- Email: [email protected]
Assessment of the in vivo mutagenicity of ethylene oxide in the tissues of B6C3F1 lacI transgenic mice following inhalation exposure
Mutation Research 391 Ž1997. 153–164
Assessment of the in vivo mutagenicity of ethylene oxide in the tissues of B6C3F1 lacI transgenic mice following inhalation exposure Susan C. Sisk 1, Linda J. Pluta, Kathy G. Meyer, Brian C. Wong, Leslie Recio
)
Chemical Industry Institute of Toxicology, 6 DaÕis DriÕe, Research Triangle Park, NC 27709, USA Received 7 October 1996; revised 10 February 1997; accepted 12 February 1997
Abstract In the present study, the lacI mutant frequency was determined in the tissues of B6C3F1 lacI transgenic mice exposed by inhalation to ethylene oxide ŽEO.. Groups of 15 male transgenic lacI B6C3F1 mice were exposed to either 0, 50, 100, or 200 ppm EO for 4 weeks Ž6 hrday, 5 daysrweek. and were sacrificed at 0, 2, or 8 weeks after the last EO exposure. The lacI transgene was recovered from lung, bone marrow, spleen, and germ cells for determination of the lacI mutant frequency. The tissues selected for analysis were tumor target site tissues in chronic bioassays Žlung tumors and lymphomas. and germ cells. The lacI mutant frequency in lung was significantly increased at 8 weeks post exposure to 200 ppm EO Ž6.2 " 2.2 vs. 9.1 " 1.5, p s 0.046.. In contrast, the lacI mutant frequency in spleen and bone marrow at 2 and 8 weeks was not significantly increased in mice exposed to 200 ppm EO. The lacI mutant frequencies in male germ cells for 200 ppm EO-exposed mice were not increased compared to air controls at 2 and 8 weeks post-exposure. In a spleen cell fraction two of three EO-exposed mice at the 200 ppm exposure level demonstrated an elevated lacI mutant frequency. The increased lacI mutant frequency in these animals was likely due to mutant siblings that contained background G:C ™ A:T transitions at CpG sites. These results demonstrate that a 4-week inhalation exposure to EO is mutagenic in lung. However, EO did not increase the frequency of mutations recovered at the lacI transgene in other tissues examined under the conditions used in the present studies. Since the mutational spectrum for EO in other systems consists of an increased proportion of large deletions, the lack of a mutagenic response in the tissues examined is likely due to the lack of recovery of large deletions in lambda-based shuttle vector systems. These data indicate that a primary mechanism of EO-induced mutagenicity in vivo is likely through the induction of deletions, not specific point mutations. Keywords: Ethylene oxide; lacI; In vivo mutation; Mutant frequency; Mutation spectrum; Inhalation exposure
1. Introduction ) Corresponding author. Tel.: q1 Ž919. 5581329; Fax: q1 Ž919. 5581300; e-mail: [email protected] 1 Present address: ClinTrials Research Inc., P.O. Box 13991, Research Triangle Park, NC 27709, USA.
Ethylene oxide ŽEO; CAS No. 75-21-8. is a major industrial intermediate and gaseous sterilant regulated by the US Environmental Protection Agency
1383-5718r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 1 3 8 3 - 5 7 1 8 Ž 9 7 . 0 0 0 6 3 - 6
154
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
ŽEPA. as a hazardous air pollutant under the Clean Air Act w1x. EO reacts directly with cellular macromolecules, including DNA, causing gene mutations and chromosome deletions in somatic cells. Studies with Bacillus subtilis, Escherichia coli, Salmonella typhimurium, and Chinese hamster ovary cells demonstrated that EO does not require metabolic activation to induce gene mutations Žreviewed in w21x.. Intraperitoneal injection of EO increases bone marrow micronuclei in rats and mice w3x and is mutagenic at hprt in the T-lymphocytes of B6C3F1 mice w4x. Increased levels of sister chromatid exchange have been observed in peripheral lymphocytes of rodents and monkeys after inhalation of EO w5–9x. In biomonitoring studies, EO-exposed workers exhibit increased frequencies of several genotoxic end points relative to unexposed workers Žreviewed in w10x.. In 2-year inhalation bioassays, EO induced exposure-related increases in gliomas, peritoneal mesotheliomas, and mononuclear leukemias in F344 rats and lymphomas and adenomasrcarcinomas of the lung, Harderian gland, and mammary gland of B6C3F1 mice w11,12x. In human epidemiological studies reviewed by the International Agency for Research on Cancer ŽIARC., mortality from lymphatic and hematopoietic cancer was marginally elevated w10x. IARC has classified EO as a Group 1 chemical Žcarcinogenic to humans. w10x. This classification was based on limited evidence in humans for carcinogenicity, sufficient evidence in experimental animals and studies demonstrating in vivo genotoxicity in humans and rodents. In addition to somatic cell genotoxicity, EO can induce transmissible genetic damage in germ cells. Intraperitoneal injection of EO induced dominant lethal mutations and reciprocal translocations in postmeiotic germ cells from male mice w13x. Related studies with inhalation-exposed mice showed that the dominant lethal response increased in a nonlinear, exposure-related manner at high exposure levels; when total dose was kept constant, higher responses were measured in animals that received high concentrations for a short duration w14,15x. DNA adduct experiments in testes suggest that this exposure rate effect is due to concentration-dependent differences in internal dose, possibly resulting from respiratory rate attenuation during the longer low-dose expo-
sures w16x. Significant increases in abnormal sperm morphologies have also been reported in male mice after inhalation exposure to EO w17x. A US EPA genetic risk assessment of EO is presented in a series of papers in an issue of EnÕironmental and Molecular Mutagenesis w2,15,18x. A reconsideration of this risk assessment for EO has been proposed that takes into account aspects of effective dose to the target tissue, mechanisms of chromosomal alterations, and the shape of the dose–response curve at low doses w19x. Transgenic animal in vivo mutation models permit the determination and molecular analysis of mutation in marker genes from the tissues of exposed animals w20x. With the introduction of transgenic mice, the induction of mutations can be determined in multiple tissues with respect to exposure, dose to target tissue, DNA adduct levels, and additional genotoxicity end points. In the study presented here, we used B6C3F1 lacI transgenic mice ŽBig Blue e . to assess the mutagenicity of a 4-week inhalation exposure to EO. This mouse was selected for study because the B6C3F1 host strain is the strain used in the National Toxicology Program ŽNTP. carcinogenicity bioassay of EO w12x. The tissues selected for analysis of the in vivo lacI mutant frequency were tumor target site tissues in chronic bioassays Žlung tumors and lymphomas. w12x and male germ cells. In this article, we report the in vivo lacI mutant frequencies determined in lung, bone marrow, spleen, and seminiferous tubule germ cells of B6C3F1 lacI transgenic mice exposed to EO by inhalation at 0, 50, 100 or 200 ppm Ž6 hrday, 5 daysrweek for 4 weeks.. We also determined lacI mutant frequencies in a T-cell enriched fraction of spleen cells Žcalled spleen cell fraction herein. from air control and EO-exposed animals as a comparison to hprt mutant frequencies determined concurrently in these same animals w21x. DNA sequence analysis was performed on lacI mutants isolated from the spleen cell fraction from 2r3 mice exposed to EO Ž200 ppm. that demonstrated an increased lacI mutant frequency. Although the group mean for these animals exposed to 200 ppm EO was not significantly increased compared to air controls, DNA sequence analysis was done on these animals Žand controls. to assess if the increased lacI mutant frequency was due to a clonal expansion of lacI mutants.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
2. Materials and methods 2.1. Chemicals Chemicals used for the present studies were purchased as follows: ethylene oxide Žpurity 99.99%, SunOx Corp.., proteinase K ŽSigma., phenolrchloroform ŽAmresco, St. Louis, MO., 5-bromo-4chloro-3-indolyl-b -D -galactoside Ž X-gal, Gold Biotechnologies, St. Louis, MO., and dimethylformamide ŽSigma.. 2.2. Animals Male B6C3F1 lacI transgenic ŽBig Blue e . mice Ž lacI mice. 4–6 weeks of age were purchased from Stratagene Cloning Systems ŽTaconic Farms, Germantown, NY.. Mice were acclimated for at least 1 to 2 weeks prior to treatment and were weighed weekly throughout the in-life part of the study. 2.3. Exposures to ethylene oxide The exposure levels of 50, 100, and 200 ppm in the present study were chosen based on the NTP carcinogenicity bioassay in which male B6C3F1 mice exposed to 50 or 100 ppm EO for 2 years showed an increased incidence of lung hyperplasia, adenoma, and carcinoma w12x. Selection of exposure time was based on evidence that EO-induced DNA adducts approach steady-state in B6C3F1 mice after 4 weeks, using a similar exposure scenario w22x. The selection of the post-exposure fixation times were based in part on data from the same study showing that the level of DNA adducts drops to a small fraction of the steady-state concentration by 2 weeks post-exposure. To further evaluate fixation time in the tissues of lacI B6C3F1 mice following exposure to EO, three time points were chosen for tissue collection: 0, 2 and 8 weeks post-exposure. Three groups of animals exposed to EO were used for the lacI mutagenicity experiment, each containing 15 male 7- to 9-week-old lacI mice. Animals were exposed to constant whole-body target exposures of 50, 100, or 200 ppm EO, respectively, for 4 weeks Ž6 hrday, 5 daysrweek.. An additional four
155
animals per exposure group Ža total of 19 mice per exposure group. were included to determine the hprt T-lymphocyte mutant frequency w21x. Another group containing 19 male lacI mice was exposed to clean air in a separate chamber in the same room. The animals in each exposure group were housed individually in hanging steel cages inside a 1-m3, Hinnersstyle chamber w23x. The mice were deprived of food during the 6-h exposure period. After each 6-h exposure, the mice were fed ŽNIH-07 certified feed. and placed in individual polystyrene shoe box cages until the next exposure period. This housing decreased the possibility that the air-only animals were exposed to EO through exhalation Žoff-gassing. of EO from exposed animals. EO concentrations in the exposure chambers were monitored with a Hewlett-Packard 5890 Series II gas chromatograph equipped with a flame ionization detector. Concentrations of EO were monitored at least once per hour from each of the four chambers. EO concentrations Žin ppm. over the entire 4-week exposure period were 0.0 Ž"0.0., 51.9 Ž"1.6., 102.4 Ž"2.5., and 196.5 Ž"4.3. for the target concentrations of 0, 50, 100, and 200 ppm, respectively. The 19 control mice were exposed to clean air of the same temperature, relative humidity, and airflow as delivered to the EO-exposed animals. Temperature and relative humidity for all exposures were maintained at 728F Ž"48. and 50% Ž"10%., respectively. 2.4. Necropsy and tissue collection Of the 19 animals from each exposure Ž0, 50, 100, or 200 ppm EO. 5 were sacrificed at 1 h or at 2 weeks after the final exposure. Tissues from the 9 remaining animals in each exposure group were collected at 8 weeks after the final exposure. The spleens from four of the 19 remaining animals at the 8-week fixation time point were used to determine the hprt T-lymphocyte mutant frequency described in an accompanying article w21x. The mice were asphyxiated using carbon dioxide and exsanguinated by cardiac puncture. Seven major organs and bone marrow excised from the animals were flash-frozen in liquid nitrogen and subsequently stored at y808C as previously
156
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
described w24x. As part of a collaborative study with Dr. V. Walker ŽWadsworth Center for Laboratories and Research, Albany, NY. to assess the hprt mutant frequency in T-lymphocytes from the same animals exposed to EO, a spleen cell fraction Žboth Band T-lymphocytes. was prepared from air control and EO-exposed mice w4x. Approximately one-half of the preparation was used to determine the hprt mutant frequency in T-lymphocytes w21x, and the remaining half was frozen in liquid nitrogen and stored at y808C for lacI mutant frequency determination. 2.5. lacI mutant frequency determinations in tissues of transgenic mice Genomic DNA was prepared according to the Stratagene Big Blue Transgenic Mouse Mutagenesis Assay System Instruction Manual ŽLa Jolla, CA.. After ethanol precipitation, the DNA was resuspended in 50 to 200 ml TE and allowed to stand at room temperature overnight before storage at 48C. To assess the lacI mutant frequency in germ cells, DNA was extracted from seminiferous tubule germ cells. The lambda shuttle vector containing the lacI transgene was recovered from bulk genomic DNA with lambda phage packaging extract ŽTranspack, Stratagene. used according to the manufacturer’s instructions. Packaged phage were preadsorbed to E. coli SCS-8 cells ŽStratagene. for 20 min at 378C, mixed with prewarmed NZY ŽN-Z-Amine A ŽCasein Enzymatic Hydrolysate.; Yeast Extract. top agar containing 1.3 mgrml X-gal Ždissolved in dimethylformamide., and poured into 22.3- or 25-cm2 square plates containing NZY agar. Plates were incubated overnight at 378C. The total number of colorless lambda phage plaques was estimated for each plate by counting three 1-cm2 sections, averaging these numbers, and multiplying by the area of the plate. Blue lacIy plaques were counted and picked into individual tubes containing 0.5 ml SM buffer and 50 ml chloroform for further characterization. All mutant plaques were confirmed by restreaking on an SCS-8 bacterial lawn on X-gal-containing NZY agar. Mosaic plaques were observed rarely and were not included in the analysis since sectored plaques are likely to be the result of lacI ex-vivo mutation during E. coli repli-
cation or repair Ždiscussed in w25x.. The lacI mutant frequencies were calculated by dividing the number of confirmed mutant plaques by the total number of plaques analyzed. Although five animals per exposure group were included during the in-life portion of the experiment ŽEO exposures., due to the large number of analysis carried in the present study 3–4 animals per data point were typically conducted. In one case Žgerm cells from air control group 8 weeks post-exposure., only two animals were analyzed for the lacI mutant frequency due to sample loss. We routinely analyzed 150 000–500 000 per animal to determine the lacI mutant frequency. 2.6. DNA sequence analysis of lacI The lacI gene was sequenced from lacIy mutants isolated from the T-cell-enriched spleen samples of air control and 200 ppm EO-exposed mice. DNA sequence analysis of lacI mutants was done as described previously w24,26x using a 373A DNA Sequencer ŽApplied Biosystems, Foster City, CA. and SeqEd 675 software for data analysis. The complete lacI gene was sequenced for each mutant, and all mutations were confirmed by two sequencing reactions. No mutants with more than one mutation were found. The first G in the GTG start codon is base 29 w27x. 2.7. Statistics The statistical analyses were conducted using Minitab and Statgraphics software. The statistical analysis were as recommended for the in vivo lacI mutagenicity assay w28x and used previously w26x. Mutant frequency data were log-transformed to normalize variance among samples. Student’s t-test was used to determine statistical significance of mutant frequency in EO-exposed mice relative to air controls in lung, spleen, bone marrow, and germ cells from EO-treated mice relative to controls. The T-cell fraction mutant frequencies that involved four exposure groups were tested by analysis of variance using the log transformed data ŽANOVA.. A value of p - 0.05 was considered significant.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
157
3. Results
3.3. Background mutations
3.1. Determination of lacI mutant frequency in tissues of mice
Twenty-two lacI spleen cell mutants were sequenced from the air control group Žanimals 201, 202, 203, and 204., and 20 unique mutations were identified. The locations and types of mutations are listed in Tables 2 and 3 and summarized in Table 4; indicated bases refer to the coding strand of the lacI gene. Single base substitutions accounted for 95% Ž19r20. of the background mutations; 68% Ž13r19. of these were transitions, and 32% Ž6r19. were transversions ŽTable 4.. Three of the base substitutions changed the original amino acids to stop codons that result in premature termination of protein synthesis. A single-base deletion at G870 that results in a protein truncated at amino acid 285, about 80% of the full-length 360 amino acid lacI protein, was the only frameshift mutation observed among the background mutants ŽTable 3.. This same site, G870 , was also mutated in animal 204, resulting in a G:C ™ A:T transition. Substitutions at G:C base pairs constituted 89% Ž17r19. of the background base substitutions. Seventy-three percent Ž8r11. of the G:C ™ A:T transitions occurred at CpG dinucleotides. G:C ™ A:T transitions at CpG nucleotides were preferentially involved as background mutations in this system w24,29,30x and likely result from deamination of 5-methyl-cytosine to thymine w31x.
The lacI mutant frequency was determined in lung, bone marrow, spleen, and germ cells and in a spleen T-cell fraction from EO-exposed male B6C3F1 lacI transgenic mice. The time from final exposure to euthanasia Žfixation time., total numbers of plaques examined, number of lacI mutant plaques confirmed, and lacI mutant frequencies for these samples are shown in Table 1. All mutant frequency data were normalized by log transformation, and then samples from animals exposed to EO were compared to air control tissue using one-sided, twosample Student’s t-test. Using this analysis, the mutant frequency in lung at 8 weeks post-exposure was significantly increased Ž p s 0.046.. The lacI mutant frequencies were not significantly increased in bone marrow at 2 Ž p s 0.24. and 8 Ž p s 0.07. weeks, in spleen at 2 Ž p s 0.11. and 8 Ž p s 0.49. weeks, or in germ cells at 2 and 8 weeks Ž p s 0.42. post-exposure. The spleen T-cell fraction mutant frequency data, which includes results from 0, 50, 100, and 200 ppm exposures, were compared using ANOVA. The T-cell mutant frequency from mice exposed to 200 ppm EO was 4-fold over controls Ž3.0 " 1.2 vs. 12.4 " 11.8; "standard deviation, SD., and yet the difference was statistically insignificant Ž p s 0.07. by ANOVA, likely due to interanimal variation as indicated by the large SD in the EO-exposed group. 3.2. DNA sequence analysis of lacI mutants from spleen cell fractions The T-cell fraction from two of three animals in the EO-exposed group showed an increased lacI mutant frequency Žanimals 214 and 215; Table 1.; DNA sequence analysis was performed on a collection of lacI mutants isolated from EO-exposed mice and air controls to determine the molecular basis for the observed mutant frequency increase. LacI mutant plaques with identical mutations isolated from a single animal were considered to be mutant siblings arising from one in vivo mutational event w26x.
3.4. Mutations in EO-exposed mice Forty-five lacI mutants were sequenced from the spleen cell fraction of the 200 ppm EO-exposed group Žanimals 213, 214, and 215.; 20 unique mutations were identified ŽTables 2 and 3.. This group was selected for sequencing because it exhibited the highest overall mutant frequency of the three exposed groups, and 2 of 3 animals in this group had increased lacI mutant frequencies. Single base substitutions accounted for 85% Ž17r20. of the mutations in the EO-exposed group; 76% Ž13r17. of these were transitions, and 24% Ž4r17. were transversions. Of the 5 G:C ™ A:T transitions, 80% occurred at CpG sites. One mutation, a G:C ™ T:A transversion, resulted in replacement of Glu 401 with a stop codon.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
158
Table 1 LacI mutant frequency in tissues of B6C3F1 lacI transgenic mice following inhalation exposure to ethylene oxide Animal Exposure Total Žppm. no. plaques
No. of Mutant mutants frequency Ž=10y5 .
Lung (8 weeks) a 11 0 12 0 13 0 14 0
542 000 27 591 000 40 201 000 8 501 000 45
56 57 58 59
478 000 341 000 286 000 206 000
200 200 200 200
47 37 23 16
Bone marrow (2 weeks) 7 0 357 000 7 8 0 316 000 17 9 0 304 000 12 10 0 242 000 2 51 52 53 54
200 200 200 200
324 000 6 337 000 7 313 000 22 240 000 16
Bone marrow (8 weeks) 11 0 285 000 12 0 262 000 13 0 254 000 14 0 225 000 56 57 58 59
200 200 200 200
8 8 8 2
275 000 9 228 000 12 293 000 8 218 000 16
Spleen (2 weeks) 7 0 8 0 9 0 10 0
529 000 12 499 000 8 594 000 41 203 000 12
51 52 53 54
277 000 247 000 242 000 243 000
200 200 200 200
Spleen (8 weeks) 11 0 12 0 13 0 14 0
12 19 36 10
541 000 24 523 000 18 237 000 9 481 000 24
5.0 6.8 4.0 9.0 mean"SD 6.2"2.2 9.8 10.9 8.0 7.8 mean"SD 9.1"1.5 b 2.0 5.4 4.0 0.8 mean"SD 3.0"2.0 1.8 2.1 7.0 6.6 mean"SD 4.4"2.8 2.8 3.0 3.2 0.9 mean"SD 2.5"1.1 3.3 5.3 2.7 7.4 mean"SD 4.7"2.1 2.3 1.6 6.9 5.9 mean"SD 4.2"2.6 4.3 7.7 14.9 4.1 mean"SD 7.8"5.0 4.4 3.4 3.8 5.0 mean"SD 4.2"0.7
Table 1 Žcontinued. Animal no.
Exposure Žppm.
Total plaques
No. of mutants
Mutant frequency Ž=10y5 .
56 57 58 59
200 200 200 200
290 000 290 000 297 000 200 000
15 24 21 2
5.2 8.3 7.1 1.0 mean"SD 5.4"3.2
Germ cells (2 weeks) 7 0 8 0 9 0 10 0
132 000 227 000 256 000 304 000
2 10 5 6
51 52 55
409 000 669 000 600 000
9 12 21
1.5 4.4 2.0 2.0 mean"SD 2.5"1.3 2.2 1.8 3.5 mean"SD 2.5"0.9
Germ cells (8 weeks) 11 0 200 000 13 0 171 000
10 3
57 59 60
200 200 200
200 200 200
197 000 237 000 254 000
8 6 5
Spleen cell fraction (8 weeks) 201 0 221 000 202 0 204 000 203 0 220 000 204 0 232 000
7 6 3 10
206 207 208
50 50 50
211 000 211 000 237 000
8 3 4
209 210 211 212
100 100 100 100
227 000 211 000 169 000 173 450
6 9 4 3
213 214 215
200 200 200
233 000 226 000 241 000
7 58 21
a b
5.0 1.7 mean"SD 3.4"2.3 4.1 2.5 2.0 mean"SD 2.9"1.1 3.2 2.9 1.4 4.3 mean"SD 3.0"1.2 3.8 1.4 1.7 mean"SD 2.3"1.3 2.6 4.3 2.4 1.7 mean"SD 2.8"1.0 3.0 25.6 8.7 mean"SD 12.4"11.8
Post-exposure fixation time following 4 weeks of EO exposure. Significantly greater than air control, Student’s t-test Ž p- 0.05..
Two deletions and one insertion mutation, all isolated from animal 213, were identified ŽTable 3.. The 11 bp deletion at base 978 in animal 213 resulted in a frameshift that encoded a product three amino acids longer than the wild-type protein. The 9
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
159
Table 2 Mutational spectrum in spleen cell fractions isolated from air control and ethylene-oxide-exposed Ž200 ppm. B6C3F1 lacI transgenic mice
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
160
Table 3 Deletions and insertions in lacI recovered from the spleen cell fraction of B6C3F1 lacI transgenic mice: air control mice and mice exposed to 200 ppm ethylene oxide ŽEO.
bp deletion at base 426 in the same mouse did not result in a frameshift, but coded for a product missing three amino acids. The 4 bp insertion in animal 213 was located within a TGGC triple tandem repeat that runs from position 621 to 632; this frameshift mutation introduced a stop codon that truncated the encoded protein at about half the size of wild-type lacI repressor. Insertions and deletions at this triple tandem repeat have been previously observed w24,29x. In the EO-exposed animals 214 and 215, 17% Table 4 LacI mutations recovered from the spleen cell fraction of B6C3F1 lacI transgenic mice: air controls and mice exposed to 200 ppm ethylene oxide Mutational type Point mutations At G:C base pairs GC ™ AT GC ™ TA GC ™CG At A:T base pairs AT ™GC AT ™CG AT ™TA Multibase alterations Tandem change Insertions Deletions Totals
Air control Ž%.
11 Ž55. 5 Ž25. 1 Ž5.
EO-exposed Ž%.
5 Ž24. 6 Ž29. 2 Ž10.
2 Ž10. 0 Ž - 5. 0 Ž - 5.
3 Ž14. 1 Ž5. 1 Ž5.
0 Ž - 5. 0 Ž - 5. 1 Ž5.
0 Ž - 5. 1 Ž5. 2 Ž10.
20 Ž100.
21 Ž100.
Ž3r18. and 55% Ž11r20. of the mutants, respectively, are presumed to represent independent mutations on the basis that lacI mutants with identical mutations isolated from the same animal tissue are mutant siblings arising from one in vivo mutational event w26x. Of the 19 mutants isolated from animal 214 that were sequenced, there are 16 Ž84%. G:C ™ A:T transitions at base 270. In animal 215, a G:C ™ A:T transition at base 777 is present in 9 Ž45%. of the 20 mutants. These G:C ™ A:T transitions at 270 and 777, are at CpG sites, and these same mutations have been observed to occur in the bone marrow of air control animals from this and our previous studies w24x. Table 5 LacI mutant frequencyrmutation frequency Ž=10y5 . in spleen cell fractions isolated from air control and 200 ppm EO-exposed B6C3F1 lacI transgenic mice Animal no.
Exposure
Mutant frequency
air air air air 200 ppm 200 ppm 200 ppm
Mutation frequency
Ž%. Ž n. a
Žppm. 201 202 203 204 213 214 215
Mutation
3.2 2.9 1.4 4.3 3.0 25.6 8.7
100 Ž5. 60 Ž5. 100 Ž2. 78 Ž9. 100 Ž4. 12 Ž18. 45 Ž20.
3.2 1.7 1.4 3.3 3.0 3.1 3.9
a Mutation frequency is the mutant frequency=the percentage of nonsibling mutants observed in each animal; ns number of lacIy mutants sequenced from each animal.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
DNA sequence analysis can be used to assess the contribution of mutant siblings to the in vivo lacI mutant frequency and permits an estimation of the lacI mutation frequency from the lacI mutant frequency w26x. The lacI mutant frequency is multiplied by the percentage of independent mutational events observed in each animal to obtain a mutation frequency. When the DNA sequence data were used to obtain a lacI mutation frequency for air control animals and lacI mice exposed to 200 ppm EO ŽTable 5., the increased lacI mutant frequency observed in animal 214 and 215 was reduced to a mutation frequency that was equivalent to the background observed in air controls.
4. Discussion We have assessed the lacI mutant frequency in somatic and germ cells of male B6C3F1 lacI transgenic mice exposed by inhalation to EO for 4 weeks. Among the tissues examined Žbone marrow, spleen, lung., an increased lacI mutant frequency was observed only in lung. Lung was a target organ for tumor induction in B6C3F1 mice in the NTP carcinogenicity bioassay of EO w12x. We have recently completed the in-life portion a 48-week EO exposure of lacI mice with interim sacrifice points to assess the lacI mutant frequency following longer term exposures. In these studies, we will analyze the lacI mutant frequency in lung to determine if the significant increase in the lung lacI mutant frequency observed in the present study can be reproduced. If elevated, the tissues from the present study and the 48-week exposure will provide the basis to determine the mutational spectrum of EO in a tumor target tissue. EO is a germ cell mutagen in mice, inducing heritable translocations, dominant lethal and specific locus mutations following inhalation exposure w15,32x. In the study presented here, mutant frequencies were assessed in germ cells from male lacI mice exposed by inhalation to 200 ppm EO Ž4 weeks., sampled at 2 and 8 weeks after the last exposure. No increase in the lacI mutant frequency was observed in treated animals over controls. The lack of a response in germ cells of lacI mice could be due to the short fixation time used to determine
161
the lacI mutant frequency after EO exposure. In l lacZ transgenic mice administered ENU Ž150 mgrkg i.p. injection., lacZ mutation occurred only in the stem cell population Ž50–100 days post-exposure. w33x. The 2- and 8-week fixation time following the 4 weeks of EO exposure used in the present study is an insufficient time period to assess EO effects at the lacI transgene in the stem cell population. Studies using longer fixation times following EO exposure than used in the present studies are required to adequately assess the effects of EO on the lacI transgene mutant frequency in testes from transgenic animals. These studies are in progress in our laboratory. In two of three animals exposed to 200 ppm EO, the lacI mutant frequency in the spleen cell fraction was increased 2.5- and 9-fold above the mutant frequency in air controls. DNA sequence analysis revealed the presence of identical mutations in mutant plaques isolated from the same animal. In both cases, a G:C ™ A:T transition at a CpG dinucleotide was the type of mutation identified in multiple isolates. The G:C ™ A:T transition ŽC 270 . identified in animal 214 accounted for 84% of the mutations observed among the lacI mutant plaques recovered. This mutation has been observed previously in bone marrow from BD-exposed mice Žw26x; 1250 ppm exposure group. and in the skin of control and DMBA-exposed mice skin w34x. The G:C ™ A:T transition ŽC 777 . identified in animal 215 accounted for 45% of the mutations among the lacI mutant plaques isolated. This mutation has been observed in two air control mice from a previous study w24x and in skin from untreated mice w34x. Identical mutations among lacI mutants recovered from a single tissue of an animal can be considered to be mutant siblings arising from one in vivo mutation w26x. The lacI mutation frequencies were equivalent in spleen cell fractions from air control and 200 ppm EO-exposed animals ŽTable 5.. These data indicate that the elevated lacI mutant frequencies determined in animals 214 and 215 ŽTable 1. were not due to an increase in the lacI mutation frequency but to clonal expansion of a cell carrying a mutated lacI transgene. Since G:C ™ A:T transitions at CpG dinucleotides occur in approximately 50% of the background mutants isolated from various tissues of B6C3F1 lacI transgenic mice w24,34,35x, the mu-
162
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
tations that were clonally expanded in these animals may likely be background mutations. Therefore the increased lacI mutant frequency in animals 214 and 215 were likely due to the clonal expansion of a background mutation following or during the EO exposure period. Alternatively, these G:C ™ A:T transitions at CpG dinucleotides may represent EO mutational hot spots. However, the nature of the mutation observed ŽG:C ™ A:T transitions at CpG sites. and the lack of detection of these mutations in multiple animals within the EO-exposed group, indicate that these are probably background mutations. The assumption of clonality for all identical mutations observed in a tissue from the same animals is the most conservative approach in estimating the mutation frequency. The available data do not permit the assessment of cellular clonality for further examination of the origin of the lacI mutants. To provide a comparison between hprt mutation in T-lymphocytes and mutation at the lacI transgene, the lacI mutant frequency was determined in the same spleen cell fraction used to determine the hprt T-lymphocyte mutant frequency w21x. Eight weeks after the final exposure to 50, 100, and 200 ppm, hprt mutant frequencies in spleen were increased approximately 1.7-, 3.1-, and 6.4-fold above background in a dose-dependent manner. It is important to note that the hprt assay is performed on T-cells that grow out from the crude fraction, while the lacI lymphocyte fraction mutant frequencies were measured in a population enriched for both B- and T-cells but not necessarily identical to the pool of cells from which hprt mutant frequencies were measured Ži.e. T-cells.. The occurrence of a detectable mutational response at hprt but not at the lacI transgene in the spleen cell fraction is likely due to the mechanism of action of EO-induced mutation. Molecular characterization of EO-induced hprt mutations in diploid human fibroblasts in vitro w36x has indicated that as many as 50% of the hprt mutations induced by EO are large deletions, often involving loss of the entire hprt gene. If large deletions are a characteristic of EO-induced mutations in vivo as well as in vitro, then differences in the hprt and lacI mutant frequencies may be due to the recovery of large deletions as part of the hprt mutant frequency but the lack of recovery of these events at the lacI transgene w37x.
The hprt mutational spectrum for EO determined in human cells in vitro indicates the presence of an increased frequency of both point mutations and large deletions. However, it is uncertain how the frequency of these mutations determined in vitro among EO-exposed cells is reflected in the mutational spectrum determined from tissues in EO-exposed mice. The present studies indicate that a primary mechanism of EO-induced mutagenicity in vivo is likely through the induction of deletions, not specific point mutations. In a comparative study of the mutant frequencies at the endogenous Dlb-1 locus and the lacI transgene after exposure to either ethyl nitrosourea ŽENU. or X-rays w38,39x the two loci responded comparably to ENU. In contrast, X-rays induced relatively few lacI mutations but many Dlb-1 mutations. The difference in response between the lacI transgene was attributed to the difference in mechanism of action for mutation; ENU is primarily a point mutagen, while X-rays can induce a large proportion of deletions that may be too large to be phage-recoverable w37–39x. From these data and others, large deletions appear to be excluded from lacI mutant frequency determination w37x. A mechanistic difference in the two mutation assays Ž hprt vs. lacI . is that the lacI transgenic mouse assay detects primarily base substitutions and small-scale deletions and insertions, while the hprt assay detects these events plus large-scale mutations as well. Therefore comparisons of lacI and hprt mutant frequencies in identically exposed animals can be complicated particularly for the case of EO, since the different mutation assays do not detect identical mutant cell populations. At present, we are conducting a long-term Ž48week. inhalation exposure to EO Žin collaboration with Dr. R.J. Preston, CIIT. in B6C3F1 mice Žgeneric and lacI transgenic., to assess both the induction of mutation at the lacI transgene and chromosomal alterations in the same tissues. In this study, we will reexamine the lacI mutant frequency in the tissues of mice with an increasing exposure time to EO to assess the role of longer term exposure on the lacI mutant frequency. Due to the findings in the present study, we will specifically examine the lacI mutant frequency in lung, spleen and germ cells at increasing exposure levels with time and if appropriate determine the lacI mutational spectrum.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
In conclusion, we determined the lacI mutant frequency in the tissues of B6C3F1 lacI transgenic mice exposed to the genotoxic carcinogen EO by inhalation. EO was mutagenic in a single tissue, the lung, and did not increase the lacI mutant frequency in other tissues examined. An increased lacI mutant frequency observed in a T-cell fraction obtained from two animals in the high-exposure group was likely due to a clonal expansion of an in vivo background mutation. These data suggest that a predominate mode of in vivo genotoxicity induced by EO Žlarge deletions. are not recoverable at the lacI transgene. The present study emphasizes the need to assess multiple genotoxic end points in the same animals to provide an understanding of in vivo mechanisms of genotoxicity by EO and the relationship between the induction of point mutations detectable at the lacI transgene or hprt vs. chromosomal alterations that occur in experimental animals and humans exposed to EO.
Acknowledgements This work was supported in part by NRSA Grant no. 1F32 CA61722-01 from the National Cancer Institute. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. We thank Drs. R. Julian Preston ŽCIIT., Tim Fennell ŽCIIT. and Dr. V. Walker ŽWadsworth Center, NY. for discussion of data and reviewing the manuscript. We thank the CIIT inhalation facility Dr. Owen Moss and Mr. Arden James for coordinating and monitoring EO exposures. We thank the CIIT animal care facility and necropsy staff. We thank Dr. Barbara Kuyper for editorial assistance.
References w1x US Environmental Protection Agency ŽUS EPA. Ž1991. Preliminary draft list of categories and subcategories under Section 112 of the Clean Air Act. Fed. Reg., 56, 28548– 28557. w2x V.L. Dellarco, W.M. Generoso, G.A. Sega, J.R. Fowle, III, D. Jacobson-Kram, Review of the mutagenicity of ethylene oxide, Environ. Mol. Mutagen. 16 Ž1990. 85–103. w3x L.E. Appelgren, G. Eneroth, C. Grant, L.E. Lanstrom, K.
w4x
w5x
w6x
w7x
w8x
w9x
w10x
w11x
w12x
w13x
w14x
w15x
w16x
w17x
163
Tenghagen, Testing of ethylene oxide for mutagenicity using the micronucleus test in mice and rats, Acta Pharmacol. Toxicol. 43 Ž1978. 69–71. V.E. Walker, T.R. Skopek, A mouse model for the study of in vivo mutational spectra: sequence specificity of ethylene oxide at the hprt locus, Mutation Res. 288 Ž1993. 151–162. J.W. Yager, R.D. Benz, Sister chromatid exchanges induced in rabbit lymphocytes by ethylene oxide after inhalation exposure, Environ. Mutagen. 4 Ž1982. 121–134. A.D. Kligerman, G.L. Erexson, M.E. Phelps, J.L. Wilmer, Sister chromatid exchange induction in peripheral blood lymphocytes of rats exposed to ethylene oxide by inhalation, Mutation Res. 120 Ž1983. 37–44. D.W. Lynch, T.R. Lewis, W.J. Moorman, J.R. Burg, D.K. Gulati, P. Kaur, P.S. Sabharwal, Sister chromatid exchanges and chromosome aberrations in lymphocytes from monkeys exposed to ethylene oxide and propylene oxide by inhalation, Toxicol. Appl. Pharmacol. 76 Ž1984. 85–95. S.E. Lewis, L.B. Barnett, C. Felton, F.M. Johnson, L.C. Skow, N. Cacheiro, M.D. Shelby, Dominant visible and electrophoretically expressed mutations induced in male mice exposed to ethylene oxide by inhalation, Environ. Mutagen. 8 Ž1986. 867–872. L.R. Ribeiro, M.N. Rabello-Gay, D.M.F. Salvadori, C.A.B. Pereira, Cytogenetic effects of inhaled ethylene oxide in somatic and germ cells of mice, Arch. Toxicol. 59 Ž1987. 332–335. International Agency for Research on Cancer ŽIARC. Ž1994., Some Industrial Chemicals, IARC, Ethylene Oxide, 60, pp. 73-159. W.M. Snellings, C.S. Weil, R.R. Maronpot, A two-year inhalation study of carcinogenic potential of ethylene oxide in Fischer 344 rats, Toxicol. Appl. Pharmacol. 75 Ž1984. 105–117. National Toxicology Program ŽNTP. Ž1988. Toxicology and carcinogenesis studies of ethylene oxide in B6C3F1 mice, in: NTP Technical Report, USDHHS, U.S. Government Printing Office, Washington, DC, vol. 326. W.M. Generoso, K.T. Cain, M. Krishna, C.W. Sheu, R. M Gryder, Heritable translocation and dominant-lethal mutation induction with ethylene oxide in mice, Mutation Res. 73 Ž1980. 133–142. W.M. Generoso, K.T. Kain, L.A. Hughes, G.A. Sega, P.W. Braden, D.G. Gosslee, M.D. Shelby, Ethylene oxide dose and dose-rate effects in the mouse dominant-lethal test, Environ. Mutagen. 8 Ž1986. 1–7. W.M. Generoso, K.T. Cain, C.V. Cornett, N.L.T. Cacherio, L.A. Hughes, Concentration–response curves for ethylene oxide-induced heritable translocations and dominant lethal mutations, Environ. Mol. Mutagen. 16 Ž1990. 126–131. G.A. Sega, P.A. Brimer, E.E. Generoso, Ethylene oxide inhalation at different exposure-rates affects binding levels in mouse germ cells and hemoglobin. Possible explanation for effect, Mutation Res. 249 Ž1991. 339–349. L.R. Ribeiro, D.M.F. Salvadori, C.A.B. Pereira, W. Becak, Activity of ethylene oxide in the sperm morphology test, Arch. Toxicol. 60 Ž1987. 331–333.
164
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
w18x L. Rhomberg, V.L. Dellarco, C. Siegel-Scott, K.L. Dearfield, D. Jacobson-Kram, Quantitative estimation of the genetic risk associated with the induction of heritable translocations at low-dose exposure: ethylene oxide as an example, Environ. Mol. Mutagen. 16 Ž1990. 104–125. w19x R.J. Preston, T.R. Fennell, A.P. Leber, R.L. Sielken, J.A. Swenberg, Reconsideration of the genetic risk assessment from ethylene oxide exposures, Environ. Mol. Mutagen. 26 Ž1996. 189–202. w20x J.C. Mirsalis, J.A. Monforte, R.A. Winegar, Transgenic animal models for measuring mutations in vivo, Crit. Rev. Toxicol. 24 Ž1994. 255–280. w21x Walker, V.E., L. Recio, S.C. Sisk and T.R. Skopek Ž1997. In vivo mutagenicity of ethylene oxide at the HPRT locus of T-lymphocytes of B6C3F1 lacI transgenic mice following inhalation exposure, submitted to Mutation Res. w22x V.E. Walker, T.R. Fennell, P.B. Upton, J.P. MacNeela, J.A. Swenberg, Molecular dosimetry of DNA and hemoglobin adducts in mice and rats exposed to ethylene oxide, Environ. Health Perspect. 99 Ž1993. 11–17. w23x R.G. Hinners, J.K. Burkart, C.L. Punte, Animal inhalation exposure chambers, Arch. Environ. Health 16 Ž1968. 194– 206. w24x S.C. Sisk, L.J. Pluta, J.A. Bond, L. Recio, Molecular analysis of lacI mutants from bone marrow of B6C3F1 transgenic mice following inhalation exposures to 1,3-butadiene, Carcinogenesis 15 Ž1994. 471–477. w25x W.W. Piegorsch, B.H. Margolin, M.D. Shelby, A. Johnson, J.E. French, R. Tennant, K.R. Tindall, Study design and sample sizes for a lacI transgenic mouse mutation assay, Environ. Mol. Mutagen. 25 Ž1995. 231–245. w26x L. Recio, K.G. Meyer, Increased frequency of mutations at A:T base pairs in the bone marrow of B6C3F1 lacI transgenic mice exposed to 1,3-butadiene, Environ. Mol. Mutagen. 26 Ž1995. 1–8. w27x P.J. Farabaugh, U. Schmeissner, M. Hofer, J.H. Miller, Genetic studies of the lacI repressor, J. Mol. Biol. 126 Ž1978. 847–863. w28x J.D. Callahan, J.M. Short, Transgenic l r lacI mutagenicity assay: statistical determination of sample size, Mutation Res. 327 Ž1995. 201–208. w29x S.W. Kohler, G.S. Provost, A. Fieck, P.L. Kretz, W.O.
w30x
w31x w32x
w33x
w34x
w35x
w36x
w37x
w38x
w39x
Bullock, J.A. Sorge, D.L. Putman, J.M. Short, Spectra of spontaneous and mutagen-induced mutations in the lacI gene in transgenic mice, Proc. Natl. Acad. Sci. USA 88 Ž1991. 7958–7962. G.S. Provost, P.L. Kretz, R.T. Hamner, C.D. Matthews, B.J. Rogers, K.S. Lundberg, M.J. Dydaico, J.M. Short, Transgenic systems for in vivo mutation analysis, Mutation Res. 288 Ž1993. 133–149. A.P. Bird, CpG-rich islands and the function of DNA methylation, Nature 321 Ž1986. 209–213. L.B. Russell, R.B. Cumming, P.R. Hunsicker, Specific-locus mutation rates in the mouse following inhalation of ethylene oxide, and application of the results to estimation of human genetic risk, Mutation Res. 129 Ž1984. 381–388. J.H. van Delft, R.A. Baan RA, Germ cell mutagenesis in lambda lacZ transgenic mice treated with ethylnitrosourea: comparison with the specific locus test, Mutagenesis 10 Ž1995. 209–214. N.J. Gorelick, J.L. Andrews, M. Gu, B.W. Glickman, Mutational spectra in skin from 7,12-dimethylbenzw axanthracenetreated and untreated transgenic mice, Mol. Carcinogen. 14 Ž1995. 53–62. A.D.P. Knoll, P.L. Jacobson, K.S. Kretz, J.M. Lundberg, J.M. Short, S.S. Sommer, Spontaneous mutations in lacIcontaining l lysogens derived from transgenic mice: the observed patterns differ in liver and spleen, Mutation Res. 311 Ž1994. 57–67. T. Bastlova, ´ B. Andersson, B. Lambert, A. Kolman, Molecular analysis of ethylene oxide-induced mutations at the HPRT locus in human diploid fibroblasts, Mutation Res. 287 Ž1993. 283–292. J.A. Gossen, H.-J. Martus, J.Y. Wei, J. Vijg, Spontaneous and X-ray-induced deletion mutations in a lacZ plasmidbased transgenic mouse model, Mutation Res. 331 Ž1995. 89–97. K.S. Tao, C. Urlando, J.A. Heddle, Comparison of somatic mutation in a transgenic versus host locus, Proc. Natl. Acad. Sci. USA 90 Ž1993. 10681–10685. K.S. Tao, C. Urlando, J.A. Heddle, Mutagenicity of methyl methanesulfonate ŽMMS. in vivo at the Dlb-1 native locus and a lacI transgene, Environ. Mol. Mutagen. 22 Ž1993. 293–296.
Assessment of the in vivo mutagenicity of ethylene oxide in the tissues of B6C3F1 lacI transgenic mice following inhalation exposure Susan C. Sisk 1, Linda J. Pluta, Kathy G. Meyer, Brian C. Wong, Leslie Recio
)
Chemical Industry Institute of Toxicology, 6 DaÕis DriÕe, Research Triangle Park, NC 27709, USA Received 7 October 1996; revised 10 February 1997; accepted 12 February 1997
Abstract In the present study, the lacI mutant frequency was determined in the tissues of B6C3F1 lacI transgenic mice exposed by inhalation to ethylene oxide ŽEO.. Groups of 15 male transgenic lacI B6C3F1 mice were exposed to either 0, 50, 100, or 200 ppm EO for 4 weeks Ž6 hrday, 5 daysrweek. and were sacrificed at 0, 2, or 8 weeks after the last EO exposure. The lacI transgene was recovered from lung, bone marrow, spleen, and germ cells for determination of the lacI mutant frequency. The tissues selected for analysis were tumor target site tissues in chronic bioassays Žlung tumors and lymphomas. and germ cells. The lacI mutant frequency in lung was significantly increased at 8 weeks post exposure to 200 ppm EO Ž6.2 " 2.2 vs. 9.1 " 1.5, p s 0.046.. In contrast, the lacI mutant frequency in spleen and bone marrow at 2 and 8 weeks was not significantly increased in mice exposed to 200 ppm EO. The lacI mutant frequencies in male germ cells for 200 ppm EO-exposed mice were not increased compared to air controls at 2 and 8 weeks post-exposure. In a spleen cell fraction two of three EO-exposed mice at the 200 ppm exposure level demonstrated an elevated lacI mutant frequency. The increased lacI mutant frequency in these animals was likely due to mutant siblings that contained background G:C ™ A:T transitions at CpG sites. These results demonstrate that a 4-week inhalation exposure to EO is mutagenic in lung. However, EO did not increase the frequency of mutations recovered at the lacI transgene in other tissues examined under the conditions used in the present studies. Since the mutational spectrum for EO in other systems consists of an increased proportion of large deletions, the lack of a mutagenic response in the tissues examined is likely due to the lack of recovery of large deletions in lambda-based shuttle vector systems. These data indicate that a primary mechanism of EO-induced mutagenicity in vivo is likely through the induction of deletions, not specific point mutations. Keywords: Ethylene oxide; lacI; In vivo mutation; Mutant frequency; Mutation spectrum; Inhalation exposure
1. Introduction ) Corresponding author. Tel.: q1 Ž919. 5581329; Fax: q1 Ž919. 5581300; e-mail: [email protected] 1 Present address: ClinTrials Research Inc., P.O. Box 13991, Research Triangle Park, NC 27709, USA.
Ethylene oxide ŽEO; CAS No. 75-21-8. is a major industrial intermediate and gaseous sterilant regulated by the US Environmental Protection Agency
1383-5718r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 1 3 8 3 - 5 7 1 8 Ž 9 7 . 0 0 0 6 3 - 6
154
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
ŽEPA. as a hazardous air pollutant under the Clean Air Act w1x. EO reacts directly with cellular macromolecules, including DNA, causing gene mutations and chromosome deletions in somatic cells. Studies with Bacillus subtilis, Escherichia coli, Salmonella typhimurium, and Chinese hamster ovary cells demonstrated that EO does not require metabolic activation to induce gene mutations Žreviewed in w21x.. Intraperitoneal injection of EO increases bone marrow micronuclei in rats and mice w3x and is mutagenic at hprt in the T-lymphocytes of B6C3F1 mice w4x. Increased levels of sister chromatid exchange have been observed in peripheral lymphocytes of rodents and monkeys after inhalation of EO w5–9x. In biomonitoring studies, EO-exposed workers exhibit increased frequencies of several genotoxic end points relative to unexposed workers Žreviewed in w10x.. In 2-year inhalation bioassays, EO induced exposure-related increases in gliomas, peritoneal mesotheliomas, and mononuclear leukemias in F344 rats and lymphomas and adenomasrcarcinomas of the lung, Harderian gland, and mammary gland of B6C3F1 mice w11,12x. In human epidemiological studies reviewed by the International Agency for Research on Cancer ŽIARC., mortality from lymphatic and hematopoietic cancer was marginally elevated w10x. IARC has classified EO as a Group 1 chemical Žcarcinogenic to humans. w10x. This classification was based on limited evidence in humans for carcinogenicity, sufficient evidence in experimental animals and studies demonstrating in vivo genotoxicity in humans and rodents. In addition to somatic cell genotoxicity, EO can induce transmissible genetic damage in germ cells. Intraperitoneal injection of EO induced dominant lethal mutations and reciprocal translocations in postmeiotic germ cells from male mice w13x. Related studies with inhalation-exposed mice showed that the dominant lethal response increased in a nonlinear, exposure-related manner at high exposure levels; when total dose was kept constant, higher responses were measured in animals that received high concentrations for a short duration w14,15x. DNA adduct experiments in testes suggest that this exposure rate effect is due to concentration-dependent differences in internal dose, possibly resulting from respiratory rate attenuation during the longer low-dose expo-
sures w16x. Significant increases in abnormal sperm morphologies have also been reported in male mice after inhalation exposure to EO w17x. A US EPA genetic risk assessment of EO is presented in a series of papers in an issue of EnÕironmental and Molecular Mutagenesis w2,15,18x. A reconsideration of this risk assessment for EO has been proposed that takes into account aspects of effective dose to the target tissue, mechanisms of chromosomal alterations, and the shape of the dose–response curve at low doses w19x. Transgenic animal in vivo mutation models permit the determination and molecular analysis of mutation in marker genes from the tissues of exposed animals w20x. With the introduction of transgenic mice, the induction of mutations can be determined in multiple tissues with respect to exposure, dose to target tissue, DNA adduct levels, and additional genotoxicity end points. In the study presented here, we used B6C3F1 lacI transgenic mice ŽBig Blue e . to assess the mutagenicity of a 4-week inhalation exposure to EO. This mouse was selected for study because the B6C3F1 host strain is the strain used in the National Toxicology Program ŽNTP. carcinogenicity bioassay of EO w12x. The tissues selected for analysis of the in vivo lacI mutant frequency were tumor target site tissues in chronic bioassays Žlung tumors and lymphomas. w12x and male germ cells. In this article, we report the in vivo lacI mutant frequencies determined in lung, bone marrow, spleen, and seminiferous tubule germ cells of B6C3F1 lacI transgenic mice exposed to EO by inhalation at 0, 50, 100 or 200 ppm Ž6 hrday, 5 daysrweek for 4 weeks.. We also determined lacI mutant frequencies in a T-cell enriched fraction of spleen cells Žcalled spleen cell fraction herein. from air control and EO-exposed animals as a comparison to hprt mutant frequencies determined concurrently in these same animals w21x. DNA sequence analysis was performed on lacI mutants isolated from the spleen cell fraction from 2r3 mice exposed to EO Ž200 ppm. that demonstrated an increased lacI mutant frequency. Although the group mean for these animals exposed to 200 ppm EO was not significantly increased compared to air controls, DNA sequence analysis was done on these animals Žand controls. to assess if the increased lacI mutant frequency was due to a clonal expansion of lacI mutants.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
2. Materials and methods 2.1. Chemicals Chemicals used for the present studies were purchased as follows: ethylene oxide Žpurity 99.99%, SunOx Corp.., proteinase K ŽSigma., phenolrchloroform ŽAmresco, St. Louis, MO., 5-bromo-4chloro-3-indolyl-b -D -galactoside Ž X-gal, Gold Biotechnologies, St. Louis, MO., and dimethylformamide ŽSigma.. 2.2. Animals Male B6C3F1 lacI transgenic ŽBig Blue e . mice Ž lacI mice. 4–6 weeks of age were purchased from Stratagene Cloning Systems ŽTaconic Farms, Germantown, NY.. Mice were acclimated for at least 1 to 2 weeks prior to treatment and were weighed weekly throughout the in-life part of the study. 2.3. Exposures to ethylene oxide The exposure levels of 50, 100, and 200 ppm in the present study were chosen based on the NTP carcinogenicity bioassay in which male B6C3F1 mice exposed to 50 or 100 ppm EO for 2 years showed an increased incidence of lung hyperplasia, adenoma, and carcinoma w12x. Selection of exposure time was based on evidence that EO-induced DNA adducts approach steady-state in B6C3F1 mice after 4 weeks, using a similar exposure scenario w22x. The selection of the post-exposure fixation times were based in part on data from the same study showing that the level of DNA adducts drops to a small fraction of the steady-state concentration by 2 weeks post-exposure. To further evaluate fixation time in the tissues of lacI B6C3F1 mice following exposure to EO, three time points were chosen for tissue collection: 0, 2 and 8 weeks post-exposure. Three groups of animals exposed to EO were used for the lacI mutagenicity experiment, each containing 15 male 7- to 9-week-old lacI mice. Animals were exposed to constant whole-body target exposures of 50, 100, or 200 ppm EO, respectively, for 4 weeks Ž6 hrday, 5 daysrweek.. An additional four
155
animals per exposure group Ža total of 19 mice per exposure group. were included to determine the hprt T-lymphocyte mutant frequency w21x. Another group containing 19 male lacI mice was exposed to clean air in a separate chamber in the same room. The animals in each exposure group were housed individually in hanging steel cages inside a 1-m3, Hinnersstyle chamber w23x. The mice were deprived of food during the 6-h exposure period. After each 6-h exposure, the mice were fed ŽNIH-07 certified feed. and placed in individual polystyrene shoe box cages until the next exposure period. This housing decreased the possibility that the air-only animals were exposed to EO through exhalation Žoff-gassing. of EO from exposed animals. EO concentrations in the exposure chambers were monitored with a Hewlett-Packard 5890 Series II gas chromatograph equipped with a flame ionization detector. Concentrations of EO were monitored at least once per hour from each of the four chambers. EO concentrations Žin ppm. over the entire 4-week exposure period were 0.0 Ž"0.0., 51.9 Ž"1.6., 102.4 Ž"2.5., and 196.5 Ž"4.3. for the target concentrations of 0, 50, 100, and 200 ppm, respectively. The 19 control mice were exposed to clean air of the same temperature, relative humidity, and airflow as delivered to the EO-exposed animals. Temperature and relative humidity for all exposures were maintained at 728F Ž"48. and 50% Ž"10%., respectively. 2.4. Necropsy and tissue collection Of the 19 animals from each exposure Ž0, 50, 100, or 200 ppm EO. 5 were sacrificed at 1 h or at 2 weeks after the final exposure. Tissues from the 9 remaining animals in each exposure group were collected at 8 weeks after the final exposure. The spleens from four of the 19 remaining animals at the 8-week fixation time point were used to determine the hprt T-lymphocyte mutant frequency described in an accompanying article w21x. The mice were asphyxiated using carbon dioxide and exsanguinated by cardiac puncture. Seven major organs and bone marrow excised from the animals were flash-frozen in liquid nitrogen and subsequently stored at y808C as previously
156
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
described w24x. As part of a collaborative study with Dr. V. Walker ŽWadsworth Center for Laboratories and Research, Albany, NY. to assess the hprt mutant frequency in T-lymphocytes from the same animals exposed to EO, a spleen cell fraction Žboth Band T-lymphocytes. was prepared from air control and EO-exposed mice w4x. Approximately one-half of the preparation was used to determine the hprt mutant frequency in T-lymphocytes w21x, and the remaining half was frozen in liquid nitrogen and stored at y808C for lacI mutant frequency determination. 2.5. lacI mutant frequency determinations in tissues of transgenic mice Genomic DNA was prepared according to the Stratagene Big Blue Transgenic Mouse Mutagenesis Assay System Instruction Manual ŽLa Jolla, CA.. After ethanol precipitation, the DNA was resuspended in 50 to 200 ml TE and allowed to stand at room temperature overnight before storage at 48C. To assess the lacI mutant frequency in germ cells, DNA was extracted from seminiferous tubule germ cells. The lambda shuttle vector containing the lacI transgene was recovered from bulk genomic DNA with lambda phage packaging extract ŽTranspack, Stratagene. used according to the manufacturer’s instructions. Packaged phage were preadsorbed to E. coli SCS-8 cells ŽStratagene. for 20 min at 378C, mixed with prewarmed NZY ŽN-Z-Amine A ŽCasein Enzymatic Hydrolysate.; Yeast Extract. top agar containing 1.3 mgrml X-gal Ždissolved in dimethylformamide., and poured into 22.3- or 25-cm2 square plates containing NZY agar. Plates were incubated overnight at 378C. The total number of colorless lambda phage plaques was estimated for each plate by counting three 1-cm2 sections, averaging these numbers, and multiplying by the area of the plate. Blue lacIy plaques were counted and picked into individual tubes containing 0.5 ml SM buffer and 50 ml chloroform for further characterization. All mutant plaques were confirmed by restreaking on an SCS-8 bacterial lawn on X-gal-containing NZY agar. Mosaic plaques were observed rarely and were not included in the analysis since sectored plaques are likely to be the result of lacI ex-vivo mutation during E. coli repli-
cation or repair Ždiscussed in w25x.. The lacI mutant frequencies were calculated by dividing the number of confirmed mutant plaques by the total number of plaques analyzed. Although five animals per exposure group were included during the in-life portion of the experiment ŽEO exposures., due to the large number of analysis carried in the present study 3–4 animals per data point were typically conducted. In one case Žgerm cells from air control group 8 weeks post-exposure., only two animals were analyzed for the lacI mutant frequency due to sample loss. We routinely analyzed 150 000–500 000 per animal to determine the lacI mutant frequency. 2.6. DNA sequence analysis of lacI The lacI gene was sequenced from lacIy mutants isolated from the T-cell-enriched spleen samples of air control and 200 ppm EO-exposed mice. DNA sequence analysis of lacI mutants was done as described previously w24,26x using a 373A DNA Sequencer ŽApplied Biosystems, Foster City, CA. and SeqEd 675 software for data analysis. The complete lacI gene was sequenced for each mutant, and all mutations were confirmed by two sequencing reactions. No mutants with more than one mutation were found. The first G in the GTG start codon is base 29 w27x. 2.7. Statistics The statistical analyses were conducted using Minitab and Statgraphics software. The statistical analysis were as recommended for the in vivo lacI mutagenicity assay w28x and used previously w26x. Mutant frequency data were log-transformed to normalize variance among samples. Student’s t-test was used to determine statistical significance of mutant frequency in EO-exposed mice relative to air controls in lung, spleen, bone marrow, and germ cells from EO-treated mice relative to controls. The T-cell fraction mutant frequencies that involved four exposure groups were tested by analysis of variance using the log transformed data ŽANOVA.. A value of p - 0.05 was considered significant.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
157
3. Results
3.3. Background mutations
3.1. Determination of lacI mutant frequency in tissues of mice
Twenty-two lacI spleen cell mutants were sequenced from the air control group Žanimals 201, 202, 203, and 204., and 20 unique mutations were identified. The locations and types of mutations are listed in Tables 2 and 3 and summarized in Table 4; indicated bases refer to the coding strand of the lacI gene. Single base substitutions accounted for 95% Ž19r20. of the background mutations; 68% Ž13r19. of these were transitions, and 32% Ž6r19. were transversions ŽTable 4.. Three of the base substitutions changed the original amino acids to stop codons that result in premature termination of protein synthesis. A single-base deletion at G870 that results in a protein truncated at amino acid 285, about 80% of the full-length 360 amino acid lacI protein, was the only frameshift mutation observed among the background mutants ŽTable 3.. This same site, G870 , was also mutated in animal 204, resulting in a G:C ™ A:T transition. Substitutions at G:C base pairs constituted 89% Ž17r19. of the background base substitutions. Seventy-three percent Ž8r11. of the G:C ™ A:T transitions occurred at CpG dinucleotides. G:C ™ A:T transitions at CpG nucleotides were preferentially involved as background mutations in this system w24,29,30x and likely result from deamination of 5-methyl-cytosine to thymine w31x.
The lacI mutant frequency was determined in lung, bone marrow, spleen, and germ cells and in a spleen T-cell fraction from EO-exposed male B6C3F1 lacI transgenic mice. The time from final exposure to euthanasia Žfixation time., total numbers of plaques examined, number of lacI mutant plaques confirmed, and lacI mutant frequencies for these samples are shown in Table 1. All mutant frequency data were normalized by log transformation, and then samples from animals exposed to EO were compared to air control tissue using one-sided, twosample Student’s t-test. Using this analysis, the mutant frequency in lung at 8 weeks post-exposure was significantly increased Ž p s 0.046.. The lacI mutant frequencies were not significantly increased in bone marrow at 2 Ž p s 0.24. and 8 Ž p s 0.07. weeks, in spleen at 2 Ž p s 0.11. and 8 Ž p s 0.49. weeks, or in germ cells at 2 and 8 weeks Ž p s 0.42. post-exposure. The spleen T-cell fraction mutant frequency data, which includes results from 0, 50, 100, and 200 ppm exposures, were compared using ANOVA. The T-cell mutant frequency from mice exposed to 200 ppm EO was 4-fold over controls Ž3.0 " 1.2 vs. 12.4 " 11.8; "standard deviation, SD., and yet the difference was statistically insignificant Ž p s 0.07. by ANOVA, likely due to interanimal variation as indicated by the large SD in the EO-exposed group. 3.2. DNA sequence analysis of lacI mutants from spleen cell fractions The T-cell fraction from two of three animals in the EO-exposed group showed an increased lacI mutant frequency Žanimals 214 and 215; Table 1.; DNA sequence analysis was performed on a collection of lacI mutants isolated from EO-exposed mice and air controls to determine the molecular basis for the observed mutant frequency increase. LacI mutant plaques with identical mutations isolated from a single animal were considered to be mutant siblings arising from one in vivo mutational event w26x.
3.4. Mutations in EO-exposed mice Forty-five lacI mutants were sequenced from the spleen cell fraction of the 200 ppm EO-exposed group Žanimals 213, 214, and 215.; 20 unique mutations were identified ŽTables 2 and 3.. This group was selected for sequencing because it exhibited the highest overall mutant frequency of the three exposed groups, and 2 of 3 animals in this group had increased lacI mutant frequencies. Single base substitutions accounted for 85% Ž17r20. of the mutations in the EO-exposed group; 76% Ž13r17. of these were transitions, and 24% Ž4r17. were transversions. Of the 5 G:C ™ A:T transitions, 80% occurred at CpG sites. One mutation, a G:C ™ T:A transversion, resulted in replacement of Glu 401 with a stop codon.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
158
Table 1 LacI mutant frequency in tissues of B6C3F1 lacI transgenic mice following inhalation exposure to ethylene oxide Animal Exposure Total Žppm. no. plaques
No. of Mutant mutants frequency Ž=10y5 .
Lung (8 weeks) a 11 0 12 0 13 0 14 0
542 000 27 591 000 40 201 000 8 501 000 45
56 57 58 59
478 000 341 000 286 000 206 000
200 200 200 200
47 37 23 16
Bone marrow (2 weeks) 7 0 357 000 7 8 0 316 000 17 9 0 304 000 12 10 0 242 000 2 51 52 53 54
200 200 200 200
324 000 6 337 000 7 313 000 22 240 000 16
Bone marrow (8 weeks) 11 0 285 000 12 0 262 000 13 0 254 000 14 0 225 000 56 57 58 59
200 200 200 200
8 8 8 2
275 000 9 228 000 12 293 000 8 218 000 16
Spleen (2 weeks) 7 0 8 0 9 0 10 0
529 000 12 499 000 8 594 000 41 203 000 12
51 52 53 54
277 000 247 000 242 000 243 000
200 200 200 200
Spleen (8 weeks) 11 0 12 0 13 0 14 0
12 19 36 10
541 000 24 523 000 18 237 000 9 481 000 24
5.0 6.8 4.0 9.0 mean"SD 6.2"2.2 9.8 10.9 8.0 7.8 mean"SD 9.1"1.5 b 2.0 5.4 4.0 0.8 mean"SD 3.0"2.0 1.8 2.1 7.0 6.6 mean"SD 4.4"2.8 2.8 3.0 3.2 0.9 mean"SD 2.5"1.1 3.3 5.3 2.7 7.4 mean"SD 4.7"2.1 2.3 1.6 6.9 5.9 mean"SD 4.2"2.6 4.3 7.7 14.9 4.1 mean"SD 7.8"5.0 4.4 3.4 3.8 5.0 mean"SD 4.2"0.7
Table 1 Žcontinued. Animal no.
Exposure Žppm.
Total plaques
No. of mutants
Mutant frequency Ž=10y5 .
56 57 58 59
200 200 200 200
290 000 290 000 297 000 200 000
15 24 21 2
5.2 8.3 7.1 1.0 mean"SD 5.4"3.2
Germ cells (2 weeks) 7 0 8 0 9 0 10 0
132 000 227 000 256 000 304 000
2 10 5 6
51 52 55
409 000 669 000 600 000
9 12 21
1.5 4.4 2.0 2.0 mean"SD 2.5"1.3 2.2 1.8 3.5 mean"SD 2.5"0.9
Germ cells (8 weeks) 11 0 200 000 13 0 171 000
10 3
57 59 60
200 200 200
200 200 200
197 000 237 000 254 000
8 6 5
Spleen cell fraction (8 weeks) 201 0 221 000 202 0 204 000 203 0 220 000 204 0 232 000
7 6 3 10
206 207 208
50 50 50
211 000 211 000 237 000
8 3 4
209 210 211 212
100 100 100 100
227 000 211 000 169 000 173 450
6 9 4 3
213 214 215
200 200 200
233 000 226 000 241 000
7 58 21
a b
5.0 1.7 mean"SD 3.4"2.3 4.1 2.5 2.0 mean"SD 2.9"1.1 3.2 2.9 1.4 4.3 mean"SD 3.0"1.2 3.8 1.4 1.7 mean"SD 2.3"1.3 2.6 4.3 2.4 1.7 mean"SD 2.8"1.0 3.0 25.6 8.7 mean"SD 12.4"11.8
Post-exposure fixation time following 4 weeks of EO exposure. Significantly greater than air control, Student’s t-test Ž p- 0.05..
Two deletions and one insertion mutation, all isolated from animal 213, were identified ŽTable 3.. The 11 bp deletion at base 978 in animal 213 resulted in a frameshift that encoded a product three amino acids longer than the wild-type protein. The 9
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
159
Table 2 Mutational spectrum in spleen cell fractions isolated from air control and ethylene-oxide-exposed Ž200 ppm. B6C3F1 lacI transgenic mice
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
160
Table 3 Deletions and insertions in lacI recovered from the spleen cell fraction of B6C3F1 lacI transgenic mice: air control mice and mice exposed to 200 ppm ethylene oxide ŽEO.
bp deletion at base 426 in the same mouse did not result in a frameshift, but coded for a product missing three amino acids. The 4 bp insertion in animal 213 was located within a TGGC triple tandem repeat that runs from position 621 to 632; this frameshift mutation introduced a stop codon that truncated the encoded protein at about half the size of wild-type lacI repressor. Insertions and deletions at this triple tandem repeat have been previously observed w24,29x. In the EO-exposed animals 214 and 215, 17% Table 4 LacI mutations recovered from the spleen cell fraction of B6C3F1 lacI transgenic mice: air controls and mice exposed to 200 ppm ethylene oxide Mutational type Point mutations At G:C base pairs GC ™ AT GC ™ TA GC ™CG At A:T base pairs AT ™GC AT ™CG AT ™TA Multibase alterations Tandem change Insertions Deletions Totals
Air control Ž%.
11 Ž55. 5 Ž25. 1 Ž5.
EO-exposed Ž%.
5 Ž24. 6 Ž29. 2 Ž10.
2 Ž10. 0 Ž - 5. 0 Ž - 5.
3 Ž14. 1 Ž5. 1 Ž5.
0 Ž - 5. 0 Ž - 5. 1 Ž5.
0 Ž - 5. 1 Ž5. 2 Ž10.
20 Ž100.
21 Ž100.
Ž3r18. and 55% Ž11r20. of the mutants, respectively, are presumed to represent independent mutations on the basis that lacI mutants with identical mutations isolated from the same animal tissue are mutant siblings arising from one in vivo mutational event w26x. Of the 19 mutants isolated from animal 214 that were sequenced, there are 16 Ž84%. G:C ™ A:T transitions at base 270. In animal 215, a G:C ™ A:T transition at base 777 is present in 9 Ž45%. of the 20 mutants. These G:C ™ A:T transitions at 270 and 777, are at CpG sites, and these same mutations have been observed to occur in the bone marrow of air control animals from this and our previous studies w24x. Table 5 LacI mutant frequencyrmutation frequency Ž=10y5 . in spleen cell fractions isolated from air control and 200 ppm EO-exposed B6C3F1 lacI transgenic mice Animal no.
Exposure
Mutant frequency
air air air air 200 ppm 200 ppm 200 ppm
Mutation frequency
Ž%. Ž n. a
Žppm. 201 202 203 204 213 214 215
Mutation
3.2 2.9 1.4 4.3 3.0 25.6 8.7
100 Ž5. 60 Ž5. 100 Ž2. 78 Ž9. 100 Ž4. 12 Ž18. 45 Ž20.
3.2 1.7 1.4 3.3 3.0 3.1 3.9
a Mutation frequency is the mutant frequency=the percentage of nonsibling mutants observed in each animal; ns number of lacIy mutants sequenced from each animal.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
DNA sequence analysis can be used to assess the contribution of mutant siblings to the in vivo lacI mutant frequency and permits an estimation of the lacI mutation frequency from the lacI mutant frequency w26x. The lacI mutant frequency is multiplied by the percentage of independent mutational events observed in each animal to obtain a mutation frequency. When the DNA sequence data were used to obtain a lacI mutation frequency for air control animals and lacI mice exposed to 200 ppm EO ŽTable 5., the increased lacI mutant frequency observed in animal 214 and 215 was reduced to a mutation frequency that was equivalent to the background observed in air controls.
4. Discussion We have assessed the lacI mutant frequency in somatic and germ cells of male B6C3F1 lacI transgenic mice exposed by inhalation to EO for 4 weeks. Among the tissues examined Žbone marrow, spleen, lung., an increased lacI mutant frequency was observed only in lung. Lung was a target organ for tumor induction in B6C3F1 mice in the NTP carcinogenicity bioassay of EO w12x. We have recently completed the in-life portion a 48-week EO exposure of lacI mice with interim sacrifice points to assess the lacI mutant frequency following longer term exposures. In these studies, we will analyze the lacI mutant frequency in lung to determine if the significant increase in the lung lacI mutant frequency observed in the present study can be reproduced. If elevated, the tissues from the present study and the 48-week exposure will provide the basis to determine the mutational spectrum of EO in a tumor target tissue. EO is a germ cell mutagen in mice, inducing heritable translocations, dominant lethal and specific locus mutations following inhalation exposure w15,32x. In the study presented here, mutant frequencies were assessed in germ cells from male lacI mice exposed by inhalation to 200 ppm EO Ž4 weeks., sampled at 2 and 8 weeks after the last exposure. No increase in the lacI mutant frequency was observed in treated animals over controls. The lack of a response in germ cells of lacI mice could be due to the short fixation time used to determine
161
the lacI mutant frequency after EO exposure. In l lacZ transgenic mice administered ENU Ž150 mgrkg i.p. injection., lacZ mutation occurred only in the stem cell population Ž50–100 days post-exposure. w33x. The 2- and 8-week fixation time following the 4 weeks of EO exposure used in the present study is an insufficient time period to assess EO effects at the lacI transgene in the stem cell population. Studies using longer fixation times following EO exposure than used in the present studies are required to adequately assess the effects of EO on the lacI transgene mutant frequency in testes from transgenic animals. These studies are in progress in our laboratory. In two of three animals exposed to 200 ppm EO, the lacI mutant frequency in the spleen cell fraction was increased 2.5- and 9-fold above the mutant frequency in air controls. DNA sequence analysis revealed the presence of identical mutations in mutant plaques isolated from the same animal. In both cases, a G:C ™ A:T transition at a CpG dinucleotide was the type of mutation identified in multiple isolates. The G:C ™ A:T transition ŽC 270 . identified in animal 214 accounted for 84% of the mutations observed among the lacI mutant plaques recovered. This mutation has been observed previously in bone marrow from BD-exposed mice Žw26x; 1250 ppm exposure group. and in the skin of control and DMBA-exposed mice skin w34x. The G:C ™ A:T transition ŽC 777 . identified in animal 215 accounted for 45% of the mutations among the lacI mutant plaques isolated. This mutation has been observed in two air control mice from a previous study w24x and in skin from untreated mice w34x. Identical mutations among lacI mutants recovered from a single tissue of an animal can be considered to be mutant siblings arising from one in vivo mutation w26x. The lacI mutation frequencies were equivalent in spleen cell fractions from air control and 200 ppm EO-exposed animals ŽTable 5.. These data indicate that the elevated lacI mutant frequencies determined in animals 214 and 215 ŽTable 1. were not due to an increase in the lacI mutation frequency but to clonal expansion of a cell carrying a mutated lacI transgene. Since G:C ™ A:T transitions at CpG dinucleotides occur in approximately 50% of the background mutants isolated from various tissues of B6C3F1 lacI transgenic mice w24,34,35x, the mu-
162
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
tations that were clonally expanded in these animals may likely be background mutations. Therefore the increased lacI mutant frequency in animals 214 and 215 were likely due to the clonal expansion of a background mutation following or during the EO exposure period. Alternatively, these G:C ™ A:T transitions at CpG dinucleotides may represent EO mutational hot spots. However, the nature of the mutation observed ŽG:C ™ A:T transitions at CpG sites. and the lack of detection of these mutations in multiple animals within the EO-exposed group, indicate that these are probably background mutations. The assumption of clonality for all identical mutations observed in a tissue from the same animals is the most conservative approach in estimating the mutation frequency. The available data do not permit the assessment of cellular clonality for further examination of the origin of the lacI mutants. To provide a comparison between hprt mutation in T-lymphocytes and mutation at the lacI transgene, the lacI mutant frequency was determined in the same spleen cell fraction used to determine the hprt T-lymphocyte mutant frequency w21x. Eight weeks after the final exposure to 50, 100, and 200 ppm, hprt mutant frequencies in spleen were increased approximately 1.7-, 3.1-, and 6.4-fold above background in a dose-dependent manner. It is important to note that the hprt assay is performed on T-cells that grow out from the crude fraction, while the lacI lymphocyte fraction mutant frequencies were measured in a population enriched for both B- and T-cells but not necessarily identical to the pool of cells from which hprt mutant frequencies were measured Ži.e. T-cells.. The occurrence of a detectable mutational response at hprt but not at the lacI transgene in the spleen cell fraction is likely due to the mechanism of action of EO-induced mutation. Molecular characterization of EO-induced hprt mutations in diploid human fibroblasts in vitro w36x has indicated that as many as 50% of the hprt mutations induced by EO are large deletions, often involving loss of the entire hprt gene. If large deletions are a characteristic of EO-induced mutations in vivo as well as in vitro, then differences in the hprt and lacI mutant frequencies may be due to the recovery of large deletions as part of the hprt mutant frequency but the lack of recovery of these events at the lacI transgene w37x.
The hprt mutational spectrum for EO determined in human cells in vitro indicates the presence of an increased frequency of both point mutations and large deletions. However, it is uncertain how the frequency of these mutations determined in vitro among EO-exposed cells is reflected in the mutational spectrum determined from tissues in EO-exposed mice. The present studies indicate that a primary mechanism of EO-induced mutagenicity in vivo is likely through the induction of deletions, not specific point mutations. In a comparative study of the mutant frequencies at the endogenous Dlb-1 locus and the lacI transgene after exposure to either ethyl nitrosourea ŽENU. or X-rays w38,39x the two loci responded comparably to ENU. In contrast, X-rays induced relatively few lacI mutations but many Dlb-1 mutations. The difference in response between the lacI transgene was attributed to the difference in mechanism of action for mutation; ENU is primarily a point mutagen, while X-rays can induce a large proportion of deletions that may be too large to be phage-recoverable w37–39x. From these data and others, large deletions appear to be excluded from lacI mutant frequency determination w37x. A mechanistic difference in the two mutation assays Ž hprt vs. lacI . is that the lacI transgenic mouse assay detects primarily base substitutions and small-scale deletions and insertions, while the hprt assay detects these events plus large-scale mutations as well. Therefore comparisons of lacI and hprt mutant frequencies in identically exposed animals can be complicated particularly for the case of EO, since the different mutation assays do not detect identical mutant cell populations. At present, we are conducting a long-term Ž48week. inhalation exposure to EO Žin collaboration with Dr. R.J. Preston, CIIT. in B6C3F1 mice Žgeneric and lacI transgenic., to assess both the induction of mutation at the lacI transgene and chromosomal alterations in the same tissues. In this study, we will reexamine the lacI mutant frequency in the tissues of mice with an increasing exposure time to EO to assess the role of longer term exposure on the lacI mutant frequency. Due to the findings in the present study, we will specifically examine the lacI mutant frequency in lung, spleen and germ cells at increasing exposure levels with time and if appropriate determine the lacI mutational spectrum.
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
In conclusion, we determined the lacI mutant frequency in the tissues of B6C3F1 lacI transgenic mice exposed to the genotoxic carcinogen EO by inhalation. EO was mutagenic in a single tissue, the lung, and did not increase the lacI mutant frequency in other tissues examined. An increased lacI mutant frequency observed in a T-cell fraction obtained from two animals in the high-exposure group was likely due to a clonal expansion of an in vivo background mutation. These data suggest that a predominate mode of in vivo genotoxicity induced by EO Žlarge deletions. are not recoverable at the lacI transgene. The present study emphasizes the need to assess multiple genotoxic end points in the same animals to provide an understanding of in vivo mechanisms of genotoxicity by EO and the relationship between the induction of point mutations detectable at the lacI transgene or hprt vs. chromosomal alterations that occur in experimental animals and humans exposed to EO.
Acknowledgements This work was supported in part by NRSA Grant no. 1F32 CA61722-01 from the National Cancer Institute. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute. We thank Drs. R. Julian Preston ŽCIIT., Tim Fennell ŽCIIT. and Dr. V. Walker ŽWadsworth Center, NY. for discussion of data and reviewing the manuscript. We thank the CIIT inhalation facility Dr. Owen Moss and Mr. Arden James for coordinating and monitoring EO exposures. We thank the CIIT animal care facility and necropsy staff. We thank Dr. Barbara Kuyper for editorial assistance.
References w1x US Environmental Protection Agency ŽUS EPA. Ž1991. Preliminary draft list of categories and subcategories under Section 112 of the Clean Air Act. Fed. Reg., 56, 28548– 28557. w2x V.L. Dellarco, W.M. Generoso, G.A. Sega, J.R. Fowle, III, D. Jacobson-Kram, Review of the mutagenicity of ethylene oxide, Environ. Mol. Mutagen. 16 Ž1990. 85–103. w3x L.E. Appelgren, G. Eneroth, C. Grant, L.E. Lanstrom, K.
w4x
w5x
w6x
w7x
w8x
w9x
w10x
w11x
w12x
w13x
w14x
w15x
w16x
w17x
163
Tenghagen, Testing of ethylene oxide for mutagenicity using the micronucleus test in mice and rats, Acta Pharmacol. Toxicol. 43 Ž1978. 69–71. V.E. Walker, T.R. Skopek, A mouse model for the study of in vivo mutational spectra: sequence specificity of ethylene oxide at the hprt locus, Mutation Res. 288 Ž1993. 151–162. J.W. Yager, R.D. Benz, Sister chromatid exchanges induced in rabbit lymphocytes by ethylene oxide after inhalation exposure, Environ. Mutagen. 4 Ž1982. 121–134. A.D. Kligerman, G.L. Erexson, M.E. Phelps, J.L. Wilmer, Sister chromatid exchange induction in peripheral blood lymphocytes of rats exposed to ethylene oxide by inhalation, Mutation Res. 120 Ž1983. 37–44. D.W. Lynch, T.R. Lewis, W.J. Moorman, J.R. Burg, D.K. Gulati, P. Kaur, P.S. Sabharwal, Sister chromatid exchanges and chromosome aberrations in lymphocytes from monkeys exposed to ethylene oxide and propylene oxide by inhalation, Toxicol. Appl. Pharmacol. 76 Ž1984. 85–95. S.E. Lewis, L.B. Barnett, C. Felton, F.M. Johnson, L.C. Skow, N. Cacheiro, M.D. Shelby, Dominant visible and electrophoretically expressed mutations induced in male mice exposed to ethylene oxide by inhalation, Environ. Mutagen. 8 Ž1986. 867–872. L.R. Ribeiro, M.N. Rabello-Gay, D.M.F. Salvadori, C.A.B. Pereira, Cytogenetic effects of inhaled ethylene oxide in somatic and germ cells of mice, Arch. Toxicol. 59 Ž1987. 332–335. International Agency for Research on Cancer ŽIARC. Ž1994., Some Industrial Chemicals, IARC, Ethylene Oxide, 60, pp. 73-159. W.M. Snellings, C.S. Weil, R.R. Maronpot, A two-year inhalation study of carcinogenic potential of ethylene oxide in Fischer 344 rats, Toxicol. Appl. Pharmacol. 75 Ž1984. 105–117. National Toxicology Program ŽNTP. Ž1988. Toxicology and carcinogenesis studies of ethylene oxide in B6C3F1 mice, in: NTP Technical Report, USDHHS, U.S. Government Printing Office, Washington, DC, vol. 326. W.M. Generoso, K.T. Cain, M. Krishna, C.W. Sheu, R. M Gryder, Heritable translocation and dominant-lethal mutation induction with ethylene oxide in mice, Mutation Res. 73 Ž1980. 133–142. W.M. Generoso, K.T. Kain, L.A. Hughes, G.A. Sega, P.W. Braden, D.G. Gosslee, M.D. Shelby, Ethylene oxide dose and dose-rate effects in the mouse dominant-lethal test, Environ. Mutagen. 8 Ž1986. 1–7. W.M. Generoso, K.T. Cain, C.V. Cornett, N.L.T. Cacherio, L.A. Hughes, Concentration–response curves for ethylene oxide-induced heritable translocations and dominant lethal mutations, Environ. Mol. Mutagen. 16 Ž1990. 126–131. G.A. Sega, P.A. Brimer, E.E. Generoso, Ethylene oxide inhalation at different exposure-rates affects binding levels in mouse germ cells and hemoglobin. Possible explanation for effect, Mutation Res. 249 Ž1991. 339–349. L.R. Ribeiro, D.M.F. Salvadori, C.A.B. Pereira, W. Becak, Activity of ethylene oxide in the sperm morphology test, Arch. Toxicol. 60 Ž1987. 331–333.
164
S.C. Sisk et al.r Mutation Research 391 (1997) 153–164
w18x L. Rhomberg, V.L. Dellarco, C. Siegel-Scott, K.L. Dearfield, D. Jacobson-Kram, Quantitative estimation of the genetic risk associated with the induction of heritable translocations at low-dose exposure: ethylene oxide as an example, Environ. Mol. Mutagen. 16 Ž1990. 104–125. w19x R.J. Preston, T.R. Fennell, A.P. Leber, R.L. Sielken, J.A. Swenberg, Reconsideration of the genetic risk assessment from ethylene oxide exposures, Environ. Mol. Mutagen. 26 Ž1996. 189–202. w20x J.C. Mirsalis, J.A. Monforte, R.A. Winegar, Transgenic animal models for measuring mutations in vivo, Crit. Rev. Toxicol. 24 Ž1994. 255–280. w21x Walker, V.E., L. Recio, S.C. Sisk and T.R. Skopek Ž1997. In vivo mutagenicity of ethylene oxide at the HPRT locus of T-lymphocytes of B6C3F1 lacI transgenic mice following inhalation exposure, submitted to Mutation Res. w22x V.E. Walker, T.R. Fennell, P.B. Upton, J.P. MacNeela, J.A. Swenberg, Molecular dosimetry of DNA and hemoglobin adducts in mice and rats exposed to ethylene oxide, Environ. Health Perspect. 99 Ž1993. 11–17. w23x R.G. Hinners, J.K. Burkart, C.L. Punte, Animal inhalation exposure chambers, Arch. Environ. Health 16 Ž1968. 194– 206. w24x S.C. Sisk, L.J. Pluta, J.A. Bond, L. Recio, Molecular analysis of lacI mutants from bone marrow of B6C3F1 transgenic mice following inhalation exposures to 1,3-butadiene, Carcinogenesis 15 Ž1994. 471–477. w25x W.W. Piegorsch, B.H. Margolin, M.D. Shelby, A. Johnson, J.E. French, R. Tennant, K.R. Tindall, Study design and sample sizes for a lacI transgenic mouse mutation assay, Environ. Mol. Mutagen. 25 Ž1995. 231–245. w26x L. Recio, K.G. Meyer, Increased frequency of mutations at A:T base pairs in the bone marrow of B6C3F1 lacI transgenic mice exposed to 1,3-butadiene, Environ. Mol. Mutagen. 26 Ž1995. 1–8. w27x P.J. Farabaugh, U. Schmeissner, M. Hofer, J.H. Miller, Genetic studies of the lacI repressor, J. Mol. Biol. 126 Ž1978. 847–863. w28x J.D. Callahan, J.M. Short, Transgenic l r lacI mutagenicity assay: statistical determination of sample size, Mutation Res. 327 Ž1995. 201–208. w29x S.W. Kohler, G.S. Provost, A. Fieck, P.L. Kretz, W.O.
w30x
w31x w32x
w33x
w34x
w35x
w36x
w37x
w38x
w39x
Bullock, J.A. Sorge, D.L. Putman, J.M. Short, Spectra of spontaneous and mutagen-induced mutations in the lacI gene in transgenic mice, Proc. Natl. Acad. Sci. USA 88 Ž1991. 7958–7962. G.S. Provost, P.L. Kretz, R.T. Hamner, C.D. Matthews, B.J. Rogers, K.S. Lundberg, M.J. Dydaico, J.M. Short, Transgenic systems for in vivo mutation analysis, Mutation Res. 288 Ž1993. 133–149. A.P. Bird, CpG-rich islands and the function of DNA methylation, Nature 321 Ž1986. 209–213. L.B. Russell, R.B. Cumming, P.R. Hunsicker, Specific-locus mutation rates in the mouse following inhalation of ethylene oxide, and application of the results to estimation of human genetic risk, Mutation Res. 129 Ž1984. 381–388. J.H. van Delft, R.A. Baan RA, Germ cell mutagenesis in lambda lacZ transgenic mice treated with ethylnitrosourea: comparison with the specific locus test, Mutagenesis 10 Ž1995. 209–214. N.J. Gorelick, J.L. Andrews, M. Gu, B.W. Glickman, Mutational spectra in skin from 7,12-dimethylbenzw axanthracenetreated and untreated transgenic mice, Mol. Carcinogen. 14 Ž1995. 53–62. A.D.P. Knoll, P.L. Jacobson, K.S. Kretz, J.M. Lundberg, J.M. Short, S.S. Sommer, Spontaneous mutations in lacIcontaining l lysogens derived from transgenic mice: the observed patterns differ in liver and spleen, Mutation Res. 311 Ž1994. 57–67. T. Bastlova, ´ B. Andersson, B. Lambert, A. Kolman, Molecular analysis of ethylene oxide-induced mutations at the HPRT locus in human diploid fibroblasts, Mutation Res. 287 Ž1993. 283–292. J.A. Gossen, H.-J. Martus, J.Y. Wei, J. Vijg, Spontaneous and X-ray-induced deletion mutations in a lacZ plasmidbased transgenic mouse model, Mutation Res. 331 Ž1995. 89–97. K.S. Tao, C. Urlando, J.A. Heddle, Comparison of somatic mutation in a transgenic versus host locus, Proc. Natl. Acad. Sci. USA 90 Ž1993. 10681–10685. K.S. Tao, C. Urlando, J.A. Heddle, Mutagenicity of methyl methanesulfonate ŽMMS. in vivo at the Dlb-1 native locus and a lacI transgene, Environ. Mol. Mutagen. 22 Ž1993. 293–296.