Michael Römer

Cross-Platform Toxicogenomics for the Prediction of Non-Genotoxic Hepatocarcinogenesis in Rat

In the area of omics profiling in toxicology, i.e. toxicogenomics, characteristic molecular profiles have previously been incorporated into prediction models for early assessment of a carcinogenic potential and mechanism-based classification of compounds. Traditionally, the biomarker signatures used for model construction were derived from individual high-throughput techniques, such as microarrays designed for monitoring global mRNA expression. In this study, we built predictive models by integrating omics data across complementary microarray platforms and introduced new concepts for modeling of pathway alterations and molecular interactions between multiple biological layers. We trained and evaluated diverse machine learning-based models, differing in the incorporated features and learning algorithms on a cross-omics dataset encompassing mRNA, miRNA, and protein expression profiles obtained from rat liver samples treated with a heterogeneous set of substances. Most of these compounds could be unambiguously classified as genotoxic carcinogens, non-genotoxic carcinogens, or non-hepatocarcinogens based on evidence from published studies. Since mixed characteristics were reported for the compounds Cyproterone acetate, Thioacetamide, and Wy-14643, we reclassified these compounds as either genotoxic or non-genotoxic carcinogens based on their molecular profiles. Evaluating our toxicogenomics models in a repeated external cross-validation procedure, we demonstrated that the prediction accuracy of our models could be increased by joining the biomarker signatures across multiple biological layers and by adding complex features derived from cross-platform integration of the omics data. Furthermore, we found that adding these features resulted in a better separation of the compound classes and a more confident reclassification of the three undefined compounds as non-genotoxic carcinogens.

Tumor promotion and inhibition by phenobarbital in livers of conditional Apc-deficient mice

Activation of Wnt/β-catenin signaling is important for human and rodent hepatocarcinogenesis. In mice, the tumor promoter phenobarbital (PB) selects for hepatocellular tumors with activating β-catenin mutations via constitutive androstane receptor activation. PB-dependent tumor promotion was studied in mice with genetic inactivation of Apc, a negative regulator of β-catenin, to circumvent the problem of randomly induced mutations by chemical initiators and to allow monitoring of PB- and Wnt/β-catenin-dependent tumorigenesis in the absence of unknown genomic alterations. Moreover, the study was designed to investigate PB-induced proliferation of liver cells with activated β-catenin. PB treatment provided Apc-deficient hepatocytes with only a minor proliferative advantage, and additional connexin 32 deficiency did not affect the proliferative response. PB significantly promoted the outgrowth of Apc-deficient hepatocellular adenoma (HCA), but simultaneously inhibited the formation of Apc-deficient hepatocellular carcinoma (HCC). The probability of tumor promotion by PB was calculated to be much lower for hepatocytes with loss of Apc, as compared to mutational β-catenin activation. Comprehensive transcriptomic and phosphoproteomic characterization of HCA and HCC revealed molecular details of the two tumor types. HCC were characterized by a loss of differentiated hepatocellular gene expression, enhanced proliferative signaling, and massive over-activation of Wnt/β-catenin signaling. In conclusion, PB exerts a dual role in liver tumor formation by promoting the growth of HCA but inhibiting the growth of HCC. Data demonstrate that one and the same compound can produce opposite effects on hepatocarcinogenesis, depending on context, highlighting the necessity to develop a more differentiated view on the tumorigenicity of this model compound.

Dysregulated serum response factor triggers formation of hepatocellular carcinoma

The ubiquitously expressed transcriptional regulator serum response factor (SRF) is controlled by both Ras/MAPK (mitogen‐activated protein kinase) and Rho/actin signaling pathways, which are frequently activated in hepatocellular carcinoma (HCC). We generated SRF‐VP16iHep mice, which conditionally express constitutively active SRF‐VP16 in hepatocytes, thereby controlling subsets of both Ras/MAPK‐ and Rho/actin‐stimulated target genes. All SRF‐VP16iHep mice develop hyperproliferative liver nodules that progresses to lethal HCC. Some murine (m)HCCs acquire Ctnnb1 mutations equivalent to those in human (h)HCC. The resulting transcript signatures mirror those of a distinct subgroup of hHCCs, with shared activation of oncofetal genes including Igf2, correlating with CpG hypomethylation at the imprinted Igf2/H19 locus. Conclusion: SRF‐VP16iHep mHCC reveal convergent Ras/MAPK and Rho/actin signaling as a highly oncogenic driver mechanism for hepatocarcinogenesis. This suggests simultaneous inhibition of Ras/MAPK and Rho/actin signaling as a treatment strategy in hHCC therapy. (Hepatology 2015;61:979–989)

Evaluation of Toxicogenomics Approaches for Assessing the Risk of Nongenotoxic Carcinogenicity in Rat Liver

The current gold-standard method for cancer safety assessment of drugs is a rodent two-year bioassay, which is associated with significant costs and requires testing a high number of animals over lifetime. Due to the absence of a comprehensive set of short-term assays predicting carcinogenicity, new approaches are currently being evaluated. One promising approach is toxicogenomics, which by virtue of genome-wide molecular profiling after compound treatment can lead to an increased mechanistic understanding, and potentially allow for the prediction of a carcinogenic potential via mathematical modeling. The latter typically involves the extraction of informative genes from omics datasets, which can be used to construct generalizable models allowing for the early classification of compounds with unknown carcinogenic potential. Here we formally describe and compare two novel methodologies for the reproducible extraction of characteristic mRNA signatures, which were employed to capture specific gene expression changes observed for nongenotoxic carcinogens. While the first method integrates multiple gene rankings, generated by diverse algorithms applied to data from different subsamplings of the training compounds, the second approach employs a statistical ratio for the identification of informative genes. Both methods were evaluated on a dataset obtained from the toxicogenomics database TG-GATEs to predict the outcome of a two-year bioassay based on profiles from 14-day treatments. Additionally, we applied our methods to datasets from previous studies and showed that the derived prediction models are on average more accurate than those built from the original signatures. The selected genes were mostly related to p53 signaling and to specific changes in anabolic processes or energy metabolism, which are typically observed in tumor cells. Among the genes most frequently incorporated into prediction models were Phlda3, Cdkn1a, Akr7a3, Ccng1 and Abcb4.

ToxDBScan: Large-scale similarity screening of toxicological databases for drug candidates.

We present a new tool for hepatocarcinogenicity evaluation of drug candidates in rodents. ToxDBScan is a web tool offering quick and easy similarity screening of new drug candidates against two large-scale public databases, which contain expression profiles for substances with known carcinogenic profiles: TG-GATEs and DrugMatrix. ToxDBScan uses a set similarity score that computes the putative similarity based on similar expression of genes to identify chemicals with similar genotoxic and hepatocarcinogenic potential. We propose using a discretized representation of expression profiles, which use only information on up- or down-regulation of genes as relevant features. Therefore, only the deregulated genes are required as input. ToxDBScan provides an extensive report on similar compounds, which includes additional information on compounds, differential genes and pathway enrichments. We evaluated ToxDBScan with expression data from 15 chemicals with known hepatocarcinogenic potential and observed a sensitivity of 88%. Based on the identified chemicals, we achieved perfect classification of the independent test set. ToxDBScan is publicly available from the ZBIT Bioinformatics Toolbox.

Xenobiotic CAR Activators Induce Dlk1-Dio3 Locus Noncoding RNA Expression in Mouse Liver

Derisking xenobiotic-induced nongenotoxic carcinogenesis (NGC) represents a significant challenge during the safety assessment of chemicals and therapeutic drugs. The identification of robust mechanism-based NGC biomarkers has the potential to enhance cancer hazard identification. We previously demonstrated Constitutive Androstane Receptor (CAR) and WNT signaling-dependent up-regulation of the pluripotency associated Dlk1-Dio3 imprinted gene cluster noncoding RNAs (ncRNAs) in the liver of mice treated with tumor-promoting doses of phenobarbital (PB). Here, we have compared phenotypic, transcriptional ,and proteomic data from wild-type, CAR/PXR double knock-out and CAR/PXR double humanized mice treated with either PB or chlordane, and show that hepatic Dlk1-Dio3 locus long ncRNAs are upregulated in a CAR/PXR-dependent manner by two structurally distinct CAR activators. We further explored the specificity of Dlk1-Dio3 locus ncRNAs as hepatic NGC biomarkers in mice treated with additional compounds working through distinct NGC modes of action. We propose that up-regulation of Dlk1-Dio3 cluster ncRNAs can serve as an early biomarker for CAR activator-induced nongenotoxic hepatocarcinogenesis and thus may contribute to mechanism-based assessments of carcinogenicity risk for chemicals and novel therapeutics.

Influence of Feature Encoding and Choice of Classifier on Disease Risk Prediction in Genome-Wide Association Studies.

Various attempts have been made to predict the individual disease risk based on genotype data from genome-wide association studies (GWAS). However, most studies only investigated one or two classification algorithms and feature encoding schemes. In this study, we applied seven different classification algorithms on GWAS case-control data sets for seven different diseases to create models for disease risk prediction. Further, we used three different encoding schemes for the genotypes of single nucleotide polymorphisms (SNPs) and investigated their influence on the predictive performance of these models. Our study suggests that an additive encoding of the SNP data should be the preferred encoding scheme, as it proved to yield the best predictive performances for all algorithms and data sets. Furthermore, our results showed that the differences between most state-of-the-art classification algorithms are not statistically significant. Consequently, we recommend to prefer algorithms with simple models like the linear support vector machine (SVM) as they allow for better subsequent interpretation without significant loss of accuracy.

Epidermal Growth Factor Represses Constitutive Androstane Receptor Expression in Primary Human Hepatocytes and Favors Regulation by Pregnane X Receptor

Growth factors have key roles in liver physiology and pathology, particularly by promoting cell proliferation and growth. Recently, it has been shown that in mouse hepatocytes, epidermal growth factor receptor (EGFR) plays a crucial role in the activation of the xenosensor constitutive androstane receptor (CAR) by the antiepileptic drug phenobarbital. Due to the species selectivity of CAR signaling, here we investigated epidermal growth factor (EGF) role in CAR signaling in primary human hepatocytes. Primary human hepatocytes were incubated with CITCO, a human CAR agonist, or with phenobarbital, an indirect CAR activator, in the presence or absence of EGF. CAR-dependent gene expression modulation and PXR involvement in these responses were assessed upon siRNA-based silencing of the genes that encode CAR and PXR. EGF significantly reduced CAR expression and prevented gene induction by CITCO and, to a lower extent, by phenobarbital. In the absence of EGF, phenobarbital and CITCO modulated the expression of 144 and 111 genes, respectively, in primary human hepatocytes. Among these genes, only 15 were regulated by CITCO and one by phenobarbital in a CAR-dependent manner. Conversely, in the presence of EGF, CITCO and phenobarbital modulated gene expression only in a CAR-independent and PXR-dependent manner. Overall, our findings suggest that in primary human hepatocytes, EGF suppresses specifically CAR signaling mainly through transcriptional regulation and drives the xenobiotic response toward a pregnane X receptor (PXR)-mediated mechanism.

ZBIT Bioinformatics Toolbox: A Web-Platform for Systems Biology and Expression Data Analysis

Bioinformatics analysis has become an integral part of research in biology. However, installation and use of scientific software can be difficult and often requires technical expert knowledge. Reasons are dependencies on certain operating systems or required third-party libraries, missing graphical user interfaces and documentation, or nonstandard input and output formats. In order to make bioinformatics software easily accessible to researchers, we here present a web-based platform. The Center for Bioinformatics Tuebingen (ZBIT) Bioinformatics Toolbox provides web-based access to a collection of bioinformatics tools developed for systems biology, protein sequence annotation, and expression data analysis. Currently, the collection encompasses software for conversion and processing of community standards SBML and BioPAX, transcription factor analysis, and analysis of microarray data from transcriptomics and proteomics studies. All tools are hosted on a customized Galaxy instance and run on a dedicated computation cluster. Users only need a web browser and an active internet connection in order to benefit from this service. The web platform is designed to facilitate the usage of the bioinformatics tools for researchers without advanced technical background. Users can combine tools for complex analyses or use predefined, customizable workflows. All results are stored persistently and reproducible. For each tool, we provide documentation, tutorials, and example data to maximize usability. The ZBIT Bioinformatics Toolbox is freely available at https://webservices.cs.uni-tuebingen.de/.

MARCARviz: Interactive web-platform for exploratory analysis of toxicogenomics data for nongenotoxic hepatocarcinogenesis

The late detection of non-genotoxic carcinogens in the drug development process can delay drug candidates for unmet medical needs from reaching the market despite considerable investments in their development. To enable faster, safer, and less expensive development of medications for patients, the MARCAR project generated a large set of transcriptomic data to investigate the underlying mechanisms of non-genotoxic hepatocarcinogenesis and to identify potential biomarkers for early detection of tumor formation in the rodent liver. The effective mining of these high-dimensional datasets is a non-trivial task that usually requires bioinformatics support to extract relevant mechanistic patterns and confirm toxicological hypotheses. Here, we present MARCARviz, a web-platform that enables biologists to (a) quickly address the most common questions associated with the MARCAR microarray data, to (b) identify relevant patterns in the data, and to (c) generate or confirm mechanistic hypotheses about non-genotoxic effects leading to cancer formation. The major advantage of MARCARviz is that there is no software or advanced technical knowledge required to perform powerful analyses and generate visualizations of the MARCAR data. MARCARviz greatly facilitates the confirmation of published MARCAR results and generation of new insights from the collected data by the greater public without the requirement for tedious pre-processing steps. MARCARviz is publicly available from https://tea.cs.uni-tuebingen.de/.