In vitro induction of micronuclei by monofunctional methanesulphonic acid esters

Chemico-Biological Interactions 164 (2006) 76–84

In vitro induction of micronuclei by monofunctional methanesulphonic acid esters Possible role of alkylation mechanisms Erwin Eder ∗ , Wolfgang K¨utt, Christoph Deininger Department of Toxicology, University of W¨urzburg, Versbacher Str. 9, 97078 W¨urzburg, Germany Received 28 June 2006; received in revised form 22 August 2006; accepted 24 August 2006 Available online 1 September 2006

Abstract Six monofunctional alkylating methanesulphonates of widely varying structures were investigated in the in vitro micronucleus assay with Syrian hamster embryo fibroblast cells. The results were compared with the alkylating activities measured in the 4(nitrobenzyl)pyridine test (NBP-test) and the N-methyl mercaptoimidazole (MMI-test) as measures for SN 2 reactivity as well as in the triflouoroacetic acid (TFA) solvolysis and the hydrolysis reaction as measures for SN 1 reactivity in order to provide insights into the role of alkylation mechanisms on induction of micronuclei. Moreover we compared the results of micronucleus assay with those of the Ames tests in strain TA 100 and TA1535 and with those of the SOS chromotest with the strains PQ37, PQ243, PM21 and GC 4798. The potency of methanesulphonates to induce micronuclei depended only to a certain degree, on the total alkylating activity (SN 1 and SN 2 reactivity). An inverse, significant correlation between the Ames test and the micronucleus assay was observed and an inverse correlation between the micronucleus assay and the SOS chromotest with the different strains. The results indicate that the primary mechanism leading to induction of micronuclei is not O-alkylation in DNA as it is the case in the Ames test with the hisG46 strains TA1535 and TA100 and not N-alkylation as with the SOS chromotest. There is evidence that protein alkylation, e.g. in the spindle apparatus in mitosis is decisive for induction of micronuclei by alkylating compounds. The structurally voluminous methanesulphonates 2-phenyl ethyl methanesulphonate and 1-phenyl-2-propyl methanesulphonate show a clear higher micronuclei inducing potency than the other tested though the bulky methanesulphonates possess a lower total alkylating activity than the others. This effect can be explained by a higher disturbance during mitosis after alkylation of the spindle apparatus with the structurally more bulky methanesulphonates. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Micronucleus assay; Alkylating compounds; Comparison with genotoxicity tests

1. Introduction The induction of micronuclei, in particular in bone marrow is a measure for cytogenetic damage and meth-

∗ Corresponding author. Tel.: +49 931 201 48926; fax: +49 931 201 48446. E-mail address: [email protected] (E. Eder).

ods to measure the micronuclei formation have been developed. [1]. The test system is now in widespread use and it is commonly referred as “micronucleus assay” [1]. The micronucleus assay can also be performed in vitro, in cell cultures, for instance in Syrian hamster embryo fibroblasts. The induction of micronuclei in this rapid and sensitive system is regarded as a qualitative measure of genotoxicity for a series of mutagens and carcinogens [2]. The mechanisms of induction of

0009-2797/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.cbi.2006.08.020

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

micronuclei are not entirely clear, however, impacts on the mitotic spindle apparatus are, in general, considered as important, initiating actions [1,2]. In a series of micronucleus tests also alkylating compounds were found positive [1,2]. The question is whether formation of DNA adducts by alkylating compounds leading to mutations is also an important mechanism in the induction of micronuclei. In this case positive correlations between the results of the Ames test with Salmonella typhimurium strains (His G46) sensitive for base pair substitution and that of the micronucleus assay should be observed when measuring monofunctional alkylating compounds. Monofunctional methanesulphonic acid esters (methanesulphonates) revealed as a well suitable class of test compounds to study the role of alkylation mechanisms in induction of genotoxicity [3–8]. Investigation of the role of structures and alkylation mechanisms upon the induction of micronuclei may provide new insights into the mechanisms of micronuclei formation. Furthermore, the importance and the position of the micronucleus assay as pre-screening test for the class of alkylating compounds and for substances which are bioactivated to alkylating, ultimate mutagens and carcinogens by metabolism can be ascertained. Another point of interest are possible residues of genotoxic alkyl mesylates in mesylate salt drug substances [9]. In order to assess the mutagenic and car-

77

cinogenic potential of these possible impurities it is necessary to test their genotoxicities in an appropriate battery of pre-screening tests [10]. For this purpose it is a prerequisite to know as much as possible about the suitability and the reliability of the different test systems for the respective classes of compounds. We therefore investigated six selected methanesulphonates of widely varying structures and varying alkylating activities in the micronucleus test in Syrian hamster embryo fibroblasts and tried to correlate the results with the SN 1 and SN 2 reactivities, with the results of the Ames test using S. typhimurium strain TA 100 and TA1535 and with the results of the SOS chromotest in Escherichia coli PQ 37, E. coli PQ 243, E. coli PM 21 and E. coli GC 4798. 2. Materials and methods 2.1. Test substances The methanesulphonates were synthesized as described earlier [3] using a modification of Crossland and Servis [11]. Table 1 shows the chemical structures, abbreviations, yield in synthesis and their purity. The purity was determined by capillary gas chromatography. The purity of substances which decomposed during gas chromatography was measured by titration and

Table 1 Structures, abbreviations, yields and purities of the synthesized methanesulphonates Alkylmethanesulphonates

Structures

Abbreviations

Yield (%)

Purity (%)

Methyl-

CH3

MMS

86

99

Cyclopropylmethyl-

CpMMS

81

98a

2-Phenylethyl-

2PhEMS

82

98

2-Propyl-(iso-propyl-)

iPMS

92

99

2-Butyl-(sec-butyl-)

sBMS

91

99

l-Phenyl-2-butyl-

lPh2PMS

83

98a

a

Purities determined by titration and estimated by 1 H NMR 96%.

78

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

estimated by 1 H NMR spectroscopy. The 1 H NMR data were presented earlier [3]. 2.2. Micronucleus assay The in vitro micronucleus assay with Syrian hamster embryo fibroblasts was performed as described earlier [2]. In brief: cell cultures were established from 13day-old embryos of Syrian hamsters. Tertiary cultures (1.5 × 105 cells) were incubated for 24 h. The cultures were treated with the test substances diluted in dimethylsulphoxide. After 5 h the substances were removed by changing the medium. After 18 h the cells were fixed, stained and scored for micronuclei microscopically. Only structures smaller than 1/3 of the nucleus were counted. Moreover, only micronuclei clearly distinct from the cell nucleus were taken into consideration. Structures similar to the nucleic vesicles were not counted. The number of cells containing micronuclei was determined among a population of 2000 cells. In untreated cells 14.1 ± 4.3 micronuclei/2000 cells were measured. In order to obtain comparable values for the genotoxic potencies of the substances specific activities (micronuclei/␮mol in 2000 cells) were determined from the initial slopes of the dose response curves (Fig. 1). Five scores were performed for each methanesulphonate and the mean values and standard deviations were presented. 2.3. Salmonella mutagenicity assay The S. typhimurium test strains TA1535 and TA 100 were kindly provided by Dr. Bruce N. Ames. The S. typhimurium strain TA 1535 carries the hisG46 missense mutation GGG instead of GAG. Therefore, O6 -alkylation leads to GC → AT transition since O6 alkylation results in backmutation of GGG to GAG in

this strain. Due to umuDC deficiency strain TA 1535 is not capable of error prone repair. Thus the promutagenic O6 -alkylation [12] is of crucial importance for the backmutation in strain TA 1535. Strain TA 100 is derived from strain TA 1535 but it contains, in addition, the plasmid pkM 101, which allows this strain error prone repair. Thus a variety of DNA-lesions can indirectly lead to backmutations in strain TA 100. This strain is also sensitive for O-alkylation but not as specific as strain TA 1535. The pre-incubation assay carried out was described by Maron and Ames [13]. The overnight culture bacterial suspension contained 2 × 109 cells/ml. Incubation time was 30 min. The colonies were counted with a Biotran III colony counter. The sensitivity of the strain was routinely checked using 1 ␮g NaN3 per plate as positive control (600 ± 36 revertants/␮g and plate with strain TA1535 and 650 ± 29 revertants/␮g and plate for strain TA 100). The spontaneous revertants were 22 ± 4 revertants for TA 1535 and 200 ± 20 revertants for TA 100. The solvent was dimethylsulphoxide. The mean values of three independent double assays were used to determine the mutagenicity in revertants/␮mol per plate. A substance was considered as significantly mutagenic if the revertants per plate were at least twice those of the background of the spontaneous revertants. 2.4. SOS-chromotest 2.4.1. PQ strains The method of Quillardet and Hofnung [14] was slightly modified as described by Eder et al. [4] using dimethylsulphoxide (DMSO) as standard solvent. The strains E. coli PQ 37 and PQ 243 were kindly provided by Dr. Philippe Quillardet, Institute Louis Pasteur, Paris, France. The strain PQ 243 is derived from PQ 37 but PQ 243 is lacking the 3-methyladenine-DNA-glycosylases I and II (TagI and TagII). The strains possess the sfiA: lacZ gen fusion [15] so that induction of the sfiA gene an early and important gene in the cascade of the SOS-response can be measured as an increase in ␤-galactosidase activity (␤-gal). As a measure of genotoxicity the SOSIP (SOS-inducing potency = I (nmol)) was calculated from the linear part of the induction factor (I)–dose response curve. I(c) values were calculated according to the following equations: R=

Fig. 1. Induction of micronuclei by methanesulphonates in dependence on the concentrations.

␤-gal units alkaline phosphatase units

I(C) =

R(C) R(0)

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

Alkaline phosphatase activity, a measure of protein synthesis, was used for toxicity correction. I(c) was the ratio of R(c) at a given concentration to the background R(0) . Substances inducing maximal I(c) values of at least 1.5 that of background are regarded as significantly genotoxic in this test. The mean values of three to four independent determinations were calculated. A SOSIP of 27 was obtained with 4-nitroquinoline N-oxide (4-NQO) as positive control in the case of the strain PQ 37 and of 43 in the case of strain PQ 243. 2.4.2. Escherichia coli K12 strains The E. coli K 12 strain AB1157 strain carries the following markers: thr-1 leu-6proA2 his-4 arg E3 thi lacY1 galK2 ara-14 xyl-5 mtl-1 tAB1157 by introduction of the sfiA::lac Z gene fusion [16]. The E. coli strain K 12 GC4798 was constructed by Boiteux et al. [17]. It carries the same markers as AB1157 and, in addition, X::Tn5 tag A1 alk A1 (p0sfiA lac) (clind). The tag gene codes for the 3-methyladenineDNA glycosylase Tag I and alk A for 3-methyladenineDNA glycosylase Tag II. The loss of the glycosylases increases the sensitivity of the strains for N-alkylation in DNA. The SOS chromotest with the K 12 strains was performed as described elsewhere [5]. The test was a modification of the procedure described by Miller [18]. The ␤-gal enzyme units (Up ) were determined as measure for the induction of the SOS repair according to the following equation: Up = 1000

A420 − 1.75A550 tVA600

where A420 is absorbance of the reaction mixture at 420 nm (␤-gal reaction with ONPG), A550 absorbance of the reaction mixture at 550 nm (correction for the light scattering by cell fragments), A600 absorbance of the mixture before adding buffer Z at 600 nm (toxicity correction, growth delay), t incubation time, V volume in ml of the bacterial culture which was used in the ␤-gal test; ONPG: o-nitrophenyl-␤-galactopyranoside The enzyme units Up per ␮mol (Up /␮mol) were determined from the linear part of the dose response curve. A substance was considered as genotoxic if the maximal value of Up , whatever the dose, was at least 1.5 times the background value (usually 65 for strain PM21 and 77 for GC4748). The mean values for at least three independent determinations were calculated. The PM21 strain and the GC4748 strain were kindly supported by Prof. Alain Favre, Institute Jacques Monod, Paris, France.

79

2.5. Determination of the alkylating activities The SN 1-reactivities of the methanesulphonates were determined using the model nucleophiles (a) H2 O and (b) trifluoroacetic acid and the SN 2-reactivities were measured with the model nucleophiles (c) 4-(p-nitrobenzyl)pyridine and (d) N-methyl mercaptoimidazole as described in detail elsewhere [3]. The SN 2 reactivities in the NBP test were presented in absorbance (extinction) per mol/l at 560 nm. The SN 2 type alkylation at the sulphur atom of the MMI was followed by 1 H NMR spectroscopy and the half-life (t1/2 ) was determined. As a measure for the alkylating activity the second order rate constant kMMI = (t1/2 × co ) was used. The SN 1 type reaction of TFA (solvolysis) is also followed by 1 H NMR-spectroscopy. The kinetic first order constant kTFA = ln 2/t1/2 was used as measure for SN 1 reactivity. The hydrolysis reaction (kH = ln 2/t1/2 ) as another measure for SN 1 reactivity was followed by titration at 37 ◦ C with an autotitrator system. For better comparison the relative values are presented. The absolute reactivities of iPMS are normalized to 1 and the values of the other substances are related to this value. 2.6. Statistical analysis The statistical analysis of the correlations was performed by the Pearson product moment correlation with the Sigma Stat program. Results with P-values below 0.050 were considered as statistically significant. 3. Results All six methanesulphonates tested clearly induced micronuclei in a concentration dependent manner. In Table 2 the mean values of five independent determinations and standard deviations of the micronuclei formation for four different concentrations are presented. The micronuclei increased with rising concentrations up to a maximum and than decreased with higher concentrations. The dose response curves for the six methanesulphonates are shown in Fig. 1. Table 3 presents the induction of micronuclei per ␮mol as determined from the initial linear slopes of the dose response curves (see Fig. 1 and methods) and, in addition, the alkylating activities, measured in four different alkylation test systems are shown. In Table 3 the NBP-test and the MMI test stands for SN 2-reactivity whereas the TFA-solvolysis and the hydrolysis represents the SN 1 reactivity. Moreover, the hydrolysis constant also indicates the life time of the respective methanesulphonate in the aqueous test

80

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

Table 2 Induction of micronuclei by methane sulphonates in dependence on concentrations [micronuclei/2000 cells, mean values ± standard deviation, background of 14 subtracted] Substance

MMS

Concentration 10 (␮mol/plate) 5 (␮mol/plate) 1 (␮mol/plate) 0.5 (␮mol/plate)

31 40.4 37 17.4

± ± ± ±

CpMMS 1.0 2.4 3,39 2.07

36.2 56.4 36.8 17.0

± ± ± ±

2PhEMS

6.83 2.51 5.63 3.16

27 31.6 42.6 17.0

± ± ± ±

5.7 11.5 4.28 3.67

iPMS 34 34.4 17.6 11.8

± ± ± ±

sBMS 10.82 10.53 3.65 3.56

22 27.6 29 16.6

± ± ± ±

1Ph2PMS 2 4.39 7.45 6.54

34.4 46.4 37 15.4

± ± ± ±

3.44 2.07 4.18 3.05

Table 3 Induction of micronuclei and alkylating activities of methanesulphonates Substance

NBP-test

MMI-test

TFA-solvolysis

Hydrolysis

Micronucleia

MMS CpMMS 2PhEMS iPMS sBMS 1Ph2PMS

145.00 153.00 3.20 1.00 1.10 0.79

1000.0 99.0 9.7 1.0 6.3 1.9

0.0030 10.4000 0.0121 1.0000 1.8600 2.8900

0.0232 17.0000 0.0001b 1.0000 2.2700 0.1140b

39.2 39.6 51.2 11.6 24.8 43.2

a b

Micronuclei/␮mol/2000 cells. Insufficient solubility in water.

systems. Relative values are presented for the alkylating activities. The absolute values of iPMS were normalized to 1 and the activities of the other substances were related to this value. The absolute NBP value, given in absorbance at 560 nm and referred to mol/l of iPMS is 11 and that of MMS is 1600. The kMMI of iPMS is 2.67 × 10−3 (l/(mol min)) and that of MMS is 2.71. The kTFA of iPMS is 0.8 (min−1 ) and that of MMS is 2.6 × 10−4 . The kH of iPMS is 2.97 (min−1 ) and that of MMS is 0.069. It has to be mentioned that due to insufficient water solubility of 2PhEMS and 1Ph2PMS the hydrolysis rate constants for these two compounds did not reflect the actual SN 1 reactivities of these compounds. Highest induction of micronuclei was observed

for 2PhEMS, followed by 1Ph2PMS. Relatively high values were also seen for CpMMS and MMS and the lowest values were observed with sBMS and iPMS. Micronucleus formation generally increased with rising SN 2 reactivities as measured by the NBP-test and the MMI-test to a certain extent (Fig. 2a and b). MMS and CpMMS, the compounds with very high SN 2 reactivities showed however a lower micronuclei induction than 2PhEMS and 1Ph2PMS with only moderate or even low SN 2 reactivities (Fig. 2a and b). When considering the SN 1 reactivities as measured in the TFA-test and the hydrolysis test there was no clear correlation with the micronucleus inducing potency (Fig. 3a and b). It is remarkable that 2PhEMS which possesses only a

Table 4 Comparison of micronuclei formation with bacterial mutagenicity in the Ames test and genotoxicity in the SOS-chromotest in different Escherichia coli strains Substance

Micronucleia

AmesTA 100b

Ames 1535b

SOS PQ37c

SOSPQ243c

SOSPM21d

SOSGC4798d

MMS CpMMS 2PhEMS iPMS sBMS lPh2PMS

39.2 39.6 51.2 11.6 24.8 43.2

493 415 458 2254 1370 189

44 39 437 2273 1350 180

21.6 n.d.e 2.5 12.2 6.6 7.7

196.0 n.d. 4.4 20.6 13.1 10.1

325.0 n.d. 2.6 70.0 27.0 3.9

2789.0 n.d. 7.1 112.0 33.5 16.0

a b c d e

Micronulei/␮mol/2000 cells. Revertants/␮mol. SOSIP × 10−3 . Up /␮mol. Not determined.

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

81

Fig. 3. (a) Correlation between induction of micronuclei and the mutagenicity in the Ames preincubation assay in strain TA 100 measured as revertants/␮mol. (×) MMS; () CpMMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS; (3) correlation between induction of micronuclei and the Ames preincubation assay in strain TA 1535 measured as revertants/␮mol. (×) MMS; () CpMMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS.

Fig. 2. (a) Relationship between induction of micronuclei and SN 2 reactivities of the methanesulphonates, tested in the NBP. Absorbance at 560 nm is a measure of SN 2 reactivity. (×) MMS; () CpMMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS; (b) relationship between induction of micronuclei and SN 2 reactivities of the methanesulphonates, tested in the MMI test as second order kinetic constant. (×) MMS; () CpMMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS; (c) relationship between induction of micronuclei and SN 1 reactivity, measured in the TFA test as first order kinetic constant of the TFA solvolysis. (×) MMS; () CpMMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS.

moderate SN 2 reactivity and a very low SN 1 reactivity exerted the highest induction of micronuclei (see Table 3). In Table 4 the results of the micronucleus assay are compared with those of other genotoxicity tests i.e. the Ames test with the S. typhimurium strains TA100 and TA1535 and the SOS chromotest with the E. coli strains PQ37, PQ243, PM21 and GC4798. We found no positive correlation between the results of the micronucleus test and any of the other genotoxicity/mutagenicity tests compared here. There is even a statistically signifi-

82

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

cant, inverse correlation between the micronucleus assay and the Ames test with strain TA 100 (correlation constant −0.943; P-value 0.0048) (Fig. 3a) and with strain 1535 (correlation coefficient −0.893; P-value 0.0012) (Fig. 3b). Again, 2PhEMS showed a deviation from the correlations between micronucleus assay and the Ames test with strains TA 100 and TA 1535. Inverse correlations were also observed between the micronucleus test and the SOS chromotest with strains PQ243 (correlation coefficient −0.969; P-value 0.031), PM21 (correlation coefficient −0.948; P-value 0.052) and GC4798 (correlation coefficient −0.907; P-value 0.093) if omitting the values of MMS. The values of MMS did not fit into these correlations (Fig. 4a and b). In Fig. 4c MMS was not included because the Up value in strain GC4798 is far out of the scale of the abscissae axis (see Table 4). CpMMS was not included in this series of SOS tests because sufficient amounts of this compound were no more available for the SOS chromotests because of decomposition due to its chemical instability. The correlation between the micronucleus assay and the SOS-chromotest with strain PQ 37 was not as distinct as with the correlations with the SOS-chromotest in the other strains, however the MMS value again did not fit into the correlation (figure not shown). 4. Discussion Most of the genotoxic carcinogens and mutagens are either directly alkylating substances or compounds, which are bioactivated to alkylating intermediates by metabolism [19]. The alkylating compounds or intermediates can react with nucleophilic sites in the critical biological targets, in particular with DNA. It has long been discussed that different alkylation mechanisms (SN 1 and SN 2) lead to different binding patterns, which are considered to decisively influence genotoxic, mutagenic and carcinogenic outcomes [5,20,21]. Methanesulphonates are an appropriate class of compounds to study the influence of alkylation mechanisms and other structural factors on genotoxic primary events leading to mutation and cancer [3,4,7,22–25]. The micronucleus assay is now a widely used test for genotoxic effects. The Fig. 4. (a) Correlation between induction of micronuclei and induction of SOS-repair measured in the SOS chromo test with strain PQ 243 as SOSIP; (×) MMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS; (b) correlation between induction of micronuclei and induction of SOS-repair in the SOS chromo test with strain PM21 as Up . (×) MMS, (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS; (c) relationship between induction of micronuclei and the induction of SOS-repair measured in the SOS chromo test with strain GC4798 as Up ; (䊉) 2PhMS, () iPMS, () sBMS and () 1Ph2PMS.

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

primary mechanisms leading to formation of micronuclei are, however scarcely known. In our series of tests all alkylating methanesulphonates clearly induced micronuclei. Thus DNA lesions by alkylation at first glance may be considered as the primary mechanism leading to induction of micronuclei as it is the case with other genotoxicity tests. When considering the relationships between the SN 1 reactivities or the SN 2 reactivities and the micronucleus induction none of these alkylation mechanisms can be attributed as specific, underlying mechanism. To a certain degree there is a general dependence of micronuclei induction on the total alkylating activity (SN 1 and SN 2reactivity) with a preference for the SN 2 mechanism. Evidently O-alkylation in DNA is not of importance for induction of micronuclei. Moreover it is questionable whether DNA alkylation actually plays a role in the induction of micronuclei. Further information is provided by the correlations with the mutagenicities in the hisG46 strains. In earlier publications we have demonstrated that O-alkylations in DNA is decisive for mutagenic activities in the S. typhimurium hisG46 strains, in particular in the TA 1535 strain [3,6–8]. Actually we do not find a positive correlation between micronucleus induction and mutagenicity in the strains TA 100 and TA1535. There are even statistically significant, inverse correlations (Fig. 3a and b). The only exception of the otherwise nearly linear inverse correlations is the structurally relatively voluminous 2PhEMS whose micronuclei inducing potency is higher than expected from the otherwise linear inverse correlation. The significant, inverse correlation between induction of micronucleus and mutagenicity in the Ames test with the hisG46 strains confirm the hypothesis that O-alkylation in DNA is not an important mechanism in the induction of micronuclei. In earlier studies we have shown that a high proportion of DNA alkylation induces SOS repair [5,8]. Inverse correlations were also found between the micronucleus assay and the SOS chromotest with the E. coli strains PQ243, PM21 and GC4798 (Fig. 4a–c) if omitting MMS, which did not fit into this correlation. These inverse correlations indicate that DNA alkylation is not the primary mechanism leading to induction of micronuclei. No clear correlation was observed between the micronucleus assay and the SOS chromotest with strain PQ 37, however MMS again deviated from the correlation. We have shown that the very high induction of the sfiA gene in the SOS chromotest by MMS depends on its high potency to form DNA adducts in bacteria [4,5,7,8]. The most probable mechanism of induction of micronuclei is alkylation of proteins, e.g. of enzymes

83

important for DNA replication, DNA repair or of mitosis. In earlier publications it was shown that typical spindle toxins are strong inducers of micronuclei [2] [1]. Alkylation can also impair the spindle apparatus, e.g. by alkylation of the proteins of the microtubuli and thus lead to induction of micronuclei. A strong alkylating activity of compounds leads to high alkylation of proteins, e.g. in the spindle apparatus. The deviation of MMS in the correlations between the micronucleus assay and the SOS chromotests can be explained by the extremely high SN 2 reactivity of MMS causing strong N-alkylation in DNA which leads to a clear induction of the SOS repair, but also to a strong disturbance of the mitosis by alkylating proteins of, e.g. the microtubuli in the spindle apparatus. Therefore MMS is expected to deviate from the inverse correlation between micronucleus formation and the bacterial genotoxicity tests. Further support that disturbance of the mitototic apparatus plays a decisive role could be provided by, e.g. kinetochor straining. Another interesting feature in the induction of micronuclei by methanesulphonates is the high efficacy of the structural more voluminous, phenyl containing compounds 2PhEMS and 1Ph2PMS. The relatively high values of these compounds in the micronuclei assay evidently depend on their bulky structure. The alkylation of the microtubuli with these voluminous methanesulphonates can cause a stronger disturbance in the spindle apparatus than the structurally less bulky methanesulphonates. On the other hand, it cannot be excluded that the very low hydrolysis rate of 2PhEMS and 1Ph2PMS (Table 3) also plays a role for the relatively high induction of micronuclei because these substances possess long life times in aqueous systems and can relatively long act in the cells. In conclusion, alkylating compounds which lead to DNA adducts and initiate cancer are well detected in mutagenicity and genotoxicity pre-screening tests. They are however also well detected in the micronucleus assay although it is unlikely that alkylation of DNA is the primary mechanism leading to induction of micronuclei as it is the case in the other tests. Alkylation at other biological sites e.g. at critical proteins which are of importance for the mitosis can lead to induction of micronuclei. Similar considerations are valid for not directly alkylating mutagens and carcinogens, which have to be bioactivated to alkylating intermediates. The advantage of the micronucleus test is that, in addition to alkylating genotoxic compounds also nonalkylating compounds and substances, which are not bioactivated to alkylating intermediates induce micronuclei and can be detected in the micronucleus assay [1,2]. The micronucleus assay there-

84

E. Eder et al. / Chemico-Biological Interactions 164 (2006) 76–84

fore is a valuable pre-screening method, which allows also detection of carcinogens whose carcinogenic mechanisms are different from DNA alkylation. Acknowledgement We thank Dr. Gaby Schmuck for providing the Syrian hamster embryo fibroblast cell cultures and for advice and assistance in the performance of the micronucleus assay. References [1] J.A. Heddle, M. Hite, B. Kirkhart, K. Mavournin, J.T. MacGregor, G.W. Newell, F. Salmone, The induction of micronuclei as a measure of genotoxicity, Mutat. Res. 123 (1983) 61–118. [2] G. Schmuck, G. Lieb, D. Wild, D. Schiffmann, D. Henschler, Characterization of an in vitro micronucleus assay with Syrian hamster embryo fibroblasts, Mutat. Res. 203 (1988) 397– 404. [3] E. Eder, W. K¨utt, The dependence of the mutagenicity of methanesulphonic acid esters in S. typhimurium TA100 on the alkylation mechanisms, Chem. Biol. Interact. 69 (1989) 45–59. [4] E. Eder, C. Deininger, W. K¨utt, Genotoxicity of monofunctional methanesulphonates in the SOS chromotest as a function of alkylation mechanisms. A comparison with the mutagenicity in S. typhimurium TA100, Mutat. Res. 211 (1989) 51–64. [5] E. Eder, A. Favre, C. Deininger, H. Hahn, W. K¨utt, Induction of SOS repair by monofunctional methanesulphonates in various Escherichia coli strains. Structure-activity relationships in comparison with mutagenicity in Salmonella typhimurium, Mutagenesis 4 (1989) 179–186. [6] E. Eder, C. Deininger, M. Wiedenmann, Methyl methanesulphonate (MMS) is clearly mutagenic in S. typhimurium strain TA1535; a comparison with strain TA100, Mutat. Res. 226 (1989) 145–149. [7] E. Eder, W. Wiedenmann, C. Deininger, W. K¨utt, The relationship between mutagenicity in His G46 Salmonella and the O6 guanine alkylation in bacterial DNA by monofunctional methanesulfonates, Toxicol. In Vitro 4 (1990) 167–174. [8] E. Eder, W. K¨utt, C. Deininger, On the role of alkylating mechanisms, O-alkylation and DNA-repair in genotoxicity and mutagenicity of alkylating methanesulfonates of widely varying structures in bacterial systems, Chem. Biol. Interact. 137 (2001) 89–99. [9] C.o. Europe, Enquiry: alkyl mesilates (methanesulphonates) impurities in mesilate salts, Pharmeuropa 12 (2000) 27.

[10] E. Eder, D. Schiffmann, K. Dornbusch, W. K¨utt, C. Hoffman, Genotoxicity of allylic compounds—a quick screening strategy based on structure–activity relationship and a battery of prescreening tests, Fd. Chem. Toxic. 24 (1986) 667–673. [11] F.K. Crossland, K.L. Servis, A facile synthesis of methanesulphonate esters, J. Org. Chem. (1970) 3195–3199. [12] A. Loveless, Possible relevance of O-6 alkylation of deoxyguanosine to the mutagenicity and carcinogenicity of nitrosamines and nitrosamides, Nature 223 (1969) 206–207. [13] D.M. Maron, B.N. Ames, Revised methods for the Salmonella mutagenicity test, Mutat. Res. 113 (1983) 173–215. [14] P. Quillardet, C. de Bellecombe, M. Hofnung, The SOS chromotest, a colorimetric assay for genotoxins: validation study with 83 compounds, Mutat. Res. 147 (1985) 79–95. ´ An inducible DNA replication-cell division [15] O. Huisman, R. DAri, coupling mechanism in E. coli, Nature 290 (1981) 797–799. [16] A.C. Caldeira de Araujo, A. Favre, Near ultraviolet DNA damage induces SOS response in Escherichia coli, EMBO J. 5 (1986) 175–179. [17] S. Boiteux, O. Huisman, J. Laval, 3-Methyladenine residues in DNA induce the SOS function sfiA in Escherichia coli, EMBO J. 3 (1984) 2569–2573. [18] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor, NY, New York, 1972. [19] H. Druckrey, H. Kruse, R. Preussmann, S. Ivankovic, C. Landsch¨utz, Cancerogene alkylierende Substanzen. III. Alkylhalogenide, sulfate, sulfonate and ringgespannte Heterocyclen, Z. Krebsforschung 73 (1970) 241–273. [20] P.D. Lawley, S.A. Shah, Reaction of alkylating mutagens and carcinogens with nucleic acids: detection and estimation of a small extent of methylating at O-6 of guanine in DNA by methyl methanesulphonate in vitro, Chem.-Biol. Interact. 5 (1972) 286–288. [21] R.C. Moschel, W.R. Hudgins, A. Dipple, Selectivity in nucleoside alkylation and alkylation in relation to chemical carcinogenesis, J. Org. Chem. 44 (1979) 3324–3328. [22] D.T. Beranek, Distribution of methyl and ethyl adducts following alkylation with monofunctional alkylating agents, Mutat. Res. 231 (1990) 11–30. [23] T.C. Reeja, N. Chandra, K.M. Marimuthu, Chromosomal break points in irradiation and methyl methane sulphonate treated leucocytes of patients with Down syndrome, Indian J. Exp. Biol. 31 (1993) 201–205. [24] M.L. Perez, T.D. Stamato, Time versus replication dependence of EMS-induced delayed mutation in Chinese hamster cells, Mutat. Res. 423 (1999) 55–63. [25] R.P. Paulin, T. Ho, H.J. Balzer, R. Holliday, Gene silencing by DNA methylation and dual inheritance in Chinese hamster ovary cells, Genetics 149 (1998) 1081–1088.