Plan Your Agenda with CDI

Monday, March 11

Exhibitor-hosted Sessions

Quantification of Cardioactive Drug Effects Using xCELLigence RTCA Cardio and Human iPS-derived Cardiomyocytes

Cardiac Toxicity Assessment Using Stem Cell-derived Cardiomyocytes

Scientific Posters

Poster: 119 | Abstract: 85
Optical Measurements of Electrical Activity from hiPSC-derived Cardiomyocytes Is a Robust and High-throughput Method for Measuring NCE-effects on the Cardiac Action Potential

Poster: 120 | Abstract: 86 Multiparameter In Vitro Assessment of Drug Effects on Cardiomyocyte Physiology Using iPS Cells

Poster: 123 | Abstract: 89 Development of an In Vitro Multiwell Cardiovascular Microelectrode Array (MEA) Toxicity Assay with Human IPS-derived Cardiomyocytes

Poster: 628 | Abstract: 363
MicroRNA Microarray Analysis of Human-induced Pluripotent Stem Cell-derived Neurons and Cardiomyocytes Following Exposure to the Organophosphate Nerve Agents Soman and VX

Tuesday, March 12

Scientific Posters

Poster: 203 | Abstract: 901
Development of an In Vitro Multiwell Microelectrode Array (MEA) Neurotoxicity Assay with Human IPS-derived Neurons

Poster: 508 | Abstract: 1063
In Vitro Assessment of Drug Induced Liver Injury (DILI) Using a High Content Cellular Imaging System

Poster: 522 | Abstract: 1465
Predicting Heart-specific Toxicity Using Two Cell Models: Human iPS-derived Cardiomyocytes and Human Liver Cells (HepaRG)

Poster: 617 | Abstract: 1556
Multi-parameter In Vitro Toxicity Testing of Crizotinib, Sunitinib, Erlotinib, and Nilotinib in Human Cardiomyocytes

Wednesday, March 13

Exhibitor-hosted Sessions

Application of Human iPSC-derived Hepatocytes Models for Toxicity Assessment

Scientific Posters

CDI Booth Poster Presentation

Characterization and Function of iPSC derived Hepatocytes for Use in Toxicity
Shannon Einhorn, Giorgia Salvagiotto Jen Luebke-Wheeler, Kristin Block, David Mann

Download PDF
Cellular Dynamics International, Madison, WI

Abstract:
Hepatoxicity is a leading cause of drug withdrawal from the market, highlighting the fact that current preclinical models of toxicity are not universally predictive of drug effects in humans. The manifestations of hepatotoxicity are highly variable and range from intrinsic toxic effects to the enzymatic production of toxic metabolites. Development of a more predictive in vitro model system that can elucidate the underlying mechanisms of hepatotoxicity early in the drug development process is critical to subverting unanticipated drug failure in the clinic. Cellular Dynamics International (CDI) has developed human induced pluripotent stem cell (iPSC)-derived hepatocytes, iCell® Hepatocytes, that exhibit high purity (>95%) and biologically relevant functions necessary for hepatotoxicity studies. To demonstrate the functional utility of iCell Hepatocytes in elucidating mechanisms of drug-induced hepatotoxicity, a variety of functional endpoints were measured in response to known hepatotoxic compounds. Responses in iCell Hepatocytes were observed for all compounds at EC50s that were comparable to those seen with primary human hepatocytes. Mechanism-based toxicity was evaluated by analyzing cell metabolism, oxidative stress, and lysosomal phospholipase activity using multiple platforms, including luminescence-based assays (Promega), High Content Analysis (ImageXpress, Molecular Devices), and the XF96 Extracellular Flux Analyzer (Seahorse Biosciences). This multiple platform approach for functional analyses demonstrated how iCell Hepatocytes provide a biologically relevant human model system for studying drug-induced hepatotoxicity.

Monday, March 11

9:30 am to12:30 pm | Exhibit Hall C-D

Poster: 119 | Abstract: 85
Optical Measurements of Electrical Activity from hiPSC-derived Cardiomyocytes Is a Robust and High-throughput Method for Measuring NCE-effects on the Cardiac Action Potential

G. Smith2, BD. Anson1, MA Craig3, and I Ghouri2

Download PDF
  1. Cellular Dynamics International, Madison, WI
  2. University of Glasgow, Glasgow, United Kingdom
  3. Clyde Biosciences, Glasgow, United Kingdom

Abstract:
Proarrhythmic assessment of new chemical entities (NCEs) is a vital component of toxicological and safety profiling. Manual patch clamp interrogation of the cardiac action potential (AP) is a gold standard for proarrhythmia assessment; however this technique is laborious, requires highly skilled scientists, and neither the technique nor traditional tissue cells have been readily amenable to higher throughput techniques. Optical measurements of electrical activity with human cardiomyocytes may offer a solution. We present data here characterizing and demonstrating the suitability of a novel optical platform used in conjunction with hiPSC-derived cardiomyocytes to assess NCE-mediated proarrhythmogenicity. Cardiac electrical activity was monitored from spontaneously beating hiPSC-cardiomyocyte synctia with di-4-ANEPPS at a 10kHz over 60 second time windows from 2-3 areas per well over 3 wells for both experimental and control values. Small molecule effects on Na+, hERG, and Ca2+ channels were assessed and compared with time matched controls. Ion channel block by all agents produced dose-dependent changes in the expected AP parameters relative to controls. Na+ channel block by 3 µM mexelitine produced a 244±86% (n=3) increase in AP rise time, hERG channel block by 30nM E4031 produced a 39±14% (n=3) increase in APD90, and Ca2+ channel block by 300nM nifedipine producing an approximate 50% shortening in APD90. Experimental measurements for a single compound were completed in less than 3 hours. Similar data was obtained across cellular manufacturing batches and experimental days. This dataset demonstrates that optical mapping of electrical activity on hIPSC-derived cardiomyocytes is a suitable methodology for assessing NCE arrhythmogenic potential with a robustness that approaches that of manual patch clamp and a throughput that is an order of magnitude greater.


Poster: 120 | Abstract: 86
Multiparameter In Vitro Assessment of Drug Effects on Cardiomyocyte Physiology Using iPS Cells

O Sirenko1, C Crittenden1, B Anson2, I Rusyn3, and EF Cromwell1

Download PDF
  1. Molecular Devices LLC, Sunnyvale, CA
  2. Cellular Dynamics International, Madison, WI
  3. University of North Carolina, Chapel Hill, NC

Abstract: Highly predictive in vitro assays suitable tor high throughput screening (HTS) tor potential cardiotoxicity are critical to drug safety testing. Adult human stem cell derived cadiomyocytes show promise for screening compounds during early drug development. We developed methods tor measuring the impact of drug candidates on the beating rate of human iPSC derived cardiomyocytes using fast kinetic fluorescence imaging. Cardiomyocyte contraction rate and pattern are characterized by monitoring changes in intracellular Ca2+ measured using calcium sensitive dyes. The assay was optimised tor HTS and allows characterization of beating profiles by using multi-parameter analysis outputs such as beating rate, peak frequency and width, or waveform irregularities. The assay is suitable tor assessment of short-term (minutes) and delayed (days) effects. Next, we tested known cardiotoxic compounds including alpha and beta blockers, hERG inhibitors, ion channel blockers, etc., as well as control drugs. IC50 values showed a significant rank correlation with published values determined by other cardiotoxicity models as well as good concordance with reported human plasma Cmax values. The assay was further tested using commercially available cardiotoxicity library ·epresenting different classes of compounds including receptor antagonists, ion channel blockers, anti-cancer and anti-inflammatory drugs, and kinase inhibitors. The estimated balanced prediction accuracy of the assay was greater than 80%, and multiparameter characterization of beating profiles allowed identification of specific patterns defining hERG or Na channel blockers. We conclude that this assay shows utility tor screening compounds tor potential to cause arrhythmic and nonarrhythmic cardiotoxicity.


Poster: 123 | Abstract: 89
Development of an In Vitro Multiwell Cardiovascular Microelectrode Array (MEA) Toxicity Assay with Human IPS-derived Cardiomyocytes
S Qin1, J Ross2, M. Brock2, J Bradley1, H Luithardt2, J Gilbert1, and C Strock1

  1. Cyprotex, Watertown, MA
  2. Axion Biosystems, Atlanta, GA

Abstract:
Cardiotoxicity represents the most common reason for attrition of compounds due to toxicity. Current in vitro assays fail to predict the many causes of cardiac toxicity. Cytotoxicity assays only identify the most overtly toxic compounds, while assays to measure more specific liabilities such as hERG or other ion channels measure only a specific risk target, potentially missing the overall physiological response a compound elicits in cardiac cells. Additionally, in vivo animal studies rely on determination of compound activity in a different species, which does not always model a human responses. Therefore, an ideal assay would screen compounds in human cardiomyocytes and provide a comprehensive assessment of their effects on electrophysiology. Here we demonstrate the successful use of multiwell MEAs to screen for cardiotoxic liabilities in human IPS derived cardiomyocytes (hsCM). MEAs measure the electrical activity in cells and therefore can be used to capture the field potential of the cardiomyocytes, producing a virtual EKG. This data can then be extracted to report effects of test compounds on the different phases of the heartbeat. Response endpoints that can be reported include: field potential duration (“QT”), conduction velocity, beat rate, as well as field potential metrics (Amplitude, Slope, etc.). The recent development of the multiwell MEA has allowed for an increased throughput for cardiotoxicity screening as well as the ability to generate dose curve responses for each compound. Here we demonstrate the different electrophysiological responses observed when we test compounds with different targets and liabilities (hERG, other ion channels, etc.). The compounds tested are E4031, Nifedipine, Sotalol, isoproterenol, Ouabain, Mexiletine, ZD7288, Amitryptiline, Cisapride, Terfenadine, and Verapamil. In summary, we show that use of multiwell MEAs with hsCM is highly effective for screening test compounds for toxic liabilities.


Poster: 628 | Abstract: 363
MicroRNA Microarray Analysis of Human-induced Pluripotent Stem Cell-derived Neurons and Cardiomyocytes Following Exposure to the Organophosphate Nerve Agents Soman and VX

EK Yego, JF Dillman, and HM Hoard-Fruchey

Research Division, US Army Medical Research Institute of Chemical Defense, Aberdeen Proving Ground, MD

Abstract:
Chemical warfare nerve agents (CWNA) are potent cholinesterase inhibitors that may also have non-cholinesterase effects. Several in vivo studies have demonstrated CWNA-induced damage in the brain and heart following CWNA exposure. The mechanisms of this damage have been a critical area of research for the development of medical countermeasures. This study utilized microRNA (miRNA) analysis to evaluate potential direct cellular effects of the nerve agents soman (O-Pinacolyl methylphosphonofluoridate) or VX (o-ethyl-s-[2 (diisopropylamino) ethyl]) on human-induced pluripotent stem cell (iPSC)-derived neurons and cardiomyocytes. This approach was taken since miRNA expression changes are stimulus specific and no previous studies of miRNA profiles have been conducted for CWNA exposure. Cells were treated with soman or VX at concentrations of 0 µM, 0.1 µM or 100 µM for either 1 hr or 6 hr. Following treatment, isolated total RNA was processed for miRNA microarray analysis and analyzed for significant changes. Soman- and VX-treated samples were analyzed separately. Principal component analysis (PCA) was used to identify major sources of variability in the dataset. PCA analysis of neurons identified differences in miRNA expression only for cells exposed for 6 hr to soman. Targets that were significantly altered under these conditions were miR-2277, miR-1910 and miR-1972. miR-2277 was significantly altered in soman- and VX-exposed neuron data sets that were analyzed by both time and dose. Minimal sample variability was observed with cardiomyocytes as determined by PCA analysis. One-way ANOVA with time as the factor identified miR-3178 as the only target that was altered significantly by both soman and VX in cardiomyocytes. This miRNA modulates several targets including complexin-1 and splicing factor-1. This study demonstrates the feasibility of using miRNA microarray analysis for the study of CWNA cellular effects.

Tuesday, March 12

9:00 am to 12:30 pm | Exhibit Hall C-D

Poster: 203 | Abstract: 901
Development of an In Vitro Multiwell Microelectrode Array (MEA) Neurotoxicity Assay with Human IPS-derived Neurons

J Bradley1, J Ross2, M Brock2, H Luithardt1, J Gilbert1, and C Strock1

  1. Cyprotex, Watertown, MA
  2. Axion Biosystems, Atlanta, GA

Abstract:
Seizurogenic neurotoxicity produces significant drug attrition during drug discovery. Current available in vitro assays fail to predict this toxicity due to the failure of general cytotoxicity assays to predict sublethal target specific electrophysiological liabilities. Ion channel and receptor activity assays can be used to predict some seizure potential, but this only focuses on specifically measured targets for prediction and may miss a response which relies on a combination of endpoints. Most evaluation of seizure inducing compounds occurs later in preclinical development in in vivo studies which have higher costs and could result in species specific results. Therefore, the development of an in vitro assay to screen compounds for seizurogenic potential in a human neural model would give the potential to screen compounds earlier at lower cost and greater reliability. Here we demonstrate the use of multiwell MEAs to screen for seizurogenic compounds in human IPS derived neurons (hsN). HsNs were seeded in 48 well MEAs and spontaneous activity began at 3 days post plating with greater activity at 7 days when the assay is performed. Neural action potentials were detected and the results were reported as mean firing rate (MFR). The seizurogenic compounds tested show dose dependent increase in MFR with changes in spike train organization, while all of the negative controls were unaffected. The seizurogenic compounds tested were Picrotoxin, Gabazine, L-­‐Glutamate, and pentylenetetrazole (PTZ). Negative compounds tested were acetaminophen, naproxen, and DMSO. To further demonstrate the responsiveness of the cells in the assay, we tested Domoic Acid, a neurotoxin known to cause amnesia, and found that it completely blocked network activity while not causing cell death. These results illustrate the power of the human Neural MEA assay for predicting compound induced neural toxicity, especially the seizurogenic response.


Poster: 508 | Abstract: 1063
In Vitro Assessment of Drug Induced Liver Injury (DILI) Using a High Content Cellular Imaging System

ML Wolfe1, S Einhorn2, V Ott2, and H Ma1

Download PDF
  1. Cellular and Molecular Biology, MPI Research, Mattawan, MI
  2. Cellular Dynamics International, Madison, WI

Abstract:
Drug-induced liver injury (DIU) is a leading cause of drugs failing during clinical trials and being withdrawn from the market. In vivo safety testing in pre-clinical species ensures that drugs which enter clinical trials do not cause reproducible and dose-dependent liver injury in man, but is of limited value for exploration of underlying mechanisms and does not assess potential to cause rare idiosyncratic DIU. Implementation of an In vitro cell-based predictive assay early in the drug discovery process would help Improve early compound attrition and develop safer drug candidates. We tested compound-treated human hepatocellular carcinoma Hep G2 cells, human induced pluripotent stem cell (iPSC)-derived hepatocytes (iCell® Hepatocytes) and primary human hepatocytes using the Thermo Scientific Toxlnsight® IVT platform and DIU Assay Cartridge to determine the hepatotoxicity risk of a compound through the measurement of multiple toxicity biomarkers inindividual cells. The compounds we investigated include a number of known hepatotoxic compounds (Ticlopidine, TrogUtazone, Naladixic acid, Mefenamic acid, Phenylbutazone and Aflatoxin B 1} and non-hepatotoxic compounds (Aspirin, Fluoxetine and Melatonin). Each compound was tested at eight concentrations in triplicate. The DILl Assay Cartridge allows for the high sensitivity and specificity for predicting hepatotoxicity by simultaneously detecting five multiplexed cellular targets and properties associated with cell loss, cellular redox stress, and mitochondrial stress. The hepatotoxicity prediction using the multiparametric data generated for the test compounds demonstrates high specificity across the three hepatocytes models but varying sensitivity for each hepatocyte model system.


1:00 pm to 4:30 pm | Exhibit Hall C-D

Poster: 522 | Abstract: 1465
Predicting Heart-specific Toxicity Using Two Cell Models: Human iPS-derived Cardiomyocytes and Human Liver Cells (HepaRG)

PC Wilga, D Keller, B Franz, and JM McKim

CeeTox, Inc, Kalamazoo, MI

Abstract:
A major reason for the failure of new drugs is due to adverse effects in the cardiovascular system. An in vitro model capable of identifying heart-specific liabilities would be of considerable value. To differentiate heart toxicity from liver toxicity, a dual cell model was developed that utilizes changes in cell health and function following exposure to a test drug. Human cardiomyocytes and normal human liver cells were used as the test system. Cardiomyocytes were derived from induced pluripotent stem cells (iPS) obtained from Cellular Dynamics International. Terminally differentiated bi-potent HepaRG cells were from Life Technologies. Each cell type was established in a 96-well format and prepared in triplicate wells for each concentration. Markers of cell health (ATP, LDH leakage, and GSH) were monitored in both cell types. Additional information was collected for beat rate (BR) in heart cells and endpoints were monitored over concentration and time. The concentration-response curves were compared using mean IC50 values. Beat rate (BR) was measured using the xCelligence RTCA Cardio system (ACEA). To test the model, hepatotoxic (camptothecin (CAMP), rotenone (ROT)) or cardiotoxic (doxorubicin (DOX), mitoxantrone (MTX)) compounds were used. These were added to cells at concentrations of 0, 0.1, 1, 5, 25, 50, and 300 µM and exposed for 24 hr. In addition, drugs known to produce QT prolongation (Terfenadine (T) and verapamil (V)) were included and the BR determined. T and V significantly reduced BR at a concentration of 0.3 µM. To determine heart (H) or liver (L) specificity, average IC50 values were obtained for each cell type and the H-to-L ratio calculated. H-to-L ratios were MTX = 0.5, DOX = 2.6, ROT = 0.02 and CAMP undetermined. Ratios < 1.0 indicate cardiotoxicity, those >1.0 but < 3 indicate toxicity in both models, while ratios >3 indicate hepatotoxicity. BR data was used to improve these predictions. The combined data provided better resolution and enabled cardiac toxicity to be determined with greater confidence.

Poster: 617 | Abstract 1556
Multi-parameter In Vitro Toxicity Testing of Crizotinib, Sunitinib, Erlotinib, and Nilotinib in Human Cardiomyocytes
K Doherty, R.L Wappel, DR Talbert, PB Trusk, DM Moran, JW Kramer AM Brown SA Shell,
S Bacus

Quintiles Transnational Corp, Westmont, IL

Abstract:
Targeted therapy has greatly improved the treatment and prognosis of multiple types of cancer. However, unexpected cardiotoxicity has arisen in a subset of patients for some of the tyrosine kinase inhibitors (TKi). For these TKi, the cardiotoxicity was not wholly predicted by pre-clinical testing, which centers around the inhibition of the human Ether-a-go-go-Related Gene (hERG) channel. Therefore, we sought to determine whether a multi-parameter panel of tests that assesses a drug's effect on cellular, molecular, and electrophysiological endpoints would more accurately predict cardiotoxicity. To do so, we examined how 4 FDA-approved drugs impacted cell viability, apoptosis, reactive oxygen species (ROS) generation, metabolic status, impedance, and ion channel function in human cardiomyocytes. The 3 drugs with known associated cardiac adverse events (crizotinib, sunitinib, and nilotinib) all proved to be cardiotoxic in our series of in vitro tests while erlotinib, a cardiac-safe drug, did not show any indications of toxicity. Crizotinib, an ALKI MET inhibitor, was the most cardiotoxic by our panel, leading to increased ROS production, caspase activation, cholesterol accumulation, and a significant disruption in cardiac cell beat rate and blockage of ion channels. The multi-targeted TKl sunitinib also demonstrated severe cardiotoxicity in our tests, showing decreased cardiomyocyte viability, inhibition of AMPK, increased lipid droplet accumulation, disrupted beat pattern, and hERG block. Nilotinib, a second generation Bcr-Abl inhibitor, led to increased ROS generation, caspase activation, hERG block, and an arrhythmic beat pattern. Thus, each drug showed a unique toxicity profile, demonstrating that a multi-parameter approach allows for a more complete assessment of the potential for drug-induced cardiotoxicity and may allow for earlier detection in the drug development process.