Toxicology Letters, 50 (1990) 179-l 88
179
Elsevier
TOXLET 02271
Lack of covalent binding to DNA of di-n-octyltin dichloride (DOTC) in vivo and in vitro
Peter Sagelsdorff, Peter Dollenmeier, Daniel Ebner, Franqoise Bieri, Samar M. Kelly, Willy Stkbli, Felix Waechter and Philip Bentley Central Toxicology Unit, CIBA-GEIGY
Ltd., Base1 (Switzerland)
(Received 29 May 1989) (Revision received 18 August 1989) (Accepted 21 August 1989) Key uords: Di-n-octyltindichloride; hamster cells
Organotin compounds; DNA binding; Liver; Thymus; V79 Chinese
SUMMARY [14ClDi-~-~tyltin dichloride ([t4C]DOTC) was administe~d by oral gavage to male and female rats. After 96 h hepatic and thymic DNA was isolated. All DNA fractions were radioactive, but analysis of DNA hydrolysates by HPLC revealed that the radioactivity was incorporated via biosynthesis and was not due to adduct formation. The limit of detection for adduct formation, expressed in units of the covalent binding index (CBI = pmol chemical bound per mol nucleotides/mmol chemical applied per kg body wt.) was approximately 0.2 for liver DNA and about 0.7 for thymus DNA. This maximum possible DNAbinding ability is about 30000 times lower than the corresponding value for the strong carcinogen, aflatoxin Br. In addition, [K?]DOTC did not bind covalently to calf thymus DNA in the presence or absence of rat liver S9 or to DNA of V79 Chinese hamster cells. This study therefore gives no indication for genotoxic activity of DOTC mediated by DNA binding.
INTRODUCTION
Organotin compounds (e.g. dioctyltin-bis-thioglycollate) are industrially used for stabihsation of polyvinylchIoride plastics. When used in plastics for food packaging, the compounds may migrate in traces into the food and consequently some human exposure may occur. In this case the compounds wouid decompose to dioctyltin dichloride (DOTC) in the acid environment of the stomach. Consequently most toxicological studies have been performed with DOTC. Feeding studies with DOTC at dieAddress for correspondence:
Peter Sagelsdorff, Central Toxicoloy Unit, CIBA-GEIGY
Basel, Switzerland. 0378-4274/90/$3.50 @I 1990 Elsevier Science Publishers B.V. (Biomedical Division)
Ltd., CH-4002
180
tary levels of 50 and 150 ppm demonstrated a dose-related reduction in thymus weight in weanling rats but not in other animal species [l]. In vitro DOTC has been shown to interact with purified DNA and to induce mutations in V79 Chinese hamster fibroblasts in the absence of an activation system [2]. In contrast, DOTC was negative in several other short-term tests for mutagenicity and genotoxic activity [3]. Organotin compounds are extensively used, thus it was considered important to obtain more information about a possible genotoxic activity of DOTC. One main characteristic of a large number of organic chemical carcinogens is their ability to undergo covalent interactions with DNA [4]. Consequently the ability of the test compound to interact covalently with DNA in vitro, in V79 cells and with liver DNA of rats after oral administration of [i4C]DOTC was investigated; in addition, a possible interaction with DNA in the thymus, which is the target organ for the toxicity of DOTC, was examined. MATERIALS AND METHODS
Test compounds 1-[‘4C]DOTC was obtained from NATEC, Hamburg, F.R.G., with a specific radioactivity of 122.8 mCi/mmol (0.3 mCi/mg). The radiochemical purity was 97.4% as determined by TLC on Silicagel 60 plates (Merck, Darmstadt, F.R.G.) with n-hexane/acetic acid (12: 1) as eluant (&=0.26). Unlabelled DOTC was obtained from CIBA-GEIGY Ltd. Animals and treatment Male and female Sprague-Dawley derived rats (Tif:RAI(f) (SPF)) were obtained from the CIBA-GEIGY breeding facility. They were kept 4 per macrolone cage at 22 f 1°C 55 + 5% humidity and a 12 h light/dark cycle. The test compound was suspended in ethanol/Tween 80/H20 (5:20:75) to yield a concentration of 1.2 mg/ml (0.36 mCi/ml). The rats received about 1 ml of the mixture by oral gavage. The exact amount of solution injected was calculated from the weight difference of the syringe before and after the administration. After administration of the radiolabelled test compound the rats were placed in all-glass metabolic cages and exhaled [‘4C]C02 was collected in ethanolamine/Z methoxyethanol(l:2 v/v) to check for the metabolic stability of the radiolabel administered. At different time-points blood was taken from the tail for estimation of radioactive DOTC. The half-life of DOTC was calculated assuming first-order kinetics for elimination from the blood. Isolation of DNA and chromatin protein After 96 h the animals were killed by open heart puncture under ether anaesthesia. The livers and thymus were excised and washed in ice-cold saline. The organs were
181
homogenized and DNA and chromatin protein was isolated and purified as described previously [5]. The highly purified DNA (containing less than 0.2% protein, as shown by incorporation of radiolabelled amino acids [6] was dissolved in 20 mM sodium succinate buffer, 8 mM CaC12 (pH 6.0) and counted for radioactivity. Repetitive purtjication of DNA The remaining DNA solution was mixed with 25 ml lysing medium and repurified [71* An&s& of the nu~leotide~ DNA in 8 mM CaC12, 20 mM sodium succinate (pH 6.0) was digested enzymatitally with micrococcal nuclease (Sigma, St. Louis MO; No. N3755; E.C. 3.1.31.1.) and spleen exonuclease (Boehringer-Mannheim, Rotkreuz, CH; No. 108 251; E.C. 3.1.16.1.) as described previously [7]. The resulting nucleotide mixture was separated by HPLC on a Nucleosil 10 pm C I8 column (250 x 8 mm) (71.The background radioactivity was determined from the average of 6 nucleotide analyses of control DNA hydrolysates. The respective standard deviation was in the worst case (fraction 16) 1.4 cpm (standard error of the mean 0.7 cpm). The limit of detection for radioactivity in a fraction was calculated on the basis of 2 standard deviations. Control experiments Background Liver DNA was isolated from an untreated rat. The radioactivity in this DNA, in comparison with historical controls, demonstrated that the DNA samples were isolated without external contamination with radiolabel. On the basis of 15 background values compiled during the last 3 months, a mean background value of 12.0 cpm with a standard deviation of 0.7 cpm (standard error of the mean 0.2 cpm) was calculated. The limit of detection for radioactivity in a vial was calculated on a level of 2 standard deviations. Contffm~nfftion control A chromatin pellet isolated from the liver of an untreated rat was incubated for 15 min at 4°C with the radiolabelled supernatant from the first chromatin precipitation step of the DNA preparation from a treated animal. DNA isolated after this incubation showed whether radiolabel present in the organ could contaminate DNA in the process of DNA isolation. DNA binding in vitro Calf thymus DNA (5 mg) in 5 ml 0.1 M phosphate buffer (pH 7.4) containing 33 mM KCI, 8 mM MgC12, 4 mM NADP and 5 mM glucose-6-phosphate was incubated at 37°C with 50 ~1 [r4C]DOTC (100 FCi, 0.33 mg) in the presence or absence of rat liver S9 (4 mg/ml) or BSA (4 mgfml) for 1 h. DNA was precipitated by the addition of 2.5 vol cold ethanol and collected by centrifugation for 30 min at x 1000 g. DNA was then resuspended in lysing medium and purified as described above.
182
DNA binding in V79 Chinese hamster cells V79 Chinese hamster cells (clone 65/3) were obtained from Dr. D. Wild, Freiburg, F.R.G. and cultivated in 25 ml growth medium (Ham’s FlO plus 10% pretested foetal calf serum supplemented with penicillin (100 U/ml) and streptomycin (100 pg/ml)) in 75 cm2 tissue-culture flasks at 92 & 3% relative humidity, 5 f 2% CO2 and 37 f 1°C. The day before treatment 2.5 x IO6 cells were plated into 75 cm2 flasks; 18 h later, the medium was replaced by 12.5 ml Ham’s medium supplemented with 3 s FCS and 125 ~1 DMSO containing [14C]DOTC to give final concentrations of 0, 2.5, 5.0 and 10 pg [t4C]DOTC/ml. After a further 21 h the treatment was terminated by washing the cell layer extensively with phosphate-buffered saline. Thereafter, the cells were suspended by trypsinization, counted and centrifuged for 10 min at 1000 x g. The cells were resuspended in lysing medium and DNA was isolated and purified as described above. RESULTS
DNA binding in vitro When calf thymus DNA was incubated with [i4C]DOTC, radioactivity was associated with purified DNA (Table I). The presence of protein either in the form of active rat liver S9 or BSA reduced the amount of radioactivity in the DNA fraction. The extent of this reduction was independent of the type of protein in the incubation. This indicates that the decrease in the amount of radioactivity in the purified DNA fraction was a consequence of passive trapping of DOTC by proteins and not the result of metabolic inactivation of DOTC or its metabolites. In addition, subsequent repurification of the DNA reduced the specific radioactivity significantly, indicating a non-covalent association of DOTC with the DNA. However, it cannot be excluded that part of the remaining radioactivity on the DNA incubated in the absence of any
TABLE
I
RADIOACTIVITY IN DNA ISOLATED AFTER INCUBATION OF CALF THYMUS DNA (1 mg/ ml) WITH [%JDOTC IN THE PRESENCE OR ABSENCE OF ACTIVE RAT LIVER S9 OR BSA None
s9
BSA
68 4.6 x 10’
51 3.4 x 10’
59 4.0 x
(dpm/mg) I. Purification 2. Purification
6740 440
2820 16
3520 23
3. Purification
250
3 12
>20
Protein
addition
Chemical dose @g/ml) Radioactive dose (dpm/ml) Specific activity
of DNA
107
183
protein was due to a covalent interaction. The following experiments demonstrate the biological insignificance of this putative interaction. DNA binding in Chinese hamster V79 cells DNA isolated from V79 cells after incubation with [i4C]DOTC was clearly radiolabelled (Table II>. A rep~~~cation of the DNA did not reduce the specific radioactivity. Nevertheless, this radioactivity is not uneq~i~~~lly an indication of a covalent interaction between the V79 cell DNA and [i4C]DOTC. The V79 cells were growing during the 21 h incubation and consequently synthesizing DNA. Thus, any i4C entering the C-l pool of the cells as a result of metabolism of the [t4C]DOTC could be incorporated into the DNA precursors biosynthetically. In such a case all the radioactivity would be present in normal DNA-bases, rather than DNA-adducts. To test this point, DNA was hydrolysed and deoxyribonucleotides were analysed by RPLC. Enzymatic degradation to the deoxyribonucleotide-~‘-monophosphates and separation of the 4 natural constituents dCp, dGp, dTp and dAp by HPLC resulted in reproducible elution profiles (Fig. 1). The 4 minor peaks eluting after the nucleotides represent small amounts of the nucleosides dC, dG, dT and dA present in the hydrolysates. Dephosphorylation of nucleotides to the corresponding nucleosides often occurs if small amounts of DNA (a few fig) are treated with an excess of micrococcal nuclease and spleen phosphodiesterase. The fractions containing the natural nuckotides and nucleosides contained between 90 and 105% of the radioactivity applied to the column These results clearly show that all radioactivity on the DNA after incubation of V79 cells with [t4C]DOTC resulted from biosynthetic incorporation. No radioactivity was detected after elution of the natural nucleotides in fractions without UV absorbance in the region known to contain the more lipophilic (poly-)deoxyribonucleotidecarcinogen adducts. DNA b~~d~~gin viw irzrat &WVaad thymtrs The levet of radioactivity in the blood at various time-points after oral administration of [‘4C]DOTC is shown in Figure 2. From the values obtained between 24 and TABLE
II
RADIOACTIVITY
IN DNA ISOLATED
CELLS WITH [‘4C]DOTC
AFTER
INCUBATION
OF V79 CHINESE
HAMSTER
._
Chemic& dox ~~/~)
10.9
5.5
Radioactive dose (dpm/mlf
7.3 x 106
3.7 x 105
2.8 I.8 x toe
-
Survival (56)
5
7
88
95
10 850
11580
11140
-
0
Constant specific activity of DNA (dpm/mg)
184
5
10
15 Fraclion
20
25
Fig. 1. HPLC elution profiles of optical density and radioactivity from deoxyribonucleotides obtained after enzymatic digestion of DNA isolated from V79 cells treated with [W]DOTC. Top: Optical density protile, representing the natural deoxyribonucleotides in the order dCp, dGp, dTp and dAp and the natural deoxyribonucleosides dC, dG, dT and dA. Center: DNA isolated from incubation with 5 @g/ml. Bottom: DNA isolated from incubation with 2.5 pg/ml.
96 h the half-life of DOTC in the blood was estimated to be SO--90h (Fig. 2). No sex difference was observed. In order to allow complete metabolism of DOTC in the DNA-binding assay, animals were sacrificed 96 h (i.e. about 1 half-life) after administration of the test compound. Radioactivity was clearly detectable in all DNA samples as shown in Tables III and IV. On the preliminary assumption that all radioactivity measured on the DNA was due to covalently bound metabolites of DOTC, the specific radioactivities were converted to the units of the covalent binding index (CBI = pmol chemical bound per mol nucleotides/mmol chemical applied per kg body wt.). Apparent CBI values of approximately 3 and 9 resulted for liver and thymus DNA, respectively. No sex difference was observed. Radioactivity on the DNA isolated from an animal that has been treated with a radiolabelled substance is not necessarily due to covalent interaction of the test compound with DNA, but could be derived from: (a) non-covalent interaction of the test
185
0
10
xl
30
40
50
Xme Fig. 2. Elimination obtained
TABLE
of [W]DOTC between
from the blood
60
70
80
90
100
;h
of female rat No. 1 and male rate No. 3. The values
24 and 96 h were used to calculate
the half-life of DOTC in the blood.
III
RADIOACTIVITY [%]DOTC
TO MALE
Chemical
IN LIVER
DNA
ISOLATED
AND FEMALE
TifiRAI(f)
96 h AFTER
ORAL
ADMINISTRATION
RATS
Female
Female
Male
Male
Control
6.6
6.6
6.6
6.4
-
4.5 x 109
4.4 x 109
4.5 x 109
4.3 x 109
-
39.5
42.2
21.2
35.7
2.1
2.9
1.9
2.6
nd.
GO.2
GO.2
n.d.
-
2980
1660
5040
2870
340
dose
(mgkg) Radioactive
dose
(dpmkg) DNA Specific radioactivity (dpm/mg) (CBI units) Covalent
binding
(CBI units) Control Protein “Derived n.d. =not
$2.7 -
data (dpm/mg) from the nucleotide determined.
analysis.
OF
186 TABLE IV RADIOACTIVITY IN THYMUS DNA ISOLATED 96 h AFTER ORAL ADMINISTRATION [14C]DOTC TO MALE AND FEMALE TifiRAI(f) RATS Female
Female
Male
Male
6.6
6.6
6.6
6.4
4.5 x 109
4.4 x 109
4.5 x 109
4.3 x
DNA Specific radioactivity (dpm/m& (CBI units)
I68 12
105 7.4
121 8.4
112 8.0
Covalent binding (CBI units)8
n.d.
60.1
n.d.
GO.6
Chemical dose (mg/kg) Radioactive dose (dpm/W
OF
109
“Derived from the nucleotide analysis. n.d. = not determined.
compound with the DNA; (b) contamination of the DNA fraction with radiolabelied protein; (c) biosynthetic incorporation of 14Cinto DNA. Control experiments, the results of which are given in the last column of Table III, demonstrated that a non-covalent interaction did not contribute to the radioactivity associated with the DNA. Table III shows that following treatment with [t4C]DOTC hepatic chromatin proteins were radiolabelled to a greater extent (4&100-fold higher) than the DNA. Previous investigations have shown that the isolation method used yields DNA preparations containing ~0.2% protein. Consequently, protein contamination could not account for more than 20% of the radioactivity in the purified DNA. Animals were maintained for 96 h after treatment with [t4C]DOTC. During this period about 3% of the administered radioactivity was exhaled in the form of COz. This indicated that ‘*C did enter the Ct-pool of the animals and could have entered the DNA precursors and been metabolically incorporated into the DNA during DNA synthesis. To check this point, DNA was hydrolysed and deoxy~bonucleotides were analysed by HPLC. The results showed that 85-100% of the total DNA-associated radioactivity was eluted together with the natural nucleotides (Fig. 3). It is very unlikely that a DOTC-nucleotide adduct would elute in the region of the natural nucleotides due to its lipophilic character, In this HPLC system even the smallest nucleotide adduct, N7-methylguanosine-3’-monophosphate, elutes after the natural nucleotides [7]. No evidence for adduct formation was obtained, therefore. The limit of detection for radioactivity in the fractions collected after the natural nucleotides,
187
30
5
10
15
20
25
5
10
15 Fraction
20
25
1
Fig. 3. HPLC after enzymatic
elution
profile, representing isolated
profiles
digestion
of optical
of liver DNA
the natural
and radioactivity
deoxyribonucleotides
from female rat No. 2. Bottom: obtained
density
from rats treated DNA isolated
with DNA isolated
orally
from deoxyribonucleotides with [Y]DOTC.
obtained
Top: Optical
in the order dCp, dGp, dTp and dAp. Center: from male rat No. 3. Similar elution
from the thymus
density DNA
profiles were
of rats Nos. 2 and 3.
which are known to contain the more lipophilic nucleotidecarcinogen adducts, was in the worst case 3.9 dpm (fraction 16) which was equal to about 10% of the applied radioactivity. The extent of maximum possible genotoxicity expressed in the units of the CBI is therefore in the order of 0.2 for liver DNA and 0.7 for thymus DNA. DISCUSSION
In agreement with the studies of Westendorf et al. [2] radioactivity was detectable in the DNA after incubation of either calf thymus DNA or V79 Chinese hamster cells with [i4C]DOTC. However, further analysis revealed that the radioactivity was non-covalently associated with calf thymus DNA or was incorporated during DNA biosynthesis in the case of the V79 cells. These results, therefore, exclude covalent DNA binding of DOTC in vitro. The reported mutagenicity of DOTC in V79 Chinese hamster cells must therefore be based upon mechanisms other than DNA-adduct formation.
188
In addition a DNA-binding study in vivo failed to demonstrate a covalent interaction of DOTC with liver or thymus DNA. The limit of detection of these studies indicated that the maximum possible CBI for DOTC was less than 0.2 in the liver and less than 0.7 in the thymus. This is about 30 000 times lower than the corresponding value for the strong hepatocarcinogen aflatoxin Bt and about 600 times lower than the value for the moderate carcinogen 2-acetylaminofluorene. The 96 h time-period between administration of DOTC and sacrifice of the animals was selected to permit complete absorption, distribution and metabolism of the test compound. Theoretically, DNA repair could have reduced the adduct level during this time-period. However, in general, bulky adducts (e.g. those from aflatoxin Bi and acetylaminofluorenene) have half-lives of several days [8,9]. Thus, it is very unlikely that repair could have reduced the adduct level to below the limit of detection within the 96 h period between DOTC application and tissue preparation. As demonstrated by Lutz [4], a good correlation exists amongst genotoxic compounds between the level of DNA binding in the liver and the carcinogenicity in a long-term rodent bioassay. CBIs in the order of 103-lo4 are found with potent carcinogens, of about 100 with moderate carcinogens and about l-10 for weak carcinogens. The CBI of less than 0.2 obtained in this experiment suggests that it is highly unlikely that DOTC exhibits any genotoxicity mediated by DNA binding. In conclusion, our study gives no indication for genotoxic activity of DOTC, either in vivo or in vitro at a very low limit of detection. REFERENCES 1 Seinen, W., Vos, J.G., Van Spanje, ganotin
compounds.
compounds
in different
Pharmacol.
42, 1977212.
2 Westendorf, 3 Internal
animal
J., Marquardt,
plastic stabilizer Report,
I., Snoek, M., Brands,
II. Comparative
species with special
H., and Marquardt,
di-n-octyltindichloride
(DOTC).
emphasis
H. (1977) Toxicity organotin
on lymphocytotoxicity.
H. (1986) DNA interaction Arzneim.-Forsch.,
of or-
and organolead Toxicol.
and mutagenicity
Appl. of the
36, 1263-1264.
Ciba-Geigy.
4 Lutz, W.K. (1986) Quantitative
evaluation
tion of direct and indirect carcinogens. 5 Sagelsdorff,
R. and Hooykaas,
in vivo and in vitro studies with various
of DNA binding
J. Cancer
P., Lutz, W.K. and Schlatter,
DNA to the carcinogenic
data for risk estimation
Res. Clin. Oncol.
C. (1983) The relevance
action of hexachlorocyclohexane
and for classifica-
112,85591.
isomers.
of covalent
binding
Carcinogenesis
to mouse liver
4, 1267-1273.
6 Caviezel, M., Lutz, W.K., Minini, U. and Schlatter, C. (1984) Interaction of estrone and estradiol with DNA and protein of liver and kidney in rat and hamster in vivo and in vitro. Arch. Toxicol. 55,97-103. 7 Sagelsdorff,
P., Lutz, W.K. and Schlatter,
C. (1988) DNA methylation
methylhydrazine, and dimethylnitrosamine. 8 Beland, F.E., Huitfeldt, H.S. and Poirier,
in rat liver by daminozide,
Fundam. Appl. Toxicol. 11,723-730. M.C. (1987) DNA adduct formation
and removal
chronic administration of a carcinogenic amine. Prog. Exp. Tumor Res. 31, 33-41. 9 Caviezel, M. (1984) Untersuchungen zur kovalenten Bindung an DNS und Protein Steroide
Cholesterin-5,6-epoxid,
Thesis, ETH Zurich,
Switzerland,
Oestron,
Oestradiol
No. 7564.
und Trenbolon,
sowie durch
l,l-diduring
in vivo durch Aflatoxin
die
Bi. Ph.D.