i.4-b.nzoqulnon. -1breaks IClastogenicity (K'-MN) , (e.g. 8-OH-dG) Fig. 5. Different mechanisms for the induction of genetic damage by I ,2.4-benzenetriol and its oxidative products. required to produce the maximal response being in the 100 vitro, accelerates the decay of tubulin-colchicine binding FM range [Yager et al., 19901. HQ, however, produces a activity, and suppresses mitogenic responses in lympho- much higher number of MN than BT. Glatt et al. [I9891 cytes [Irons and Neptun, 1980; Irons et al.,, 1981; Pfeifer found HQ and 1,4-benzoquinone (p-BQ) to be more effi- and Irons, 198I]. Recently, using an immunocytochemical cient than BT at producing MN in Chinese hamster V79 staining assay, we have shown that BT also disrupts micro- cells with BT being more comparable to CAT. In the same tubule assembly in HL60 cells (Zhang et al., 1993). study, SCEs were analyzed and BT and CAT were found to The oxidation of BT generates reactive oxygen species, be more efficient than the other metabolites tested. This is in which is stimulated by transition metal ions such as copper contrast to results obtained in human lymphocytes, where (Cu', Cu2+)and iron (Fe2+, Fe'+). Thus, BT in combina- :4" . .. BT was the least efficient at producing SCEs except for PH tion with metal ions provides a model to investigate the role [Erexson et al., 19851. Overall, BT appears to consistently of these active oxygen species in BT-induced DNA damage. test positive as a genotoxic agent, leaving little doubt that it Previous studies have shown that BT induces DNA strand may contribute to the genotoxicity observed following ben- breaks, which are increased in the presence of Cu2+ and zene exposure. inhibited by the metal chelating agent, bathocuproine ':. .. Our investigations were continued using HL60 cells, a [Kawanishi et al.. 19891. In addition, Cu2+ was more effi. human myeloid leukemia cell line, representative of early cient than Fe3+ at stimulating BT-induced DNA strand df lineage myelocytic bone marrow-derived cells [Gallagher et breaks [Kawanishi et al., 19891, catalyzing the autoxidation al., 19791. Similar to bone marrow, HL60 cells contain an of BT [Zhang and Divison, 19901, and causing DNA base appreciable amount of MPO, which facilitates the bioactiva- damage in the presence of H,O, [Aruoma et al., 19911. In tion of BT enzymatically [Subrahmanyam et al., 19921. In this study, we therefore tested the effect of Cu2+ on BT- addition, they can be readily cultured in vitro. In HL60 induced DNA damage in two ways: first, by measuring its cells, BT not only increased the frequency of MN but also effect on the clastogenic potential of BT using K--MN as induced a high proportion of K+-MN, which indicates that the marker, and second, by measuring the formation of the chromosomes are not being properly segregated at mitosis. hydroxylated DNA base, 8-OH-dG. This may result from the disruption of the mitotic spindle We found that BT alone produced only a small proportion microtubules by the oxidative products of BT, 2-OH-BQ of K--MN, but when Cu2+was present, not only were more and/or s-Q' . BT inhibits rat brain tubulin polymerization in MN formed but also the proportion of K--MN increased C P r .. . '::, ..., . .... c Genotoxicity of Benzenetriol in Lymphocytes and HL60 Cells 347 from 27% to 85%. The fact that Cu2+changes the pattern of MN formation from K+-MN to K--MN in HL60 cells is consistent with our previous finding in a chemical system where Cu*+ changes BT-initiated free radical chain reactions from superoxide-propagated to Cu*+-mediated [Zhang and Davison, 19901. In this cell-free system, superoxide dismustase normally inhibited the autoxidation of BT, but when Cu2+ was present, it did not do so. Thus. Cuz+ will stimulate the oxidation of BT in cells containing superoxide dismutase and produce more active oxygen species, which is probably the primary cause of K--MN formation. In partic- ular, Cu2+ catalyzes continual formation of . OH via the +Fenton reaction (H,02 -. OH OH-), which would cause DNA base modification, strand breaks, and DNA or deoxyribose fragmentation [Halliwell and Aruoma, 19911. The ability of BT to produce oxidative DNA damage was evaluated using a sensitive HPLC technique to determine 8-OH-dG in DNA. 8-OH-dG represents the covalent addi- tion of . OH or '0,to guanine in DNA [Kasai and Nish- imura, 1984; Floyd et al., 1986, 1989; Devasagayam et a]., 19911 and has become a useful marker of oxidative DNA damage both in vitro and in vivo [Floyd et al., 1986; Roy et al., 19911. 8-OH-dG can be misread at the modified and adjacent base residues in a DNA-polymerase reaction in vitro, which is potentially mutagenic [Kuchino et al., 19871. Recently, Loeb and his coworkers [Cheng et al., 19921 have demonstrated that 8-OH-dG causes DNA base substitutions leading to point mutations. We found that exposure of HL60 cells to BT increased the level of 8-OH-dG. This increase was transient, however, and a decrease was observed during prolonged cell incubation. A possible explanation is that HL60 cells have an efficient repair mechanism, which may function to maintain a low 8-OH-dG level in the cells. In this regard, Kasai et al. [ 19861found that 8-OH-dG formation in DNA isolated from liver of y-irradiated mice rapidly decreased with time and speculated the presence of repair enzymes for 8-OH-dG removal [Kasai et al., 19861. The presence of CU*+ potentiated the BT-induced increase in 8-OH-dG, which did not return to background levels at 60 min. This most likely reflects the fact that Cu'+-binding sites on macromolecules serve as centers for repeated pro- duction of .OH radicals that are generated via the Fenton reaction. In conclusion, our data demonstrate that BT is genotoxic in both human lymphocytes and HL60 cells by increasing the frequency of MN and levels of 8-OH-dG. BT-induced MN most likely result either from the disruption of microtubules by its corresponding quinones or from oxidative DNA damage produced by reactive oxygen species generated during its oxidation. BT, by itself, causes mainly aneuploidy as indicated by the predominant induction of K+-MN. However, in combination with Cu*+ or other transition metal ions, BT causes mainly clastogenicity and point mutations as shown by the formation of K--MN and 8-OH-dG. Thus, BT may cause both numerical and structural chromosome aberrations as well as point mutations by two different mechanisms. BT's high potency suggests that it may play a role in benzene-induced genetic damage and perhaps human leukemogenesis. ACKNOWLEDGMENTS The authors are grateful to Dr. Vangala Subrahmanyam for his critical reading of this manuscript. L. Zhang is an awardee of The William and Ada Isabel1 Steel Memorial Graduate Scholarship from Simon Fraser University, B.C., Canada. M.L.R. and P.K. are trainees of the Health Effects Component of the University of California Toxic Substances Program. This work was financially supported by NIH grants P42-ES04705 and P30-ESOl896. REFERENCES Aruoma 01, Halliwell B, Gajewski E. Dizdaroglu M (1991): Copper-iondependent damage to the bases in DNA in the presence of hydrogen peroxide. Biochem J 273:601-604. Bandy B, Moon J . Davison AJ (1990): Multiple actions of superoxide dismutase: Why can it both inhibit and stimulate reduction of oxygen by hydroquinones'?Free Radical Biol Med 9: 143-148. Cheng KC. Cahill DS, Kasai H, Nishimura S, Loeb LA (1992): 8-Hydroxyguanine. an abundant from of oxidative DNA damage, causes G T and A C substitutions. J Biol Chem 267: 166172. Degrassi F. Tanzarella C (1988): Immunofluorescent staining of kinetochores in micronuclei: A new assay for the detection of aneuploidy. Mutat Res 203:339-345. Devasagayam TPA, Steenken S, Obendorf MSW, Schulz WA, Sies H ( I99 I ) : Formation of 8-hydroxy(deoxy)guanosine and generation of strand breaks at guanine residues in DNA by singlet oxygen. Biochemistry 30:6283-6289. Eastmond DA, Tucker JD ( 1989): Identification of aneuploidy-inducing agents using cytokinesis-blocked human lymphocytes and an antikinetochore antibody. Environ Mol Mutagen 13:34-43. Erexson GL. Wilmer JL, Kligerman AD (1985): Sister chromatid exchange induction in human lymphocytes exposed to benzqne and its metabolites in vitro. Cancer Res 45:247 1-2477. Fenech M. Morley AA (1985): Measurement of micronuclei in lymphocytes. Mutat Res 147:29-36. Floyd RA. West MS. Eneff KL, Schneider JE (1989): Methylene blue plus light mediates 8-hydroxyguanine formation in DNA. Arch Biochem Biophys 273:1061I1. Floyd RA. Watson JJ, Wong PK, Altmiller DH. Rickard RC (1986): Hydroxyl free radical adduct of deoxyguanosine: sensitive detection and mechanisms of formation. Free Rad Res Comms I : 163-172. Gallagher R . Collins S . Trujillo J , McCredie K , Ahearn M, Tsai S . Metzgar R, Aulakh G , Ting R, Ruscetti F, Gallo R (1979): Characterization of the continuous differentiating myeloid cell line (HL-60) from a patient with acute promyelocytic leukemia. Blood 54:713733. Glatt H , Padykula R, Berchtold GA, Ludewig G, Platt KL. Klein J , Oesch F (1989): Multiple activation pathways of benzene leading to products with varying genotoxic characteristics. Environ Health Perspect 8 2 8 1-89. Goldstein BD ( 1977): Hematotoxicity in humans. J Toxicol Environ Health (SUPPI.)2:69-105. Greenlee WF, Sun JD. Bus IS (1981):A proposed mechanism of benzene toxicity: Formation of reactive intermediates from polyphenol metabolites. Toxicol Appl Pharmacol 59: 187-195. .. - 348 Zhang et al. Gudi R, Sandhu S S , Athwal RS (1990): Kinetochore identification in 8-hydroxydeoxyguanosine at the modified base and at adjacent res- micronuclei in mouse bone-marrow erythrocytes: An assay for the idues. Nature 327:77-79. detection of aneuploidy-inducing agents. Mutat Res 234:263-268. Lewis JG, Stewart W. Adams DO (1988): Role of oxygen radicals in Gupta RC (1984): Nonrandom binding of the carcinogen N-hydroxy-2- induction of DNA damage by metabolites of benzene. Cancer Res acetylaminofluorene to repetitive sequences of rat liver DNA in 48:47624765. vivo. Proc Natl Acad Sci 8 1:6943-6947. Meyer K B . Subrahmanyarn V V , Kolachana P. Smith MT (1991): Bioacti- Halliwell B, Aruoma 01(1991): DNA damage by oxygen-derived species: vation and cytotoxicity of hydroquinone in human promyelocytic Its mechanism and measurement in mammalian systems. FEBS Lett leukemia (HL60) cells. Toxicologist I I: 180. 28 I :9-19. Pellack-Walker P. Blumer JL (1986): DNA damage in L5178YS cells Huff JE, Haseman JK, DeMarini DM, Eustis S. Maronpot RR, Peters AC, following exposure to benzene metabolites. Mol Pharmacol 30: Persing RL, Chrisp CE, Jacobs AC (1989): Multiple-site carcinoge- 42-47. nicity of benzene in Fisher 344 rats and B6C3FI mice. Environ Pfeifer RW, Irons RD (1981): Inhibition of lectin-stimulated lymphocyte Health Perspect 82: 125-163. agglutination and mitogenesis by hydroquinone: Reactivity with IARC (1988): IARC Monographs on the Carcinogenic Risk of Human. In intracellular sulfhydryl groups. Exp Mol Pathol 35:189-198. Fishbein L, O'neill IK (eds): "Methods of Analysis and Exposure Robertson ML, Eastmond DA. Smith MT (1991): Two benzene metabo- 0 Measurement-Benzene and Alkylated Benzenes." Vol IO. Lyon: lites, catechol and hydroquinone, produce a synergistic induction of International Agency for Researech on Cancer, pp 10.3-10.18. micronuclei and toxicity in cultured human lymphocytes. Mutat Res lnoue 0,Seiji K , Ikeda M (1989a): Pathways for formation of catechol and 249:201-209. I ,2,4-benzenetriol in rabbits. Bull Environ Contam Toxicol Roy D, Floyd R A , Liehr JG (1991): Elevated 8-hydroxydeoxyguanosine 43:220-224. levels in DNA of dietliylstilbestrol-treatedsyrian hamsters: covalent lnoue 0,Seiji K. Nakatsuka H. Watanabe T. Yin SN. Li GL. Cai SX, Jin DNA damage by free radicals generated by redox cycling of dieth- C, Ikeda M (1989b): Excretion of I .2,4-benzenetriol in the urine of ylstilbestrol. Cancer Res 51:3882-3885. ..... .... workers exposed to benzene. Br J Ind Med 4639-565. Rusch GM, Leong BKJ, Laskin S (1977): Benzene metabolism. J Toxicol .! Inoue 0, Seiji K, Nakatsuka H, Watanabe T. Yin SN, Li GL, Cai SX, Jin Environ Health (Suppl.) 2:23-36. C. lkeda M (1989~)U: rinary t,t-muconic acid as an indicator of Smith MT, Yager JW, Steinmetz KL, Eastmond DA (1989): Peroxidase- exposure to benzene. Br J Ind Med 46:122-127. dependent metabolism of benzene's phenolic metabolites and its Irons RD (1985): Quinones as toxic metabolites of benzene. J Toxicol potential role in benzene toxicity and carcinogenicity. Environ Environ Health 16:673678. Health Perspect 82:23-29. Irons RD, Neptun DA (1980): Effects of the principal hydroxy-metabolites Stadtman ER, Oliver CN (1991): Metal-catalyzed oxidation of proteins. J of benzene on microtubule polymerization. Arch Toxicol 45:297- Biol Chem 266:2005-2008. 305. Subrahmanyam VV. Kolachana P, Smith MT (1992): Metabolism of I ,2,4- Irons RD. Neptun DA, Pfeifer RW (1981): Inhibition of lymphocyte trans- benzenetriol by human myeloperoxidase: Implications in benzene- formation and microtubule assembly by quinone metabolites of induced myelotoxicity. Toxicologist 1259. benzene: Evidence for a common mechanism. J Reticuloendothel Vig BK, Swearngin SE (1986): Sequence of centromere separation: Kine- SOC 30:359-372. tochore formation in induced laggards and micronuclei. Mutagene- Johnson GD. Nogueira Araujo GMdeC (198I ) : A simple method of reduc- sis 1:461-465. ing the fading of immunofluorescence during microscopy. J Immu- Yager JW, Eastmond DA, Robertson ML. Paradisin WM, Smith MT no1 Methods 43:349-350. ( 1990):Characterization of micronuclei induced in human lympho- Kasai H, Nishimura S ( 1984): Hydroxylation of deoxyguanosine at the C-8 cytes by benzene metabolites. Cancer Res 50:393-399. position by ascorbic acid and other reducing agents. Nucleic Acids Zhang L, Davison AJ (1990): The effect of metals on the autoxidation of Res I2:2 137-2 145. benzenetriol: Cu changes the free radical propagated chain reaction Kasai H. Crain PF, Kuchino Y, Nishimura S. Ootsuyama A, Tanooka H but Fe does not. Free Radical Bio Med (Suppl. 1)9:185. (1986): Formation of 8-hydroxyguanine moiety in cellular DNA by Zhang L.. Venkatesh P. Smith MT (1993): Use of fluorescence in situ agents producing oxygen radicals and evidence for its repair. Car- hybridization to detect aneuploidy induced by I ,2.4-benzenetriol in cinogenesis 7: 1849-185l . HL60 cells. Proceedings of the American Association for Cancer Kawanishi S, lnoue S, Kawanishi M (1989):Human DNA damage induced Research. 34 (in press). by I ,2,4-benzenetriol, a benzene metabolite. Cancer Res 49: 164- 168. Kuchino Y , Mori F, Kasai H, lnoue H, lwai S, Miura K, Ohtsuka E, Accepted by - Nishimura S (1987): Misreading of DNA templates containing A. Morley