In vitro toxicological characterisation of the antifungal compound soybean toxin (SBTX).

Soybean toxin (SBTX) is a protein isolated from soybean seeds and composed of two polypeptide subunits (17 and 27 kDa). SBTX has in vitro activity against phytopathogenic fungi such as Cercospora sojina, Aspergillus niger, and Penicillium herguei, and yeasts like Candida albicans, C. parapsilosis, Kluyveromyces marxiannus, and Pichia membranifaciens. The present study aimed to analyze in vitro whether SBTX causes any side effects on non-target bacterial and mammalian cells that could impede its potential use as a novel antifungal agent. SBTX at 100 µg/mL and 200 µg/mL did not hinder the growth of the bacteria Salmonella enterica (subspecies enterica serovar choleraesuis), Bacillus subtilis (subspecies spizizenii) and Staphylococcus aureus. Moreover, SBTX at concentrations up to 500 µg/mL did not significantly affect the viability of erythrocytes, neutrophils, and human intestinal Caco-2 cells. To study whether SBTX could induce relevant alterations in gene expression, in vitro DNA microarray experiments were conducted in which differentiated Caco-2 cells were exposed for 24 h to 100 µg/mL or 200 µg/mL SBTX. SBTX up-regulated genes involved in cell cycle and immune response pathways, but down-regulated genes that play a role in cholesterol biosynthesis and platelet degranulation pathways. Thus, although SBTX did not affect bacteria, nor induced cytotoxity in mammalian cells, it affected some biological pathways in the human Caco-2 cell line that warrants further investigation.

Detection and profiling of diarrheic marine biotoxins in shellfish by mRNA analysis of exposed Caco-2 cells using qRT-PCR and multiplex magnetic bead-based assays.

The mouse bioassay for the detection of marine biotoxins in shellfish products is 40 years old and still in use. A full ban or total replacement of this in vivo test has been postponed because of the fear that current chemical-based detection methods could miss a new emerging toxin. In order to fully replace the mouse bioassay, more efforts are needed on the search for functional assays with specific endpoints. Gene expression elicited by diarrheic shellfish poisons in Caco-2 cells allowed us to determine three 'DSP profiles', i.e. OA/DTX, AZA-YTX and PTX profiles. Twelve marker genes were selected to envision the three profiles. qRT-PCR is relatively cheap and easy, and although its multiplex capacity is limited to 5 genes, this turned out to be sufficient to show the three expected profiles. The use of the multiplex magnetic bead-based assay turned out to be even a slightly better alternative, allowing the use of all twelve selected marker genes and 2 reference genes, and resulting in clear profiles with for some genes even higher induction factors as obtained by qRT-PCR. When analysing blank and contaminated shellfish samples with this multiplex magnetic bead-based assay, the contaminated samples could easily be distinguished from the blank samples, showing the expected profiles. This work is one step further on the final replacement of the mouse bioassay, e.g. by combining the neuro-2a bioassay for screening and detection with analytical chemical analyses and the multiplex magnetic bead-based assay for confirmation of known and unknown toxins respectively.

A Strategy to Replace the Mouse Bioassay for Detecting and Identifying Lipophilic Marine Biotoxins by Combining the Neuro-2a Bioassay and LC-MS/MS Analysis.

Marine biotoxins in fish and shellfish can cause several symptoms in consumers, such as diarrhea, amnesia, or even death by paralysis. Monitoring programs are in place for testing shellfish on a regular basis. In some countries testing is performed using the so-called mouse bioassay, an assay that faces ethical concerns not only because of animal distress, but also because it lacks specificity and results in high amounts of false positives. In Europe, for lipophilic marine biotoxins (LMBs), a chemical analytical method using LC-MS/MS was developed as an alternative and is now the reference method. However, safety is often questioned when relying solely on such a method, and as a result, the mouse bioassay might still be used. In this study the use of a cell-based assay for screening, i.e., the neuro-2a assay, in combination with the official LC-MS/MS method was investigated as a new alternative strategy for the detection and quantification of LMBs. To this end, samples that had been tested previously with the mouse bioassay were analyzed in the neuro-2a bioassay and the LC-MS/MS method. The neuro-2a bioassay was able to detect all LMBs at the regulatory levels and all samples that tested positive in the mouse bioassay were also suspect in the neuro-2a bioassay. In most cases, these samples contained toxin levels (yessotoxins) that explain the outcome of the bioassay but did not exceed the established maximum permitted levels.

Whole genome mRNA transcriptomics analysis reveals different modes of action of the diarrheic shellfish poisons okadaic acid and dinophysis toxin-1 versus azaspiracid-1 in Caco-2 cells.

A study with DNA microarrays was performed to investigate the effects of two diarrhetic and one azaspiracid shellfish poison, okadaic acid (OA), dinophysistoxin-1 (DTX-1) and azaspiracid-1 (AZA-1) respectively, on the whole-genome mRNA expression of undifferentiated intestinal Caco-2 cells. Previously, the most responding genes were used to develop a dedicated array tube test to screen shellfish samples on the presence of these toxins. In the present study the whole genome mRNA expression was analyzed in order to reveal modes of action and obtain hints on potential biomarkers suitable to be used in alternative bioassays. Effects on key genes in the most affected pathways and processes were confirmed by qPCR. OA and DTX-1 induced almost identical effects on mRNA expression, which strongly indicates that OA and DTX-1induce similar toxic effects. Biological interpretation of the microarray data indicates that both compounds induce hypoxia related pathways/processes, the unfolded protein response (UPR) and endoplasmic reticulum (ER) stress. The gene expression profile of AZA-1 is different and shows increased mRNA expression of genes involved in cholesterol synthesis and glycolysis, suggesting a different mode of action for this toxin. Future studies should reveal whether identified pathways provide suitable biomarkers for rapid detection of DSPs in shellfish.

Screening for the presence of lipophilic marine biotoxins in shellfish samples using the neuro-2a bioassay.

The neuro-2a bioassay is considered as one of the most promising cell-based in vitro bioassays for the broad screening of seafood products for the presence of marine biotoxins. The neuro-2a assay has been shown to detect a wide array of toxins like paralytic shellfish poisons (PSPs), ciguatoxins, and also lipophilic marine biotoxins (LMBs). However, the neuro-2a assay is rarely used for routine testing of samples due to matrix effects that, for example, lead to false positives when testing for LMBs. As a result there are only limited data on validation and evaluation of its performance on real samples. In the present study, the standard extraction procedure for LMBs was adjusted by introducing an additional clean-up step with n-hexane. Recovery losses due to this extra step were less than 10%. This wash step was a crucial addition in order to eliminate false-positive outcomes due to matrix effects. Next, the applicability of this assay was assessed by testing a broad range of shellfish samples contaminated with various LMBs, including diarrhetic shellfish toxins/poisons (DSPs). For comparison, the samples were also analysed by LC-MS/MS. Standards of all regulated LMBs were tested, including analogues of some of these toxins. The neuro-2a cells showed good sensitivity towards all compounds. Extracts of 87 samples, both blank and contaminated with various toxins, were tested. The neuro-2a outcomes were in line with those of LC-MS/MS analysis and support the applicability of this assay for the screening of samples for LMBs. However, for use in a daily routine setting, the test might be further improved and we discuss several recommended modifications which should be considered before a full validation is carried out.

Big data in food safety: An overview.

Technology is now being developed that is able to handle vast amounts of structured and unstructured data from diverse sources and origins. These technologies are often referred to as big data, and open new areas of research and applications that will have an increasing impact in all sectors of our society. In this paper we assessed to which extent big data is being applied in the food safety domain and identified several promising trends. In several parts of the world, governments stimulate the publication on internet of all data generated in public funded research projects. This policy opens new opportunities for stakeholders dealing with food safety to address issues which were not possible before. Application of mobile phones as detection devices for food safety and the use of social media as early warning of food safety problems are a few examples of the new developments that are possible due to big data.

Effects of Digested Onion Extracts on Intestinal Gene Expression: An Interspecies Comparison Using Different Intestine Models.

Human intestinal tissue samples are barely accessible to study potential health benefits of nutritional compounds. Numbers of animals used in animal trials, however, need to be minimalized. Therefore, we explored the applicability of in vitro (human Caco-2 cells) and ex vivo intestine models (rat precision cut intestine slices and the pig in-situ small intestinal segment perfusion (SISP) technique) to study the effect of food compounds. In vitro digested yellow (YOd) and white onion extracts (WOd) were used as model food compounds and transcriptomics was applied to obtain more insight into which extent mode of actions depend on the model. The three intestine models shared 9,140 genes which were used to compare the responses to digested onions between the models. Unsupervised clustering analysis showed that genes up- or down-regulated by WOd in human Caco-2 cells and rat intestine slices were similarly regulated by YOd, indicating comparable modes of action for the two onion species. Highly variable responses to onion were found in the pig SISP model. By focussing only on genes with significant differential expression, in combination with a fold change > 1.5, 15 genes showed similar onion-induced expression in human Caco-2 cells and rat intestine slices and 2 overlapping genes were found between the human Caco-2 and pig SISP model. Pathway analyses revealed that mainly processes related to oxidative stress, and especially the Keap1-Nrf2 pathway, were affected by onions in all three models. Our data fit with previous in vivo studies showing that the beneficial effects of onions are mostly linked to their antioxidant properties. Taken together, our data indicate that each of the in vitro and ex vivo intestine models used in this study, taking into account their limitations, can be used to determine modes of action of nutritional compounds and can thereby reduce the number of animals used in conventional nutritional intervention studies.

Inhibition of CXCL12-mediated chemotaxis of Jurkat cells by direct immunotoxicants.

Directional migration of cells to specific locations is required in tissue development, wound healing, and immune responses. Immune cell migration plays a crucial role in both innate and adaptive immunity. Chemokines are small pro-inflammatory chemoattractants that control the migration of leukocytes. In addition, they are also involved in other immune processes such as lymphocyte development and immune pathology. In a previous toxicogenomics study using the Jurkat T cell line, we have shown that the model immunotoxicant TBTO inhibited chemotaxis toward the chemokine CXCL12. In the present work, we aimed at assessing a novel approach to detecting chemicals that affect the process of cell migration. For this, we first evaluated the effects of 31 chemicals on mRNA expression of genes that are known to be related to cell migration. With this analysis, seven immunotoxicants were identified as potential chemotaxis modulators, of which five (CoCl2 80 µM, MeHg 1 µM, ochratoxin A 10 µM, S9-treated ochratoxin A 10 µM, and TBTO 100 nM) were confirmed as chemotaxis inhibitor in an in vitro trans-well chemotaxis assay using the chemokine CXCL12. The transcriptome data of the five compounds together with previously obtained protein phosphorylation profiles for two out of five compounds (i.e., ochratoxin A and TBTO) revealed that the mechanisms behind the chemotaxis inhibition are different for these immunotoxicants. Moreover, the mTOR inhibitor rapamycin had no effect on the chemotaxis of Jurkat cells, indicating that the mTOR pathway is not involved in CXCL12-mediated chemotaxis of Jurkat cells, which is opposite to the findings on human primary T cells (Munk et al. in PLoS One 6(9):e24667, 2011). Thus, the results obtained from the chemotaxis assay conducted with Jurkat cells might not fully represent the results obtained with human primary T cells. Despite this difference, the present study indicated that some compounds may exert their immunotoxic effects through inhibition of CXCL12-mediated chemotaxis.

Protein phosphorylation profiling identifies potential mechanisms for direct immunotoxicity.

Signaling networks are essential elements that are involved in diverse cellular processes. One group of fundamental components in various signaling pathways concerns protein tyrosine kinases (PTK). Various toxicants have been demonstrated to exert their toxicity via modulation of tyrosine kinase activity. The present study aimed to identify common cellular signaling pathways that are involved in chemical-induced direct immunotoxicity. To this end, an antibody array-based profiling approach was applied to assess effects of five immunotoxicants, two immunosuppressive drugs and two non-immunotoxic control chemicals on the phosphorylation of 28 receptor tyrosine kinases and 11 crucial signaling nodes in Jurkat T-cells. The phosphorylation of ribosomal protein S6 (RPS6) and of kinases Akt, Src and p44/42 were found to be commonly regulated by immunotoxicants and/or immunosuppressive drugs (at least three compounds), with the largest effect observed upon RPS6. Flow cytometry and Western blotting were used to further examine the effect of the model immunotoxicant TBTO on the components of the mTOR-p70S6K-RPS6 pathway. These analyses revealed that both TBTO and the mTOR inhibitor rapamycin inactivate RPS6, but via different mechanisms. Finally, a comparison of the protein phosphorylation data to previously obtained transcriptome data of TBTO-treated Jurkat cells resulted in a good correlation at the pathway level and indicated that TBTO affects ribosome biogenesis and leukocyte migration. The effect of TBTO on the latter process was confirmed using a CXCL12 chemotaxis assay.

Exploration of new functional endpoints in neuro-2a cells for the detection of the marine biotoxins saxitoxin, palytoxin and tetrodotoxin.

Marine neurotoxins accumulate in seafood and therewith represent a threat for consumers. At the European level, the use of in vivo bioassays is banned from 2015 onwards, except for the control of production areas. Cytotoxicity in the neuro-2a assay has been shown a promising in vitro alternative. However, given that cytotoxicity may be sensitive to confounding factors the current study investigates the suitability of functional endpoints as alternatives to cytotoxicity for the detection of marine neurotoxins. Microarray analyses were performed following exposure of neuro-2a cells to three marine neurotoxins (palytoxin (PlTx), saxitoxin (STX) and tetrodotoxin (TTX)) to identify genes up- or down-regulated that can be used as biomarkers for screening purposes. In addition to microarrays, the voltage dependent fluorescent probe bisoxonol was used to assess changes in cellular membrane potential. Biomarkers based on mRNA expression were detected for PlTx but not for STX and TTX. STX and TTX decreased the fluorescence of bisoxonol while PlTx showed no effect. When using cytotoxicity as the read out the neuro-2a assay detects these three neurotoxins at similar concentrations. Therefore it is concluded that the newly investigated endpoints in the neuro-2a assay are not preferred over cytotoxicity in a suitable broad and sensitive bioassay for the detection of marine neurotoxins in real practice.

Bioavailability and biodistribution of differently charged polystyrene nanoparticles upon oral exposure in rats.

The likelihood of oral exposure to nanoparticles (NPs) is increasing, and it is necessary to evaluate the oral bioavailability of NPs. In vitro approaches could help reducing animal studies, but validation against in vivo studies is essential. Previously, we assessed the translocation of 50 nm polystyrene NPs of different charges (neutral, positive and negative) using a Caco-2/HT29-MTX in vitro intestinal translocation model. The NPs translocated in a surface charge-dependent manner. The present study aimed to validate this in vitro intestinal model by an in vivo study. For this, rats were orally exposed to a single dose of these polystyrene NPs and the uptake in organs was determined. A negatively charged NP was taken up more than other NPs, with the highest amounts in kidney (37.4 µg/g tissue), heart (52.8 µg/g tissue), stomach wall (98.3 µg/g tissue) and small intestinal wall (94.4 µg/g tissue). This partly confirms our in vitro findings, where the same NPs translocated to the highest extent. The estimated bioavailability of different types of NPs ranged from 0.2 to 1.7 % in vivo, which was much lower than in vitro (1.6-12.3 %). Therefore, the integrated in vitro model cannot be used for a direct prediction of the bioavailability of orally administered NPs. However, the model can be used for prioritizing NPs before further in vivo testing for risk assessment.

In vitro gastrointestinal digestion increases the translocation of polystyrene nanoparticles in an in vitro intestinal co-culture model.

The conditions of the gastrointestinal tract may change the physicochemical properties of nanoparticles (NPs) and therewith the bioavailability of orally taken NPs. Therefore, we assessed the impact of in vitro gastrointestinal digestion on the protein corona of polystyrene NPs (PS-NPs) and their subsequent translocation across an in vitro intestinal barrier. A co-culture of intestinal Caco-2 and HT29-MTX cells was exposed to 50 nm PS-NPs of different charges (positive and negative) in two forms: pristine and digested in an in vitro gastrointestinal digestion model. In vitro digestion significantly increased the translocation of all, except the "neutral", PS-NPs. Upon in vitro digestion, translocation was 4-fold higher for positively charged NPs and 80- and 1.7-fold higher for two types of negatively charged NPs. Digestion significantly reduced the amount of protein in the corona of three out of four types of NPs. This reduction of proteins was 4.8-fold for "neutral", 3.5-fold for positively charged and 1.8-fold for one type of negatively charged PS-NPs. In vitro digestion also affected the composition of the protein corona of PS-NPs by decreasing the presence of higher molecular weight proteins and shifting the protein content of the corona to low molecular weight proteins. These findings are the first to report that in vitro gastrointestinal digestion significantly affects the protein corona and significantly increases the in vitro translocation of differently charged PS-NPs. These findings stress the importance of including the in vitro digestion in future in vitro intestinal translocation screening studies for risk assessment of orally taken NPs.

Translocation of differently sized and charged polystyrene nanoparticles in in vitro intestinal cell models of increasing complexity.

Intestinal translocation is a key factor for determining bioavailability of nanoparticles (NPs) after oral uptake. Therefore, we evaluated three in vitro intestinal cell models of increasing complexity which might affect the translocation of NPs: a mono-culture (Caco-2 cells), a co-culture with mucus secreting HT29-MTX cells and a tri-culture with M-cells. Cell models were exposed to well characterized differently sized (50 and 100 nm) and charged (neutral, positively and negatively) polystyrene NPs. In addition, two types of negatively charged NPs with different surface chemistries were used. Size strongly affected the translocation of NPs, ranging up to 7.8% for the 50 nm NPs and 0.8% for the 100 nm NPs. Surface charge of NPs affected the translocation, however, surface chemistry seems more important, as the two types of negatively charged 50 nm NPs had an over 30-fold difference in translocation. Compared with the Caco-2 mono-culture, presence of mucus significantly reduced the translocation of neutral 50 nm NPs, but significantly increased the translocation of one type of negatively charged NPs. Incorporation of M-cells shifted the translocation rates for both NPs closer to those in the mono-culture model. The relative pattern of NP translocation in all three models was similar, but the absolute amounts of translocated NPs differed per model. We conclude that for comparing the relative translocation of different NPs, using one intestinal model is sufficient. To choose the most representative model for risk assessment, in vivo experiments are now needed to determine the in vivo translocation rates of the used NPs.

Successful validation of genomic biomarkers for human immunotoxicity in Jurkat T cells in vitro.

Previously, we identified 25 classifier genes that were able to assess immunotoxicity using human Jurkat T cells. The present study aimed to validate these classifiers. For that purpose, Jurkat cells were exposed for 6 h to subcytotoxic doses of nine immunotoxicants, five non-immunotoxicants and four compounds for which human immunotoxicity has not yet been fully established. RNA was isolated and subjected to Fluidigm quantitative real time (qRT)-PCR analysis. The sensitivity, specificity and accuracy of the screening assay as based on the nine immunotoxicants and five non-immunotoxicants used in this study were 100%, 80% and 93%, respectively, which is better than the performance in our previous study. Only one compound was classified as false positive (benzo-e-pyrene). Of the four potential (non-)immunotoxicants, chlorantraniliprole and Hidrasec were classified immunotoxic and Sunset yellow and imidacloprid as non-immunotoxic. ToxPi analysis of the PCR data provided insight in the molecular pathways that were affected by the compounds. The immunotoxicants 2,3-dichloro-propanol and cypermethrin, although structurally different, affected protein metabolism and cholesterol biosynthesis and transport. In addition, four compounds, i.e. chlorpyrifos, aldicarb, benzo-e-pyrene and anti-CD3, affected genes in cholesterol metabolism and transport, protein metabolism and transcription regulation. qRT-PCR on eight additional genes coding for similar processes as defined in ToxPi analyzes, supported these results. In conclusion, the 25 immunotoxic classifiers performed very well in a screening with new non-immunotoxic and immunotoxic compounds. Therefore, the Jurkat screening assay has great promise to be applied within a tiered approach for animal free testing of human immunotoxicity.

Characterization of the modes of action of deoxynivalenol (DON) in the human Jurkat T-cell line.

Deoxynivalenol (DON) is one of the most abundant mycotoxins worldwide and mostly detected in cereals and grains. As such, DON poses a risk for many adverse health effects to human and animals. In particular, immune cells are very sensitive to DON, with the initiating step leading to toxicity being a binding to the eukaryotic 60S ribosomal subunit and induction of ribotoxic stress. The present study aimed to: (1) extend insight into the mechanism of action (MOA) of DON in immune cells; and (2) understand why immune cells are more sensitive to DON than most other cell types. Previously published microarray studies have described the effects of DON on immune cells. To build upon these findings, here, immunocytological and biochemical studies were performed using human T-lymphocyte Jurkat cells that were exposed for 3 h to 0.5 µM DON. Induction of ER stress by DON was confirmed by immunocytology demonstrating increased protein expression of two major ER stress markers ATF3 and DDIT3. T-cell activation was confirmed by induction of phosphorylation of protein kinases JNK and AKT, activation of NF-κB (p65), and increased expression of NFAT target gene NUR77; each of these are known inducers of the T-cell activation response. Induction of an oxidative stress response was also confirmed by monitoring the nuclear translocation of major oxidative stress markers NRF2 and KEAP1, as well as by changes (i.e. decreases) in cell levels of reduced glutathione. Lastly, this study showed that DON induced cleavage of caspase-3, an event known to mediate apoptosis. Taken together, these results allowed us to formulate a potential mechanism of action of DON in immune cells, i.e. binding to eukaryotic 60S ribosomal subunit → ribotoxic stress → ER stress → calcium release from the ER into cytoplasm → T-cell activation and oxidative stress → apoptosis. It is proposed that immune cells are more sensitive to DON than other cell types due to the induction of a T-cell activation response by increased intracellular calcium levels.

In vitro detection of cardiotoxins or neurotoxins affecting ion channels or pumps using beating cardiomyocytes as alternative for animal testing.

The present study investigated if and to what extent murine stem cell-derived beating cardiomyocytes within embryoid bodies can be used as a broad screening in vitro assay for neurotoxicity testing, replacing for example in vivo tests for marine neurotoxins. Effect of nine model compounds, acting on either the Na(+), K(+), or Ca(2+) channels or the Na(+)/K(+) ATP-ase pump, on the beating was assessed. Diphenhydramine, veratridine, isradipine, verapamil and ouabain induced specific beating arrests that were reversible and none of the concentrations tested induced cytotoxicity. Three K(+) channel blockers, amiodarone, clofilium and sematilide, and the Na(+)/K(+) ATPase pump inhibitor digoxin had no specific effect on the beating. In addition, two marine neurotoxins i.e. saxitoxin and tetrodotoxin elicited specific beating arrests in cardiomyocytes. Comparison of the results obtained with cardiomyocytes to those obtained with the neuroblastoma neuro-2a assay revealed that the cardiomyocytes were generally somewhat more sensitive for the model compounds affecting Na(+) and Ca(2+) channels, but less sensitive for the compounds affecting K(+) channels. The stem cell-derived cardiomyocytes were not as sensitive as the neuroblastoma neuro-2a assay for saxitoxin and tetrodotoxin. It is concluded that the murine stem cell-derived beating cardiomyocytes provide a sensitive model for detection of specific neurotoxins and that the neuroblastoma neuro-2a assay may be a more promising cell-based assay for the screening of marine biotoxins.

Detection of marine neurotoxins in food safety testing using a multielectrode array.

SCOPE: At the European level, detection of marine neurotoxins in seafood is still based on ethically debated and expensive in vivo rodent bioassays. The development of alternative methodologies for the detection of marine neurotoxins is therefore of utmost importance. We therefore investigated whether and to what extent a multielectrode array (MEA) approach can be used as an in vitro alternative for screening of marine neurotoxins potentially present in seafood. METHODS: This MEA approach utilizes rat cortical neurons comprising a wide range of ion channels/pumps and neurotransmitter receptors targeted by marine neurotoxins. We tested the effects of neurotoxic model compounds, pure marine neurotoxins, and extracts from contaminated seafood on neuronal activity of rat cortical neurons cultured on commercial 48-well plates to increase throughput. CONCLUSION: We demonstrate that the MEA approach has a sensitivity of 88% (7/9 model compounds, 6/6 pure marine neurotoxins, and 2/2 marine neurotoxins present in seafood extracts were correctly identified) and a good reproducibility compared to existing in vitro alternatives. We therefore conclude that this MEA-based approach could be a valuable tool for future food safety testing.

Model steatogenic compounds (amiodarone, valproic acid, and tetracycline) alter lipid metabolism by different mechanisms in mouse liver slices.

Although drug induced steatosis represents a mild type of hepatotoxicity it can progress into more severe non-alcoholic steatohepatitis. Current models used for safety assessment in drug development and chemical risk assessment do not accurately predict steatosis in humans. Therefore, new models need to be developed to screen compounds for steatogenic properties. We have studied the usefulness of mouse precision-cut liver slices (PCLS) as an alternative to animal testing to gain more insight into the mechanisms involved in the steatogenesis. To this end, PCLS were incubated 24 h with the model steatogenic compounds: amiodarone (AMI), valproic acid (VA), and tetracycline (TET). Transcriptome analysis using DNA microarrays was used to identify genes and processes affected by these compounds. AMI and VA upregulated lipid metabolism, whereas processes associated with extracellular matrix remodelling and inflammation were downregulated. TET downregulated mitochondrial functions, lipid metabolism, and fibrosis. Furthermore, on the basis of the transcriptomics data it was hypothesized that all three compounds affect peroxisome proliferator activated-receptor (PPAR) signaling. Application of PPAR reporter assays classified AMI and VA as PPARγ and triple PPARα/(ß/δ)/γ agonist, respectively, whereas TET had no effect on any of the PPARs. Some of the differentially expressed genes were considered as potential candidate biomarkers to identify PPAR agonists (i.e. AMI and VA) or compounds impairing mitochondrial functions (i.e. TET). Finally, comparison of our findings with publicly available transcriptomics data showed that a number of processes altered in the mouse PCLS was also affected in mouse livers and human primary hepatocytes exposed to known PPAR agonists. Thus mouse PCLS are a valuable model to identify early mechanisms of action of compounds altering lipid metabolism.

Transcriptome-based functional classifiers for direct immunotoxicity.

Current screening methods for direct immunotoxic chemicals are mainly based on general toxicity studies with rodents. The present study aimed to identify transcriptome-based functional classifiers that can eventually be exploited for the development of in vitro screening assays for direct immunotoxicity. To this end, a toxicogenomics approach was applied in which gene expression changes in human Jurkat lymphoblastic T cells were investigated in response to a wide range of compounds, including direct immunotoxicants, immunosuppressive drugs, and non-immunotoxic control chemicals. On the basis of DNA microarray data previously obtained by the exposure of Jurkat cells to 31 test compounds (Shao et al. in Toxicol Sci 135(2):328-346, 2013), we identified a set of 93 genes, of which 80 were significantly regulated (|numerical ratio| ≥1.62) by at least three compounds and the other 13 genes were significantly regulated by either one single compound or compound class. A total of 28 most differentially regulated genes were selected for qRT-PCR verification using a training set of 44 compounds consisting of the above-mentioned 31 compounds (23 immunotoxic and 8 non-immunotoxic) and 13 additional immunotoxicants. Good correlation between the results of microarray and qRT-PCR (Pearson's correlation, R ≥ 0.69) was found for 27 out of the 28 genes. Redundancy analysis of these 27 potential classifiers led to a final set of 25 genes. To assess the performance of these genes, Jurkat cells were exposed to 20 additional compounds (external verification set) followed by qRT-PCR. The classifier set of 25 genes gave a good performance in the external verification: accuracy 85 %, true positive rate (sensitivity) 88 %, and true negative rate (specificity) 67 %. Furthermore, on the basis of the gene ontology annotation of the 25 classifier genes, the immunotoxicants examined in this study could be categorized into distinct functional subclasses. In conclusion, we have identified and validated classifier genes that can be used for the development of an in vitro assay for the identification and initial characterization of hazards for direct immunotoxicity of chemicals and drugs. This assay promises to complement animal-free toxicity testing approaches within the field of direct immunotoxicity.

DON shares a similar mode of action as the ribotoxic stress inducer anisomycin while TBTO shares ER stress patterns with the ER stress inducer thapsigargin based on comparative gene expression profiling in Jurkat T cells.

Previously, we studied the effects of deoxynivalenol (DON) and tributyltin oxide (TBTO) on whole genome mRNA expression profiles of human T lymphocyte Jurkat cells. These studies indicated that DON induces ribotoxic stress and both DON and TBTO induced ER stress which resulted into T-cell activation and apoptosis. The first goal of the present study was to provide final proof for these mode of actions by comparing the effects of 6 h exposure to DON and TBTO on mRNA expression to those of positive controls of ribotoxic stress (anisomycin), ER stress (thapsigargin) and T cell activation (ionomycin). Genes affected by anisomycin and the majority of genes affected by thapsigargin were affected in the same direction by DON and TBTO, respectively, confirming the expected modes of action. Pathway analysis further sustained that DON induces ribotoxic stress and both DON and TBTO induce unfolded protein response (UPR), ER stress, T cell activation and apoptosis. The second goal was to assess whether DON and/or TBTO affect other pathways above those detected before. TBTO induced groups of genes that are involved in DNA packaging and heat shock response that were not affected by thapsigargin. DON did not affect other genes than anisomycin indicating the effect of DON to be restricted to ribotoxic stress. This study also demonstrates that comparative gene expression analysis is a very promising tool for the identification of modes of action of immunotoxic compounds.