- Email: [email protected]
Dose-dependent intracellular accumulation of endogenous bile acids in cyclosporine A-treated HepaRG hepatocytes
Abstracts / Toxicology Letters 238S (2015) S56–S383
chondria and lysosome) in cultured ovarian cells exposed to low concentrations of Patulin (Pat) (0.025, 0.05 and 0.1 M) and Sterigmatocystin (Ste) (0.078, 0.78 and 3.12 M) similar to those detected in food. Both, Pat and Ste reduce cell viability with an IC50 equal to 2.82 and 25 M respectively after 24 h of exposure. ROS measurement every 5 min during 2 h showed that ROS increased for the two mycotoxins compared to control. Pat reduced mitochondrial membrane potential (MMP) at all concentration tested with a reduction of 20% respect to control while the range of reduction for STE is 25–40%. Besides, the two mycotoxins increased the lysosomal membrane stability (LMS). Our results suggest that ROS, MMP and LMS may be incriminated in cell viability reduction. http://dx.doi.org/10.1016/j.toxlet.2015.08.862
P13-083 Evidence of hormesis on human neuronal SK-N-BE cells treated with sodium arsenate: Impact at the mitochondrial level W. Kharroubi University, Biochemistry, Monastir, Tunisia Exposure of human neuronal SK-N-BE cells to sodium arsenate (AsV) (0.1–400 M; 48 h) induced a biphasic toxic effect evoking hormesis. Indeed, at low concentrations, AsV stimulates cell proliferation visualized by phase contrast microscopy, whereas at high concentrations an induction of cell death associated with a loss of cell adhesion and an increase number of cells floating in the culture medium is observed. These side effects were confirmed with crystal violet test, cell cycle analysis, evaluation of the percentage of Ki67 positive cells and staining with propidium iodide which are complementary methods allowing to evaluate cell growth and cell death induction. The impact of AsV on mitochondrial functions, which was determined by the MTT assay, the measurement of mitochondrial transmembrane potential with DiOC6(3), and the rate of mitochondrial ATP, also support an hormesis process. In addition, in the presence of high concentrations of AsV, a significant decrease of the protein expression of OXPHOS complexes of the respiratory chain was observed by Western blot supporting that AsV-induced cell death is associated with mitochondrial alterations. Therefore, there are some evidences of hormesis on AsV-treated SK-N-BE cells, and at high concentrations the mitochondria are a target of toxicity induced by AsV.
S301
human HepaRG cell line within the framework of the European project MIP-DILI. Previous data from our laboratory have shown that cholestatic drugs can cause either constriction or dilatation of bile canaliculi associated with deregulation of the Rho-kinase pathway in HepaRG cells (Sharanek et al. submitted). In the present work we have analyzed the effects of 8 cholestatic and 8 hepatotoxic (non-cholestatic) or non-toxic drugs on the dynamics of bile canaliculi using phase-contrast imaging, the activity of the Rhokinase pathway by western blotting and the expression/activity of canalicular and basolateral transporters. The 8 cholestatic drugs were correctly identified as causing either constriction or dilatation of bile canaliculi while 7/8 non-cholestatic drugs were ineffective. Only one behaved as a false positive. Moreover, 7/7 cholestatic drugs modulated the phosphorylation level of the myosin light chain, a downstream effector of the Rho-kinase pathway. 6/6 of the tested non-cholestatic drugs did not alter the phosphorylation level of the myosin light chain. Transcripts levels of six major basolateral and canalicular transporters involved in uptake or efflux of bile acids (BSEP, NTCP, OATP-B, MRP2, MRP3, MRP4) were also determined after a 24 h treatment. A decrease in NTCP and BSEP expression was found for 6/6 cholestatic compounds. Moreover, 2/2 non-cholestatic drugs were ineffective. Furthermore, BSEP (efflux) and NTCP (influx) activities were measured after 2 and 24 h treatments: they showed a dose-dependent decrease for 6/6 and 5/5 cholestatic drugs respectively. By contrast, 2/2 non-cholestatic drugs had no effect on these activities. In summary, our preliminary results suggest that analysis of alterations of bile canaliculus structures associated with modulation of the Rho-kinase pathway are promising predictive markers of drug-induced cholestasis. Financial support: European Community [Contract MIP-DILI115336] and ANRT [no. 2013/0112] to MB. http://dx.doi.org/10.1016/j.toxlet.2015.08.864
P13-085 Dose-dependent intracellular accumulation of endogenous bile acids in cyclosporine A-treated HepaRG hepatocytes A. Sharanek 1,∗ , A. Burban 1 , L. Humber 2 , P. Bachour-El Azzi 1 , N. Felix-Gomes 1 , D. Rainteau 2 , A. Guillouzo 1 1
INSERM U991, Rennes, France 2ERL Inserm U1157/UMR7203, Faculté de Médecine Pierre et Marie Curie, Site Saint Antoine, Paris, France 2
http://dx.doi.org/10.1016/j.toxlet.2015.08.863
P13-084 Selective bile canalicular changes induced by cholestatic drugs M. Burbank 1,2,∗ , A. Sharanek 1 , A. Burban 1 , R.J. Weaver 2 , C. Guguen-Guillouzo 1,3 , A. Guillouzo 1 1
INSERM U991, Rennes 1 University, 35043 Rennes, France International Recherche Institute Servier, Suresne 92150, France 3 Biopredic International, St Grégoire, 35760 Rennes, France 2
Many drugs have been reported to induce cholestasis in a dosedependent or independent (idiosyncratic) manner. Intrahepatic cholestasis represents around 40% of drug-induced injuries and up to 40% of all cases remain unpredictable. Our aim is to investigate the mechanisms underlying drug-induced cholestasis and to improve its prediction in humans using the metabolic competent
Cholestasis is one of the most common manifestations of druginduced liver injury (DILI). Its accurate prediction represents a major challenge since up to now it is unpredictable in 40% of all cases. Various in vitro cell models are used to investigate uptake and biliary excretion of bile acids (BAs) aiming to predict cholestatic injury. However, previous studies showed that no human liver cell line did exhibit normal production of BAs. Furthermore, no study has reported yet intracellular accumulation of endogenous BAs in hepatocyte cultures following treatment with a cholestatic drug. We have evaluated the capacity of the human HepaRG cell line to produce BAs and analyzed intracellular changes in BA content and profiles after treatment by the cholestatic drug cyclosporine A (CsA). Our data show that HepaRG cells synthesized, conjugated and excreted normal primary BAs in a serum-free medium at daily levels comparable to those measured in conventional primary human hepatocytes. A 4-h treatment with CsA in serum-free medium resulted in a concentration-dependent intracellular accumulation and changes in the profiles of endogenous BAs associated with occurrence of cholestatic features while after 24 h BAs were
S302
Abstracts / Toxicology Letters 238S (2015) S56–S383
decreased in the cellular layer and increased in the supernatant. The latter effects resulted from inhibition of BSEP and NTCP, the main BA efflux and uptake transporters, inhibition of CYP7A1, CYP8B1 and CYP27A1 which are involved in the initial steps of BA synthesis and enhanced expression of the basolateral transporters MRP3 and MRP4 which is recognized as a compensatory mechanism of BA excretion. Importantly, when CsA-treated HepaRG cells were incubated in a medium containing 2% bovine serum they did not accumulate endogenous BAs; however, after 24-h they showed a dose-dependent accumulation of bovine serum lithocholic acid in a non-sulfoconjugated form resulting from an inhibition of SULT2A1. In summary, our work bring the first demonstration that an in vitro human liver cell model, the HepaRG cell line, is able to produce, conjugate and secrete BAs and that a transient accumulation of endogenous BAs concomitantly to occurrence of various other cholestatic features can be observed following treatment with a cholestatic drug. Our data should help in the development of screening methods for early prediction of drug-induced cholestatic side-effects. Financial support: European Community, Contract MIP-DILI115336. http://dx.doi.org/10.1016/j.toxlet.2015.08.865
P13-086 Assessment of cytotoxicity and metabolism of salinomycin in two mammalian cell cultures M. Olejnik ∗ , L. Radko National Veterinary Research Institute, Department of Pharmacology and Toxicology, Pulawy, Poland Salinomycin (SAL) is widely used in veterinary medicine as an antiprotozoal to control coccidiosis. In the last decade, in vitro studies have shown the potential of its applicability in human medicine. The toxicity of SAL is species dependent and seems to be related to the metabolism of the drug. The present study was performed to determine cytotoxic effects of salinomycin on human hepatoma (HepG2) and mouse fibroblast (Balb/c 3T3) cell lines. The cells were incubated with SAL and its combination with tiamuline and prednisolone for 24 and 48 h. The cell metabolism (MTT assay), lysosomal activity (NRU assay), cellular protein content (TPC assay) and membrane integrity (LDH assay) were used as endpoints. Additionally, biotransformation products of salinomycin were determined using liquid chromatography coupled with mass spectrometry. The lowest EC50 -values for salinomycin in the cell cultures were detected by NRU test, confirming the mode of toxic action of a drug (interfering with transmembrane cation transport). Fibroblasts were more sensitive to SAL than HepG2 cells. This could be partly explained by the biotransformation of the drug to non-toxic metabolites. However, the metabolic activity of human hepatoma cells does not seem to be the only reason for their lower susceptibility to SAL toxicity. Although sixteen potential metabolites of salinomycin were identified, they represented not more than 10% of SAL. The interaction between salinomycin and tiamulin or prednisolone showed synergic, additive or antagonistic mode depending on the cell line, the concentration of salinomycin and the kind of measured end-point. The presented research was performed in the frame of the project funded by National Science Centre, Poland (2012/07/D/NZ7/0338). http://dx.doi.org/10.1016/j.toxlet.2015.08.866
P13-087 Translational analysis of c-kit kinase to justify inclusion as a screen in early discovery to aid bone marrow risk assessment L. Rosenbrier Ribeiro 1,∗ , M. Wagoner 2 , J. Kelsall 3 , C. Walker 3 , M. Crosby 2 , C. Sadler 3 1 AstraZeneca, Drug Safety & Metabolism, Cambridge, United Kingdom 2 AstraZeneca, Drug Safety & Metabolism, Waltham, United States 3 AstraZeneca, Drug Safety & Metabolism, Macclesfield, United Kingdom
Bone marrow toxicity, particularly with oncology kinase inhibitors, can cause dose-limiting anaemia, thrombocytopenia or neutropenia in the clinic that needs to be managed (Dy and Adjei, 2013). The bone marrow toxicity observed with kinase inhibitors is often attributed to pharmacology of the primary target, however off target liabilities may cause or contribute to the adverse impact on the bone marrow. As pathways of haematopoiesis within the bone marrow are now increasingly well understood we can propose a number of potential targets, for example c-kit kinase, which if disrupted would likely cause effects on the bone marrow (Edling and Hallberg, 2007). Using predictive in vitro screens for such targets could be included as part of a strategy for early bone marrow toxicity identification. The aim of this study was to examine the in vitro to in vivo correlation between activity at c-kit kinase and bone marrow toxicity to build the weight of evidence to support inclusion of an early screen as part of a strategy for detecting bone marrow toxicity. The compound set for the analysis comprised of three known marketed c-kit inhibitors (imatinib, sunitinib and dasatinib), and a set of compounds from an in-house X-kinase inhibitor project known to have bone marrow toxicity in vivo. All compounds were tested to establish the inhibitory activity (IC50 ) in an in vitro human c-kit enzyme activity assay and an in vitro human CD34+ stem cell bone marrow toxicity assay. Data showed good correlation between ckit inhibition and in vitro human CD34+ stem cell bone marrow toxicity. In vivo bone marrow toxicity observed in humans for imatinib, sunitinib and dasatinib and in rats for three of the in-house inhibitors, was also found to occur at drug exposures which cover the c-kit IC50 . This data builds confidence in the linkage between ckit inhibition and bone marrow toxicity and supports the inclusion of an early rapid, cost-effective screen for c-kit activity and rationale for omitting activity within chemical series to reduce the risk of bone marrow toxicity caused by this off-target liability. References Dy, G.K., Adjei, A.A., 2013. CA: Cancer J. Clin. 63 (4), 249–279. Edling, C.E., Hallberg, B., 2007. Int. J. Biochem. Cell Biol. 39 (11), 1995–1998.
http://dx.doi.org/10.1016/j.toxlet.2015.08.867
chondria and lysosome) in cultured ovarian cells exposed to low concentrations of Patulin (Pat) (0.025, 0.05 and 0.1 M) and Sterigmatocystin (Ste) (0.078, 0.78 and 3.12 M) similar to those detected in food. Both, Pat and Ste reduce cell viability with an IC50 equal to 2.82 and 25 M respectively after 24 h of exposure. ROS measurement every 5 min during 2 h showed that ROS increased for the two mycotoxins compared to control. Pat reduced mitochondrial membrane potential (MMP) at all concentration tested with a reduction of 20% respect to control while the range of reduction for STE is 25–40%. Besides, the two mycotoxins increased the lysosomal membrane stability (LMS). Our results suggest that ROS, MMP and LMS may be incriminated in cell viability reduction. http://dx.doi.org/10.1016/j.toxlet.2015.08.862
P13-083 Evidence of hormesis on human neuronal SK-N-BE cells treated with sodium arsenate: Impact at the mitochondrial level W. Kharroubi University, Biochemistry, Monastir, Tunisia Exposure of human neuronal SK-N-BE cells to sodium arsenate (AsV) (0.1–400 M; 48 h) induced a biphasic toxic effect evoking hormesis. Indeed, at low concentrations, AsV stimulates cell proliferation visualized by phase contrast microscopy, whereas at high concentrations an induction of cell death associated with a loss of cell adhesion and an increase number of cells floating in the culture medium is observed. These side effects were confirmed with crystal violet test, cell cycle analysis, evaluation of the percentage of Ki67 positive cells and staining with propidium iodide which are complementary methods allowing to evaluate cell growth and cell death induction. The impact of AsV on mitochondrial functions, which was determined by the MTT assay, the measurement of mitochondrial transmembrane potential with DiOC6(3), and the rate of mitochondrial ATP, also support an hormesis process. In addition, in the presence of high concentrations of AsV, a significant decrease of the protein expression of OXPHOS complexes of the respiratory chain was observed by Western blot supporting that AsV-induced cell death is associated with mitochondrial alterations. Therefore, there are some evidences of hormesis on AsV-treated SK-N-BE cells, and at high concentrations the mitochondria are a target of toxicity induced by AsV.
S301
human HepaRG cell line within the framework of the European project MIP-DILI. Previous data from our laboratory have shown that cholestatic drugs can cause either constriction or dilatation of bile canaliculi associated with deregulation of the Rho-kinase pathway in HepaRG cells (Sharanek et al. submitted). In the present work we have analyzed the effects of 8 cholestatic and 8 hepatotoxic (non-cholestatic) or non-toxic drugs on the dynamics of bile canaliculi using phase-contrast imaging, the activity of the Rhokinase pathway by western blotting and the expression/activity of canalicular and basolateral transporters. The 8 cholestatic drugs were correctly identified as causing either constriction or dilatation of bile canaliculi while 7/8 non-cholestatic drugs were ineffective. Only one behaved as a false positive. Moreover, 7/7 cholestatic drugs modulated the phosphorylation level of the myosin light chain, a downstream effector of the Rho-kinase pathway. 6/6 of the tested non-cholestatic drugs did not alter the phosphorylation level of the myosin light chain. Transcripts levels of six major basolateral and canalicular transporters involved in uptake or efflux of bile acids (BSEP, NTCP, OATP-B, MRP2, MRP3, MRP4) were also determined after a 24 h treatment. A decrease in NTCP and BSEP expression was found for 6/6 cholestatic compounds. Moreover, 2/2 non-cholestatic drugs were ineffective. Furthermore, BSEP (efflux) and NTCP (influx) activities were measured after 2 and 24 h treatments: they showed a dose-dependent decrease for 6/6 and 5/5 cholestatic drugs respectively. By contrast, 2/2 non-cholestatic drugs had no effect on these activities. In summary, our preliminary results suggest that analysis of alterations of bile canaliculus structures associated with modulation of the Rho-kinase pathway are promising predictive markers of drug-induced cholestasis. Financial support: European Community [Contract MIP-DILI115336] and ANRT [no. 2013/0112] to MB. http://dx.doi.org/10.1016/j.toxlet.2015.08.864
P13-085 Dose-dependent intracellular accumulation of endogenous bile acids in cyclosporine A-treated HepaRG hepatocytes A. Sharanek 1,∗ , A. Burban 1 , L. Humber 2 , P. Bachour-El Azzi 1 , N. Felix-Gomes 1 , D. Rainteau 2 , A. Guillouzo 1 1
INSERM U991, Rennes, France 2ERL Inserm U1157/UMR7203, Faculté de Médecine Pierre et Marie Curie, Site Saint Antoine, Paris, France 2
http://dx.doi.org/10.1016/j.toxlet.2015.08.863
P13-084 Selective bile canalicular changes induced by cholestatic drugs M. Burbank 1,2,∗ , A. Sharanek 1 , A. Burban 1 , R.J. Weaver 2 , C. Guguen-Guillouzo 1,3 , A. Guillouzo 1 1
INSERM U991, Rennes 1 University, 35043 Rennes, France International Recherche Institute Servier, Suresne 92150, France 3 Biopredic International, St Grégoire, 35760 Rennes, France 2
Many drugs have been reported to induce cholestasis in a dosedependent or independent (idiosyncratic) manner. Intrahepatic cholestasis represents around 40% of drug-induced injuries and up to 40% of all cases remain unpredictable. Our aim is to investigate the mechanisms underlying drug-induced cholestasis and to improve its prediction in humans using the metabolic competent
Cholestasis is one of the most common manifestations of druginduced liver injury (DILI). Its accurate prediction represents a major challenge since up to now it is unpredictable in 40% of all cases. Various in vitro cell models are used to investigate uptake and biliary excretion of bile acids (BAs) aiming to predict cholestatic injury. However, previous studies showed that no human liver cell line did exhibit normal production of BAs. Furthermore, no study has reported yet intracellular accumulation of endogenous BAs in hepatocyte cultures following treatment with a cholestatic drug. We have evaluated the capacity of the human HepaRG cell line to produce BAs and analyzed intracellular changes in BA content and profiles after treatment by the cholestatic drug cyclosporine A (CsA). Our data show that HepaRG cells synthesized, conjugated and excreted normal primary BAs in a serum-free medium at daily levels comparable to those measured in conventional primary human hepatocytes. A 4-h treatment with CsA in serum-free medium resulted in a concentration-dependent intracellular accumulation and changes in the profiles of endogenous BAs associated with occurrence of cholestatic features while after 24 h BAs were
S302
Abstracts / Toxicology Letters 238S (2015) S56–S383
decreased in the cellular layer and increased in the supernatant. The latter effects resulted from inhibition of BSEP and NTCP, the main BA efflux and uptake transporters, inhibition of CYP7A1, CYP8B1 and CYP27A1 which are involved in the initial steps of BA synthesis and enhanced expression of the basolateral transporters MRP3 and MRP4 which is recognized as a compensatory mechanism of BA excretion. Importantly, when CsA-treated HepaRG cells were incubated in a medium containing 2% bovine serum they did not accumulate endogenous BAs; however, after 24-h they showed a dose-dependent accumulation of bovine serum lithocholic acid in a non-sulfoconjugated form resulting from an inhibition of SULT2A1. In summary, our work bring the first demonstration that an in vitro human liver cell model, the HepaRG cell line, is able to produce, conjugate and secrete BAs and that a transient accumulation of endogenous BAs concomitantly to occurrence of various other cholestatic features can be observed following treatment with a cholestatic drug. Our data should help in the development of screening methods for early prediction of drug-induced cholestatic side-effects. Financial support: European Community, Contract MIP-DILI115336. http://dx.doi.org/10.1016/j.toxlet.2015.08.865
P13-086 Assessment of cytotoxicity and metabolism of salinomycin in two mammalian cell cultures M. Olejnik ∗ , L. Radko National Veterinary Research Institute, Department of Pharmacology and Toxicology, Pulawy, Poland Salinomycin (SAL) is widely used in veterinary medicine as an antiprotozoal to control coccidiosis. In the last decade, in vitro studies have shown the potential of its applicability in human medicine. The toxicity of SAL is species dependent and seems to be related to the metabolism of the drug. The present study was performed to determine cytotoxic effects of salinomycin on human hepatoma (HepG2) and mouse fibroblast (Balb/c 3T3) cell lines. The cells were incubated with SAL and its combination with tiamuline and prednisolone for 24 and 48 h. The cell metabolism (MTT assay), lysosomal activity (NRU assay), cellular protein content (TPC assay) and membrane integrity (LDH assay) were used as endpoints. Additionally, biotransformation products of salinomycin were determined using liquid chromatography coupled with mass spectrometry. The lowest EC50 -values for salinomycin in the cell cultures were detected by NRU test, confirming the mode of toxic action of a drug (interfering with transmembrane cation transport). Fibroblasts were more sensitive to SAL than HepG2 cells. This could be partly explained by the biotransformation of the drug to non-toxic metabolites. However, the metabolic activity of human hepatoma cells does not seem to be the only reason for their lower susceptibility to SAL toxicity. Although sixteen potential metabolites of salinomycin were identified, they represented not more than 10% of SAL. The interaction between salinomycin and tiamulin or prednisolone showed synergic, additive or antagonistic mode depending on the cell line, the concentration of salinomycin and the kind of measured end-point. The presented research was performed in the frame of the project funded by National Science Centre, Poland (2012/07/D/NZ7/0338). http://dx.doi.org/10.1016/j.toxlet.2015.08.866
P13-087 Translational analysis of c-kit kinase to justify inclusion as a screen in early discovery to aid bone marrow risk assessment L. Rosenbrier Ribeiro 1,∗ , M. Wagoner 2 , J. Kelsall 3 , C. Walker 3 , M. Crosby 2 , C. Sadler 3 1 AstraZeneca, Drug Safety & Metabolism, Cambridge, United Kingdom 2 AstraZeneca, Drug Safety & Metabolism, Waltham, United States 3 AstraZeneca, Drug Safety & Metabolism, Macclesfield, United Kingdom
Bone marrow toxicity, particularly with oncology kinase inhibitors, can cause dose-limiting anaemia, thrombocytopenia or neutropenia in the clinic that needs to be managed (Dy and Adjei, 2013). The bone marrow toxicity observed with kinase inhibitors is often attributed to pharmacology of the primary target, however off target liabilities may cause or contribute to the adverse impact on the bone marrow. As pathways of haematopoiesis within the bone marrow are now increasingly well understood we can propose a number of potential targets, for example c-kit kinase, which if disrupted would likely cause effects on the bone marrow (Edling and Hallberg, 2007). Using predictive in vitro screens for such targets could be included as part of a strategy for early bone marrow toxicity identification. The aim of this study was to examine the in vitro to in vivo correlation between activity at c-kit kinase and bone marrow toxicity to build the weight of evidence to support inclusion of an early screen as part of a strategy for detecting bone marrow toxicity. The compound set for the analysis comprised of three known marketed c-kit inhibitors (imatinib, sunitinib and dasatinib), and a set of compounds from an in-house X-kinase inhibitor project known to have bone marrow toxicity in vivo. All compounds were tested to establish the inhibitory activity (IC50 ) in an in vitro human c-kit enzyme activity assay and an in vitro human CD34+ stem cell bone marrow toxicity assay. Data showed good correlation between ckit inhibition and in vitro human CD34+ stem cell bone marrow toxicity. In vivo bone marrow toxicity observed in humans for imatinib, sunitinib and dasatinib and in rats for three of the in-house inhibitors, was also found to occur at drug exposures which cover the c-kit IC50 . This data builds confidence in the linkage between ckit inhibition and bone marrow toxicity and supports the inclusion of an early rapid, cost-effective screen for c-kit activity and rationale for omitting activity within chemical series to reduce the risk of bone marrow toxicity caused by this off-target liability. References Dy, G.K., Adjei, A.A., 2013. CA: Cancer J. Clin. 63 (4), 249–279. Edling, C.E., Hallberg, B., 2007. Int. J. Biochem. Cell Biol. 39 (11), 1995–1998.
http://dx.doi.org/10.1016/j.toxlet.2015.08.867