Document 1JG9YX8nDpQV609goNGYORza

MUT 04994 209 R.V. 0027-5107 /`$03.50 201 Two benzene metabolites, catechol and hydroquinone, produce a synergistic induction of micronuclei and toxicity in CUItured human lymphocytes Moire L. Robertson, David A. Eastniond * and Martyn T. Smith Depurlniriit ./Bioinedicml ciiid Eiwiroiimi~iiiulI leirlih .Sciences, School o/ Puhlic Ilealih, lJiiiotv-.si/)o~/ C ' d i J o I i i r ( r , l l o - L ~ + ~ :C ` A 94720 (iJ..S A , ) (Received 23 July 1990) (Revision received 21 November 1990) Isenzene; Catec Sirmriiary A mixture of two benzene metabolites, hydroquinone and catechol, produces a striking synergistic gen ot oxic response in cu 1t u red h u nian Iy mphocy tes. `111is was demonstrated 11sing an ;in t i -k inetochore antibody modification of the micronucleus assay. Treatment with hydroquinone alone or in combination with phenol produced a 3-fold increase in micronucleated cells over background. Treatment with catechol or phenol alone and in combination produced only minor increases in the number of micronucleated cells. In contrast, simultaneous treatment with equimolar (75 pM)concentrations of hydroquinone and catechol resulted in a greater than 16-fold induction of micronucleated cells. Given an additivity model, 20 additional micronucleated cells would be expected (after correcting for background frequencies), yet 140 were observed. Further analysis revealed that over 90% of the micronucleated cells stained positively for kinetochores, indicating a high probability that these micronuclei contain entire chromosomes. This synergistic response appears to occur only at equimolar levels of hydroquinone and catechol. These results suggest that these metabolites are acting together to disrupt the mitotic spindle and interfere with chromosome segregation. These data provide further support for the hypothesis that multiple nietabolites acting in concert are involved i n the benzene-induced genotoxicity and leukemia i n humans. Nonrandom ch roninsomal aberra Lions, pri nci- pally deletions, traiislocations and the gain or loss of entire chromosomes, are associated with a * Present address: Environmcnt;il Toxicology Graduate Pro- yrarn. I l n i v c r x r t y of C ~ i i l i f t ) r n i i i ,Ilivcrhidc, (`A Y2521 (I1.S.A.). Correspondence: Dr. Martyn T. Smith, Department of R i o nietlical and Environmental l k a l t h Sciences, Scl~oolof I'uhlic I lc;ilI11,Clnivcrhity of ~ ` i i l ~ f ~ i rInlc~rL; c~l,ey. ('A 93720 ( U . S . A . ) . variety of human cancers, including leukemia (Yunis, 1983; Rowley, 1984). Many of these changes are thought to be primary events critical t o the onset of neoplasia (Klein, 1981; Cavanee et al., 1983; Tsutsui et al., 1983). The common industrial solvent, benzene, causes the induction of tmt 11 st r u c turn1 ilntl numerical chromosome changes (Erdogan and Aksoy, 1973; Ding et al., 1983) in the peripheral blood cells o f highly exposed workers ;is well as niyelotoxici ty and 202 leukemia (IARC, 1988). Benzene, itself, is unlikely to be the actual toxic species, but is converted within the liver to species which exert myelotoxic effects. The primary metabolites are phenol (PH). hydroquinone (IIQ), catechol (CAT) and t r m ~ , trans-muconic acid (Gad-El-Karin1 et al., 1985; Cooper and Snyder, 1988). Both HQ and CAT have been shown to accumulate in rodent bone marrow (Rickerr et al., 1979). These phenolic metabolites may then be oxidked t o highly toxic quinones by peroxidase enzymes such as myeloperoxidase, present in the bone marrow (Smith et al., 1989). However, the metabolite(s) ultimately responsible for the myelotoxicity and leukemia produced by benzene remains unknown. In older t o study the ability o f different benzene metabolites t o induce certain types of struct- ural and numerical chroniosomal changes, namely chromosorrie breakage o r mi lscgrega t ion, in human blood cells, we have used a modified niicroniicleus assay 111 isolated peripheral human lymphocytes. This technique employs cytochalasin B to inhibit cytokinesis resulting in the formation of multi-nucleated cells which can easily be scored for the presence or absence of micronuclei (Fenech and Morley, 1985). Chromosome breakage and malsegregatiori events can be differentiated utilizing an anti-kinetochore antibody whereby kinetochore-positive micronuclei have a high probability of containing an entire chromosome. and kinetochore-negative micronuclei contain only ch romosoma I fragnien ts (Eastmond and 'I'uck er ~ 1989). Using this technique, we recently denionstrated that several of the phenolic metabolites of benzene and the HQ oxidation product, 1,4-benzoquinone (BQ), induce micronuclei in human lymphocytes (Yager et al., 1990). Based on the slope of the dose-response curves, the relative potencies for niicronuclei induction were as fol- lows: BQ > HQ > C A T > PH (Smith et al., 1990). Previous studies in our laboratory have shown that both P l i and C A T will stimulate the per- oxidase-dependent oxidation of HQ to BQ (Smith et al., 1989) and that the simultaneous adniinistration of PH and HQ produces myelotoxicity siniilar to that produced by benzene (Eastmond et al., 1987). This work, and that of others (Snyder et al., 1989; Barale et ai., IYYO), suggests that a combination of metabolites, rather than a single metabo- lite, is responsible for the myelo- and genotoxic effects of benzene. We have therefore investigated the ability o f different combinations o f benzene metabolites to induce micronuclei in liunian Iym p11 ocytes. A s1r i k i ng synergistic i nd tict ion of micronuclei following treatment with 1-I() and CAT is reported here. Materials :iacI nielliocls Cell culture arid ireuiriierit conditions. Peripheral human lymphocytes were isolated from the same healthy adult male using Ficoll-Paque (Pharmacia, Piscataway. N J ) density gradients and cultured as described previously (Yager et al , 1990). Briefly, lymphocytes a t a density of 5 x I O ' cells/ml were cultured for 72 11 at 37C in a 5% CO, atmosphere. The total culture volume before acldiriori of the chemicals was 2 nil. Culture medium consisted of R P M l 1640 supplemented with 10% fetal bovine serum (llyclnne, Logan. U'r), 2 m M r.-glutaniine, 100 units/ml penicillin. 100 pg/ml streptomycin (all from Gibco, Grand Island, NY), and 1.5% phytohemagglutinin (PMA) ( H A 15, Burroughs-Wellcome, Greenville, NC). P l i (C'AS NO. 108-95-2), C A T (C'AS NO. 120- 80-9) and HQ (CAS No. 123-31-9), all purchased from Aldrich (Milwaukee, WI), were diluted in phosphate-burrered saline (PIE) (Val ' and Mg2' free) imtnediately prior to treatment at 24 h in complete medium. Total volumes added did not exceed YO 111. For experiments with mixtures, the order of addition was as follows: PH, C A T then HQ. Viability was determined at 72 h using the trypan blue dye exclusion technique. Micronucleus assa-y. PHA stimulated lymphocytes were cultured for 44 h before the addition of cytochalasin 13 (Sigma, St. Louis, MO) (3 pg/ml final concentration). Cells were harvested at 72 11 onto glass slides using a cytocentriluge (Shandon. Sewickley, PA). After allowing the slides to air dry, they were fixed in methanol for 15 min. dried and stored desiccated, under a N, atmosphere al - 20 C until use. /niniurtoffu~~re.rc.eilc.eS. lides were stained with an anti-ki netochore an ti body (Antibodies I tic., Davis, CA) essentially as described (Eastnioncl and at 3 anticon ti for ; anti-I After 4',6-( IANI~ Nogu were SC, scorec epiflu rescei barrie at 40( nn,). , that 1 scorec numb1 locate, ahsenc fluore! sure c scorinl Table StUl of the (ANO' MacIn Results The nucleat alone a and c'/ creases respect I decreas HQ (7: and 28 backgrc decreas lrea tine appreci; oxic ated Eene man 1 of and iphthe lque and al., io5 5% fore ture ited :an, Ilin, and {A) 1120[sed I in g2+ I in not the hen the Iho1 of /ml 2 11 Ion, air -ied : at ith X., )nd and Tucker, 1989). Briefly, slides were incubated at 37C in a humidified atmosphere with the anti-kinetochore antibody for 1 h, washed in PBS containing the detergent Tween 20, and incubated for another hour using a fluoresceinated goat anti-human IgG (Antibodies Inc., Davis, CA). After washing, the slides were then stained with 4',641i a 111id in0 - 2-pI1eny I i ndole (I>A1'1 ) (Sigma, St. Louis, MO) in an antifade solution (Johnson and Nogueira Araujo, 1981) and coverslipped. Slides were stored refrigerated until use. Scoring. Randomized and coded slides were scored using a Nikon microscope equipped with epifluorescent illumination using filters for fluo- rescein (excitation at 470 nm,dichroic at 510 nm, barrier at 520-560 nm) and quinacrine (excitation at 400-440 nm, dichroic at 455 nm, barrier a t 470 nm). At least 1000 binucleate lymphocytes (those that have undergone one mitotic division) were scored, 500 from each duplicate culture, for the number of micronuclei. When ii micronucleus was located using the quinacrine filter, the presence or abscnce of ii kinetochore spot was noted using the fluorescein filter. fkplicative index (R.I.), a measure o f cell division kinetics, was calculated by scoring 400 cells per dose (200 per duplicate) (see Table 1 legend). Starrsricd cinu!ysrs. The statistical significance of the results was assessed by analysis o f variance (ANOVA), using the Super Anova" program for a Mac1ntosh. Residts `The induction of micronuclei and microi1licleiltetl cells by primary metitbolites o f benzene alone and in combination is shown i n Table 1 . I'H and CAT alone did not produce significant increases over the PBS control at 1 mM and 75 pM, respectively. While lymphocyte viability was not decreased, CAT treatment lowered the R.I. t o 1.4. H Q (75 p M ) treatment produced 30 micronuclei a i d 28 il1icronucleiiteci cells, ;I 3-fold incrciisc over background. This treatment also produced slight decreases in both viability and R.1. Simultaneous treatment with PI-l and CAT did n o t prodiice appreciable increases in micronuclei or micro- 203 TABLE 1 INDUCTION OF MICRONUCLEI IN CYTOKINESIS- BLOCKED HUMAN LYMPHOCYTES FOLLOWING TREATMENT WITH BENZENE METABOLITES ALONE A N D IN COMBINATION Number per 1000 BN ab R.I. E Viability Micro- Micronuclei nucleated cells Control (PBS) 10 9 (2) ( 1 ) 2.0 94 Phenol (1 mM) 1 3 12 (4) (4) 1.9 96 Cdtechol (75 PM) 1 1 10 ( 5 ) (4) 1.4 94 Hydroquinone (75 P M ) 30 28 (4) (4) 1.7 83 +Catechol phenol (75 pM t I mM) 13 13 (4) (4) 1.s 88 +Ilydroquinone phenol (75 pM + 1 pM) 35 28 (2) (4) 1.6 SO Cdtechol + hydroquinone 200 149 1.3 62 (75 pM +75 p M ) (3) (20) Catechol + +hydroquinone 221 149 . 1.3 62 phenol (80) (45) (75 1M + 75 p M + 1 m M ) - * Data represent the mean from duplicate experiments. Num- bers in parentheses are the standard deviation. `' BN is the number of binucleate cells. K . I . , the replicative index, is defined as: [(I X %! mono- + +nucleated cells) (2 X % binucleated cells) ( 3 X % tri and > trinucleated cells)]/lOO. nucleated cells over hiickground or their respective controls. However, the addition of PH with HQ produced a borderline significant (described below) increase in the number of micronuclei over that observed from 110 treatment alone. Neither of these mixtures (PI3 with C A T o r P1-l. with HQ) significantly altered the R . I . o r viability of the lymphocytes from their respective C A T or I IQ controls. Treatment with equimolar (75 pM) levels of C A T and 10,however, produced a synergistic increase in both micronuclei and micronucleated 204 CONTROL (PBS) PH ( I m M ) HQ ( 7 5 uM) CAT ( 7 5 uM) PH + CAT PH + HQ HQ + CAT cells. SI CA'J' p r hi nuclea over hat di tivi ty, would b, cells a h nkronuc ground. TAf1L.f; 2 DIST'R I BI, NUCLEUS 0 100 200 MICRONUCLEATED C.ELLS PER 1000 l , i R . I . lti(Jitt.lioiiof k i i i ( . l ( K , l i t > i c * povtivt. :iml iiq!;iIiv~* i r i i c . r l i t i i t c . l ~ . : i ~ ~ ~t.tt.l11.; i t i c.ylcAiiiris 1)Ioc.kvil I i i i ~ i t : ~li iy ~ ~ i p l i c ~ ~ v Il Ic~~I .I ;I W ~ I I ~ trcatinent with I,eii~errc t ~ i c t ; ~ l ~ o l ~iiltocisie iiiitl it1 cotiiliiiiatioti. '1 liesc Cliilii itre ii subset o f the tliita preseriled in 'I'ahle 1. D represents the nuinher or micronucleated cells containing at least one kinetochore-positive micronucleus. represents the nuinher o f micronucleated cells containing a t least one kinetocllore-neRative micronucleus. E Z 3 3 represents the nuinher of iiitcrotiuclci unscol-;ihlc f o r kinctocliol-cs Pf3S cciritrol CONTROL (PSS) CAT ( 5 uM) CAT (25 uM) CAT ( 7 5 uM) HQ ( 7 5 uM) HQ + CAT (5uM) HQ +CAI' (75 pM + 75 p 0 2 0 4 0 60 80 100 MICRONUCLEATED CELLS PER 1000 -Fig. 2. Induction of kinetochore-positive and -negative micronucleated cells in cytokinesis-blocked human lymphocytes following treatment with 75 p M hydroquinone and varying concentration of catechol. represents the number of micronucleated cells containing at least one kinetochore-positive micronucleus. hB%% represents the nomher of micronucleated cells containing at leas1 one kinetochore-negative micronucleus. represents the number o f micronuclei unscorable for kinetochores. Only 964 BN cells scored, results normali7.ed t o 1000 BN cells. ' Data represe BN. binticlea MN, inicront KIN. kinetoc e U. unscorablc cells. Simultaneous administration of 1IQ and CAT produced 149 micronucleated cells per 1000 binucleate cells, a greater than 16-fold increase over background. Given .a model of simple additivity, simultaneous exposure ta HQ and CAT would be expected to produce 20 micronucleated cells above background, yet an additional 140 micronucleated cells were observed over background. This yields a 7-fold increase over that 205 expected given an adclitivity inodel. While vii~bilily was decreased to 62% with this treatment, a similar decrease was not observed in the R.1. Addition of PI3 to the mixture of HQ and CAT produced a small increase of total micronuclei, but did not increase the number of niicroniicleated cells. The addition of PH to the mixture also did not affect viability or the K.1. An anti-kinetochore antibody modification of TAB1.E 2 DISTRIBUTION OF MICRONUCLEI A N D FREQlJENC'Y OF KlNETOCI1ORES CONTAINED WITHIN EACH MICRO- NUC'I.EUS ' PBS control CAT (75 pM) IjQ(75 P M ) IiQ -+ C A T (75p M -t 75 p M ) Nuniber o f BN hcell~ with 1 M N 4 1 13 2 1 10 6 2 1 2 2 I I 2 I 1 IO 26 13 2 I 3 3 2 3 1 2 I 2 Nunibrr of KIN per M N 0 3 0 1 2 0 1 2 5 7 8 9 IO II 13 U' 0 1 2 3 4 5 6 7 x 10 11 14 U Number of BN cells with 2 M N N unihrr or K I N per M N Number of BN cells with >/=3MN Number of KIN per M N 0, I 1. I 1 2.2,1 1 u, u, u 3 1. 1, I , 3 3 0.1. 2, 2,3 3 I 1 1 2 1 I I 1 1 1 206 the micronucleus test (Eastniond and Tucker, 1989) was used to distinguish between micronuclei con raining chromosome fragments (kinetochorenegative) and micronuclei containing entire clironiosomes (kinetocliore-positive). Fig. 1 shows t h a t for treatment of lymphocytes with the metabolites alone, the proportion of kinetochore-positive vs. -negative did not differ significantly from the control, except for t-IQ (see below). `Treatment o f lymphocytes with 1'11 and I-IQor P i ant1 C A T also did n o t alter the proportion of kinetochoreposi tive and -negative micronucleated cells relative to the controls. Treatment with the equimolar mixture of H Q and CAT, however, produced mostly kinetochore-positive rnicronucleated cells, with > 90% of the total M N containing kinetochores. This treatment has no effect on the niimber of kinetochore-negative micronucleated cells. The synergistic genotoxic effect of this combinat i o n on lympliocytcs can he expl;iined almost cntirely in terms of an increase in kinetochore-positive micronuclei. The dose-dependency of micronucleus formation on CAT concentration was investigated by varying the CAT concentrations from 5 to 75 p M while maintaining 1lQ constant at 75 p M (Fig. 2). The synergistic increase in micronucleated cells was not observed until equimolar concentrations of HQ and CAT were reached. This non-linear effect indicates that a protective mechanism, operational at the lower doses of CAT, may be overwhelmed with equimolar concentrations of CAT and HQ. Utilizing a subset of the data shown in Fig. 2, the number of kinetochores contained within each micronucleus is shown i n Table 2. These data illustrate that the equimolar dose of I-IQ and CAT result in many cells containing multiple micronuc.Iei onti t h;it t.;ich ruicrnnuclziis frequen 11). con- t ;I i11s I I I I I 1t I 1' 1T i1, 11c.11 \i1 1 1C~Y. X n anal\ si3 of \ ariance \\ as perfurnied after pooling all the data presented in Table I and Figs. 1 : i i i L i :. P t l <'I- < - . . \ I - t'\ t h c i i i s d \ c s . cir thcir conibinatiuns. ha\ e no significant effect on an\ o f the results. HQ alone increases significantl?. the nuiiilwi- of riiic.roiiiii.lc.;~tt.~eid l s ( p = 0 . 0 2 2 ) atid o f kinetootiore-negative micronucleated cells ( p < 0.0001). Although borderline significant ( p = 0.053). a n interaction between PH and 1Q is the second most important effect at explaining the number of micronuclei formed. The most important cffect is ;I non-linear interaction between CAT and HQ ( p < 0.0001). `This interaction explains niosl of the results f o r the number of niicronuclei, micronucleated cells and kjnetochore-posilive micronucleated cells. I t has n o effect on the number of kinetochore-negative micronucleated cells. Discussion Here we report that simultaneous exposure o l human lymphocytes to two benzene metabolites. CAT and HQ, results in a striking synergislic induction of micronuclei in human lymphocytes. Greater than 90% of the micronuclei stained positively for kinetochores, indicating that they arose ;IS:I result o f clironiosonic lag evcnts. These data provide considerable support to the growing acceptance that benzene-induced myelotoxicity and genotoxicity is not the effect of a single metabolite acting at a single target site (Tice et al., 1982; Goldstein, 1989; Snyder et al., 1989; Barale et al., 19901, but rather multiple metabolites acting at several target sites, all o f wliich may be critical t o the onset of neoplasia. A considerable amount o f evidence exists demonstrating that the metabolites of benzene can produce biochemical lesions both individually and in combination. Individually, each of the primary metabolites have been shown to be capable of producing genotoxic activity in vitro (Erexson et al.. 1985; Glatt et al., 1989; Yager et ai., 1990). Additional evidence supports a role f o r mixtures of metabolites. Irons and associates (Eastmond et al., 1987) demonstrated that cn-admiiiisrr;ition of P t l and HQ resulted in sign i fie-an t r-edUC-I icms o f h i e mnrron' cellularity. $ I I \ dcr ci A I . 1 Wa) I1it.n sh{wt d t k i t c-kwibiniitions of PH and H Q or H Q and rruns.1run.smuconaldeh>-dewere s?-nerpisticin decreasing iron i i i L . ~ \ r ~ ~ c ~ r i~ll~[(>I i r~c~dI iblcxki LTIIS. ;\Iost rcx-zntl!. Barale et al. ( IYYO) reported that the in L . i t - 0 cnadministration of 3 mixture of PH and HQ 13I c ~I ilc*e~l ;I s!.rie rpi st i c i n d11c t ion of niicronucleated erythrocytes in mice. We also observed an interaction between PH and HQ, but of borderI i ne signif i carice. Addi t i ona I experi men ts are warranted in our test `these tl together in for benzer administra JiQf,ails characteri s Cancer In: possible ex hc inetabo clifferen tly nously fron Another lik be that sub ing benzenc ministratio] and coworl Since CAT oxidation o zene adniiI mainly cor 1983). i t is following b the critical observed ir benzene ( R achieved by the markec CAT descr changes ob exposed to 1983) may f o r ma t i o n posure to I- N otable blue d > e e. ment with However, I derestiniirtc dying as a to analy\is is thc ni;i measure o rently undt natulc of t doses beloi only niicrc tion of the ng the st im::tween on exmicroe-posion the :lealed ;ure of >olites, :rgistic ocytes. 1 posi1 arose ,e data rowing oxicity single : et al., Barale acting xi tical mount ,etaboI esions lividu: been activ, 1989; mrts a associd that in sigilarity. nbin a,transig iron ;ently, 1 vivo d HQ trricr.0served ) U t of nts are warranted to verify the presence of this interaction in our test system. These data implicate that PH and HQ acting together in the bone marrow may be responsible for benzene-induced carcinogenicity. However, administration of PH, a metabolic precursor to HQ, fails to reproduce the pancarcinogenicity characteristic of benzene exposure (National Cancer Institute, 1980; t t u f f et al., 1989). One possible explanation is that administered PH may be metabolized and distributed within the body differently than PH which is generated endogenously from benzene (Subrahmanyam et al., 1990). Another likely explanation for this difference may be that substantially more C A T is formed following benzene administration than following PH administration, as recently demonstrated by Inoue and coworkers using rabbits (Inoue et al., 1989). Since C A T is formed primarily ( > 95%) from the oxidation of clihydrodiol precursors following henzene administration (Billings, 1985), and PH is mainly converted t o I iQ (Sawahata and Neal, 1983), i t is likely that more C A T will be formed following benzene than P I I ~iclministration`.Fhus, the critical levels of CA`I' relative to HQ which are observed in the bone marrow of rats exposed t o benzene (Rickert et al., 1979), are unlikely to be achieved by the administration o f 1'1-1 alone. Given the marked synergistic genotoxicity of IlQ and CAT described here, the numerical chromosome changes observed in the lymphocytes ol` workers exposed to benzene (Forni et al., 1971; Ding et al., 1983) may therefore be due, at least in part, to the Formation o f both HQ and C A T following exposure to benzene. Notable toxicity (38%), as assessed by trypan blue dye exclusion, was only apparent after treatment with equimolar levels of HQ and CAT. Ilowever, this measure o f toxicity is likely to u n derestimate the true overall toxicity since cells dying as a result of treatment at 24 h may be lost to analysis a&72 h. A further indication of toxicity is the marked decrease in replicative index, a measure of cell cycle kinetics. Studies are currently underway in o u r laboratory focusing on the r i ; i ~ i ~oi f~ cllic i r i ~ c r ~ ~ i cb~cJlowrc~crr I I() i i l d < ` A ' I `;it doses below 75 pM. These studies will include not only micronucleus formation, but also investigution of the time course of toxicity. 207 Several mechanisms can be proposed for the observed synergism between H Q and CAT. The most likely explanation is interference with microtubule integrity. Microtubules, composed primarily of tubulin polymers, are essential for the proper segregation of chromosomes into the daughter nuclei during cell division. I n vitro treatment with HQ or its oxidation product BQ to isolated tubulin inhibits polymerization by binding at or near the GTP binding site (Irons and Neptun, 1980). This disruption could not be shown with C A T or PH. HQ, BQ and C A T also inhibit lymphocyte mitogenesis and agglutination, processes dependent on microtubule and sulfhydryl integrity (Sherline and Mundy, 1977; Chaplin and Wedner, 1978; Pfeifer and Irons, 1981; Epe et a]., 1989). Microtubule disruption can result from chemical blockage of at least 2 of the multiple (4-11) titratable sulfhydryl moieties (Kuriyama and Sakai, 1974; Mellon and Rebhun, 1976; Iketla and Steiner, 1978). Such sulfhydryl groups provide highly reactive nucleophilic centers for conjugate addition reactions with electrophilic compounds such as 1IQo r IIQ ( I'leifer and Irons, 1983). Thus, these findings suggest that the synergistic induction of kinetochore-positive micronuclei may be occurring from mechanisms whereby one metabolite facilitates the binding of the other to tubulin sull`hydryl moieties. Studies exploring this hypothesis are currently underway in our laboratory. Alternative molecular mechanisms which may be responsible for the observed synergistic effect include the formation of a multi-polar spindle (Sellitto and Kuriyama, 1988), disruption of microtubule-associa ted proteins and chromosome-topole movement (Paschal and Valee, 1987; Koshland et al., 1988) or interference with kinetochore capture of microtubules (Mitchison and Kirschner, 1985). Cleavage of kinetochores, as is seen with caffeine (Brinkley et al., 1988), would result in poleward movement of kinetochores detached from the chromosomes yielding predominately kinetochore-negative micronuclei. Since kinetochores were not observed i n the cytoplasm and because the majority of micronuclei observed i n iliis stiitly wcrc kinctoc.l~ol-c-posilivc;iiitl coiltainecl multiple kinetochores ('Fable2), kinetochore cleavage is unlikely to be a plausible mechanism. C`on siste nt , no n -r;i nd om st riict ii T i l I and n11meri - cal clironiosoninl cliatiges li;i\fe IYXII ohservcd in various forms of leukemia (Yunis. 1983; Rowley, 1984). The potential role for nurnerical clircitiiosoni;tl c1i;ingcs in c;ircinogcncsis is hccotiiiiig niore apparent a s the importance o f c1i;itig~'s i n (toxicokinctic\ itntl the molecitl;ir nspccts nf benzene toxicity), in: M . Aksoy (Ed.). l3enz.ene Carcinogenicity. C'RC Press. Hcr;t Raton. FL. pp. 3?~-SX. I)ing. X , 1.i. Y . , Ihrig. Y . . ;itid 11, Yang (1983)C.hroninsonie clt;tnpcs i n patictits w i t h clirottic hcnictic pciiscuiing. C'hin. Med. J.. Yh. 6x1- - 6 X 5 . gene dosage and balance, expression of recessive liastinond. D . A . , Smith, M.T.. arid R.D.Irons (19x7) An mutations and changes in genetic stability become renlized (Osliimura and Harrett. 1986: I'pstciti. 1988). The role for aneuploidy a s 21 priniary event leading to carcinogenesis is also being clarified. inter;tction of henzene metabolites reproduces the myelotoxirily produced w i t h bcn7enc exposure. 'I'oxicol. Appl. l'li:irin:tcol,. 01, X5 95 l::tstnicind. I1.A.. ;itid 1.1). 'Iucker (19x9) Identification of aricuploid~-itidttcirlgagents using cytokinesis-blocked hu- especially with the recent evidence for the involve- n t i t n Ivniphocytes antl ;in iititikinetoctiore ;intibndy. En- ment of chromosomal loss in the cascade of events viron. Mol. Mutagen.. 13. 34-43. leading t o colon carcinogenesis in humans (Vogel- llpe. 13.. U.II. llnrhig. I). Scliiffniann and M. Metzler (1989) h4icrotuhular proteins as cellular targets fnr carcinogenic stcin et al., 19x9). The syncrgistic ~itic~ii~l(~i[ly-iii- c\ticigc'iis ;itit1 othcr ciiicinogenF. in: M.A. Rcsnicl ;iiid ducing effect of I IQ and C A I ' described liere may H.K. Vig (lids.). Meclianisriis o f C'liromosorne 1)istrihution therefore play a key role i n the tleveloprnent of and Aneuplnidy. Liss. New York, pp. 345-351. benzene-induced leukemia. 1-pstein. C.1. (1988)Mechanisms of the effects of aneuploidy in ni:imiiiitls. Annu. Rev. Genet.. 22,S 1 - 75. Acknowledgenierit s I:i-dog:rn. ( i . . and M. Aksoy (1973) Cytogenetic studies in thirteen patients with pancytopenia and leukaemia associ- ;ttctl with long-tcrrn cxposurc to henretie. New Istiinhd The authors are indebted to Dr. Janice W. C'ontrib. Clin. Sci.. IO, 230-247. Yager for her expert advice on the micronucleus assay, to Dr. Frederic Bois f o r the statistical m a l ysis and to Mr. William M. Parasidin for technical assistance. M.L.R. is a trainee of the Health Effects Component of the U.C. Toxic Substances Itrexson. G.L., J.L. Wilnier and A.D. Kligerrnan (1985)Sister chromatid exchange induction in human lymphocytes expnsed to benzene and its riietaholites iii vitrn. Cancer Res.. 45.2471 -2477. Fenech. M..antl A.A. Morley (1985)Measurement of micronuclei in lymphocytes, Mutation Res.. 147. 29-36. i'rograni. Suppor tcd hy N11 I pranls 1'42 l!S04705 and 1'30 ESOI 896. l'ciini, A.. (';ippelliiii. A , . h c i f i c o . I:,.:iiitl I:. Vipli;ttii (1971) ( 'litottiosonie cli:tiiges :itid tlicir cvolutioit iii subjects with past exposure 10 benzeric. Arch. finviroti. I le:iltti, 23. 7XS References 391. (iad-15I-Karini. M.M.. V.M. Sadagopocla Ramanujam and M.S. Legator ( 1985) / r u m ,fmris-Muconic acid, a n open-chain Rarale, R., A. Marrazzini, C. Hetti, V. Vangelisti, N. Loprieno urinary nietaholite of benzene in mice. Quantification by and I. Rarrai (1990) Genotoxicity of two metabolites of high pressure liquid chromatography. Xenobiotica, IS, heii7enc: phenol and hydroquinone sliow strong synergistic 21 I 220. effects in vivo, Mutation Res., 244. 15-20, (;I:itt. I I . . i';idykul:t, R..Ilerchtold. G.A.. Ludewig. ( i . . Plait, Billings, R.E. (1985) Mechanisms of catechol formation from K.L.. Klein. J. and F. Oesch (1989) Multiple activation aromatic compounds in isolated rat hepatocytes, Drug pathways of henzene leading to products with varying Metab. Dispos.. 13. 287-290. genotoxic characteristics. Environ. Iiealth Perspect., 82, 13rinkley. 13.R.. Zinkowski. R.P.. Mollon. W.I,.. Davis, F.M., X I - XY. Pisegna. M A . , M. Pershouse and P.N. Rao (1988) Move- (ioltlstein. 13.1). ( 1989) Introduction: Occam's razor is dull, ment and segregation of kinetochores experimentally de- Fmviron. Health Perspect.. 82,3-6. tached from mammalian chromosomes, Nature (London), H u f f . J.H., J.K. Haseman. D . M . DeMarini. S. Eustis, R.R. 336, 251-254. Maronpot. A.C. Peters. R.L. Pershing. C.C. Chrisp and Cavanee, W.K.. Dryja, T.P., Phillips, R.A., Renedict, W.F.. A.C. Jacobs (1989)Multiple-site carcinogenicity of benz,ene Gobout. R.. Gallic. R.L., Murphee, A.L.. L.C. Strong and in Fisher 344 rats and B6C3F1 mice, Environ. Health R.L. White (1983)Expression of recessive alleles by chro- Perspect.. 82. 125-163. tiiosom;tl ntcch:itiisms i n retinohl;istotn;t, Nature (I.oticlon). l A R ( ' (IYXX) Ihll(' Monogr:iphs o t i the flvitlit;ition of the 305. 77%784. C'arcinogetiic Risk o f Chemicals to Ilunians. Vol. IO, L. Chaplin. D.D..and H . J . Wedner (1978)Inhibition of lectin-in- Fishbein and I.K. O'Neill (Eds.), Environmental Carcino- duced lymphocyte activation by dianiitle and other sulf- gens; Methods of Analysis and Exposure Measurement- hydryl reagents, Cell. Immunol.. 36, 303-31 I . Benzene and Alkylated Benzenes. Lyon, International Cooper. K.R. and R. Snyder (1988) Benzene metabolism Agency for Research on Cancer, pp. 3-18. 1ked:i. ' liii: Bioc Illouc. 4 tion Ell\ Iron\. I Iiyh r.;iIi I Johnscii SIlllF CCllC 350. Klein. C tioti: Koshlar I'<ik depc Ku riya11 grou 651 Mellon. vitrc. Mi tclii s( kine penc Nationa ble [ Heal 203. Oshimu :111c11 sigtii l~;iscll~ the (Lon Pfeifer, stimt hydr prnii' I'lei fer, funct rupti COl.. Rickert. Iron: by in Row ley. .oxicCRC some Zhin. I An yelo\PPI. 1 of huEn- 989) enic and tion y in i in ociibul ster exes., :ro- 71) lith 15 - IS. ain by 15, Ill, on ng 32, Ill, R. nd ne th Ikeda, Y.. and M. Steiner (1978) 'Sulfhydryls of platelet tubulin: their role in polymerization and colchicine binding, Biochemistry, 17, 3454-3464. Inoue, O., K. Seiji and M. Ikeda (1989) Pathways for fornia- , tion o f cateclic)l and 1,2,4-benzenetriol in rabbits, Ijull Environ. Contam. Toxicol., 43, 220-224. I I Irons, R.D.. and 1J.A. Neptuii (1980) Effects of the principal liydroxy-metabolites of benzene o n rnicrotubule polymerization, Arch. Toxicol.. 45, 297-305. Johnson. G.D., and G . M . de C . Noguiera Araujo (1981) A simple method of reducing the fading of immunofluorescence during niicroscopy, J. Immunol. Methods, 43, 349- 350. Klein, G . (1981) The role of gene dosage and genetic transposi- tions in carcinogenesis, Nature (I-andon). 294, 313-316. Koshland. D.E., T.J. Mitchison and M.W. Kirschner (1988) Polewards chromosome movement driven by microtubule depolymerization in vitro, Nature (Lmdon), 33 I , 409-504. Kuriyania, K., and H. Sakai (1974) Role of tubulin - S t i groups in polymerization to microtubules, J. Hiocheni., 76, 651---654. Mellon, M.G., and L.1. Itebhun (1976) Sulfhydryls and the i l l vitro polyrnerization of tubulin, J . Cell R i o l . , 70, 226-238. Mitchison, 'T.J., and M.W. Kirschner (1985) Properties o f the kinetochore in v i t r o , 11. Microtubule capture and ATP-dependent translocation. J . Cell Biol., 101, 766-777. National Cancer Institute (1980) Bioassay of phenol f o r possible carcinogenicity, N T P N o . 80-1 5, U.S. Ikpartment of Health and Human Services, Technical Iteport Series, Vol. 203. Oshimura, M., and J.C. Barrett (1986) Chemically induced aneuploidy in manimnlian cella: Mechanisms and biological significance in cancer. Environ. Mutagen., 8, 129-159. Pascli;il, 13.M.. and 11.11. Vallce ( 19x7) 1tetrogr;itle tranaporl b y the microtubule associated protein M A P I C , N a t u r c (London). 330, 181-183. Pfeifer, R.W., and R.D. Irons (1981) Inhibition of lectin- I stimulated lyniphrxyte agglutination and niitogeiiesis by hydroquinone: Reactivity with intracellular sulfhydryl s groups, Exptl. Mol. Pathol.. 35, 189--198. Pfeifer. I t . W . , and I<.[).Irons (1983) Alterntiori o f lymphocyte function by quinones through a sulfhydryl-depeiident disruption of microtubule assembly, I n t . J. Inirnunopharmacol., 5, 463--470. Rickert, I-).l<B.a,ker, T.S., h s . J.S., liarrow, C.S. and 11.1). I Irons (IY79) fknzene diqwsitioii i n ttic rat ;ifter exposure by inhalation, Toxicol. Appl. l'harrnacol., 49, 417--423. Kowley, J.D. ( 1984) Biological implications of consistent cliro- lie L. o- a1 209 inosoine rearrangements in leukemia and lymphoma, Cancer Res., 44, 3159-3161. Sawahata. T., and K.A. Neal (1983) Biotransformation of phenol to hydroquinone and catechol by rat liver micro- ~ ( i n i e s M. ol. Pharmacol., 23, 453 460. Sellitto, ('., and It. Kuriyama (1988) Distribution of peri- centriolar material in multipolar spindles induced by col- cernid treatment jn Chinese hamster ovary cells. J. Cell Sci., 89, 57--65. Sherline. P., and G.R. Mundy (1977) Kole of the tubulin-mi- crotuhule system in lymphocyte activation. J. Cell. Biol.. 74, 37 1-376. Sniitli. M.T.. Yirger, J W., Steinrnek, K.M. and [).A. tastniontl ( 1989) Peroxiciase-dependent rrietabolism of benzene's phe- nolic metabolites and its potential role in benzene toxicity and carcinogenicity, Environ. Health Perspect., 82, 23-29. Smith, M.T.. Robertson, M.L., Yager. J.W. arid D.A. East- mond ( 1990) Role o f inetabolisni in benzene-induced inyelotoxicity and leukemogenesis, in: M.L. Mendelsohn and K.J. Albertini (Eds.). Mutation and the Environment; Part H: Metabolism, Testing Methods, and Chromosomes. Wiley-l.iss, New Y o r k , pp. 125- 136. Snyder. It.. Iknitriadis, E., Guy, It., H u , P., Cooper. K.. Hauer, H.,Witz, ( 3 . and B.D. Goldstein (1989) Studies on the niechanism of benzene toxicity, Environ. Health Per- 5pect.. 82. 31- 35. Subrahnianyam. V.V.. Doane-Setzer, P., Steinmetz, K.L., Ross, I).a i d M.7'. Smith (1990) Plienol-induced stiinulirtion of hydroquinone bioactivation in iiiouse bone marrow in vivo: I'ossi ble iniplicetions i n benzene niyelotoxicity, Toxicology, 62, 107-116. Tice, lt.lt., TI-.Vogt iind Dl.. C'osta (1982) C'ymgcnetic effects of inhaled benzene in murine bone marrow, in: R . K . 'l'isc.'l..l:. V o g ~miti K.M. Scliaich (PAIS.).(ienotoxic Effects o f Airborne Agents, I'leiiuni. New Y o r k . pp. 257-275. Tsutsui, T., Maizumi, 1.. McLachlan, J.A., and J.C. Barrett ( 1983) Aneuploidy induction and cell transformation by diethylstilbestrol: A possible chromosomal mechanism in carcinogenesis, Cancer Res., 43, 3814-3821. Vogelstein, H., E.K. Fearon, S.E. Kern, S.R. Hamilton, A.C. I'reisinger. Y . Nakitniurii and R. White (1989) hllelotype of colorectal carcinoiniix, Science, 244, 207-21 I . Yager, J . W . , Eastmond, 11.A.. Robertson, M.L., Paradisin, W.M. and M.T. Smith (1990) Characterization of niicro- nuclei induced i n huiiiiin lymphocytes by h e n m i e metabo- lites. (`;inter Res.?50, 393 ~ 399. Y unis, J.J. (1983) 'The chromosomal basis of human neoplasia, Science, 22 1 227-236. ~ I