Cong. Anom., 28: 15 7 16 7, 1988 Original

Effects of Antimutagens on the Teratogenicity of N-Methyl­

N'-Nitro-N-Nitrosoguanidine in Mice

Yasunobu MORITA and Masahiro MIZUTANI Hatano Research Institute, Food and Drug Safety C'rntcr, Ochiai, Hadano, Kana­ gawa 257, Japan

ABSTRACT We examined whether vanillin (VA) and CoC'l 2 -6H 2 0(CoC1 2 ), anti­ mutagens, which have mutation suppressing effect, i.e., promotion of cellular repair function in vitro, can modify the teratogenicity in mice caused by N-methyl­ N'-nitro-N-nitrosoguanidine (MNNG), a direct-acting monofunctional alkylating agent. ICR mice were treated with MNNG alone (single IP dose of 40 or 60 mg/kg) or in combination with the antimutagen on day 11 of gestation. Embryotoxicity and teratogenicity were examined at term. The incidence of MNNG-induced syndactyly in the fore- and hindlimbs was significantly decreased by VA (50 mg/kg, IP) or

CoC1 2 (IO mg/kg, IV) and a tendency to decrease in the incidence of oligodactyly was noted as well. On the other hand, the incidence of MNNG-induced brachy­ dactyly was increased by VA or CoC1 2 •

Though the mechanism of the modifying effects of both VA and CoC'l 2 on MNNG-induced malformations could not be delineated in the present study, the results indicate that the antimutagens which stimulate DNA recombination repair in vitro modify the manifestation of malformations caused by teratogens that attack the fetal DNA in the initial teratogenic mechanism. Key words: N-methyl-N'-nitro-N-nitrosoguanidine, antirnutagens, vanillin, CoCl 2 , fetal development, malformations, modification, mice

N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) is a direct-acting alkylating agent that possesses mutagenic activity (Mandell and Greenberg, 1960; Hsu et al., 1977) and carcinogenic activity (Sugi­ rnura et al., 1966 ). Further study by Inouye and Murakami (I 978) has demonstrated that this compound is also teratogenic in mice. Maternal intraperitoneal administration of 60 mg/kg of MNNG on day 11 of gestation in mice induced brain, facial, vertebral, rib and limb malformations in the offspring. Digital defects were induced also at a higher incidence (Manson and Marian, 1983). At this dose (60 mg/kg), MNNG induced less lethal damage to the fetus compared to fetal mortality caused by other alkylating agents

Received February 3, 1988 i"iilll*'f,t, Jkti-iE"(L (11-t Ht,\/,'l,\/,'ii:·'t'-t:;,, ?-40HJf'}):riJr, 9'257 4-f'l'frfi{f;{:-n\) :i 158 Y. Morita and M. Mizutani such as MMS (t11ethyl methanesulfonate). Recently. antimutagens capable of suppressing cellular mutagenesis in the bacterial system have hcen reported ( Kada and Kanematsu, 1978 ). Among those antimutagens cobaltous chloride (CoCl 2 •

<,11 2 0) ( Kiuchi ct al.. I 984) and vanillin (Ohta ct al., 1986) are categorized as bio-antimutagens that suppress t11utagcncsis by promoting cellular repair function after inducing primary DNA damage (Kada ct al.. 1986). MNNG treatment (0.25 µg/ml) of cultured FM3A cells induced mutation result­

ing in 6-thioguaninc resistant clones. When CoCl 2 -6H 2 0 was present (at a concentration of 1.0 µg/ml) in the culture medium of FM3A cells during the whole mutation expression time, after treat­ t11cnt with MNNG for 2 hours. the frequency of 6-thioguanine resistant clones was slightly reduced ( Unpublished data by authors). More recently. it has been reported that post-treatment of vanillin reduced the formation of hrcakagc-type chromosome aherrations of mitomycin C (MMC)-trcated Chinese hamster ovary cells. despite the increase in frequency of MMC-induced SC Es (Sasaki et al., 1987). It is generally accepted that mutagenicity and carcinogenicity arc highly correlated with each other in terms of short-term assay for carcinogenicity (Kawachi et al., 1980). However there is no evident correlation between either of these two forms of toxicity and teratogcnicity due to much more complex factors involved in the initial process of abnormal development. Nevertheless, alkylat­ ing agents found to be both mutagcnic and carcinogenic have also been proven to be teratogenic when appropriately tested (Wilson, 1972). As for the teratogenic mechanism. especially of alkylating agents, it has been proposed that genetically transmitted mutational mechanisms not involving cell death can profoundly influence the course of morphogenesis. although cell death plays an important role in the occurrence of birth defects (Manson, 1981; Neubert, 1980; Platzck et al., 1982). In this study, we examined whether substances which have been shown to be antimutagenic in vitro (e.g., CoCJ 2 -6H 2 O and vanillin) can modify the tcratogenicity caused by MNNG in mice.

MATERIALS AND METHODS

The animals used in this study were Sic: ICR mice purchased from Shizuoka Agricultural Co­ operative Association for Laboratory Animals. The mice were housed in plastic cages with wood chip bedding and kept in a temperature (24±1 °C) - and humidity (55±5 %) - controlled room with a fixed 12-hour lighting schedule throughout the year. Food (laboratory chow CA-I, CLEA Japan, Inc.) and water were allowed ad libitum. Estrous females (9-13 weeks old) were housed overnight with males in pairs. The next morning, female mice with copulation plugs were designated to be on day O of gestation. Treatment: To determine the effect of antimutagens on fetal development, mice on day 11 of ges­ tation at noon were separately given a single intraperitoneal dose of 50 mg/kg of vanillin (VA, Tokyo

Kasei Industries. Inc.) or single intravenous dose of 10 mg/kg of CoCl 2 -6H2 O(CoC1 2 , Wako Pure Chemical Industries. Inc.). In another experiment, 2 groups of mice on day 11 of gestation (at noon) were given intraperi­ toneal dose of 40 mg/kg and 60 mg/kg of N-methyl-N'-nitro-N-nitrosoguanidine (MNNG, Aldrich Chemical Company. Inc.). respectively. Each group of MNNG-treated mice was divided into 2 sub- Effects of antimutagcns on the teratogenicity of MNNG 159 groups. A single dose (50 mg/kg, IP) of VA was administered to one sub-group (MNNG+V A group) and CoC1 2 (IO mg/kg, IV) to the other sub-group (MNNG+CoC1 2 group) one hour after MNNG treatment. Saline was used as diluent for MNNG, VA and CoC1 2 • Control groups for this study were set up as follows: one control group was administered saline only (NSS, 10 ml/kg, IP) and the other control group was given saline (same dose, IP) one hour after MNNG treatment (MNNG+saline group). Observations: On day 18 of gestation, the dams were humanely killed and the number of im­ plants, resorptions, dead fetuses and live fetuses were counted. Live fetuses were weighed and ex­ amined for external malformations under a dissecting microscope, and then cleared and stained by means of Dawson's technique (Dawson, 1926) for skeletal examination. Furthermore, for examina­ tion of the phalanges stained cartilagious and ossified components in the fore- and hindlimbs, some of the specimens stained with alizarin red S were subsequently stained with toluidine blue according to the modified method of Burdi (Burdi, 1965). Statistics: Data obtained herein were calculated on the basis of the litters and analyzed by means of the following statistical methods. Fetal body weight and number of live fetuses were analyzed using student's t-test. Fetal mortality and incidence of fetal malformations and variations were analyzed using Wilcoxon's rank sum test.

RESULTS

The effects of the· antimutagens, VA and CoC1 2 , on fetal development in mice are shown in Table I. Effects of VA and CoCl2 were comparable to that of the saline control with respect to the number of live fetuses, fetal body weight, fetal mortality and incidence of external and skeletal abnormalities. The influence of these antimutagens on MNNG-induced fetal defects is shown in Table 2. Body weight of fetuses from dams to which doses of 60 mg/kg of MNNG had been administered showed very considerable reduction with or without the antimutagens. Cleft palate and digital malformations of the fore- and hindlimbs were observed with a correlation to dose rate of MNNG. At a dose of 60 mg/kg of MNNG, these malformations appeared in a high frequency. As a brain malformation, flatness at the top of the head was noted in fetuses of all treated groups following the administration of 60 mg/kg of MNNG. This malformation appeared to correspond to microcephaly. The incidence of external malformations in MNNG+VA group was comparable to that in MNNG+saline group. The incidence of microcephaly and facial defects (cleft palate, microtia etc.), however, showed a tendency to decrease compared to MNNG (60 mg/kg)+saline group. In MNNG (60 mg/kg)+CoCl2 group, fetuses with external malformations were less than those in MNNG (60 mg/kg)+saline group. As for individual malformations, the incidence of microcephaly and facial defects (cleft palate, microtia etc.) was significantly decreased and that of digital malformations in the fore- and hindlimbs showed a tendency to decrease compared to MNNG (60 mg/kg)+saline group. Fig. 1 shows the influence of VA and CoCl2 on the incidence of MNNG-induced cleft palate and digital malformations (oligodactyly, brachydactyly and syndactyly) in the fore- and hindlimbs. The incidence of syndactyly in the fore- and hindlimbs was significantly decreased both in MNNG (60 mg/kg)+ VA or MNNG (60 mg/kg)+ CoCl2 groups, with a decrease in oligodactyly also being 160 Y. Morita and M. Mizutani

Table l Influence of vanillin and CoC1 1 on fetal development in mice

External abnormalities (% )a Skeletal abnormalities (%)a No. Mean no. Fetal Fetal Ex- Treat- of of live body mortal- ternal a ment litters fetuses weighta ity abn. Brain Face Others Malfor- Variationse (g) (%)a (%)a mations

Saline 8 13.6 1.34 5.1 0 0 0 0 l.Oc 34.4 ± 0.6 ±0.06 ±1.8 ±l.0 ± 6.0

VA 3 13.7 1.42 9.6 0 0 0 0 0 25.7 so ± 1.8 ±0.06 ±4.8 ± 17.4 mg/kg

1.6b CoCl 1 4 12.8 1.39 4.6 1.6 0 0 4_9d 51.4 10 ± 2.7 ±0.05 ± 1.6 ±1.6 ± 1.6 0 ±l.1 ± I 2.5 mg/kg

Vanillin (VA) or CoC11 was administered on day 11 of gestation. a: Data were calculated on the basis of the litter as sample unit and given as mean ± S.E. b: Cleft palate c: Fusion of the cervical vt. arches d: llypoplasia of the palatine bone, Hypoplasia of the cervical vt. arches e: Cervical ribs, Wavy ribs, Lumbar ribs, Variations in number of the lumbar vertebrae, Epiphyscal bifurcation of the cervical vt. arch, A supernumerary sternebra

noted. In MNNG (40 mg/kg)+CoCl 2 group, decrease of syndactyly as well as oligodactyly and brachydactyly was also observed. In MNNG (60 mg/kg)+VA group, the incidence of cleft palate was also significantly decreased, while that of brachydactyly in the fore- and hindlimbs was higher than in MNNG (60 mg/kg)+saline group. Oligodactyly (Od) was observed predominantly in the fifth digit and occasionally in the first digit, and syndactyly (Sd) predominantly in the axial digits. Brachydactyly (Bd) existed in nearly all five digits. It is known that syndactyly induced by teratogen usually occures in association with other digital malformations such as polydactyly and oligodactyly. In the present study, syndactyly was noted to be associated with oligodactyly. Fig. 2 shows the incidence of syndactyly (Sd only), syndactyly with other digital malformations (+Od, +Bd, +Od and Bd) and digital malformations except for syndactyly (Od only, Bd only, Od + Bd) in the hindlimbs in the groups treated with MNNG (60 mg/kg)+saline, +VA and +CoCl2. The incidence of syndactyly associated with oligo­ dactyly was significantly decreased in MNNG+V A and MNNG+CoCl2 groups (Fig. 2 B, C), while a decrease in the incidence of oligodactyly alone was noted in MNNG+VA group (Fig. 2 E). The incidence of brachydactyly alone showed significant increase in MNNG+V A group (Fig. 2 E), and a similar tendency was also noted in MNNG+CoCl2 group (Fig. 2 F). Skeletal abnormalities (malformations and variations) in fetuses from dams administered saline, VA or CoCl2 after receiving 40 mg/kg of MNNG are shown in Table 3. In MNNG (40 mg/kg)+CoCl 2 group, the incidence of skeletal malformations including thinning of the metacarpal bones and Effects of antimutagens on the teratogenicity of MNNG 161

Table 2 Influence of post-treatment with vanillin and CoCI, on MNNG-induced teratogencsis in mice

External abnormalities (%)a No. Mean no. Fetal Fetal External Treatment of of live body mortal- abn. a litters fetuses weight(g)a ity (%)a (%)a Brainb Facec Forelimbd Hindlimbe Other/

MNNG Saline 6 13.2 1.23 12.1 34.1 0 6.9 24.9 28.7 0 40 mg/kg ± 1.0 ±0.05 ± 3.6 ± 6.6 ±4.5 ± 7.1 ± 6.2

VA 5 13.6 1.21 10.2 30.4 0 4.0 27.5 27.7 1.3 50 mg/kg ± 0.6 ±0.04 ± 2.6 ±10.2 ±2.7 ± 8.4 ±10.1 ± 1.3

CoCI, 7 11.9 1.27 14.0 25.8 0 6.0 17.2 21.0 0 10 mg/kg ± I. I ±0.03 ± 5.5 ± 7.7 ±2.3 ± 6.6 ± 6.8

MNNG Saline 5 13.6 0.94 14.8 63.2 11.3 38.9 52.5 55.7 34.8 60 mg/kg ± 1.2 ±0.09 ± 9.1 ±10.5 ± 4.1 ± 6.4 ±10.6 ± 9.7 ± 6.8

VA 5 12.4 1.06 11.5 67.5 1.8 25.6 56.2 67.5 26.4 50 mg/kg ± 0.5 ±0.06 ± 1.8 ± 11.2 ±1.8 ± 6.1 ± 10.2 ± 11.2 ± 9.9

CoCl2 9 13.0 0.90 11.8 48.6 4.6* 19.3** 37.7 44.7 13.2** 10 mg/kg ± 0.9 ±0.04 ± 4.0 ± 7.8 ± 1.9 ± 5.8 ± 6.1 t 7.5 t 4.2

MNNG was administered on day 11 of gestation and VA or CoCl 2 was administered one hour later. *· Significantly different from MNNG+saline group, p < 0.05. **· Significantly different from MNNG+saline group, p < 0.01. a: Data were calculated on the basis of the litter as sample unit and given as mean ± S.E. b: Microcephaly (Flatness at the top of the head) c: Cleft palate, Microtia, Open eyelids d: Oligodactyly, Brachydactyly, Syndactyly e: Micromelia, Adactyly, Oligodactyly, Brachydactyly, Syndactyly, Maldirection of digit f: Short tail, Anal atresia, Constriction of anus

retarded ossification of the phalanges was decreased in contrast to MNNG (40 mg/kg)+salinc group. In MNNG (40 mg/kg)+VA group, no decreasing effect on the incidence of skeletal malformations and retarded ossifications was observed. Syndactyly observed externally in the fore- and hindlimbs was classified into parallel fusion of adjacent phalanges and narrow interdigital space without bony fusion. Fig. 3 shows the incidence of skeletal malformations of digits examined in bone-cartilage specimens double-stained with alizarin

red Sand toluidine blue in both MNNG (60 mg/kg)+CoCl2 and MNNG (60 mg/kg)+saline groups.

In MNNG (60 mg/kg)+CoCl2 group, the incidence of narrow interdigital space without bony fusion was decreased only in the hindlimbs, in contrast to MNNG (60 mg/kg)+saline group. No Parallel fusion of adjacent phalanges was observed in the forelimbs and its incidence in the hindlimbs was significantly decreased (1.1%) compared to that in MNNG (60 mg/kg)+saline group (24.2%). 162 Y. Morita and M. Mizutani

MNNG 40 mg/kg 30 Forelimb Hlndlimb -'I? - 20 Cp Od Sd Sd "'C .2 cu - 10 ..E ,2 5 ii E MNNG 60 mg/kg -0 * Q) 70 u C 60 Q) "0 0d ·u 50 Sd Sd .5 40 30 20 10 ** ** D Saline VA 50m91k9 CoClz 10mg/kg

Fig. I Influence of vanillin and CoCl2 on MNNG-induced cleft palate and digital malformations in mice. MNNG was administered at an intraperitoneal dose of 40 or 60 mg/kg on day 11 of gestation. One hour after

) MNNG-treatment, saline, vanillin (VA) IP 50 mg/kg or CoCl2 ·6H2 O(C0Cl2 IV 10 mg/kg was administered. The fetuses were examined on day 18 of gestation. Cp: cleft palate, Od: oligodactyly, Bd: brachydactyly, Sd: syndactyly •: Significantly different from MNNG+saline group, P < 0.05 .. : Significantly different from MNNG+saline group, P < 0.01

DISCUSSION

In this study, the incidence of MNNG-induced syndactyly in the fore- and hindlimbs was markedly decreased by administration of vanillin (VA) or CoC1 2 , and also noted decrease in other malforma­ tions, such as microcephaly, cleft palate and oligodactyly. The suppressive effect of CoCI 2 was more remarkable than that of VA.

The suppressive effect of VA and CoCl 2 on the incidence of both syndactyly and oligodactyly supports the assumption that these two types of digital malformation are pathogenetically related each other. In the limb buds of fetuses from dams administered MNNG on day 11 of gestation, mesenchymal cells appeared necrotic, particularly in the distal region of the limb, and mesenchymal condensation was disrupted, predisposing abnormality of the digital rays (Manson and Marian, 1983 ). Effects of antimutagens on the teratogenicity of MNNG 163

30 A D MN NG-Saline 20

10

B MN NG-VA E '#- ** : 40 = Q •••••.. -... •···• E 30 :•:•: Q ····· :::... ::::: E ·•··••··•· .; 20 :-:-:·=·=· -~~ ·=·=· "Cl ···•••••••••··• ~ 10 ... ····· u ·•·••~ g O O ** 0 0 * ::::: ~ 0-1--,.___,______._.....,•Lo•Lo-~i-_._...__...._~.~ ..... _ ...... _._~ u .s C F MNNG-CoCl2

20

10

** 0 -'----'-..;_...__...mwa0 _ _.___.__..._....___.__0 0 Sd only Sd + 0d Sd + Bd Sd + 0d + Bd 0d only Bd only 0d + Bd

Fig. 2 Detailed classification of MNNG (60 mg/kg)-induced digital malformations in the hindlimbs showed in Fig. I. The left section of this figure shows the incidence of syndactyly with or without other digital malformations in the hindlimbs of fetuses in each treated group, and the right section shows the incidence of another digital malformations except for syndactyly. Sd: syndactyly, Od: oligodactyly, Bd: brachydactyly *: Significantly different from MNNG+saline group, P < 0.05 *"'· Significantly different from MNNG+saline group, P < 0.01

The administration of VA or CoC)z might possibly suppress the MNNG-induced mesenchymal tissue damage, causing the approach of normal morphogenesis of the digital rays. Findings on phalanges with stained cartilagious and ossified components in the fore- and hindlimbs 164 Y. Morita and M. Mizutani

Table 3 lntluence of vanillin and CoCl2 on the incidence of MNNG-induced fetal skeletal malformations in mice

MNNG 40 mg/kg

+ Saline +VA50mg/kg + CoCl 2 IO mg/kg

No. of litters 6 5 7 No. of specimens 78 68 82 Malformations ('if,)a 33.9 ± 3.2 39.8 ± 9.4 26.4 ± 7.5

Types and frequencies (%)a Hypoplasia of the palatine bone 3.6 ± 3.6 1.3 ± 1.3 2.5 ± 1.7 Thinning of the ribs 3.3 ± 3.3 0 0.9 ± 1.2 Hypoplasia of the ishium 2.9 ± 1.8 0 1.6 ± 1.6 Thinning of the metacarpal bones 23.0 ± 4.6 24.6 ± 9.4 9.3 ± 3.5 Thinning of the metatarsal bones 28.2 ± 5.2 36.9 ±9,7 24.0 ± 6.2 b Others 2.4 ± 2.4 0 0

Retarded ossification Unossified cervical vertebral body (%)a 10.2 ± 6.9 5.5 ± 2.5 2.6 ± 1.7 Unossificd middle and proximal phalanges in the hindlimbs (%)a 20.4 ± 4.9 15.5 ± 9.8 9.7 ± 3.4 No. of ossified sacral and caudal vertebrae 10.5 ± 0.7 10.5 ± 0.7 10.9 ± 0.5

a: Data were calculated on the basis of the litter as sample unit and given as mean ± S.E. b: Fusion of the sterncbrae, Absence of the fibula

indicated the incidence of hypoplasia of phalanges in MNNG+CoCl2 group to be essentially the same as that in MNNG+saline group (60 mg/kg of MNNG). Fetal body weight in MNNG+VA and MNNG+ CoCl2 groups was almost the same as that of MNNG+saline group, though all groups showed remark­ able growth retardation. It therefore appears that the administration of VA or CoCI 2 could not recover effectively the MNNG-induced cell deficit in the limb buds. The administration of VA or CoCl2 following treatment with 60 mg/kg of MNNG caused a greater incidence of brachydactyly, in contrast to the suppression of syndactyly and oligodactyly. This increase of brachydactyly was more predominant by post-treatment of VA than CoCl2. Day 11 of gestation in mice has been known to correspond to the later period of limb bud de­ velopment, including the sensitive period for reduction deformity of the limbs. The mechanism of teratogen-induced brachydactyly has yet to be determined. Takehira and Kameyama ( 1981) reported that the sensitive period for 5-FU and Ara-C induced brachydactyly was about one day latter than the period for syndactyly and oligodactyly, and the brachydactyly was generally characterized by symphalangism with shortness or absence of the middle phalange. Though phalangial abnormalities could not be definitely classified in this study, the observed hypoplasia of phalanges may possibly involve thinning of ph_alanges, shortness or the absence of middle phalange and symphalangism. Additional study should be conducted for confirmation of this point. When mice were treated with 60 mg/kg of MNNG on day 12 of gestation, the resulting digital malformations were mostly brachy- Effects of antimutagens on the teratogenicity of MNNG 165

Forelimb

60

::,!:! -0 50 -Cl) 40 C 0 30 =IQ E... 20 .E 10 i6 E 0 Absence of Hypoplasia Parallel fusion Narrow interdigital Thinning of Retarded ouificetion 0 phalanges of phalanges of adjacent space without the metacarpal of phalanges - phalanges bony fusion bones Q) Hindlimb 0 60 C Q) 50 :2 0 .E 40 30

20

10

0 Absence of Hypoplasia Parallel fusion Narrow lnterdigital Thinning of Retarded o,sification phalanges of phalanges of adjacent space without the metatarHI of phalanges phalanges bony fusion bones

Fig. 3 Influence of CoCl2 ·6H2 0 on MNNG-induced skeletal malformations of digits. The bone-cartilage specimens stained with alizarin red S and toluidine blue in both groups received admin­

istration of saline ( D ) and CoCl 2 ( ~) after treatment with 60 mg/kg of MNNG were examined for phalangial abnormalities. *: Significantly different from MNNG+saline group, P < 0.05 dactyly, which incidence exceeded that in the group treated on day 11 of gestation (Unpublished data). The increase of brachydactyly by VA or CoCl 2 treatment in the fetuses whose dams were treated with MNNG on day 11 suggests a possibility that VA and CoCl 2 might delay MNNG-induced DNA or cell damage.

The dose of VA (IP dose of 50 mg/kg) and CoCl 2 (IV dose of IO mg/kg) approximately corre­ sponds to one-tenth of their respective LD 50 values [Vanillin IP LD 50 = 475 mg/kg; CoCl 2 .6H2 0 IV LDso = 162 mg/kg (Llobet et al., 1985)]. It was shown in this study that administration of vanillin and CoCl 2 , separately, at the dose described above, on day 11 of gestation, did not cause any em­ bryo toxicity. It is known that MNNG injected intraperitoneally to pregnant mice enters rapidly through the placenta into embryos within I hour (Inouye, 1977). Previous studies demonstrated that treatment of cultured mammalian cells with MNNG resulted in rapid methylation of nucleic acid with N- 7 position being the most frequently alkylated site and 0-6 position an important alkylated site for mutagenicity (Lawly and Thatcher, 1970; Manson, 1981). It therefore appears that VA and CoCl 2 manifest their effects after MNNG-induced fetal DNA damage started. Inouye et al. (1987) reported that mitomycin C-induced micronuclei in mice was suppressed by 166 Y. Morita and M. Mizutani post-treatment with vanillin. In this study, it is of considerable interest that post-treatment with VA or CoCl 2 , which stimulates DNA recombination repair in vitro, caused the incidence of MNNG­ induced syndactyly and oligodactyly to decrease while that of MNNG-induced brachydactyly to increase. The mechanism of the modifying effects of both VA and CoC1 2 on MNNG-induced digital malformations should be investigated in further study.

ACKNOWLEDGMENTS

The authors wish to thank Mr. Koh Isobe for his technical assistance.

REFERENCES

Burdi, A.R. (1965) Toluidine blue-alizarin red S stain­ genicity in Japan. Intl. Agency Res. Cancer Sci., ing of cartilage and bone in whole-mount skeletons Publ. 27, (IARC), Lyon, 323-330. in vitro. Stain. Technol., 40: 45-48. Kiuchi, II., Inoue, T., Kada, T., Makino, 0., Shibata, Dawson, A.B. ( 1926) Note on staining of the skeleton T. and Ando, T. (1984) Effect of an antirnuta· of cleared specimens with alizarin red S. Stain. genie metal compound, cobaltous chloride, on E. Technol., J: 123-124. coli Rec A protein in vitro. Mutat. Res., 130: Hsu, T.C., Collie, C.J., Lusby, A.F. and Johnston, 367-368. D.A. (I 977) Cytogenetic assays of chemical clas­ Lawley, P.D. and Thatcher, C.J. (I 970) Methylation togens using mammalian cells in culture. Mutat. of deoxyribonucleic acid in cultured mammalian Res., 45: 233-247. cells by N-methyl-N'-nitro-N-nitrosoguanidine. Inouye, M. (I 977) Effect of N-methyl-N'-nitro-N­ Biochem. J., I 16: 693-707. nitrosoguanidinc on the fetal development in mice. Llobet, J.M., Domingo, J.L. and Corbclla, J. (1985) (The 2nd report). Ann. Rep. Res. Inst. Environ. Comparison of antidotal efficacy of chelating Med. Nagoya Univ., 28: 204-205. (Japanese) agents upon acute toxicity of cobalt in mice. Res. Inouye, M. and Murakami, U. ( 1978) Teratogenic ef­ Commun. Chern. Pathol. Pharmacol., 50: 305- fect of N-methyl-N'-nitro-N-nitrosoguanidine in 308. mice. Teratology, 18: 263-268. Mandell, J.D. and Greenberg, J. (1960) A new chemi­ Inouye, T., lmanishi, H., Sasaki, Yu F., Watanabe, M. cal mutagen for bacteria, 1-methyl-3-nitro-1-nitro­ and Ohta, T. (I 987) The 16th Annual Meeting of soguanidinc. Biochem. Biophys. Res. Commun .• 3: the Environmental Mutagen Society of Japan, Kyo­ 575-577. to (Japan), 9: 102. (Japanese) (Abst.) Manson, J.M. ( 1981) Developmental toxicity of alkyl­ Kada, T. and Kanematsu, N. (1978) Reduction of ating agents: Mechanism of action. The Biochemi­ N-methyl-N'-nitro-N-nitrosoguanidine-induced mu­ cal Basis of Chemical Teratogenesis, (l\lont, R.J. tations by cobalt chloride in Escherichia coli. ed.), New York, 95- I 36. Proceedings of the Japan Academy, 54: 234-237. Manson,J.M. and Marian, L.M.(1983) Contribution of Kada, T., Inoue, T., Ohta, T. and Shirasu, Y. (1986) mescnchymal cell death and mitotic alteration to Antimutagens and their modes of action. Anti­ asymmetric limb malformations induced by MNNG. mutagenesis and Anticarcinogenesis Mechanisms, Teratogenesis Carcinog. Mutagen., 3: 335-353. (Shankel, D.M., Hartman, P.E., Kada, T. and Neubert, D. (I 980) Teratogenicity: any relationship Hollaender, A. ed.), Plenum Press, New York, to carcinogenicity? Molecular and Cellular Aspects 181-196. of Carcinogen Screening Tests, (Montesano, R., Kawachi, T., Yahagi, T., Kada, T., Tazima, Y., lshidate, Bartsh, H. and Tomatis, L. ed.) !ARC Sci. Publ. 27, M., Sasaki, M. and Sugiyama, T. (1980) Coopera­ Lyon, 169-178. tive programme on short-term assays for carcino- Ohta, T., Watanabe, K. and Shirasu, Y. (1986) In- Effects of antimutagens on the teratogenicity of MNNG 167

hibitory effects of flavourings on mutagcnesis in­ Sugimura, T., Nagao, M. and Okada, Y. (1966) Car­ duced by chemicals in bacteria. rood Chem. cinogenic action of N-nwthyl-N'-nitro-N-nitrosogu­ Toxic., 24: 51-54. anidine. Nature (London), 210: 962--963. Platzek, T., Bochert, G., Schneider, W. and Neubert, Takehira, Y. and Kameyama, Y. ( 1981) Morphogene­ D. ( 1982) Embryotoxicity induced by alkylating sis of reductional malformations of digit in rat agents: I. Ethylmethanesulfonate as a teratogen in fetuses induced by cytosine arabinosidc or 5- mice. Arch. Toxicol., Sl: 1-25. tluorouracil. Cong. Anom., 21: I 05--118. Sasaki, Yu F., lmanishi, H., Ohta, T. and Shirasu, Y. Wilson, J.G. (1972) lntem::lations between l'arcino· ( 1987) Effects of vanillin on sister-chromatid ex­ gcnicity, mutagenicity and teratog.:uicity. Muta· changes and chromosome aberrations induced by genie Effects on Environmental Cont,11ninants, rnitornycin C in cultured Chinese hamster ovary (Sutton, H.E. and Harris, M.I. .:d.) Academic cells. Mutat. Res., 191: 193-200. Press, New York, 185--195.