Helga Gerets

Characterization of primary human hepatocytes, HepG2 cells, and HepaRG cells at the mRNA level and CYP activity in response to inducers and their predictivity for the detection of human hepatotoxins

In the pharmaceutical industry, improving the early detection of drug-induced hepatotoxicity is essential as it is one of the most important reasons for attrition of candidate drugs during the later stages of drug development. The first objective of this study was to better characterize different cellular models (i.e., HepG2, HepaRG cells, and fresh primary human hepatocytes) at the gene expression level and analyze their metabolic cytochrome P450 capabilities. The cellular models were exposed to three different CYP450 inducers; beta-naphthoflavone (BNF), phenobarbital (PB), and rifampicin (RIF). HepG2 cells responded very weakly to the different inducers at the gene expression level, and this translated generally into low CYP450 activities in the induced cells compared with the control cells. On the contrary, HepaRG cells and the three human donors were inducible after exposure to BNF, PB, and RIF according to gene expression responses and CYP450 activities. Consequently, HepaRG cells could be used in screening as a substitute and/or in complement to primary hepatocytes for CYP induction studies. The second objective was to investigate the predictivity of the different cellular models to detect hepatotoxins (16 hepatotoxic and 5 nonhepatotoxic compounds). Specificity was 100% with the different cellular models tested. Cryopreserved human hepatocytes gave the highest sensitivity, ranging from 31% to 44% (depending on the donor), followed by lower sensitivity (13%) for HepaRG and HepG2 cells (6.3%). Overall, none of the models under study gave desirable sensitivities (80–100%). Consequently, a high metabolic capacity and CYP inducibility in cell lines does not necessarily correlate with a high sensitivity for the detection of hepatotoxic drugs. Further investigations are necessary to compare different cellular models and determine those that are best suited for the detection of hepatotoxic compounds.

The Use of Real-Time Cell Analyzer Technology in Drug Discovery Defining Optimal Cell Culture Conditions and Assay Reproducibility with Different Adherent Cellular Models

The use of impedance-based label-free technology applied to drug discovery is nowadays receiving more and more attention. Indeed, such a simple and noninvasive assay that interferes minimally with cell morphology and function allows one to perform kinetic measurements and to obtain information on proliferation, migration, cytotoxicity, and receptor-mediated signaling. The objective of the study was to further assess the usefulness of a real-time cell analyzer (RTCA) platform based on impedance in the context of quality control and data reproducibility. The data indicate that this technology is useful to determine the best coating and cellular density conditions for different adherent cellular models including hepatocytes, cardiomyocytes, fibroblasts, and hybrid neuroblastoma/neuronal cells. Based on 31 independent experiments, the reproducibility of cell index data generated from HepG2 cells exposed to DMSO and to Triton X-100 was satisfactory, with a coefficient of variation close to 10%. Cell index data were also well reproduced when cardiomyocytes and fibroblasts were exposed to 21 compounds three times (correlation >0.91, p < 0.0001). The data also show that a cell index decrease is not always associated with cytotoxicity effects and that there are some confounding factors that can affect the analysis. Finally, another drawback is that the correlation analysis between cellular impedance measurements and classical toxicity endpoints has been performed on a limited number of compounds. Overall, despite some limitations, the RTCA technology appears to be a powerful and reliable tool in drug discovery because of the reasonable throughput, rapid and efficient performance, technical optimization, and cell quality control.

Selection of cytotoxicity markers for the screening of new chemical entities in a pharmaceutical context: A preliminary study using a multiplexing approach

The present study was undertaken to validate a battery of cytotoxicity assays performed in a multiplex format to screen pharmaceutical compounds at an early stage of drug development. Two experiments were performed on HepG2 cells and the parameters were measured in 96-well plates. Biological and technical triplicates were performed to evaluate the reproducibility of the assay. In the first experiment, HepG2 cells were exposed to tamoxifen, staurosporine, phenobarbital and triton X-100 for 2 and 24h. The following nine cytotoxicity parameters were analyzed, cell viability, lactate dehydrogenase (LDH), adenosine triphosphate (ATP), caspase-3/7, aspartate aminotransferase (AST), glutamate dehydrogenase (GLDH), alanine aminotransferase (ALT), alkaline phosphatase (ALP) and alpha-glutathione-S-transferase (alpha-GST). In the second experiment, HepG2 cells were exposed to doxorubicin, t-butyl hydroperoxide, ferrous sulfate and sulfamoxole for 2 and 24h. Based on the results of the first experiment, six cytotoxicity parameters were selected for further evaluation (cell viability, ATP, LDH, caspase, AST and GLDH). ALT (activity always below detection limit), ALP (no response to drug treatment) and alpha-GST (too labor intensive and not possible to multiplex) were eliminated. The analysis of the data revealed that the reproducibility of the assays was accurate according to principal component analysis. Our data also clearly indicated that the potential of this battery of selected assays measured in a multiplex format not only made it possible to rank and select the most promising drug candidates based on their cytotoxic potential, but also to gather information that may help to understand some of the toxic events occurring in the cells.

Predictivity of dog co-culture model, primary human hepatocytes and HepG2 cells for the detection of hepatotoxic drugs in humans

Drug induced liver injury (DILI) is a major cause of attrition during early and late stage drug development. Consequently, there is a need to develop better in vitro primary hepatocyte models from different species for predicting hepatotoxicity in both animals and humans early in drug development. Dog is often chosen as the non-rodent species for toxicology studies. Unfortunately, dog in vitro models allowing long term cultures are not available. The objective of the present manuscript is to describe the development of a co-culture dog model for predicting hepatotoxic drugs in humans and to compare the predictivity of the canine model along with primary human hepatocytes and HepG2 cells. After rigorous optimization, the dog co-culture model displayed metabolic capacities that were maintained up to 2 weeks which indicates that such model could be also used for long term metabolism studies. Most of the human hepatotoxic drugs were detected with a sensitivity of approximately 80% (n=40) for the three cellular models. Nevertheless, the specificity was low approximately 40% for the HepG2 cells and hepatocytes compared to 72.7% for the canine model (n=11). Furthermore, the dog co-culture model showed a higher superiority for the classification of 5 pairs of close structural analogs with different DILI concerns in comparison to both human cellular models. Finally, the reproducibility of the canine system was also satisfactory with a coefficient of correlation of 75.2% (n=14). Overall, the present manuscript indicates that the dog co-culture model may represent a relevant tool to perform chronic hepatotoxicity and metabolism studies.

Evaluation of Impedance-Based Label-Free Technology as a Tool for Pharmacology and Toxicology Investigations

The use of label-free technologies based on electrical impedance is becoming more and more popular in drug discovery. Indeed, such a methodology allows the continuous monitoring of diverse cellular processes, including proliferation, migration, cytotoxicity and receptor-mediated signaling. The objective of the present study was to further assess the usefulness of the real-time cell analyzer (RTCA) and, in particular, the xCELLigence platform, in the context of early drug development for pharmacology and toxicology investigations. In the present manuscript, four cellular models were exposed to 50 compounds to compare the cell index generated by RTCA and cell viability measured with a traditional viability assay. The data revealed an acceptable correlation (ca. 80%) for both cell lines (i.e., HepG2 and HepaRG), but a lack of correlation (ca. 55%) for the primary human and rat hepatocytes. In addition, specific RTCA profiles (signatures) were generated when HepG2 and HepaRG cells were exposed to calcium modulators, antimitotics, DNA damaging and nuclear receptor agents, with a percentage of prediction close to 80% for both cellular models. In a subsequent experiment, HepG2 cells were exposed to 81 proprietary UCB compounds known to be genotoxic or not. Based on the DNA damaging signatures, the RTCA technology allowed the detection of ca. 50% of the genotoxic compounds (n = 29) and nearly 100% of the non-genotoxic compounds (n = 52). Overall, despite some limitations, the xCELLigence platform is a powerful and reliable tool that can be used in drug discovery for toxicity and pharmacology studies.

The automated micronucleus assay for early assessment of genotoxicity in drug discovery.

Recent publications on the automated in vitro micronucleus assay show predictive values higher than 85% for the classification of in vitro aneugens, clastogens and non-genotoxic compounds. In the present work, the CHO-k1 micronucleus assay in combination with cellular imaging was further evaluated. Firstly, the effect of a range of S9 concentrations on micronucleus formation and cytotoxicity was investigated. Subsequently, the reproducibility and predictivity of the micronucleus assay on CHO-k1 cells was investigated with a set of four compounds. Then, a larger set of compounds (n=44) was tested on CHO-k1 cells and inter-laboratory correlation was calculated. Finally, cellular imaging was compared with flow cytometry for in vivo assessment of micronucleus formation. The concentration of S9 had a significant impact on micronucleus formation and cytotoxicity. In addition, calculations of relative cell count (RCC) and cytokinesis-block proliferation index (CBPI) showed to be complementary to cytotoxicity assessment. The CHO-k1 micronucleus assay correctly classified the four reference compounds, with a dose-response relationship and low variability. Based on a larger set of compounds, the assay proved to be reliable with a sensitivity of 94% (n=31) and a specificity of 85% (n=13). A correlation coefficient of 97% was obtained when the lowest observable adverse effect levels (LOAELs) from our study were compared with those published by Diaz et al. (2007) [10]. In conclusion, the in vitro CHO-k1 micronucleus assay combined with cellular imaging is a predictive assay appropriate for genotoxicity screening at early stages of drug development. In addition, for in vivo assessment of micronucleus formation, we preferred to use flow cytometry rather than cell imaging.

Multiplexing Cell Viability Assays

Today, obtaining mechanistic insights into biological, toxicological, and pathological processes is of upmost importance. Researchers aim to obtain as many as possible data from one cell sample to understand the biological processes under study. Multiplexing, which is the ability to gather more than one set of data from the same sample, fulfills completely this objective. Obviously, multiplexing has several advantages compared to single plex experiments and probably the most important one is that data on various parameters at exactly the same time point on the same cells or group of cells can be obtained and consequently this may contribute to saving time and effort and a reduction of the costs.In this chapter, different endpoints were measured starting from two-seeded multiwell plates, namely, cell viability, caspase-3/7 activity, lactate dehydrogenase (LDH), adenosine triphosphate (ATP), aspartate aminotransferase (AST), and glutamate dehydrogenase (GLDH) measurements. These -different endpoints were analyzed together to determine the cytotoxic properties of pharmaceutical compounds and/or reference compounds. A 96-well plate was designed to allow appropriate measurement of five doses of a compound in triplicate to determine the effect of the compound on the six different endpoints. The first four endpoints (cell viability, caspase-3/7 activity, LDH, and ATP) are discussed in detail in this chapter. AST and GLDH measurements are not discussed in detail as these are fully automatic measurements and thus behind the scope of this chapter.As an illustrating example, the reference compound tamoxifen was used to evaluate its cytotoxic properties using the hepatocellular carcinoma cell line HepG2 cells.

Comparison of a genomic and a multiplex cell imaging approach for the detection of phospholipidosis

Phospholipidosis (PLD) is a topic of concern in drug development because it may be associated with toxicological consequences. The aim of the study was to determine the best method to screen proprietary compounds with regard to their potential to induce PLD. Two in vitro approaches, a genomic method previously evaluated in our laboratory and a fluorescent cell based approach to detect PLD were compared using HepG2 cells. The same set of reference compounds (15 PLD inducing, 7 non-PLD inducing and 4 additional compounds) were used to compare both approaches. The same sensitivity (15/15) and similar specificity (7/7 versus, 6/7 for the genomic approach) were obtained. In addition, 11 proprietary compounds were tested in 4-day exploratory rat toxicity studies as well as in both in vitro approaches. Two of the 11 compounds induced alveolar foamy macrophages and lung vacuolization in vivo and were considered as PLD inducers. Sensitivity (2/2) and specificity (7/9) were better with the fluorescent method than with the genomic approach (1/2 and 3/9, respectively). In conclusion, compared to the genomic approach, the fluorescent method is the test of choice for screening compounds at an early stage of drug development. Indeed, the fluorescent method is more adapted to medium throughput, detects the positive reference compounds at lower (8/15) or equal (7/15) concentrations, allows multiplexing and is associated with higher sensitivity and specificity to predict lung foamy macrophages and vacuolization in vivo. Nevertheless, to confirm our conclusion, it would be necessary to compare the predictivity of both in vitro approaches by using a wide range of reference and proprietary compounds, with information on their potential to induce PLD under in vivo conditions.

Investigative safety strategies to improve success in drug development

Understanding and reducing attrition rate remains a key challenge in drug development. Preclinical and clinical safety issues still represent about 40% of drug discontinuation, of which cardiac and liver toxicities are the leading reasons. Reducing attrition rate can be achieved by various means, starting with a comprehensive evaluation of the potential safety issues associated to the primary target followed by an evaluation of undesirable secondary targets. To address these risks, a risk mitigation plan should be built at very early development stages, using a panel of in silico , in vitro , and in vivo models. While most pharmaceutical companies have developed robust safety strategies to de-risk genotoxicity and cardiotoxicity issues, partly driven by regulatory requirements; safety issues affecting other organs or systems, such as the central nervous system, liver, kidney, or gastro-intestinal system are less commonly addressed during early drug development. This paper proposes some de-risking strategies that can be applied to these target organ systems, including the use of novel biomarkers that can be easily integrated in both preclinical and clinical studies. Experiments to understand the mechanisms’ underlying toxicity are also important. Two examples are provided to demonstrate how such mechanistic studies can impact drug development. Novel trends in investigative safety are reviewed, such as computational modeling, mitochondrial toxicity assessment, and imaging technologies. Ultimately, understanding the predictive value of non-clinical safety testing and its translatability to humans will enable to optimize assays in order to address the key objectives of the drug discovery process, i.e., hazard identification, risk assessment, and mitigation.

In vitro screening of cell bioenergetics to assess mitochondrial dysfunction in drug development

Drug-induced mitochondrial toxicity is considered as a common cellular mechanism that can induce a variety of organ toxicities. In the present manuscript, 17 in vitro mitochondrial toxic drugs, reported to induce Drug-Induced Liver Injury (DILI) and 6 non-mitochondrial toxic drugs (3 with DILI and 3 without DILI concern), were tested in HepG2 cells using a bioenergetics system. The 17 mitochondrial toxic drugs represent a wide variety of mitochondrial dysfunctions as well as DILI and include 4 pairs of drugs which are structurally related but associated with different DILI concerns in human. Cell bioenergetics were measured using the XF96e analyzer which simultaneous monitor oxygen consumption rate (OCR) and extracellular acidification rate (ECAR), indirect measurements of oxidative phosphorylation and glycolysis, respectively. OCR associated with ATP production, maximal respiration, proton leak and spare respiratory capacity, were also assessed. Duplicate experiments resulted in a sensitivity of 82% (14/17) and specificity of 83% (5/6). The addition of stressors improved specificity considerably. Cut-offs, statistics and rules are clearly discussed to facilitate the use of this assay for screening purposes. Overall, the authors consider that this assay should be part of the battery of safety screening assays at early stages of drug development.