Identification of a novel set of biomarkers for evaluating phospholipidosis-inducing potential of compounds using rat liver microarray data measured 24-h after single dose administration

In Vitro Models for Studying Chronic Drug-Induced Liver Injury

Drug-induced liver injury (DILI) is a major clinical problem in terms of patient morbidity and mortality, cost to healthcare systems and failure of the development of new drugs. The need for consistent safety strategies capable of identifying a potential toxicity risk early in the drug discovery pipeline is key. Human DILI is poorly predicted in animals, probably due to the well-known interspecies differences in drug metabolism, pharmacokinetics, and toxicity targets. For this reason, distinct cellular models from primary human hepatocytes or hepatoma cell lines cultured as 2D monolayers to emerging 3D culture systems or the use of multi-cellular systems have been proposed for hepatotoxicity studies. In order to mimic long-term hepatotoxicity in vitro, cell models, which maintain hepatic phenotype for a suitably long period, should be used. On the other hand, repeated-dose administration is a more relevant scenario for therapeutics, providing information not only about toxicity, but also about cumulative effects and/or delayed responses. In this review, we evaluate the existing cell models for DILI prediction focusing on chronic hepatotoxicity, highlighting how better characterization and mechanistic studies could lead to advance DILI prediction.

Predicting volume of distribution with decision tree-based regression methods using predicted tissue:plasma partition coefficients

BACKGROUND

Volume of distribution is an important pharmacokinetic property that indicates the extent of a drug's distribution in the body tissues. This paper addresses the problem of how to estimate the apparent volume of distribution at steady state (Vss) of chemical compounds in the human body using decision tree-based regression methods from the area of data mining (or machine learning). Hence, the pros and cons of several different types of decision tree-based regression methods have been discussed. The regression methods predict Vss using, as predictive features, both the compounds' molecular descriptors and the compounds' tissue:plasma partition coefficients (Kt:p) - often used in physiologically-based pharmacokinetics. Therefore, this work has assessed whether the data mining-based prediction of Vss can be made more accurate by using as input not only the compounds' molecular descriptors but also (a subset of) their predicted Kt:p values.

RESULTS

Comparison of the models that used only molecular descriptors, in particular, the Bagging decision tree (mean fold error of 2.33), with those employing predicted Kt:p values in addition to the molecular descriptors, such as the Bagging decision tree using adipose Kt:p (mean fold error of 2.29), indicated that the use of predicted Kt:p values as descriptors may be beneficial for accurate prediction of Vss using decision trees if prior feature selection is applied.

CONCLUSIONS

Decision tree based models presented in this work have an accuracy that is reasonable and similar to the accuracy of reported Vss inter-species extrapolations in the literature. The estimation of Vss for new compounds in drug discovery will benefit from methods that are able to integrate large and varied sources of data and flexible non-linear data mining methods such as decision trees, which can produce interpretable models. Graphical AbstractDecision trees for the prediction of tissue partition coefficient and volume of distribution of drugs.

Putative biomarker for phospholipid accumulation in cultured cells treated with phospholipidosis-inducing drugs: alteration of the phosphatidylinositol composition detected using high-performance liquid chromatography-tandem mass spectrometry

We developed a high-performance liquid chromatography-tandem mass spectrometric method for phospholipid biomarker discovery and applied it to a cell-based assay system for the screening of phospholipidosis-inducing drugs. We studied the compositions of phospholipid molecules exceeding 100 species in cultured cells and found a characteristic alteration in the composition by treatment with cationic amphiphilic drugs possessing phospholipidosis-inducing potency. The compositions of phosphatidylinositol in RAW264 cells were significantly affected by the drug treatment. Similar alterations were also found in THP-1 cells. These phenomena were not observed when cells were treated with warfarin, which does not have phospholipidosis-inducing potency. Structural analysis of the altered phosphatidylinositols by a product ion scan revealed the presence of certain fatty acyl chains. Based on our findings, we proposed a prediction parameter (PP) for phospholipid accumulation calculated from the relative compositions of phosphatidylinositol species. As the dosage of imipramine (a cationic amphiphilic drug) increased, both the PP and cellular phospholipid content increased. Our results suggest that PP has potency as a biomarker for phospholipid accumulation in cells treated with drugs.

Glucosylceramide and lysophosphatidylcholines as potential blood biomarkers for drug-induced hepatic phospholipidosis

Drug-induced phospholipidosis is one of the major concerns in drug development and clinical treatment. The present study involved the use of a nontargeting lipidomic analysis with liquid chromatography-mass spectrometry to explore noninvasive blood biomarkers for hepatic phospholipidosis from rat plasma. We used three tricyclic antidepressants (clomipramine [CPM], imipramine [IMI], and amitriptyline [AMT]) for the model of phospholipidosis in hepatocytes and ketoconazole (KC) for the model of phospholipidosis in cholangiocytes and administered treatment for 3 and 28 days each. Total plasma lipids were extracted and measured. Lipid molecules contributing to the separation of control and drug-treated rat plasma in a multivariate orthogonal partial least squares discriminant analysis were identified. Four lysophosphatidylcholines (LPCs) (16:1, 18:1, 18:2, and 20:4) and 42:1 hexosylceramide (HexCer) were identified as molecules separating control and drug-treated rats in all models of phospholipidosis in hepatocytes. In addition, 16:1, 18:2, and 20:4 LPCs and 42:1 HexCer were identified in a model of hepatic phospholipidosis in cholangiocytes, although LPCs were identified only in the case of 3-day treatment with KC. The levels of LPCs were decreased by drug-induced phospholipidosis, whereas those of 42:1 HexCer were increased. The increase in 42:1 HexCer was much higher in the case of IMI and AMT than in the case of CPM; moreover, the increase induced by IMI was dose-dependent. Structural characterization determining long-chain base and hexose delineated that 42:1 HexCer was d18:1/24:0 glucosylceramide (GluCer). In summary, our study demonstrated that d18:1/24:0 GluCer and LPCs are potential novel biomarkers for drug-induced hepatic phospholipidosis.

Drug safety testing paradigm, current progress and future challenges: an overview

Early assessment of the toxicity potential of new molecules in pharmaceutical industry is a multi-dimensional task involving predictive systems and screening approaches to aid in the optimization of lead compounds prior to their entry into development phase. Due to the high attrition rate in the pharma industry in last few years, it has become imperative for the nonclinical toxicologist to focus on novel approaches which could be helpful for early screening of drug candidates. The need is that the toxicologists should change their classical approach to a more investigative approach. This review discusses the developments that allow toxicologists to anticipate safety problems and plan ways to address them earlier than ever before. This includes progress in the field of in vitro models, surrogate models, molecular toxicology, 'omics' technologies, translational safety biomarkers, stem-cell based assays and preclinical imaging. The traditional boundaries between teams focusing on efficacy/ safety and preclinical/ clinical aspects in the pharma industry are disappearing, and translational research-centric organizations with a focused vision of bringing drugs forward safely and rapidly are emerging. Today's toxicologist should collaborate with medicinal chemists, pharmacologists, and clinicians and these value-adding contributions will change traditional toxicologists from side-effect identifiers to drug development enablers.

Application of genomics, proteomics and metabolomics in drug discovery, development and clinic

Genomics, proteomics and metabolomics are three areas that are routinely applied throughout the drug-development process as well as after a product enters the market. This review discusses all three 'omics, reporting on the key applications, techniques, recent advances and expectations of each. Genomics, mainly through the use of novel and next-generation sequencing techniques, has advanced areas of drug discovery and development through the comparative assessment of normal and diseased-state tissues, transcription and/or expression profiling, side-effect profiling, pharmacogenomics and the identification of biomarkers. Proteomics, through techniques including isotope coded affinity tags, stable isotopic labeling by amino acids in cell culture, isobaric tags for relative and absolute quantification, multidirectional protein identification technology, activity-based probes, protein/peptide arrays, phage displays and two-hybrid systems is utilized in multiple areas through the drug development pipeline including target and lead identification, compound optimization, throughout the clinical trials process and after market analysis. Metabolomics, although the most recent and least developed of the three 'omics considered in this review, provides a significant contribution to drug development through systems biology approaches. Already implemented to some degree in the drug-discovery industry and used in applications spanning target identification through to toxicological analysis, metabolic network understanding is essential in generating future discoveries.

Identification of drugs inducing phospholipidosis by novel in vitro data

Drug-induced phospholipidosis (PLD) is a lysosomal storage disorder characterized by the accumulation of phospholipids within the lysosome. This adverse drug effect can occur in various tissues and is suspected to impact cellular viability. Therefore, it is important to test chemical compounds for their potential to induce PLD during the drug design process. PLD has been reported to be a side effect of many commonly used drugs, especially those with cationic amphiphilic properties. To predict drug-induced PLD in silico, we established a high-throughput cell-culture-based method to quantitatively determine the induction of PLD by chemical compounds. Using this assay, we tested 297 drug-like compounds at two different concentrations (2.5 μM and 5.0 μM). We were able to identify 28 previously unknown PLD-inducing agents. Furthermore, our experimental results enabled the development of a binary classification model to predict PLD-inducing agents based on their molecular properties. This random forest prediction system yields a bootstrapped validated accuracy of 86 %. PLD-inducing agents overlap with those that target similar biological processes; a high degree of concordance with PLD-inducing agents was identified for cationic amphiphilic compounds, small molecules that inhibit acid sphingomyelinase, compounds that cross the blood-brain barrier, and compounds that violate Lipinski's rule of five. Furthermore, we were able to show that PLD-inducing compounds applied in combination additively induce PLD.

Phospholipogenic pharmaceuticals are associated with a higher incidence of histological findings than nonphospholipogenic pharmaceuticals in preclinical toxicology studies

While phospholipidosis is thought to be an adaptive response to chemical challenge, many phospholipogenic compounds are known to display adverse effects in preclinical species and humans. To investigate the link between phospholipogenic administration and incidence of preclinical histological signals, an internal AstraZeneca in vivo toxicology report database was searched to identify phospholipogenic and nonphospholipogenic compounds. The datasets assembled comprised 46 phospholipogenic and 62 nonphospholipogenic compounds. The phospholipogenic potential of these compounds was confirmed by a pathologist's interpretation and was supported by well-validated in silico and in vitro models. The phospholipogenic dataset contained 37 bases, 4 neutral compounds, 3 zwitterions, and 1 acid, whereas the nonphospholipogenic dataset contained 9 bases, 34 neutrals, 1 zwitterion, and 18 acids. Histologic findings were tracked for adrenal gland; bone marrow; kidney; liver; lung; lymph node; spleen; thymus; and reproductive organs. On average, plasma exposures were higher in animals dosed with the nonphospholipogenics. Phospholipogenics yielded proportionally more histologic changes (exclusive of phospholipidosis itself) in all organs. Statistically significant higher frequencies of liver necrosis, alveolitis/pneumonitis, as well as lymphocytolysis in the thymus, lymph nodes, and spleen occurred in response to phospholipogenics compared to that for nonphospholipogenics.