Physical,chemical, and biological factors affecting sister-chromatid exchange induction in human lymphocytes exposed to mitomycin C prior to culture

In experiments to assess the effects of several biological, chemical, and physical variables on sister-chromatid exchange (SCE) induction in cultured lymphocytes exposed to mitomycin C (MMC) before PHA stimulation we observed: (1) high SCE frequencies in female cells, and normal SCE frequencies in Y-bearing metaphases in mixed cultures containing equal numbers of MMC-treated female lymphocytes and untreated male lymphocytes; (2) small, but statistically significant, decreases in SCEs with increasing pH after G₀ exposure in the pH range 6.6–7.6; (3) pronounced reductions in MMC-induced SCEs in lymphocytes exposed at 4°C vs. 37°C. In other studies, SCE induction was evaluated in cultures exposed during G₀ to MMC concentrations ranging from 0.25 to 2.5 μg/ml for varying time intervals ranging from 5 min to 24 h. For all concentrations tested SCE induction varied as a linear function of G₀ exposure time. To compare SCE induction between cultures, we calculated the mean frequencies of SCEs induced per metaphase/unit dose MMC/unit G₀ exposure time (SCE/μg/h). A mean frequency of 20.7 ± 4.8 SCE/μg/h was observed for 41 lymphocyte cultures suggesting that a single term adequately describes the rate of SCE induction following G₀ exposure to a 10-fold range in concentration of MMC for time intervals of 30 min to 24 h.     

Induction of sister-chromatid exchanges by 2-aminofluorene in cultured human lymphocytes with and without erythrocytes.

2-Aminofluorene (2-AF), an indirect mutagen reported to be metabolically activated by erythrocytes in the Salmonella mutagenicity test, was studied for the induction of sister-chromatid exchanges (SCEs) in human lymphocytes in vitro with (whole-blood cultures) and without erythrocytes (isolated lymphocyte cultures). 2-AF (0.025-0.8 mM) was present in the cultures for the last 48 h of 72-h cultures. In both types of culture, SCEs increased in a dose-dependent manner, with a statistically significant elevation already at the lowest concentration of 2-AF tested and maximum responses of 2.4-fold (whole blood) and 2.1-fold (isolated lymphocytes), in comparison with mean SCEs/cell in control cultures, at 0.4 and 0.2 mM concentrations (respectively). Thus, the induction of SCEs by 2-AF was not dependent on the presence of erythrocytes. Styrene (2 mM), a positive control chemical known to require erythrocytes for efficient SCE induction in vitro, was shown to produce a 4.9-fold increase in SCEs in whole-blood cultures, but only a slight (1.3-fold) effect in isolated lymphocyte cultures. The results suggest that leukocytes, but not erythrocytes, are important in the metabolic activation of 2-AF in the human lymphocyte SCE assay.     

Sister-chromatid exchanges induced by vinyl esters and respective carboxylic acids in cultured human lymphocytes.

Vinyl acetate--an efficient inducer of sister-chromatid exchanges (SCEs)--is known to be hydrolyzed in mammalian cells into acetic acid and acetaldehyde, the latter being the likely metabolite responsible for the SCE induction. As similar hydrolysis to acetaldehyde and to a carboxylic acid is also expected for other vinyl esters, five such compounds--vinyl formate, vinyl chloroformate, vinyl propionate, vinyl crotonate and vinyl-2-ethylhexanoate--and five carboxylic acids--formic acid, acetic acid, propionic acid, crotonic acid and 2-ethylhexanoic acid--were tested for their ability to induce SCEs in cultured (72 h) human lymphocytes with a 48-h treatment, starting at 24 h after culture initiation. Vinyl formate, vinyl propionate and vinyl crotonate induced a clear dose-dependent increase in the number of SCEs/cell at concentrations of 0.125-0.5 mM and vinyl chloroformate at 0.063-1 mM, i.e., at roughly the same concentration range as vinyl acetate and acetaldehyde. Vinyl-2-ethylhexanoate required slightly higher concentrations (0.25-4 mM) for SCE induction. All of the carboxylic acids tested also elevated SCEs, but only slightly. Formic acid and crotonic acid produced some SCE increase at a concentration of 10 mM, acetic acid at 5 and 10 mM and propionic acid at 2.5 mM. 2-Ethylhexanoic acid induced SCEs at a lower concentration range (0.63-2.5 mM) than the other acids. The positive concentrations of the first three carboxylic acids lowered the pH of the culture medium immediately after the treatment by 0.5-1.0 pH unit (lowest observed pH 6.53). The pH differences from the control cultures became smaller in measurements done 24 h and 48 h after the beginning of treatment. Propionic acid and 2-ethylhexanoic acid affected medium pH only slightly (maximum drop 0.2 pH units) at the concentrations that induced SCEs. The results lend support to the idea that the efficient SCE induction observed with the vinyl esters results from the formation of acetaldehyde, with carboxylic acids--with the possible exception of 2-ethylhexanoic acid--playing no significant role. The slight SCE induction obtained with the carboxylic acids cannot be explained by lowered pH alone.     

Developmental and cytogenetic effects of caffeine on mouse blastocysts, alone or in combination with benzo(a)pyrene

Mouse blastocysts were treated with caffeine and/or benzo(a)pyrene (BP), and the effects on development and on induction of sister chromatid exchanges (SCEs) were examined. Caffeine interfered with blastocyst development in a dose-related manner. At 4 mM, the highest concentration tested, caffeine interfered with development of blastocysts to all four endpoints: hatching, trophoblast outgrowth, inner cell mass (ICM) growth, and two-layer (primary endoderm and ectoderm) differentiation of ICMs. At 2 mM, caffeine reduced the incidence of both ICM growth and differentiation but did not affect hatching or formation of trophoblast outgrowths. At 1 mM, caffeine interfered only with ICM differentiation. Cell proliferation was least sensitive to caffeine and was reduced at concentrations of greater than or equal to 2 mM. Induction of SCEs was most sensitive to caffeine exposure; an increase in SCE frequency was observed at 0.1 and 0.5 mM. When caffeine was added to cultures with BP (1 microM, a concentration that was not embryotoxic and did not induce SCEs), both embryotoxic effects and SCE frequency were increased. The enhancing effect on SCE induction was particularly marked; as little as 0.1 mM caffeine was sufficient to cause doubling of induced SCE frequencies when added to cultures with BP.     

BCNU-induced sister chromatid exchanges are increased by X irradiation

We have studied the effect on sister chromatid exchange (SCE) induction in 9L rat brain tumor cells caused by combination treatment with BCNU and X rays. Over the dose and concentration ranges used in these experiments, BCNU induced relatively large numbers of SCEs, while X rays induced few SCEs. When cells were X irradiated immediately after BCNU treatment, the number of SCEs induced was greater than the number of SCEs expected by adding the number of SCEs induced by each agent alone; the number of SCEs induced as a result of this BCNU-X-ray interaction increased as the concentration of BCNU and/or dose of X rays increased. When the addition of bromodeoxyuridine was delayed from 0 to 16 hr after BCNU treatment, the number of SCEs induced declined to control levels by 16 hr. If X irradiation was delayed for up to 16 hr after BCNU treatment the same pattern of decrease was observed; the number of SCEs induced at each time point, however, was greater than that induced by BCNU and X rays alone. X irradiation from 0-16 hr before BCNU treatment produced the same number of SCEs as that produced by BCNU alone. Thus the SCE assay is capable of detecting a drug-X-ray interaction in mammalian cells and provides a sensitive means of studying the sequencing and timing that leads to the interaction.     

Effect of sodium selenite and methyl methanesulfonate or N-hydroxy-2-acetylaminofluorene co-exposure on sister-chromatid exchange production in human whole blood cultures.

Sodium selenite (Na2SeO3) was tested for its sister-chromatid exchange (SCE)-inducing ability in human whole blood cultures and for the effect of its co-exposure with methyl methanesulfonate (MMS) or N-hydroxy-2-acetylaminofluorene (N-OH-AAF) on SCE frequency. Long exposure times (77 h and 96 h) to 3.95 X 10(-6) M Na2SeO3 resulted in cell death as measured by mitotic indices, but mitotic figures were present after exposure to higher concentrations for a shorter time (19 h). High Na2SeO3 concentrations (7.90 X 10(-6) and 1.19 X 10(-5) M) resulted in a three-fold increase in the SCE frequency above background level (6--7 SCEs/cell). Exposure of lymphocytes to 1 X 10(-4) M MMS for the last 19 h of culture yielded an average SCE frequency of 30.17 +/- 0.75 while a similar exposure to 2.7 X 10(-5) M N-OH-AAF resulted in 13.61 +/- 0.43 SCEs/cell. Simultaneous addition of the high Na2SeO3 concentrations and MMS or N-OH-AAF to the cultures resulted in SCE frequencies that were 25--30% and 11--17%, respectively, below the sum of the SCE frequencies produced by the individual compounds.     

SCE evaluations in human lymphocytes after G0 exposure to mitomycin C Lack of expression of MMC-induced SCEs in cells that have undergone greater than two in vitro divisions

We conducted a series of experiments designed to determine whether DNA damage induced in G₀ lymphocytes by mitomycin C (MMC) would be expressed as sister-chromatid exchanges during the second and third post-treatment cell cycles. Lymphocytes from normal donors were exposed to MMC for 2 h prior to culture in the presence of phytohemagglutinin. MMC-treated and control cells were subsequently exposed to bromodeoxyuridine (BrdUrd) for the entire culture period (i.e. 48 h or 72 h) or for the terminal 24 h of 72-h cultures. We observed a 3–4-fold increase in SCEs in MII metaphases from lymphocytes treated with MMC and cultured in the presence of BrdUrd for the entire culture period. In contrast, in replicate cultures of MMC-treated lymphocytes that were exposed to BrdUrd for the terminal 24 h only, the SCE frequency in uniformly harlequinized metaphases was not significantly different from that observed in control cultures. We interpret these data as providing evidence that MMC-induced lesions (or alterations) in the DNA of G₀ lymphocytes are probably expressed as SCEs during the first period of mitogeninduced DNA synthesis, and that these lesions do not persist and give rise to SCEs in subsequent cell divisions.     

Do the frequencies of sister chromatid exchanges in endoreduplieated mitoses provide a measure for lesion persistence and repair?

Endoreduplication was induced in V 79 cells using Colcemid. The concentration of Colcemid necessary to induce endoreduplication is about 1000 times higher than that needed to arrest mitoses or to induce ordinary tetraploid cells. Diplochromosomes with sister chromatid differentiation were obtained by adding BrdU for the duration of one cell cycle prior to the induction of endoreduplication. The induction of endoreduplication with Colcemid had no influence on the frequency of sister chromatid exchanges (SCEs). Treating the cultures with mitomycin C (MMC) before adding BrdU increased the percentage of endoreduplieated mitoses and also led to marked SCE induction. In the diplochromosomes, the frequencies of both twin SCEs (first cycle) as well as single SCEs (second cycle) were increased. It was also found that the SCE frequencies in mitoses after endoreduplication were lower than the values found in diploid and ordinary tetraploid metaphases of the same preparation. The possible conclusions concerning the lifetime of SCE-inducing lesions and the influence of repair processes are discussed.     

Induction of sister-chromatid exchanges in Chinese hamster ovary cells by the biotic ketoaldehyde methylglyoxal

The number of sister-chromatid exchanges (SCEs) per metaphase was determined in Chinese hamster ovary cells after 16 h exposure to methylglyoxal (MG) concentrations ranging from 0.1 to 0.75 mM. MG produced an increase of SCE frequency that proved to be dose-dependent, and to reach a maximum of 2 X baseline at the highest nontoxic concentration (0.5 mM).     

Determination of the baseline sister chromatid exchange frequency in human and mouse peripheral lymphocytes using monoclonal antibodies and very low doses of bromodeoxyuridine

We measured the frequency of sister chromatid exchanges (SCEs) in human and mouse peripheral lymphocytes using doses of bromodeoxyuridine (BrdU) ranging from 30 nM to 100 microM (human) and from 10 nM to 10 microM (mouse). Heparinized peripheral blood was obtained from five healthy nonsmokers and from six C57B1/6 male mice. The blood was stimulated with PHA (human) or lipopolysaccharide (LPS, mouse) and grown for the first of two cell cycles in BrdU. Metaphase chromosomes were denatured and exposed to a monoclonal antibody reactive to single-stranded DNA containing BrdU. A second antibody was used to label the first antibody with fluorescein, and propidium iodide was used as a counterstain. Second-division metaphases were thus differentially stained red to indicate DNA content and yellow-green to indicate the presence of BrdU. The results indicate that the baseline SCE frequency in human and mouse peripheral lymphocytes is 3.6 and 2.4 SCEs per cell per generation, and that in the human these frequencies are invariant at the lowest BrdU levels. This suggests that SCEs are an integral part of DNA replication, even in the absence of agents known to induce SCEs. The distribution of SCEs per chromosome was analyzed and found to be Poisson-distributed in all 24 murine cultures and in 25 of 36 human cultures. The distribution of SCEs per chromosome may be due to either species-specific chromosome packaging or to karyotypic differences between the species.     

Chromosomal aberrations in patients with primary biliary cirrhosis

Summary Chromosomal aberrations in untreated lymphocyte cultures, bleomycin (BLM)-induced aberrations and sister chromatid exchanges (SCE) in the peripheral blood lymphocytes of 11 patients suffering from primary biliary cirrhosis (PBC) and 14 matched control individuals were analysed. The lymphocytes of the PBC patients had on average a lower mitotic index (2.3) compared with controls (3.5) in the untreated cultures. The mean baseline rate of aberrations of the cultured lymphocytes of the patients was 5.3 aberrations per 100 metaphases (%); this was significantly different (P=0.0291) from that of the controls with a mean of 2.3%. In lymphocytes of the patients and controls, most of the aberrations observed took the form of gaps; there was an almost equal breakage rate in both groups (0.5% and 0.4%, respectively). The average number of mitoses with aberrations in the PBC patients studied was double that of the controls (4.9% and 2.3% respectively, P=0.0323). The mean number of the BLM-induced aberrations was 54.0% and 27.7% for the lymphocytes of the patients and controls, respectively. The mean number of the aberrant mitoses in the BLM cultures was 6 times higher than that of the untreated cultures for both groups, 25.7% and 14.6% respectively (P=0.018). The chromosomal distribution of baseline and induced aberrations was not random. The PBC patients had a mean number of 8.7 SCE per mitosis, which was significantly higher than the SCEs in the controls (6.3 SCE per mitosis; P=0.0156). The evidence suggests that the chromosomes of the lymphocytes of PBC patients may be less stable than those of the control individuals in this study.     

Sister-chromatid exchange in childhood in relation to age and sex

Small children have been found to have a lower SCE/cell than adults and in recent reports females have had higher SCEs/cell than males. We here describe the relationship between SCE/cell and age and sex in 46 girls and 39 boys with an age range of 1.4-19.2 years and 2.6-18.7 years, respectively. For the calculation a transformation y = (sum SCE)1/2 + (sum SCE + 1)1/2 was used. The best fit to our material was represented by the equation y = b0 + b1 X log age. A common slope (b1) could be used for the boys and girls. This slope was significantly different from zero (P less than 0.0005). The levels of the regression lines for the two sexes were different (P = 0.0006). The girls had a 0.55-0.7 higher SCE/cell than the boys, depending on age. The following equations were found: Girls: y = 22.49 + 6.53 X log age. Boys: y = 21.11 + 6.53 X log age. By this model 43% of the variation in y could be explained. As a consequence of the result it is absolutely essential, when planning studies of children, to use age-matched groups to decrease the variability of the test system.     

Failure to produce greater than additive sister chromatid exchange induction with X rays and BCNU

In 1984, Tofilon and Deen reported that X irradiation of 9L cells immediately after treatment with 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) produced greater than additive induction of sister chromatid exchanges (SCEs) compared to the sum of SCEs induced by each agent alone [Radiat. Res. 97, 171-177 (1984)]. We (C.P.S., D.F.D.) repeated experiments conducted by Tofilon and Deen, scored SCEs in a blind manner, and were unable to reproduce the original findings. Instead, we found that X rays and BCNU induced SCEs in an additive manner. The slides prepared in the original study are extant; they were coded and recounted blind. Data obtained in this reevaluation do not substantiate the previous report. We conclude that the original findings were apparently the result of bias introduced in the SCE scoring process. Our experience with the SCE assay emphasizes the need to recode SCE slide preparations and count SCEs blind.     

SCE variability in lymphocytes and fibroblasts

Summary To determine whether the sister chromatid exchange (SCE) distributions obtained in lymphocytes and fibroblasts from different individuals are comparable, a controlled study was set up. Peripheral blood and skin biopsies were taken on the same day from five individuals living for years under the same environmental conditions. All samples were treated in the same fashion, and the SCEs were scored in 50 metaphases of peripheral blood lymphocytes and of skin fibroblasts in an early and in a late passage. A repeat blood sample was taken from the same five indivuduals 1 year later. Based on the results obtained in this first part of the study, five randomly chosen healthy blood donors were sampled at different times and studied in the same fashion. Each chromosome was identified, and the SCE scores were tabulated per chromosome over 50 metaphases. The statistical analysis consisted of fitting log linear models to these scores and examining the best fit by determining the exceedance probabilities (observed significance level). For lymphocytes, the results indicated that the SCE distributions depended only on the chromosome examined, and not on BrdU-exposure time, individuals, or time of sampling. Treatment with ethyl methane sulfonate (EMS) increased the number of SCEs proportionally on all chromosomes. Analysis of the SCE scores on lymphocytes and fibroblasts of the five individuals and on their low and high passage fibroblast cultures revealed the necessity of including higher order interactions in order to fit a suitable model to the data. Therefore comparison of the SCE scores of lymphocytes with those of fibroblasts or comparison of scores on fibroblasts from different individuals could not be done. In practice, to compare samples or individuals, it suffices to score the SCE on a limited number of chromosomes (e.G., the A group) of 50 metaphases.     

Demonstration of sister chromatid differentiation in human amniotic fluid cells after partial synchronization with BrdU or dT surplus

Summary Protocols are compared demonstrating sister chromatid differentiation (SCD) in human amniotic fluid (AF) cells with and without partial synchronization. Partial synchronization both with an excess of 5-bromodeoxyuridine (BrdU) and an excess of thymidine leads to an increase of metaphases with SCD. Compared with unsynchronized cells, the rate of sister chromatid exchanges (SCE) is not increased. Studies on the late replicating X chromosome of female cells showed that the addition of mitomycin C (MMC) after releasing the thymidine block preferentially induces SCEs in late replicating regions. The partial synchronization with thymidine surplus provides a good basis for SCE experiments with AF cells and facilitates the prenatal diagnosis of diseases characterized by changes in the SCE rate.     

Sister-chromatid exchanges and cell-cycle kinetics in human lymphocyte cultures exposed to alkylating mutagens: apparent deformity in dose-response relationships

Experiments have been carried out using human whole-blood cultures to determine the effects of sampling times and of the duration of 5-bromodeoxyuridine (BrdUrd) treatment before fixation on sister-chromatid exchange (SCE) frequencies following exposure to mitomycin C (MMC). Cells were pulse treated for 1 h with 3 X 10(-6) M MMC at G1, and then sampled at 4-h intervals up to 88 h after stimulation of cultures with phytohemagglutinin (PHA). Results showed that this MMC treatment induced a 5-6 h proliferation delay per cell cycle, and that SCE frequencies first increased with time of fixation, peaking at 68 h, and then decreased. When cells were similarly treated with MMC, but subsequently exposed to BrdUrd for various times before fixation of cultures at 72 h, the SCE frequencies markedly increased with increasing durations of BrdUrd incubation times. These data indicate that, in mutagen-treated cultures, lymphocytes having relatively longer cell-cycle times show a higher mean frequency of SCEs. In a subsequent experiment, cells were treated for 1 h with increasing doses of MMC or 4-nitroquinoline 1-oxide (4NQO) at 0, 24, or 48 h, and then fixed at 72 h after PHA stimulation. Results showed that the optimal treatment times at which the agents could most efficiently produce SCEs were different for MMC and 4NQO, and that the dose-response curves tended to 'bend down' at very high doses; that is, treatments with very high doses induced smaller than expected numbers of SCEs. However, cells similarly treated with very high doses showed a higher, expected frequency of SCEs when sampled at 84 h, but again had a lower than expected SCE frequency when fixed at 96 h. The results indicate that there is an optimal time for sampling at which one can observe the maximum increase in SCE frequencies following mutagen exposure, and strongly suggest that the higher the dose, the later the optimal sampling time. Because of the apparent deformity of dose-response curves obtained after various treatments and sampling times, it seems necessary that extra fixation-time points be included in test protocols so as to avoid false negatives or confirm possible positives.     

Cell-stage dependence of mutagen-induced sister-chromatid exchanges in human lymphocyte cultures

The induction of sister-chromatid exchanges (SCEs) was studied in phytohemagglutinin (PHA)-stimulated human lymphocytes exposed for 1 h to mitomycin C (MMC, 3 X 10(-6) M), ethyl methanesulphonate (EMS, 2 X 10(-2) M), or 4-nitroquinoline-1-oxide (4NQO, 3 X 10(-5) M) at various cell-cycle stages of 72-h cultures. The doses of the chemical were chosen to give about 20 SCEs per cell when treated at Go. The SCE frequency increased almost linearly with MMC or EMS treatments at later times after PHA stimulation, peaking with those at 36 h (at around the first G1/S boundary in the 2 consecutive cell cycles, which was revealed by concomitant experiments), and then decreased with subsequent treatment times. Cell-cycle kinetics and the cell stages at which the cells were treated were measured by autoradiography and sister-chromatid differential staining. The data show that MMC and EMS produce larger numbers of SCEs when treated at stages closer to the beginning of S, and that the most efficient time of treatment is the G1/S boundary in the first cell cycle of the two consecutive cycles before sampling. Pulse treatment with EMS caused about 3 times larger inductions of SCEs when done at late G1/early S(G1/S boundary) in the first cell cycle compared to that at G0/early G1, whereas identical exposure to MMC at the first G1/S boundary produced only 1.5 times larger numbers of SCEs than that at G0/early G1. EMS and MMC both, however, induced 30-40% larger numbers of SCEs when treated at the G1/S boundary in the first cell cycle than when treated at the second cell cycle before sampling. On the contrary, treatment with 4NQO led to the induction of about the same numbers of SCEs even when treated at different cell-cycle stages before the second G1/S boundary. The SCE frequency in 4NQO-treated cells then decreased with subsequent treatment times.     

Cytogenetic evaluations in human lymphocytes exposed to methyl acetimidate, a lysine-specific protein crosslinking agent

The cytogenetic effects of methyl acetimidate (MAI), a lysine-specific protein crosslinking reagent, were investigated using human peripheral lymphocytes in culture. Lymphocytes were treated with the chemical either prior to PHA exposure or 2-3 days following mitogenic stimulation and assessed for perturbations in cellular proliferation and induction of SCEs. Severe reductions in the mitotic index (MI) and pronounced decreases in the proportion of metaphases proceeding beyond M(1) were observed following G0 exposure to MAI concentrations of as low as 2 mM; with complete suppression of mitotic activity in all cultures exposed to levels of 3 mM MAI or greater. Concentrations resulting in severe depression in MI caused only moderate increases in SCEs. Cells exposed to less than 10 mM MAI during the late S-G2 stages of the cell cycle and harvested at the first metaphase following treatment exhibited profound mitotic delay, impaired prophase to metaphase transitions and abnormal mitotic configurations. These findings demonstrate that protein-specific crosslinking agents may induce a wide spectrum of adverse cytogenetic outcomes in both cycling and noncycling lymphocytes.     

Sister chromatid exchanges in bromine incorporation into DNA cytosine nucleotides. III. 5-bromodeoxycytidine as a DNA thymine precursor. The characteristics of the exchanges]

Cell movement through the mitotic cycle and sister chromatid exchanges (SCE) were studied in human blood lymphocytes cultured in the presence of 5-bromodeoxycytidine (BrdC, 0.05 mM) plus thymidine (dT 0.4, 0.8, and 1.0 mM). In controls, lymphocytes were cultivated in the presence of 5-bromodeoxyuridine (BrdU, 0.05 mM) and deoxycytidine (0.1 mM), or BrdC alone. All nucleosides were added to the cultures 28 hours prior to fixation and were maintained in the medium for 16 hours. As determined from percentage of metaphases of 1st to 3rd divisions, BrdC did not release from thymidine block. This fact leads us to conclude that BrdC in contrast to deoxycytidine does not serve as a cytosine precursor. No significant differences in the frequency of SCE and their distribution among chromosomes were found between cultures treated with BrdC and with BrdU.     

Cytogenetic damage in workers exposed to ethylene oxide

Sister-chromatid exchanges (SECs) and chromosomal aberrations (CAs) were detected in the peripheral lymphocytes of 41 sanitary workers exposed to ethylene oxide (EO) in the sterilizing units of 8 hospitals in the Venice Region. The first group (19 workers) was exposed to 10.7 +/- 4.9 ppm EO, expressed as the time-weighted average concentration for an 8-h working day (TWA/8 h conc.), and the second group (22 workers) to 0.35 +/- 0.12 ppm. Each exposed worker was paired with a control of similar age and smoking habits. A highly significant (P less than 0.001) increase in the mean frequency of SCEs was found in the higher exposure group, 14 (74%) exposed subjects having significantly increased levels of SCEs compared to their matched controls. In the lower exposure group, the increase in mean frequency of SCEs was lower, though still significant (P less than 0.05): 7 (33%) exposed subjects had higher and 1 (5%) had a lower SCE level than the matched controls. From the first group, 10 subjects, 7 of whom had increased SCE levels, were reanalysed 12-18 months after their exposure had been lowered or interrupted: in only 2 of them the SCE level was significantly decreased. A statistically significant correlation between SCE frequency and level of EO exposure (TWA/8 h conc.), as well as a multiple correlation between SCE level and EO exposure, smoking and age were found. However, no interaction could be detected between EO exposure and smoking in the induction of SCEs. In controls, SCE frequency was correlated with smoking and age. In the higher exposure group, the number of both chromatid- and chromosome-type aberrations, independent of gaps, was significantly increased, whereas in the lower exposure group only the frequency of chromosome-type aberrations, excluding gaps, was statistically higher than in controls. The level of CAs remained to a great extent unchanged in the 10 subjects re-examined at a later stage after lowering or halting exposure. Taking the group as a whole, the frequency of cells with total CAs was found to be weakly (P = 0.05) correlated with EO exposure, and was not correlated with smoking, age or SCE frequency.