Mutation Research, 150 (1985) 383-392 383 Elsevier

MTR 02010

Concomitant observations of UDS in the liver and micronuclei in the bone marrow of rats exposed to cyclophosphamide or 2-acetylaminofluorene

J o h n A s h b y * a n d Bri ta Beije ** Imperial Chemical Industries PLC, Central Toxicology Laboratory, Alderley Park, Nr. Maccleshield, Cheshire SKIO 4TJ (Great Britain)

(Received 18 December 1984) (Accepted 4 January 1985)

Summary

Oral dosing of between 5-30 mg/kg of cyclophosphamide (CP) to Alderley Park rats induced micronuclei in the bone marrow between 12 and 36 h after dosing, but failed to induce unscheduled DNA synthesis (UDS) in the liver at similar dose levels and treatment periods. Dose levels of > 30 mg/kg were toxic to the liver. In contrast, 2-acetylaminofluorene (2AAF) induced UDS in the rat liver between 4-36 h after dosing, but gave only a weak response in the bone marrow assay at dose levels between 0.5 and 2 g/kg. Selected observations were made for each chemical using both tissues of the same test animal.

It is concluded that an assessment of the genotoxicity in vivo of chemicals defined as genotoxic in vitro will contribute to an assessment of their possible mammalian carcinogenicity, and that these should involve assays conducted using both the bone marrow and the liver of rodents. Due to its relative ease of commission, the bone marrow micronucleus assay will usually be conducted first; in the case of negative results it is recommended that a liver genotoxicity assay should also be conducted. The case for employing in vivo short-term genotoxicity tests to predict the possible organotropic carcinogenicity or germ cell mutagenicity of a new in vitro genotoxin is discussed.

In an earlier paper, it was argued that short-term genotoxicity assays conducted in vivo had a criti- cal role to play in discerning which in vitro genotoxins are likely to be either carcinogenic or mutagenic in vivo (Ashby, 1983). It was concluded that no single assay in vivo could be relied upon for this purpose since each is associated with a particular tissue or organ, while chemical carcino- gens often show marked organotropic selectivity. It was therefore suggested that such assessments in vivo should commence with a bone marrow micro-

* To whom requests for reprints should be addressed. ** Present address: Division of Cellular Toxicology, Environ-

mental Toxicology Unit, WaUenberg Laboratory, Univer- sity of Stockholm, S-10691 Stockholm (Sweden).

nucleus assay, due to its general availability (Hed- die, 1973; Heddle et al., 1983), and proceed to a liver genotoxicity assay in the case of negative results. We have recently adopted the in vivo liver DNA-repair (UDS) assay described by Mirsalis and Butterworth (1980) for the latter purpose (Ashby et al. 1985). This liver assay was itself based on an earlier in vivo liver UDS test de- scribed by Stich and Kieser (1974) and its in vitro counterpart subsequently described by Williams (1976).

In order to assess the importance of the routine availability within a genotoxicology laboratory of at least 2 organotropically distinct short-term in vivo genotoxicity assays we have evaluated the bone marrow genotoxin cyclophosphamide (CP)

0027-5107/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

384

and the hepatogenotoxin 2-acetylaminofluorene (2AAF) in the same strain of rat using both of the assays discussed above. Selected determinations were made using both genetic end-points in the same animal. We elected to employ a rat bone marrow micronucleus assay, rather than its more generally encountered mouse counterpart, to ob- viate the problem of species-specific sensitivity when comparing the liver and bone marrow data.

Mater ia l s and m e t h o d s

Chemicals Cyclophosphamide (CP) was used as received

(Aldrich) and 2-acetylaminofluorene (2AAF) was as previously described (de Serres and Ashby, 1981). CP was dissolved in water and 2AAF was either dissolved or suspended in corn oil. For the higher dose levels of 2AAF employed it was finely ground and maintained as a suspension in corn oil for 4 h at 70°C. Before dosing the mixture was cooled with concomitant sonication, and finally shaken before gavage dosing.

Animals and dosing Male Alderley Park (AP) rats (9-13 w; 190-250

g) were maintained as described earlier (Ashby et al., 1985; Beije and Ashby, 1985). Dosing was via oral gavage at 9 pm (12 and 36 h exposure), 9 am (24 h exposure) or at 7 am (4 h exposure). Animals were starved for 5 h before dosing and subse- quently overnight in the case of the 12 h experi- ments. Water was available ad libitum in all cases. The concentration of the test chemical in the test vehicle was adjusted to allow dosing at 10 ml/kg bodyweight in all cases. Control animals received the same volume of either water or corn oil.

Bone marrow micronucleus assays These assays were conducted as originally de-

scribed by Schmid (1975). A single femur was aspirated with foetal calf-serum (fcs) and the sus- pension pelleted at 1000 g for 5 min. The pellet was re-suspended in fcs by gentle agitation with a thin glass rod. Following a further centrifugation the pellet was washed with fcs and resuspended in - 0.1 ml of fcs. Smears were prepared and stained with Giemsa using a Haematech automatic stain- ing machine (Ames stain pack). Fluorescent tech-

niques for scoring micronuclei in rats, which avoid the potential problem of confusing mast cell gran- ules with micronuclei, have been described (Mac- Gregor et al., 1983), but these were not employed here. MacGregor et al. demonstrated that assess- ment of the incidence of rat bone marrow micro- nuclei by light microscopy can be as accurate as fluorescent analysis, but in our experience a period of practice is required before the light technique can be relied upon; thus, the fluorescent assay is probably to be encouraged. 1000 polychromatic erythrocytes (PCE) were assessed for micronuclei per animal.

Liver UDS assays These were conducted exactly as described

earlier (Ashby et al., 1985; Beije and Ashby, 1985). Slides were read by eye. Unless indicated other- wise in the Tables, 50 cells of normal morphology were assessed from each of 3 slides per test animal. The present experiments were conducted in paral- lel to those described earlier (Ashby et al., 1985).

Grouping of assay determinations Liver UDS tests were conducted using between

8-12 rats per experiment. In the case of the CP studies, 2AAF acted as the test positive control. A negative control (vehicle-treated) rat was included in each experiment to monitor the technical integ- rity of the assay. The data shown in the Results section therefore represents the collected findings of several separate experiments. In particular, the standard deviations shown reflect both inter-ex- periment and inter-animal assay variability. Simi- lar comments apply to the bone marrow cyto- genetic data. In some instances, the femur was removed from a rat during the perfusion of its liver. In such cases two genotoxic results are re- ported per individual animal (Table 3).

Period of exposure to the test chemicals and control obseroations

UDS in the rat liver is usually assessed between 4 and 12 h after dosing (Mirsalis and Butterworth, 1980; Ashby et al., 1985). In contrast, the presence of micronuclei in PCEs of the bone marrow is usually assessed at 24 h and a later period after dosing (Heddle et al., 1983). These divergent de- mands led to determinations being made between

385

4 and 36 h in the present experiments. Both the liver UDS data and the bone marrow

micronucleus data have been assessed against cumulative control values. In particular, our pre- sent database indicates that control liver net grain counts (NG) are independent of the period of exposure selected or the test vehicle selected (data not shown). Likewise, the incidence of micronuclei in PCEs of the bone marrow of rats treated with either water or corn oil, or sacrified between 12 and 36 h after dosing, does not Vary significantly. We have therefore presented both cumulative con- trol data for both assays derived in the course of the present study, and isolated examples of specific control groups in Table 2 (Results).

marrow of treated rats. 2AAF gave a weak positive response in this assay, but only at elevated dose levels and at the 36 h sampling time (Table 1). In contrast, 2AAF was active in the liver UDS assay over a range of dose levels and exposure periods (Table 2). CP proved toxic to the liver at dose levels > 30 mg/kg, but at this and lower dose levels no evidence of UDS activity was observed (Table 2). Instances where the same treated animal was assessed for both liver UDS and bone marrow micronucleus responses are shown separately in Table 3; these data have also been absorbed into Tables 1 and 2. The combined test data are repre- sented diagrammatically in Fig. 1 (micronucleus assay) and Fig. 2 (liver UDS assay).

Results Discussion

CP gave a dose and time-related increase in the incidence of micronucleated PCEs in the bone

The present experiments have confirmed that CP is a potent bone marrow clastogen when ad-

TABLE 1

INCIDENCE OF M I C R O N U C L E A T E D PCEs IN THE BONE M A R R O W OF RATS T R E A T E D WITH CP OR 2AAF AT THE DOSE LEVELS A N D EXPOSURE PERIODS SHOWN

Significance levels 0.01 (**) and 0.001 (***) by a one-sided Student's t test.

Compound Exposure Dose Number of MN/1000 PCEs (h) (mg/kg) animals (mean _+ SD)

Corn oil 24/36 10 m l / k g 26 1.42 _+ 1.14 Water 1 2 / 2 4 / 3 6 10 m l / k g 11 1.36 _+ 0.92

CP a 12 10 6 5.3 ±1.03 20 6 5.2 ± 1.94

24 10 5 4.8 ± 2.8 15 6 9.8 ±3.9 20 5 8.4 _+ 3.5 30 5 6.2 ± 4.97

36 5 5 3.4 + 0.89 10 11 14.6 ±6.23 15 6 15.5 ±6.12 20 5 9.4 ± 3.9 30 5 5.0 ± 2.35

2AAF 24 2000 6 2.0 ± 1.41

36 50 5 1.6 ±1.14 100 5 2.6 ±2.51 250 8 2.13 ± 0.99 500 19 3.5 5:2.29***

1000 10 3.0 5:1.41"* 2000 10 4.8 +2.04***

a All of the CP data were significant at the 0.001 level.

386

TABLE 2

I N D U C T I O N OF UDS IN HEPATOCYTES ISOLATED F R O M RATS TREATED WITH CP OF 2AAF AT THE DOSE LEVELS A N D EXPOSURE PERIODS SHOWN

Cells in repair are those with a NG value of 5 or more.

Compound Length Dose Number Number Individual Number Cumulative data

of ex- (mg/kg) of of animal of (NG)+-SD % repair (NG) for cells posure cells slides data animals + SD in repair

(h) N G % repair ( N - C >~ 5)+- SD

Water 4 10 m l / k g 150 3 - 3 . 7 7.3 1 - 3 . 7 7.0 10.82

12 10 m l / k g 150 3 - 2 . 0 4.0 1 - 2 . 0 4.0 7.17

Corn oil 12 10 m l / k g 150 3 - 2 . 8 2.7 150 3 - 3.2 1.3 100 2 - 2 . 5 1.0 3 -2 .8+- 0.35 2+- 0.88 5.7+- 0.66

24 10 m l / k g 150 3 - 2 . 6 2.0 150 3 - 2 . 7 2.0 2 - 2 . 6 2.0 7.5

36 10 m l / k g 125 3 - 2 . 4 0.0 150 3 - 1 . 8 5.3 150 3 - 4 . 6 0.7 150 3 - 2 . 4 5.3 4 -2.8_+ 1.24 3+ 2.9 6.0+- 0.89

Historical controls (Corn oil; 4 -24 h) 40 -2 .1+- 1.09 3+ 3.72 6.3+- 1.83

2AAF 4

12

100 150 3 9.6 71.3 75 2 8.9 64.0

0.1 150 3 - 0 . 4 2 150 3 - 0 . 3 0 100 2 0.0 3

1 150 3 0.8 12 125 3 0.1 4 100 2 0.3 4

5 150 3 3.8 41.3 150 3 2.4 24.7 100 2 7.0 66 100 2 10.3 70 150 3 14.7 82.6

50 1 9.7 60

50 100 2 40.5 100 150 3 37.8 97.3 150 3 45.7 98.7 100 2 45.3 96.0 150 3 44.2 96.7 100 2 48.2 100 150 3 43.7 99.3 100 2 40.9 98 100 2 31.8 94

100 100 3 40.5 100 100 3 35.0 100 100 2 64.4 100 150 3 57.7 100 150 3 62.8 100

2 9.2 68 14.1

3 -0 .2+- 0.23 2_+ 1.5 5.9+- 0.2

3 0.3+- 0.58 7+- 4.6 7.4+- 1.4

6 8.0+ 4.5 57+_20.9 12.0+_ 4.2

9 42.0+ 4.97 98__+ 2.00 42.9+ 4.67

5 52.1 + 13.48 100 52.1 + 13.48

TABLE 2 (continued)

387

Compound Length of ex- posure (h)

Dose (mg/kg)

Number of cells

Number of slides

Individual animal data

NG % repair

Number of animals

Cumulative data

(NG) + SD % repair (NG) for cells + SD in repair

( N - C >/5):t:SD

2AAF 24

36

50 150 3 23.1 100 2 18.1 150 3 17.3 150 3 18.4 150 3 19.4

500 150 3 21.6 150 3 21.2

50 150 3 31.3 175 3 21.1 150 3 24.4 150 3 17.9 150 3 28.0

75 3 37.0 75 3 42.8 75 3 25.2

100 75 3 27.2 75 3 42.5

125 2 30.5

250 75 3 31.8 75 3 36.0

500 125 3 41.6 75 3 43.2 75 3 34.8 75 3 50.0

94 96 89.3 92 93.5

96 88.7

93.3 94.9 95.3 88 98

100 100 96

100 100 100

100 100

94 100 98.7

100

5 19.3+ 2.27 93+_ 2.48 20.6-t- 2.26

2 21.4 92 23.2

8 28.5+ 8.28 96_+ 3.93 29.5+ 7.71

3 33.5+ 8.06 100 33.5+ 8.06

2 33.9+ 2.98 100 33.9+ 2.98

4 42.4+ 6.24 98+ 2.85 43.2+ 6.08

CP 4 20 150 3 - 2.2 150 3 -2 .1 150 3 -2 .4 150 3 - 2 . 4 150 3 - 1.4 100 3 -2 .6 150 3 -2 .1 150 3 - 2.0

12 10 150 3 -1 .9 150 3 - 1.7 150 3 - 1.7 150 3 -1 .1

20 50 1 - 0.4 150 3 -2 .2 150 3 - 1.8 50 1 - 2.5

36 5 100 2 - 3.3 150 3 - 1.9

10 75 3 - 2.9 75 3 -2 .5

150 3 -2 .5

9.3 6.7 6.0 2.0 6.7 3.0 5.3 7.3

11.3 2.0 2.0 4.7

10.0 1.3 6.0 0.0

3.0 4.7

0.0 1.3 7.3

8 - 2.1 + 0.38 6_+ 2.36 7.3_+ 0.58

4 - 1 . 6 + 0.34 5+ 4.4 6.5+ 1.02

4 - 1 . 7 + 0.94 4+ 4.57 6.3+ 0.81

2 -2 .6 4 6.6

388

TABLE 2 (continued)

Compound Length of ex- posure (h)

Dose (mg/kg)

Number of cells

Number of slides

Individual animal data

NG % repair

Number Cumulative data

of (NG)+ SD % repair animals ± S D

(NG) for cells in repair ( N - C >/5)+SD

CP 36 10 150 3 - 4.1 150 3 - 5.4 150 3 - 9.2

15 100 3 - 7.0 25 1 - 5.6

30 100 2 - 1.9 50 1 - 1.1

4.0 0.7 9.5 6 - 4 . 5 ± 2.59 4± 3.88 6.0± 1.06

1.0 0.0 2 -6 .3 0.5 5.0

1.0 0.0 2 -1 .5 0.5 7.0

TABLE 3

BONE MARROW MICRONUCLEUS DATA AND LIVER UDS RESPONSE OBSERVED IN THE SAME ANIMAL AT THE TIMES AND DOSE LEVELS SHOWN

These data-points are included as separate observations in the micronucleus and UDS Tables (1 and 2). It may be significant that the negative NG values shown for CP at 15 mg/kg, which are abnormally low by reference to our historical controls, occurred in rats showing a maximal bone marrow cytogenetic response

Test Length Dose Liver Bone marrow chemical of (mg/kg) UDS (MN/1000 PCEs)

exposure (NG) (h)

Corn 12 10 ml /kg -2 .1 2 Oil 36 10 ml /kg -1 .8 0

CP 12 10 -1 .9 5 -1 .7 5 -1 .7 4 -1.1 5

20 - 0.4 7 -2 .2 4 -1 .8 7 -2 .5 6

36 5 - 3.3 3 -1 .9 4

10 -2 .5 9 -4 .1 18 - 5 . 4 22 -9 .1 13

15 -7 .0 13 -5 .6 25

30 - 1.9 7 -1 .1 8

2AAF 36 50 25.2 0

500 41.6 1

ministered to rats by oral gavage (Cihak, 1979). However, it has been shown here to be inactive in the rat liver in vivo UDS assay at similar dose levels. Its hepatotoxicity precluded the conduct of UDS experiments at dose levels above 30 mg/kg . In contrast, these experiments have confirmed 2AAF is a potent inducer of UDS in the rat liver (Mirsalis et al., 1982; Ashby et al., 1985) and established it as only weakly active as a clastogen in the rat bone marrow micronucleus assay under similar conditions of test. This confirms that if short-term assays conducted in rodents in vivo are to be employed to discern significant mammalian genotoxins from among agents defined as geno- toxic in vitro, then at least two organotropically

Fig. 1. Diagrammatic representation of the rat bone marrow micronucleus data for CP and 2AAF at the dose levels and exposure periods indicated (see Table 1). Data for 2AAF are shown in red, for CP in green and the controls in blue. The blue control level has been extended across the test data to indicate the magnitude of the induced genotoxic response. The dose level scale adopted was imposed by space restraints and fails to emphasize adequately the relatively high dose levels at which 2AAF gave a positive assay response.

Fig. 2. Diagrammatic representation of the rat liver UDS data for CP and 2AAF at the dose levels and exposure periods indicated (Table 2). Data for 2AAF are shown in red (see Fig. 1), for CP in green (see Fig. 1) and for the controls in blue. In contrast to Fig. 1, UDS data are not assessed by reference to control levels, but by whether a mean NG value of 3 or greater is observed per datapoint. This arbitrary test criterion (Ashby et al., 1985) is shown as a yellow line across the positive 2AAF

data.

16

I CP

F

ig.

1

I P

AA

F

I C

on

tro

l

14

v)

- C

on

tro

l le

vel

; 12

g 10

76

f 6

60

55

50

45

40

g 35

w 30

g

'5

25

b 20

5 z

15

10 5 0

-5

-10

r 1

CP

Fi

a.

2

i P

AA

F

I C

on

tro

l

Po

siti

ve c

rite

rio

n

(+3

NG

)

I _

390

distinct in vivo assays may be required (Ashby, 1983). Practical problems may accompany at- tempts to assess two genetic endpoints in the same treated animal. In the case of the present two assays their (presumed) different optimum sam- pling times was anticipated to prove a problem, but such did not prove true. Multi-endpoint in vivo assays remain a viable proposition for further study.

It can prove demanding of resources to confirm a negative response in an in vivo assay. For exam- ple, it is possible that were CP to be evaluated in the liver UDS assay using weanling rats (which possess a mitotically active liver), or if 2AAF were to be evaluated for clastogenicity in the bone marrow assay following its intraperitoneal injec- tion (to list two of many possible examples), then each might show a different response. However, the protocols adopted for the present experiments are representative of the general conduct of each test, and therefore, of current short-term in vivo genotoxicity data.

Once the decision is made to evaluate the genotoxicity of a chemical in more than one tissue or organ of rodents in vivo, then the subsidiary question is posed of whether any organ-specific acute responses observed will correlate with chronic organ-specific genotoxic effects such as carcino- genicity. Unfortunately, the data required to assess this question are scant. However, ethyl methane- sulphonate (EMS) and CP are known to induce UDS in rat spermatocytes (Working and Butter- worth, 1984) and each is a germ cell mutagen in rodents (Adler, 1983) (Table 4). In contrast, 2AAF and dimethylnitrosamine (DMN) are reported to be inactive as UDS-initiating agents in spermato- cytes (Working and Butterworth, 1984) and in the rodent forestomach (Furihata et al., 1984), but both are potent UDS-initiating agents in the rat liver (Mirsalis et al., 1982); again, this correlates with their reported carcinogenic profiles. Finally, N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) is reported to elicit a positive UDS response in the rat forestomach (Furihata et al., 1984), but not in the rat liver (Mirsalis et al., 1982), and this corre- lates with its organ-specific carcinogenicity (Taka- yama et al., 1971; Tomatis et al., 1978). These combined reports suggest that the use of short- terms genotoxicity assays conducted in vivo may

TABLE 4

DATA FROM BOTH THE LITERATURE AND THE PRE- SENT PAPER WHICH INDICATE A POSSIBLE LINK BE- TWEEN THE SITES OF RODENT CARCINOGENESIS, THE INDUCTION OF RODENT GERM CELL MUTA- TIONS AND THE OCCURRENCE OF ORGAN-SPECIFIC UDS ACTIVITY IN RATS IN VIVO

MNNG, N-methyl-N'-nitro-N-nitrosoguanidine; 2AAF, 2- acetylaminofluorene; DMN, dimethylnitrosamine; CP, cyclophophamide and EMS, ethyl methanesulphonate.

e: Liver Glandular Germ cells stomach

Cherm'cal '~ C--~ncer UDS Cancer UDS Mutation UDS

M N N G - - ~ + + - 2AAF + + _ _ ?b _

DMN + + . . . . CP - - - + + EMS + +

These negative data (Mirsalis et al., 1982) are worthy of more detailed confirmation. The negative dominant lethal response for DMN is described in Propping et al. (1978). Possible evidence for the germ-cell mutagenicity of 2AAF has recently been described (Harnasch and Stumpf, 1984). Activity was specific to the meiotic division stage, in particu- lar, spermatogonia and spermatocytes were insensitive. No dominant lethal data for 2AAF have been described. The negative UDS response recorded for 2AAF (Working and Butterworth, 1982) was determined on mature spermatocytes, and this is consistent with the observations of Harnasch and Stumpf (1984). The extreme movelty of the histocompatibil- ity mutation assay, coupled to the unusual temporal aspects of the data presented, led us not to regard 2AAF as an established germ-cell mutagen herein.

have a useful role to play when predicting chronic organotropic genotoxic responses (Table 4). Such knowledge could assist the selection of appropriate test species/strains of rodents for cancer bioassays and aid priority setting for agents to be evaluated for germ-cell mutagenic activity.

The reported non-genotoxicity of M N N G to both the mouse bone marrow (Janssen and Ramel, 1980) and rat liver (Mirsalis et al., 1982) points to a weakness of the micronucleus/liver UDS battery of in vivo assays discussed herein. However, the marked organotrophy of this agent as a carcinogen [skin tumours in the case of its topical application (Takayama et al., 1971), glandular stomach tumours in the case of its oral gavage (Tomatis et al., 1978)] is probably due to its hydrolytic insta-

bility (Ashby et al., 1982), a property that can be readily ascertained for a new test chemical in advance of its bioassay in vivo.

It is concluded that genotoxic carcinogens can be discerned in vivo by the sequential use of the rodent bone marrow micronucleus assay and the rat-liver UDS assay, and that these and other acute in vivo assays may have a useful role to play in predicting organotropic genotoxicity: as ex- pressed by Stich et al. (1981) - - 'Particular atten- tion should be focussed on the possibility of adapt- ing UDS assays to estimate the action of carcino- gens in vivo'. Such endeavours would be optimised by the general adoption of a common tester species of rodent. These propositions can be assessed by testing selected chemicals in appropriate in vivo genotoxicity assays: instances for further study are MNNG in the rat spermatocyte UDS assay and CP in the rat forestomach UDS test; negative results would be anticipated by reference to Table 4.

Acknowledgements

We are grateful to Brian Burlinson for permis- sion to present the 0.1 and 1 mg/kg 2AAF liver UDS data, and to Sue Rodger and Jane Steele for secretarial assistance. Brian Elkins provided the artwork. One of us (BB) acknowledges financial support from the Swedish Work Environment Fund.

References

Adler, I.D. (1983) Comparison of types of chemically induced genetic changes in mammals, Mutation Res., 115, 293-321.

Ashby, J. (1983) The unique role of rodents in the detection of possible human mutagens and carcinogens, Mutation Res., 115, 177-213.

Ashby, J., P.A. Lefevre, B.M. Elliott and J.A. Styles (1982) An overview of the chemical and biological reactivity of 4CMB and structurally related compounds, Mutation Res., 100, 417-433.

Ashby, J., P.A. Lefevre and B. Burlinson (1985) An assessment of the in vivo rat hepatocyte DNA repair assay, Mutation Res., in press.

Beije, B., and J. Ashby (1985) Use of an in vivo/in vitro rat liver DNA repair assay to predict the relative rodent hepatocarcinogenic potency of 3 new azo mutagens, Carcinogenesis, in press.

Cihak, R. (1979) Evaluation of benzidine by the micronucleus test, Mutation Res., 67, 383-384.

391

de Serres, F.J., and J. Ashby (Eds.) (1981) Evaluation of Short-Term Tests for Carcinogens, Progress in Mutation Research, Vol. 1, Elsevier, New York.

Ehling, U.H., R.B. Cumming and H.V. Mailing (1968) Induc- tion of dominant lethal mutations by alkylating agents in male mice, Mutation Res., 5, 417-428.

Furihata, C., Y. Yamasaki, S.S. Jin, H. Moriya, K. Kodama, T. Matsushima, T. Ishikawa, S. Takayama and M. Nakadate (1984) Induction of UDS in rat stomach mucosa by glandu- lar stomach carcinogens, J. Natl. Cancer Inst., 72, 1327-1333.

Harnasch, D., and R. Stumpf (1984) Studies on the inductions of histocompatibility gene mutations in germ cells of mice by chemical mutagens, Mutation Res., 126, 279-295.

Heddle, J.A. (1973) A rapid in vivo test for chromosomal damage, Mutation Res., 18, 187-190.

Heddle, J., M. Hite, B. Kirkhart, K. Mavournin, J.T. MacGre- gor, G.W. Newell and M.F. Salamone (1983) The induction of micronuclei as a measure of genotoxicity, Mutation Res., 123, 61-118.

Janssen, D., and C. Ramel (1980) The micronucleus test as part of a short-term mutagenicity test program for the prediction of carcinogenicity evaluated by 143 agents tested, Mutation Res., 75, 191-202.

MacGregor, J.T., C.M. Wehr and R.L. Langlois (1983) A simple fluorescent staining procedure for micronuclei and RNA in erythrocytes using H33258 and pyronin Y, Muta- tion Res., 120, 269-275.

Mirsalis, J.C., and B.E. Butterworth (1980) Detection of un- scheduled DNA synthesis in hepatocytes isolated from rats treated with genotoxic agents, Carcinogenensis, 1, 621-625.

Mirsalis, J.C., C.K. Tyson and B.E. Butterworth (1982) Detec- tion of genotoxic carcinogens in the in vivo/in vitro hepatocyte DNA repair assay, Environ. Mutagen., 4, 553-562.

Propping, P., G. R6hrborn and W. Buselmaier (1972) Com- parative investigations on the chemical induction of point mutations and dominant lethal mutations in mice, Mol. Gen. Genet., 117, 197-209.

Schmid, W. (1975)The micronucleus test, Mutation Res., 31, 9-15.

Stich, H.F., and D. Kaiser (1974) Use of DNA repair synthesis in detecting organotropic actions of chemical carcinogens, Proc. Soc. Exp. Biol. Med., 145, 1339-1342.

Stich, H.F., R.H.C. San and H. Freeman (1981) DNA repair synthesis (UDS) as an in vitro and in vivo bioassay to detect precarcinogens, ultimate carcinogens and organo- tropic carcinogens, in: H.F. Stich and R.H.C. San (Eds.), Short-Term Tests for Chemical Carcinogens, Springer, New York, Chapter 7.

Takayama, S., N. Kuwabara, Y. Azama and T. Sugimura (1971) Skin tumours in mice painted with N-methyl-N'- nitro-N-nitrosoguanidine and N-ethyl-N'-nitro-N-nitro- soguanidine, J. Natl. Cancer Inst., 46, 973-980.

Tomatis, L., C. Agthe, H. Bartsch, J. Huff, R. Montesano, R. Saracci, E. Walker and J. Wilbourne (1978) Evaluation of the carcinogenicity of chemicals: A review of the mono- graph program of the International Agency for Research on Cancer (1971-1977), Cancer Res., 38, 877-885.

392

Williams, G.M. (1976) Carcinogen-induced DNA repair in primary rat liver cell cultures: a possible screen for chemical carcinogens, Cancer Lett., 1,231-236.

Working, P.K., and B.E. Butterworth (1984) An assay to detect chemically induced DNA repair in rat spermatocytes, En- viron. Mutagen., 6, 273-286.