Ryanodine receptors (RyR) and IP3 receptors (IP3R) are Ca2+ release channels expressed on the endoplasmic/sarcoplasmic reticulum (ER/SR) membrane in various cell types. Both the spatial localization and the distinct gating properties of these channels contribute to the diverse cellular functions controlled by intracellular Ca2+ signaling. It is known that both RyR2s and IP3R2s are expressed on the SR membrane of ventricular cardiomyocytes and that the expression of IP3R2s on the SR is increased in cardiac diseases such as heart failure (HF), and evidence that Ca2+ release through IP3R2s can influence RyR2-mediated Ca2+ release in excitation-contraction coupling has been described. However, despite the suggested functional role for crosstalk between RyR2s and IP3R2s, especially under pathologic conditions, most previous mathematical models of cardiomyocyte Ca2+ signaling have accounted for only RyR2s in isolation. We hypothesized that the combined effects of (1) fragmentation and dispersion of RyR2s within calcium release units (CRUs) and (2) increased expression of IP3R2s that occur in HF promote pro-arrhythmic Ca2+ spark behavior, which may contribute to increased risk of arrhythmogenic Ca2+ wave formation and incidence of ventricular arrhythmias. We built a stochastic mathematical model of local SR Ca2+ release events—Ca2+ sparks—that incorporates both RyR2s and IP3R2s. This model considers the spatial arrangement of RyR2s and IP3R2s relative to one another based on published immunohistochemistry studies and the arrangement of RyR2s under HF and healthy control conditions based on super-resolution microscopy data. RyR2 and IP3R2 gating are modeled based on single channel patch clamp studies which show that (1) RyR2 gating is stochastic and depends on local cytosolic [Ca2+], JSR [Ca2+], and allosteric coupling, (2) IP3R2 gating is stochastic and depends primarily on local cytosolic [Ca2+] and [IP3], and the (3) RyR2 has a larger single channel Ca2+ current than the IP3R2. Our simulations show that Ca2+ spark probability increases with increasing IP3R2 expression in HF CRUs and IP3R2 expression mitigates differences in mean duration of and mean total Ca2+ released during Ca2+ sparks observed in simulations in which HF is modeled as fragmentation and dispersion of RyR2s within CRUs alone. Overall, this mathematical modeling study suggests that increased IP3R2 expression in the context of HF may contribute to pro-arrhythmic Ca2+ signaling via increased Ca2+ spark frequency but may also serve a compensatory function by countering changes in Ca2+ spark morphology that arise due to RyR2 remodeling within CRUs in HF.

The Library of Integrated Network-based Cellular Signatures (LINCS), an NIH Common Fund program, has cataloged and analyzed cellular function and molecular activity profiles in response to >80,000 perturbing agents that are potentially disruptive to cells. Because of the importance of proteins and their modifications to the response of specific cellular perturbations, four of the six LINCS centers have included significant proteomics efforts in the characterization of the resulting phenotype. This manuscript aims to describe this effort and the data harmonization and integration of the LINCS proteomics data discussed in recent LINCS papers.

Knowledge in cardiac development, heart disease and drug-induced toxicity has steadily progressed for centuries, but the most recent decades have seen an explosion in technological advancements that have benefited cardiac research. In particular, the development of induced pluripotent stem cells (iPSCs) derived from accessible human adult tissues, as well as lineage-specific cell cultures differentiated from these iPSCs, has led to the rapid growth of the iPSC-derived cardiomyocyte (iPSC-CM) as a promising in vitro model. However, major differences in iPSC-CM phenotype have been observed across studies. This variability may be attributed to differences in cardiomyocyte differentiation protocols, maturation efficiency, or iPSC donor genetic background. While phenotypic heterogeneity is an important aspect of modelling a population as diverse as humans, it can also confound research study interpretation and reproducibility. Computational models of iPSC-CM physiology provide a potential avenue for assessing the mechanisms behind varied phenotypes and responses without sacrificing the valuable information this heterogeneity provides. Recently, new developments in the calibration of mechanistic models have aided in the generation of patient- or cell line-specific computational models, which hold potential in benchmarking iPSC-CM preparations. In this review, we summarize recent literature on iPSC-CM heterogeneity and computational model calibration, and we emphasize the utility of integrating computational ('dry lab') models with information from experimental ('wet lab') datasets in future iPSC-CM studies.

Human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) have gained traction as a powerful model in cardiac disease and therapeutics research, since iPSCs are self-renewing and can be derived from healthy and diseased patients without invasive surgery. However, current iPSC-CM differentiation methods produce cardiomyocytes with immature, fetal-like electrophysiological phenotypes, and the variety of maturation protocols in the literature results in phenotypic differences between labs. Heterogeneity of iPSC donor genetic backgrounds contributes to additional phenotypic variability. Several mathematical models of iPSC-CM electrophysiology have been developed to help to predict cell responses, but these models individually do not capture the phenotypic variability observed in iPSC-CMs. Here, we tackle these limitations by developing a computational pipeline to calibrate cell preparation-specific iPSC-CM electrophysiological parameters. We used the genetic algorithm (GA), a heuristic parameter calibration method, to tune ion channel parameters in a mathematical model of iPSC-CM physiology. To systematically optimize an experimental protocol that generates sufficient data for parameter calibration, we created in silico datasets by simulating various protocols applied to a population of models with known conductance variations, and then fitted parameters to those datasets. We found that calibrating to voltage and calcium transient data under 3 varied experimental conditions, including electrical pacing combined with ion channel blockade and changing buffer ion concentrations, improved model parameter estimates and model predictions of unseen channel block responses. This observation also held when the fitted data were normalized, suggesting that normalized fluorescence recordings, which are more accessible and higher throughput than patch clamp recordings, could sufficiently inform conductance parameters. Therefore, this computational pipeline can be applied to different iPSC-CM preparations to determine cell line-specific ion channel properties and understand the mechanisms behind variability in perturbation responses.

BACKGROUND: Germline HRAS gain-of-function pathogenic variants cause Costello syndrome (CS). During early childhood, 50% of patients develop multifocal atrial tachycardia, a treatment-resistant tachyarrhythmia of unknown pathogenesis. This study investigated how overactive HRAS activity triggers arrhythmogenesis in atrial-like cardiomyocytes (ACMs) derived from human-induced pluripotent stem cells bearing CS-associated HRAS variants. METHODS: HRAS Gly12 mutations were introduced into a human-induced pluripotent stem cells-ACM reporter line. Human-induced pluripotent stem cells were generated from patients with CS exhibiting tachyarrhythmia. Calcium transients and action potentials were assessed in induced pluripotent stem cell-derived ACMs. Automated patch clamping assessed funny currents. HCN inhibitors targeted pacemaker-like activity in mutant ACMs. Transcriptomic data were analyzed via differential gene expression and gene ontology. Immunoblotting evaluated protein expression associated with calcium handling and pacemaker-nodal expression. RESULTS: ACMs harboring HRAS variants displayed higher beating rates compared with healthy controls. The hyperpolarization activated cyclic nucleotide gated potassium channel inhibitor ivabradine and the Na v 1.5 blocker flecainide significantly decreased beating rates in mutant ACMs, whereas voltage-gated calcium channel 1.2 blocker verapamil attenuated their irregularity. Electrophysiological assessment revealed an increased number of pacemaker-like cells with elevated funny current densities among mutant ACMs. Mutant ACMs demonstrated elevated gene expression (ie, ISL1 , TBX3 , TBX18 ) related to intracellular calcium homeostasis, heart rate, RAS signaling, and induction of pacemaker-nodal-like transcriptional programming. Immunoblotting confirmed increased protein levels for genes of interest and suppressed MAPK (mitogen-activated protein kinase) activity in mutant ACMs. CONCLUSIONS: CS-associated gain-of-function HRAS G12 mutations in induced pluripotent stem cells-derived ACMs trigger transcriptional changes associated with enhanced automaticity and arrhythmic activity consistent with multifocal atrial tachycardia. This is the first human-induced pluripotent stem cell model establishing the mechanistic basis for multifocal atrial tachycardia in CS.

Drug-induced gene expression profiles can identify potential mechanisms of toxicity. We focus on obtaining signatures for cardiotoxicity of FDA-approved tyrosine kinase inhibitors (TKIs) in human induced-pluripotent-stem-cell-derived cardiomyocytes, using bulk transcriptomic profiles. We use singular value decomposition to identify drug-selective patterns across cell lines obtained from multiple healthy human subjects. Cellular pathways affected by cardiotoxic TKIs include energy metabolism, contractile, and extracellular matrix dynamics. Projecting these pathways to published single cell expression profiles indicates that TKI responses can be evoked in both cardiomyocytes and fibroblasts. Integration of transcriptomic outlier analysis with whole genomic sequencing of our six cell lines enables us to correctly reidentify a genomic variant causally linked to anthracycline-induced cardiotoxicity and predict genomic variants potentially associated with TKI-induced cardiotoxicity. We conclude that mRNA expression profiles when integrated with publicly available genomic, pathway, and single cell transcriptomic datasets, provide multiscale signatures for cardiotoxicity that could be used for drug development and patient stratification. Using a new computational pipeline for identification of drug-selective transcriptomic responses and FAERS data, the authors identified potential pathways and genomic variants indicative of cancer drug cardiotoxicity in iPSC-derived cardiomyocytes.

Human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) have gained traction as a powerful model in cardiac disease and therapeutics research, since iPSCs are self-renewing and can be derived from healthy and diseased patients without invasive surgery. However, current iPSC-CM differentiation methods produce cardiomyocytes with immature, fetal-like electrophysiological phenotypes, and the variety of maturation protocols in the literature results in phenotypic differences between labs. Heterogeneity of iPSC donor genetic backgrounds contributes to additional phenotypic variability. Several mathematical models of iPSC-CM electrophysiology have been developed to help understand the ionic underpinnings of, and to simulate, various cell responses, but these models individually do not capture the phenotypic variability observed in iPSC-CMs. Here, we tackle these limitations by developing a computational pipeline to calibrate cell preparation-specific iPSC-CM electrophysiological parameters. We used the genetic algorithm (GA), a heuristic parameter calibration method, to tune ion channel parameters in a mathematical model of iPSC-CM physiology. To systematically optimize an experimental protocol that generates sufficient data for parameter calibration, we created simulated datasets by applying various protocols to a population of in silico cells with known conductance variations, and we fitted to those datasets. We found that calibrating models to voltage and calcium transient data under 3 varied experimental conditions, including electrical pacing combined with ion channel blockade and changing buffer ion concentrations, improved model parameter estimates and model predictions of unseen channel block responses. This observation held regardless of whether the fitted data were normalized, suggesting that normalized fluorescence recordings, which are more accessible and higher throughput than patch clamp recordings, could sufficiently inform conductance parameters. Therefore, this computational pipeline can be applied to different iPSC-CM preparations to determine cell line-specific ion channel properties and understand the mechanisms behind variability in perturbation responses.

Cardiac Purkinje fibers form the most distal part of the ventricular conduction system. They coordinate contraction and play a key role in ventricular arrhythmias. While many cardiac cell types can be generated from human pluripotent stem cells, methods to generate Purkinje fiber cells remain limited, hampering our understanding of Purkinje fiber biology and conduction system defects. To identify signaling pathways involved in Purkinje fiber formation, we analyzed single cell data from murine embryonic hearts and compared Purkinje fiber cells to trabecular cardiomyocytes. This identified several genes, processes, and signaling pathways putatively involved in cardiac conduction, including Notch signaling. We next tested whether Notch activation could convert human pluripotent stem cell-derived cardiomyocytes to Purkinje fiber cells. Following Notch activation, cardiomyocytes adopted an elongated morphology and displayed altered electrophysiological properties including increases in conduction velocity, spike slope, and action potential duration, all characteristic features of Purkinje fiber cells. RNA-sequencing demonstrated that Notch-activated cardiomyocytes undergo a sequential transcriptome shift, which included upregulation of key Purkinje fiber marker genes involved in fast conduction such as SCN5A, HCN4 and ID2, and downregulation of genes involved in contractile maturation. Correspondingly, we demonstrate that Notch-induced cardiomyocytes have decreased contractile force in bioengineered tissues compared to control cardiomyocytes. We next modified existing in silico models of human pluripotent stem cell-derived cardiomyocytes using our transcriptomic data and modeled the effect of several anti-arrhythmogenic drugs on action potential and calcium transient waveforms. Our models predicted that Purkinje fiber cells respond more strongly to dofetilide and amiodarone, while cardiomyocytes are more sensitive to treatment with nifedipine. We validated these findings in vitro, demonstrating that our new cell-specific in vitro model can be utilized to better understand human Purkinje fiber physiology and its relevance to disease.

Determining whether an ectopic depolarization will lead to a self-perpetuating arrhythmia is of critical importance in determining arrhythmia risk, so it is necessary to understand what factors impact substrate vulnerability. This study sought to explore the impact of cell-to-cell heterogeneity in ion channel conductance on substrate vulnerability to arrhythmia by measuring the duration of the vulnerable window in computational models of one-dimensional cables of ventricular cardiomyocytes. We began by using a population of uniform cable models to determine the mechanisms underlying the vulnerable window phenomenon. We found that in addition to the known importance of GNa, the conductances GCa,L and GKr also play a minor role in determining the vulnerable window duration. We also found that a steeper slope of the repolarizing action potential during the vulnerable window correlated with a shorter vulnerable window duration in uniform cables. We applied our understanding from these initial simulations to an investigation of the vulnerable window in heterogeneous cable models. The heterogeneous cables displayed a great deal of intra-cable variation in vulnerable window duration, highly sensitive to the cardiomyocytes in the local environment of the ectopic stimulus. Coupling strength modulated not only the magnitude of the vulnerable window duration but also the extent of intra-tissue variability in vulnerable window duration.NEW & NOTEWORTHY We investigate the impact of cell-to-cell heterogeneity in ion channel conductance on substrate vulnerability to arrhythmia by measuring the vulnerable window duration in computational cardiomyocyte cable models. We demonstrate a wide range of intra-cable variability in vulnerable window duration (VWD) and show how this is changed by ion channel block and coupling strength perturbations.

Introduction: Tyrosine kinase inhibitor drugs (TKIs) are highly effective cancer drugs, yet many TKIs are associated with various forms of cardiotoxicity. The mechanisms underlying these drug-induced adverse events remain poorly understood. We studied mechanisms of TKI-induced cardiotoxicity by integrating several complementary approaches, including comprehensive transcriptomics, mechanistic mathematical modeling, and physiological assays in cultured human cardiac myocytes.Methods: Induced pluripotent stem cells (iPSCs) from two healthy donors were differentiated into cardiac myocytes (iPSC-CMs), and cells were treated with a panel of 26 FDA-approved TKIs. Drug-induced changes in gene expression were quantified using mRNA-seq, changes in gene expression were integrated into a mechanistic mathematical model of electrophysiology and contraction, and simulation results were used to predict physiological outcomes.Results: Experimental recordings of action potentials, intracellular calcium, and contraction in iPSC-CMs demonstrated that modeling predictions were accurate, with 81% of modeling predictions across the two cell lines confirmed experimentally. Surprisingly, simulations of how TKI-treated iPSC-CMs would respond to an additional arrhythmogenic insult, namely, hypokalemia, predicted dramatic differences between cell lines in how drugs affected arrhythmia susceptibility, and these predictions were confirmed experimentally. Computational analysis revealed that differences between cell lines in the upregulation or downregulation of particular ion channels could explain how TKI-treated cells responded differently to hypokalemia.Discussion: Overall, the study identifies transcriptional mechanisms underlying cardiotoxicity caused by TKIs, and illustrates a novel approach for integrating transcriptomics with mechanistic mathematical models to generate experimentally testable, individual-specific predictions of adverse event risk.

Ryanodine receptors (RyR) and IP3 receptors (IP3R) are Ca2+ channels expressed on the endoplasmic/sarcoplasmic reticulum (ER/SR) membrane in various cell types. Both the spatial localization and the distinct gating properties of these channels contribute to the diverse cellular functions controlled by intracellular Ca2+ signaling. It is known that both RyRs and IP3Rs are expressed on the SR membrane of ventricular cardiomyocytes and that the expression of IP3Rs on the SR at dyadic junctions is increased in cardiac diseases such as hypertrophy and heart failure (HF), and evidence that Ca2+ release through IP3Rs can influence RyR-mediated Ca2+ release has been described. However, despite the suggested functional role for crosstalk between RyRs and IP3Rs, especially under pathological conditions, most previous mathematical models of cardiomyocyte Ca2+ signaling have accounted for only RyRs in isolation. Here, we propose a mathematical model of intracellular Ca2+ signaling that incorporates both RyRs and IP3Rs and can be used to develop quantitative predictions about crosstalk between the two channels. This model considers the spatial arrangement of RyRs and IP3Rs relative to one another based on published immunohistochemistry co-localization studies and simulates the stochastic opening and closing of individual receptors based on previously published models of Ca2+ sparks and puffs. In this model, RyR gating depends on local cytosolic [Ca2+], JSR [Ca2+], and allosteric coupling to neighboring RyRs while IP3R gating depends primarily on local cytosolic [Ca2+] and [IP3]. Based on preliminary results, we hypothesize that the subcellular spatial remodeling and increased expression of IP3Rs that occur in hypertrophy and HF will promote slowed, uncoordinated diastolic Ca2+ sparks, which may contribute to increased risk of arrhythmogenic Ca2+ wave formation and incidence of ventricular arrhythmias observed clinically in patients with hypertrophy and HF.

STRUCTURE: Mavacamten, 3-(1-methylethyl)-6-[[(1S)-1-phenylethyl]amino]-2,4(1H,3H)-pyrimidinedione, is a reversible small-molecule allosteric inhibitor of cardiac myosin ATPase with a molecular formula of C15H19N3O2 and molecular weight of 273.33 g/mol. It belongs to the pyrimidinedione drug class.

Pluripotent stem-cell-derived cardiomyocytes (PSC-CMs) provide an unprecedented opportunity to study human heart development and disease, but they are functionally and structurally immature. Here, we induce efficient human PSC-CM (hPSC-CM) maturation through metabolic-pathway modulations. Specifically, we find that peroxisome-proliferator-associated receptor (PPAR) signaling regulates glycolysis and fatty acid oxidation (FAO) in an isoform-specific manner. While PPARalpha (PPARa) is the most active isoform in hPSC-CMs, PPARdelta (PPARd) activation efficiently upregulates the gene regulatory networks underlying FAO, increases mitochondrial and peroxisome content, enhances mitochondrial cristae formation, and augments FAO flux. PPARd activation further increases binucleation, enhances myofibril organization, and improves contractility. Transient lactate exposure, which is frequently used for hPSC-CM purification, induces an independent cardiac maturation program but, when combined with PPARd activation, still enhances oxidative metabolism. In summary, we investigate multiple metabolic modifications in hPSC-CMs and identify a role for PPARd signaling in inducing the metabolic switch from glycolysis to FAO in hPSC-CMs.

Drug Toxicity Signature Generation Center (DToxS) at the Icahn School of Medicine at Mount Sinai is one of the centers for the NIH Library of Integrated Network-Based Cellular Signatures (LINCS) program. Its key aim is to generate proteomic and transcriptomic signatures that can predict cardiotoxic adverse effects of kinase inhibitors approved by the Food and Drug Administration. Towards this goal, high throughput shotgun proteomics experiments (308 cell line/drug combinations +64 control lysates) have been conducted. Using computational network analyses, these proteomic data can be integrated with transcriptomic signatures, generated in tandem, to identify cellular signatures of cardiotoxicity that may predict kinase inhibitor-induced toxicity and enable possible mitigation. Both raw and processed proteomics data have passed several quality control steps and been made publicly available on the PRIDE database. This broad protein kinase inhibitor-stimulated human cardiomyocyte proteomic data and signature set is valuable for prediction of drug toxicities.

Abstract Background Human-induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CM) are a promising disease model, even though hiPSC-CMs cultured for extended periods display an undifferentiated transcriptional landscape. MiRNA–target gene interactions contribute to fine-tuning the genetic program governing cardiac maturation and may uncover critical pathways to be targeted. Methods We analyzed a hiPSC-CM public dataset to identify time-regulated miRNA–target gene interactions based on three logical steps of filtering. We validated this process in silico using 14 human and mouse public datasets, and further confirmed the findings by sampling seven time points over a 30-day protocol with a hiPSC-CM clone developed in our laboratory. We then added miRNA mimics from the top eight miRNAs candidates in three cell clones in two different moments of cardiac specification and maturation to assess their impact on differentiation characteristics including proliferation, sarcomere structure, contractility, and calcium handling. Results We uncovered 324 interactions among 29 differentially expressed genes and 51 miRNAs from 20,543 transcripts through 120 days of hiPSC-CM differentiation and selected 16 genes and 25 miRNAs based on the inverse pattern of expression (Pearson R-values < − 0.5) and consistency in different datasets. We validated 16 inverse interactions among eight genes and 12 miRNAs (Person R-values < − 0.5) during hiPSC-CMs differentiation and used miRNAs mimics to verify proliferation, structural and functional features related to maturation. We also demonstrated that miR-124 affects Ca2+ handling altering features associated with hiPSC-CMs maturation. Conclusion We uncovered time-regulated transcripts influencing pathways affecting cardiac differentiation/maturation axis and showed that the top-scoring miRNAs indeed affect primarily structural features highlighting their role in the hiPSC-CM maturation.

Both experimental and modeling studies have attempted to determine mechanisms by which a small anatomical region, such as the sinoatrial node (SAN), can robustly drive electrical activity in the human heart. However, despite many advances from prior research, important questions remain unanswered. This study aimed to investigate, through mathematical modeling, the roles of intercellular coupling and cellular heterogeneity in synchronization and pacemaking within the healthy and diseased SAN. In a multicellular computational model of a monolayer of either human or rabbit SAN cells, simulations revealed that heterogenous cells synchronize their discharge frequency into a unique beating rhythm across a wide range of heterogeneity and intercellular coupling values. However, an unanticipated behavior appeared under pathological conditions where perturbation of ionic currents led to reduced excitability. Under these conditions, an intermediate range of intercellular coupling (900-4000 MΩ) was beneficial to SAN automaticity, enabling a very small portion of tissue (3.4%) to drive propagation, with propagation failure occurring at both lower and higher resistances. This protective effect of intercellular coupling and heterogeneity, seen in both human and rabbit tissues, highlights the remarkable resilience of the SAN. Overall, the model presented in this work allowed insight into how spontaneous beating of the SAN tissue may be preserved in the face of perturbations that can cause individual cells to lose automaticity. The simulations suggest that certain degrees of gap junctional coupling protect the SAN from ionic perturbations that can be caused by drugs or mutations.

As one of three journals within the ASCPT family of publications, CPT:PSP is committed to publishing top-quality work that applies quantitative methods to important issues in drug discovery and development. As the full name of the journal suggests, pharmacometrics and systems pharmacology were originally envisioned as the two central approaches;1 however, recent years have seen the journal’s scope expand to include methods beyond those core approaches, including advanced statistical analysis, construction of biological networks, and machine learning (ML)/artificial intelligence.2 With respect to applications, articles published in CPT:PSP cover a wide range of drugs, both approved and in development, and a wide range of disease and treatment modalities. Because the content included in the journal is quite broad, it is useful to occasionally collect articles that share a common theme, in order to gain an appreciation of new developments within a particular field. CPT:PSP has long done this through the creation of “virtual issues” that are posted online, and more recently with the publication of special issues related to a particular topic. After 2021’s highly successful inaugural special issue on Pharmacometrics and Statistics,3 we are happy to present the second annual special issue, which addresses “Quantitative Approaches to Drug Safety.” The Drug Safety special issue contains a variety of original research articles that collectively demonstrate how several different quantitative strategies can be applied to understand, predict, and ultimately prevent a wide range of adverse events. These are complemented by a review article that provides useful “big picture” context,4 and two brief Perspectives that describe somewhat specialized issues that are sometimes overlooked in classical toxicity studies. These Perspectives cover drugs potentially carried in breast milk by nursing mothers5 and toxicity caused by snake bites.6 A brief description of the research articles, grouped thematically, is intended to provide a useful overview of the various techniques and applications that are showcased in our special issue. It has long been appreciated that drugs that cause beneficial effects at the correct doses may cause extremely undesirable effects at excessively high concentrations. After all, it was in the 16th century that Paracelsus, sometimes referred to as the “Father of Toxicology,” said, “Solely the dose determines that a thing is not a poison.” Accordingly, a substantial amount of work in drug safety is aimed at determining dosing schedules that will maximize efficacy while avoiding potential adverse events. Because population pharmacokinetics and pharmacodynamics (PopPK and PopPD) are explicitly designed to address dosing questions, it should come as no surprise that several articles in the special issue employ such a strategy. For example, Araki et al.7 used PopPK/PopPD to address the effects of TAS-114, a compound in development that is intended to improve the therapeutic index of capecitabine. The latter drug is a chemotherapeutic that can sometimes cause hand-foot syndrome, a form of dermatitis, and the authors explore how concurrent treatment with TAS-114 may improve tolerance of capecitabine in some patients. Similarly, Keutzer et al.8 use PopPK modeling to examine bedaquiline, a drug with a long terminal half-life that is used for treatment of drug-resistant tuberculosis. When bedaquiline is reintroduced after dose interruption, concerns exist about the risk of QT-prolongation and potential ventricular arrhythmias, and the results presented by the authors demonstrate a potential strategy for safe reintroduction of the drug. Along similar lines, Marco-Ariño et al.9 use a PopPD model to address pupillary reflex dilation caused by remifentanil, an opioid used as an analgesic during surgery, and Dosne et al.10 use PK/PD modeling for individualized dosing of erdafitinib, a drug that can treat urothelial carcinoma but can cause acute hyperphosphatemia in some patients. Together, these articles demonstrate the centrality of pharmacometrics approaches in ensuring that highly useful drugs do not cause adverse events due to improper dosing. Although PK/PD models remain the gold standard when a drug has been administered to many patients and target concentration ranges have been established, sometimes, particularly early in the drug development pipeline, investigators are unsure about the potential for a new compound to cause adverse events because biological mechanisms may be incompletely understood. In such cases, quantitative systems pharmacology (QSP) models, which incorporate cellular mechanisms and frequently also consider interactions between organ systems, can be highly useful for suggesting adverse event mechanisms and guiding future work intended to minimize toxicity while maintaining efficacy.11, 12 The special issue also contains strong examples of how models based on mechanism may be used to distinguish between alternative compounds, thereby nudging drug development into a direction that is more likely to be successful. For example, Gadkar et al.13 present a QSP model that simulates how T-cell proliferation may affect epithelial barrier integrity in the gastrointestinal (GI) tract, potentially leading to adverse events, such as colitis. By using this model to examine PI3-kinase inhibitors that have differential selectivity for various PI3-kinase isoforms, they suggest ways to avoid colitis in patients through targeting of particular isoforms. Similarly, Fu et al.14 expand on previous research from their group by presenting a QSP model of the cardiovascular system that accounts for interactions among relevant biomarkers, such as heart rate, arterial blood pressure, and cardiac output. Through validating the model with a proof-of-concept drug, the β-blocker atenolol, they establish a platform that can be used to minimize the possibility that developmental compounds may cause adverse cardiac events. Although the studies mentioned above illustrate the wide range of potential drug toxicities that can be explored, readers of CPT:PSP are fortunate that two interesting articles addressed thrombocytopenia, or depletion of blood platelets, which can result from several types of cancer therapeutics. This provides an opportunity to compare and contrast the approaches taken by the authors of the two papers. For example, Krishnatry et al.15 used PK/PD modeling to address how molibresib, a bromodomain inhibitor, may lead to thrombocytopenia, QT prolongation, and GI events. The strategy taken by Lignet et al.,16 who were interested in the effects of three pan-proteasome inhibitors, shared significant similarity with respect to the treatment of drug PKs but included somewhat more detail with respect to how the drugs may affect both proliferation of platelet progenitor cells and platelet budding. The comparison between the two studies therefore reveals how the specific mechanisms included within a model depend on the questions being addressed, consistent with the philosophy that a model should be as simple as possible, but no simpler. Finally, it is worthwhile to highlight a couple of publications that discuss and use methods that are not typically encountered in CPT:PSP. The review article by Soldatos et al.4 advocates for the construction of biological networks through the integration of molecular data (such as target lists or drug-induced changes in gene expression) with adverse event reports, often obtained through sophisticated text mining strategies. Along similar lines, in the paper by Jeong et al.,17 the authors use mechanistic simulation results to build a classifier using a convolutional neural network. This strategy of combining mechanistic and ML approaches is currently receiving considerable attention,18, 19 and the Jeong et al. study provides a nice example of the potential benefits. Both approaches—network analysis and ML classifiers—are currently active areas of research that are likely to become increasingly important in the coming years as our understanding improves. When considered collectively, therefore, the articles in the Drug Safety special issue demonstrate a wide range of approaches to a variety of adverse events caused by several different drugs. Together the research studies, along with the review and the two perspectives,5, 6 show how creative quantitative approaches are becoming increasingly important in all phases of drug development, not only for preclinical and clinical predictions of absorption, distribution, metabolism, and excretion and efficacy, but also for questions related to drug safety. The author declares that no conflicts of interest exist. Research in Dr. Sobie’s laboratory is funded by the National Heart Lung and Blood Institute (U01 HL 136297 and R44HL139248) and the US Food and Drug Administration (75F40119C10021).

Drug-induced gene expression profiles are an important source for the characterization of drug-specific mechanisms of action that might also indicate potential mechanisms of drug toxicity. Unfortunately, drug-induced transcriptomic signatures are often the sum of multiple responses, such as cell line or cell-type specific responses, and can include false positive results. Both issues complicate the identification of drug-specific effects. To unmask drug-specific effects in drug-induced transcriptomic signatures, we use singular value decomposition to characterize shared transcriptomic responses induced in multiple cell lines treated with the same drug. Six different cardiomyocyte cell lines that were generated from induced pluripotent stem cells obtained from six healthy human subjects were stimulated with 25 protein kinase inhibitors, 4 monoclonal antibodies against protein kinases and 25 other cardiac- and non-cardiac acting drugs. Searching for drug-specific subspaces that characterize drug-specific responses in the original data, we could identify highly similar drug responses for multiple drugs. Pathway enrichment analysis of those responses predicts reasonable effects for multiple drugs. Revealing drug-specific responses from transcriptomic signatures might allow an easier characterization of drug-induced toxicity, such as cardiotoxic side effects induced by multiple protein kinase inhibitors and monoclonal antibodies against protein kinases.

During morphogenesis, molecular mechanisms that orchestrate biomechanical dynamics across cells remain unclear. Here, we show a role of guidance receptor Plexin-B2 in organizing actomyosin network and adhesion complexes during multicellular development of human embryonic stem cells and neuroprogenitor cells. Plexin-B2 manipulations affect actomyosin contractility, leading to changes in cell stiffness and cytoskeletal tension, as well as cell-cell and cell-matrix adhesion. We have delineated the functional domains of Plexin-B2, RAP1/2 effectors, and the signaling association with ERK1/2, calcium activation, and YAP mechanosensor, thus providing a mechanistic link between Plexin-B2-mediated cytoskeletal tension and stem cell physiology. Plexin-B2-deficient stem cells exhibit premature lineage commitment, and a balanced level of Plexin-B2 activity is critical for maintaining cytoarchitectural integrity of the developing neuroepithelium, as modeled in cerebral organoids. Our studies thus establish a significant function of Plexin-B2 in orchestrating cytoskeletal tension and cell-cell/cell-matrix adhesion, therefore solidifying the importance of collective cell mechanics in governing stem cell physiology and tissue morphogenesis.