Mutation Research, 192 (1987) 191-201 191 Elsevier

MTRL 061

Genotoxic effects of fluoride evaluated by sister-chromatid exchange

Yiming Li, Nyla A. Heerema a, Ann J. Dunipace and George K. Stookey

Oral Health Research Institute, Indiana University School of Dentistry, 415 Lansing Street, Indianapolis, IN 46202 (U.S.A.) and "Department of Medical Genetics, Indiana University School of Medicine, 702 Barnhill Drive, Indianapolis, IN 46223 (U.S.A.)

(Accepted 15 July 1987)

Keywords." Fluoride; Sister-chromatid exchange; Genetoxic potential.

Summary

The purpose of this investigation was to study the genotoxic potential of fluoride (in the form of sodium fluoride, NaF) using in vitro and in vivo sister-chromatid exchange (SCE) assays with Chinese hamster cells. The NaF concentrations used in cultures of Chinese hamster ovary (CHO) cells ranged from 0 to 6.3 mM, both with and without $9 activation. Fluoride analysis of the culture medium demonstrated that it contained little indigenous fluoride, and the concentration of added fluoride was not affected by the components of the medium or the $9 mix. The CHO cells cultured in 6.3 mM NaF almost vanished, and at the concentration of 5.3 mM NaF in cultures without $9 microsome, only MI cells were observed. In in vivo studies, Chinese hamsters were intubated with NaF dosages of 0, 0.1, 1.0, 10, 60 and 130 mg/kg, and the bone marrow (CHBM) cells were examined for SCE frequencies. Bone fluoride data showed that the intubated NaF was effectively absorbed. Death occurred in 3 of the 8 animals given 130 mg NaF/kg. The results indicated that NaF, in dosages up to 5.3 mM in CHO cell cultures and 130 mg/kg in in vivo CHBM cells, did not significant- ly increase the SCE frequencies over those observed in the negative (distilled water) controls. However, exam- ination of the cell cycle revealed an inhibitory effect of NaF on cell proliferation with doses of NaF at or greater than 1.0 mM in cultured CHO cells and at or greater than 60 mg NaF/kg in in vivo CHMB cells. The results of the present study indicated an inhibition of the cell cycle and death of the cells with increasing concentrations of fluoride but not effect of fluoride on SCE frequency in CHO and CHBM cells.

There are 3 different viewpoints regarding the genotoxic effects of fluoride in the scientific literature: (1) fluoride has no genotoxic effects (Kihlman, 1957; Kram et al., 1978; Martin et al., 1979, 1982; Thomson et al., 1985; Skare et al., 1986; Li et al., 1987a,b,c,d); (2) fluoride is a

Correspondence: Dr. Yiming Li, Oral Health Research In- stitute, Indiana University School of Dentistry, 415 Lansing Street, Indianapolis, IN 46202 (U.S.A.).

mutagenic agent and has cytogenetic effects which cause DNA and chromosome damage (Moham- med, 1968; Gerdes, 1971; Mohammed and Chandler, 1977; He et al., 1983; Tsutsui et al., 1984a,b,c); or (3) fluoride has synergistic or an- tagonistic effects with certain known mutagens (Mukherjee and Sobels, 1968; Vogel, 1973; Obe and Slacik-Erben, 1973; Slacik-Erben and Obe, 1976; Luchnik et al., 1985). These contradictory conclusions concerning the potential genotoxic ef-

0165-7992/87/$ 03.50 (© 1987 Elsevier Science Publishers B.V. (Biomedical Division)

192

fects of fluoride demonstrate the need for objec- tive reappraisal of this issue (Smith, 1985).

Because of its well-documented cariostatic ef- ficacy, fluoride has been increasingly used as an ef- fective public health measure to combat dental caries. Adjustment of the fluoride content in drinking water has become a common practice worldwide. It was estimated that in 1984 more than 260 million people in the world were living in com- munities with artificially fluoridated water, and in addition, about as many people received drinking water with a natural fluoride content of 1 ppm or more (WHO, 1984). At present, more than half the American population is consuming naturally or ar- tificially fluoridated water, and it has been a dental health objective to increase this proportion (CDC, 1985). In addition, there is an increasing use of fluoride in other delivery systems for caries preven- tion, including dentifrices, mouthrinses, profes- sional topical applications, pediatric fluoride supplements, and various dental restorative materials. These dental care products may contain fluoride in concentrations as high as 12 300 ppm.

In view of the contradictory evidence regarding the genotoxic effects of fluoride and with recogni- tion of the practical significance of fluoride for dental caries prevention, we have been conducting a series of experiments to clarify this controversial issue. This article reports the results of in vivo and in vitro sister-chromatid exchange (SCE) assays in which the cytogenetic effects of fluoride (in the form of sodium fluoride) on the chromosomes of Chinese hamster cells were investigated.

Materials and methods

Chemicals and reagents Sodium fluoride (NaF, CAS No. 7681-49-4) was

obtained from Matheson Coleman and Bell (Nor- wood, OH). Ethyl methanesulphonate (EMS, CAS No. 62-50-0), cyclophosphamide monohydrate (CP, CAS No. 6055-19-2), 5-bromo-2'-deoxyuri- dine (BrdU, CAS No. 59-14-3), nicotinamide adenine dinucleotide phosphate (NADP, CAS No. 53-59-8), o-glucose 6-phosphate, monosodium salt (G6P, CAS No. 54010-71-8), trypan blue (CAS

No. 72-57-1), Hoechst 33258 (CAS No. 23491-44-3), and colchicine (CAS No. 64-86-8) were from Sigma Chemical Co. (St. Louis, MO). Giemsa (CAS No. 51811-82-6) was obtained from Harleco (Gibbstown, N J). Minimum essential medium (Eagle, MEM), fetal bovine serum (FBS), penicillin-streptomycin solution, L-glutamine, MEM non-essential amino acids, trypsin, and col- cemid were purchased from Gibco Laboratories (Grand Island, NY). Rat-liver $9, which was in- duced with Aroclor 1254, was from Litton Bione- tics (Charleston, SC). The purchased $9 was examined for activity and sterility in a pilot study before this investigation, and the data indicated that the overall quality of the commercial $9 was satisfactory.

In vitro SCE assays in Chinese hamster ovary (CHO) cells

The CHO cells were obtained from American Culture Collection (ATCC CCL 61, Rockville, MD). The toxic effect of NaF on the cells was determined by culturing the cells in 10 ml MEM medium, supplemented with FBS (20%), L- glutamine (0.1 ml 200 mM), non-essential amino acids (0.1 ml 10 × ), streptomycin (100 #g/ml) and penicillin (100 U/ml), with NaF concentrations of 0, 0.6, 1.2, 2.4, 6.0 and 12.0 mM for 24 h and ex- amining the cell density with a hemacytometer and trypan blue. For the cultures containing 6.0 mM NaF a 77% inhibition in cell growth as determined by cell counts was observed. Consequently, the SCE assays were conducted at NaF doses of 6.30, 5.30, 4.20, 2.10, 1.00, 0.50 and 0.05 raM. Distilled, deionized water (DDW) served as a negative (sol- vent) control, and known mutagens were included as positive controls. The experiments were per- formed in triplicate, both with and without $9 ac- tivation, for each chemical dosage.

The cells were cultured in 25-cm 2 culture flasks (Coming Glass Works, Corning, NY) containing 10 ml complete MEM medium, which was sup- plemented with FBS (20%, L-glutamine (0.1 ml 200 raM), non-essential amino acids (0.1 ml 10×), streptomycin (100 #g/ml) and penicillin (100 U/ml). The cells were allowed to grow for 20 h to

reach a population of approximately 6 x 105, which was determined with trypan blue exclusion and a hemacytometer. In assays without $9 activa- tion, the predetermined doses of the agents were added to the cultures followed by a 2-h incubation period at 37°C. For assays with $9 activation, the medium was aspirated and the cell layer was wash- ed with prewarmed phosphate-buffered saline (PBS). The $9 mix was prepared by adding 31.5 mg NADP, 15.2 mg G6P and 1 ml $9 microsome to 100 ml MEM medium just before use. For each culture, 5 ml $9 mix were delivered into the flask immediately followed by the chemical agent. After incubating at 37°C for 2 h, the $9 mix with the test agent was removed. The cell layer was washed twice with prewarmed PBS, and replenished with 10 ml fresh complete MEM medium. For assays, both with and without $9 activation, BrdU was then added to a final concentration of 10 /~M (3 #g/ml) , and the culture was gassed with CO2 at 4 1/min for 40 sec under a dim yellow light. The flasks were wrapped with aluminum foil, and in- cubation continued at 37°C for another 34 h. 2 h before the termination of culturing, colcemid was added to the culture to a final concentration of 2 x 10-7 M while keeping the flasks foil-wrapped.

The cells were harvested 36 h after the addition of BrdU by briefly treating the cell layer with 2 ml 0.250/o trypsin. The cell suspension was then cen-

trifuged at 1000 rpm for 5 min. The ceils were resuspended in 0.075 M KC1 hypotonic solution and incubated at 37°C for 20 rain. After recen-

trifuging, the ceils were fixed by dropwise addition of 5 ml fresh fixative of 3:1 methanol and acetic acid. 20 min later the fixing procedure was repeated 2 more times. The fixed cells were then dropped on a clean slide and air-dried overnight. For assays, both with and without $9 microsome, the medium containing the test agents was saved for fluoride analysis.

The fluorescence plus Giemsa method was used for staining the chromosomes, and the procedure essentially followed that described by Goto et al. (1978). All slides were randomized with code numbers so that the scoring was performed without knowledge of treatment. The

193

chromosomes were examined under a light

microscope with a green filter. The influence of the test chemicals on the cell cycle was evaluated by recording the percentage of cells in the first (M0,

second (M2) and third (M3) cell division for 100 cells f rom each culture. The Mz chromosomes were further examined and only the cells with 22 well- spread chromosomes were scored for SCE. A total of 25 cells f rom each culture and 75 cells for each dosage were examined for SCE frequency. A two- tailed t-test was used for statistical analysis of SCE data.

In vivo SCE assay in Chinese hamster bone mar-

row (CHBM) cells

Male Chinese hamsters f rom Cambridge Diagnostic, Toxikon (Norwood, MA) were obtain- ed at 3 months of age and kept in quarantine until tested at approximately 4 months. Upon arrival, animals were identified by ear punch and were housed individually in suspended wire-grid bot- tomed cages with daily care in a pathogen-free en- vironment. Throughout the experiment, feeding was ad libitum with distilled water and a low fluoride diet [high protein casein, 6.5%; skin milk, 21.6%; whole wheat flour, 56.9%; soybean oil, 10.0°70; Jones-Foster mineral mix (ICN Nutritional Biochemicals, Cleveland OH), 2.0%; NaC1, 1%; and Teklad vitamin mix (Teklad, Madison, WI), 2.0%; F < 0.2 ppm].

At the initiation of the experiment, the hamsters were lightly anesthetized by ether inhalation, and a

BrdU tablet was implanted subcutaneously in the lower lateral abdominal region. The BrdU tablets were prepared with a pellet press and a 0.178 in. (4.5 mm) diameter punch and die set (Parr Instru- ment Co., Moline, IL) and were coated with agar following the method described by Allen et al.

(1978) and King et al. (1982). Tablets were selected which weighed to 50-60 mg.

The test agents were administered 6 h after im- plantat ion of the BrdU tablets. The animals were randomly assigned to seven groups. Group I con- tained 8 hamsters and the other groups each had 3 animals. Group I received NaF by intubation at the dose of 130 mg/kg , which was the maximum

194

tolerable dosage (MTD) for the hamsters and was

expected to result in the death of some animals. Groups II through V were intubated with NaF at dosages of 60, 10, 1 and 0.1 mg per kg body weight, respectively. Group VI served as a negative control and was intubated with distilled water. The positive control, Group VII, received CP at a dosage of 25 mg per kg body weight by in- traperitoneal injection. Sodium fluoride solutions were prepared with distilled water, and cyclophosphamide was dissolved in physiologic saline. The volume for intubation was 15 ml/kg, and for i.p. it was 10 ml/kg. The animals received colchicine (0.6 mg/kg , i.p.) 2 h before sacrifice.

All hamsters were sacrificed by cervical disloca- tion under ether anesthesia 24 h after implantation of BrdU tablets. The femurs were removed and freed of muscle. Both the distal and proximal ends of the femurs were cut with sharp scissors until the marrow canal opened. The needle (25 g) of a sy- ringe containing 3 ml 0.075 M KC1 hypotonic solu- tion (prewarmed to 37°C) was inserted into the iliac end of the shaft and the bone marrow was flushed into a 25-cm 2 culture flask. The bone mar-

row cells f rom both femurs of an animal were pooled. The cells were suspended in the KCI solu- tion by gently shaking the flask and were then in- cubated at 37°C for 30 min. The procedures for fixing the cells, preparing, staining, randomizing and scoring of the slides and statistical analysis of the data were the same as those for in vitro SCE assays with CHO cells. The humera were saved for

analysis of fluoride content by the method of Taves (1968) in order to monitor the absorption of fluoride following the administration of NaF by intubation. The results o f fluoride content in the bones were statistically analysed by one way analysis of variance (ANOVA) and Newman- Keul's multiple comparison test.

R e s u l t s

Because the cells in media containing 6.3 mM NaF had almost vanished, these cultures were not examined for cell cycle and SCE frequency. For the cultures with 5.3 mM NaF without $9 activa- tion, no M2 or Ms cells were found and thus their SCE were not scorable. In the in vivo study, death occurred in 3 of 8 animals which received the MTD of NaF (130 mg/kg) .

The influence of NaF on the cell cycle is presented in Tables 1, 2 and 3. The data show that for both in vitro and in vivo assays, the percentage of M~ cells increased and that of M3 cells decreased as the NaF concentration increased. As illustrated

in Figures 1, 2 and 3, this effect was most marked in cultured CHO cells without microsome activa- tion and was least in in vivo CHBM cells. The number of M2 cells also decreased with the increase of NaF concentration in the medium. However, it seemed that the influence of NaF on the percentage of M2 cells in the in vivo assay was limited. Inhibi- tion of cell division by NaF was not observed until the dosages reached 1.0 mM NaF in cultured CHO

TABLE 1 CELL CYCLE OF CHO CELLS AS INFLUENCED BY FLUORIDE IN IN VITRO SCE ASSAYS W I T H O U T $9 ACTIVATION

Cell cycle/100 cells

Ml M: Ms

Group Dosage (raM NaF)

I 6.30 11 5.30 100.0 _+ 0.0 0.0 + 0.0 0.0 _+ o.oa[

!

_ _ _ O.OJ Ill 4.20 87.0 + 5.0 13.0 + 5.0 0.0 + _ _ 4 . 0 | 2 . 3 + 1.5 IV 2.10 73.0 + 2.6 24.7 +

33.0 + 5.6.] 7.3 _+ 2.1 V 1.00 59,7 _+ 4.0 -] VI 0.50 49,0 _+ 6.1~ 44.0 _+ 4 .4] 7.0 +_ 1.7

4 .9J ~ 7.7 + 2.5 VII 0.05 42,3 + 50.0 + 7.0 _ VIII 0.00 47,0 + 7.5 42.7 + 4.5 10.3 + 3.1

aMean _+ SD. The values within brackets are not significantly different (P>0.05) .

T A B L E 2

C E L L C Y C L E OF C H O

195

C E L L S AS I N F L U E N C E D BY F L U O R I D E IN IN V I T R O SCE A S S A Y S W I T H $9 A C T I V A T I O N

Group Dosage Cell cyc le /100 cells (mM NaF)

Ml M2 M3

1 6.30 - - -

11 5.30 51.7 ± 5.9 45.3 ± 4.0 3.0 ± 2.0 ~

_ 6.7 67.3 + 7.2 7.3 + 1.1

IV 2.10 17.3 + _ 5.9 72.0 + 5.3 10.7 + 0.6

V 1.00 17.3 + _ 4.0 l l . 0 + 2.0 4.5 71.7 ÷

Vl 0 .50 15.0 + _ 1.7 73.3 + 4.2 11.7 + 2.5

VII 0.05 13.0 + 2.6 | 73.7 + 2.3 13.3 + 2.9

VIII 0.00 14.7 + 1.1 73.3 + 2.3 12.0 + 2.0

aMean _+ SD. The values within brackets are not significantly different ( P > 0 . 0 5 ) .

T A B L E 3

C E L L C Y C L E O F C H I N E S E H A M S T E R B O N E M A R R O W C E L L S AS I N F L U E N C E D BY NaF IN IN VIVO SCE A S S A Y S

Group NaF dosage Cell cyc le /100 cells

(mg/kg) Ml Mz M3

1 130.0 a 39.3 ± 2.5 48.0 + 5.2 7

11 60.0 26.3 ± 5.7 42.0 + 5.2

_ 2.1 / 43.3 + 1.5 I l l 10.0 17.3 +

IV 1.0 15.0 + 3 . 6 J 41.3 _+ 3.5 V 0.1 16.0 + 2.6 37.3 _ + 2.5

VI - 14.3 ± 1.5 36.3 _+ 2.1

12.7 ± 5.5 b

39.3 _+ 3.1

43.7 ± 5.9

46.7 ± 5.0

49.3 ± 1.2

aDeath occur red in 3 out o f 8 an imals .

bMean ± SD. The values within brackets are not significantly different ( P > 0 . 0 5 ) .

cells and 60 mg NaF/kg in in vivo CHBM cells (Tables 1, 2 and 3). There is no data in this study which makes it possible to relate the in vivo 60 mg NaF/kg dosage to the resulting fluoride concentra- tion at the in vivo cellular level; however, it should

be noted that the in vitro fluoride concentrations equal to or greater than 1.0 mM NaF which in- hibited cell proliferation in CHO cell cultures far exceeded those which occur in in vivo mammalian body fluids bathing cells even following dosage

100 I ° ° T

! [].i e.2 [].3 • I 8O

8O

6o ~ 60

40 w

2 2O

i o 0 1 10 20 40 80 100

FLUORIDE (pom)

Fig. 1. Inf luence of NaF on the cell cycle of C H O cells in in

vitro SCE assays wi thou t $9 activation.

[ ] MI • M2 [] M3

0 1 10 20 40 80 100

FLUORIDE (ppm)

Fig. 2. Inf luence of NaF on the cell cycle of C H O cells in in

vitro SCE assays wi th $9 activation.

196

100

v 60

[ ] Ml [ ] M2 [ ] M3

0 0 O.l i , ( ] I0 60 130

DOSAGE (NoF mg/Kg)

Fig. 3. Influence of NaF on the cell cycle of Chinese hamster

bone marrow cells in in vivo SCE assays.

with 60 mg NaF/kg or greater. The frequencies of SCE for both in vitro and in

vivo assays in Chinese hamster cells are summariz-

ed in Tables 4, 5 and 6. The values are the average number of SCE observed in 75 M2 cells, each with 22 chromosomes. Statistical analysis indicated that

the positive controls in all assays had a significant- ly higher incidence of SCE than both the negative controls and NaF groups (p < 0.05). The SCE counts for the NaF-treated cultures, with and without $9 activation, and for the animals which received doses of NaF up to the MTD, were not statistically different from the negative controls. In the 3 assays, the average frequency of SCE was

TABLE 4

FREQUENCIES OF SISTER CHROMATID EXCHANGES IN CULTURED CHO CELLS WITHOUT $9 ACTIVATION

Group Treatment system SCE/cell

Agent a Dosage (mM) Range Mean _+ SD

1 NaF 6.30 -

11 NaF 5.30

111 NaF 4.20 0- 18

IV NaF 2.10 2- 16

V NaF 1.00 2- 19

VI NaF 0.50 1- 18

VII NaF 0.05 1- 19

VIII DDW 0- 19

IX EMS 2.80 68-143

p

b 8.93 +_ 3.96

8.75 _+ 3.40

9.08 -4-- 3.80

8.52 _+ 3.68

8.63 _+ 3.65

8.56 _+ 4.04

97.03 _+ 16.20

aDDW, distilled deionized water; EMS, ethyl methanesulphonate.

bValues within brackets are not statistically different (P>0.05) .

TABLE 5

FREQUENCIES OF SISTER CHROMATID EXCHANGES 1N CULTURED CHO CELLS WITH $9 ACTIVATION

Group Treatment system SCE/cell

Agent a Dosage (raM) Range Mean _+ SD

1 NaF 6.30 -

11 NaF 5.30 5-21

I 11 NaF 4.20 6-20

IV NaF 2.10 5-18

V NaF 1.00 5-18

VI NaF 0.50 6-18

VII NaF 0.05 5-18

VIII DDW 6-21

IX CP 0.05 34-73

aDDW, distilled deionized water; CP, cyclophosphamide monohydrate.

bValues within brackets are not statistically different (P>0.05) .

11.77 +_ 3.38" ~

11.56 _+ 3.50

11.41 _+ 3.121

11.28 _+ 3.06

11.55 + 3.01

11.47 + 3.15

11.63 + 3.28

51.75 _+ 9.67

TABLE 6

FREQUENCIES OF SISTER CHROMATID EXCHANGES IN CHINESE HAMSTER MARROW CELLS

197

Group Agent a Dosage

(mg/kg)

SCE/cell

Range Mean ± SD

1 NaF 130.0 b

11 NaF 60.0

I11 NaF 10.0 IV NaF 1.0

V NaF 0.1

VI DDW -

VII CP 25.0

0-11 5.04 ± 2.30] ~

0-12 5.08 ± 2.52 I 0-15 4.80 + 2.84[

0-12 4.76 + 2.44 I

0-13 4.80 + 2.58[

0-11 4.76 + 2.57..1

29-91 58.47 ± 12.88

aDDW distilled deionized water; CP, cyclophosphamide monohydrate. bDeath occurred in 3 of 8 animals. CValues within brackets are not statistically different (P>0.05).

lowest in in vivo CHBM cells, higher in the in vitro CHO cells without $9, and highest in cultured CHO cells with $9 activation.

Analysis of fluoride content in the media from in vitro assays, both with and without $9 microsome, indicated that the media and prepared $9 mix contained little indigenous fluoride. In ad- dition, the data showed that the concentration of fluoride was accurate, and the incorporated fluoride was not significantly affected by the com- ponents of the medium and $9 mix (Table 7).

The humera fluoride data obtained from the in vivo study are summarized in Table 8. The bone fluoride content increased with the increase in NaF

dosage. One-way ANOVA and Newman-Keul's multiple comparison tests showed that the bone fluoride was significantly different in the animals receiving 130 and 60 mg NaF/kg (P < 0.01) and was higher than the groups which received 10 mg NaF/kg or less (P < 0.01). The bone fluoride con- tent of the animals treated with 10 mg NaF/kg or less, including negative and positive controls, was lower and not significantly different.

Discussion

The occurrence of sister-chromatid exchange is the result of reciprocal interchange of DNA be-

TABLE 7

FLUORIDE CONTENT OF CULTURE MEDIA

Group Agent a Dosage

(F ppm)

Fluoride in media (ppm)

Without $9 With $9

1 NaF 120

11 NaF 100

111 NaF 80 IV NaF 40

V NaF 20

VI NaF 10

VII NaF 1 VIII DDW 0

IX PC 0

m

107.4 + 5.4 b 100.8 ± 6.5

82.6 ± 3.6 80.3 ± 4.2

41.4 ± 2.3 40.1 ± 4.2

18.9 ± 1.2 20.3 + 2.3

9.2 ± 0.4 9.6 ± 0.3 0.9 ± 0.2 1.0 + 0.1

0.02 _+ 0.0 0.02 ± 0.01

0.04 ± 0.01 0.03 ± 0.01

aDDW, distilled deionized water; PC, positive controls, ethyl methanesulphonate monohydrate for cultures with $9.

bMean ± SD.

for cultures without $9, and cyclophosphamide

198

TABLE 8

FLUORIDE CONTENT OF CHINESE HAMSTER HUMORA

Group Agent a Dosage

(mg/kg)

1 NaF

11 NaF

Ill NaF IV NaF

V NaF

VI DDW

VII CP

130.0 1325 _+ 161

60.0 986 _+ 86

10.0 748 +_ 108~ ~ !

1.0 646 + 83|

/ 0.1 649 + 108

578 + 139|

25.0 605 + lOlJ

~DDW, distilled deionized water; CP, cyclophosphamide monohydrate.

h#g F/g bone ash.

~Mean + SD. The values within brackets are not significantly different (P>0.05).

Humora fluoride (ppm) t'

tween sister-chromatid segments at homologous loci (Latt et al., 1981). Although its molecular mechanism and biological significance still remain to be defined, SCE analysis has been widely used to assess the mutagenic potential of chemicals (Wolff, 1977; Abe and Motomichi, 1982; Perry and Thomson, 1984). Numerous studies have reported a direct relationship between the mutagenicity of various substances and their effect on increasing the incidence of baseline SCE, although the relative efficiency of SCE and muta- tion induction varied with the compounds tested (Latt, 1974; Solomon and Bobrow, 1975; Perry and Evans, 1975; Carrano et al., 1978; Latt et al., 1981). Implicit in these studies is the assumption that SCE formation is in some way related to DNA damage and repair. Perry and Evans (1975) have shown that the SCE assay can efficiently detect significant effects of mutagens on chromosome structure at doses that are usually much lower than those necessary to produce marked increases in chromosome aberrations. For most chemicals tested, a dose of mutagen which caused a 10-fold increase in baseline SCE had little or no effect on chromosome aberration frequency. Therefore, it has been concluded that SCE frequency is a far more sensitive index of chromosome damage than gross chromosomal aberrations, and the SCE test should form part of any mutagen testing program as a complement to other assay systems (Perry and Evans, 1975; L a t t e t al., 1981; Perry and Thom- son, 1984).

A number of investigators have utilized the SCE test to study the genotoxicity of fluoride (Kram et al., 1978; He et al., 1983; Tsutsui et al., 1984a; Kishi and Tonomura, 1984; Thomson et al., 1985). The results from in vivo mouse bone marrow cells and cultured human lymphocytes indicated that no increase in SCE frequencies was observed in mice maintained for 7 generations on a diet containing 1.2 mM NaF (Kram et al., 1978) or in human lym- phocytes treated with up to 4 mM NaF (Kishi and Tonomura, 1984) and 3 mM NaF (Thomson et al., 1985). However, He et al. (1983) and Tsutsui et al. (1984a) reported that NaF at a concentration of 3.0 mM significantly increased the incidence of SCE in cultured Red Muntjac cells and in cultured Syrian hamster embryo cells at dosage levels of 0.5, 1.0 and 1.9 mM, respectively. Close examination of these studies showed that Kram et al. (1978) and Thomson et al. (1985) used the random coding system for scoring but this was not indicated in the reports by He et al. (1983) and Tsutsui et al. (1984a). Scoring without knowledge of treatment is considered essential for assays which require somewhat subjective judgement, including SCE analysis ( L a t t e t al., 1981; Perry and Thomson, 1984). In addition, He et al. (1983) found that the SCE frequency in the cultures treated with 0.3 mM NaF was not significantly different from that of the negative control; the cells exposed to 3.0 mM NaF had significantly higher SCE compared with the negative control. The mean SCE values of these groups were rather close (10.4 for the

negative control group, 11.1 for 0.3 mM NaF, and 12.8 for 3.0 mM NaF). However, the standard deviations were unusually small when compared to those reported by other investigators (Kram et al., 1978; Thomson et al., 1985).

The results of the present investigation support the point of view that fluoride does not cause chromosomal damage as indicated by SCE fre- quency. However, the data clearly show that fluoride is a toxic substance which inhibits cell pro- liferation and, at high doses, kills cells and even- tually results in animal death. The toxic action of fluoride was particularly evident in cultured C H O cells without $9 activation which were exposed to fluoride for 36 h. When compared with the cultures treated with NaF at, or lower than, 0.5 mM, the percentage of M1 cells was significantly higher in those with NaF concentrations at, or higher than, 1.0 mM (Table 1). For the cell cultures with $9 microsome, the increase of M1 cells was not significant until the NaF level reached approx- imately 4.2 mM. The inhibitory action of fluoride on cell proliferation in systems with and without

$9 cannot be compared because the cells were ex- posed to fluoride for different time periods: 36 h

for cultures without $9 and 2 h for those with $9. A major difference in procedure for these cultures was that the cultures with $9 were replenished with medium chemical treatment, which was performed 24 h after the initiation of the cell culture. Whether or not the exposure of cells to $9 mix affects the susceptibility of C H O cells to NaF toxicity remains unclear. However, it should be noted that even in the cultures without fluoride, the percentage of M~ cells was much higher in cultures without $9 than in those with $9. Further studies are necessary to confirm and clarify the findings in the present study.

For in vivo studies, oral administration was chosen as the route for providing NaF in the pres- ent investigation because it simulates the physiological exposure to fluoride in human be- ings. The analysis of bone fluoride indicated the effective absorption of fluoride following stomach intubation and, therefore, the route of oral ad- ministration of NaF for hamster bone marrow

199

SCE tests is justified.

The cytogenetic activity of a chemical in humans may involve complicated biological actions. It would not be appropriate to establish definite con- clusions regarding the genotoxic effects of fluoride based on one or two assays, which is the situation in most previous studies. A series of investigations consisting of various cytogenetic and mutagenetic assays in different systems is necessary to provide sound scientific evidence for clarifying this con- troversy. Our laboratory has been conducting ad- ditional studies using Ames Salmonella assays, the mouse sperm morphology test and the mouse bone marrow micronucleus test in order to obtain reliable conclusions regarding this important issue (Li et al., 1987a,b,c,d). Results obtained in the pre- sent studies indicate that fluoride in increasing concentrations inhibits cell cycle progression and at high dosages, results in cell and animal death but does not increase SCE frequency.

Acknowledgements

The authors thank Dr. Wu Zhang, Department of Biochemistry, and Donna Stump and Rosalie Roach, Department of Medical Genetics, Indiana University School of Medicine, for their skillful technical assistance and also gratefully acknowledge the cooperation and assistance pro- vided by the Animal Research Facility at Indiana University School of Dentistry. Critical review of the manuscript by Dr. Catherine G. Palmer, Department of Medical Genetics, Indiana Univer- sity School of Medicine, is also greatly ap- preciated.

References

Abe, S., and S. Motomichi (1982) SCE as an index of mutagenesis and/or carcinogenesis, in: A. Sandberg (Ed.), Sister Chromatid Exchange, Liss, New York, pp. 461-514.

Allen, J.W., C.F. Shuler and S.A. Latt (1978) Bromodeoxy- uridine tablet methodology for in vivo studies of DNA syn- thesis, Somat. Cell Genet., 4, 393-405.

Carrano, A.B., L.H. Thompson, P.A. Lindl and J.L. Minkler (1978) Sister chromatid exchanges as an indicator of mutagenesis, Nature (London), 9, 551-553.

200

Centers for Disease Control (1985) Dental caries and communi- tY water fluoridation trends - United States, Morbidity and Mortality Weekly Report, 34, 77-80.

Gerdes, R.A. (1971) The influence of autospheric hydrogen fluoride on the frequency of sex-linked lethals and sterility in Drosophila rnelanogaster, Fluoride, 4, 25-29.

Goto, K., S. Maeda, Y. Kano and T. Sugiyama (1978) Factors involved in differential Giemsa-staining of sister chromatids, Chromosoma, 66, 351-359.

He, W., A. Liu, H. Bao, Y. Wang and W. Cao (1983) Effect of sodium fluoride and fluoroacetamide on sister chromatid exchanges and chromosomal aberrations in cultured Red Muntjac (Muntjacus muntjak) cells, Acta Sci. Circumst., 3, 94-100.

Kihlman, B.A. (1957) Experimentally induced chromosome aberrations in plants, 1. The production of cbromsome aber- rations by cyanide and other heavy metal complexing agents, J. Biophys. Biochem. Cytol., 3, 363-380.

King, M.T., D. Wild, E. Grocke and K. Eckhardt (1982) 5-Bromodeoxyuridine tablets with improved depot effect for analysis in vivo of sister chromatid exchanges in bone mar- row and spermatogonial cells, Mutation Res., 97, 117-129.

Kishi, K., and A. Tonomura (1984) Cytogenetic effects of sodium fluoride, Mutation Res., 130, 367.

Kram, D., E.L. Schneider, L. Singer and G.R. Martin (1978) The effects of high and low fluoride diets on the frequencies of sister chromatid exchanges, Mutation Res., 57, 51-55.

Latt, S.A. (1974)Sister chromatid exchanges, indices of human chromosome damage and repair: detection by fluorescence and induction by mitomycin C, Proc. Natl. Acad. Sci. (U.S.A.), 71, 3162-3166.

Latt, S.A., J. Allen, S. Bloom, A. Carrano, E. Falke, D. Kram, E. Schneider, R. Schreck, R. Tice, B. Whitfield and S. Wolff (1981) Sister chromatid exchanges: a report of the Gene-Tox Program, Mutation Res., 87, 17-62.

Li, Y., A. Dunipace and G.K. Stookey (1987a) Absence of mutagenic and antimutagenic activities of fluoride in Ames Salmonella assays, Mutation Res., 190, 229-236.

Li, Y., A. Dunipace and G.K. Stookey (1987b) Effects of fluoride on the mouse sperm morphology test, J. Dent. Res., in press.

Li, Y., A. Dunipace and G.K. Stookey (1987c) Mutagenic and antimutagenic activities of fluoride in Ames Salmonella assays, J. Dent. Res., 66, 342 (abstract No. 1882).

Li, Y., A. Dunipace and G.K. Stookey (1987d) Cytogenetic ef- fects of fluoride in mouse sperm morphology test, J. Dent. Res., 66, 342 (abstract No. 1883).

Luchnik, N.V., N.A. Poryadkova and N.M. lzmailova (1985) The influence of inhibitors of cellular respiration on the pro- duction of structural mutations in human lymphocytes ir- radiated during different stages of the mitotic cycle, Genetika, 21, 252-261.

Martin, G.R., K.S. Brown, D.W. Matheson, H. Lebowitz, L. Singer and R. Ophaug (1979) Lack of cytogenetic effects in mice or mutations in Salmonella receiving sodium fluoride, Mutation Res., 66, 159-167.

Martin, G.R., K.S. Brown, L. Singer, R. Ophaug and D. Jacobson-Kram (1982) Cytogenetic and mutagenic assays on fluoride, in: J.L. Shupe, H.B. Peterson and N.C. Leone (Eds.), Fluorides - Effects on Vegetation, Animals and Humans, Proceedings of an International Symposium on Fluorides at Utah State University, Logan, UT, USA, May, 1982, pp. 271-279.

Mohammed, A.H. (1968) Cytogenetic effects of hydrogen fluoride treatment in tomato plants, J. Air Pollution Control Assoc., 18, 395-398.

Mohammed, A.H., and M.E. Chandler (1977) Cytological ef- fects of sodium fluoride on mice, presented during hearings, the National Cancer Program, Part 2. Fluoridation of Public Drinking Water, Sept. 21 and Oct. 12., pp. 42-60.

Mukherjee, R.N., and F.A. Sobels (1968) The effects of sodium fluoride and iodoacetamide on mutation induction by X- irradiation in mature spermatozoa of Drosophila, Mutation Res., 6, 217-225.

Obe, G., and R. Slacik-Erben (1973) Suppressive activity by fluoride on the induction of chromosome aberrations in human cells with alkylating agents in vitro, Mutation Res., 19, 369-371.

Perry, P., and H.J. Evans (1975) Cytological detection of mutagen-carcinogen exposure by sister chromatid exchange, Nature (London), 258, 121-125.

Perry, P., and E.J. Thomson (1984) The methodology of sister chromatid exchanges, in: B.J. Kilbey, M. Legator, W. Nichols and C. Ramel (Eds.), Handbook of Mutagenicity Test Procedures, 2rid edn., Elsevier, Amsterdam, pp. 495-529.

Skare, J.A., K.R. Schrotel and G.A. Nixon (1986) Lack of DNA-strand breaks in rat testicular cells after in vivo treat- ment with sodium fluoride, Mutation Res., 170, 85-92.

Slacik-Erben, R., and G. Obe (1976) The effect of sodium fluoride on DNA synthesis, mitotic indices and chromosomal aberrations in human leukocytes treated with Trenimon in vitro, Mutation Res., 37, 253-266.

Smith, G.E. (1985) A surfeit of fluoride? Sci. Prog. Oxf., 69, 429-442.

Solomon, E., and M. Bobrow (1975) Sister chromatid ex- changes - a sensitive assay of agents damaging human chromosomes, Mutation Res., 30, 273-278.

Taves, D.R. (1968) Separation of fluoride by rapid diffusion us- ing hexamethyldisiloxane, Talanta, 15, 969-974.

Thomson, E.J., F.M. Kilanowski and P.E. Perry (1985) The ef- fect of fluoride on chromosome aberration and sister chromatid exchange frequencies in cultured human lym- phocytes, Mutation Res., 144, 89-92.

Tsutsui, T., N. Susuki and M. Ohmori (1984a) Sodium fluoride-induced morphological and neoplastic transforma- tion, chromosome aberrations, sister chromatid exchanges, and unscheduled DNA synthesis in cultured Syrian hamster embryo cells, Cancer Res., 44, 938-941.

Tsutsui, T., N. Suzuki, M. Ohmori and H. Maizumi (1984b) Cytotoxicity, chromosome aberrations and unscheduled

DNA synthesis in cultured human diploid fibroblasts induced by sodium fluoride, Mutation Res., 139, 193-198.

Tsutsui, T., K. lde and H. Maizumi (1984c) Induction of unscheduled DNA synthesis in cultured human oral keratinocytes by sodium fluoride, Mutation Res., 140, 43-48.

Vogel, E. (1973) Strong antimutagenic effects of fluoride on mutation induction by Trenimon and l-phenyl-3,3-dimethyl-

201

triazene in Drosophila melanogaster, Mutation Res., 20, 339-352.

Wolff, S. (1977) Sister chromatid exchange, Annu. Rev. Genet., 11, 183-201.

World Health Organization (1984) Fluorine and Fluorides, En- vironmental Health Criteria 36, WHO, Geneva, pp. 1-136.

Communicated by R.J. Preston