Andreas Zeller

Assessment of the Genotoxic Potential of Azidothymidine in the Comet, Micronucleus, and Pig-a Assay

The genotoxic potential of azidothymidine (Zidovudine, AZT), chosen as a model compound for nucleotide analogs, was comprehensively assessed in vivo for gene mutation, clastogenicity, and DNA breakage endpoints. Male Wistar rats were treated by oral gavage over 7 days with AZT at dose levels of 2×0 (control), 2×250, 2×500, and 2×1000mg/kg/day with a final single dose given on day 8. DNA damage was then evaluated with the comet assay in liver, stomach, and peripheral blood and with the micronucleus test in bone marrow and peripheral blood (by flow cytometry) in the same animals. After a treatment-free period of upto 42 days, the Pig-a gene mutation assay was performed in peripheral blood of the high-dose animals. In the comet assay as well as the micronucleus test, AZT caused a considerable dose-dependent increase in DNA damage in all tissues evaluated and was highly cytotoxic to bone marrow and peripheral blood cells. These data are well in line with published results. Surprisingly, AZT did not significantly increase the number of Pig-a mutant cells. We speculate that two factors likely contributed to this negative result: a predominance of large deletions caused by AZT, and the relatively low statistical power of the first-generation scoring method used for this study.

Comparison of different cytotoxicity measures for the in vitro micronucleus test (MNVit) in L5178Y tk+/− cells: Summary of 4 compounds (Mitomycin C, Cyclophosphamide, Colchicine and Diethylstilboestrol) with clastogenic and aneugenic mode of action

This study summarises the results of four different chemicals evaluated for induction of micronuclei (MN) in L5178Y tk(+/-) cells in the absence of cytochalasin B. All four chemicals (the tubulin polymerisation inhibitor Colchicine, Diethylstilboestrol which inhibits both tubulin polymersation as well as depolymerisation, the cross-linking agent Mitomycin C and Cyclophosphamide which requires metabolism to form the ultimate mutagen) showed biologically and statistically significant induction in MN frequency compared to concurrent controls. Irrespective of whether the measure of cytotoxicity was based on relative cell count (RCC), relative increase in cell count (RICC) or relative population doubling (RPD), micronucleus induction was observed at or below the targeted toxicity of 55±5%. Therefore, all measures of cytotoxicity in the absence of cytochalasin B proved to be equally acceptable to select the top-dose without missing micronucleation activity for any of the four compounds.

Interlaboratory evaluation of a multiplexed high information content in vitro genotoxicity assay: Multiplexed High Information Content Assay

We previously described a multiplexed in vitro genotoxicity assay based on flow cytometric analysis of detergent‐liberated nuclei that are simultaneously stained with propidium iodide and labeled with fluorescent antibodies against p53, γH2AX, and phospho‐histone H3. Inclusion of a known number of microspheres provides absolute nuclei counts. The work described herein was undertaken to evaluate the interlaboratory transferability of this assay, commercially known as MultiFlow® DNA Damage Kit—p53, γH2AX, Phospho‐Histone H3. For these experiments, seven laboratories studied reference chemicals from a group of 84 representing clastogens, aneugens, and nongenotoxicants. TK6 cells were exposed to chemicals in 96‐well plates over a range of concentrations for 24 hr. At 4 and 24 hr, cell aliquots were added to the MultiFlow reagent mix and following a brief incubation period flow cytometric analysis occurred, in most cases directly from a 96‐well plate via a robotic walk‐away data acquisition system. Multiplexed response data were evaluated using two analysis approaches, one based on global evaluation factors (i.e., cutoff values derived from all interlaboratory data), and a second based on multinomial logistic regression that considers multiple biomarkers simultaneously. Both data analysis strategies were devised to categorize chemicals as predominately exhibiting a clastogenic, aneugenic, or nongenotoxic mode of action (MoA). Based on the aggregate 231 experiments that were performed, assay sensitivity, specificity, and concordance in relation to a priori MoA grouping were ≥ 92%. These results are encouraging as they suggest that two distinct data analysis strategies can rapidly and reliably predict new chemicals’ predominant genotoxic MoA based on data from an efficient and transferable multiplexed in vitro assay. Environ. Mol. Mutagen. 58:146–161, 2017.

Quantitative assessment of the dose–response of alkylating agents in DNA repair proficient and deficient ames tester strains

Mutagenic and clastogenic effects of some DNA damaging agents such as methyl methanesulfonate (MMS) and ethyl methanesulfonate (EMS) have been demonstrated to exhibit a nonlinear or even “thresholded” dose–response in vitro and in vivo. DNA repair seems to be mainly responsible for these thresholds. To this end, we assessed several mutagenic alkylators in the Ames test with four different strains of Salmonella typhimurium: the alkyl transferases proficient strain TA1535 (Ogt+/Ada+), as well as the alkyl transferases deficient strains YG7100 (Ogt+/Ada−), YG7104 (Ogt−/Ada+) and YG7108 (Ogt−/Ada−). The known genotoxins EMS, MMS, temozolomide (TMZ), ethylnitrosourea (ENU) and methylnitrosourea (MNU) were tested in as many as 22 concentration levels. Dose–response curves were statistically fitted by the PROAST benchmark dose model and the Lutz‐Lutz “hockeystick” model. These dose–response curves suggest efficient DNA‐repair for lesions inflicted by all agents in strain TA1535. In the absence of Ogt, Ada is predominantly repairing methylations but not ethylations. It is concluded that the capacity of alkyl‐transferases to successfully repair DNA lesions up to certain dose levels contributes to genotoxicity thresholds. Environ. Mol. Mutagen. 55:15–23, 2014.

In vitro genotoxicity of neutral red after photo-activation and metabolic activation in the Ames test, the micronucleus test and the comet assay.

Neutral red (Nr) is relatively non-toxic and is widely used as indicator dye in many biological test systems. It absorbs visible light and is known to act as a photosensitizer, involving the generation of reactive oxygen species (type-I reaction) and singlet oxygen (type-II reaction). The mutagenicity of Nr was determined in the Ames test (with Salmonella typhimurium strains TA1535, TA97, TA98, TA98NR, TA100, and TA102) with and without metabolic activation, and with and without photo-activation on agar plates. Similarly to the situation following metabolic activation, photo-mutagenicity of Nr was seen with all Salmonella strains tested, albeit with different effects between these strains. To our knowledge, Nr is the only photo-mutagen showing such a broad action. Since the effects are also observed in strains not known to be responsive to ROS, this indicates that ROS production is not the sole mode of action that leads to photo-genotoxicity. The reactive species produced by irradiation are short-lived as pre-irradiation of an Nr solution did not produce mutagenic effects when added to the bacteria. In addition, mutagenicity in TA98 following irradiation was stronger than in the nitroreductase-deficient strain TA98NR, indicating that nitro derivatives that are transformed by bacterial nitroreductase to hydroxylamines appear to play a role in the photo-mutagenicity of Nr. Photo-genotoxicity of Nr was further investigated in the comet assay and micronucleus test in L5178Y cells. Concentration-dependent increases in primary DNA damage and in the frequency of micronuclei were observed after irradiation.

Possibility of methodical bias in analysis of comet assay studies RE: DNA damage detected by the alkaline comet assay in the liver of mice after oral administration of tetrachloroethylene. (Mutagenesis, 25, 133-138, 2010)

Cederberg et al. (1) recently published an in vivo genotoxicity study on tetrachloroethylene using the alkaline comet assay. The authors concluded that the observed weak, but doserelated, increase in DNA damage following treatment was statistically significant and biologically plausible. The study had been performed at Covance laboratory and Cederberg et al. reported that the interpretation of the test outcome by the study director differed from their own assessment: the study director concluded that the test result was not indicative of a DNA damaging activity of tetrachloroethylene. Subsequent letters to the editor [Lillford et al. (2) and Lovell (3)] provided detailed insights into the different statistical approaches, with their merits and limitations in the analysis of comet assay data. While this exchange of opinions provided very stimulating discussions, we would like to address a different aspect of the study performance, which we believe could shed some light onto the interpretation of the study data. One of the prominent aspects reported by Cederberg et al. (1) was an apparently monotonous—albeit weak—increase of tail intensity (TI) values from control over low to high dose (Figure 1). It so happened that shortly before the publication by Cederberg et al., we had contracted a comet assay study in rats to the same laboratory and observed a similar apparently dose-related increase of TI values for male as well as for female rat liver (Figure 1, compound A). All TI values of the three treatment groups were well within the historical control range and there was unanimous agreement that the data did not indicate a DNA damaging property of the test item. There was, however, some discomfort because of the following simplistic statistical consideration: the probability for four data points to show a steadily increasing trend solely by chance is P5 1/4 1/3 1/25 1/24, which is just below the significance level of 5%. Noticing the parallels between our data and those on tetrachloroethylene reported by Cederberg et al. (1), we were prompted to plot all the liver comet assay data of studies contracted by us to Covance into one graph, together with the tetrachloroethylene data (Figure 1). While the monotonous increases over dose groups were not fully reproduced for the two other test compounds (B and C), it was apparent that, overall, the treated groups showed higher TI values than the respective control group in 17 of 18 instances (the remaining value was identical to the control group). Note that all our studies were concluded to show no DNA damaging activities of the test compounds. Consequently, the question arises whether the increased TI values are treatment related or whether an unidentified experimental bias could reproducibly cause the observed positive trends. Since the comet assay is known to be sensitively affected by different experimental variables, e.g. temperature and time span during the isolation of organs and preparation of single cells or during DNA unwinding and electrophoresis, the possibility of inadvertently introduced artifactual increases should definitely be considered. Specifically, we noted that the experimental workup was done starting from the animals of the control group and then proceeding to the treated groups in a dose group-related manner with the intention to avoid possible carry-over of test substance. We hypothesised that this process could introduce an experimental bias leading, seemingly, to dose-related, but in reality, to workup-related trends. In our most recent study (with compound D) performed at the same laboratory, we specifically asked to do the workup in a modified order, namely: ‘high dose/ control / low dose / intermediate dose’. As seen in Figure 2A, no upward trend was apparent any more when TI values are plotted against dose. However, when plotting according to the order of workup (Figure 2B), all groups processed subsequently to the first group showed increased TI values. As observed in the previous studies (Figure 1), the group processed first showed the lowest TI values. We see this observation as a support to the hypothesis that an experimental bias is inadvertently introduced when keeping a fixed workup order starting with the control animals and proceeding in a dose-wise fashion (as, to our knowledge, is generally done in all laboratories performing the comet assay). We do want to emphasise that this conclusion is not aimed at criticising the work of the laboratory since we know that the colleagues indeed possess high expertise and perform efficient and dependable work. We do, however, wish to point out that a technically demanding test such as the comet assay poses considerable challenges and we speculate that the discussed workup order might be a reason for apparently dose-related trends. While we conclude that the weak monotonous increase in TI values is most likely an experimental artifact and not related to primary DNA damage induced by tetrachloroethylene, we much appreciate the scientific exchanges this study has elicited (2,3) and we hope that our contribution will be leading to further discussions.

GADD45α induction in the greenscreen HC indicator assay does not occur independently of cytotoxicity

Mammalian chromosomal integrity assays are influenced by cytotoxicity, a phenomenon which impacts data interpretation, assay specificity, and regulatory testing guidelines. Concordance of the GADD45α GreenScreen HC indicator assay to established in vitro and in vivo genetic toxicological assays has previously been described, yet a detailed description in the manner by which cytotoxicity influences its performance has not. Here we present a post‐hoc analysis of a previously tested set of 91 proprietary and nonproprietary compounds investigating the influence of cytotoxicity on GADD45α induction and how varying assay cutoff criteria impacts assay performance. Significant cytotoxicity was identified to accompany the majority (72%) of compounds classified as genotoxic by GADD45α induction. Decreasing the GADD45α genotoxic induction criteria (from a 50 to a 30% increase over solvent controls) resulted in an increased assay sensitivity (from 30 to 68%) and concordance (from 55 to 68%), though a concomitant decrease in specificity was also observed (from 97 to 68%). We conclude that GADD45α induction in the GreenScreen HC indicator assay is influenced by cytotoxicity and that assay performance can be improved if different cutoff criteria are implemented. Environ. Mol. Mutagen. 52:28–34, 2011.

A proposal for a novel rationale for critical effect size in dose-response analysis based on a multi-endpoint in vivo study with methyl methanesulfonate.

Methyl methanesulfonate, a well-known direct-acting genotoxicant, was assessed in a multi-endpoint study in rats using six closely spaced dose levels. The main goal of the study was to investigate the genotoxic response at very low doses and to analyse this response with dedicated statistical tools in order to find a Point of Departure (PoD) and related metrics. Software packages like PROAST or EPA-BMDS require the toxicologist to define a so-called critical effect size (CES) or benchmark response (BMR) and this choice has a large impact on the result of the PoD calculation. Currently, increases of 5%, 10% or 1 standard deviation over concurrent vehicle controls have been proposed for CES/BMR, values that may or may not be suited for all genotoxicity endpoints. Based on the data obtained in this study, we propose an endpoint specific CES approach that reflects the typical evaluation process of a regulatory acceptable genotoxicology study. However, we are aware that this ratio-based CES strategy will need to be more fully developed with additional experimentation and should be mainly seen as a starting point for scientific discussion.

Genotoxicity testing of peptides: folate deprivation as a marker of exaggerated pharmacology.

The incidence of micronucleated-cells is considered to be a marker of a genotoxic event and can be caused by direct- or indirect-DNA reactive mechanisms. In particular, small increases in the incidence of micronuclei, which are not associated with toxicity in the target tissue or any structurally altering properties of the compound, trigger the suspicion that an indirect mechanism could be at play. In a bone marrow micronucleus test of a synthetic peptide (a dual agonist of the GLP-1 and GIP receptors) that had been integrated into a regulatory 13-week repeat-dose toxicity study in the rat, small increases in the incidence of micronuclei had been observed, together with pronounced reductions in food intake and body weight gain. Because it is well established that folate plays a crucial role in maintaining genomic integrity and pronounced reductions in food intake and body weight gain were observed, folate levels were determined from plasma samples initially collected for toxicokinetic analytics. A dose-dependent decrease in plasma folate levels was evident after 4 weeks of treatment at the mid and high dose levels, persisted until the end of the treatment duration of 13-weeks and returned to baseline levels during the recovery period of 4 weeks. Based on these properties, and the fact that the compound tested (peptide) per se is not expected to reach the nucleus and cause DNA damage, the rationale is supported that the elevated incidence of micronucleated polychromatic erythrocytes is directly linked to the exaggerated pharmacology of the compound resulting in a decreased folate level.

DNA Damage Response of 4-Chloro-Ortho-Toluidine in Various Rat Tissues

The potential of 4-chloro-ortho-toluidine (4-CloT), an aromatic amine substituted on the ortho- and para-position of the amine function, to induce DNA damage in male Wistar rats was evaluated with the micronucleus test (peripheral blood), Pig-a (peripheral blood), and comet assay (peripheral blood, liver, urinary bladder, jejunum) at several time points. In addition to those markers of DNA damage, ie, gene mutation and clastogenicity, standard hematology, including methemoglobin, histopathology and immunohistochemistry of γ-H2AX and Ki-67 in liver, jejunum, and urinary bladder were performed. 4-CloT was administered orally over 28 consecutive days (days 1-28), followed by a 28-day treatment-free (days 29-56), and a second dosing phase of 3 days (days 57-59). 4-CloT showed some effects on the integrity of the DNA as measured by the comet assay in liver and urinary bladder but not in peripheral blood or jejunum. However, for liver and urinary bladder histopathological changes were observed. An increase in the frequency of micronuclei in peripheral blood was seen in parallel to a dose-dependent increase of reticulocytes and methemoglobin. Therefore, impact from a compensatory erythropoiesis on micronucleation cannot be excluded. Interestingly, no increase in the frequency of RETCD59- and RBCCD59- was observed in the Pig-a assay.