article 3

Abstracts from the Gesellschaft fur Umwelt-Mutationsforschung (GUM) 16 th Annual Meeting in Basel, Switzerland 22-25 April 1997

'ALTERNATIVES TO THE CLASSICAL CHROMOSOME ANALYSIS IN VITRO'

Comparative evaluation of the in vitro micronucleus test and the in vitro chromosome aberration test: Industrial experience.

Silvio Albertini ¹, Beate Miller ¹, Franziska Locher², Veronique Thybaud ³ and Elisabeth Lorge⁴

1Pharma Division, Preclinical Research, Department of Toxicology, F. Hoffmann-La Roche Ltd, CH-4070 Basel (Switzerland), ²Safety-Division, Sandoz Pharma AG, CH-4002 Basel (Switzerland), ³Rhone-Poulenc Rorer - Recherche et Developpement, Departement de la Securite du Medicament (D.S.M.), F-94403 Vitry-sur-Seine Cedex (France), ⁴Institut de Recherches Internat. Servier, F-92415 Courbevoie (France)

This contribution analyses a database assembled in four industrial laboratories using the in vitro micronucleus assay in routine screening for detection of chromosomal aberrations. Besides different Chinese hamster cell lines (CHO-K5, CHO-K1, V79), in one laboratory also human peripheral blood lymphocytes were used for the assay. A comparison of data from the in vitro micronucleus assay with data of standard chromosome aberration assay strongly supports that the in vitro micronucleus test is a suitable alternative for the in vitro chromosome aberration test. The in vitro micronucleus test has several advantages: simplicity of the method, rapidness of the assay, less artificial and less prone to artifacts, potential for automation of scoring. The overall concordance for the dataset of 56 chemicals tested both in the in vitro MNT and CA was about 82% (25 compounds negative in both assays; 21 compounds positive in both assays). The available data indicate, that the in vitro micronucleus test is somewhat more sensitive than the CA (lowest effective doses tested lower for positive compounds). The very high concordance is remarkable in particular taking into account that the compounds evaluated (except for the gyrase and topoisomerase inhibitors) are only weak inducers in general. The in vitro micronucleus test has the potential of not only detecting clastogens but in addition aneuploidy inducing chemicals.

Keywords: MNT in vitro; CA in vitro; correlation; industrial experience

Genotoxic effects of known and suspected aneugens: a comparison of data from the micronucleus, chromosome aberration and gene mutation test in vitro

Peter Arni*, Natasha Lane and Bernhard Ogorek

Novartis Crop Protection Inc., R-1058.3.68, Postfach, CH-4002 Basel, Switzerland

The well known aneugen colchicine and the suspected aneugens chloral hydrate, hydroquinone and thimerosal are known to induce micronuclei in vitro, but data on the induction of chromosomal aberrations, polyploidy and gene mutation are inconclusive or not available. In order to correlate the micronucleus data with other genotoxic effects, these compounds were tested in three different mammalian in vitro systems. For the micronucleus and the chromosome aberration test clone K-5 from cell line CCL 61 (Chinese hamster ovary cells) established in our laboratory was used. The cells were treated with the test compounds for 24 hours in the micronucleus assay and for 18, 24, 42 or 48 hours in the chromosome aberration assay. In the mouse lymphoma test (microtitre plate method) L5178Y TK^+/- cells were treated for 4 hours. Colchicine induced micronuclei at concentrations around 0.01 ug/ml, but not chromosomal aberrations. The compound was clastogenic at high concentrations, induced polyploidy and was negative in the mouse lymphoma test. With chloral hydrate a high percentage of micronuclei and chromosomal aberrations was found around 400 m g/ml. The compound induced also polyploidy, but was negative in the gene mutation assay. Hydroquinone was slightly positive in the micronucleus and the chromosome aberration assay and it revealed distinct effects in the mouse lymphoma test. Thimerosal showed clear-cut effects in both, the micronucleus and the chromosome aberration assay, and it induced polyploidy. It was, however, negative in the mouse lymphoma test. Micronucleus induction does not parallel the formation of structural chromosomal aberrations with colchicine. However, with chloral hydrate, hydroquinone and thimerosal both effects occurred at comparable concentrations.

Keywords: Micronuclei, chromosome aberrations, gene mutations, aneugens

Different DNA damaging species produced by radical reactions of aliphatic aldehydes with cu(ll)

Thomas W. Becker* , Gabriele Krieger, lrene Witte

Carl-von-Ossietzky Universitat Oldenburg, ICBM/FB7, PB 2503, D-26111 Oldenburg, F.R.G.

The transition metal copper is able to induce redox reactions via one electron transfers. Thereby a well-known mechanism is the copper(ll) catalyzed oxidation of hydroquinones followed by a one electron transfer from the reduced copper(l) ion to 0₂. The resulting reactive oxygen species, especially via Fenton-type reactions generated *OH-radicals, are seen as causative DNA damaging species in the Cu(ll)/hydroquinone reaction (1). We found that aldehydes are oxidable by Cu(ll) too, resulting in excessive DNA single and double strand breakage (2). In contrast to the Cu(ll)/hydroquinone system, for aldehydes the DNA strand breaks can not be explained in general by *OH-radicals. The inhibition of strand break formation by catalase in the Cu(ll)/iso-butyraldehyde reaction as well as the formation of 8-0HdG points to *OH-radicals as DNA-damaging species. The reaction of the isomer n-butyraldehyde with Cu(ll) is not to explain in this way. Cu(ll)/n-butyraldehyde did not produce 8-0HdG, which is seen as the specific base modification induced by *OH-radicals or ¹0₂, respectively. In addition this was confirmed by the ineffectivity of catalase to prevent DNA-strand breakage. The formation of excited species others than ¹0₂points to fundamentally different reaction mechanism in Cu(ll) driven oxidations as known so far.

(1) Li Y. and Trush M.A. (1993): DNA damage resulting from the oxidation of hydroquinone by copper: role for a Cu(ll)/Cu(l) redox cycle and reactive oxygen generation. Carcinogenesis, 14 (7), 1303-1311

(2) Becker T.W., Krieger G. and Witte I. (1996): DNA single and double strand breaks induced by aliphatic and aromatic aldehydes in combination with copper(ll). Free Radical Res., 24 (5), 325-332

Keywords: copper toxicity, aldehydes, *OH-radicals, singulet oxygen, triplet carbonyl compounds

Transgenic mice as a tool to analyse the role of DNA primary lesions and 0⁶-methylguanine-DNA methyltransferase in multistage skin carcinogenesis

Klaus Becker*¹, Jorg Dosch², Cornelia Gregel¹ and Bernd Kaina²

1DNA Repair Group, Institute of Plant Genetics, 06466 Gatersleben. ²Division of Applied Toxicology, University of Mainz, 55131 Mainz, Germany

In order to study in vivo the contribution of 0⁶-methylguanine (O⁶meG) adducts to chemical skin carcinogenesis and to define the role of the DNA repair protein 0⁶-methylguanine-DNA methyltransferase (MGMT) in protection against tumor formation during multistage carcinogenesis, we generated transgenic mice that overexpress the human MGMT cDNA in epidermal cells by virtue of bovine cytokeratin III and IV promoter elements. To evaluate both aspects in the process of alkylation-induced tumor initiation, transgenic and nontransgenic mice were topically treated with a single dose of 20 and 50 m mol MNU, respectively, followed by repeated application of the tumor promoter TPA. A significant and dose-dependent skin-tumor response was observed in nontransgenic mice, whereas transgenic mice showed an approx. 6-fold reduction of the number of papillomas per mouse. When DMBA, which induces mutations in mouse skin by other lesions than O⁶alkG was used for tumor initiation, no difference in tumor response of both groups of mice was observed. These data provide clear evidence that MGMT protects cells from alkylation-induced tumor initiation without affecting tumor promotion. G?A base substitutions in codon 12 of the Ha-ras gene were shown to be the predominating mutation responsible for MNU-induced papilloma formation. When the chloroethylating agent ACNU (5 m mol) was used for tumor initiation, CkMGMT transgenic mice showed an approx. 2.5-fold lower papilloma incidence as compared to nontransgenic controls. Sequencing of Ha-ras mutations in ACNU-induced papillomas revealed a higher percentage of genetic alterations outside codon 12 as compared to MNU-induced skin tumors. Moreover, experiments are currently being performed to gain more insights into the relevance of O⁶meG base damage and the protective function of MGMT in malignant progression of papillomas to squamous cell carcinomas. (Supported by DFG grants Be1484/1-2 and Ka724/4-2 to K.B and B.K., resp.)

Keywords: DNA repair, 0⁶-methylguanine-DNA methyltransferase, multistage carcinogenesis, skin, transgenic mice, alkylating agents

Transgenesis and targeted mutagenesis in the mouse

Bluthmann Horst, F.

Hoffmann-La Roche AG, Basel (Schweiz)

Microinjection of recombinant genes into the mouse zygote usually results in the gain of a novel function in transgenic offspring. The promoter of the recombinant transgene determines its transcription pattern and may lead to the ectopic or over expression of a normal gene or to the de novo expression of a mutant gene in specific tissues thus recapitulating human pathophysiological processes in mouse models. The regulated expression of transgenes by tetracycline-controllable transactivation may offer additional clues to the physiological or pathophysiological function of normal or mutant genes in vivo. The target mutagenesis of endogenous genes by homologous recombination in mouse embryonic stem cells allows the inactivation of specific genes in situ and the subsequent analysis of the loss of function in 'knock-out' mice as a tool for basic research or in an attempt to mimic human diseases. Targeted mutagenesis by homologous recombination can be further refined by Cre/loxP-mediated site-specific recombination leading to the subtle mutation of an endogenous gene and the analysis of the change of function in the resulting 'knock-out' mice.

The in vitro micronucleus test as a tool to study resistance to compounds used in tumor therapy

Gunther Boos*¹, Ulrike Diener¹, Frank Gieseler², Helga Stopper¹

*¹Department of Toxicology and ²Medizinische Poliklinik, Medical School, University of Wurzburg, D-97078 Wurzburg

A fast and reliable in vitro test to detect resistance of a tumor against therapeutic compounds before treatment would be of great value. We have used the human leucemic cell line CCRF-CEM and sublines (ACT D 400, ADR 5000, ADR 5000 RCV) for a comparison of the two endpoints of cell growth and micronucleus induction for this purpose. These cell lines are known to exhibit resistancies to several therapeutic drugs with different mechanisms of activity to a different extend. As model compounds for this investigation we used etoposide, mitoxantrone, daunoblastin, and idarubicin. We found a dose-dependent induction of effects in both systems at similar substance concentrations for each of the cell lines. However, the sensitivities for the compounds varied between the cell lines to a great degree (10 to 100-fold). Of both assays, the micronucleus assay in general showed a response at lower substance concentrations than the cell growth assay. Thus, although much research in this area will be needed, it seems that the micronucleus test is a promising tool for the in vitro analysis of resistance to chemotherapeutic compounds that could be tested before therapy.

Keywords: in vitro micronucleus test, cell growth, tumor therapy

Modification of the HPRT-assay for photomutagenicity-testing

Susanne Y. Brendler-Schwaab, B.A. Herbold, E. von Keutz and G. Schluter.

Toxicology, Bayer AG, P.O.Box 10 17 09, D-42096 Wuppertal, Germany

The protocol of the V79/HPRT-assay was modified to evaluate chemicals for photomutagenic effects. UV-irradiation was performed with a low dose (100 J/m² UVA plus 5 J/m² UVB) and a high dose (1O x low dose) using a solar simulator. The photomutagenic potential of Bay Y 3118, a photoreactive fluoroquinolone, was investigated in five protocol variations: 1. Start of treatment with test substance and UV-irradiation at the same time (standard protocol). 2. Preincubation of the cells with test substance in the dark prior to irradiation. Using both protocols, Bay Y 3118 induced significant increases in mutation frequency over the concurrent irradiated vehicle controls. 3. Irradiation of cells and subsequent treatment with non-irradiated substance. 4. Exposure of non-irradiated cells to test substance irradiated in culture medium. 5. Reseeding of cells immediately after treatment and irradiation. No photomutagenic effects were evident using protocols 3-5. Due to these results we conclude: The photomutagenic effect of Bay Y 3118 is not due to- a repair-inhibition of UV-induced DNA lesions - the formation of a stable photomutagenic product after UV-irradiation. The question whether UV-induced radical formation is responsible for the effects seen in the standard protocol is currently under investigation using radical scavengers. The combined use of protocols 1 to 5 makes it possible to detect photomutagens with high sensitivity and to approach the mechanism of the photomutagenic effect.

Keywords: Photomutagenicity, Photo-HPRT-Assay, Fluoroquinolone

Mutagenicity of biodiesel exhausts

J�rgen B�nger*, J�rgen Krahl**, Hans-Ulrich Franke***, Ernst Hallier*

*Zentrum Umwelt- und Arbeitsmedizin, Universitat Gottingen, Waldweg 37, D-37073 Gottingen **Institut fur Biosystemtechnik, Bundesforschungsanstalt fur Landwirtschaft, Bundesallee 50, D-38116 Braunschweig ***Institut fur Maschinenmeb technik und Kolbenmaschinen, Universitat Magdeburg, Universitatsplatz 2, D-39106 Magdeburg

Diesel engine emissions (DEE) act as mutagens in various in vitro test systems and are classified as probable carcinogenic agents to humans by the International Agency for Research on Cancer (IARC 1989). The mutagenic effects of particulate extracts of DEE from a VW Vento TDI using rape seed oil methylester (RME) as fuel were directly compared to DEE of fossil diesel fuel (DF) in the Salmonella typhimurium/mammalian microsome assay (AMES-test). in the tester strain TA 100 the following results were obtained: In the tester strain TA98 we found similar elevated reversion rates. Testing with activated liver S9-fraction induced a slight additional increase. TA97a and TA102 showed no significant enhancement of spontaneous mutation rates. These results indicate a higher mutagenic potency of DEE of DF compared to RME. This was probably due to the lower content of PAH and soot, although the emitted masses of RME were higher in the MVEG-A run cycles used in this study.

Keywords: Diesel engine exhausts, biodiesel, rape seed oil methylester, Salmonella typhimurium/mammalian microsome assay

Induction of DNA damage by hyperbaric oxygen therapy

Claudia Dennog and Gunter Speit,

Universitat Ulm, Abteilung Medizinische Genetik, 89069 Uim

Hyperbaric oxygen therapy is clinically used for the treatment of a variety of diseases. Using the comet assay we have recently shown that HBO treatment under therapeutic conditions (i.e. inhalation of 100% oxygen under a pressure of 1,5 ATA for 60 minutes) clearly and reproducibly induces DNA damage in leukocytes which obviously is due to an increased production of reactive oxygen species (ROS)(1). Therefore, HBO therapy is an excellent system for the investigation of DNA damage induced by oxidative stress in vivo. ROS can attack DNA, thus producing distinctive patterns of DNA alterations including the oxidized base 8-oxoguanine which is a premutagenic lesion and seems to be involved in the induction of cancer. HBO induced DNA damage in leucocytes of all test subjects investigated. DNA damage was found directly after the treatment but could not be detected 24 h later. Significant oxidative DNA base damage was found by converting oxidized DNA bases to strand breaks by using bacterial formamidopyrimidine-DNA glycosylase (FPG), a DNA repair enzyme, which specifically nicks DNA at sites of 8-oxo-guanines and formamidopyrimidines. Direct determination of 8-oxo-guanine by HPLC is now in progress to obtain information about the types of oxidative base damage induced by HBO treatment. As DNA damage was detected only after the first treatment but not after further treatments under the same conditions, an increase of antioxidative defences has been suggested. We are now investigating changes in the antioxidant status (i.e. gluthathion peroxidase, catalase, superoxide dismutase and vitamins) after HBO treatment. To elucidate the biological significance of the DNA effects seen in the comet assay, micronuclei are evaluated to see if the observed DNA breakage is relevant for chromosomal damage.

This work was supported by the program `Environment and Health' (PUG) at the Forschungszentrum Karlsruhe with funds of the Department for Environment Baden-Wurttemberg, Germany.

(1) Dennog et al. (1996) Mutagenesis, 11(6) pp 605-609.

Keywords: comet assay, reactive oxygen species, 8-oxo-guanine.

MES microsuspension assay: mutagenicity test with small amounts of test material

Eckhard Deparade* and Peter Arni

Novartis Crop Protection Inc., R-1058.3.42, Postfach, CH-4002 Basel, Switzerland

The AMES assay, a bacterial genotoxicity test generally required for registration, is a rapid and convenient method to get information on the mutagenic potential of a chemical. For a full registration study up to 2 grams of test substance are needed. In early compound development, when often a first indication of the possible genotoxic potential is required, the available amount of substance may be limited. To reduce substantially the required amount of test substance, a scaled-down version of the AMES test, the "AMES microsuspension assay" (originally proposed by Kado et al. (1)), was adapted to the needs of a routine testing laboratory. A number of known bacterial mutagens (4-nitroquinoline-N-oxide, mitomycin-C, 2-nitrofluorene, 9-aminoacridine, 2-aminoanthracene, cyclophosphamide) were tested in both, the conventional AMES test and the AMES microsuspension assay. In the latter 0.5 to 1 x l0⁹ bacteria together with up to 1000 m g test substance and 100 m l S9 mix or buffer were incubated in a total volume of 0.21 ml for 90 min. on a shaker at 37 'C. Thereafter this mixture was added to 2 ml of soft agar and plated on minimum agar plates. The plates were incubated and counted as in the conventional AMES test. Strains Salmonella typhimurium TA 98, TA 100, TA 102, TA 1535, TA 1537 and E. coli WP2uvrA were used. For a full screening test 25 mg of test compound are needed. The results obtained indicate that the sensitivity of the microsuspension assay is at least that of the conventional AMES assay. The advantage of this screening system in comparison to others is that the same bacterial strains are used as in the normal AMES assay used for registration.

Keywords: AMES test, microsuspension assay, screening

(1) Kado et al. (1983) Mutation Res. 121, 25-32

MSH2 mismatch repair activity of CHO cells is growth regulated and inversely related to resistance of cells to methylating agents

Jorg Dosch* and Bernd Kaina

Division of Applied Toxicology, Medical Faculty, University of Mainz, Obere Zahlbacher Str. 67, D-55131 Mainz

Mammalian cells aquire resistance to methylating mutagens and carcinogens by expression of the DNA repair protein alkyltransferase (MGMT). We have generated another class of cells which are methylation resistant without expressing MGMT. The resistance pertained the end points cytotoxicity and chromosomal aberration formation. These cells obviously have gained the ability to tolerate the genotoxic DNA lesion 0⁶-methylguanine which is, in Mex+ cells, repaired by MGMT (tolerance phenotype). Here we show that methylation resistance of the tolerant cells we have generated correlates with reduced G/T mismatch binding and lower level of expression of the mismatch repair protein MSH2. The MSH2 level and G/T binding activity was also shown to correlate inversely with growth rate. Thus, reduction of serum concentration lowered MSH2 and G/T binding whereas serum stimulation induced the expression. These findings are taken to indicate that MSH2-dependent mismatch repair is growth regulated and can undergo quantitative changes which ultimately affect the level of methylating drug resistance.

Keywords: Mismatch repair, alkylating agents, drug resistance

Gentoxicity-test of non-concentrated surface waters - Ames-test -

Matthias Durr*, Lothar Erdinger, Hans-Gunther Sonntag

Universitat Heidelberg, Hygiene-lnstitut, 1m Neuenheimer Feld 324, 69120 Heidelberg, Deutschland

The Ames-Test is one of a number of methods used to test the genotoxic potential of pollutants in surface waters. Because many surface water samples contain pollutants at trace levels, it is often necessary to concentrate the sample prior to testing. However, preconcentration techniques can in some cases induce chemical changes to the micropollutants within the sample. As part of a wider BMBF network plan to assess the suitability of methods developed to test the genotoxic potential of unconcentrated surface water samples, this study has focused on enhancing the sensitivity of the Ames-Test. This has been achieved by evaluating and optimising various components of the Ames-Test including type of bacterial tester, sample size, strain, inoculum, contact time, and S9 concentration. Six strains of the new TA7000 line, used to indicate the exchange of specific base pairs were evaluated. TA7005 was found to be significantly more sensitive than TA100. It was also found that by concentrating the agar, the maximum sample size for each experiment could be raised from 5Om L up to 4mL. From these results the following scheme was derived: Tests should be carried out using 4mL of water sample and pre-concentrated agar. Samples should be assessed using the Plate Incorporation Assay in conjunction with tester strains TA98, TA100 and TA7005. Metabolic activation should be achieved using S9 at 8% or where metabolic activation is not used, a preincubation time of 90 minutes should be employed. Under these conditions, using tester strain TA100 the detection limit for the directly acting standard reference mutagens Nitroquinolin-N-oxide and Nitrofurantoin were found to be 5m g/L and 10m g/L respectively. Detection limits using tester strain TA98 for two substances yielding acting metabolises were found to be 125m g/L for Benz(a)pyrene and 1000m g/L for N-Acetylaminoflurene.

We thank the BMBF for sponsoring this project (sponsoring sign 02 WU 9557/2).

Keywords: Ames-Test, surface water, sensitivity, TA7000

Aneugen induced micronuclei containing the selectable gene do not lead to mutagenesis

lnge Eckert*¹, William J. Caspary² and Helga Stopper¹

1Department of Toxicology, University of Wurzburg, Versbacher Str, 9, 97078 Worzburg, Germany. ²NIEHS/NIH, Research Triangle Park, NC 27709, USA,

Aneuploidy is one of the genetic changes observed in many tumors. However, it is not known whether loss of a single chromosome mediated by micronucleus formation is a viable mechanism leading to genetic desease. Previously we treated L5178Y mouse lymphoma cells with four aneugens and found that, although the induced micronuclei containing predominantly whole chromosomes, they did not induce micronuclei at the tk locus in these cells under the same non-toxic conditions where they induced micronuclei. This suggested that the induction of micronuclei containing whole chromosomes was not an early event leading to phenotypically expressed mutations in these cells. However, it is possible that chromosome 11, on which the tk-gene resides, may be under-represent in the micronuclei. To ascertain the frequency of induction of micronuclei containing chromosome 11, we applied fluorescence in situ hybridization, using a chromosome 11 paint, to colcemide and vinblastine induced micronuclei. We found that the numbers of micronuclei containing chromosome 11 are more than sufficient to be detectable as mutations if these micronuclei lead to viable mutants. We conclude that cells with micronuclei containing whole chromosomes are not precursors of viable mutants.

Keywords: L5178Y mouse lymphoma cells, micronuclei, mutation

Role of reactive oxygen species in photomutagenesis

B. Epe*, M. Pflaum, C. Kielbassa, D. Ballmaier and 0. Will

Institute of Pharmacy, University of Mainz, D-55099 Mainz, Germany

According to an analysis by means of various repair endonucleases, the DNA damage induced by many photosensitizers plus light (acridine orange, riboflavin and porphyrins) in cultured mammalian cells consists to a large extent of 8-hydroxyguanine (8-oxoG) residues (recognized by Fpg protein) and only few other DNA modifications (pyrimidine modifications, AP sites and strand breaks). This damage profile is consistent with a reaction of the cellular DNA with singlet oxygen or with the excited sensitizer itself, but excludes the involvement of hydroxyl radicals. Interestingly, natural sunlight induces a similar DNA damage profile in otherwise untreated HaCaT cells (derived from human keratinocytes) due to endogenous photosensitizers with an absorption maximum between 400 and 450 nm. Pyrimidine dimers (caused by direct UV-B absorption of DNA) are only tenfold more frequent than Fpg-sensitive modifications under these conditions. The mutagenicity observed in the gpt locus of AS 52 cells after treatment with various photosensitizers is well correlated with the number of Fpg-sensitive base modifications induced. This suggests that 8-oxoG is not only a suitable marker lesion, but the most relevant premutagenic DNA modification in these cases, Its premutagenic potential (number of mutants per modification) is similarly high as that of pyrimidine dimers induced by UV-B in this system. The Fpg-sensitive base modifications (8-oxoG) induced by photosensitization are efficiently repaired (^T1/2 1-2 h at 37'C) in several cell lines tested.

Keywords: Photosensitization, 8-Hydroxyguanine, DNA damage, mutagenicity, DNA repair

Evaluating recombinagenicity of genotoxic agents in the somatic mutation and recombination test of drosophila

Hansjorg Frei*, Friedrich E. Wurgler

Institute of Toxicology, ETH and University of Zurich, Schorenstrasse 16, CH-8603 Schwerzenbach, Switzerland

The wing spot test of Drosophila is a Somatic Mutation and Recombination Test (SMART) which detects a vast array of mutagens and promutagens. In this test, two strains are outcrossed to produce larval progeny which are trans-heterozygous for two recessive wing cell markers, i.e. multiple wing hairs (mwh) and flare (flr), both located on the left arms of chromosome 3. In larvae exposed to genotoxic agents, genetic changes are induced in the cells of the wing primordial They become manifest on the metamorphosed trans-heterozygous wings as mutant spots (mwh and/or flr clones). Spot induction is due to loss of heterozygosity and covers several genetic endpoints, in particular somatic mutation, deletion and recombination. In addition to the trans-heterozygous mwh/flr progeny, the outcross carried out in the wing spot test also produces 50% of individuals which are heterozygous for a multiply inverted chromosome 3. In such inversion-heterozygous individuals, mitotic crossing-over is suppressed, and wing spots recovered are due to mechanisms other than recombination, The contribution of recombination to total spot induction can be estimated from a comparison of the mwh clone frequency in inversion-free and inversion-heterozygous flies. Alkylating agents, topoisomerase inhibitors, antimetabolites and DNA synthesis inhibitors have been studied for genotoxicity and recombinagenicity. For most compounds, recombination is the major but not the only mechanism by which wing spots are generated: azaserine (72% recombination), streptozotocin (49%), mitoxantrone (68%), ellipticine (>99%), etoposide (59%), teniposide (71%), camptothecin (88%), aminopterin (84%), 6-azauracil (47%), ribavirin (72%), acyclovir (83%), azidothymidine (79%), dideoxycytidine (83%), trifiuorothymidine (94%). Among these, ellipticine is exceptional as it is an (almost) exclusive recombinagen,

Keywords: Somatic Mutation and Recombination Test (SMART), wing spot test, Drosophila, genotoxicity, recombinagenicity

The automatic analysis of the in vitro micronucleus test on V79 cells

Wilfried Frieauff

Sandoz Pharma Ltd., CH-4002 Basel

The in vitro micronucleus test has been well established for early screening of new chemical entities in industrial toxicology (1). For the assessment of the mutagenic potential of a test compound, the induction of micronuclei in cells has been shown to be a useful parameter. Being used as a measure for numerical and structural chromosome aberrations, the in vitro micronucleus test consists of determining the frequency of micronucleated cells in a representative fraction of the cells. To replace the tedious labor of counting cells and micronucleated cells manually, an automation of the in vitro micronucleus scoring on V79 cells by image analysis has been recently introduced into routine screening in Genetic Toxicology in Sandoz Pharma. The comparison of manual vs automatic micronucleus analysis shows the high degree of concordance of the micronucleus scores for both approaches. The speed advantage due to fully automatic overnight analysis of 16 slides, the good reproducibility of the results and the lack of subjectivity in the micronucleus assessment make automatic image processing a powerful tool for the in vitro micronucleus analysis.

Miller et al: Comparative evaluation of the in vitro micronucleus test and the in vitro chromosome aberration test: industrial experience. In press.

Keywords: in vitro micronucleus test, image analysis

Induction of chromosome aberrations in rad52 mutant cells requires yku70, the yeast Ku70 homologue

Anna A. Friedi, Markus Kiechle, Friederike Eckardt-Schupp

GSF-Forschungszentrum fur Umweit und Gesundheit, institut fur Strahlenbiologie, Postfach 1129, D-85758 Oberschleissheim, Germany

Radiation-induced chromosome aberrations are thought to result from misrepair of DNA double-strand breaks (DSB). However, the relationship between individual pathways of DSB repair and aberration formation is not clear. Therefore, we have analysed the induction of stable aberrations in the yeast Saccharomyces cerevisiae by electrophoretic karyotyping of single-cell clones. In yeast, several mechanisms of DSB repair have been genetically discriminated and mutants are easily avaible. Preferentially, DSB are repaired by homologous recombination controlled by the RAD52 epistasis group of genes; non-homologous end-joining is of minor importance in yeast, but of high importance in mammalian cells. For this pathway, the end-binding protein Ku70, a subunit of the DNA protein kinase, and supposedly its homologue in yeast, yKu70 (� HDF1), play a role. We have analysed ⁶⁰Co-y-induced aberrations in hapioid yeast cells mutated for the RAD52 gene, the yKu70 gene or both genes. We found low and comparable frequencies of aberrational events in wildtype and yku70 mutants. In the rad52 strain, a fivefold increase was detected; at least 50 % of these aberrations seem to represent exchange type events probably caused by illegitimate end-joining. No aberration was found in the rad52 yku70 double mutant. Hence, yKU70 is required for the generation of stable aberrations in a rad52 mutant background, presumably by a non-homologous end rejoining pathway. Thus, for the first time, genes have been identified which play a role in the molecular process of chromosomal aberration formation in eukaryotes.

Keywords: chromosome aberrations, DNA double-strand break repair, homologous recombination, non-homologous end-joining, Ku70 endbinding protein, yeast.

An electrophoretic approach to investigations on induction and repair of DNA breaks and characterization of chromosomal aberrations in yeast

Anna A. Friedl,

GSF-Institut fur Strahlenbiologie 85758 Oberschleibheim, Germany

DNA double-strand breaks (DSBS) are biologically very important among the large variety of DNA lesions. DSBs can be induced by ionizing radiation or various chemical compounds. To understand the biological effectiveness of various DSB-inducing treatments and for the analysis of DSB repair pathways, it is necessary to know the spatial distribution and frequency of chromosomal breaks. We developed a method based on length-dependent separation of chromosomal molecules and fragments by PFGE. Staining of the gel and measurement of the staining signal along the lanes yields DNA mass profiles. The. shape of these profiles is determined by the lengths of the intact molecules, the frequency and spatial distribution of the breaks, and the electrophoretic behaviour of the molecules. Accounting for these parameters, similar profiles can be computed and compared to the observed profiles, thus allowing to determine, whether given assumptions on the parameters of the system (i.e. frequency and distribution of DSB) are consistent with the experimental observations. Using this method, it was possible to demonstrate that the distribution of DSBs induced by sparsely ionizing radiation is consistent with the assumption of a random breakage model, thus allowing an accurate quantification of breaks. The distribution of heavy ion-induced DSBs, however, is overdispersed, which makes an accurate quantification impossible at the moment. In wildtype. yeast cells and most mutants analysed so far, also residual breaks remaining after repair incubation are distributed randomly and the impact of mutations on the repair kinetics can be analysed. However, double mutant cells deficient for the Rad52 and the Hdfl protein exhibit secondary degradation of broken molecules, suggesting that at least one of both proteins is required to avoid degradation. Furthermore, we. found that the simulation procedure. is a valuable tool for the identification and characterization of stable chromosome aberrations in yeast.

High sensitivity of the cytochalasin-block micronucleus test in human biomonitoring

T. Gebel

Medical Institute of General Hygiene and Environmental Health, University of Gottingen, Windausweg 2, D-37073 Gottingen, FRG

The cytochalasin-block micronucleus test, the sister chromatid exchange test and the alkaline elution technique were compared in their ability to detect genetic damage in human peripheral lymphocytes in a case of an elevated occupational exposure with antineoplastic agents. The putatively elevated exposure of the study subjects was caused by a malfunction of safety hoods resulting in leakage of air during antineoplastic drug infusion preparation. Two months after new safety hoods were installed the frequencies of micronuclei and sister chromatid exchanges of exposed nurses (n=10) were still significantly elevated when compared to a matched control (p<0.01 and p<0.05 in the one-sided Wilcoxon-test, respectively). in a second examination seven months later, the frequency of micronuclei was significantly decreased to control values (p<0.05, one-sided Wilcoxon-test, n=6). When comparing the study subjects according to smoking status, it was observed that smokers (n=8) had significantly elevated frequencies of micronuclei and sister chromatid exchanges (p<0.01 and p<0.05 in the one-sided U-test, respectively). No differences in the rate of DNA damage could be detected with the alkaline elution technique suggesting the sensitivity of this method to be low. In conclusion, the cytochalasin-block micronucleus test seems to be a rapid, highly sensitive and specific tool for biological exposure monitoring in comparison to the classically established sister chromatid exchange test and alkaline elution.

Keywords: Human monitoring, cytostatic drugs, cytochalasin-block micronucleus test, sister chromatid exchange, alkaline elution

Detection by chromosome painting and role in defining individual radiosensitivity of complex chromosomal rearrangements

Erich Gebhart*, Susann Neubauer, Jurgen Dunst¹

Universitat Erlangen-Nurnberg, lnstitut f. Humangenetik, Schwabachanlage 10, D-91054 Erlangen, Germany ¹Universitat Halle, Strahlenklinik, Dryanderstr. 4-7, D-06110 Halle, Germany

Chromosome painting has been used for a fast analysis of chromosomal rearrangements in cancer patients exposed to therapeutic irradiation and for in vitro testing of individual radiosensitivity. A varying portion of the observed aberrations were complex chromosomal rearrangements (CCR). A three-colour "FISH-painting" technique was used for their detection in 81 samples of peripheral blood lymphocytes from 66 cancer patients, from 7 healthy controls, and from 3 patients with Ataxia telangiectasia (A.t.). In vitro irradiation of aliquots of the blood samples with 0.7 and 2 Gy of X-rays served as a means for defining individual radiosensitivity. CCR are a very rare event in non-irradiated cells. Lymphocytes of patients who had just undergone radiotherapy did not only exhibit rather high "basic" frequencies of CCR but also responded to in vitro irradiation with a more drastic increase of those than did lymphocytes of non-exposed patients. A high inter-individual variability of thin reaction to in vitro irradiation could be generally stated. The lymphocytes of patients with clinical signs of an outstanding radiosensitivity as well as the lymphocytes of A.t. patients responded with an unusual high frequency of CCR. The presence and frequency of CCR detected by chromosome painting, in addition to the general frequency of aberrations found by this technique, is not only of use for defining individual radiohypersensitivity but it also helps reducing the efforts of cytogenetic analysis of mutagenic influences.

Keywords: Chromosome painting, complex chromosomal rearrangements, radiosensitivity, past radiation exposure

Urinary excretion of 8-hydroxydeoxyguanosine: no influence of smoking

Dietmar Germadnik, Alexander Pilger, Hugo W. Rudiger

Department of Occupational Medicine, University of Vienna, Wahringer Gurtel 18-20, 1090 Vienna, Austria

Endogenous formation of reactive oxygen species has been proposed as an important factor of cumulative cancer risk, Smoking can enhance the generation of free oxygen radicals and is a possible source of radicalinduced cellular damage. 8-Hydroxydeoxyguanosine (80HdG) is a major oxidative DNA modification and has earned much interest due to its miscoding potential and implication in carcinogenesis. 80HdG is repaired efficiently via enzymatic pathways and is excreted into urine undergoing no further metabolism. To assess the effect of smoking on the formation and repair of 80HdG, the urinary 80HdG levels of 82 smokers have been measured and compared to the values of healthy non-smokers. No statistical significant difference in the mean excretion levels of 80HdG between smokers and non-smokers was observed (2.06 � 1.22 m mol 80HdG / mob creatinine in smokers and 1.95 � 1.24 m mol 80HdG 1 mol creatinine in non-smokers). Therefore, by measuring 80HdG concentrations in spontaneous urine, we could not confirm smoking induced oxidative DNA damage.

Keywords: Smoking, 8-Hydroxydeoyguanosine, Oxidative DNA damage

Gum arbeitsgruppe "mutagenicity testing of biotechnology products"

Elmar Gocke F.

Hoffmann-La Roche AG Pharma Preclinical R & D, Toxicology CH-4070 Basel

The evaluation of mutagenic potential has since many years become an integral part of the toxicological investigations of low-molecular-weight chemicals. The possibility of a "direct" electrophilic attack on DNA has been and still is - the main reason for concern. Indirect interactions with eg enzymes of DNA metabolism (precursor synthesis, replication, repair....) and with the processes of chromosome distribution have also to be taken into consideration. Regarding risk assessment the various modes of action might lead to different risk scenarios. The potential of biotechnology products, in particular of high-molecular-weight recombinant proteins, to elicit DNA damage in somatic or hereditary cells cannot be compared to that of the low-molecular-weight chemicals. Electrophilic reactions are hardly imaginable, while indirect reactions via DNA metabolism and growth regulation seem possible for very specific proteins. The capacity of the standard mutagenicity tests to recognise such modes of action appears to be highly doubtful. Since no general approach/guideline has been adopted the extend of testing has been rather variable. In order to obtain a basis for further discussions into the merits of performing genotoxicity tests a questionnaire was send to pharmaceutical companies and contract laboratories asking for their experiences. Up to now test results have been reported for some 30 compounds with a total of about 90 tests, comprising almost exclusively the standard systems. The overwhelming majority of the tests gave no indication of genotoxic activities. A few cases of positive responses have been variably attributed to contaminants, artefacts (e.g. liberation of histidine) or indirect action related to the pharmacologic activity. It will be discussed what extend of testing might be recommended for proteinatious compounds. In addition, an outlook on potential problems with other products of biotechnology (eg antisense oligonucleotides) will be presented.

Keywords: guideline, questionnaire, biotechnology, test battery, relevance

Evaluation of a modified Ames assay to measure mutagenic potency of vapor-phase compounds

Waltraud Goggelmann*, Astrid Winkler, Paul E. Kreuzer, Johannes G. Filser, GSF - National Research Center for Environment and Health, Institute of Toxicology, D-85764 Neuherberg

A vaporization technique to measure the mutagenic potential of gases and volatile liquid compounds in the Ames assay was developed. Ethylene oxide (E0) acid propylene oxide (PO) were tested with Salmonella strains TA 100 and TA 1535 using 6.4 I desiccators with GC septum for dosing and chemical analysis. Three petri dishes with agar and bacteria were placed in each of the desiccators and different concentrations of EO or PO were injected to the gas phase of individual desiccators. Immediately after dosing the desiccators were placed in an incubator at 37'C. The concentration of EO or PO in the atmosphere of the individual desiccators was monitored by taking small air samples and periodically analysing them with a gas chromatograph. Within 60 min an initial decline of EO or PO in the atmosphere was observed. After this time the concentrations of EO or PO in the air remain constant for the next hours which means that equilibrium between atmospheric EO or PO and agar has been achieved. From the experimental data and initial concentrations the agar:air partition coefficients (K_eq) of EO or PO were calculated. With these partition coefficients it was possible to calculate the different concentrations of EO or PO in the agar. Initial atmospheric concentrations of 16 - 78 ppm EO resulted in doses of 27 - 133 m g EO in the agar layer of one petri dish, whereas 8 - 42 ppm PO corresponded to 16 - 88 m g PO per plate. EO and PO caused a clear dose-dependent increase in the number of revertants in the Salmonella strains TA 100 and TA 1535. In strain TA 100, the highest concentration tested of EO or PO increased the number of revertants by a factor of about 7 or 5 beyond the spontaneous revertant numbers, The mutagenicity was much higher with strain TA 1535 where the same amounts of EO or PO induced a 75-fold or 50-fold increase of revertant numbers as compared to control values. This vaporization technique together with GC analysis allows chemical and biological analysis of gases and volatile compounds. Moreover, with this technique it is possible to calculate in the agar layer the concentration of test compound which induces the reversion of his- to HIS⁺ bacteria.

Keywords: Ames assay, vaporization technique, GC analysis

Skin-targeted expression of the human MGMT protein protects against chloroethylnitrosourea-induced skin tumor initiation.

Cornelia Gregel*1, Klaus Becker¹ and Bernd Kaina²

1DNA Repair Group, Institute of Plant Genetics 06466 Gatersleben. ²Division of Applied Toxicology, University of Mainz, 55131 Mainz,Germany

Chloroethylnitrosoureas (CNU's) represent an important class of chemotherapeutic drugs for treatment of many types of cancer. These bifunctional alkylating agents produce monoadducts, such as N⁷guanine and 0⁶guanine alkylations, exocyclic adducts and inter- as well as intrastrand crosslinks. The latter class of DNA lesions is considered to be the ultimate cytotoxic lesion responsible for the therapeutic effect of the agents. CNU's also possess mutagenic and carcinogenic activity that eventually leads to secondary malignancies after CNU-chemotherapy. It is, therefore, important to gain insight into the main precarcinogenic lesions induced by different CNU's and the mechanism of protection against CNU-induced tumorigenicity mediated by 0^6-methylguanine-DNA methyltransferase (MGMT), which was shown to remove chloroethyl groups from the 0⁶position of guanine and to protect cells against the cytotoxic effect of CNU'S. For this purpose CkMGMT-transgenic mice, that overexpress the human MGMT protein in epidermal cells and nontransgenic controls were treated with a single topical dose of 5 m mol ACNU (Nimustines^R) followed by repeated application of the tumor promoter TPA for 28 weeks. These two-stage skin carcinogenesis experiments showed a moderate tumor response after ACNU initiation in nontransgenic mice. MGMT over-expression offered transgenic individuals the ability to repair a subset of precarcinogenic DNA lesions, as demonstrated by a 2.5 times lower papilloma incidence. Screening for oncogenic mutations in exon 1 of the Ha-ras gene of papillomas of both groups by RFLP analysis and sequencing revealed G?A transitions in codon 12 (25%) as well as base substitutions in codon 13 (20%) comprising G?T transversions and GC?TT tandem substitutions.

Keywords: Chloroethylnitrosourea, skin carcinogenesis, transgenic mice, 0⁶-methylguanine-DNA methyltransferase, Ha-ras gene

C-FOS is protectively involved in cell killing and chromosomal aberration formation induced by UV light and chemical mutagens

Simone Haas*, Holger Kappes and Bernd Kaina

Division of Applied Toxicology, Medical Faculty, University of Mainz, Obere Zahlbacher Str. 67, D-55131 Mainz

The proto-oncogene c-fos is induced by a variety of DNA damaging agents such as UV, N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) and methyl methanesulfonate (MMS). The immediate-early induction of c-fos by genotoxic agents leads to the hypothesis that c-fos induction is essential for cellular defense against genotoxic stress. We show, using c-fos knockout cells, that lack of c-Fos protein results in hypersensitivity of cells for UV-induced chromosomal aberrations and cell killing. The overall nucleotide excision repair capacity of the cells was not affected. To answer the question of whether hypersensitivity of c-fos deficient cells is a more general phenomenon or limited to UV, c-fos deficient cells were treated with ionizing radiation (IR), the methylating agents MNNG and MMS and the antineoplastic drug melphalan. For all S-Phase dependent mutagens, but not IR, c-fos deficient cells showed increased aberration frequencies and cell death, which was due to necrosis and apoptosis. The activity of various repair and detoxifying enzymes (MGMT, MPG, APendo, GST) was not altered in c-fos deficient fibroblasts compared to the corresponding wild-type cells, indicating that defective DNA repair or detoxification is not the underlying cause of c-fos hypersensitivity. Mutagen-induced DNA replication was stronger inhibited and recovery from mutagen induced block to replication was incomplete in c-fos deficient cells, compared to the wild type. We conclude that c-Fos is required for abolition of the mutagen-induced block to DNA replication and thus is involved in cell cycle checkpoint control.

Keywords: c-fos, genotoxic agents, genomic instability

Disturbance of DNA damage recognition during nucleotide excision repair by nickel(ll) and cadmium(ll)

Andrea Hartwig*, Maike Hartmann

University of Bremen, Department of Biology and Chemistry, Postfach 330440, 28334 Bremen, Germany

Nucleotide excision repair (NER) is essential for the removal of bulky DNA lesions induced by a variety of environmental mutagens. One important step of the repair process is the recognition of DNA damage, mediated by different proteins including XPA and RPA. In the present study we prepared nuclear extracts from HeLa cells and applied a gelmobility-shift-assay to (i) analyse DNA-protein interactions with damaged DNA and (ii) to investigate the effects of nickel(ll) and cadmium(ll) on this DNA damage recognition process, since in previous studies compounds of both metals had been demonstrated to interfere with nucleotide excision repair by inhibiting the incision step of the repair process. By this approach, we identified a protein which binds specifically to a UV-irradiated synthetic oligonucleotide. When investigating nuclear extracts from nickel-treated HeLa cells, the DNA damage specific protein binding was diminished in a dose-dependent manner, starting at non-cytotoxic concentrations as low as 50 pM nickel(ll). This inhibition was partly reversed by the addition of magnesium( ll to the gel-shift reaction. In the case of cadmium(ll), a dose-dependent inhibition of DNA-protein interactions was detected at 0.5 pM and higher, which was largely reversible by the addition of zinc( ll to the gel-shift buffer. Therefore, compounds of both metals disturb DNA-protein interactions essential for the initiation of nucleotide excision repair most likely by the displacement of essential metal ions.

Keywords: nucleotide excision repair, DNA damage recognition, nickel(ll), cadmium(ll), gel mobility shift assay

Assessment of induction and repair of DNA damage in mutant V79 chinese hamster cells with the comet assay

Rainer Helbig² and Gunter Speit²

1University of Ulm, Department of Medical Genetics, D-89069 Ulm, Germany, ²present address: CCR, Cytotest Cell Research GmbH & Co, KG, D-64380 Rossdorf, Germany

Using the alkaline comet assay, we studied the induction and repair of DNA damage induced by chemical mutagens in the repair-deficient Chinese hamster cell lines V-E5 and XR-V15B. XR-V15B belongs to a group of X-ray-sensitive V79 mutants, which were found to be defective in the repair of DNA double strand breaks (DSB). The ataxia telangiectasia-like mutant cell line V-E5 is characterized by a radioresistant DNA synthesis, chromosomal instability and cross-sensitivities to different radiomimetic agents. We investigated the genotoxic and cytotoxic effects of the radiomimetic mutagen neocarzinostatin (NCS) and the monofunctional alkylating agent methyl methanesulfonate (MMS). Effects in the comet assay were analyzed directly after treatment as well as after a postincubation period in mutagen-free medium to get insight into the DNA repair capacities of the mutant cell lines in relation to different primary DNA lesions. Both mutagens caused a concentration-related increase in DNA strand breakage in both cell lines and in the normal parental cell lines. Persistence of MMS-induced DNA damage after postincubation was similar in normal and mutant cell lines, indicating similar repair capacities for MMS-induced lesions. However, a decreased repair capacity was observed in XR-V15B cells after NCS treatment. XR-V15B cells only repaired about 30% of the NCS-induced DNA damage, while in the parental V79 cell line about 70% of the strand breaks were repaired. This difference seems to be due to a significant portion of DSB as main type of NCS-induced DNA damage. Compared to previously studied effects on gene mutations and chromosome aberrations, detection of NCS-induced DNA damage with the comet assay was less sensitive and only occurred under strong cytotoxic conditions. Taken together the comet assay has been shown to be a useful tool for studying induction and repair of DNA strand breaks in mutant cell lines. However, it seems to be less suited as a sensitive genotoxicity test for the detection of agents producing DSB as the main DNA damage.

Keywords: Single cell gel (SCG) assay, DNA strand breaks, genotoxicity, cytotoxicity, neocarzinostatin, methyl methanesulfonate

DNA damage repair in mammals

Jan H.J. Hoeijmakers

MGC Dept of Cell Biology & Genetics, Erasmus Univ. PO Box 1738, 3000 DR Rotterdam, The Netherlands

Nature has equipped all organisms with an intricate network of DNA repair pathways, to cope with damage induced by genotoxic agents. These pathways ensure genome stability and prevent carcinogenesis. Examples of multi-step damage repair processes are: nucleotide excision repair (NER, for removal of a wide variety of lesions, including UV) and recombination repair (for elimination of the very genotoxic radiation induced double strand breaks). The NER pathway is understood in great detail and is associated with three human syndromes characterized by marked sun sensitivity: xeroderma pigmentosum (XP), Cockayne syndrome (CS), and trichothiodystrophy (TTD). XP patients show an over 1000x increased risk of skin cancer, in contrast to CS and TTD. At least 25 proteins are involved some are also implicated in other cellular processes, explaining puzzling features associated with defects in these genes. Recombination repair is much less understood. However, recently a number of genes has been cloned based on sequence homology with yeast genes. For both DNA repair processes mouse mutants have been generated, that permit evaluation of the biological impact of these systems.

Simultaneous detection of structural and numerical chromosome aberrations in human sperm by multi-color FISH,

Paul Van Hummelen*, Xiu Lowe, Andrew J Wyrobek

Biology and Biotechnology Research Program, Lawrence Livermore National Lab., PB 808, Livermore, CA, 94550

Structural and numerical chromosome aberrations transmitted via germ cells are major contributors to pregnancy loss and congenital abnormalities. Although the parental origin of autosomal aneuploidies are predominantly maternal, the contribution of the father is substantial in sex chromosome aneuploidies and de novo structural aberrations. Therefore there is a general concern about the hazards of paternal exposure to mutagenic agents via environment, occupation or lifestyle. Recently, methods have been developed to assess chromosomal damage in sperm using fluorescence in situ hybridization (FISH). The sperm nucleus presents a special challenge for in situ hybridization because of the tight packaging of DNA which interferes with penetration by DNA probes. To detect aneuploidy, methodologies were developed that progressively increased the number of simultaneous detection of centromeric specific DNA probes in sperm. In the most recent method.) up to four probes could simultaneously be labeled in different colors. The chromosomes under investigation in our laboratory are X, Y, 16, 18, and 21 representing the majority of aneuploid abortions or offspring in humans. Since most of the mutagenic agents are chromosome breaking agents an additional method was developed to assess structural aberrations in chromosome 1p, using probes labeling both the centromere and its telomere in different colors. Baseline frequencies for healthy donors have been determined and excellent agreements were obtained with the human-sperm/hamster-egg technique for sperm cytogenetics indicating the validity of the hybridization measurements for both structural aberrations and aneuploidy detection. These assays were used to detect increases of structural and numerical aberrations in Hodgkin's disease patients undergoing chemotherapy and in cigarette smokers. We propose to apply these newly developed methods to detect and characterize effects of exposure to mutagens and to evaluate host factors that may predispose individuals to produce chromosomally defective sperm.

[Work was performed under the auspices of the U.S. DOE by the Lawrence Livermore National Laboratory under contract W-7405-ENG-48]

Activation of nitrosamines in metabolically proficient V79 cells to induce cytotoxic, genotoxic and mutagenic effects.

Christine Janzowski*, Monika Hofer, Tanja Reichert, Gerhard Eisenbrand

University, Dept. of Chemistry, Food chemistry & Environmental Toxicology, Erwin Schrodingerstr. 52, D 67663 Kaisersiautern

We investigated activation of environmentally relevant nitrosamines by rat CYP2B1 and human CYP2E1, stably expressed in transfected V79 cells. Activation of nitrosamines was measured comparing cytotoxicity (neutral red uptake, sulforhodamin B staining) and induction of mutations at the HPRT locus in transfected and parental cells (incubation: 24 h). Induction of DNA damage and formation of DNA adducts were monitored by alkaline elution and ³²P-postlabelling, respectively. Dimethylnitrosamine (DMN: 25 - 100 m M) induced strong cytotoxic effects cells, mutations at the HPRT locus, 0⁶-methyidG and DNA single strand breaks in CYP 2E1 cells. CYP 2E1 activation was also observed with nitrosomorpholine, nitrosopyrrolidine and with diethanolnitrosamine. Dibutylnitrosamine (DBN) induced cytotoxic, mutagenic and genotoxic effects and caused 0⁶- butyldG formation in r2B1 cells. The longer chain analogue dioctyinitrosamine, however, was not found to be a substrate for CYP2B1; cytotoxicity was equally observed in parental and r2B1 cells at concentration > 100m M. In conclusion, isoenzyme-specific activation of nitrosamines is strongly determined by structural characteristics.

The gift of transfected V79 cells by Drs. W.A. Schmalix, J. Doehmer is gratefully acknowledged.

Keywords: nitrosamines, V79 cells, CYP2B1, CYP2E1, hepatocytes, HPRT, DNA damage, 0⁶-aikyldG adducts

Induction of functions involved in the repair of DNA alkylation damage - cause of adaptive response

Bernd Kaina*, Thomas Grombacher, Sabine Kiesel, Gerhard Fritz and Kirsten Ochs

Division of Applied Toxicology, Medical Faculty, University of Mainz, Obere Zahlbacher Str. 67, D-55131 Mainz

DNA alkylation damage is repaired by 0⁶-methylguanine-DNA methyltransferase (MGMT) and base excision repair enzymes. MGMT removes alkylations from the 0⁶-position of guanine and thus protects cells from the genotoxic effects of alkylating agents. N-methylation lesions are repaired by base excision accomplished by N-methylpurine-DNA glycosylase (MPG), apurinic endonuclease (APR), polymerase b (pol b) and DNA ligase. Down-modulation of various of these enzymes by antisense or gene targeting techniques sensitized the affected cells, whereas overexpression of yeast APE protected cells against the toxic and clastogenic effect of various alkylating agents. From this we conclude that both MGMT and the base excision repair pathway is important in cellular defense against alkylating drugs. In order to elucidate whether the adaptive response of cells to alkylating agents is related to elevation of repair activity, we studied the inducibility of alkylation repair genes. It is shown that MGMT is inducible,, in rat liver cells, upon DNA damaging treatments on the level of MRNA expression and promoter activity which results in increased cellular repair capacity. The induction of MGMT gave rise to reduced gene mutation frequencies indicating the existence of an adaptive response in mammalian cells which is due to stimulated expression of repair gene(s). Induction of MGMT also occurred upon glucocorticoid treatment. It was not accompanied by significant elevated expression of base excision repair genes indicating that, in mammalian cells, MGMT and base excision repair genes are not under coordinate control.

Keywords: DNA repair, alkylating agents, adaptive response

The micronucleus test in vitro with V79 cells comparison of three different methodologies

S. Kalweit¹*, D. Utesch², S. Ertelt^1, B. Fissler¹, A. Grun², B. Hehl², P.J. Kramer², S. Madle¹

1Federal Institute for Health Protection of Consumers and Veterinary Medicine (BgW), Thielallee 88-92, D-14195 Berlin, Germany ²Merck KGaA, Institute of Toxicology, D-64271 Darmstadt, Germany

The micronucleus test (MNT) in vitro is a useful assay for the detection of mutagenic events on both the chromosomal and genomic level. In this investigation three different methodological approaches of the MNT in vitro were compared: The standard assay using an asynchronically growing mixed cell population, the cytokinesis block (CB) assay and a modified MNT, the so called mitotic shake off (MSO) method. V79 cells were thus treated for various time periods with two known aneugens (colcemide and griseofulvin) and two clastogens (mitomycin C and cyclophosphamide). The cultures of the CB assay were additionally exposed to cytochalasin B which inhibits cell but not nucleus division. After treatment the cells were harvested and analyzed for the appearance of micronuclei (MN). All three assays yielded positive results for both, the aneugens and clastogens tested. The LOEDs for all substances in all assays were very similar. However inconsistent results were evident in the CB assay in so far as an increase in MN occurred not only in the binucleated but also in the mononucleated cell fraction. Since the latter represents cells that have not undergone mitosis, a prerequisite for the establishment of mutagenic effects, the increase in MN in the mononucleated cells cannot be explained on a mechanistic basis. Therefore, the CB assay seems to be inappropriate, especially for the testing of aneugens, when both the test compound and cytochalasin B influence microtubuli assembly.

Keywords: Micronucleus test, V79 cells, Cytochalasin B (CB)

Coumarin, a rodent liver carcinogen, shows no genotoxic activity in rat and mouse hepatocytes in vitro

Peter Kasper*, Karin Gindler, Kerstin Tegethoff, Birgit Kersten, Lutz Muller

Bundesinstitut fur Arzneimittel und Medizinprodukte, Seestr. 10, 13353, Berlin , Deutschland

The use of coumarin as a flavouring substance in food was prohibited by the FDA in 1954 following reports of hepatotoxic effects in rats and dogs. Within the EU, coumarin itself is not approved as a food additive. However, it is a constituent of natural flavouring source materials widely used in a variety of foods and alcoholic beverages. The general limit in such materials is currently 2 mg coumarin/kg. Coumarin is also used as a pharmaceutical in the medical treatment of lymphoedema. Beside its hepatotoxic effects in rats and dogs coumarin has been shown in long-term studies to induce adenoma and carcinoma of the liver and bile duct in rats and adenomas and carcinomas of the lung and liver adenomas in mice. Several investigations into the metabolism of coumarin suggest that a cytochrome-P450 generated metabolise, probably the 3,4-epoxide intermediate in the oxidation of coumarin to 3-hydroxycoumarin, may be involved in the tumor formation via a genotoxic mechanism of action. However, available genotoxicity data are limited and especially liver/hepatocyte-specific studies are missing. In the present study the genotoxicity of coumarin was investigated using the UDS DNA repair test and the micronucleus assay with primary rodent hepatocytes. In hepatocytes from male Wistar rats and from male and female NMRI or B6C3F1 mice, coumarin did not induce DNA repair activity. In agreement with this result, no clastogenic effects were observed in the rat hepatocyte micronucleus assay. However, a concentration-dependent increase in mitotic cells may indicate a hepatomitogenic potential of coumarin. In summary, the negative results with coumarin in two hepatocyte-based genotoxicity assays do not suggest that the liver carcinogenicity of coumarin is due to a genotoxic mechanism of action.

Keywords: coumarin, rodent liver carcinogen, hepatocyte UDS test, hepatocyte micronucleus assay

Genotoxicity of naturally occurring carcinogens in the micronucleus test with human lymphocytes and in the sos-chromotest with E. coli pq37

S. Kevekordes, A. Knievel, J. Spielberger, K. Kleinschmidt, Ch. Burghaus, H. Dunkelberg

Medical Institute of General Hygiene and Environmental Health, University of Gottingen, Windausweg 2, D-37073 Gottingen, FRG

In the present investigation, six naturally occurring carcinogens were tested for their genotoxicity in the cytokinesis-block micronucleus test with human lymphocytes and in the SOS-chromotest, a modified laboratory protocol of the Escherichia coli PQ37 genotoxicity assay. Both tests were employed in the presence and/or absence of an erogenous metabolizing system from rat liver S9-mix. In the presence of this system, b -asarone caused a significantly elevated genotoxicity in the micronucleus test (MN_CB27; NDI 1.42) and the SOS-chromotest (If_max x 6.3). In its absence, monocrotaline (If_max 1.8) and b -asarone (If_max 1.6) were classified as marginal genotoxic in the SOS-chromotest. And in the presence and absence of the metabolizing system aristolochic acid caused a significantly elevated genotoxicity in the micronucleus test and the SOS-chromotest; [in the presence (MN_CB33; NDI 1.49 - lf_max 9.8) and absence (MN_CB 29; NDI 1.51 lf_max 20.8) of rat liver S9]; emodine, reserpine and safrole did not show genotoxic effects in the micronucleus test the SOS-chromotest; and moncrotaline was not genotoxic in the micronucleus test. (MN_CB=mean number of micronuclei in binucleate cells of two or three independent experiments; NDI=nuclear division index; lf_max=maximum induction factor at highest non-cytotoxic dose).

Keywords: naturally occurring carcinogens, genotoxicity, SOS-chromotest, micronucleus test, human lymphocytes, S9-mix

Study of clastogenic and aneugenic effects of synthetic polycyclic musk flavors in the micronucleus test with human lymphocytes in vitro and the human hepatoma cell line hep g2

S. Kevekordes, M. Diez, K. Kleinschmidt, R. Suchenwirth, T. Gebel, H. Dunkelberg

Medical Institute of General Hygiene and Environmental Health, University of Gottingen, Windausweg 2, D-37073 Gottingen, FRG

The synthetic polycyclic musk flavor compounds comprising 1,3,4,6,7,8-hexahydro-4,6,6,7,8,8-hexa-methyl-cyclopenta-(g)-2-benzopyrane (Galaxolide), 7-acetyl-1,1,3,4,4,6-hexamerthyltetraline (Tonalide), 4-acetyl-1,1-dimethyl-6-tert.butylindane (Celestolide), 6-acetyl-1,1,2,3,3,5-hexamethylindane (Phantolide), 6,7-dihydro-1,1, 2,3,3-pentamethyl-4-(5H) indanone (Cashmeran) and 5-acetyl-1,1, 2,6-tetramethyl-3-isopropylindane (Traseolide) are used as fragrance ingredients in perfumes, lotions and detergents. These also were detected in freshwater fish, human milk and fat samples. Due to the high concentrations reported in the literature it is vital to investigate the bioaccumulation of this class of compounds as well as their genotoxicity. The aim of the study was to evaluate the genotoxic activity of synthetic polycyclic musk flavor compounds in the micronucleus test with human lymphocytes in vitro and the human hepatoma cell line Hep G2. The cell line Hep G2 was originally established from human liver tumour biopsy. The morphological characteristics and epithelial cell shape are compatible with those of liver parenchymal cells. The polycyclic musk flavors tested for their genotoxic activity were employed up to cytotoxic doses and indicated by the NDI (nuclear division index). The musk flavor compounds revealed no significant increase in the frequency of micronuclei in human lymphocytes nor in the human hepatoma cell line Hep G2.

Keywords: synthetic polycyclic musk flavors, micronucleus test,human lymphocytes, Hep G2, S9-mix

DNA repair within genes in drug-sensitive testis tumour cell lines

Beate Koeberle*, John A Hartley, John RW Masters

University College London, Institute of Urology and Nephrology, and Department of Oncology, 67 Riding House Street, London WIP 7PN, UK

Metastatic cancer is normally incurable. However, over 80% of the patients with metastatic testis tumours can be cured using cisplatin-based combination therapy. Continuous cell lines derived from human testis tumours retain hypersensitivity to DNA damaging agents in vitro thus providing a model system to investigate the molecular basis of this drug sensitivity. Previous investigations measuring DNA repair in the whole genome indicated that the hypersensitivity of testis tumour cell lines may be related to a low capacity for DNA repair. Using quantitative PCR (QPCR) we measured repair within the actively transcribed N-ras and the non-transcribed CD3 gene in 6 continuous cell lines derived from testis (cisplatin-sensitive) and bladder (cisplatin-resistant) tumours. The amount of initial binding of cisplatin adducts was similar for both genes in the two cell types. In agreement with the data for the whole genome we found that all bladder tumour cell lines were repair proficient whereas 2 out of 3 testis tumour cell lines showed little capacity for repair within specific regions of both genes. These findings indicate that the sensitivity of testis tumour cell lines to DNA damaging agents is in part due to a reduced capacity to repair induced DNA damage.

Keywords: Drug sensitivity, gene-specific DNA repair

Role and significance of oncogenetic alterations in skin cancer:the p53 paradigm

Coen F. van Kreijl

RIVM, Lab of Health Effects Research, Dpt Carcinogenesis, Mutagenesis and Genetics, P.O. Box 1, 3720 BA Bilthoven, The Netherlands

Non-melanoma skin cancer (NMSC) is the most common type of human cancers. It comprises basal cell carcinomas (BCC) and squamous cell carcinomas (SCC), which mainly develop from the typical precursor lesion actinic keratosis (AK). A causative role for solar UV radiation, implicated by epidemiological studies, has been further substantiated by recent molecular analysis. Frequent and also early alterations observed in the p53 gene serve as paradigm in this respect. Basically two types of animal models have been used to study human NMSC. Experimental chemical carcinogenesis on mouse skin with a variety of chemicals (initiators, promoters) has been the general model for human (epithelial) squamous cancer for a number of decades. UVB radiation-induced skin cancer in the mouse, especially the albino hairless mice (SKH-HR) is the best studied other type of rodent model for human NMSC. However, major differences between the pathogenesis and oncogenetic events seem to exist between the chemically-induced and UVB-induced murine SCC. These concern not only the observed precursor lesions (respectively papillomas and AK) but also the significance of certain oncogenetic alterations. An overview of these, with emphasis on the significance of p53 and ras gene mutations will be presented. Using transgenic mouse strains, deficient in either DNA repair or p53 function, we have further investigated the significance of p53 alterations for UV-induced skin carcinogenesis. The results obtained led to new and unexpected insights. These concern the process of UV-induced skin cancer, especially the role of precursor lesions, the significance of p53 mutations as early (genetic) events, as well as their dependence on UV dose (amount of DNA damage). Finally, our results may also help to bridge the gap between chemically-induced and UVB-induced murine skin cancer.

Keywords: human skin cancer, mouse models, chemical carcinogens, UVB, p53, ras genes, nucleotide excision repair.

Fluorescence in situ hybridisation (FISH) with chromosome-specific centromeric probes: a sensitive method to detect aneuploidy

Richard R Marshall¹, Morna Murphy¹, David J Kirkland¹* and Karin S Bentley²

1Covance Laboratories, Harrogate, UK ²E I Dupont de Numours & Co, Newark, USA

Cytochalasin B-blocked binucleate human lymphocytes from female donors have been used to measure micronucleus induction and other aneuploidy events after treatment with colchicine, vinblastine or carbendazim. For the aneuploidy events, centromeric probes for 6 selected chromosomes (1, 8, X, 11, 17, 18) were used to measure chromosome loss, addition and non-disjunction in the interphase nuclei of these binucleate cells. The chromosomes were probed in pairs using Cy-3 (red) and FITC (green) labels for the 2 different centromeric regions. For colchicine, the total non-disjunction frequencies for chromosomes 1 and 8 were similar to the total micronucleus frequencies, but were detected as significant at lower concentrations. For vinblastine (chromosomes 1 and 8) and carbendazim (all 6 chromosomes) the frequencies of non-disjunction far exceeded (7 and >2-fold respectively) the peak frequencies of micronucleus induction. Although most chromosomes exhibited similar sensitivity in all the aneuploidy events measured, there was an indication that chromosome X was more than susceptible to non-disjunction that the other chromosomes. We believe that measurement of non-disjunction in binucleate human lymphocytes using chromosome specific centromeric probes offers a sensitive method for detection of aneuploidy and is particularly appropriate for the establishment of thresholds.

Keywords: aneuploidy, centromeres, probes, colchicine, vinblastine, carbendazim

High spontaneous mutation frequency at a transgenic Mouse locus

Hans-Jorg Martus¹*, Michael E.T.I. Boerrigter², Martijn E.T. Dolle² and Jan Vijg²

1Sandoz Pharma Ltd., Drug Safety/Toxicology, CH-4002 Basel, Switzerland ²Beth Israel Hospital, Gerontology Division, Department of Medicine, and Harvard Medical School, Boston, MA 02215, USA

Recently, a new transgenic mouse system, based on the stable integration of the lacZ- containing plasmid pUR288 in the genome of C57B1/6J mice, was introduced. Mutation analysis utilizes magnetic beads with an immobilized regressor protein for separation of the transgene from total genomic DNA, followed by electrotransformation of suitable host bacteria for mutant scoring and analysis. Three homozygous transgenic mouse lines were generated, designated lines 61 30 and 60. Experiments with line 60 have demonstrated that this novel system is a sensitive tool for studying mutation induction by various mutagens, including clastogens. In the organs studied so far, two lines, 30 and 60, had background frequencies comparable to other transgenic, e.g. bacteriophage lambda based, systems. However, in line 6, an enormous elevation in the spontaneous mutant frequency, up to 1 00-fold of the other lines, was found in lung, liver, spleen and brain. In addition, restriction analyses of 288 mutants, isolated from lung and brain tissues of line 6, revealed that they almost exclusively contained large size changes. About 30-40% of them extended into the mouse genome, as demonstrated by southern blot hybridization with a nontransgenic total genomic mouse DNA probe. To our knowledge this is the third report of such an elevation in the spontaneous mutant frequency at a transgenic shuttle vector locus, and the highest values found so far. Based on circumstantial evidence, we propose that, in line 6, the transgene integration occurred at a highly unstable genomic position, or that the transgene integration itself generated such an unstable site. These experiments demonstrate that, besides an application in genetic toxicology, transgenic shuttle vectors represent valuable tools for probing the integrity of the mammalian genome.

Keywords: transgenic mice, shuttle vector, plasmid, mutation, deletion, genomic instability

Characterization of SV40-transformed xeroderma pigmentosum cell lines for HPRT mutagenicity studies

Oliver Merk and Gunter Speit

Universitat Ulm, Abteilung Medizinische Genetik, D-89069 Ulm

Six SV40-transformed Xeroderma pigmentosum (XP) cell lines were characterized for their suitability for the HPRT mutagenicity test. Three of them were from the complementation group A (XP 2 OS, XP 12 BE, XP 12RO-SV), two were from the complementation group C (XP 4 PA-SV, XP 8 CAC-SV) and one was from the rare complementation group G (XP 3 BR/12SV). Analysis of the chromosome frequency distribution revealed that the karyotypes of all cell lines were hyperdiploid and heteronuclear. Using fluorescence-in-situ-hybridization (FISH) techniques, we were able to demonstrate that most of the cell lines contained additional X chromosomes (and consequently additional HPRT alleles) in the majority of cells. Only one male cell line (XP 8 CAC-SV) contained only one X chromosome in 75% of the cells. A subclone of this cell line was established (XP 8 CAC-SV-C1), which showed a near diploid karyotype and had only one X chromosome in 94% of the cells. This subclone was shown to have the same typical UV-hypersensitivity as the parental XP cell line. Multiplex-PCR and sequence analysis revealed no change in the HPRT gene structure and the wild-type DNA sequence. Consequently this cell line should exhibit a normal HPRT activity and should be suited for the analysis of mutant frequencies and mutational spectra of the HPRT gene. However, the cell line had a low plating efficiency (<5%), which could not be increased significantly by using different culture conditions. In two independent experiments with 5-bromdesoxyuridine as mutagen we found a concentration-related increase in the mutant frequencies, but a high variability between the experiments. The calculated mutant frequencies in both experiments were mainly influenced by the variable decrease in the plating efficiencies and less by an increase in the absolute number of mutants. Taken together, the results demonstrate the necessity for a cytogenetic and molecular characterization of transformed cell lines. Furthermore, it is shown that the usability of transformed human cells for HPRT mutagenicity studies depends on both the genetic constitution and growth characteristics of the cell lines.

Keywords: Xeroderma pigmentosum, HPRT, FISH,5-bromdesoxyuridine, mutant frequency

High spontaneous mutation frequency at a transgenic Mouse locus

Hans-Jorg Martus¹*, Michael E.T.I. Boerrigter², Martijn E.T. Dolle² and Jan Vijg²

1Sandoz Pharma Ltd., Drug Safety/Toxicology, CH-4002 Basel, Switzerland ²Beth Israel Hospital, Gerontology Division, Department of Medicine, and Harvard Medical School, Boston, MA 02215, USA

Keywords: transgenic mice, shuttle vector, plasmid, mutation, deletion, genomic instability

Different protein-binding affinities of naturally occurring anthraquinones modulate their genotoxicity**

Stefan 0. Muller*, Werner K. Lutz, Helga Stopper

Department of Toxicology, University of Wurzburg, Versbacher Str. 9, D-97078 Wurzburg, Germany

We have reported that the plant-derived 1,8-dihydroxyanthraquinone derivatives, emodin and danthron, were clearly genotoxic in mouse lymphoma L5178Y cells, using the micronucleus, tk^+/- mutation and comet assay (Muller et al., Mutation Research, in press). In these cell culture systems danthron was more effective than emodin. When we investigated non-covalent DNA binding measured by the reduction of Hoechst 33342 DNA-bound fluorescence with isolated DNA and in the living cell emodin was more potent. This inversion of the ranking prompted us to investigate the underlying mechanism. Since emodin shows a high serum-protein binding affinity, horse serum used as a media-supplement in the mouse lymphoma assays was analyzed for a potential selective scavenging of emodin. Therefore, we tested noncovalent DNA binding in mouse lymphoma L5178Y cells in absence or presence of serum. In the presence of 10% serum, the DNA binding of emodin was markedly reduced and lower than that of danthron. To determine the role of serum-protein in a mouse-lymphoma-cell genotoxicity assay, we modified the tk^+/- mutation assay and used only 0.5% horse serum in the medium, i.e., a concentration at which no reduction of non-covalent DNA binding of emodin relative to danthron was observed. Now, emodin was more mutagenic than danthron. It is concluded that the genotoxicity of emodin is strongly modulated by the cell culture medium supplement horse serum. Moreover, for the assessment of a particular compounds' genotoxicity in cell culture systems, interactions with cell culture supplements can play a critical role in potency evaluations and comparison.

Keywords: Genotoxicity, mouse lymphoma L5178Y cells, protein binding, non-covalent DNA binding, anthraquinones, potency.

**This work was supported by the Swiss Federal Office of Public Health (BAG grant No. FE 316.95.0500).

Die bedeutung von chromosomenaberrationen und chromo-someninstabilitat beim menschen

Hansjakob Muller, Universitatskinderklinik Basel, Romergasse 8, 4005 Basel

Seit Beginn der 60er Jahre sind Chromosomenanalysen ein fester Bestandteil der kiinischen Diagnostik. Je nach Lebensphase haben zytogenetische Anomalien im Hinblick auf Morbiditat und Mortalitat eine beachtliche Bedeutung. In den ersten Schwangerschaftssmonaten sind Aneuplodien verschiedener Chromosomen die h;aufigste Ursache eines Spontanabortes. Auch bei einem grossen Anteil von Totgeburten, resp. Neugeborenen ohne Ueberlebenschance liegen Chromosomenaberrationen, wie die Trisomien 18 oder 21, vor. Etwa eines von 200/250 Neugeborenen ist Trager einer nicht-balancierten Aberrationen mit klinischen Konsequenzen. lhretwegen resultieren schwere geistige Retardierung, Dysmorphien und/oder Fehibildungen. Wegen ausbleibender oder gestorter Pubertat wird man auf das Vorliegen von Aberrationen der Geschiechtschromosomen aufmerksam. Fertilitatsprobleme oder wiederholte Aborte konnen die Folge einer Chromosomenaberration bei einem Partner sein. Zu einem Durchbruch in der Diagnostik von Chromosomenaberrationen kam es wahrend der letzten Jahre dank der Fiuoreszenz-in situ-Hybridiserung (FISH)- und weiteren Techniken, die den Nachweis von Chromosomenaberrationen wahrend der lnterphase und vor allem von lichtmikroskopisch bisher kaum entdeckbaren Deletionen kieinster Chromosomenabschnitte ermoglichen. Dies fohrte zur Entdeckung einer neuen Gruppe von Erbkrankheiten, namiich den "contiguous gene syndromes". Die Mikrodeletionen haben auch zu einem besseren Verstandnis verschiedener genetischer Phanomenen wie des "genomic imprinting" gefuhrt. Einige monogene Erbkrankheiten (Ataxia telangiectasia oder die Fanconi Anaemie) sind durch eine erhohte Bruchigkeit der Chromosomen, resp. durch eine erhohte Rate von Schwesternchromatid-Austauschen (Bloom-Syndrom) charakterisiert. Bestimmte Malignome treten typischerweise damit assoziiert auf. Zudem konnen bei einzeinen Personen umschriebene fragile Stellen vorkommen. Mit wenigen Ausnahmen (fragiles X-Syndrom) sind Trager von solchen umschriebenen bruchigen Stellen phanotypisch unauffallig. Chromosomenanomalien in der Keimbahn konnen neu entstanden sein und gelegentlich weitervererbt werden. Sie entstehen aber auch nur in Korperzeilen. So hat auch die zytogenetische Untersuchung von entarteten Zeiien kiinische Bedeutung. Moiekuiargenetische Marker erlauben die Feststellung der Herkunft der Chromosomen und eroffnen Einblicke in die Mechanismen ihrer Entstehung.

Photogenotoxicity testing as a predictor of photocarcinogenicity

Lutz Muller, Peter Kasper

Federal Institute for Drugs and Medical Devices, Seestr. 10, D-13353 Berlin (Germany)

The tumorogenic effects of UV-radiation (UVR) are well known. Specifically, the premutagenic lesions of UVB (290-320 nm), i.e. pyrimidine-dimers and 6-4 photoproducts, are known to be the most important molecular events in UVR tumorigenicity, In contrast, the less carcinogenic UVA (320-400 nm) mainly generates oxidative damage in the DNA via photodynamic generation of active oxygen species involving endogenous or erogenous photosensitizers. This oxidative damage leads mainly to thymidine glycol and 8-oxo-2'-deoxyguanosine, Mammalian cells possess effective repair mechanisms for oxidative damage, photoproducts and dimers. However, these repair mechanisms may be overloaded. Especially, the repair capacity to remove oxidative DNA damage can be overstressed by exposure to photosensitizers. This may lead to gene mutations or chromosomal damage in mammalian cells and eventually tumor induction in the skin. In fact, it was shown for a number of photoinstable compounds such as phenotiazides, fucocumarines (e.g. 8-methoxypsoralene) and fluoroquinolones that they are very efficient inducers of chromosomal damage in mammalian cells in culture, Photocarcinogenicity testing in hairless mice has clearly shown tumor induction in the skin for furocoumarines and several fluoroquinolones. The data base on other phototoxic and potentially photomutagenic compounds is quite limited up to now. However, the data allow the hypothesis that photogenotoxicity testing in mammalian cells in vitro may be a good predictor of photocarcinogenic potential. Whereas the damage to the DNA caused by UVR+photosensitizers may not be the only causative factor of photocarcinogenicity, it is probably an important one and a factor that can easily be assessed. Animal models of photocarcinogenicity testing in hairless mice or XPA (repair) deficient mice are costly, artificial, of doubtful predictivity and cruel to animals. Alternatively, an integrated approach using photostability data, phototoxicity testing (in vitro and in vivo) and photomutagenicity test systems may give the same level of safety information for a pharmaceutical or cosmetic product when assessing their effects in the preserve of UVR.

Analysis of micronuclei by flow cytometry and FISH

Michael Nusse,* Carmela Fimognari**, Silvia Viaggi***, Sabine Sauer-Nehls*, Herbert Braselmann+, Jan Grawe'++

*GSF-Flow Cytometry Group, 85764 Neuherberg, Germany, **University of Bologna, Dept. of Pharmacology, Bologna, Italy, ***IST, Lab. Mutagen. Genova, Italy, +GSF-Institute for Radiation Biology, ++Dept. of Genetcs, Uppsala University, Uppsala, Sweden

DNA content distributions of micronuclei (MN) induced in cells by ionizing radiation and several chemicals were measured using flow cytometry. MN with increasing DNA content were sorted and analyzed for the presence of centromeric signals using a centromeric a -satellite probe. Radiation-induced MN were found to be produced mainly by chromosome fragments, whereas MN induced by the tear gas chlorobenzylidene malonitrile (CS) were found to be produced mainly by whole chromatids. In contrast, MN induced by vinblastine (VBL) were produced mainly by whole chromosomes and by combinations of two or more whole chromosomes. Computerized random breakage of chromosomes and random combination of chromosome fragments, whole chromatids and whole chromosomes was used to simulate successfully the measured DNA content distributions of MN. Whole chromosome painting probes (chromosome 1, 7, 11, 14, 17 and 21) in combination with a human pancentromeric a -satellite probe were used to analyze the presence of specific chromosomal material in radiation- induced MN. The purpose was to study if the fraction of paint-positive MN is proportional to the relative DNA content of the respective chromosomes which might indicate a random breakage of chromosomes. Flow sorted MN and MN in binucleated cells were analyzed. The fraction of paint-positive MN was found to increase linearly with the DNA content of the chromosomes. About 13% radiation-induced MN were found to contain centromeric signals independent of the presence of specific chromosome painting signals. The data obtained on flow sorted MN and MN in binucleated cells agreed well indicating that flow sorted MN can be used for studying their chromosomal content with the FISH technique. If it is assumed that the chromosomal content of MN reflects radiation-induced damage, then these results support a random model of radiation-induced cytogenetic damage in human lymphocytes for the six chromosomes studied.

The DNA-crosslinking potency of chemotherapeutics investigated by the alkaline comet assay in vitro and in vivo

S. Pfuhler*, S. Deutschenbaur and H.U. Wolf

Universitat Uim, Abt. Pharmakologie und Toxikologie, Albert-Einstein-Allee 11, D 89069 Ulm , Germany

The drugs cisplatin, carboplatin, mitomycin c and mechiorethamine are frequently used in chemotherapy for the treatment of cancer. All of them induce DNA crosslinks, and these crosslinks are known to be the main reason for their cytostatic effect. Recently we have shown that the alkaline comet assay is able to detect crosslinking agents very fast and easily (1). By irradiating cells with gamma-radiation prior to electrophoresis, DNA strand breaks are induced to get a standardized amount of DNA fragments. The migration of these DNA fragments on the agarose-coated slides is reduced by DNA-DNA and/or DNA-protein crosslinks. This methodology was used to investigate the DNA-crosslinking potency of four chemotherapeutics in vitro. Isolated human lymphocytes were treated with the drugs for two hours, and, after embedding the cells in agarose, the slides were irradiated with 3 Grey to set the standard damage. All drugs were tested in three independent experiments and we found a very strong and dose-related reduction of radiation-induced DNA migration in the comet assay. The lowest doses which were able to significantly reduce DNA-migration were 33 m mol/L for cisplatin, 100 m mol/L for carboplatin, 50 m mol/L for mitomycin C and 3 m mol/L for mechiorethamine. To examine whether cisplatin-induced crosslinks could be detected in vivo, blood was taken from 11 cancer patients before and 20 hours after treatment with cisplatin (mean dose: 85 mg/patient), embedded in agarose and irradiated. Additional slides were set up without irradiating the cells. We found a small but statistically not significant reduction of the mean tail moment after the treatment with cisplatin for the irradiated (29.37 vs. 28.10) as well as for the non-irradiated (6.60 vs. 5.68) cells. On the individual level. four r)atients showed significant lower tail moment values after cisplatin treatment - three of them without irradiation, one after irradiating the cells.

(1) Pfuhler and Wolf, Environ. Mol. Mutagen. 27:196-201 (1996)

Keywords: Comet Assay, DNA-Crosslinks, Chemotherapy

Habitual long distance running does not enhance urinary excretion of 8-hydroxydeoxyguanosine

Alexander Pilger¹, Dietmar Germadnik¹, Dieter Formanek², Hartmut Zwick³Hugo W. Rudiger¹

1Department of Occupational Medicine, University of Vienna, Wahringer Gurtel 18-20, 1090 Vienna, Austria; ²Sanatorium Liebhartstal, Vienna, Austria; ³Department of Pulmology, Hospital Lainz, Vienna, Austria

The energy demand during physical exercise causes an increased oxygen uptake and supply to active tissues, which may increase the rate of free oxygen radical production and thereby affect the capacity of endogenous cellular defense systems. This could result in DNA base modifications among which 8-Hydroxydeoxyguanosine (80HdG) is one of the most important and has widely been used as a marker for in vivo oxidative lesions. Therefore we examined the effect of regular running exercise on the urinary levels of 80HdG in 32 long distance runners and in a group of untrained healthy subjects. The range of 8OHdG in urine was 0.12-6.45 m mol/mol creatinine in both groups and no significant difference in the mean excretion levels between runners and control probands was observed. This gives no reason to believe that physical exercise in trained individuals may induce a disturbance of the oxidant-to-antioxidant-balance.

Keywords: Exercise, 8-Hydroxydeoxyguanosine, Oxidative DNA damage

Induction of genotoxic and mutagenic effects by laser pyrolysis products

Ulla Plappert*, Bernd Stocker, Theodor M. Fliedner

Department of Clinical Physiology, Occupational and Social Medicine, University of Ulm, Germany.

The use of lasers in medical applications has grown enormously in the last few years. There is evidence that aerosols generated by pyrolytic decomposition of tissue could be health hazards. Thus, much attention is paid to the physicochemical analysis of laser plume. However, assessing potential health hazards to the operating team and the patients requires a comprehensive insight into the cyto- and genotoxic capacity of laser pyrolysis products. Therefore we analysed the cyto- and genotoxic effect of laser pyrolysis products form different types of porcine tissue. The generated aerosols were sampled on glass fiber filters. Then human peripheral blood and V79 Chinese hamster cells were incubated with the eluated pyrolysis products. Afterwards, these exposed cells were subjected to the comet assay or the hprt gene mutation test. In addition to the mutant frequencies the hprt mutation spectra were determined by PCR analysis. The ability to induce cytotoxic and genotoxic effects turns out to be strongly dependent on the type of tissue that has been irradiated during laser treatment. Genotoxic properties of the aerosols analysed with the comet assay can be aligned as follows: adipose tissue < skin < striated muscle << liver. Whereas mutagenic effects in the hprt test can be aligned as follows: skin, adipose tissue < striated muscle << liver. In conclusion our experiments showed that the laser pyrolysis products originating from porcine tissues induced very potent cytotoxic and genotoxic effects as well as mutations and therefore they are potential health hazards for humans.

Keywords: laser pyrolysis products, comet assay, hprt gene mutation test

Genotoxicity-test of unconcentrated surface waters - luminometric umu-test -

Georg Reifferscheid*, Alexandra Becker, Petra Waldmann, Rudolf-Karl Zahn

AMMUG, Universitat Mainz, Obere Zahlbacher Str. 63,55101 Mainz, Germany

The umu-test, the first german standardized method for the detection of genotoxicity of waste water (DIN 38415-3), is based on the ability of DNAdamaging agents to induce the expression of the umu-operon in bacteria. In connection with other damage inducible genes the umu-operon is essentially involved in bacterial mutagenesis via the so-called SOS-pathway. All strains used carry a plasmid which bears the umu-operon fused to lacZ. The induction of the mutator gene umuC by DNA-damaging agents is detected by measuring intracellular b -galactosidase levels colorimetrically. In the course of a BMBF-sponsored cooperative project we developed a Iuminometric umu-test version which allows incubation of test samples using very low amounts of test bacteria and increased sample sizes in order to enhance sensitivity of the test system and to avoid preconcentrations of surface water samples. Luminometric detection is possible by using dioxetane-galactosides as substrate for b -galactosidase. The luminescent product of the enzyme reaction can be accumulated over a period of time and flashes light by enhanced stimulation shifting pH to 12. Studies on standard reference genotoxins (NQO, Nitrofurantoin, Acetyl-aminofluorene + S9, Aminoanthracene + S9) indicate considerate sensitivity enhancement compared to the DIN-version of the test. Depending on the used inoculum size, detection limits of the mutagens were 0,2 - 2 m g/L for NQO (10 m g/l in the DIN-version), 7 m g/L for NF (20), 10 m g/L for AAF (22) and 0,4 m g/L for AA (18). At these concentrations either no or only weak cytotoxicity is observed. These results indicate that the luminometric umu-test is highly sensitive and should be appropriate to measure genotoxicity of unconcentrated surface waters. As well as the DIN-test the luminometric version shows excellent reproducibility of its results. Genotoxicity and cytotoxicity are measured simultaneously in microplates at moderate costs and low S9 consumption.

The authors wish to thank the BMBF for supporting this study (grant No. 02 - WU9562/0)

Key-words: umu-test, luminometric detection, genotoxicity, surface water

Measurement of micronuclei with flow cytometric techniques

Jens Rickert, Herbert G. Miltenburger

Technische Hochschule Darmstadt, Fachbereich Biologie, Schnittspahnstr. 3, D-64287 Darmstadt, Deutschland

Flow cytometric detection of micronuclei from eukaryotic cells is an established method since long (Nusse et al. 1984, Cytometry 5, 2025). However, so far the method was not used routinely for genotoxicity studies because the expensive equipment was not available in most laboratories. We have tried to establish a method with a cheaper equipment. A problem while identifying micronuclei by flow cytometry in a suspension of macronuclei and micronuclei is debris particles that may give the same fluorescence signals as micronuclei. Up to now the methods for discrimination of micronuclei and debris by size are based on two parameter measurements of fluorescence and scatter signals applying laser excitation. This technique requires a relatively expensive equipment. In comparison to this we used a Partec PAS 11 cytometer with an inexpensive HBO-100 mercury high pressure lamp. Although it is not possible to quantify scatter signals with this UV-source we could very well discriminate micronuclei and debris with a DNA/protein double staining (DAPI/SR 101). Furthermore, it is possible to remove most of the debris by passing the suspension through a syringe filter before flow cytometry is performed. Both methods make it possible to reliably determine micronuclei frequencies. In our experiments with monolayer cells of the Chinese hamster cell fine V79 and the rainbow trout cell line RTG-2 induction of micronuclei by chemicals from several classes could be quantified. As RTG-2 cell line is a candidate for routine cytotoxicity studies with waste water it is of interest to evaluate its potential for genotoxicity testing. Using genotoxic chemicals we could demonstrate the induction of micronuclei in RTG-2 cells. As compared to the induction of micronuclei in V79 cells the frequency was lower at the same concentrations of the chemicals applied and at the same preparation intervals. A likely explanation is the quite long population doubling interval of about 50 hours (V79 cells: 12-14 hours).

Keywords: Micronuclei, flow cytometry, DAPI/SR101, V79, RTG-2

Detection of apoptotic cells by the comet assay

Jens Rickert, Herbert G. Miltenburger

Technische Hochschule Darmstadt, Fachbereich Biologie, Schnittspahnstr. 3, D-64287 Darmstadt, Deutschland

The comet assay is a fast and sensitive indicator test for the detection of DNA strand breaks in eukaryotic cells. It is increasingly applied as test on the genotoxic potential of chemicals. As demonstrated by our data it can be used for the detection of cytotoxic effects as well. In addition, the assay can detect the occurrence of apoptotic processes. We performed experiments with cells of the human cell line MOLT-4, the rainbow trout cell line RTG-2 and the Chinese hamster cell line V79 for comparing the effects of chemicals from several classes in this test system. Results with MOLT-4 cells showed that treatment with camptothecine, actinomycin D and dexamethasone can induce apoptosis (microscopic observation, DNA ladder and sub-G, peak measurement with FCM). Data on the kinetics of the apoptotic process (sub-G, peak measurement with FCM) showed that the short treatment interval of 2 hours usually applied in comet assay studies is sufficient to induce apoptosis. Such cells undergoing apoptosis are responsible for an increase of the mean tail moment. Investigations with apoptotic cells demonstrated the appearance of more comets the shape of which is different from those obtained from non-apoptotic cells: hedgehog comets with enlarged tails. The appearance of hedgehog comets is regarded as air indication of apoptosis. This observation and the data described above indicate that the use of cell lines with apoptotic properties of different extent may have consequences on the interpretation of genotoxicity data. Therefore, for the evaluation of the genotoxic potential of chemical agents the use of cell lines with low or no apoptotic tendency is advisable. The V79 cell line seems to fulfill this requirement as our data demonstrate.

Keywords: Apoptosis, Comet-Assay, V79, Molt-4, RTG-2

Evaluation of a new flow cytometric analysis for in vitro chemically induced micronuclei

Danielle Roman*, Franziska Locher, Willi Suter, Maria Bobadilla

Department of Drug Safety, Sandoz Pharma AG, Postfach 4002, Basel, Switzerland

The measurement of the frequency of micronuclei induced in cells by ionizing radiation or by treatment with chemicals is widely used for analysis of cytogenetic damage. Microscopic scoring of micronuclei is a tedious and time-consuming procedure. Some attempts have therefore been made to automate micronuclei scoring by image analysis or by flow cytometry. A new flow cytometric analysis procedure for chemically induced micronuclei in V79 cells was establish in our laboratory. Non-specific debris was discriminated from micronuclei by a new gating procedure using the area and width fluorescence of the stained suspension of micronuclei and nuclei. To test the sensitivity and the specificity of this improved flow cytometric analysis, five well-known mutagenic compounds were tested. The frequency of chemical induced micronuclei measured and analyzed with this technique agreed well with results obtained by conventional microscopy. Additionally, a large series of negative controls and weak, middle and strong micronuclei inducers was used for assessing the correction factor needed for flow cytometric measurements to discriminate between toxic and Non toxic compounds. This procedure for flow cytometric screening of micronuclei represents a quick, reliable and relatively simple method for routine tests in drug safety assessment.

Keywords: Micronuclei, Flow cytometry, in vitro

Studies on the in vitro metabolism of dictamnine and on the photomutagenicity of metabolites in Chlamydomonas reinhardtii

Oskar Schimmer*, Bernhard Klier, Gisela Hofmann

Institut fur Botanik und Pharmazeutische Biologie, Universitat Erlangen-Nurnberg, Staudtstrab e 5, D 91058 Erlangen, Deutschland

The furoquinoline alkaloid dictamnine exhibits a remarkable photomutagenicity (1, 2). It can also: be activated into a mutagen by erogenous metabolization (3). In order to clarify the metabolism in vitro dictamnine was incubated with S9 mix or microsomal fractions from Pb-induced rat liver. Dictamnic acid, demethyldictamnine and a hydroxyderivative were identified by g.c.-m.s. and by synthesis. Further metabolises were characterized only by their mass spectra. The ultimate mutagen was postulated to be an unstable epoxide which could not be isolated. The proposed pathway of metabolization is similar to that of furocoumarins (4). The influence on photomutagenicity of the metabolization was studied using an algal reversion assay (2). In the presence of S9 mix or microsomes the photomutagenicity was quickly reduced with incubation time. The identified metabolises did not show any photomutagenicity.

1.Ashwood-Smith, M.J. et al., Mutation Res. 102. 401-412 (1982)

2.Schimmer, 0.. Kuhne, I., Mutation Res. 249, 105-110 (1991)

3.Hafele, F., Schimmer, 0., MUTAGENESIS 3, 349-353 (1988)

4.Mays, D.C. et al., Drug Metabol. Disposit. 15. 318-328 (1987)

Keywords: Dictamnine, metabolism, S9, photomutagenicity

In vitro toxicological evaluation of environmentally relevant compounds with estrogenic activity

Manfred Schlager*¹ , Franz Krassnigg² and Peter M. Eckl¹

1Institute of Genetics and General Biology, University of Salzburg, Helibrunnerstr. 34, A-5020 Salzburg and ²Federal Environment Agency, lnnsbrucker Bundesstrab e 47, A-5020 Salzburg, AUSTRIA

A number of environmentally relevant chemicals are capable of stimulating cell proliferation in the liver including liver tumor promoters such as dioxins, poiychiorinated biphenyls, certain peroxisome proliferators (i.e. plasticizers) and among others chemicals exhibiting estrogenic properties. Reports of abnormal sexual development of fish species, the findings indicating a reduction of male fertility during the last decades together with results implicating the possible relation to an increased breast and prostate cancer rate have focused common interest in such compounds, particularly aikyiphenols, persistent products of the biodegradation of aikyiphenol-poiyethoxyiates (non-ionic surfactants) and certain pesticides such as alpha-endosuifan, dieldrin and chiordane. Compared to the natural 17b -estradiol the estrogenic potencies are relatively low , however, combinations of estrogenic chemicals have been demonstrated to enhance the activation of estrogen receptors. Since estrogens are also implicated in liver carcinogenesis we are currently evaluating the toxicological potential of selected chemicals (4nonyiphenol, a.-endosuifan, chiordane, DDT, atrazine, dieldrin and tamoxifen) and combinations in primary rat hepatocytes. The endpoints analysed are: cytotoxicity, induction of micronuclei and chromosomal aberrations and mitogenic effects as determined by DNA synthesis and mitotic index analysis. Furthermore, we investigate the induction of vitellogenin synthesis in fish hepatocytes as a well-characterized method for the evaluation of estrogenicity. Preliminary results indicate a trend to dose-dependent increases of DNA synthesis and cell division. These together with the findings on genotoxicity and vitellogenin induction will be presented.

Keywords: estrogenic activity, rat hepatocytes, cyto- and genotoxicity, mitogenicity, fish hepatocytes, vitellogenin synthesis

Application of transgenic animal systems in mutation research

Peter Schmezer, Ute. M. Liegibel, Claudia Eckert, and H. Bartsch

Division of Toxicology and Cancer Risk Factors, German Cancer Research Centre (DKFZ), lm Neuenheimer Feld 280, 69120 Heidelberg, Germany

In vivo mutagenicity assays play a critical role in the risk assessment for mutagenic agents. The advent of transgenic technology has led to the development of novel in vivo mutagenicity assays and there is a broad potential of this technology to contribute to mutation research. While other systems are still in varying stages of development, two commercially available mouse gene mutation assays based on LacZ and Lacl shuttle vector systems have been used by a considerable number of investigators including our laboratory, An overview on both assays will be given with special focus on experimental issues, e.g. treatment regimen or organ/tissue specificity of mutagenic effects. From the data analysis, it can be concluded that transgenic rodent mutation assays offer a practical method to investigate mutagenic activity and mutational mechanisms in vivo both in somatic and germ cell DNA. The latter aspect of research has been the result of a recently performed international collaborative study in which seventeen laboratories have been involved (Mutation Research 388, 97, 1997).

Keywords: Transgenic rodent assays, in vivo mutagenicity, lacZ, Lacl, organ/tissue specificity

Activation of indirectly acting genotoxic substances in primary cultures of liver and gill cells from zebrafish (Brachydanio rerio).

A. Schnurstein*, E. Leist, A. Froschauer and T. Braunbeck

Department of Zoology 1, University of Heidelberg, Im Neuenheimer Feld 230, D-69120 Heidelberg, Federal Republic of Germany

Genotoxic substances represent serious threats to biota. Thus, there is an urgent need for the development of screening assays for mutagens in surface and drinking waters. For such purposes, the measurement of DNA fragmentation by means of the comet assay (single cell gelelectrophoresis, SCG) in primary cultures of fish hepatocytes and gill cells appear promising systems, since these cells represent eukaryotic systems provided with biotransformation capacity. In order to study reactions of zebrafish to common mutagens, cells were isolated and treated in vitro for 2 h with hydrogen peroxide, N-methyl-N-nitro-N-nitrosoguanidine, benzo[a]pyrene, dimethylnitrosamine, 2-acetylaminofluorene, 4-nitroquinoline-N-oxide and nitrofurantoin. Directly acting agents such as hydrogen peroxide and N-methyl-N-nitro-N-nitrosoguanidine induced massive time- and dose-dependent genotoxicity. 4-Nitroquinoline-N-oxide and nitrofurantoin were endogenously activated by cellular xanthine oxidase and nitroreductases, respectively, to metabolises inducing similar genotoxic effects (primarily via oxidative stress). In contrast, genotoxicants requiring activation by cytochrome P450 1A such as benzo[a]pyrene, 2-acetylaminofluorene and dimethylnitrosamine failed to induce any damage, unless they were pre-incubated with P450 1A-competent S9 preparations (from aroclor-stimulated rat). However, even after erogenous activation, genotoxic effects were lower than those induced by hydrogen peroxide, N-methyl-N-nitro-N-nitrosoguanidine, 4-nitroquinoline-N-oxide and nitrofurantoin. In general, hepatocytes reacted more sensitively than gill cells. Results suggest that, provided there is adequate activation of agents, the comet assay with fish cells is sensitive to the mutagens tested, and at least isolated fish hepatocytes represent a suitable system to investigate DNA damage.

Keywords: In vitro genotoxicity, comet assay, bioactivation, hepatocytes, gill cells, zebrafish

Analysis of the mutagenic potencies of methacrylate-based dental resin compounds in V79 cells.

Helmut Schweikl*, Gottfried Schmalz, Kirsten Rackebrandt

Poliklinik fur Zahnerhaltung und Parodontologie, Universitat Regensburg, D- 93042 Regensburg

Methacrylate esters are widely used in the production of polymers and resins including dental resin composite filling materials. Human exposure is frequent for both professional and medical reasons. Here we report on the analysis of mutagenic activities of unpolymerized dental resin compounds in the quantitative V79/hprt mutation assay. The experiments were carried out both in the presence and absence of a metabolically active microsomal fraction from rat liver (S9 fraction) Bisphenol A-diglycidyl methacrylate (Bis-GMA), urethane dimethacrylate (UDMA), tri ethylenglykol dimethacrylate (TEGDMA), methyl methacrylate (MMA), 2-hydroxyethyl methacrylate (HEMA), glycidly methacrylate (GMA), and Bisphenol A were dissolved in dimethyl sulfoxide to a concentration of 1 Mol/l. The stock solutions were then serially diluted in cell culture medium before testing. Initial range finding experiments indicated cytotoxic concentrations of the test substances in V79 Chinese hamster lung fibroblasts after a 24h exposure period from 30 m Mol/l (Bis-GMA) to about 20 mMol/l (MMA). GMA and TEGDMA were mutagenic in V79 cells at sub-toxic concentrations. The numbers of mutant cells induced by increasing concentrations of both substances were dose-related. Mutant frequencies were enhanced about 10-fold by 0.2 mMol/l GMA and 0.5 mMol/l TEGDMA, respectively. The numbers of GMA- and TEGDMA-induced mutants clearly decreased after an exposure period of 4h in the presence of a metabolically active microsomal fraction from rat liver. Methyl methacrylate (MMA) induced a very slight increase of mutant numbers at high concentrations (20 mMol/l). Bis-GMA, UDMA, HEMA, and Bisphenol A were not shown to be mutagenic in the present investigation. Our findings contribute to the growing numbers of various biologic effects of dental resin components at non-toxic concentrations including genotoxicity and mutagenicity. Supported by the Deutsche Forschungsgemeinschaft (DFG) (Schw 431/6-1).

Key words: Methacrylates, mutagenicity, V79/HPRT, dentistry

A combination of mutagenicity and recombinogenicity may confer aflatoxin B1 a potent liver carcinogen

Christian Sengstag

Institute of Toxicology, ETH and University of Zurich, Schorenstr. 16, 8603 Schwerzenbach, Switzerland

The potent liver carcinogen aflatoxin B1 (AFB1) (TD50 = 0.9 microgram/ kg/day for rat) is mutagenic in prokaryotes and eukaryotes. In the human p53 tumor suppressor gene it introduces G to T transversions predominantly at the hot spot codon 249. Apart from that, metabolically activated AFB1 is highly recombinogenic in engineered yeast strains where it induces gene conversion and chromosomal translocation events (1-3). In yeast, the recombinogenic activity of AFB1 by far exceeds its mutagenic activity, and various data in the literature are in agreement with AFB1-induced recombination also occurring in mammalian cells. A combination of the mutagenic activity with the recombinogenic activity might confer AFB1 a powerful genotoxin, since a de novo mutation introduced into p53 or another tumor suppressor gene will be rendered homozygous by the recombinogenic activity of AFB1.

1. Sengstag, C., B. Weibel and M. Fasullo (1996) Genotoxicity of Aflatoxin B1: evidence for a recombination-mediated mechanism in Saccharomyces cerevisiae, Cancer Res, (in press, December issue)

2. Sengstag, C. (1997) Working Hypothesis: The Molecular Mechanism of Aflatoxin B1-Induced Liver Cancer: Is Mitotic Recombination Involved?, Mol Carcinogenesis, (submitted)

3. Sengstag, C. and F.E. Wurgler (1994) DNA recombination induced by aflatoxin B-1 activated by cytochrome P450 1A enzymes, Molecular Carcinogenesis, 11, 227-235.

Keywords: aflatoxin B1, loss of heterozygosity, tumor suppressor gene, genotoxicity, chromosomal translocation, gene conversion

Evaluation of the comet assay as a routine in vitro genotoxicity test for surface waters

Andrea Sokolowski, Rainer Helbig and Herbert G. Miltenburger

CCR, Cytotest Cell Research GmbH & Co. KG, In den Leppsteins- wiesen 19, D-64380 Rossdorf, Germany

The aim of the BMBF-project is the development and evaluation of the alkaline version of the comet assay as a fast and reliable routine in vitro genotoxicity test for surface waters. In order to determine an appropriate and sensitive test system we used the Chinese hamster cell line V79 and the rainbow trout cell line RTG-2. After adaptation and optimisation of the comet assay protocol for both cell lines we studied in a first approach the short-term induction of DNA damage by genotoxic chemicals 2-acetylaminofluorene (2-AAF), benzo[a]pyrene (B[a]P), dimethylnitrosamine (DMN), nitrofurantoin (NF), and 4-nitroquinoline N-oxide (4-NQO), that might be components in waste waters. Activation of B[a]P, 2-AAF and DMN to their genotoxic metabolises was achieved by addition of Aroclor 1254-induced S9 rat liver fraction to the culture medium. To investigate whether DNA effects are accompanied by cytotoxic effects, cell viability was determined using the fluorochrome-mediated viability assay (FDA). Also survival was measured by the colony-forming ability (plating efficiency) of V79 cells or by a calorimetric assay (XTT-test) for RTG-2, respectively. In both cell lines the comet assay revealed a clear and concentration dependent increase in DNA strand breakage for all mutagens except 2-AAF. A significant increase in tail moment as analysed by statistical calculations was detected at concentrations which did not reduce viability and plating efficiency indicating the high sensitivity of the comet assay. However, the two cell lines differed markedly in their sensitivity to the mutagens. RTG-2 cells exhibited a significantly higher amount of DNA strand breakage after treatment with DMN, NF and 4-NQO as compared to V79 cells. For example, V79 cells had to be treated with 400 nM of 4-NQO to induce significant DNA damage, whereas similar effects were already caused by 12.5 nM in RTG-2 cells. Preliminary results from experiments assaying DNA damage after a postincubation period in mutagen-free medium indicate a prolonged lesion persistence in RTG-2 cells. These different repair capacities of the two cell lines might be responsible for the observed differences in sensitivities to various mutagens.

Keywords: Single cell gel (SCG) assay, DNA strand breaks, genotoxicity, cytotoxicity, V79, RTG-2.

Evaluation of the in vitro micronucleus test (mnt) as an alternative to the in vitro chromosomal aberration assay (ca): position of the gum working group "in vitro mnt"

Helga Stopper*, Beate Miller¹, Franziska Loche², Angelika Seelbach³, Dietmar Utesch⁴, Stephan Madle³

*Department of Toxicology, University of Wurzburg, D-97078 Wurzburg; ¹F. Hoffmann-La Roche Ltd., CH-4070 Basel; ²Sandoz Pharma Ltd., CH- 4002 Basel; ³BgVV, D-14191 Berlin; ⁴E. Merck AG, D-64271 Darmstadt;

For licensing of pharmaceuticals or chemicals, compounds must be tested for several genotoxicity endpoints including the induction of chromosomal aberrations in vitro. A working group of the GUM now evaluated published data on the in vitro MNT with the aim to judge the suitability of the MNT as a replacement for the in vitro CA test. After the application of strict rejection criteria, a data base of 97 publications and 34 compounds was obtained. For 30 of the compounds data on both tests were available. For 24 of these, concordant results were obtained in both test systems (80% correlation). The discordant results of 6 compounds can be explained by suspect or known aneugenic potential of these compounds. Considering that cell types and test protocols were extremely heterogeneous, this correlation is rather encouraging. Comparison of the different protocols and discussions within the working group yielded several recommendations for the routine use of the MNT. Although many cell lines are suitable, those most often used in genotoxicity testing (such as CHL, CHO, V79, HULY, L5178Y) are recommended. Cytochalasin B (CB) may be used in the case of HULY, but the possibility of interactions with aneugenic test compounds should be considered. For other cell lines, CB is not recommended by the working group. There seems to be flexibility in the choice of treatment and sampling times, although the average generation time of the cell line of choice as well as the mechanism of genotoxicity of the test compound should be taken into account for determination of sampling time. The use of appropriate toxicity tests is strongly recommended. Although studies on some parameters of the MNT protocol may be useful, the introduction of the MNT into genotoxicity testing/guidelines should not be delayed. Even in the present state the MNT is a reliable genotoxicity test. Compared to the CA, it is more reliable for the detection of aneugens, faster and easier to perform, has more statistical power and the potential for automation.

Keywords: in vitro micronucleus test, in vitro chromosomal aberration test, genotoxicity testing, GUM

The in vitro rat hepatocyte micronucleus assay in comparison with standard cytogenetic assays using V79 cells

Kerstin Muller-Tegethoff*, Birgit Kersten, Renata Schleicher, Anne Wolbert, Peter Kasper, Lutz Muller

Federal Institute for Drugs and Medical Devices, Seestrab e 10, 13353 Berlin, Germany

In vitro proliferating rat hepatocytes can serve as target cells for the investigation of micronucleus induction. In comparison with the usually performed micronucleus or chromosomal aberration assays with permanent cell lines (e.g. V79 or CHO cells), where external metabolization systems (mostly S9-mix) have to be added, the hepatocyte micronucleus assay has the advantage that the use of intact, metabolically highly competent hepatocytes ensures a more adequate metabolization of the test compounds. Additionally, no transfer of activated metabolises via the culture medium is necessary since the metabolizing and target cells are identical. In a comprehensive validation study with about 40 chemicals of different classes the hepatocyte micronucleus assay has shown to be a reliable system for the detection of mutagens and genotoxic carcinogens (Muller-Tegethoff et al., Mutation Res. 1996, in press). This study compares the sensitivities of the rat hepatocyte micronucleus assay and of standard cytogenetic tests using V79 cells in detecting the clastogenic potential of three liver carcinogens, the pyrrolizidine alkaloids retrorsine, monocrotaline and isatidine. In primary hepatocytes all three alkaloids gave clear positive results, showing a strong and dose dependent increase in the number of micronucleated hepatocytes. In V79 cells with S9-mix for metabolization only retrorsine and monocrotaline showed clastogenic effects, however only at 30 times higher concentrations than in the hepatocyte micronucleus assay. In the V79/S9-mix system isatidine gave no clastogenic effects at all. This study shows that misleading results can be generated when S9-mix is used for metabolization and it underlines the importance of using intact hepatocytes for clastogenicity studies, especially for the investigation of liver specific acting carcinogens.

Keywords: rat hepatocytes, micronucleus induction, V79 cells, S9-mix, pyrrolizidine alkaloids

Complementary roles of nucleotide excision repair and photoreactivation in chromosomes

Fritz Thoma

lnstitut fur Zellbiologie, ETH-Honggerberg, CH-8093 Z0rich

Cyclobutane pyrimidine dimers (CPDs) are a major class of stable DNA lesions generated by UV light. RNA-polymerase 11 is blocked by CPDs on the transcribed strand but not by CPDs on the non transcribed strand, In yeast S. cerevisiae, CPDs are repaired by nucleotide excision repair (NER) or photolyase in presence of light, Packaging of chromosomes into nucleosomes and higher order chromatin structures restricts the accessibility of the DNA for proteins and therefore is likely to affect transcription and DNA-repair, We analyzed the effect of chromatin structure and transcription on repair of CPDs by photolyase and NER in yeast genes and minichromosomes with defined chromatin structures. Photoreactivation is tightly regulated by chromatin structure; open promoter regions of active genes are repaired within 15 minutes, nucleosomes are repaired in 2 hours. In active genes of NER deficient strains, Photoreactivation was slow in the transcribed strand, but fast in the non transcribed strand. In inactive genes, both strands were repaired at similar rates, These results suggest that stalled RNA-polymerase 11 prevents accessibility of CPDs to photolyase. In NER proficient strains, fast photoreactivation in the non transcribed strand can match fast repair by NER in the transcribed strand (transcription coupled repair). A combination of both repair pathways allows efficient restoration of damaged genes after UV irradiation,

Keywords: pyrimidine dimer, photolyase, nucleotide excision repair, chromatin, transcription

Genotoxicity-test of unconcentrated surface waters- alkaline elution -

Petra Waldmann*, lmke Metz, Georg Reifferscheid, Patrizia Mohr and RudolfKarl Zahn

AMMUG, Universitat Mainz, Obere Zahlbacher Str. 63, 55101 Mainz, Germany

The alkaline elution is a powerful and sensitive method for detection of genotoxic potentials inducing different types of DNA-damage. Especially, the possibility to adapt this method to all kind of cells and organisms is an important advantage for environmental monitoring . environmental samples can be tested exposing organisms to their natural habitat, In order to detect genotoxic potentials in native fresh-water samples we adapted the alkaline elution method to the fresh-water clam Corbicula fluminea, an invertebrate which shows a. high bioaccumulation of xenobiotica. The use of this filter feeder enabled us to measure genotoxic effects of non-concentrated, native waters from the river rhine and other rivers in Germany. In the course of a BMBF-sponsored cooperative project we developed a mikroplate version of the alkaline filter elution assay to simplify and fasten routine measurements. In addition we adapted this assay to a primary producing organism - the algae Chlamydomonas reinhardtii. Studies on sensitivity to standard reference genotoxins (4-Nitroquinoline-N-oxid, NQO; 2-Acetylaminofluorene, AAF; Methyl,-N-Nitro-N-Nitroso-Guanidine, MNNG; Nitroso-Dimethylamin, NDA) with both aquatic organisms showed pronounced effects at low concentrations of the genotoxins. Significant DNA-fragmentation could be detected at concentrations of 1 m g/1 NCO 2 m g/1 AAF, 3.5 m g/1 NDA and 14 m g/1 MNNG in the test with Corbicula fluminea. The genotoxin Nitrofurantoin (NF), an antibiotic agent, which is dominantly acting in bacteria, induced - as expected - only a low enhancement of single-strand breaks at high concentrations (mg/1) in the gills of Corbicula fluminea. The OECs of MNNG, NQO, NF and NDA using the alkaline elution with Chlamydomonas reinhardtii were at a level of 14 m g/1 (MNNG), 20 m gA (NQO) 20/1 (NF) and 50 m g/1 (NDA). AAF showed no genotoxic effect on Chlamydomonas reinhardtii performing the same procedure.