Molly Vallant

Predicting the hepatocarcinogenic potential of alkenylbenzene flavoring agents using toxicogenomics and machine learning.

Identification of carcinogenic activity is the primary goal of the 2-year bioassay. The expense of these studies limits the number of chemicals that can be studied and therefore chemicals need to be prioritized based on a variety of parameters. We have developed an ensemble of support vector machine classification models based on male F344 rat liver gene expression following 2, 14 or 90 days of exposure to a collection of hepatocarcinogens (aflatoxin B1, 1-amino-2,4-dibromoanthraquinone, N-nitrosodimethylamine, methyleugenol) and non-hepatocarcinogens (acetaminophen, ascorbic acid, tryptophan). Seven models were generated based on individual exposure durations (2, 14 or 90 days) or a combination of exposures (2+14, 2+90, 14+90 and 2+14+90 days). All sets of data, with the exception of one yielded models with 0% cross-validation error. Independent validation of the models was performed using expression data from the liver of rats exposed at 2 dose levels to a collection of alkenylbenzene flavoring agents. Depending on the model used and the exposure duration of the test data, independent validation error rates ranged from 47% to 10%. The variable with the most notable effect on independent validation accuracy was exposure duration of the alkenylbenzene test data. All models generally exhibited improved performance as the exposure duration of the alkenylbenzene data increased. The models differentiated between hepatocarcinogenic (estragole and safrole) and non-hepatocarcinogenic (anethole, eugenol and isoeugenol) alkenylbenzenes previously studied in a carcinogenicity bioassay. In the case of safrole the models correctly differentiated between carcinogenic and non-carcinogenic dose levels. The models predict that two alkenylbenzenes not previously assessed in a carcinogenicity bioassay, myristicin and isosafrole, would be weakly hepatocarcinogenic if studied at a dose level of 2 mmol/kg bw/day for 2 years in male F344 rats; therefore suggesting that these chemicals should be a higher priority relative to other untested alkenylbenzenes for evaluation in the carcinogenicity bioassay. The results of the study indicate that gene expression-based predictive models are an effective tool for identifying hepatocarcinogens. Furthermore, we find that exposure duration is a critical variable in the success or failure of such an approach, particularly when evaluating chemicals with unknown carcinogenic potency.

Transcriptional profiling of the left and median liver lobes of male F344/N rats following exposure to acetaminophen

The liver is a common organ for transcriptional profiling because of its role in xenobiotic metabolism and because hepatotoxicity is a common response to chemical exposure. To explore the impact that sampling different lobes may have on transcriptional profiling experiments we have examined and compared gene expression profiles of the left and median lobes of livers from male F344 rats exposed to toxic and nontoxic doses of acetaminophen. Transcript profiling using micorarrays revealed clear differences in the response of the left and median liver lobes of F344 rats to acetaminophen exposure both at low doses as well as doses that caused hepatotoxicity. Differences were found in the total number of differentially expressed genes in the left and median lobes, the number and identity of genes that were differentially expressed uniquely only in the left or median lobe, and in the patterns of gene expression. While it is not possible to generalize these results to compounds other than acetaminophen or other strains of rat, these results highlight the potential impact of sampling differences on the interpretation of gene expression profiles in the liver.

Characterization of polybrominated diphenyl ether toxicity in Wistar Han rats and use of liver microarray data for predicting disease susceptibilities.

The toxicity of polybrominated diphenyl ethers (PBDEs), flame-retardant components, was characterized in offspring from Wistar Han dams exposed by gavage to a PBDE mixture (DE71) starting at gestation day 6 and continuing to weaning on postnatal day (PND) 21. Offspring from the dams underwent PBDE direct dosing by gavage at the same dose as their dams from PND 12 to PND 21, and then after weaning for another thirteen weeks. Liver samples were collected at PND 22 and week 13 for liver gene expression analysis (Affymetrix Rat Genome 230 2.0 Array). Treatment with PBDE induced 1,066 liver gene transcript changes in females and 1,200 transcriptional changes in males at PND 22 (false discovery rate < 0.01), but only 263 liver transcriptional changes at thirteen weeks in male rats (false discovery rate < 0.05). No significant differences in dose response were found between male and female pups. Transcript changes at PND 22 coded for proteins in xenobiotic, sterol, and lipid metabolism, and cell cycle regulation, and overlapped rodent liver transcript patterns after a high-fat diet or phenobarbital exposure. These findings, along with the observed PBDE-induced liver hypertrophy and vacuolization, suggest that long-term PBDE exposure has the potential to modify cell functions that contribute to metabolic disease and/or cancer susceptibilities.

An Ethanolic Extract of Black Cohosh Causes Hematological Changes but Not Estrogenic Effects in Female Rodents

Black cohosh rhizome (Actaea racemosa) is used as a remedy for pain and gynecological ailments; modern preparations are commonly sold as ethanolic extracts available as dietary supplements. Black cohosh was nominated to the National Toxicology Program (NTP) for toxicity testing due to its widespread use and lack of safety data. Several commercially available black cohosh extracts (BCE) were characterized by the NTP, and one with chemical composition closest to formulations available to consumers was used for all studies. Female B6C3F1/N mice and Wistar Han rats were given 0, 15 (rats only), 62.5 (mice only), 125, 250, 500, or 1000 mg/kg/day BCE by gavage for 90 days starting at weaning. BCE induced dose-dependent hematological changes consistent with a non-regenerative macrocytic anemia and increased frequencies of peripheral micronucleated red blood cells (RBC) in both species. Effects were more severe in mice, which had decreased RBC counts in all treatment groups and increased micronucleated RBC at doses above 125 mg/kg. Dose-dependent thymus and liver toxicity was observed in rats but not mice. No biologically significant effects were observed in other organs. Puberty was delayed 2.9 days at the highest treatment dose in rats; a similar magnitude delay in mice occurred in the 125 and 250 mg/kg groups but not at the higher doses. An additional uterotrophic assay conducted in mice exposed for 3 days to 0.001, 0.01, 0.1, 1, 10, 100 and 500 mg/kg found no estrogenic or anti-estrogenic activity. These are the first studies to observe adverse effects of BCE in rodents.

Repeated dose toxicity and relative potency of 1,2,3,4,6,7-hexachloronaphthalene (PCN 66) 1,2,3,5,6,7-hexachloronaphthalene (PCN 67) compared to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) for induction of CYP1A1, CYP1A2 and thymic atrophy in female Harlan Sprague-Dawley rats

In this study we assessed the relative toxicity and potency of the chlorinated naphthalenes 1,2,3,4,6,7-hexachloronaphthalene (PCN 66) and 1,2,3,5,6,7-hexachloronaphthalene (PCN 67) relative to that of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Chemicals were administered in corn oil:acetone (99:1) by gavage to female Harlan Sprague-Dawley rats at dosages of 0 (vehicle), 500, 1500, 5000, 50,000 and 500,000 ng/kg (PCN 66 and PCN 67) and 1, 3, 10, 100, and 300 ng/kg (TCDD) for 2 weeks. Histopathologic changes were observed in the thymus, liver and lung of TCDD treated animals and in the liver and thymus of PCN treated animals. Significant increases in CYP1A1 and CYP1A2 associated enzyme activity were observed in all animals exposed to TCDD, PCN 66 and PCN 67. Dose response modeling of CYP1A1, CYP1A2 and thymic atrophy gave ranges of estimated relative potencies, as compared to TCDD, of 0.0015-0.0072, for PCN 66 and 0.00029-0.00067 for PCN 67. Given that PCN 66 and PCN 67 exposure resulted in biochemical and histopathologic changes similar to that seen with TCDD, this suggests that they should be included in the WHO toxic equivalency factor (TEF) scheme, although the estimated relative potencies indicate that these hexachlorinated naphthalenes should not contribute greatly to the overall human body burden of dioxin-like activity.

Variation in the Hepatic Gene Expression in Individual Male Fischer Rats

A new tool beginning to have wider application in toxicology studies is transcript profiling using microarrays. Microarrays provide an opportunity to directly compare transcript populations in the tissues of chemical-exposed and unexposed animals. While several studies have addressed variation between microarray platforms and between different laboratories, much less effort has been directed toward individual animal differences especially among control animals where RNA samples are usually pooled. Estimation of the variation in gene expression in tissues from untreated animals is essential for the recognition and interpretation of subtle changes associated with chemical exposure. In this study hepatic gene expression as well as standard toxicological parameters were evaluated in 24 rats receiving vehicle only in 2 independent experiments. Unsupervised clustering demonstrated some individual variation but supervised clustering suggested that differentially expressed genes were generally random. The level of hepatic gene expression under carefully controlled study conditions is less than 1.5-fold for most genes. The impact of individual animal variability on microarray data can be minimized through experimental design.

Developmental Neurotoxicity of 3,3',4,4'-Tetrachloroazobenzene with Thyroxine Deficit: Sensitivity of Glia and Dentate Granule Neurons in the Absence of Behavioral Changes

Thyroid hormones (TH) regulate biological processes implicated in neurodevelopmental disorders and can be altered with environmental exposures. Developmental exposure to the dioxin-like compound, 3,3’,4,4’-tetrachloroazobenzene (TCAB), induced a dose response deficit in serum T4 levels with no change in 3,5,3’- triiodothyronine or thyroid stimulating hormone. Female Sprague-Dawley rats were orally gavaged (corn oil, 0.1, 1.0, or 10 mg TCAB/kg/day) two weeks prior to cohabitation until post-partum day 3 and male offspring from post-natal day (PND)4-21. At PND21, the high dose showed a deficit in body weight gain. Conventional neuropathology detected no neuronal death, myelin disruption, or gliosis. Astrocytes displayed thinner and less complex processes at 1.0 and 10 mg/kg/day. At 10 mg/kg/day, microglia showed less complex processes, unbiased stereology detected fewer hippocampal CA1 pyramidal neurons and dentate granule neurons (GC) and Golgi staining of the cerebellum showed diminished Purkinje cell dendritic arbor. At PND150, normal maturation of GC number and Purkinje cell branching area was not observed in the 1.0 mg/kg/day dose group with a diminished number and branching suggestive of effects initiated during developmental exposure. No effects were observed on post-weaning behavioral assessments in control, 0.1 and 1.0mg/kg/day dose groups. The demonstrated sensitivity of hippocampal neurons and glial cells to TCAB and T4 deficit raises support for considering additional anatomical features of brain development in future DNT evaluations.

Effects of the PPARα Agonist and Widely Used Antihyperlipidemic Drug Gemfibrozil on Hepatic Toxicity and Lipid Metabolism

Gemfibrozil is a widely prescribed hypolipidemic agent in humans and a peroxisome proliferator and liver carcinogen in rats. Three-month feed studies of gemfibrozil were conducted by the National Toxicology Program (NTP) in male Harlan Sprague-Dawley rats, B6C3F1 mice, and Syrian hamsters, primarily to examine mechanisms of hepatocarcinogenicity. There was morphologic evidence of peroxisome proliferation in rats and mice. Increased hepatocyte proliferation was observed in rats, primarily at the earliest time point. Increases in peroxisomal enzyme activities were greatest in rats, intermediate in mice, and least in hamsters. These studies demonstrate that rats are most responsive while hamsters are least responsive. These events are causally related to hepatotoxicity and hepatocarcinogenicity of gemfibrozil in rodents via peroxisome proliferator activated receptor-α (PPARα) activation; however, there is widespread evidence that activation of PPARα in humans results in expression of genes involved in lipid metabolism, but not in hepatocellular proliferation.

Toxicology and Carcinogenesis Study of Senna in C3B6.129F1-Trp53 tm1Brd N12 Haploinsufficient Mice

Senna is a pod or leaf of Senna alexandrina P. Mill and is used as a stimulant laxative. In the large intestine, bacterial enzymes reduce sennosides to rhein-9-anthrone, the active form for the laxative effect. To determine the potential toxic effects of senna, a 5-week dose range finding study in the C57BL/6N mouse and a 40-week toxicology and carcinogenesis study in the C3B6.129F1-Trp53 tm1Brd N12 haploinsufficient (p53+/−) mouse were conducted. In the 5-week study, C57BL/6N mice were exposed to up to 10,000 ppm senna in feed. Increased incidences of epithelial hyperplasia of the cecum and colon were observed in males and females exposed to 5,000 or 10,000 ppm senna. These intestinal lesions were not considered to be of sufficient severity to cause mortality and, thus, in the p53+/− mouse 40-week study, the high dose of 10,000 ppm was selected. Significant increases in the incidences of epithelial hyperplasia of the colon and cecum were observed at 10,000 ppm in p53+/− males and females, and the incidence of hyperplasia of the colon was significantly increased at 3,000 ppm in females. In conclusion, the large intestine was the major target of senna-induced toxicity in both wild-type and the p53+/− mouse model. There was no neoplastic change when senna was administered to p53+/− mouse.

Comparative phenotypic assessment of cardiac pathology, physiology, and gene expression in C3H/HeJ, C57BL/6J, and B6C3F1/J mice.

Human cardiomyopathies often lead to heart failure, a major cause of morbidity and mortality in industrialized nations. Described here is a phenotypic characterization of cardiac function and genome-wide expression from C3H/HeJ, C57BL/6J, and B6C3F1/J male mice. Histopathologic analysis identified a low-grade background cardiomyopathy (murine progressive cardiomyopathy) in eight of nine male C3H/HeJ mice (age nine to ten weeks), but not in male C57BL/6J and in only of ten male B6C3F1/J mice. The C3H/HeJ mouse had an increased heart rate and a shorter RR interval compared to the B6C3F1/J and C57BL/6J mice. Cardiac genomic studies indicated the B6C3F1/J mice exhibited an intermediate gene expression phenotype relative to the 2 parental strains. Disease-centric enrichment analysis indicated a number of cardiomyopathy-associated genes were induced in B6C3F1/J and C3H/HeJ mice, including Myh7, My14, and Lmna and also indicated differential expression of genes associated with metabolic (e.g., Pdk2) and hypoxic stress (e.g. Hif1a). A novel coexpression and integrated pathway network analysis indicated Prkaa2, Pdk2, Rhoj, and Sgcb are likely to play a central role in the pathophysiology of murine progressive cardiomyopathy in C3H/HeJ mice. Our studies indicate that genetically determined baseline differences in cardiac phenotype have the potential to influence the results of cardiotoxicity studies.