- Email: [email protected]
An improved method for the mouse liver micronucleus test
Mutation Research 391 Ž1997. 49–55
An improved method for the mouse liver micronucleus test Miyuki Igarashi ) , Hiroyasu Shimada Daiichi Pharmaceutical Co. Ltd., 16-13 Kita-Kasai 1-Chome, Edogawa-ku, Tokyo 134, Japan Received 18 April 1996; revised 16 January 1997; accepted 23 January 1997
Abstract We have developed a very practical method for performing the liver micronucleus test in mice. Using this method, we evaluated 11 different types of mutagens, including, 2-acetylaminofluorene, amsacrine, benzene, cyclophosphamide, diethylnitrosamine, 4-dimethylamino-3X-methylazobenzene, N-ethyl-N-nitrosourea, fluorouracil, mitomycin C, potassium chromate ŽVI. and selenious acid. In order to assess the sensitivity of our method, the peripheral blood reticulocyte micronucleus test was performed in the same mouse. Animals were given test chemicals once and underwent partial hepatectomy ŽPH. 24 h later in order to induce mitotic stimulation. Peripheral blood was sampled 0, 24, 48 and 72 h after treatment. The incidence of micronucleated hepatocytes was determined 5 days after PH. As a result, diethylnitrosamine and 4-dimethylamino-3X-methylazobenzene, known as liver carcinogens, increased the incidence of micronucleated cells in the liver only. Positive reactions for benzene, on the other hand, were found in the peripheral blood reticulocytes only. The other chemicals showed positive reaction in the liver and peripheral blood reticulocytes with almost the same maximum response of micronucleus induction. Our method was found to have the advantage over Cliets’ liver micronucleus test in that it required much less time and was easier to perform procedures and highly sensitive in detecting clastogens. It can be used in combination with the peripheral blood reticulocyte micronucleus test to evaluate test chemicals in two tissues, the liver and the bone marrow, in the same animal. We propose a method of combining this test with the peripheral blood reticulocyte micronucleus test for efficient screening for the clastogenic potential of new chemicals in vivo. Keywords: Mouse; Liver micronucleus test; Reticulocyte micronucleus test; Mutagen; Heavy metal; Anticancer drug
1. Introduction The mouse bone marrow micronucleus ŽBMM. test is generally recognized as a screening assay for clastogenic potential in vivo. However, Shimada w1x reported that the carcinogenicity test showed a lower positive concordance with the BMM test Ž47%. than with any of the other mutagenicity tests. This low sensitivity must be due to the target organ, bone
) Corresponding author. Tel.: q81 Ž3. 5696-8294; Fax: q81 Ž3. 5696-8335.
marrow, in which little metabolism occurs, and to which only small amounts of foreign substances are distributed. Actually, in in vivo evaluation of the clastogenic potential in the BMM test, we often obtained the opposite result to the in vitro chromosome aberration test. When a negative response in the BMM test and a positive response in the chromosome aberration test were obtained, mutagenic potential in the liver must be considered. Because almost all chemicals reach the liver and are metabolized to their active forms. However, mutagenicity tests using liver, liver micronucleus ŽLM. test and unscheduled DNA synthesis assay, require much time and labor
1383-5718r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 1 2 1 8 Ž 9 7 . 0 0 0 3 1 - 1
50
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
or use of the radioactive substances. Therefore, we attempted to develop a simple micronucleus test, using the mouse liver, suitable for routine application. The LM test, using mice, was developed by Cliet et al. w2x on the basis of the rat LM test w3x. It consists of 4 phases: chemical treatment, partial hepatectomy ŽPH., isolation of hepatocytes and slide preparation. The Cliet model has two advantages: Ž1. unstable mutagens or those which generate shortlived metabolites can be detected; and Ž2. the results can be readily compared with that of the BMM test. This assay has been hard to perform for routine applications, because it required the cannulation of the thin, breakable portal vein of mice and careful liver perfusion for yielding high viability. We improved these two disadvantages and developed a more practical LM test that meets the following requirements: Ž1. an easy technique for anyone used to the BMM test; Ž2. a short execution time; and Ž3. satisfactory sensitivity, equal to or better than that of the BMM test. We then established the optimal protocol for treatment, PH and sampling. After that, we performed the LM test with 11 mutagens, 2-acetylaminofluorene Ž2-AAF., amsacrine Žm-AMSA., benzene ŽBEN., cyclophosphamide ŽCP., diethylnitrosamine ŽDEN., 4-dimethylamino3X-methylazobenzene Ž3X-Me-DAB ., N-ethyl-Nnitrosourea ŽENU., 5-fluorouracil Ž5-FU., mitomycin C ŽMMC., potassium chromate ŽVI. ŽK 2 CrO4 . and selenious acid ŽH 2 SeO 3 .. Peripheral blood was also sampled in the same animal to monitor the clastogenic effect in bone marrow. As a result, DEN and 3X-Me-DAB were detected in the liver only, BEN was detected in the peripheral blood reticulocytes only. The other chemicals showed positive reactions in the liver and peripheral blood reticulocytes. These results show that the liver micronucleus test, combined with the PBRM test, has a sufficiently high sensitivity to detect clastogens in vivo.
2. Materials and methods 2.1. Animals Male, 6-week-old ddY strain mice were purchased from Japan SLC Inc. ŽHamamatsu, Japan..
They were acclimatized to the environment Žtemperature, 23 " 28C; humidity, 55 " 15%; 12-h lightrdark cycle. for 2 weeks prior to the start of the study. Commercial pellet diet ŽF-2, Funabashi Farm, Funabshi. and tap water were available ad libitum. 2.2. Chemicals and treatment The following chemicals were used: 2-AAF Ž5396-3; Wako Pure Chemical Industries Ltd., Osaka., m-AMSA Ž51264-14-3; Sigma, St. Louis., BEN Ž7143-2; Wako Pure Chemical Industries Ltd.., CP Ž5018-0; Nacalai Tesque Inc., Kyoto., DEN Ž55-18-5; Sigma., 3X-Me-DAB Ž55-80-1; Nacalai Tesque Inc.., ENU Ž759-73-9; Nacalai Tesque Inc.., 5-FU Ž51-218; Wako Pure Chemical Industries Ltd.., MMC Ž5007-7; Kyowa Hakkou Kogyo Co., Ltd., Tokyo., K 2 CrO4 Ž7789-00-6; Wako Pure Chemical Industries Ltd.. and H 2 SeO 3 Ž7783-00-8; Wako Pure Chemical Industries Ltd... CP, DEN, 5-FU, MMC, K 2 CrO4 and H 2 SeO 3 were dissolved in saline; ENU was dissolved in phosphate buffer ŽpH 6.0. and 2-AAF, m-AMSA, BEN and 3X-Me-DAB were suspended in olive oil. Five mice per group were administered the test chemicals by a single i.p. injection, except in the case of BEN, which was administered orally. The dose of each chemical was selected as the highest dose that mice could survive for at least 5 days after PH. The dose volume was a uniform 10 mlrkg b.wt. The control animals received no chemical, but underwent PH. 2.3. Partial hepatectomy In order to stimulate the hepatocytes into mitosis, two-thirds of the liver was resected under ether anesthesia according to the method of Kobayashi and Inoue w4x. The abdominal region was shaved and disinfected with tincture of iodine. Incisions of the midline ventral abdominal skin and muscles, less than 1.5 cm long, were made from just above the xiphoid cartilage. Sterile gauze was placed along the edge of the skin incision to prevent blood from dripping into the abdominal cavity, and the median and left lateral lobes of the liver were then squeezed out of the abdominal cavity through the incision by pushing the lobes gently with the thumb and forefin-
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
ger of each hand. A ligature with a double reef knot was placed tightly around the two lobes and their blood vessels at their base. The gauze was folded over the lobes, which were then raised, placed under slight tension and severed with blunt-ended curved scissors. The incisions were closed with ligatures for muscle and staples for skin. No special therapy was required. The animals usually awoke within 15 min of the operation. 2.4. Isolation of hepatocytes The animals were killed by cervical dislocation. The regenerated liver was removed from the abdominal cavity. Ten ml of 0.019% EGTA Ždissolved in PBSy. was gently injected into the inferior vena cava using a microtip connected with a syringe. A 25-mgrml collagenase type IV ŽSigma. solution Ždissolved in phosphate buffer. was injected continuously in the same way. These solutions were maintained at 378C in advance. Immediately after small cracks appeared on the surface of the liver, the serous membrane was peeled off and the liver was cut into several blocks. Shaking these blocks with fetal bovine serum ŽFBS. in a bottle quickly released hepatocytes from the tissue. FBS prevented the collagenase from having an excessive effect and maintained the normal shape of the hepatocytes. The cell suspension was filtered through a 60-mm nylon mesh and centrifuged at 700 rpm for 2 min. The pellet was then resuspended in FBS and rinsed 3 times in the same way. The final pellet was diluted in FBS and smeared on slides using the same technique as for the BMM test. After being air-dried, slides were fixed with methanol for 5 min and stained in 0.007% acridine orange in pH 6.4 Sorensen phosphate buffer for 3 ¨ min. The slides were then rinsed twice with the same buffer for 5 min. At the analysis of micronucleus, we did not distinguish between single and bi-nucleated cells, which are generally contained in mouse hepatocyte population at 5 days after the PH.
51
to 800 nm long pass.. One thousand intact hepatocytes of normal shape were examined for each animal. For the identification of micronuclei, the following criteria were used: Ž1. a diameter smaller than one-quarter of that of the nucleus; Ž2. a clearly marked perimeter; and Ž3. the same color or luster as the nucleus. 2.6. Micronucleus test with peripheral blood reticulocytes (PBRM test) The incidence of micronuclei in the peripheral blood reticulocytes was determined using the acridine orange ŽAO.-coated slide method w5x. Peripheral blood was obtained by piercing a tail blood vessel and collection with a micropipette. Five ml of blood was placed in the center of an AO-coated glass slide and covered with a 24 = 40 mm coverslip. After 24 h, 1000 peripheral blood reticulocytes with red fluorescence were observed in each animal. Micronuclei were scored only when they were round and exhibited a strong yellow-green fluorescence. 2.7. Statistical analysis Five thousand cells were scored for each dose. The Kastenbaum and Bowman method w6x was used to assess the clastogenic potential. Significance of differences from control was determined at 5% level. 2.8. Experimental design 2.8.1. Experiment 1 To determine the optimal protocol, two procedures were compared by evaluation of clastogenicity of MMC and CP at various interval days. In the pre-PH treatment method, the animals received chemicals 1 day before PH. After 4, 5, 6 or 7 days, hepatocytes were isolated and analyzed. The experimental schedule is shown in Fig. 1A. In the post-PH treatment method, the animals received the chemicals 3, 4 or 5 days after PH, and hepatocytes were isolated 1 day after the chemical treatment ŽFig. 1B..
2.5. Microscopic obserÕation The slides were observed under a fluorescent microscope with blue excitation Žfrom 450 to 480 nm. and a yellow-to-orange barrier filter Žfrom 510
2.8.2. Experiment 2 In order to evaluate the 11 mutagens, we used the following experimental design ŽFig. 1C.: animals were dosed 1 day before PH, and regenerated hepa-
52
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
3. Results 3.1. Experiment 1
Fig. 1. Experimental design. A: Expt. 1. In the pre-PH treatment method, the animals received the chemicals 1 day before PH and hepatocytes were sampled 4, 5, 6 or 7 days after treatment. B: Expt. 1. In the post-PH treatment method, the animals received the chemicals 3, 4 or 5 days after PH and hepatocytes were sampled 1 day after treatment. C: Expts. 2 and 3. For evaluation of 11 mutagens, the animals received the chemicals 1 day before PH. Hepatocytes were sampled 5 days after PH and the peripheral blood was sampled at 0, 24, 48 and 72 h after treatment.
tocytes were isolated 5 days after PH. The incidence of micronucleated hepatocytes ŽMNHPC. was determined as described above. Five ml of peripheral blood was collected from each animal at 0, 24, 48 and 72 h after treatment, and the incidence of micronucleated peripheral blood reticulocytes ŽMNPBR. was determined as described above. To determine the spontaneous incidence of MNHPC, 20 mice were subjected to PH without administration of any of the chemicals. On the 5th day after PH, hepatocytes were isolated. One thousand hepatocytes were taken at random from each animal, and the number of cells containing micronuclei was counted. 2.8.3. Experiment 3 We studied the dose-dependent changes in the incidence of MNHPC on administration of 2-AAF, 3X-Me-DAB or BEN. For each chemical, the maximum tolerated dose against PH and one half of that dose were employed.
In order to determine which experimental design would yield the maximum incidence of micronucleated hepatocytes, two procedures, one with pre-PH, the other with post-PH treatment, were compared by evaluation of the clastogenicity of MMC and CP after various time intervals. The highest incidence of MNHPC following either MMC or CP treatment was observed with the pre-PH treatment method on examination 5 days after PH, slight decreases being seen on the 6th day, and major 7th day decreases occurring to below the level of the 4th day ŽFig. 2.. In the post-PH treatment method, there were no significant changes in the incidence of MNHPC at any sampling time. The maximum incidence of MNHPC after MMC or CP treatment was lower than that of the pre-PH treatment method ŽFig. 2.. We therefore employed the pre-PH treatment method, in which there is a 5-day interval between PH and hepatocyte isolation. 3.2. Experiment 2 We studied the clastogenic potential of 11 mutagens in two tissues, the liver and the peripheral blood in the same animal. The results are summarized in
Fig. 2. Determination of optimal experimental design ŽExpt. 1.. A: pre-PH treatment method. Mice received a single i.p. injection of MMC Ž1 mgrkg, gray bars; or CP Ž20 mgrkg, hatched bars. 1 day before PH. Regenerated hepatocytes were isolated 4, 5, 6 and 7 days after PH. B: post-PH treatment method. Mice received a single i.p. injection of MMC Ž1 mgrkg, gray bars. or CP Ž20 mgrkg, hatched bars. 3, 4 and 5 days after PH. Regenerated hepatocytes were isolated 1 day after treatment Ž4, 5 and 6 days after PH.. MNHPC, micronucleated hepatocytes. Error bars represent SD.
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
Fig. 3. The spontaneous incidence of MNHPC was 0.41 " 0.22%. All chemicals that were positive in the peripheral blood showed their maximum incidence at 48 h after treatment. Administration of DEN and 3X-Me-DAB, the liver carcinogens, induced no significant increase in the incidence of micronuclei in the peripheral blood, but significantly increased the incidence of MNHPC. BEN, which is myelotoxic and causes myelogenous leukemia, was positive in the peripheral blood following 2000 mgrkg p.o., but not in the liver. The other chemicals, 2-AAF, mAMSA, CP, ENU, 5-FU, MMC, K 2 CrO4 and
53
H 2 SeO 3 , were detected in both tissues, and there was no significant difference in the maximum incidence of micronuclei between the peripheral blood and the hepatocytes. 3.3. Experiment 3 In order to determine if this improved method would be sensitive to dose change, we evaluated lower doses of 2-AAF, 3X-Me-DAB and BEN ŽTable 1.. The higher dose was found to induce higher incidence of MNHPC than the half dose: 1.6 times as
Fig. 3. The incidence of micronucleated cells in peripheral blood reticulocytes Žgray bars. and liver Žhatched bars.. Evaluation of 11 mutagens. A single i.p. injection of each chemical was given to groups of 5 mice. From each animal, peripheral blood was sampled at 0, 24, 48 and 72 h after treatment, and the liver was sampled on the 5th day after partial hepatectomy Ž6 days after treatment.. Error bars represent SD.
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
54
Table 1 Dose-dependent changes in the liver micronucleus test Compound Control 2-AAF X
3 -Me-DAB BEN
Dose Žmgrkg. 100 200 100 200 1000 2000
No. of animals
MNHPC Ž%, mean"SD.
20 5 5 5 5 5 5
0.41"0.22 1.33"0.49 a 2.20"0.55 a 1.81"0.30 a 3.39"1.38 a 0.38"0.15 0.53"0.29
a
Significant difference from controls ŽKastenbaum and Bowman method w6x, p- 0.05..
high for 2-AAF, 1.9 times for 3X-Me-DAB and 1.4 times for BEN.
4. Discussion Many improvements of the liver micronucleus test were reported in the 1980s, but none recently. However, the merit of the LM test is still valid, which is as a means for resolving conflicting results of the BMM test with other mutagenicity tests. Therefore, we developed a simple LM test which is easy to perform for routine application. In order to establish the experimental schedule, we used MMC and CP. Our choice resulted from our suspicion that the optimal intervals might be different between direct mutagens like MMC and indirect mutagens like CP. However, these chemicals induced the maximum responses in the incidence of MNHPC on the 5th day after PH. Nevertheless, the optimal intervals between PH and sampling may differ according to the chemical, because of differences in pharmacokinetics, metabolism and detoxification. In the pre-PH treatment, test chemicals were distributed and metabolized in healthy animals for 24 h. This is enough time for chemicals to induce preclastogenic lesions in DNA, because clastogens, including active metabolites, react with DNA very quickly after treatment. This new method employed the simple techniques of isolation and slide preparation. Although Tates et al. w3x stated that rough handling of hepatocytes led to cell rupture, resulting in a considerable loss of scorable cells, we never experienced such a problem.
The difference may have been due to differences in collagenase treatment time and in FBS rinsing. With our method, we isolated hepatocytes more quickly than by perfusion with a perista pump, and FBS can prevent excessive collagenase action and maintain the shape of the hepatocytes. However, our method yielded very low viability Žapproximately 2%, data not shown. of isolated hepatocytes, measured by dye-exclusion test using Trypan blue, micronucleus test needs intact cells, but not high viability, because isolated cells are not cultured in the LM test. Another laborious technique is partial hepatectomy. PH was performed for stimulating hepatocytes to proliferate which allows the expression of preclastogenic lesions as micronuclei. In order to save labor and reduce the stress on the animals, Ashby and Braithwaite w7x used 4-acetylaminofluorene Ž4-AAF. instead of PH. They reported that 4-AAF, used as a mitogen, elevated the incidence of MNHPC seen in untreated and in the positive control, dimethylnitrosamine. Another chemical mitogen, carbon tetrachloride, was used by Uryvaeva and Delone w8x, and they showed that the effects of carbon tetrachloride on the spontaneous incidence of MNHPC did not differ from those of PH. Chemical mitogens and PH have merits and demerits, respectively. PH requires considerable time and labor, but it is necessary to avoid the use of any agent other than the test chemical that may affect the results. For the same reason, we used ether anesthesia for PH. The use of chemical mitogen is good for control of hepatocytes proliferation rate. The results of Expt. 1 showed that the incidence of MNHPC was affected by the proliferation rate that changed dependent on the interval days after PH. We compared the sensitivity of our method with that of the BMM test using the PBRM test. The PBRM test enables the time-dependent clastogenic effect to be monitored in bone marrow, with a delay of 24 h w9,10x. Cliet et al. w2x showed the LM test can detect many clastogens for which the BMM test was negative. We obtained the same results in the groups treated with DEN and 3X-Me-DAB. The other 8 chemicals showed almost the same maximum response of micronucleus induction in the liver and peripheral blood reticulocytes, although the amount distributed to the liver and that of the bone marrow must be different. The incidence of MNHPC would
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
be determined by clastogenic potential of test chemical, proliferating rate of target cells or tissue, pharmacokinetics and the other complicated factors. This study showed that some chemicals that have been metabolized in special ways in terminal or specified restricted tissue, such as BEN, cannot be detected by the LM test alone. We suppose this is the limitation of the LM test. In conclusion, our results show that a combination of the liver and the peripheral blood reticulocyte micronucleus test is currently the best way to evaluate clastogenic potential in vivo. Furthermore, our method allowed one experienced person to isolate hepatocytes and prepare slides in 5 h for 30 mice. When conflicting results for the in vitro assay and the in vivo BMM test are obtained, our simple liver micronucleus test is valuable.
References w1x H. Shimada, Mutagenicity test battery for toxicity studies of drugs: status quo and perspective, Environ. Mutagen. Res. Commun. 15 Ž1. Ž1993. 109–121. w2x I. Cliet, E. Fournier, C. Melcion, A. Cordier, In vivo micronucleus test using mouse hepatocytes, Mutation Res. 216 Ž1989. 321–326.
55
w3x A.D. Tates, I. Neuteboom, M. Hofker, L. den Engelse, A micronucleus technique for detecting clastogenic effects of mutagensrcarcinogens ŽDEN, DMN. in hepatocytes of rat liver in vivo, Mutation Res. 74 Ž1980. 11–20. w4x Kobayashi, Y. and T. Inoue Ž1981. Partial hepatectomy, in: T. Inoue and K. Matsumoto ŽEds.., Illustrated Techniques of Animal Experiments, Kyouritsu Syuppan, Tokyo, pp. 34–42. w5x M. Hayashi, T. Morita, Y. Kodama, T. Sofuni, M. Ishidate Jr., The micronucleus assay with mouse peripheral blood reticulocytes using acridine orange-coated slides, Mutation Res. 245 Ž1990. 245–249. w6x M.A. Kastenbaum, K.O. Bowman, Tables for determining the statistical significance of mutation frequencies, Mutation Res. 9 Ž1970. 527–549. w7x J. Ashby, I. Braithwaite, A non-invasive micronucleus assay in the rat liver, Mutation Res. 203 Ž1988. 23–32. w8x I.V. Uryvaeva, G.V. Delone, An improved method of mouse liver micronucleus analysis: an application to age-related genetic alteration and polyploidy study, Mutation Res. 334 Ž1995. 71–80. w9x J.T. MacGregor, C.M. Wehr, D.H. Gould, Clastogen-induced micronuclei in peripheral blood erythrocytes: the basis of an improved micronucleus test, Environ. Mutagen. 2 Ž1980. 509–514. w10x J.T. MacGregor, C.M. Wehr, P.R. Henika, M.D. Shelby, The in vivo erythrocyte micronucleus test: measurement at steady state increases assay efficiency and permits integration with toxicity studies, Fund. Appl. Toxicol. 14 Ž1990. 513–522.
An improved method for the mouse liver micronucleus test Miyuki Igarashi ) , Hiroyasu Shimada Daiichi Pharmaceutical Co. Ltd., 16-13 Kita-Kasai 1-Chome, Edogawa-ku, Tokyo 134, Japan Received 18 April 1996; revised 16 January 1997; accepted 23 January 1997
Abstract We have developed a very practical method for performing the liver micronucleus test in mice. Using this method, we evaluated 11 different types of mutagens, including, 2-acetylaminofluorene, amsacrine, benzene, cyclophosphamide, diethylnitrosamine, 4-dimethylamino-3X-methylazobenzene, N-ethyl-N-nitrosourea, fluorouracil, mitomycin C, potassium chromate ŽVI. and selenious acid. In order to assess the sensitivity of our method, the peripheral blood reticulocyte micronucleus test was performed in the same mouse. Animals were given test chemicals once and underwent partial hepatectomy ŽPH. 24 h later in order to induce mitotic stimulation. Peripheral blood was sampled 0, 24, 48 and 72 h after treatment. The incidence of micronucleated hepatocytes was determined 5 days after PH. As a result, diethylnitrosamine and 4-dimethylamino-3X-methylazobenzene, known as liver carcinogens, increased the incidence of micronucleated cells in the liver only. Positive reactions for benzene, on the other hand, were found in the peripheral blood reticulocytes only. The other chemicals showed positive reaction in the liver and peripheral blood reticulocytes with almost the same maximum response of micronucleus induction. Our method was found to have the advantage over Cliets’ liver micronucleus test in that it required much less time and was easier to perform procedures and highly sensitive in detecting clastogens. It can be used in combination with the peripheral blood reticulocyte micronucleus test to evaluate test chemicals in two tissues, the liver and the bone marrow, in the same animal. We propose a method of combining this test with the peripheral blood reticulocyte micronucleus test for efficient screening for the clastogenic potential of new chemicals in vivo. Keywords: Mouse; Liver micronucleus test; Reticulocyte micronucleus test; Mutagen; Heavy metal; Anticancer drug
1. Introduction The mouse bone marrow micronucleus ŽBMM. test is generally recognized as a screening assay for clastogenic potential in vivo. However, Shimada w1x reported that the carcinogenicity test showed a lower positive concordance with the BMM test Ž47%. than with any of the other mutagenicity tests. This low sensitivity must be due to the target organ, bone
) Corresponding author. Tel.: q81 Ž3. 5696-8294; Fax: q81 Ž3. 5696-8335.
marrow, in which little metabolism occurs, and to which only small amounts of foreign substances are distributed. Actually, in in vivo evaluation of the clastogenic potential in the BMM test, we often obtained the opposite result to the in vitro chromosome aberration test. When a negative response in the BMM test and a positive response in the chromosome aberration test were obtained, mutagenic potential in the liver must be considered. Because almost all chemicals reach the liver and are metabolized to their active forms. However, mutagenicity tests using liver, liver micronucleus ŽLM. test and unscheduled DNA synthesis assay, require much time and labor
1383-5718r97r$17.00 Copyright q 1997 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 1 2 1 8 Ž 9 7 . 0 0 0 3 1 - 1
50
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
or use of the radioactive substances. Therefore, we attempted to develop a simple micronucleus test, using the mouse liver, suitable for routine application. The LM test, using mice, was developed by Cliet et al. w2x on the basis of the rat LM test w3x. It consists of 4 phases: chemical treatment, partial hepatectomy ŽPH., isolation of hepatocytes and slide preparation. The Cliet model has two advantages: Ž1. unstable mutagens or those which generate shortlived metabolites can be detected; and Ž2. the results can be readily compared with that of the BMM test. This assay has been hard to perform for routine applications, because it required the cannulation of the thin, breakable portal vein of mice and careful liver perfusion for yielding high viability. We improved these two disadvantages and developed a more practical LM test that meets the following requirements: Ž1. an easy technique for anyone used to the BMM test; Ž2. a short execution time; and Ž3. satisfactory sensitivity, equal to or better than that of the BMM test. We then established the optimal protocol for treatment, PH and sampling. After that, we performed the LM test with 11 mutagens, 2-acetylaminofluorene Ž2-AAF., amsacrine Žm-AMSA., benzene ŽBEN., cyclophosphamide ŽCP., diethylnitrosamine ŽDEN., 4-dimethylamino3X-methylazobenzene Ž3X-Me-DAB ., N-ethyl-Nnitrosourea ŽENU., 5-fluorouracil Ž5-FU., mitomycin C ŽMMC., potassium chromate ŽVI. ŽK 2 CrO4 . and selenious acid ŽH 2 SeO 3 .. Peripheral blood was also sampled in the same animal to monitor the clastogenic effect in bone marrow. As a result, DEN and 3X-Me-DAB were detected in the liver only, BEN was detected in the peripheral blood reticulocytes only. The other chemicals showed positive reactions in the liver and peripheral blood reticulocytes. These results show that the liver micronucleus test, combined with the PBRM test, has a sufficiently high sensitivity to detect clastogens in vivo.
2. Materials and methods 2.1. Animals Male, 6-week-old ddY strain mice were purchased from Japan SLC Inc. ŽHamamatsu, Japan..
They were acclimatized to the environment Žtemperature, 23 " 28C; humidity, 55 " 15%; 12-h lightrdark cycle. for 2 weeks prior to the start of the study. Commercial pellet diet ŽF-2, Funabashi Farm, Funabshi. and tap water were available ad libitum. 2.2. Chemicals and treatment The following chemicals were used: 2-AAF Ž5396-3; Wako Pure Chemical Industries Ltd., Osaka., m-AMSA Ž51264-14-3; Sigma, St. Louis., BEN Ž7143-2; Wako Pure Chemical Industries Ltd.., CP Ž5018-0; Nacalai Tesque Inc., Kyoto., DEN Ž55-18-5; Sigma., 3X-Me-DAB Ž55-80-1; Nacalai Tesque Inc.., ENU Ž759-73-9; Nacalai Tesque Inc.., 5-FU Ž51-218; Wako Pure Chemical Industries Ltd.., MMC Ž5007-7; Kyowa Hakkou Kogyo Co., Ltd., Tokyo., K 2 CrO4 Ž7789-00-6; Wako Pure Chemical Industries Ltd.. and H 2 SeO 3 Ž7783-00-8; Wako Pure Chemical Industries Ltd... CP, DEN, 5-FU, MMC, K 2 CrO4 and H 2 SeO 3 were dissolved in saline; ENU was dissolved in phosphate buffer ŽpH 6.0. and 2-AAF, m-AMSA, BEN and 3X-Me-DAB were suspended in olive oil. Five mice per group were administered the test chemicals by a single i.p. injection, except in the case of BEN, which was administered orally. The dose of each chemical was selected as the highest dose that mice could survive for at least 5 days after PH. The dose volume was a uniform 10 mlrkg b.wt. The control animals received no chemical, but underwent PH. 2.3. Partial hepatectomy In order to stimulate the hepatocytes into mitosis, two-thirds of the liver was resected under ether anesthesia according to the method of Kobayashi and Inoue w4x. The abdominal region was shaved and disinfected with tincture of iodine. Incisions of the midline ventral abdominal skin and muscles, less than 1.5 cm long, were made from just above the xiphoid cartilage. Sterile gauze was placed along the edge of the skin incision to prevent blood from dripping into the abdominal cavity, and the median and left lateral lobes of the liver were then squeezed out of the abdominal cavity through the incision by pushing the lobes gently with the thumb and forefin-
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
ger of each hand. A ligature with a double reef knot was placed tightly around the two lobes and their blood vessels at their base. The gauze was folded over the lobes, which were then raised, placed under slight tension and severed with blunt-ended curved scissors. The incisions were closed with ligatures for muscle and staples for skin. No special therapy was required. The animals usually awoke within 15 min of the operation. 2.4. Isolation of hepatocytes The animals were killed by cervical dislocation. The regenerated liver was removed from the abdominal cavity. Ten ml of 0.019% EGTA Ždissolved in PBSy. was gently injected into the inferior vena cava using a microtip connected with a syringe. A 25-mgrml collagenase type IV ŽSigma. solution Ždissolved in phosphate buffer. was injected continuously in the same way. These solutions were maintained at 378C in advance. Immediately after small cracks appeared on the surface of the liver, the serous membrane was peeled off and the liver was cut into several blocks. Shaking these blocks with fetal bovine serum ŽFBS. in a bottle quickly released hepatocytes from the tissue. FBS prevented the collagenase from having an excessive effect and maintained the normal shape of the hepatocytes. The cell suspension was filtered through a 60-mm nylon mesh and centrifuged at 700 rpm for 2 min. The pellet was then resuspended in FBS and rinsed 3 times in the same way. The final pellet was diluted in FBS and smeared on slides using the same technique as for the BMM test. After being air-dried, slides were fixed with methanol for 5 min and stained in 0.007% acridine orange in pH 6.4 Sorensen phosphate buffer for 3 ¨ min. The slides were then rinsed twice with the same buffer for 5 min. At the analysis of micronucleus, we did not distinguish between single and bi-nucleated cells, which are generally contained in mouse hepatocyte population at 5 days after the PH.
51
to 800 nm long pass.. One thousand intact hepatocytes of normal shape were examined for each animal. For the identification of micronuclei, the following criteria were used: Ž1. a diameter smaller than one-quarter of that of the nucleus; Ž2. a clearly marked perimeter; and Ž3. the same color or luster as the nucleus. 2.6. Micronucleus test with peripheral blood reticulocytes (PBRM test) The incidence of micronuclei in the peripheral blood reticulocytes was determined using the acridine orange ŽAO.-coated slide method w5x. Peripheral blood was obtained by piercing a tail blood vessel and collection with a micropipette. Five ml of blood was placed in the center of an AO-coated glass slide and covered with a 24 = 40 mm coverslip. After 24 h, 1000 peripheral blood reticulocytes with red fluorescence were observed in each animal. Micronuclei were scored only when they were round and exhibited a strong yellow-green fluorescence. 2.7. Statistical analysis Five thousand cells were scored for each dose. The Kastenbaum and Bowman method w6x was used to assess the clastogenic potential. Significance of differences from control was determined at 5% level. 2.8. Experimental design 2.8.1. Experiment 1 To determine the optimal protocol, two procedures were compared by evaluation of clastogenicity of MMC and CP at various interval days. In the pre-PH treatment method, the animals received chemicals 1 day before PH. After 4, 5, 6 or 7 days, hepatocytes were isolated and analyzed. The experimental schedule is shown in Fig. 1A. In the post-PH treatment method, the animals received the chemicals 3, 4 or 5 days after PH, and hepatocytes were isolated 1 day after the chemical treatment ŽFig. 1B..
2.5. Microscopic obserÕation The slides were observed under a fluorescent microscope with blue excitation Žfrom 450 to 480 nm. and a yellow-to-orange barrier filter Žfrom 510
2.8.2. Experiment 2 In order to evaluate the 11 mutagens, we used the following experimental design ŽFig. 1C.: animals were dosed 1 day before PH, and regenerated hepa-
52
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
3. Results 3.1. Experiment 1
Fig. 1. Experimental design. A: Expt. 1. In the pre-PH treatment method, the animals received the chemicals 1 day before PH and hepatocytes were sampled 4, 5, 6 or 7 days after treatment. B: Expt. 1. In the post-PH treatment method, the animals received the chemicals 3, 4 or 5 days after PH and hepatocytes were sampled 1 day after treatment. C: Expts. 2 and 3. For evaluation of 11 mutagens, the animals received the chemicals 1 day before PH. Hepatocytes were sampled 5 days after PH and the peripheral blood was sampled at 0, 24, 48 and 72 h after treatment.
tocytes were isolated 5 days after PH. The incidence of micronucleated hepatocytes ŽMNHPC. was determined as described above. Five ml of peripheral blood was collected from each animal at 0, 24, 48 and 72 h after treatment, and the incidence of micronucleated peripheral blood reticulocytes ŽMNPBR. was determined as described above. To determine the spontaneous incidence of MNHPC, 20 mice were subjected to PH without administration of any of the chemicals. On the 5th day after PH, hepatocytes were isolated. One thousand hepatocytes were taken at random from each animal, and the number of cells containing micronuclei was counted. 2.8.3. Experiment 3 We studied the dose-dependent changes in the incidence of MNHPC on administration of 2-AAF, 3X-Me-DAB or BEN. For each chemical, the maximum tolerated dose against PH and one half of that dose were employed.
In order to determine which experimental design would yield the maximum incidence of micronucleated hepatocytes, two procedures, one with pre-PH, the other with post-PH treatment, were compared by evaluation of the clastogenicity of MMC and CP after various time intervals. The highest incidence of MNHPC following either MMC or CP treatment was observed with the pre-PH treatment method on examination 5 days after PH, slight decreases being seen on the 6th day, and major 7th day decreases occurring to below the level of the 4th day ŽFig. 2.. In the post-PH treatment method, there were no significant changes in the incidence of MNHPC at any sampling time. The maximum incidence of MNHPC after MMC or CP treatment was lower than that of the pre-PH treatment method ŽFig. 2.. We therefore employed the pre-PH treatment method, in which there is a 5-day interval between PH and hepatocyte isolation. 3.2. Experiment 2 We studied the clastogenic potential of 11 mutagens in two tissues, the liver and the peripheral blood in the same animal. The results are summarized in
Fig. 2. Determination of optimal experimental design ŽExpt. 1.. A: pre-PH treatment method. Mice received a single i.p. injection of MMC Ž1 mgrkg, gray bars; or CP Ž20 mgrkg, hatched bars. 1 day before PH. Regenerated hepatocytes were isolated 4, 5, 6 and 7 days after PH. B: post-PH treatment method. Mice received a single i.p. injection of MMC Ž1 mgrkg, gray bars. or CP Ž20 mgrkg, hatched bars. 3, 4 and 5 days after PH. Regenerated hepatocytes were isolated 1 day after treatment Ž4, 5 and 6 days after PH.. MNHPC, micronucleated hepatocytes. Error bars represent SD.
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
Fig. 3. The spontaneous incidence of MNHPC was 0.41 " 0.22%. All chemicals that were positive in the peripheral blood showed their maximum incidence at 48 h after treatment. Administration of DEN and 3X-Me-DAB, the liver carcinogens, induced no significant increase in the incidence of micronuclei in the peripheral blood, but significantly increased the incidence of MNHPC. BEN, which is myelotoxic and causes myelogenous leukemia, was positive in the peripheral blood following 2000 mgrkg p.o., but not in the liver. The other chemicals, 2-AAF, mAMSA, CP, ENU, 5-FU, MMC, K 2 CrO4 and
53
H 2 SeO 3 , were detected in both tissues, and there was no significant difference in the maximum incidence of micronuclei between the peripheral blood and the hepatocytes. 3.3. Experiment 3 In order to determine if this improved method would be sensitive to dose change, we evaluated lower doses of 2-AAF, 3X-Me-DAB and BEN ŽTable 1.. The higher dose was found to induce higher incidence of MNHPC than the half dose: 1.6 times as
Fig. 3. The incidence of micronucleated cells in peripheral blood reticulocytes Žgray bars. and liver Žhatched bars.. Evaluation of 11 mutagens. A single i.p. injection of each chemical was given to groups of 5 mice. From each animal, peripheral blood was sampled at 0, 24, 48 and 72 h after treatment, and the liver was sampled on the 5th day after partial hepatectomy Ž6 days after treatment.. Error bars represent SD.
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
54
Table 1 Dose-dependent changes in the liver micronucleus test Compound Control 2-AAF X
3 -Me-DAB BEN
Dose Žmgrkg. 100 200 100 200 1000 2000
No. of animals
MNHPC Ž%, mean"SD.
20 5 5 5 5 5 5
0.41"0.22 1.33"0.49 a 2.20"0.55 a 1.81"0.30 a 3.39"1.38 a 0.38"0.15 0.53"0.29
a
Significant difference from controls ŽKastenbaum and Bowman method w6x, p- 0.05..
high for 2-AAF, 1.9 times for 3X-Me-DAB and 1.4 times for BEN.
4. Discussion Many improvements of the liver micronucleus test were reported in the 1980s, but none recently. However, the merit of the LM test is still valid, which is as a means for resolving conflicting results of the BMM test with other mutagenicity tests. Therefore, we developed a simple LM test which is easy to perform for routine application. In order to establish the experimental schedule, we used MMC and CP. Our choice resulted from our suspicion that the optimal intervals might be different between direct mutagens like MMC and indirect mutagens like CP. However, these chemicals induced the maximum responses in the incidence of MNHPC on the 5th day after PH. Nevertheless, the optimal intervals between PH and sampling may differ according to the chemical, because of differences in pharmacokinetics, metabolism and detoxification. In the pre-PH treatment, test chemicals were distributed and metabolized in healthy animals for 24 h. This is enough time for chemicals to induce preclastogenic lesions in DNA, because clastogens, including active metabolites, react with DNA very quickly after treatment. This new method employed the simple techniques of isolation and slide preparation. Although Tates et al. w3x stated that rough handling of hepatocytes led to cell rupture, resulting in a considerable loss of scorable cells, we never experienced such a problem.
The difference may have been due to differences in collagenase treatment time and in FBS rinsing. With our method, we isolated hepatocytes more quickly than by perfusion with a perista pump, and FBS can prevent excessive collagenase action and maintain the shape of the hepatocytes. However, our method yielded very low viability Žapproximately 2%, data not shown. of isolated hepatocytes, measured by dye-exclusion test using Trypan blue, micronucleus test needs intact cells, but not high viability, because isolated cells are not cultured in the LM test. Another laborious technique is partial hepatectomy. PH was performed for stimulating hepatocytes to proliferate which allows the expression of preclastogenic lesions as micronuclei. In order to save labor and reduce the stress on the animals, Ashby and Braithwaite w7x used 4-acetylaminofluorene Ž4-AAF. instead of PH. They reported that 4-AAF, used as a mitogen, elevated the incidence of MNHPC seen in untreated and in the positive control, dimethylnitrosamine. Another chemical mitogen, carbon tetrachloride, was used by Uryvaeva and Delone w8x, and they showed that the effects of carbon tetrachloride on the spontaneous incidence of MNHPC did not differ from those of PH. Chemical mitogens and PH have merits and demerits, respectively. PH requires considerable time and labor, but it is necessary to avoid the use of any agent other than the test chemical that may affect the results. For the same reason, we used ether anesthesia for PH. The use of chemical mitogen is good for control of hepatocytes proliferation rate. The results of Expt. 1 showed that the incidence of MNHPC was affected by the proliferation rate that changed dependent on the interval days after PH. We compared the sensitivity of our method with that of the BMM test using the PBRM test. The PBRM test enables the time-dependent clastogenic effect to be monitored in bone marrow, with a delay of 24 h w9,10x. Cliet et al. w2x showed the LM test can detect many clastogens for which the BMM test was negative. We obtained the same results in the groups treated with DEN and 3X-Me-DAB. The other 8 chemicals showed almost the same maximum response of micronucleus induction in the liver and peripheral blood reticulocytes, although the amount distributed to the liver and that of the bone marrow must be different. The incidence of MNHPC would
M. Igarashi, H. Shimadar Mutation Research 391 (1997) 49–55
be determined by clastogenic potential of test chemical, proliferating rate of target cells or tissue, pharmacokinetics and the other complicated factors. This study showed that some chemicals that have been metabolized in special ways in terminal or specified restricted tissue, such as BEN, cannot be detected by the LM test alone. We suppose this is the limitation of the LM test. In conclusion, our results show that a combination of the liver and the peripheral blood reticulocyte micronucleus test is currently the best way to evaluate clastogenic potential in vivo. Furthermore, our method allowed one experienced person to isolate hepatocytes and prepare slides in 5 h for 30 mice. When conflicting results for the in vitro assay and the in vivo BMM test are obtained, our simple liver micronucleus test is valuable.
References w1x H. Shimada, Mutagenicity test battery for toxicity studies of drugs: status quo and perspective, Environ. Mutagen. Res. Commun. 15 Ž1. Ž1993. 109–121. w2x I. Cliet, E. Fournier, C. Melcion, A. Cordier, In vivo micronucleus test using mouse hepatocytes, Mutation Res. 216 Ž1989. 321–326.
55
w3x A.D. Tates, I. Neuteboom, M. Hofker, L. den Engelse, A micronucleus technique for detecting clastogenic effects of mutagensrcarcinogens ŽDEN, DMN. in hepatocytes of rat liver in vivo, Mutation Res. 74 Ž1980. 11–20. w4x Kobayashi, Y. and T. Inoue Ž1981. Partial hepatectomy, in: T. Inoue and K. Matsumoto ŽEds.., Illustrated Techniques of Animal Experiments, Kyouritsu Syuppan, Tokyo, pp. 34–42. w5x M. Hayashi, T. Morita, Y. Kodama, T. Sofuni, M. Ishidate Jr., The micronucleus assay with mouse peripheral blood reticulocytes using acridine orange-coated slides, Mutation Res. 245 Ž1990. 245–249. w6x M.A. Kastenbaum, K.O. Bowman, Tables for determining the statistical significance of mutation frequencies, Mutation Res. 9 Ž1970. 527–549. w7x J. Ashby, I. Braithwaite, A non-invasive micronucleus assay in the rat liver, Mutation Res. 203 Ž1988. 23–32. w8x I.V. Uryvaeva, G.V. Delone, An improved method of mouse liver micronucleus analysis: an application to age-related genetic alteration and polyploidy study, Mutation Res. 334 Ž1995. 71–80. w9x J.T. MacGregor, C.M. Wehr, D.H. Gould, Clastogen-induced micronuclei in peripheral blood erythrocytes: the basis of an improved micronucleus test, Environ. Mutagen. 2 Ž1980. 509–514. w10x J.T. MacGregor, C.M. Wehr, P.R. Henika, M.D. Shelby, The in vivo erythrocyte micronucleus test: measurement at steady state increases assay efficiency and permits integration with toxicity studies, Fund. Appl. Toxicol. 14 Ž1990. 513–522.