Detection of hyperdiploidy and chromosome breakage in interphase human lymphocytes following exposure to the benzene metabolite hydroquinone using multicolor fluorescence in situ hybridization with DNA probes

Genetic Toxicology

ELSEVIER

Mutation Research 322 (1994) 9-20

Detection of hyperdiploidy and chromosome breakage in interphase human lymphocytes following exposure to the benzene metabolite hydroquinone using multicolor fluorescence in situ hybridization with DNA probes David A. Eastmond *, D.S. Rupa, Leslie S. Hasegawa Environmental Toxicology Graduate Program, Department of Entomology, University of California, Riverside, Riverside, CA 92521, USA (Received 28 July 1993; revision received 13 January 1994; accepted 24 January 1994)

Abstract

Increased frequencies of structural and numerical chromosomal aberrations have been observed in the lymphocytes of benzene-exposed workers. Similar aberrations occurring in bone-marrow cells may contribute to the increased incidence of leukemia seen in these populations. Fluorescence in situ hybridization with chromosomespecific D N A probes is a relatively new technique which shows promise for the identification of aneuploidy-inducing agents. In these studies, fluorescence in situ hybridization with several chromosome-specific D N A probes was used to investigate the ability of the benzene metabolite hydroquinone to induce hyperdiploidy in interphase human lymphocytes. Using a classical satellite probe specific for human chromosome 9, a significant dose-related increase in the frequency of cells containing 3 or more hybridization regions was observed following the in vitro exposure of lymphocytes to hydroquinone at concentrations from 75 to 150 p~M. At the 100-~M concentration of hydroquinone, the frequency of nuclei containing 3 or more hybridization regions was determined using probes for chromosomes 1, 7 and 9. Significantly higher frequencies of affected nuclei were observed using the chromosome 1 and 9 probes when compared to the chromosome 7 probe. To establish whether this difference was due to the nonrandom involvement of these chromosomes in hydroquinone-induced hyperdiploidy or to chromosomal breakage within the chromosomal region targeted by these probes, a multicolor fluorescence in situ hybridization approach was developed using probes to two adjacent regions on chromosome 1. Using this tandem-labeling approach, the frequency of nuclei with multiple hybridization regions and the origin of the regions was determined by scoring slides labeled simultaneously with the chromosome 7 alpha satellite probe and the adjacent alpha and classical satellite probes for chromosome 1. The results of these studies confirmed that hydroquinone exposure resulted in a significant increase in hyperdiploid nuclei, but indicated that the different frequency of nuclei containing 3 or more hybridization regions observed using the chromosome 1 and 7 probes, was due to breakage within the chromosomal region targeted by the chromosome 1 classical satellite probe. These results indicate that hydroquinone may contribute significantly to the numerical and structural aberrations observed in benzene-exposed workers. In

* Corresponding author, Tel. 9097874497; Fax 9097873087. 0165-1218/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0165-1218(94)00006-O

10

D.A. Eastrnond et al. /Mutation Research 322 (1994) 9-20

addition, the multicolor fluorescence in situ hybridization approach utilized in these studies promises to be a powerful technique for the detection of chromosomal breakage occurring in interphase human cells. Key words: Hydroquinone; Benzene; Aneuploidy; Chromosome aberrations; Fluorescence in situ hybridization;

Interphase; Lymphocyte, human

1. Introduction

Benzene, a widely used industrial chemical and ubiquitous environmental pollutant, is an established leukemia-inducing agent in humans and a multi-organ carcinogen in animals (IARC, 1987; Huff et al., 1989). In spite of extensive research, the mechanism(s) involved in leukemogenesis remain unknown. Previous research has generally indicated that benzene is weakly active or inactive in standard gene-mutation assays and, in contrast to most carcinogenic agents, benzene exhibits poor DNA-binding ability (Dean, 1985; Lutz, 1986; Waters et al., 1988; Reddy et al., 1989). These results suggest that other types of genetic alterations such as translocations, deletions, recombination or aneuploidy may play an important role in the carcinogenesis induced by this agent. Studies of the blood and bone-marrow cells of benzene-exposed workers have shown that these individuals exhibit elevated frequencies of structural and numerical aberrations (Ding et al., 1983; Aksoy, 1988; Sasiadek, 1992; Eastmond, 1993). Similar chromosome-damaging effects have been observed in animals following exposure to benzene (Tice et al., 1980; Rithidech et al., 1987; Ciranni et al., 1991). In recent years, a growing body of molecular and cytogenetic evidence has indicated that these structural and numerical chromosomal alterations play an important and possibly essential role in neoplastic development (Yunis, 1983; Oshimura and Barrett, 1986; Hansen and Cavanee, 1987; Fearon and Vogelstein, 1990; Solomon et al., 1991). For the past several years, our research has focused on the mechanisms by which benzene induces its myelotoxic and genotoxic effects. In vitro studies on the ability of benzene's phenolic metabolites to induce centromere-containing and centromerelacking micronuclei have indicated that hydro-

quinone was effective in inducing both chromosomal loss and breakage in human peripheral blood lymphocytes (Yager et al., 1991; Rupa and Eastmond, unpublished results). The aneuploidy-inducing and chromosome damaging ability of hydroquinone has also been reported in a recent series of in vitro and in vivo studies (Natarajan, 1993; Adler, 1993). In addition to the aberrations observed in workers, elevated frequencies of chromosomal alterations have been observed in patients with pancytopenia, myelodysplastic syndromes and leukemia resulting from benzene exposure (Erdogan and Aksoy, 1973; Aksoy, 1988). Although the types of aberrations vary considerably between patients, trisomy of a C-group chromosome (chromosomes 6-12 and X) has been frequently reported to occur in these patients (Erdogan and Aksoy, 1973; Wolman, 1977; Aksoy, 1988). On several occasions the chromosome involved in this C-group trisomy was reported to be chromosome 9 (Forni and Moreo, 1967; Erdogan and Aksoy, 1973). Fluorescence in situ hybridization (FISH) with chromosome-specific DNA probes is a recently developed molecular cytogenetic technique which shows promise for the rapid identification of aneuploidy in human cells. The use of DNA sequences (probes) that hybridize to regions of repetitive DNA on specific chromosomes results in the staining of compact chromosomal regions which can readily be detected in both metaphase and interphase cells (Pinkel et al., 1986). Recently Eastmond and Pinkel (1990) demonstrated the usefulness of this technique to detect hyperdiploidy in interphase cells following the in vitro exposure of human lymphocytes to various aneuploidy-inducing agents. One advantage of this approach over conventional cytogenetics is that the frequency of hyperdiploidy for a chromosome of interest can be determined rapidly for a large

D.A. Eastmond et al. / Mutation Research 322 (1994) 9-20

number of cells, allowing relatively weak hyperdiploidy-inducing agents to be detected. As a result of these observations, we decided to investigate the ability of hydroquinone to induce hyperdiploidy in human lymphocytes following in vitro exposure using FISH techniques. Initial studies were performed using single-color FISH with a DNA probe specific for the heterochromatin of human chromosome 9. Later studies were expanded using probes for other human chromosomes and multicolor FISH.

2. Methods

Culture and treatment conditions Heparinized whole blood was obtained by venipuncture from a male donor whose lymphocytes had previously been shown to be sensitive to the chromosome-damaging effects of hydroquinone (Yager et al., 1990). Lymphocytes were isolated using Ficoll-Paque density gradients (Pharmacia, Piscataway, N J) and were cultured at 37°C for 72 h in a 5% CO z atmosphere at an initial density of 0.5 x 106 cells per ml. The culture medium was comprised of RPMI 1640 supplemented with 2 mM L-glutamine, 100 units of penicillin, 100 /zg/ml streptomycin (all from Gibco, Grand Island, NY), 10% fetal bovine serum (Hyclone, Logan, UT) and 1.5% phytohemagglutinin (HA 15, Burroughs-Wellcome, Greenville, NC). Hydroquinone (99 + %; CAS 123-31-9; Aldrich, Milwaukee, WI) was solubilized in phosphate-buffered saline and added to duplicate cultures at 24 h after culture initiation. At 72 h following the initiation of the culture, the cells were treated with a 0.075 M KCI hypotonic solution for 30 min, fixed 3 times with methanol:acetic acid (3:1) and dropped onto slides. The slides were stored at -20°C under a nitrogen atmosphere until use. Cells prepared using these procedures have been shown to be suitable for hybridization studies after 5 years of storage. Fluorescence in situ hybridization procedures The probes, nick translation and hybridization conditions for the single-color FISH studies were essentially those described previously (Eastmond

11

and Pinkel, 1990; Trask and Pinkel, 1990). In these studies, probes for 4 different chromosomal regions were used: an alpha satellite probe (pSD21-1) and a classical satellite probe (pUC1.77) for chromosome 1, an alpha satellite probe (p7atet) for chromosome 7 and a classical satellite probe (pHuR98) for chromosome 9. During the initial stages of these investigations, these probes were labeled by nick translation with biotin-ll-dUTP according to the instructions of the supplier (Bethesda Research Laboratories, Gaithersburg, MD). As these probes became commercially available, the biotinylated alpha satellite probes for chromosomes 1 and 7 and the biotinylated classical satellite probe for chromosome 9 were also purchased from Oncor Inc., Gaithersburg, MD. The classical satellite probe for chromosome 1 used in the multi-probe hybridizations was digoxigenin-labeled using nick translation. Labeled forms of this probe are now commercially available (Boehringer-Mannheim, Indianapolis, IN). The use of multicolored FISH with more than one probe necessitated a number of significant changes in the protocol. These are outlined as follows: Slides previously hybridized with the chromosome 7 probe were washed twice in PN buffer (0.1 M phosphate buffer, pH 8.0, containing 0.5% NP-40) for 10 rain and destained in methanol for 2 h. These slides were then denatured in 70% formamide/2 x SSC for 2 min, dehydrated in a 70%, 85% and 100% ethanol series and placed on a slide warmer at 37°C. The hybridization mix containing 1 p.1 (5-20 ng) digoxigen-labeled classical satellite probe for chromosome 1, 1/zl (5-20 ng) of the biotinylated alpha satellite probe for chromosome 1, 1 /zl (1 /zg) sonicated herring sperm DNA (Sigma) and 7 /zl Master Mix 2.1 (Trask and Pinkel, 1990) was denatured at 70°C for 2 min and placed on each slide. Following an overnight incubation at 37°C, the slides were washed 3 times in 60% formamide/2 X SSC and one time in 2 x SSC for 5 rain each at 45°C. The slides were then rinsed twice in PN buffer at room temperature. The digoxigenin-labeled classical satellite probe was detected using a mouse anti-digoxigenin antibody (3.2 ~ g / m l in PN buffer containing 5% nonfat

12

D,A. Eastmond et al. / Mutation Research 322 (1994) 9-20

milk supernatant [PNM]) followed by a Texas red-conjugated goat anti-mouse antibody (10/zg/ ml in PNM). The biotinylated alpha satellite probe for chromosome 1 and the previously hybridized alpha satellite probe for chromosome 7 were detected using fluoresceinated avidin (5 /zg/ml in PNM) followed by an amplification step consisting of biotinylated anti-avidin antibody (5/zg/ml in PNM) and another layer of fluoresceinated avidin. For the multi-probe hybridizations, 4',6diamidino-2-phenylindole (DAPI) at 2.5 /~g/ml in an antifade solution was used to counterstain the cell nuclei.

Microscopy and scoring criteria The slides were scored using a Nikon fluorescence microscope at a magnification of 1250. For the single label hybridizations, a blue filter (Nikon B-2A; excitation at 450-490 nm, emission at 520 nm) was used to visualize the fluorescein-labeled probes and the orange-red propidium iodidestained nuclei. For the multi-labeled hybridizations, a triple-band-pass filter (excitation at 375395 nm, 470-500 nm and 560-590 nm, emission at 445-470 nm, 515-555 nm and 610-660 nm, respectively; Omega, Battlemore, VT) was used to simultaneously visualize the fluorescein-labeled alpha satellite chromosome 1 and 7 probes, the Texas red-labeled classical satellite chromosome 1 probe and the blue DAPI-stained nucleus. In situations where the fluorescein-labeled probes were difficult to see due to a strong Texas red signal, the blue filter was used in conjunction with the triple-band-pass filter to allow the fluorescein signals to be more clearly identified. For all experiments, scoring was performed on coded slides. For the single-labeled probes, the previously described scoring criteria was used (Eastmond and Pinkel, 1990). However, for the multi-probe hybridizations the scoring criteria were slightly modified. Nuclei were simultaneously scored for the number of copies of the chromosome 1 probes and the chromosome 7 probe. For chromosome 1, hybridization regions in a nucleus comprised of a Texas red-labeled hybridization region (classical satellite probe) adjacent to a small yellow spot (alpha satellite probe) were scored as indicating the presence of an

intact chromosome 1. A clear green hybridization signal corresponding the chromosome 7 alpha satellite probe was scored as indicating the location of each chromosome 7. However, a nucleus containing 3 hybridization regions in which 2 were comprised of adjacent red and yellow fluorochromes and a third region containing only a Texas red-labeled region was scored as containing two copies of chromosome 1 with a breakage event having occurred in the chromosomal region targeted by the Texas red-labeled classical satellite probe. For both the single- and multi-labeled experiments, hybridization regions appearing as doublets or diffused signals were scored as one hybridization region.

Statistical analyses A variety of statistical tests were used to analyze the data. The Cochran-Armitage test for trend in binomial proportions was performed to determine whether the frequency of nuclei containing 3 or more hybridization regions exhibited a significant dose-related trend (Margolin and Risko, 1988). Following a positive response in the trend test, a 1-tailed Fisher exact test was employed to compare each treatment with the control. The Fisher exact test was also used to compare the hyperdiploidy and chromosome breakage results obtained using multicolored FISH. Analysis of variance was utilized to compare the number of nuclei containing various numbers of hybridization regions when probes for chromosomes 1, 7 and 9 were used (Abacus, 1987). Following a significant analysis of variance result, Scheffe's F-test was used as a post hoc test to compare the results of the individual chromosomal probes. Critical values were determined using a 0.05 probability of Type I error.

3. Results and discussion

Single label hybridizations To determine the hyperdiploidy-inducing effects of hydroquinone, human peripheral blood lymphocytes were treated in culture with hydroquinone at concentrations ranging from 25 to 150 /~M. FISH with a classical satellite probe for

D.A. Eastmond et al. /Mutation Research 322 (1994) 9-20

~

25-

"~ ~

20 ~

~o ~=oo

[] •

~dment I Ex~ment 2

•

[] o ~ ~_o

•

10.

-_~ 0

~

:

0

0

2's

s'o

~'s

i~0

i~5

i~o

Hydroqulnone Concentratlon (p.M) Fig. 1. Frequency of lymphocyte nuclei containing 3 or more hybridization regions using fluorescence in situ hybridization with a chromosome 9-specific DNA probe following in vitro treatment with various concentrations of hydroquinone. The mean frequencies of the results from two separate experiments are shown.

chromosome 9 was used to determine the frequency of nuclei containing 3 or more hybridization regions. 1000 or more cells were scored from two experiments at each concentration of hydroquinone and in the control. As shown in Fig. 1, a significant dose-related increase in the frequency of nuclei exhibiting 3 or more hybridization regions was induced by hydroquinone (p < 0.05, Cochran-Armitage binomial trend test). The frequency of affected nuclei increased linearly up to the 100-125-/zM concentrations of hydroquinone. The small reduction in the frequency of affected nuclei apparent at the highest (150/~M) concentration of hydroquinone is likely due to an effect of this agent on cell viability or cell kinetics. Significant increases in nuclei containing 3 or more hyperdiploid regions were observed at concentrations ranging from 75 to 150 /xM when compared with the control (p < 0.05, 1-tailed Fisher exact test). Significant dose-related increases in the frequency of nuclei containing 0 and 1 hybridization regions were also observed (p < 0.05; Cochran-Armitage binomial trend test). Nuclei with 0 or 1 hybridization regions increased from a combined frequency of 88 per 1000 in the 0-/.~M hydroquinone concentration to 141 per 1000 at the 125-/zM hydroquinone con-

13

centration. These relatively high frequencies of nuclei containing 0 and 1 hybridization regions reflect the influence of probe overlap and hybridization efficiency as well as chemical exposure. The influence of these factors on fluorescence in situ hybridization has been discussed previously (Eastmond and Pinkel, 1990). While scoring these slides, a relatively high frequency of micronucleus-containing cells was observed in which the micronucleus was labeled with the chromosome 9 classical satellite probe. These observations suggested that chromosome 9 might be involved in a non-random fashion in the aneuploidy-inducing effects of hydroquinone. In order to ascertain whether non-random alterations in chromosome 9 were induced by hydroquinone, the frequency of nuclei containing 0, 1, 2, 3, 4 or more hybridization regions was determined using probes for chromosomes 1, 7 and 9 in lymphocytes treated with the 100-~M concentration of hydroquinone. 1000 nuclei from each of 6 slides were scored using classical satellite probes for 1 and 9 and an alpha satellite probe for chromosome 7. All analyses were performed using slides from one experiment to eliminate experiment to experiment variability. The results of these analyses are presented in Table 1. Using the 3 probes, no significant differences were seen in the frequency of nuclei containing 0, 1 or 2 Table 1 Number of hybridization regions in interphase nuclei using various chromosome-specific DNA probes following in vitro exposure to hydroquinone Chromo- N a

Number of regions/1000 nuclei

some probe

0

1

2

3

4

5+

1

6 Mean 11.5 94.7 (Std. dev.) (3.9) (7.9)

877 16 1.7 0.2 (10.3) (3.3) (1.0) (0.4)

7

6 Mean 19.5 95.5 875 8.7 b 2.0 0 (Std. dev.) (9.7) (19.7) (28.1) (1.5) (1.5) (0)

9

6 Mean 14.2 107 861 14.3 3.3 0 (Std. dev.) (7.0) (11.7) (17.6) (4.5) (2.2) (0)

a Number of sets of 1000 cells scored from different slides from the same experiment. b Differs significantly from the hyperdiploid frequencies observed using probes for chromosomes 1 and 9 ( p < 0.05; analysis of variance followed by Scheffe's F test).

14

NONDItSJUNCTIOH/

D.A. Eastmond et al. / Mutation Research 322 (1994) 9-20

hybridization regions. However, significant differences in the frequency of nuclei containing 3 or more hybridization regions were observed between the 3 probes (p < 0.05, analysis of variance). The frequency of nuclei containing 3 or more hybridization regions using the chromosome 1 and 9 probes was significantly greater than that of the chromosome 7 probe (p < 0.05, Scheffe's F test). The observed difference between the probes was primarily due to differences in the frequency of nuclei containing 3 hybridization regions. Although these results could indicate the nonrandom involvement of certain chromosomes in the aneuploidy-inducing effects of hydroquinone, other explanations for these observations were possible. One possible explanation was that due to its somewhat smaller size, the lower frequency of nuclei with 3 or more hybridization regions observed for chromosome 7 might be related to a somewhat reduced ability to see the chromosome 7 probe. However, this type of hybridization problem would likely result in relatively high frequencies of nuclei with 0 and 1 hybridization regions. Since the frequencies of nuclei with 0 and 1 hybridization regions were comparable between the 3 probes, we concluded that this possibility was unlikely. Another possible explanation was that chromosomal breakage was occurring within the large classical satellite regions targeted by the chromosome 1 and 9 probes and that this breakage event was being incorrectly identified as an additional chromosome. These false hyperdiploid events should be a minor problem if breakage had occurred randomly throughout the genome. However, chromosomal breakage induced by a number of chemical agents exhibits highly nonrandom patterns (Brogger, 1977). For certain chromosomal regions, such as the chromosome lq12 and 9q12 heterochromatin regions targeted by these classical satellite probes, elevated frequencies of chromosomal breakage have been observed using conventional cytogenetic techniques both in untreated lymphocytes and in lymphocytes treated with clastogenic agents (Brogger, 1977; Eastmond and Pinkel, 1990). For example, 13.4% of the breakpoints observed in human lymphocyte chro-

t Fig, 2. In situ hybridization with two adjacent DNA probes to distinguish chromosome breakage from hyperdiploidy in interphase cells.

mosomes following treatment with the clastogenic agent triethylenemelamine occurred within the 9q12 region (Meyne et al., 1979). Our previous research had also indicated that this could be a potential problem when FISH was used to assess the aneuploidy-inducing ability of strong clastogenic agents (Eastmond and Pinkel, 1990).

Multi-label hybridizations In order to determine whether the increases in nuclei containing multiple hybridization regions were due to an increase in hyperdiploid cells or simply were the result of breakage within the chromosomal region targeted by the classical satellite probes, we designed a strategy for distinguishing between these two genotoxic events. We have termed this approach "tandem labeling". The principle underlying this approach is illustrated in Fig. 2. In this application, we used a classical satellite probe which hybridizes to the large heterochromatin region located in the pericentric region of chromosome 1. By placing a different fluorochrome on an alpha satellite probe that targets a small region adjacent to the heterochromatin probe, but which is much less prone to breakage, we are able to distinguish hyperdiploid cells from cells with breakage within the targeted heterochromatin region. The presence of an interphase nucleus containing 3 alpha satellite probes adjacent to 3 classical satellite probes

D~A. Eastmond et al. / Mutation Research 322 (1994) 9-20

indicates a nucleus which has 3 copies of chromosome 1. However, if a similar interphase nucleus contains only two alpha satellite probes adjacent to two of the 3 classical satellite probes, this indicates that a breakage event has occurred within the chromosomal region targeted by one of the classical satellite probes. This tandem-labeling approach was used to determine the contribution of breakage within the regions targeted by the chromosome 1 classical satellite probe to the frequency of nuclei containing 3 or more hybridization regions. Hydroquinone-treated cells, in which the number of hybridization regions had been previously determined using the alpha satellite probe for chromosome 7, were destained, denatured and hybridized a second time with a hybridization cocktail containing both the alpha and classical satellite probes for chromosome 1. Some examples of triple-labeled nuclei are illustrated in Fig. 3. Fig. 3a shows a metaphase nucleus and Fig. 3b an interphase nucleus following hybridization with the alpha satellite probe for chromosome 7 and the adjacent classical and alpha satellite probes

15

on chromosome 1. Using a recently developed triple-band-pass filter, the 3 probes and the DNA counterstain can be visualized simultaneously. Under these conditions the DAPI-stained nucleus appears blue, the fluorescein-labeled chromosome 7 alpha satellite probe appears green and the Texas red-labeled chromosome 1 classical satellite probe appears red. However, when the fluorescein-labeled alpha satellite probe for chromosome 1 is adjacent to the Texas redlabeled classical satellite probe, a shift in the emission wavelength occurs for the fluoresceinlabeled probe so that it appears yellow (Tkachuk et al., 1990). As a result, the 3 separate probes as well as the nucleus are visible simultaneously. Using this triple-label approach, the frequency of nuclei containing 0, 1, 2, 3, 4 or more hybridization regions for each of the 3 probes was determined. The results for the hyperdiploid and breakage frequencies for both the control and the 100-/zM hydroquinone treatment are presented in Table 2. A significant increase in the frequency of nuclei hyperdiploid for chromosomes 1 and 7 was observed at the 100-/zM hydroquinone con-

Table 2 Frequency of nuclei containing 3 or more hybridization regions for various D N A probes following in vitro exposure to hydroquinone N u m b e r of hybridization regions within the nucleus for the indicated D N A probe

Affected nuclei per 1000 cells at the indicated hydroquinone concentration

Chromosome 1 classical

Chromosome l alpha

Chromosome 7 alpha

3 2 3 4 3 3 3 5

3 2 3 4 2 1 2 2

2 3 3 4 2 2 3 2

Cells scored

0/zM a 1,6 2.0 0,3 0,3 1.0 0 0 0 3056

100 ~ M 10.2 c,d,e 5.1 d,e 0.9 d 2.6 d 6.0 c 0.9 0 d 0 1171

125/zM a 7.1 c,d,e 11.0 c,d,e 2.7 c,d 0 d 12.6 c 0 0.5 a 0.5 1822

Combined b 8.4 8.7 2.0 1.0 10.4 0.3 0.3 0.3

c,d,e c,d,e a d c a

2993

a Represents the pooled results of two separate experiments. No significant differences in the frequencies of nuclei with 3 or more hybridization regions were observed between the two experiments. b Represents the combined data from the 100-/zM and the 125-/~M treatments of hydroquinone. e Differs significantly from the corresponding control (0-/xM), p < 0.05, l-tailed Fisher exact test. d The hyperdiploid frequency for this chromosome (total from rows 1, 3 and 4 for chromosome 1 and rows 2 - 4 and 7 for chromosome 7) differs significantly from that of the corresponding control, p < 0.05, 1-tailed Fisher exact test. e The hyperdiploid frequency and the frequency of nuclei containing 3 copies of chromosome 1 does not differ significantly from those of chromosome 7, p > 0.05, 1-tailed Fisher exact test.

16

D.A. Eastmond et aL / Mutation Research 322 (1994) 9-20

Fig. 3. Multicolor fluorescence in situ hybridization with 3 centromeric DNA probes: (a) human metaphase lymphocyte following hybridization with a fluorescein-labeled alpha satellite probe for chromosome 7 (green color) and a Texas red-labeled classical satellite probe for chromosome 1 (red color) adjacent to a fluorescein-labeled alpha satellite probe for chromosome 1 (yellow color). The nuclear DNA was counterstained with DAPI (blue color). (b) A untreated human lymphocyte interphase nucleus following hybridization with the 3 probes described above. (c) A hydroquinone-treated interphase human lymphocyte exhibiting 3 copies each of the chromosome 7 probe and the two tandemly labeled chromosome 1 probes. (d) A hydroquinone-treated interphase human lymphocyte exhibiting 3 hybridization regions for the Texas red-labeled classical satellite probe for chromosome 1 but only 2 copies of the chromosome 1 alpha satellite probe and the chromosome 7 alpha satellite probe. This pattern indicates that a break has occurred in the region of chromosome 1 targeted by the classical satellite probe.

D,,4. Eastmond et al. / Mutation Research 322 (1994) 9-20

centration. An example of a nucleus exhibiting 3 copies of chromosomes 1 and 7 is shown in Fig. 3c. In addition, a significant number of nuclei containing two yellow hybridization regions adjacent to only two of 3 Texas red-labeled regions was observed indicating that breakage within the chromosomal region targeted by the classical satellite probe had occurred. An example of a nucleus containing a break within the lq12 region is illustrated in Fig. 3d. Breakage apparently accounted for the higher frequencies of nuclei containing 3 or more hybridization regions that were seen in the single-probe hybridizations as there was no significant increase in the frequency of nuclei hyperdiploid for chromosome 1 when compared to that of chromosome 7 (p > 0.05; 1-tailed Fisher exact test). To confirm the results at the 100-/zM concentration, nuclei on slides from the 125-~M hydroquinone concentration from the same experiment as well as an additional experiment were also hybridized using this triple-label approach (Table 2). As seen previously, significant increases in hyperdiploidy for chromosomes 1 and 7 as well as breakage within the chromosome 1 heterochromatin were also observed. No significant difference in the frequency of hyperdiploid nuclei was seen when comparing the results for chromosome 1 and 7. oC

25

N .~ _~

20

¢D

ts

~

Lge w/In cs 1 ~ J l o n esCSlor7 es of cs 1 a n d 7 es of cs 1 and 7

E ~

== o

o.

10

=_o

0 CS 1

Untreated

cS 7

CS 1

0S 7

Hydroqulnone-treated

Fig. 4. The classification of nuclei containing 3 or more hybridization regions following triple labeling with the chromosome 1 and 7 probes. The hydroquinone-treated results represent the combined frequencies of affected nuclei at the 100 and 125-/zM concentrations of hydroquinone presented in Table 2 whereas the untreated results are derived directly from the 0-/,~M data in that table. To simplify the presentation, only categories with 2 or more observations have been presented (i.e. rows 1-5 from Table 2).

17

The results of these triple-label multicolor FISH studies are summarized in Fig. 4. The resuits from the 100-~M and the 125-/xM hydroquinone concentrations have been combined to simplify the presentation of the results. These triple-label analyses confirm that hydroquinone induced a significant increase in nuclei with 3 or more hybridization regions and that significant differences in the frequency of these nuclei were observed when comparing the results obtained with the chromosome 1 and 7 probes. However, as is apparent from Fig. 4, the difference in hybridization frequencies using the chromosome 1 and 7 probes was due primarily to breakage within the region targeted by the chromosome 1 classical satellite probe. These results also indicate that hydroquinone induced a significant increase in breakage within the lq12 heterochromatin region. Discussion The results of these experiments indicate that the in vitro treatment of human lymphocytes with hydroquinone induces significant increases of both hyperdiploidy and chromosomal breakage. The clastogenic and aneuploidy-inducing effects of hydroquinone have recently been the subject of a number of investigations (Crebelli et al., 1987; Yager et al., 1990; Gudi et al., 1992; Adler, 1993; Natarajan, 1993). In general, our results are consistent with these reports in which hydroquinone has been shown to interfere with normal chromosome segregation and induce chromosomal breakage. However, in these reports the effectiveness of hydroquinone to induce numerical and structural aberrations varies considerably. Much of this variability is undoubtedly related to differences in test systems, laboratory procedures and scoring criteria. Another important factor to consider is the instability of hydroquinone which readily autoxidizes in aqueous solutions to 1,4benzoquinone (Greenlee et al., 1981). 1,4-Benzoquinone is capable of binding covalently to proteins and other macromolecules which are likely to be present in the test system (Lunte and Kissinger, 1983). Furthermore, we have also observed considerable variability in the chromosome-altering effects of hydroquinone on lympho-

18

D.A. Eastmond et al. / Mutation Research 322 (1994) 9-20

cytes isolated from various donors (Rupa and Eastmond, unpublished data). Although the effects of hydroquinone on chromosome breakage and chromosome segregation are observed in the lymphocytes of most donors, the magnitude of response can vary significantly from donor to donor. The lymphocytes used in these experiments were obtained from a donor whose lymphocytes were previously shown to be sensitive to the chromosome-damaging effects of hydroquinone (Yager et al., 1990). However, we have also detected significant increases in hyperdiploidy and breakage within the lq12 region induced by hydroquinone using lymphocytes from another donor. These experiments demonstrate that FISH with chromosome-specific DNA probes can be effectively used to investigate the chromosome-altering effects of a xenobiotic agent. One advantage of these techniques is that the hyperdiploidy-inducing effects of a chemical can be assessed rapidly on thousands of cells, allowing relatively weak aneuploidy-inducing chemicals to be detected. Although the ability of these FISH techniques to detect hyperdiploidy following chemical exposure in vitro has been previously reported (Pinkel and Eastmond, 1990; De Sario et al., 1990), our results also indicate that FISH techniques can be used to detect chromosomal breakage occurring in interphase cells following exposure to clastogenic agents. In this case we have used multicolor FISH with two adjacent probes to distinguish hyperdiploidy from chromosome breakage. The ability of the benzene metabolite hydroquinone to induce a significant increase in nuclei with a hybridization pattern characteristic of breakage, combined with the relatively small size of this region in comparison to the human genome, suggests that hydroquinone may be inducing nonrandom breakage within the heterochromatin region of chromosome 1. Additional studies from our laboratory have supported these observations. Following the in vivo administration of hydroquinone to mice, a highly non-random frequency of breakage within the heterochromatin regions of mouse chromosomes has been observed (Chen and Eastmond, unpublished resuits). As mentioned previously, awide variety of

chemical and physical agents have been shown to induce non-random breakage within the heterochromatin regions of human chromosomes 1 and 9 (Brogger, 1977; Eastmond and Pinkel, 1990). These results indicate that caution should be used in interpreting the results of FISH studies, particularly those using the classical satellite probes for human chromosomes 1 and 9. The ability of the benzene metabolite hydroquinone to induce chromosome breakage and aneuploidy suggests that this metabolite is likely to contribute to the structural and numerical aberrations that have been observed in the lymphocytes of workers following benzene exposure (Ding et al., 1983; Aksoy, 1988; Sasiadek, 1992; Eastmond, 1993). The occurrence of these chromosomal alterations at early, intermediate and late stages of benzene-induced bone-marrow dysfunction (Forni and Moreo, 1967; Erdogan and Aksoy, 1973; Ding et al., 1983; Aksoy et al., 1988), combined with the involvement of nonrandom structural and numerical aberrations in acute nonlymphocytic leukemias (Yunis, 1983; Rowley and Le Beau, 1989; Rowley, 1990) - - the cancers primarily associated with benzene exposure - indicate that these genetic alterations may play an important role in benzene-induced neoplastic development. In summary, the use of multicolor FISH with chromosome-specific DNA probes following the in vitro exposure of human lymphocytes to hydroquinone indicates that this benzene metabolite is capable of inducing both hyperdiploidy and chromosomal breakage. Additional studies are currently under way in our laboratory to investigate the mechanisms underlying hydroquinone's genotoxic effects and their role in the genotoxic and leukemogenic effects of benzene.

Acknowledgements We would like to acknowledge the assistance of Moire Robertson Creek with the lymphocyte culturing. This research was supported in part by funds provided by the Cigarette and Tobacco Surtax Fund of the State of California through the Tobacco-Related Disease Research Program

D.A. Eastmond et al. / Mutation Research 322 (1994) 9-20

of the University of California, grant number 1KT36 and the Universitywide Energy Research Group.

References Abacus (1987) StatView II, Abacus Concepts, Inc., Berkeley, CA. Adler, I.D. (1993) Synopsis of the in vivo results obtained with the 10 known or suspected aneugens tested in the CEC collaborative study, Mutation Res., 287, 131-137. Aksoy, M. (1988) Benzene Carcinogenicity, in: M. Aksoy (Ed.), Benzene Carcinogenicity, CRC Press, Boca Raton, FL, pp. 113-148. Brogger, A. (1977) Non-random localization of chromosome damage in human cells and targets for clastogenic action, Chromosomes Today, 6, 297-306. Ciranni, R., R. Barale and I.D. Adler (1991) Dose-related clastogenic effects induced by benzene in bone-marrow cells and in differentiating spermatogonia of Swiss CD1 mice, Mutagenesis, 6, 417-421. Crebelli, R., G. Conti and A. Carere (1987) On the mechanism of mitotic segregation induction in Aspergillus nidulans by benzene hydroxy metabolites, Mutagenesis, 2, 235238. Dean, B.J. (1985) Recent findings on the genetic toxicology of benzene, toluene, xylenes and phenols, Mutation Res., 154, 153-181. De Sario A., P. Vagnarelli and L. De Carli (1990) Aneuploidy assay on diethylstilbestrol by means of in situ hybridization of radioactive and biotinylated DNA probes on interphase nuclei, Mutation Res., 243, 127-131. Ding, X., Y. Li, Y. Ding and H. Yang (1983) Chromosome changes in patients with chronic benzene poisoning, Chinese Med. J., 96, 681-685. Eastmond, D.A. (1993) Induction of micronuclei and aneuploidy by the quinone-forming agents benzene and ophenylphenol, Toxicol. Lett., 67, 105-118. Eastmond, D.A., and D. Pinkel (1990) Detection of aneuploidy and aneuploidy-inducing agents in human lymphocytes using fluorescence in situ hybridization with chromosome-specific DNA probes, Mutation Res., 234, 303-318. Erdogan, G., and M. Aksoy (1973) Cytogenetic studies in thirteen patients with pancytopenia and leukaemia associated with long-term exposure to benzene, New Istanbul Contrib. Clin. Sci., 10, 230-247. Fearon, E.R., and B. Vogelstein (1990) A genetic model for colorectal tumorigenesis, Cell, 61, 759-767. Forni, A., and L. Moreo (1967) Cytogenetic studies in a case of benzene leukaemia, Eur. J. Cancer, 3, 251-255. Greenlee, W.F., J.D. Sun and J.S. Bus (1981) A proposed mechanism of benzene toxicity: Formation of reactive intermediates from polyphenol metabolites, Toxicol. Appl. Pharmacol., 59, 187-195.

19

Gudi, R., J. Xu and A. Thilagar (1992) Assessment of the in vivo aneuploidy/micronucleus assay in mouse bone marrow cells with 16 chemicals, Environ. Mol. Mutagen., 20, 106-116. Hansen, M., and W. Cavenee (1987) Genetics of cancer predisposition, Cancer Res., 47, 5518-5527. Huff, J.E., J.K. Haseman, D.M. DeMarini, S. Eustis, R.R. Maronpot, A.C. Peters, R.L. Persing, C.E. Chrisp and A.C. Jacobs (1989) Multiple-site carcinogenicity of benzene in Fischer 344 rats and B6C3F1 mice, Environ. Health Perspect., 82, 125-163. IARC (1987) Benzene, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs, Vols. 1-42, Suppl. 7, pp. 120-122. Lunte, S.M., and P.T. Kissinger (1983) Detection and identification of sulfhydryl conjugates of p-benzoquinone in microsomal incubations of benzene and phenol, Chem.-Biol. Interact., 47, 195-212. Lutz, W.K. (1986) Quantitative evaluation of DNA binding data for risk estimation and for classification of direct and indirect carcinogens, J. Cancer Res. Clin. Oncol., 112, 85-91. Margolin, B.H., and K.J. Risko (1988) The statistical analysis of in vivo genotoxicity data: Case studies of the rat hepatocyte UDS and mouse bone marrow micronucleus assays, in: J. Ashby, F.J. de Serres, M.D. Shelby, B.H. Margolin, M. Ishidate Jr. and G.C. Becking (Eds.), Evaluation of Short-Term Tests for Carcinogens, Vol. 1, Cambridge University Press, Cambridge, pp. 1.29-1.42. Meyne, J., L.H. Lockhart and F.E. Arrigh (1979) Nonrandom distribution of chromosomal aberrations induced by three chemicals, Mutation Res., 63, 201-209. Natarajan A.T. (1993) An overview of the results of testing of known or suspected aneugens using mammalian cells in vitro, Mutation Res., 287, 113-118. Oshimura, M., and J.C. Barrett (1986) Chemically induced aneuploidy in mammalian cells: Mechanisms and biological significance in cancer, Environ. Mutagen., 8, 129-159. Pinkel, D., T. Straume and J.W. Gray (1986) C~togenetic analysis using quantitative, high-sensitivity, fluorescence hybridization, Proc. Natl. Acad. Sci. (U.S.A.), 83, 29342938. Reddy, M.V., G.R. Blackburn, C.A. Schreiner, M.A. Mehlman and C.R. Mackerer (1989) 32p Analysis of DNA adducts in tissues of benzene-treated rats, Environ. Health Perspect., 82, 253-257. Rithidech, K., W.W. Au, V.M.S. Ramanujam, E.B. Whorton Jr. and M.S. Legator (1987) Induction of chromosome aberrations in lymphocytes of mice after subchronic exposure to benzene, Mutation Res., 88, 135-140. Rowley, J.D. (1990) Molecular cytogenetics: Rosetta stone for understanding cancer - - twenty-ninth G.H.A. Clowes memorial award lecture, Cancer Res., 50, 3816-3825. Rowley, J.D., and M.M. Le Beau (1989) Cytogenetic and molecular analysis of therapy-related leukemia, Ann. N.Y. Acad. Sci., 567, 130-140.

20

D.A. Eastmond et al. / Mutation Research 322 (1994) 9-20

Sasiadek, M. (1992) Nonrandom distribution of breakpoints in the karyotypes of workers occupationally exposed to benzene, Environ. Health Perspect., 97, 255-257. Solomon, E., J. Borrow and A.D. Goddard (1991) Chromosomal aberrations and cancer, Science, 254, 1153-1160. Tice, R,R., D.L. Costa and R.T. Drew (1980) Cytogenetic effects of inhaled benzene in murine bone marrow: Induction of sister chromatid exchanges, chromosomal aberrations, and cellular proliferation inhibition in D B A / 2 mice, Proc. Natl. Acad. Sci. (U.S.A.), 77, 2148-2152. Tkachuk, D.C., C.A. Westbrook, M. Andreeff, T.A. Donlon, M.L, Cleary, K. Suryanarayan, M. Homage, A. Redner, J. Gray and D. Pinkel (1990) Detection of bcr-abl fusion in chronic myelogenous leukemia by in situ hybridization, Science, 250, 559-562.

Trask, B., and D. Pinkel (1990) Fluorescence in situ hybridization with DNA probes, Meth. Cell Biol., 33, 383-400. Waters, M.D., H.F. Stack, A.L. Brady, P.H.M. Lohman, L. Haroun and H. Vainio (1988) Use of computerized data listings and activity profiles of genetic and related effects in the review of 195 compounds, Mutation Res., 205, 295-312. Wolman, S.R. (1977) Cytologic and cytogenetic effects of benzene, J. Toxicol. Environ. Health, Suppl. 2, 63-68. Yager, J.W., D.A. Eastmond, M.L. Robertson, W.M. Paradisin and M.T. Smith (1990) Characterization of micronuclei induced in human lymphocytes by benzene metabolites, Cancer Res., 50, 393-399. Yunis, J.J. (1983) The chromosomal basis of human neoplasia, Science, 221, 227-236.