Annual Meeting of the GTA
May 11-12, 2017
John M. Clayton Hall Conference Center
University of Delaware
Poster AbstractsPoster #1
Enhancing the Leadscope genetox expert alerts supporting ICH M7
Tetyana Kobets1, Michael J. Iatropoulos1, Jian-Dong Duan1, Klaus D. Brunnemann1, Dumitry A. Iacobas1, Sanda Iacobas1, Esther Vock2, Ulrich Deschl2, Gary M. Williams1
1New York Medical College, Valhalla, NY, USA; 22Boehringer Ingelheim Pharma GmbH&Co. KG, Biberach an der Ri
A variety of genotoxic carcinogens tested in the Chicken Egg Genotoxicity Assay (CEGA), which assesses liver DNA breaks and adducts, also interfered with fetal liver proliferation, differentiation and migration, leading to hepatocellular dysplasia and distortion of trabecular pattern. The present study assesses cell proliferation and gene expression in fetal chicken livers, since these processes are involved in embryo-fetal development and their dysregulation can lead to neoplastic development. Groups of at least 12 white leghorn chicken eggs were administered the vehicles for CEGA, deionized water (DW) and 20% aqueous solution of Solutol HS15 (HS15), in 3 daily injections on days 9 - 11 of incubation. The group used as control did not receive any injections. Three hours after the last injection, half of the livers were collected for gene expression analysis. One day (day 12 of incubation) and one week (day 18) after dosing was discontinued, the remaining livers were collected for histologic analysis. At day 12, in sections stained with hematoxylin and eosin, in all groups, nascent trabecular hepatocellular pattern was established and normal liver cells and extracellular matrix elements were present. At day 18, the hepatocellular trabecular pattern was completed. Immunohistochemical staining for proliferating cell nuclear antigen (PCNA) showed no significant differences in cell proliferation between the groups, the percentage of proliferating cells at 12 and 18 days was in the range of 61% to 65%. Gene expression analysis using chicken 44K Agilent microarray, revealed that DW had minimal effect on the expression of genes involved in the regulation of cell cycle and proliferation, while downregulation of two tumor suppressor genes p53 and APC was observed. HS15 significantly deregulated multiple genes involved in cell cycle and proliferation pathways. Among genes upregulated were Rad21, p300, Mdm2, Mad1. Downregulated genes included APC, CycA, CycB, TGFÎ², p53, GADD45 and PCNA. Thus, cellular proliferation in fetal chicken livers was not affected by DW and HS 15. Despite alterations in the expression of genes involved in cell cycle and proliferation produced by HS15, fetal chicken hepatocytes evidently maintained their proliferative status.
Category: Early Stage Investigator
Back to Top ↑
Testing of Flavor and Fragrance Materials in Turkey Egg Genotoxicity Assay (TEGA) and Comparison of the Results In Ovo, In Vitro and In Vivo
Gary M. Williams1, Tetyana Kobets1, Jian-Dong Duan1, Klaus D. Brunnemann1, Michael J. Iatropoulos1, Sylvain Etter2, Christina Hickey2, Benjamin Smith2
1New York Medical College, Valhalla, NY; 2Firmenich Inc, Plainsboro, NJ
The genotoxic potential of 19 diverse flavor and fragrance (F & F) agents was assessed in the Turkey Egg Genotoxicity Assay (TEGA) using 32P-nucleotide postlabeling (NPL) and comet assays to detect hepatic DNA adducts and strand breaks, respectively. The compounds were selected for testing based on their chemical structures and results in the GADD45a-Gluc BlueScreen™ HC (BSHC) genotoxicity and the Ames mutagenicity assays. Groups of at least 10 fertilized medium white turkey eggs, containing 22 to 24 day old fetuses, received 3 daily injections of test substances at various dose levels and were terminated 3 hours after the last dose. Livers were collected for analyses. BSHC and Ames positive quinoline (QUI), at 15 mg/egg was positive in both NPL and comet assays. Among F & F agents, p-tert-butyldihydrocinnamaldehyde (BDHCA) at 20 mg/egg and methyl eugenol (MEU) at 4 mg/egg produced DNA adducts. Additionally, BDHCA at 20 mg/egg, p-t-butyl-a-methylhydrocinnamic aldehyde (BMHCA) at 10 mg/egg, trans-2-hexenal (HEX) up to 10 mg/egg and maltol (MAL) up to 20 mg/egg produced DNA strand breaks. Fourteen other tested compounds were negative in both NPL and comet assays. Based on reports of oxidative DNA damage induction by MAL and 4-hydroxy-2.5-dimethyl-3(2H) furanone (HDMF), these compounds were tested in the enhanced comet assay using repair enzymes to identify oxidative DNA damage. In the enhanced comet assay positive comet findings for MAL were not confirmed, and only equivocal evidence of oxidative damage was found. Thus, this compound was considered to have equivocal genotoxicity in TEGA. Meanwhile, HDMF, produced positive results in the enhanced comet assay with formamidopyrimidine DNA glycosylase (FPG) enzyme digestion, indicative of its ability to produce DNA strand breakage via oxidative damage. When compared to in vitro test results, TEGA had good specificity and moderate positive predictivity. Differences with in vitro genotoxicity could be due to differences in endpoints measured by TEGA and in vitro tests, or due to false-positive results in the in vitro systems. In contrast, findings in TEGA, with few exceptions, were concordant with the results of the genotoxicity testing of the chemical set in vivo. Thus, TEGA is an attractive alternative screening model for assessment of genotoxic potential of chemicals in vivo.
Back to Top ↑
Extending the MultiFlow™ Assay from Hazard ID to Mechanistic Insights About DNA Reactivity
Derek T. Bernacki, Steven Bryce, Jeffrey Bemis, Stephen Dertinger
The MultiFlow™ assay is based on flow cytometric analysis of detergent-liberated nuclei that simultaneously provides information about gH2AX, phospho-H3, p53 status and cytotoxicity. Previous work with TK6 cells has shown that this multiplexed assay is useful for genotoxic hazard identification, with the additional benefit of discriminating chromosome-damaging agents as acting through clastogenic or aneugenic modes of action. The current work was conducted to evaluate whether this platform could be adapted to provide additional information about clastogenesis, that is, whether it is caused by DNA-reactivity or is occurring through non-DNA mechanisms. For these experiments, TK6 cells were treated with each of 30 clastogens that were predicted as such in a first-tier, continuous exposure assay. After 4 hrs cells were removed from MultiFlow analysis, and the remainder were washed free of chemical and then allowed to recover for an additional 20 hrs at which time cells were analyzed as before. Based on our a priori classifications of DNA-reactivity, Random Forest, a machine-learning tool, was used to predict each chemical’s group membership. Among 4 useful biomarkers, gH2AX at 24 hrs was found to be a particularly predictive, as it tended to be maintained or even increased at the later time point in the case of DNA-reactive compounds, and was at baseline or near-baseline values for non-reactive chemicals. The full model’s concordance with a priori categories was >90%. Cross-validation, using a 25% hold-out fraction, provide evidence that the model is not overfitting the data. Collectively these results suggest MultiFlow data, in combination with a Random Forest model, reliably predict clastogens’ DNA-reactivity. Further work with external test set chemicals (i.e., those not used in model construction) is in progress to assess the generalizability of these tools.
Category: Early Stage Investigator - Award Recipient
to Top ↑
Discrimination of Micronuclei Inducing Aneugens by Aurora Kinase Inhibition or by Tubulin Binding
Stephanie Coffing, Pamela Heard, Maik Schuler
Pfizer Worldwide Research and Development
Aneugens represent a large portion of the compounds found to be positive in the in vitro micronucleus assay. The vast majority of compounds tested in the pharmaceutical industry induce aneugenicity either through the inhibition of aurora kinases or the interference with tubulin polymerization. To investigate the two mechanisms of action, TK6 cells were treated with five known aurora kinase inhibitors and five known disruptors of tubulin polymerization and were analyzed for micronucleus induction, biomarkers of aneugencity and clastogenicity, aurora kinase inhibition and tubulin polymerization. The results from these studies show that micronucleus induction for aurora kinase inhibitors happens at concentrations that induce significant polyploidy and correlate well with the IC50 for aurora kinase inhibition as measured in a phospho-H3 dephosphorylation assay. By comparison, tubulin inhibitors induce predominantly micronuclei at lower concentrations and mostly polyploidy at higher concentrations. In addition, tubulin inhibitors increase the mitotic index at 4-hours while aurora kinases do not. The IC50 in a cell free tubulin polymerization assay allows for tubulin inhibitors to be rank ordered by potency. Based on the results from these studies, we will propose a Genotoxic Mode of Action Pathway (GMAP) for both mechanisms and identify a simple matrix on how to distinguish both mechanisms.
Back to Top ↑
Complementarity of Cellular Biomarkers Quantification for Characterizing Genotoxic Mechanisms of Action in the HepG2 Cell Line
Marc Audebert, Kopp, Dario, Zalko
INRA, TOXALIM UMR1331 Toxalim, Research Centre in Food Toxicology
The development of fast and specific high-throughput in vitro genotoxicity assays is a major priority, notably in the context of the REACH program. Currently in vitro genotoxicity tests are sensitive, but the occurrence of irrelevant positive results is high compared to in vivo studies. Moreover, most in vitro genotoxicity assays are not able to discriminate aneugenic from clastogenic compounds, and cytotoxicity could be a confounding factor. Furthermore, assays generally rely on cell lines with limited metabolic capabilities. The phosphorylated histone H2AX (gH2AX) is a global marker of genotoxicity. Coupled with the quantification of the phosphorylation of histone H3 (pH3), it also allows an efficient discrimination between aneugenic and clastogenic compounds1. However, these two biomarkers do not permit to deduct the specific mechanisms involved in the action of clastogenic compounds. The aim of this study was to investigate other possible biomarkers allowing differentiating clastogenic properties. For this purpose, we analyzed gH2AX and pH3, but also six other biomarkers involved in the DNA damage signaling pathway (p-p53(S15); p21; p53; p-Chk1(S345); p-ATM(S1981) and p-Chk2(T68)) in HepG2 cells treated with nine clastogens exhibiting different mechanisms of action, as well as one aneugen. All compounds were tested at various concentrations and with kinetics of 2, 6, 24 and 48 h. No specific biomarker signature of mechanisms of clastogenic action was observed. However, our results demonstrate the activation of the investigated biomarkers by the tested compounds in a time and concentration dependent manner. In addition, multiparametric analysis demonstrates a strong correlation between gH2AX and p-p53(S15). The latter biomarker is expected to improve the previous proposed strategy for accurately classifying clastogenic and aneugenic compounds.
1Khoury, L., et Al. (2016). Arch. Tox., 90(8):1983-95.
Back to Top ↑
Application of the ToxTracker Assay in a MOA-Based Genotoxicity Classification of Mutagenic, Clastogenic and Aneugenic Compounds
Giel Hendriks1, R.S. Derr1, P.I. Racz1, L. Zeijdel1, H. Vrieling2
1Toxys B.V., Leiden, the Netherlands; 2Leiden University Medical Center, The Netherlands
ToxTracker is a mammalian stem cell-based reporter assay that detects activation of specific cellular signalling pathways upon exposure to unknown compounds (Hendriks et al, Tox Sci 2016). ToxTracker contains six different GFP-tagged reporters that allows discrimination between induction of DNA damage, oxidative stress and protein damage in a single test.
In an extensive validation study using 300 reference chemicals and >200 proprietary compounds, ToxTracker classified the genotoxic compounds with a sensitivity of 94% specificity of 95%. Interestingly, various compounds that give misleading positive results in the conventional in vitro genotoxicity assays did not activate the DNA damage reporters by did induce high levels of oxidative stress or protein damage in ToxTracker Next we investigated if ToxTracker could provide insight into the mode of action of genotoxic compounds. By assessing the differential induction of the two DNA damage reporters, ToxTracker was able to discriminate between a mutagenic and clastogenic mechanism of genotoxicity. The Bscl2-GFP reporter is activated in response to DNA replication stress following induction of bulky, promutagenic DNA lesions. Clastogenicity is indicated by activation of the Rtkn-GFP reporter following induction of DNA double strand breaks.
Furthermore, we found that the assay could discriminate between a clastogenic and aneugenic mode by the selectively induction the Rtkn-GFP reporter. Induction of the Rtkn-GFP reporter was significantly slower (>12 h) for the mitotic spindle poisons compared to clastogenic compounds (8 h).
Finally, we developed an extension of ToxTracker to identify aneugenicity by cell cycle kinase inhibitors. The ToxTracker reporter cell lines are simultaneously stained with antibodies against phosphorylated histone H3 and with a DNA stain. Decreased levels of phospho-H3 and polyploidy are both hallmarks for aneugenicity by kinase inhibitors. Together, the integrative approach of the ToxTracker assay provides a powerful tool for in vitro carcinogenic hazard identification of chemicals by unveiling activation of specific cellular signalling pathways upon exposure and delivers insight into the underlying mechanism of toxicity.
Back to Top ↑
Optimization of Random Forest Analysis to Predict Genotoxic Mode of Action Based on MultiFlow™ Assay Data
Steven M. Bryce, Derek Bernacki, Jeffrey Bemis, Stephen Dertinger
MultiFlow™ is a multiplexed add-and-read flow cytometry assay that simultaneously evaluates cells for chemical-induced changes to gH2AX, phospho-histone H3, nuclear p53 content, polyploidization, and cytotoxicity. Previous work with the MultiFlow assay and TK6 cell from 7 laboratories, 84 chemicals, and 231 experiments explored the use of global evaluation factors and multinomial logistic regression to predict chemicals’ genotoxic MoA. The current report extends this work by investigating the utility of a machine learning strategy known as Random Forest, an ensemble-based approach that is considered a powerful tool for classification problems. First, the following variables were optimized: number of trees in the forest, numbers of terms sampled per split, and minimum size split to best categorize a training set of chemicals (n=84) as either clastogenic, aneugenic, or non-genotoxic. Performance metrics, including R2 values and misclassification rates, indicate that 50 trees, 3 terms sampled per split, and a minimum split size of 1 optimized the analysis. Sensitivity and specificity values > 95% were observed for the full training set. Cross-validation, using a 25% hold-out fraction, provides evidence that the model is not overfitting the data. Collectively these results suggest MultiFlow data, in combination with a Random Forest model, reliably predict chemicals’ predominant genotoxic MoA. Further work with external test set chemicals (i.e., those not used in model construction) is in progress to assess the generalizability of these toolsy.
Back to Top ↑
The TGx-28.65 Biomarker to Detect DNA Damage-Inducing Substances: Validation and Provision of a Reference Dataset using Quantitative RT-PCR and Next Generation Sequencing
Eunnara Cho1, Julie K. Buick2, Andrew Williams2, Heng Hong Li3,4, Albert J. Fornace Jr.3,4, Jiri Aubrecht5, Carole L. Yauk1,2
1Department of Biology, Carleton University, Ottawa, Ontario, Canada; 2Environmental Health Science and Research Bureau, Health Canada, Ottawa, Ontario, Canada; 3Dept. of Biochemistry and Molecular and Cellular Biology, Georgetown University Medical Center, Washington, District of Columbia; 4Department of Oncology, Georgetown University Medical Center, Washington, District of Columbia; 5Drug Safety Research and Development, Pfizer Inc., Groton, Connecticut
Transcriptional responses provide insight into the mode of action (MoA) of toxicants. Indeed, toxicogenomic signatures can be used as predictive mechanistic markers in chemical risk assessment. To this end, TGx-28.65, a genomic biomarker consisting of 63 genes, was derived from expression profiles of 28 training (reference) chemicals in TK6 human lymphoblastoid cells to identify DNA damage-inducing (DDI) agents. The biomarker was originally developed using DNA microarray technology. To date, the TGx-28.65 biomarker has demonstrated strong performance in discriminating DDI from non-DDI agents in both internal and external validation exercises.
In the present study, the TGx-28.65 biomarker was reproduced and tested on two additional gene expression platforms, quantitative RT-PCR and RNA sequencing (RNA-seq), to broaden the utility and facilitate the application of the biomarker in genotoxicity testing. QPCR is the most widely available tool for gene expression studies, while RNA-seq is a newer technology that generates high-content data on the whole transcriptome. TK6 cells were treated with the same training chemicals at the same concentrations used initially in the microarray study. The biomarker gene expression changes were measured using TaqMan qPCR assays in 96-well plates, and the transcriptome was sequenced and quantified using the Illumina NextSeq system. Both the qPCR and RNA-seq profiles accurately classified 27 of the 28 training agents as DDI or non-DDI. Methotrexate, a DNA/RNA antimetabolite, was misclassified as non-DDI by both platforms. Furthermore, external validation of the qPCR profiles using 24 additional chemicals demonstrated conservation of classification capabilities of the biomarker between the qPCR and microarray platforms. External validation of the RNA-seq profiles is currently in progress.
The TGx-28.65 biomarker provides mechanistic insight and can readily be integrated into the current standard in vitro genotoxicity test battery. The addition of qPCR and RNA-seq reference datasets extends the application of the biomarker to laboratories that are interested in applying more focused or modern technologies.
Back to Top ↑
Improving the Performance of Alkyl Halide and Aromatic Amine Mutagenicity Prediction
Glenn J Myatt, David A. Bower, Kevin P. Cross, Catrin Hasselgren, Scott Miller, Donald P. Quigley
In support of the ICH M7 guideline, this poster will describe how information derived from proprietary databases has been used to improve the performance of public bacterial mutagenicity expert alerts, without revealing confidential compounds or data using a SAR fingerprint methodology. This methodology was run against over 35,000 compounds with mutagenicity data. The results have been used to improve the overall performance of the Leadscope expert alerts system as well as continued improvements to specific classes including aromatic amines, aromatic amides, aryl boronic acids, and alkyl halides. This poster describes how this analysis can be used to support the efficient adherence to the ICH M7 guideline, including any expert review.
Back to Top ↑
Methods to Handle Out-of-Domain Results for ICH M7
Catrin Hasselgren, David A. Bower, Kevin P. Cross, Scott Miller, Donald P. Quigley, Glenn J. Myatt
The ICH M7 guideline specifies that the “…absence of structural alerts…” from the two defined (Q)SAR methodologies is sufficient to assign the impurity to class 5. Many (Q)SAR approaches will predict an impurity to be positive or negative, in addition to generating out-of-domain or indeterminate results. This poster will outline a range of approaches to handling out-of-domain results. These approaches will cover data and knowledge sharing schemes, expert review options, and rebuilding of the (Q)SAR model. The advantages and disadvantages of the different approaches will be assessed. This will also include a discussion on the assignment of out-of-domain results to class 4 or 5 as defined by ICH M7 guideline.
Back to Top ↑
Flow Cytometric Scoring of Rat Liver Micronuclei Progress Towards a High Throughput High Information Content Method
Stephen Dertinger, Svetlana Avlasevich, Sumee Khanal, Priyanka Singh, Jeffrey Bemis
A preferred endpoint for characterizing chemicals’ in vivo chromosomal damage potential is micronucleus formation, usually evaluated in hematopoietic cells. Here we describe progress towards automatically scoring the incidence of rat hepatocyte micronuclei via flow cytometry. The method currently takes the following form: excision of non-perfused left lateral lobe; slicing liver tissue into thin sections; clearing erythrocytes; exposing to collagenase; and a stepwise addition of two solutions. These solutions serve to lyse outer membranes and thereby liberate micronuclei and nuclei, clear the chromatin of membranous debris, stain chromatin with fluorescent dye(s) and label dividing cells with a fluorescent antibody against Ki-67. Proof-of-concept data are provided from experiments with 4 week-old Sprague Dawley rats that were exposed to the prototypical liver genotoxicant diethylnitrosamine (DEN) for two consecutive days (0 or 50 mg/kg/day). Four days after cessation of treatment left lateral lobes were collected and processed for flow cytometric analysis. Changes to Ki-67, ploidy status and micronucleus frequencies were observed for all DEN-treated animals. Whereas the fraction of Ki-67-positive nuclei appears to denote recent cellular division, the frequency of 4n+ nuclei conveys information about division history over a longer period of time. While negative control animals tended to exhibit micronucleus frequencies in the range of 0.1 to 0.3%, DEN-exposed rats showed frequencies of approximately 2%. Parallel microscopic scoring of hepatocytes stained with acridine orange confirmed the accuracy of these micronucleus frequencies. Similar results were observed for 6 week-old animals treated for two days with DEN, an unexpected result given our expectation of less cell division in these somewhat older animals. Collectively, these results are encouraging, as they suggest an important tissue compartment and endpoint will be amenable to an objective, automated scoring technique that provides higher information content relative to current microscopy-based approaches. Additional work will be needed to further optimize the methods, evaluate the suitability of several important in-life considerations such as rodent age and treatment schedules, and test the transferability of the method.
Back to Top ↑
Long-Term Persistence of Polycyclic Aromatic Hydrocarbon (PAH)-DNA Adducts in post-Mortem Whale and Mouse Gastrointestinal Tissues
Mehnaz Ali1, Alyssa Damon1, Frederick A. Beland2, Kathy Divi1, Michelle Vanlandingham2, Daniel Martineau3, Stephanie Lair3, Miriam C. Poirier1
11LCBG, CCR, National Cancer Institute, Bethesda, MD; 2Division of Biochemical Toxicology, NCTR, FDA, Jefferson, AR; 3U. Montreal College of Veterinary Medicine, St. Hyacinthe, PQ, Canada
Some PAHs, a family of compounds formed by partial combustion of organic materials, are carcinogenic and bind to DNA, providing a critical link between PAH exposure and tumor induction. In the St. Lawrence River Estuary (SLE) in Quebec, Canada, high levels of PAH contamination in the Saguenay river sediment are associated with high rates of gastrointestinal cancer in the resident beluga whales. Using immunohistochemical (IHC) staining, with an antiserum specific for DNA modified with several carcinogenic PAHs, we found PAH-DNA adduct formation in intestines of beluga whales found stranded on the shores of the SLE. Because whales are often not autopsied for several days post-mortem, there was concern that PAH-DNA adducts would be unstable during that time. To investigate the stability of PAH-DNA adducts in post-mortem tissue, we stained small intestine and forestomach taken from B6C3F1 mice injected intraperitoneally with 0 or 2.0 mg benzo[a]pyrene (BP), a carcinogenic PAH. Twenty-four hours later (0 hr), mice were euthanized and tissues from 3 mice/group were harvested, while the remaining mouse carcasses were refrigerated (4°C) and tissues were taken at 4, 8, 12, 24, 48, and 168 hr. IHC of paraffin-embedded sections showed intact nuclei with strong DNA adduct staining in BP-exposed mouse forestomach up to 168 hr, and weaker DNA staining, with more autolysis, in BP-exposed mouse intestine for up to 48 hr. Overall, the study showed that PAH-DNA adducts are stable long-term in nuclei of post-mortem whale and mouse gastrointestinal tissues. Additionally, the high levels of DNA adducts in mouse forestomach at 0 hr suggest a rationale for the susceptibility of the forestomach to tumor induction.
Back to Top ↑
The Liver Delivers: Understanding the Value of Cultured Muta™Mouse Primary Hepatocytes for Mutagenicity Assessment
Julie A Cox1, Raeesa E. Matadar2, Edwin P. Zwart3, Mirjam Luijten3, Paul A. White1,4
1Department of Biology, University of Ottawa, Ottawa, ON; 2Faculty of Science, University of Ottawa, Ottawa, ON; 3RIVM, Bilthoven, The Netherlands; 4EHSRB, Health Canada, Ottawa, ON
Large-scale initiatives, such as ToxCast, promote a shift in the paradigm for regulatory evaluations of new and existing substances; specifically, away from time-consuming in vivo assays towards predictive, short-term in vitro assays. Unfortunately, the assays included in such initiatives are ill-equipped to assess chemically-induced genetic damage and mutation. Moreover, many currently used mammalian cell genotoxicity assays generate an unacceptably high frequency of false or irrelevant positive results with respect to in vivo mutagenicity and/or carcinogenicity. A novel in vitro gene mutation assay utilizing primary hepatocytes from the transgenic Muta™Mouse was been developed to address the shortfalls of existing in vitro mutagenicity assays. In order to assess the utility of the Muta™Mouse primary hepatocyte assay, the cells were extensively characterized. Freshly isolated cells were found to have a hepatocyte-like morphology, with a large proportion of binucleated cells. After 24 hours in culture on a collagen-coated substrate, the hepatocytes began to lose their cuboidal morphology and gain a more fibroblast-like appearance. This de-differentiation continued over the 5 day culture time course. Fluorescent immunocytochemistry demonstrated that Muta™Mouse primary hepatocytes maintain hepatocyte-specific markers, such as albumin and cytokeratin 18, for up to 5 days in culture. Markers for other hepatic cell types (i.e., fibroblasts, stellate cells, Küppfer cells, and bile duct cells) were studied but not found in the primary hepatocyte cultures, indicating that the cultures are not contaminated with other hepatic cells. Karyotype analysis revealed a normal murine karyotype; the majority of the cells are polyploid, a well-known characteristic of hepatocytes. Fluorescence in situ hybridization (FISH) analysis confirmed the presence of the lambda shuttle vector on chromosome 3, a defining characteristic of the Muta™Mouse. The rate of nuclear division of the cultured hepatocytes was measured using a modified flow cytometry technique, and the doubling time was determined to be 32.7 hours. The basal ethoxyresorufin O-deethylase (EROD) activity, a measure of cytochrome P450 (CYP) 1A1 and 1A2, maintained a level of 9.6 ± 1.2 pmol resorufin/mg protein/min over time in culture, and was highly inducible. The expression of 84 Phase I murine metabolic enzyme genes was assessed using quantitative real-time PCR arrays following exposure to 33 µM β-naphthoflavone and 500 µM phenobarbital for 24 hours. The CYP 1A1 gene showed a 7750-fold induction over control, and several other noteworthy CYPs (i.e., CYP 1A2 and members of the CYP 2C family) were also significantly induced. Concentration-dependent increases in lacZ mutant frequency have been observed following exposures to benzo[a]pyrene (BaP), 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), aflatoxin B1 (AFL), and dimethylnitrosamine (DMN). Following this extensive characterization of morphology, karyotype, metabolic capacity and inducibility, proliferative ability, and performance for detection of chemical mutagens, it is apparent that primary hepatocytes from the Muta™Mouse show considerable promise for in vitro mutagenicity assessment.
Back to Top ↑
QSAR Modeling for Japan NIHS QSAR International Collaborative Study to Predict Ames Mutagenicity
Suman Chakravarti and Roustem Saiakhov
ICH-M7 is the first guideline to address the use of QSAR in place of an actual experiment for assessing carcinogenic risk of impurities in pharmaceuticals. Therefore, robust and highly predictive QSAR models are now a necessity for the prediction of Ames mutagenicity. The Division of Genetics and Mutagenesis, National Institute of Health Sciences, Japan, is collaborating with twelve QSAR builders around the world by providing high quality Ames mutagenicity data as training sets to improve QSAR models. MultiCASE Inc is a participant and we are reporting various aspects of model building and prediction using NIHS data. So far three batches of data has been released and after combining with our existing training data we were able to build models with very high prediction accuracy, e.g. we obtained 82% and 88% sensitivity and specificity respectively in predicting phase III chemicals. We will also discuss the importance of classification thresholds, ROC plots and effect of adding the new data to our existing training sets on prediction accuracy.
Back to Top ↑
Quantitative Relationships Between lacZ Mutagenic Potency and DNA Adduct Genotoxic Potency in Muta™Mouse Tissues Following Sub-Chronic Oral Exposure to Polycyclic Aromatic Hydrocarbons
Alexandra S. Long1, John W. Wills2, Christine L. Lemieux3, Volker M. Arlt4, Paul A. White2
1Department of Biology, University of Ottawa, Ottawa, ON; 2Mechanistic Studies Division, Health Canada, Ottawa, ON; 3New Substances Assessment Division, Health Canada, Ottawa; 4King’s College London, London, UK
It is well recognized that polycyclic aromatic hydrocarbons (PAHs) are mutagenic carcinogens that can induce mutations via the formation and processing of stable bulky DNA adducts. However, there is a paucity of information regarding the quantitative, empirical relationship between DNA adducts formation and mutation establishment across various tissues; likely due to the fact that it is very challenging to simultaneously make the required measurements (i.e., chemically-induced adduct and mutation frequency). In this study, we employed the transgenic Muta™Mouse to simultaneously quantify the frequency of DNA adducts and lacZ mutants across 5 somatic tissues, following sub-chronic oral exposure to 9 PAHs and 6 PAH mixtures. The frequency of stable DNA adducts was assessed in bone marrow, glandular stomach, small intestine, liver, and lung, and we had previously determined the frequency of lacZ mutants in these same animals and tissues. We then applied the benchmark dose (BMD)-approach to estimate the BMD100 (i.e., the doubling dose) and associated 90% confidence intervals for each compound/tissue combination. Using this metric, we examined the empirical relationship between mutagenic potency (i.e., lacZ BMD) and genotoxic potency (i.e., adduct BMD) across tissues, and compared the inverse of the slope of the linear regression between tissues, a metric termed the mutagenic efficiency of total adducts (i.e., the conversion of total bulky DNA adducts determined via 32P-postlabelling into lacZ mutants). We found a significant positive relationship between mutagenic potency and genotoxic potency when all tissues were combined. When each tissue was examined independently, we found significant positive relationships for liver, glandular stomach, and lung, and did not observe any significant differences in mutagenic efficiency between these three tissues (i.e., no significant difference in slopes of the mutation-adduct relationships). We did not find a significant quantitative relationship between mutagenic potency and genotoxic potency for bone marrow or small intestine. The work demonstrates significant quantitative relationships between the mutagenic potency and genotoxic potency of PAHs; however, it also highlights important tissue-specific differences in the mutagenic efficiency of PAHs. Although the work can be considered a significant contribution to improved mechanistic understanding regarding the formation PAH-induced mutations in vivo, it also raises numerous questions regarding tissue- and compound-specific differences in damage processing and mutation establishment.
Category: Student - Travel Award Recipient
Back to Top ↑
Validation of an In Vitro Mutagenicity Assay Based on Pulmonary Epithelial Cells From the Transgenic Muta™Mouse
Joleen Hanna1, Rebecca Maertens2, Paul A. White2
1Department of Biology, University of Ottawa, Ottawa, Ontario; 2Environmental Health Science and Research Bureau, Health CAN
Validation of toxicological test methods includes several discrete steps to determine assay performance and reliability. Specifically, protocol transferability studies, as well as inter- and intra-laboratory validation studies. Collectively, these efforts provide evidence of the assay’s reproducibility, reliability and applicability domain (i.e., the range and/or classes of test chemicals that can be reliably evaluated). Novel tools for genetic toxicity assessment are increasingly based on cultured cells, and the objective of this work is to validate the performance of an in vitro transgene mutagenicity assay based on an immortalised cell line, denoted FE1, derived from Muta™Mouse lung tissue. FE1 cells retain concatenated copies of the lacZ transgene in a lambda shuttle vector that can be retrieved for scoring chemically-induced mutations. Intra-laboratory screening of ECVAM (European Centre for the Validation of Alternative Methods) reference chemicals has appropriately identified known mutagens such as benzo[a]pyrene and N-ethyl-N-nitrosourea, yielding mutant frequency values up to 32-fold and 13-fold above background levels, respectively. False positive compounds such as phthalic anhydride and t-butylhydroquinone, as well as the known non-mutagens D-mannitol and ampicillin trihydrate, have also been appropriately classified. Mutation frequency values generated across different operators and test dates show little variability in benchmark dose (BMD) values, indicating that the assay is reliable and reproducible. The assay also demonstrates exceptional sensitivity and specificity for identification of mutagens, including those that require metabolic conversion to DNA reactive substances. In order to characterize the ability of FE1 cells to metabolize different types of compounds, the expression of 168 genes involved in Phase I and Phase II xenobiotic metabolism was assessed using quantitative real-time PCR following exposure to the known mutagens 1-methylpyrene (1-MP), 2-acetylaminofluroene (2-AAF), aflatoxin B1 (AFB1), 7, 12-dimethylbenz[a]anthracene (DMBA) and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK). Phase I cytochrome P450 (Cyp) isozymes are specifically required for the metabolic activation of DMBA, NNK and AFB1; the Phase II sulfotransferase and glucuronosyltransferases are required for activation of 1-MP and 2-AAF. Results indicate that FE1 cells possess the Phase I and Phase II metabolic capacity required for activation of some mutagenic compounds (i.e., DMBA), and specifically retain significant levels of Cyp’s, sulfotransferases, glucuronosyltransferases and glutathione S-transferases. The addition of Aroclor 1254 induced rat liver S9 is necessary for the metabolic activation of 2-AAF. This is likely due to deficiency of Cyp1a2 activity. The data obtained to date indicate that the FE1 cell transgene mutagenicity assay constitutes a reliable alternative to currently-used mammalian cell-based assays, and a logical candidate for an internationally-accepted OECD test guideline (Organization for Economic Co-operation and Development).
Category: Student - Travel Award Recipient
Back to Top ↑
Progress Toward the Development of an In Vitro Pig-a Gene Mutation Assay
Jeffrey C. Bemis1, Carson Labash1, Svetlana Avlasevich1, Page McKinzie2, Javier Revollo2, Vasily Dobrovolsky2, Stephen Dertinger1
1Litron Laboratories, Rochester NY, USA; 2National Center for Toxicological Research US FDA Jefferson
Our laboratories developed in vivo methods for the detection of glycosylphosphatidylinositol (GPI) anchor deficiency as a reporter of mutagenic potential of chemicals. As an extension of this work at Litron, we created an analogous in vitro methodology using mouse lymphoma L5178Y cells analyzed by flow cytometry. For our initial work, fluorescent antibodies against GPI-anchored cell surface protein CD90.2 were used to indicate Pig-a mutation, and fluorescent antibodies against CD45 guarded against non-specific cell surface alterations. Flow cytometry was used to evaluate cells for the presence or absence of CD45 and CD90, where CD45+/CD90- cells were considered mutant phenotype cells and CD45+/CD90+ cells were defined as wildtype. Analysis of L5178Y cells showed maximal recovery of genotoxicant-induced mutant phenotype cells at approximately 7 days post-exposure. The sensitivity of the assay was studied with the DNA-reactive genotoxicants N-ethyl-N-nitrosourea, ethyl methanesulfonate (EMS), methyl methanesulfonate, 1,3-propane sultone, 4-nitroquinoline 1-oxide and cisplatin. Specificity was considered with the presumed non-mutagenic agents phenformin hydrochloride, D-mannitol, dexamethasone, diethanolamine and cycloheximide. The five reference mutagens were observed to increase the frequency of mutant phenotype cells in a concentration-dependent manner, whereas none of the non-mutagenic compounds caused reproducible increases. As a second iteration of the method, we added DRAQ7 as a dead cell marker in order to eliminate these cells from analysis, and included use of an isotype control to guide instrument calibration. Our most recent work focused on confirmation of the genotype contributing to the GPI anchor deficiency. L5178Y cells were treated with EMS and 38 CD90.2-negative clones were isolated. These presumed mutant clones, along with a sample of wildtype cells, were sent to NCTR for gene sequencing which revealed mutations in the Pig-a gene for all CD90.2-negative clones studied, thus supporting the link between specific mutation at the Pig-a locus and the GPI-negative phenotype in mutant cells. These results show that in vitro Pig-a assays could represent a useful complement to established genotoxicity assays.
Back to Top ↑
Comparative In Vitro Genotoxicity of Multiple Samples of a Popular Botanical, Black Cohosh Extract
Stephanie L. Smith-Roe1, Carol D. Swartz2, Cheryl A. Hobbs2, Chad R. Blystone1, Suramya Waidyanatha1, Stavros Garantziotis3, Steven M. Bryce4, Stephen D. Dertinger4, Kristine L. Witt1
1Division of the National Toxicology Program, NIEHS, RTP, NC; 2Genetic and Molecular Toxicology Program, ILS, Inc., RTP, NC; 3Clinical Research Unit, NIEHS, RTP, NC; 4Litron Laboratories, Rochester, NY
Products containing black cohosh extract (BCE) are marketed for relief of gynecological ailments, yet the toxicity of BCE has not been well characterized. The National Toxicology Program (NTP) exposed female B6C3F1/N mice and female Wistar Han rats to BCE for 90 days via gavage. Significant, dose-related increases in micronucleated reticulocytes (MN-RET) and mature erythrocytes (MN-E) were observed in the peripheral blood of mice administered 62.5 - 1000 mg BCE/kg/day, and a significant increase in MN-RET was observed in the peripheral blood of female Wistar Han rats administered 15 - 1000 mg BCE/kg/day. Also, a dose-dependent, non-regenerative macrocytic anemia was observed in mice and rats. Follow-up studies confirmed significant dose-related increases in MN-RET and MN-E in female mice after 90 days or one year of exposure. Chromatographic screening of commercially available lots of BCE by the NTP demonstrated a diversity of chemical signatures. Due to this heterogeneity in composition, we used an in vitro micronucleus assay in human TK6 lymphoblastoid cells to examine whether genotoxicity is inherent to various cohoshes. The lot of BCE that was genotoxic in vivo was also genotoxic in vitro. Lots from other suppliers were also found to induce MN, and were genotoxic whether they had chemical signatures consistent with BCE, or a mixture of BCE and Chinese cohosh extract. A BCE standard reference material also induced MN in TK6 cells, albeit at higher concentrations than were required for the lot used by the NTP for in vivo testing. Both the lot of BCE that was genotoxic in vivo and the BCE standard reference material showed signatures indicative of aneugenicity in the Litron MultiFlowâ„¢ assay. Both MN and non-regenerative macrocytic anemia can result from disruption of the folate metabolism pathway. To determine whether the effects seen in mice and rats are also detected in humans, the NTP is assessing a variety of hematological endpoints including MN-RET frequencies, and folate and cobalamin levels in women who use BCE products.
Back to Top ↑
Oxidative Stress Contributes to the Interactive Genotoxicity and Hepatotoxicity of Titanium Dioxide Nanoparticles and Zinc Oxide Nanoparticles in Mice
Opeoluwa Fadoju1, Ogunsuyi Olusegun Ifeoluwa1, Akanni Olubukola Oyebimpe2, Alabi Okunola Aderenle3, Alimba Chibuisi Gideon1, Adaramoye Oluwatosin Adekunle2, Bakare Adekunle Akeem1
1Department of Zoology, University of Ibadan, Nigeria; 2Department of Biochemistry, University of Ibadan, Nigeria; 3Federal University of Technology Akure, Ondo State, Nigeria
Titanium dioxide (TiO2) nanoparticles (NPs) and Zinc oxide (ZnO) NPs are one of the most highly manufactured and utilized metal oxide NPs due to their novel physicochemical properties which include photocatalytic, anticorrosive and semi-conductive properties. Human exposures to these NPs do not only occur to the individual but also to the heterogeneous forms. There is dearth of information on in vivo interactive effects of TiO2 NPs and ZnO NPs regarding the genotoxicity, hepatotoxicity and mechanism of DNA damage. This study was therefore designed to investigate the genotoxicity and hepatotoxicity of the individual nanoparticles and their mixture (1:1) in mice. The LD50 was determined in male mice (6 - 8 weeks) according to OECD guidelines. Genotoxicity was carried out using the bone marrow micronucleus assay at concentrations of 9.38 - 150.00 milligram per kilogram body weight (mg/kg b.wt) for 5- and 10- day exposure periods via intraperitoneal exposure. Hepatotoxicity was evaluated via histopathology and assessment of the coefficient of liver, serum biochemical parameters [aspartate aminotransferase (AST) and alanine aminotransferase (ALT)] and oxidative stress parameters [superoxide dismutase (SOD), catalase (CAT), reduced glutathione (GSH) and malondialdehyde (MDA)]. The LD50 for ZnO NPs was 333.33 mg/kg b.wt and indeterminate for TiO2 NPs and their mixture. Only TiO2 NPs induced a significant (p < 0.05) increase in micronucleated polychromatic erythrocytes (MNPCE) at the two highest concentrations compared to the negative controls (distilled water) for the exposure periods. A significant difference in MNPCE induction was observed between the two exposure periods for TiO2 NPs and their mixture only. Concurrently, the frequency of normochromatic erythrocytes (NCE) in TiO2 NPs, ZnO NPs and their mixture showed a significant (p < 0.05) increase at all concentrations for both exposure periods. There was a significant difference in the induction of NCE between the two exposure periods for the NPs and their mixture. The coefficient of liver decreased across all concentrations at both exposure periods. Both NPs and their mixture induced a significant increase in AST at both exposure periods while ALT levels significantly (p < 0.01) increased at the 5- day and decreased at the 10- day exposure period. The effects of TiO2 NPs, ZnO NPs and their mixture on hepatic SOD, CAT, GSH and MDA activities were significantly (p < 0.001) altered compared to the control indicating oxidative stress. There were multiple foci of single-cell hepatocellular necrosis and dense aggregates of mononuclear inflammatory cells in the liver sections. The interaction factor for both NPs indicated synergism for the activities of SOD, CAT and GSH, and antagonism for the MDA levels at the 5- day exposure. In addition, the interaction factor indicated antagonism for the activities of SOD, CAT and MDA, and synergism for the GSH levels at the 10- day exposure. TiO2 NPs, ZnO NPs and their mixture may have the potential of interacting with the DNA, thereby inducing DNA damage and hepatic injury, which might be attributed to oxidative stress.
Back to Top ↑
Metal Accumulation, Cytogenotoxicity and Systemic Toxicity Assessment in Rats (Rattus Norvegicus) Exposed to Emissions and Groundwater from Olusosun Municipal Landfill Site, Nigeria
Adeyinka M. Gbadebo1, Okunola A. Alabi2, Chibuisi G. Alimba3, Adekunle A. Bakare3
1Ecology and Environmental Biology Unit, Department of Zoology, University of Ibadan, Ibadan, Nigeria; 2Department of Biology, FUTA, Akure, Nigeria; 3Department of Zoology, University of Ibadan, Ibadan, Nigeria
Human exposure to emissions from landfills in Nigeria has become a national issue due to inappropriate solid waste disposal into unsanitary landfills. The study herein investigated systemic toxicity and genome instability in rats exposed to emissions and underground water from a municipal landfill in Nigeria. Male Wistar albino rats (5/point/duration) were exposed at three different points to emissions and groundwater from the landfill for 4, 8, 12, 16, 20 and 24 weeks. Rats concurrently sited at about 17 km from the landfill site served as control. Metal (Pb, Cd, Cr, Cu and Zn) concentrations in blood, lungs, liver and kidneys; alterations in haematological indices and biochemical parameters in serum and liver homogenates, histopathology and bone marrow micronucleus formation were analysed. Pb, Cd, Cr, Cu and Zn (mg/L) increased in the examined blood and organs of rats at all points on the landfill compared to control. Haematological parameters (RBC, Hgb, PCV, WBC, lymphocyte and Neutrophil) were not significantly altered in exposed rats compared to control. Levels of AST, ALT, creatinine, urea, CAT, GSH and MDA significantly (p<0.05) increased, with decreased SOD in the exposed rats compared to the control. Frequencies of micronucleated polychromatic erythrocytes significantly (p<0.05) increased with decrease in polychromatic erythrocyte/normochromatic erythrocytes ratio at all exposure points compared to the control. Oedema, hyperplasia, inflammations, fibrosis, congestion, liver steatosis and sinusoid dilations were observed in at least one of the lungs, liver and kidneys of exposed rats. The observed alterations in the exposed rats may be related to the toxic effects of metals from emissions and groundwater and other unanalysed pollutants from the landfill site. This is of environmental and public health importance in Nigeria and other countries where populations are at risk due to occupational and residential exposure to landfill emissions.
Category: StudentBack to Top ↑
DNA Damage and Systemic Toxicity Induced by Silver and Copper Oxide Nanoparticles and Their Mixture in Clarias Gariepinus (Burchell, 1822)
Olusegun I. Ogunsuyi1, Fadoju, Opeoluwa M.1, Akanni, Olubukola O.2, Alabi, Okunola A.3, Alimba, Chibuisi G.1, Adaramoye, Oluwatosin A.3, Bakare, Adekunle A.1
1Department of Zoology University of Ibadan, Ibadan, Nigeria; 2Dept. of Biochemistry,University of Ibadan, Ibadan,Nigeria; 3Dept. of Biology, Federal University of Tech. Akure, Nigeria
Nanotechnology has recorded a remarkable advancement in designs and patronage which has led to tremendous increase in the production of nanoparticles (NPs). Primarily due to their small size (< 100nm), NPs have found application in virtually all spheres of human life which is evident in numerous biomedical, agricultural and consumer products as well as in environmental remediation processes. Silver (Ag) and copper oxide (CuO) nanoparticles are among the highly produced NPs due to their antimicrobial potentials. However, there are scientific concerns that their potential release into the environment may cause adverse human and environmental health effect. Also their co-existence as contaminant in the environment may lead to synergistic, antagonistic or additive toxicity. Hence, the DNA damaging effect of AgNPs, CuONPs and their mixture (1:1) was assessed in the mud catfish, Clarias gariepinus using the peripheral blood micronucleus (MN) assay. Cytotoxicity and probable mechanism of DNA damage were also examined via haematological analysis, histopathology of the liver and assessment of hepatic oxidative stress biomarkers [malondialdehyde (MDA), catalase (CAT), reduced glutathione (GSH) and superoxide dismutase (SOD)]. Juvenile C. gariepinus were sub-acutely exposed to five sub - lethal concentrations (6.25, 12.5, 25, 50 and 100 mg/L) of these NPs and their mixture for 28 days in a semi-static renewal system. AgNPs, CuONPs and their mixture caused significant (p < 0.05) increase in MN frequency as well as increase in other nuclear abnormalities (blebbed, binucleated and notched) compared to the control (dechlorinated tap water). Haematological analysis revealed reduction in red blood cell counts, haemoglobin concentration, packed cell volume and an increase in white blood cell counts in peripheral blood of exposed C. gariepinus. In addition, histopathology of the liver showed architectural distortions such as diffuse vacuolation of hepatocytes, vacuolar degeneration and atrophy as well as necrosis in concentration dependent manner. Concomitantly, the levels of MDA, GSH and SOD increased while CAT decreased in in the liver. Interaction analysis of data indicates antagonistic DNA damage, oxidative damage (MDA), GSH, and SOD of the two NPs while CAT was synergistic. These results suggest adverse genetic and toxic effects of exposure of aquatic organisms to AgNPs, CuONPs and their mixture. This is of environmental health importance and therefore there is need for proper regulation of the use of the NPs. Category: Student - Travel Award Recipient Back to Top ↑
Effect of Slide Positioning in Electrophoresis Chamber over Comet Assay Results
Marise Roy, Annie Hamel, Marilyn Registre, Renato Cardoso
Charles River Laboratories
The single cell gel electrophoresis (aka. comet assay), is commonly used in the regulatory battery of genotoxicity testing for different industries (eg. pharmaceutical, medical device, chemical, etc). Several parameters can influence the variability of the results and therefore can potentially impact the conclusion. Some examples of parameters that have been described to interfere with comet assay results include: agarose concentration, duration of alkaline incubation, and electrophoresis time, temperature, and voltage gradient, etc. Due to the voltage gradient, slides positioned in different sectors of the electrophoresis chamber could be subjected to different voltage. The present study was performed to evaluate how the position of the slide in the chamber impacted the tail intensity. Slides were prepared using lung cells from male Sprague-Dawley rats. Water (10 mL/kg/day) was used as the negative control and ethylmethanesulfonate (200 mg/kg/day) as the positive control. The route of administration was oral gavage. The electrophoresis box area was divided in two sides and three sections (top, middle and bottom). At least four slides of negative and positive control group were placed in each side or section of the box and 75 cells were analyzed for each slide. The average tail intensity was determined for each slide and the average tail intensity of all slides was used as a standard for comparison against the sections. To simulate the variability of a typical study design, permutations of 3 slides per section were used to determine the minimum (min) and maximum (max) values for region and the average deviation of the min and max values to the standard. A substantial difference was observed in the average tail intensity for the negative control slides positioned in different sections of electrophoresis chamber. The percent deviation to the standard was as follows: left side (30%), right side (-28%), top (20%) middle (3%) bottom (-35%). In the case of the positive control, the deviation to the standard was within Â±10% for all areas. To determine if distributing the slides of a group through the chamber would suffice to minimize deviations, permutations with mixed regions were performed. The mixed groups of 3 slides consisted of one slide from each section and from one opposing side. The average deviation of the group to the standard was 1.7% for negative control and -3.6% for positive control. In conclusion, the variability between positive controls was not big enough to impact the determination of a positive result, however the variability of a negative control (30% to -35%) may deteriorate the confidence of negative or equivocal response. Ensuring the heterogeneous distributions of slides from different treatment groups was sufficient to drop the deviation to acceptable levels.
Back to Top ↑
Current Activities of the HESI Genetic Toxicology Technical Committee
Jennifer Y. Tanir1, Maik Schuler2, Jan van Benthem3, Paul White4
1ILSI Health and Environmental Sciences Institute; 2Pfizer, Inc.; 3RIVM; 4Health Canada
The Genetic Toxicology Technical Committee (GTTC) of the International Life Sciences Institute (ILSI) Health and Environmental Sciences Institute (HESI) has a mission to advance the field of genetic toxicology and human risk assessment through the international collaboration of experts. The projects center around three main objectives. The first objective is to integrate genetic toxicology into risk assessment and decision-making for protection of human health. The group has worked on a conceptual framework for a next-generation testing strategy for assessment of genomic damage and case studies are being compiled for further illustration. In another project, dose-response modeling tools for practical use have been developed, which allow for the quantitative evaluation of genetic toxicology data and could be integrated into regulatory decision making. A new project is evaluating and recommending mode of action (MOA) approaches for regulatory decision making that identify genotoxic mechanisms in mammalian cells. The second objective is to improve new and existing test guidelines and to provide guidance on strategies as well as interpretation of results. For example, another new project aims to increase our knowledge of which in vivo tests to choose to follow-up in vitro positive results. Data for 91 chemicals are being compared from the transgenic rodent, in vivo comet, and cancer assay. With relation to the Pig-a gene mutation assay, a Detailed Review Paper (DRP) is being drafted with the ultimate goal of developing an OECD test guideline. The last objective is to examine the genomic risks associated with novel entities like nanomaterials and novel technologies like targeted genome editing.
Back to Top ↑
Stem Cell Divisions, Somatic Mutations, Cancer Etiology, and Cancer Prevention
Lu Li1, Bert Vogelstein3, Cristian Tomasetti11,2
1Johns Hopkins University School of Medicine; 2Johns Hopkins University Bloomberg School of Public Health; 3Johns Hopkins Kimmel Cancer Center
The role of environmental factors (E) and heredity (H) in cancer causation has been confirmed in multiple studies. We recently hypothesized that mutations occurring randomly as a result of the normal processes associated with cellular replication (R) are also an important contributor to cancer and can explain why cancers occur much more commonly in some tissues than others.
To determine what fractions of cancer-causing mutations result from E, H, or R, we developed a novel approach based on the integration of genome-wide sequencing and epidemiological data. When normalized for the incidence of each of 32 cancer types, the application of this methodology yields that 29% of the mutations in cancers occurring in the UK are attributable to E, 5% of the mutations are attributable to H, and therefore 66% are presumably due to R. The estimate for the proportion that may be attributed to R varies considerably: it is less than 40% in cancers such as those of the lung, esophagus, and skin and 80% or more in cancers such as those of the prostate, brain, and breast.
Given the mathematical relationship between cancer etiology and cancer preventability, the proportion of mutations caused by environmental factors is always less than the proportion of cancers preventable by avoidance of these factors. Thus, our estimate that a maximum of 29% of the mutations in these cancers are due to E is perfectly compatible with the estimate that 42% of these cancers are preventable by avoiding known risk factors.
Back to Top ↑
Retrospective Evaluation of Genetic Toxicity Assays Performed in WuXi AppTec
Liwen Gao, Chunhua Xu, Xueping Xu, Biao Xu, Millie Chen
WuXi AppTec (Suzhou) Co., Ltd., Suzhou, China
When DNA is exposed to particular chemicals, mutations and other damage can occur leading to cancer and/or teratogenic effects. The severity of these effects then necessitates examining whether new or existing chemicals intended for human use have any effect on DNA. Since no single test is capable of detecting all relevant genotoxic endpoints, a battery of in vitro and in vivo tests for genotoxicity is recommended by regulatory agencies. In WuXi AppTec, we offered screening and regulatory genotoxicity assays. Screening battery includes mini-Ames assay and microwell micronucleus assay (MNvit); regulatory battery usually includes Ames, in vitro chromosomal aberration (CAA) and in vivo micronucleus assays. To better understand these assays, we retrospectively evaluated the genetic toxicology data generated in WuXi AppTec.
In our laboratory, of 358 different classes of drug candidates tested mini-Ames assay, about 12.0% were positive or equivocal. 22 compounds were tested in both mini-Ames and Ames assays, with concordant results for 21 of them. Of 124 different classes of drug candidates tested in MNvit in Chinese hamster ovary (CHO-WBL) cells, about 12.0% were positive or equivocal, the ratio is close to the overall frequency of positive and equivocal results (13.6%) in CAA in CHO-WBL cells.
The overall frequency of positive results in Ames was 5.1%, comprising 4 positive results of a total of 78 compounds tested, 3 of positive were anti-cancer candidates and another has altering structures.
The overall frequency of positive and equivocal results in CAA in CHO-WBL cells was 13.6%, comprising 4 positive and 2 equivocal results of a total of 44 compounds tested, 3 of the positive results could be explained based on known mechanisms, i.e., PARP inhibitors; one of the equivocal results was for anti-cancer candidate. Only one of a total of 22 compounds tested in CAA in HPBL was positive with unknown mechanism. Of 54 candidates tested in rat bone marrow micronucleus, about 9.3% were positive, all positive results could be explained based on known mechanisms such as PARP inhibitors and cytotoxic agents, targeting on microtubule.
Based on the above analysis, the screening battery has a good predictive capability with those of the standard assays. In the regulatory battery performed in WuXi AppTec, most of studies are negative or positive/equivocal but can be explained based on known mechanisms, even for in vitro cytogenetic assay, less than 5% studies need to be followed up with the second in vivo assays such as Comet.
Category: Early Stage Investigator
Back to Top ↑
De-risking a Positive Mouse Lymphoma Assay: Implications for Clinical Development
John B. Phelps, Vincent L. Reynolds, Todd J. Page
Eli Lilly and Company
A small molecule pharmaceutical agent intended for a chronic use, non-life-threatening indication was previously evaluated in Phase 1 clinical trials outside the US. Results of completed genetic toxicology studies included negative findings in a gene mutation assay in bacteria, an in vitro chromosomal aberration assay, a rodent bone marrow micronucleus test, and an in vivo unscheduled DNA synthesis (UDS) assay in rat liver. A positive response occurred in the mouse lymphoma assay (MLA). Following the filing of an IND to enable Phase 2 clinical studies, FDA requested an in vivo comet assay to de-risk the positive MLA; notably, this was not a clinical hold issue. The comet assay was conducted in rats and evaluated the liver, stomach, and peripheral blood lymphocytes (PBLs). Results from the comet assay included small, but statistically significant increases in tail intensity and moment in the liver in all treated groups. This finding resulted in the ongoing clinical trial being halted. In a confirmatory trial evaluating liver only, small, statistically significant increases in tail intensity and tail moment occurred but only at the high dose. Because the range of values for tail intensity and moment from the high dose overlapped with the concurrent control, the comet study was concluded to be equivocal but not biologically relevant in the liver; negative in the stomach; and negative in PBLs. Several factors can have a role in the clinical development of a molecule, in this case, a positive MLA finding that was not a clinical hold issue, leading to a clinical hold based on the initial Comet data, and then the subsequent equivocal Comet outcome were all significant events.
Back to Top ↑
Development of the In Vivo Alkaline Comet Assay for Use With Sprague-Dawley Rat Skin
Farida Merah, G. Marceau, M Ghoussoub, S. Lavallée, N Boudjadja, C. Gauthier, A Nelson
ITR Laboratories Canada Inc., Montreal, Canada
The in vivo alkaline comet assay (Single cell Gel Electrophoresis) is used for the detection of DNA strand breaks in the nuclei of cells isolated from different animal tissues. It is part of the genetic toxicology battery for the detection of potential genotoxic effect of substances. Here we report on the reproducibility of the comet assay in rat skin tissue and building a historical control database.
Five independent experiments were performed with Sprague-Dawley rats. Each experiment included two groups. One group was treated with water (negative control) and a second group was treated twice with a positive control substance, ethyl methanesulphonate (200 mg/kg, 10 mL/kg), at 24 and 3 hours before euthanasia. All the animals were treated via oral gavage. At termination a section (3x3 cm) of dorsal skin was shaved and removed. The skin was sliced into strips, immersed in a cold EDTA/PBS solution, transferred into a trypsin/HBSS solution overnight at 4Â°C and then rinsed again in cold PBS. The epidermis was peeled from the dermis and transferred into a cold RPMI with 10% FBS solution. The epidermal strips were gently stirred to prepare single cell suspensions, which were then passed through cell strainers. The isolated cells were laid over agar-coated slides, lysed, and subjected to electrophoresis. The prepared slides were then stained with ethidium bromide and evaluated under a fluorescent microscope. The DNA migration was measured with the Comet Assay IV scoring system (Perceptive Instruments) and % DNA intensity was compared between groups within each experiment.
The results obtained indicated that there were significant increases in DNA migration in the positive controls versus the negative controls (P< 0.01); and the results were reproducible between the five experiments. The mean values of the negative and positive controls were (19.91, 37.74) and (13.03, 32.47), (7.37, 33.22), (12.49, 44.29) and (15.75, 39.33) in the five experiments.
In summary, this technique can be used to prepare single cell population to perform Comet assay on skin for dermal studies in rats.
Back to Top ↑
Implementation of the Comet Assay in Eight Target Organs and Tissues of the Sprague-Dawley Rat
Farida Merah, G. Marceau, M. Ghoussoub, S. Lavallée, N. Boudjadja, C. Gauthier, A. Nelson
ITR Laboratories Canada Inc., Montreal, Canada
The comet assay utilizes single-cell gel electrophoresis for the detection of DNA strand breaks in the nuclei of cells isolated from a variety of animal tissues. It is used as part of the genetic toxicology battery of tests to identify potential genotoxic compounds in pre-clinical research. Various target organs may be assessed for genotoxic analysis as required by the nature of the studied test item, target tissue, dosing method and metabolism. Here we detail the implementation of the comet assay in eight target organs or tissues of rats for the purpose of in vivo genotoxicity assessment.
Two groups of Sprague-Dawley rats were treated, in six independent replicates at 24 hours intervals of each other. In each replicate, the first group of animals was treated twice by oral gavage with ethyl methanesulphonate (EMS) at 200 mg/kg and the second group of animals treated with water by the same route. Tissues were collected (liver, stomach, kidney and skin in the first three replicates; and brain, spleen, lung and bladder in the last three replicates) in mincing buffer. Single cell suspensions were prepared, embedded in agarose on glass slides, lysed, and then subjected to alkaline unwinding and electrophoresis. The migrated DNA slides were then randomized, and stained with ethidium bromide before scoring. The DNA migration was measured with Comet Assay IV (Perceptive Instruments) scoring system and % DNA intensity was compared between groups within each replicate.
The mean percentage values of DNA intensity obtained in this study were significantly increased (p<0.01) between negative (water) controls (from 6.2% to 11.9%) and EMS-treated tissues (from 33.2% to 38.9%). More specifically, DNA intensity increased in the liver from 7.1% to 33.2%, in the lung from 6.2% to 36.0%, in the stomach from 7.6% to 33.4%, in the kidney from 6.6% to 33.3%, in the brain from 7.6% to 35.5%, in the skin from 11.9% to 38.9%, in the bladder from 8.1% to 36.1% and in the spleen from 7.6% to 35.0%.
The procedures used in this study are reliable and the organs selected are sensitive to genotoxic compounds such as EMS at 200 mg/kg which induced positive response in all these organs.
Back to Top ↑
High-Throughput Platform for Detection of DNA Adducts Induced by Metabolic Activation of Xenobiotics
Le P. Ngo1, John Winters2, Carol Swartz2, Jing Ge1, Leslie Recio2, Leona Samson1, Bevin Engelward1
11Massachusetts Institute of Technology; 2Toxicology Program, Integrated Laboratory Systems, Inc.
The comet assay is a well-known sensitive method used for measuring DNA damage that can lead to or can be converted to strand breaks in DNA. The disadvantages of the comet assay lie in its low-throughput nature, poor reproducibility, and image processing and analysis methods that are prone to human errors. Recently, CometChip was invented in the Engelward laboratory to address the above challenges of the conventional in vitro comet assay. CometChip increase throughput by ~100 fold and also increases sensitivity. In addition, the CometChip prevents bias that can be introduced when comets are selected by eye.
One blind spot of the current application of the comet assay is its poor sensitivity to detect bulky DNA adducts, such as DNA adducts induced by the environmental procarcinogen benzo(a)pyrene (B[a]P). Here, we describe work aimed at optimizing conditions for detection of bulky DNA damage by the use of DNA repair synthesis inhibitors (hydroxyurea and cytosine arabinoside) to prevent completion of nucleotide excision repair (NER) of bulky DNA adducts. Inhibition of NER enables the accumulation of the strand-break intermediates that are readily detectable with alkaline CometChip (the HU-AraC approach).
A major drawback of current in vitro genotoxicity assays is the lack of an appropriate cell model that can provide human-relevant metabolic activity. In the human body, xenobiotics are extensively metabolized by hepatocytes in the liver. This process can result in reactive intermediates that can form adducts with DNA, which may lead to mutations and tumorigenesis. It is, therefore, essential to assess genotoxic potential of chemicals in human-relevant metabolic conditions. The hepatoma cell line HepaRG expresses broad-spectrum physiologically relevant levels of Phase I and Phase II enzymes that are key in human xenobiotic. HepaRG is potentially an appropriate human cell model for genotoxic assessment.
We have applied the HU-AraC approach using the HepaRG cells to enable the detection of bulky DNA adducts induced by bioactivation. We have optimized conditions for our method using benzo[a]pyrene (B[a]P) and aflatoxin B1 (AFB1) as model DNA damaging agents that require bioactivation to be reactive with DNA. Using CYP450 inhibitors, we have validated that DNA damage detected using our approach is dependent upon metabolic activation of B[a]P and AFB1. Importantly, inhibiting NER initiation resulted in a reduction in the detected signal, confirming the mechanism of DNA damage detection of our approach is via NER strand-break intermediates.
Back to Top ↑
A Comparison of BALB/c 3T3 and CHO-WBL Cell Lines in the Neutral Red Uptake Assay
Sara B. Hurtado, Leon F. Stankowski, Vincent Y. Kwok
Charles River Laboratories, Skokie, IL
Charles River - Skokie, has established methods for the Neutral Red Uptake Assay (NRU) using BALB/c 3T3 cells in accordance with the OECD Test Guideline 129 as well as Chinese Hamster Ovary (CHO) - WBL cells in accordance with the Health Canada Test Guidance T-502. The NRU Assay is used to generate in vitro cytotoxicity data for a variety of chemicals, cigarette smoke fractions, e-liquids and medical devices.
Briefly, BALB/c 3T3 and CHO-WBL cells are exposed to a range of test article concentrations in a 96-well plate for 24 or 48 hours respectively. At the end of exposure, the test article is removed and neutral red (NR) is added to the cells for an incubation of 3-hours. Viable cells will incorporate and bind the NR, such that cytotoxicity is expressed as a concentration-dependent reduction in the uptake of NR. Following incubation, the NR media is removed, cells are washed, and a desorb media is added to extract the NR from the cells. Absorption is measured at 540 nm. For comparison; the data generated is fitted with a 4-parameter curve, and the IC50, the concentration at which 50% cytotoxicity is observed, is calculated. Absorbance fractions were also calculated as a percent of the negative control to provide the relative absorbance in accordance with the Health Canada Methods.
Wells were seeded at a density of 1x104 cells/well. BALB/c 3T3 cells were exposed for 48 hours to the positive control, sodium lauryl sulfate (SLS). To accommodate the faster doubling time (12-14 hours for CHO-WBL compared to 18-22 hours for BALB/c 3T3), the exposure period of the CHO-WBL cells was 24 hours. Both cell lines underwent approximately two doublings in the presence of SLS. Eight concentrations of SLS were examined in the BALB/c 3T3 cell line ranged from 6.8 to 100 µg/mL using a 1.21 dilution factor. To obtain a similar cytotoxicity curve as with the BALB/c 3T3 cells the SLS in the CHO-WBL NRU assay was increased to 13.6 to 200 µg/mL. In total over 10 trials in each cell line were conducted.
The concentrations tested produced a cytotoxicity fitted dose-response curve with an R2 (coefficient of determination) of ≥ 0.85 for the Hill model fit. Based on this model the two cell lines produced significantly different IC50 values for SLS. BALB/c 3T3 cells were more sensitive to the effects of SLS with an average IC50 of 19.99 ± 3.35 µg/mL. CHO-WBL cells provided a more robust response with an average IC50 of 89.15 ± 3.38 µg/mL. The average absorbance displayed in the vehicle control wells varied by trial, but overall was approximately the same between the two cell lines, suggesting similar growth, final cell density and similar uptake of the NR dye across the two cell lines.
Back to Top ↑
Genotoxicity Assessment of Azoxymethane Using Micronucleus and the Pig-a Assay
Javed A. Bhalli, Cameron Tebbe, Michelle Chan, Jessica Noteboom, Demetria Thomas
Covance Laboratories Inc, 671 S. Meridian Rd, Greenfield, IN
Azoxymethane (AOM) is a potent genotoxic carcinogen. It is a DNA alkylating agent used in biological research. Azoxymethane (AOM) induces colon cancer in experimental animals; in a mechanism that is mediated by glutathione (GSH) depletion in colonic cells of rats.
To assess the Azoxymethane genotoxicity in rodents, Pig-a and micronucleus assay was performed in rats. Four groups of six Sprague Dawley male rats were dosed i.p. for three consecutive days with vehicle (0.9% saline), 2.50, 5.00 or 10.0 mg/kg/day of Azoxymethane. Doses were selected based on the dose range finding assay and 10.0 mg/kg/day dose was identified as the maximum tolerated dose (MTD). Approximately 100 µl of blood was collected for the Pig-a assay on Days -1, 15, 29, and 42 and for the micronucleus assay on Day 4.
Analysis of micronucleated reticulocytes (MN-RETs) by flow cytometry showed a statistically significant increase compared to the concurrent vehicle control at the highest tested dose of 10 mg/kg/day. No statistically significant increase in the RBCs and RETs mutant frequencies was observed at any tested dose level at any evaluated time point.
In conclusion, Azoxymethane was tested positive in the micronucleus assay and negative in the Pig-a assay when evaluated in rats up to the MTD levels.
Back to Top ↑
Bhas-42 Cell Transformation Assay Historical Control Data
Shannon W. Bruce, Kamala Pant, Jamie E. Sly, Michelle Klug LaForce, Edgar Andrus, Rohan Kulkarni
1Carcinogen-DNA Interactions Section, LCBG, CCR; 2Methods and Technologies Branch, DCPC, NCI, BioReliance by SAFC, Rockville, MD, USA
The Bhas-42 cell transformation assay (CTA) is a sensitive short term system for predicting chemical carcinogenicity. In recent years this assay has become popular in the testing of tobacco products. Bhas-42 cells were established from BALB/c 3T3 cells by the transfection of v-Ha-ras gene and postulated to be initiated in the two-stage carcinogenesis theory. The assay protocol consists of two parts, the initiator assay (Sasaki, et.al. 1988, 1990, Asada, et.al. 2005) and promoter assay (Ohmori, et.al.2004, 2005) to detect tumor-initiating and promoting activity, respectively, of chemical carcinogens. The Bhas-42 CTA uses the Bhas-42 cells and measures induction of morphologically transformed (MT) foci that show invasive growth into the monolayer of surrounding contact-inhibited cells. There are many improvements in the Bhas-42 CTA over the earlier version of BALB/c 3T3 CTA. The timing of the assay as well as the number of replicates needed have been reduced. The procedures for the initiator and promoter assays are different and utilize different number of cells. The treatment duration is three days in the initiator assay and ten days in the promoter assays. The positive control used for the Bhas-42 initiator assay is 3-methylcholanthrene (3MCA) at 1.0 µg/mL and for the Bhas-42 promoter assay is 12-O-tetradecanoylphorbol-13-acetate (TPA) at 50 ng/mL. Vehicle controls include, but are not limited to, cell growth medium, dimethyl sulfoxide (DMSO) and acetone. BioReliance has been performing this assay for more than a decade and has accumulated an extensive historical database for the Bhas-42 CTA. In the Bhas-42 initiator assay the frequency of mean MT foci are 4.4 and 26.2 for the vehicle and positive controls, respectively. In the Bhas-42 promoter assay the mean values are 5.1 and 22.99 for the vehicle and positive controls, respectively. These data are from the years 2009 through 2016. In this poster, we will present the vehicle and positive control data with 95% control limits to show consistent results over the years for both the initiator and promoter Bhas-42 CTAs.
Category: Early Stage Investigator
Back to Top ↑