Title: Rule based classifier for the analysis of gene-gene and gene-environment interactions in genetic association studies

Calcineurin inhibitor (CI) therapy has been associated with chronic nephrotoxicity, which limits its long-term utility for suppression of allograft rejection. In order to understand the mechanisms of the toxicity, we analyzed gene expression changes that underlie the development of CI immunosuppressant-mediated nephrotoxicity in male Sprague-Dawley rats dosed daily with cyclosporine (CsA; 2.5 or 25 mg/kg/day), FK506 (0.6 or 6 mg/kg/day), or rapamycin (1 or 10 mg/kg/day) for 1, 7, 14, or 28 days. A significant increase in blood urea nitrogen was observed in animals treated with CsA (high) or FK506 (high) for 14 and 28 days. Histopathological examination revealed tubular basophilia and mineralization in animals given CsA (high) or FK506 (low and high). We identified a group of genes whose expression in rat kidney is correlated with CI-induced kidney injury. Among these genes are two genes, Slc12a3 and kidney-specific Wnk1 (KS-Wnk1), that are known to be involved in sodium transport in the distal nephrons and could potentially be involved in the mechanism of CI-induced nephrotoxicity. The downregulation of NCC (the Na-Cl cotransporter coded by Slc12a3) in rat kidney following CI treatment was confirmed by immunohistochemical staining, and the downregulation of KS-Wnk1 was confirmed by quantitative real-time-polymerase chain reaction (qRT-PCR). We hypothesize that decreased expression of Slc12a3 and KS-Wnk1 could alter the sodium chloride reabsorption in the distal tubules and contribute to the prolonged activation of the renin-angiotensin system, a demonstrated contributor to the development of CI-induced nephrotoxicity in both animal models and clinical settings. Therefore, if validated as biomarkers in humans, SLC12A3 and KS-WNK1 could potentially be useful in the early detection and reduction of CI-related nephrotoxicity in immunosuppressed transplant patients when monitoring the health of kidney xenographs in clinical practice.

Objectives The aim of this analysis was to show the applicability of a newly developed algorithm to assess the influence of genetic variants and other covariates on nevirapine’s drug disposition. The algorithm combines high-throughput genotyping data and nonlinear mixed effects modeling methods. Methods Patients, who participated in the 2NN pharmacokinetic sub study, were reconsented and reenrolled into a clinical trial for genotyping analysis. Overall, 198 single nucleotide polymorphisms located in 45 absorption, distribution, metabolism, and elimination related genes were genotyped using the Illumina BeadArray technology. Data analysis was performed using NONMEM VI and SAS 9.1.3. Results Overall, 1260 nevirapine plasma concentrations were obtained from 271 genotyped patients. Plasma concentration−time profiles of nevirapine were best described by a one-compartment model with auto-induced first-order elimination process. Nevirapine clearance was 19.4% reduced in Asian/Black patients, compared with Caucasian/Hispanic patients. For single nucleotide polymorphism rs3745274 (CYP2B6 516G>T) heterozygous patients (GT) showed a 15.3% reduced clearance; patients with homozygous CYP2B6 516TT alleles showed a 30.6% reduced clearance compared to patients with homozygous 516GG alleles. Patients carrying the homozygote genotype of rs12768009 (CYP2C19 8403AA), highly linked to rs4244285 (CYP2C19*2), showed a 26.8% reduced clearance compared with patients with CYP2C19 8403 AG and GG alleles. Conclusion By integration of high-throughput genotyping data into a pharmacometric analysis of nevirapine, the impact of the CYP2B6 516G>T polymorphism on nevirapine’s exposure was confirmed and quantified. In addition, a new hypothesis with regard to CYP2C19 involvement in nevirapine metabolism has been generated. The analysis presented might help to optimize and individualize the therapy for patients treated with nevirapine to add to their therapeutic benefit.

Background Several methods have been presented for the analysis of complex interactions between genetic polymorphisms and/or environmental factors. Despite the available methods, there is still a need for alternative methods, because no single method will perform well in all scenarios. The aim of this work was to evaluate the performance of three selected rule based classifier algorithms, RIPPER, RIDOR and PART, for the analysis of genetic association studies. Methods Overall, 42 datasets were simulated with three different case-control models, a varying number of subjects (300, 600), SNPs (500, 1500, 3000) and noise (5%, 10%, 20%). The algorithms were applied to each of the datasets with a set of algorithm-specific settings. Results were further investigated with respect to a) the Model, b) the Rules, and c) the Attribute level. Data analysis was performed using WEKA, SAS and PERL. Results The RIPPER algorithm discovered the true case-control model at least once in >33% of the datasets. The RIDOR and PART algorithm performed poorly for model detection. The RIPPER, RIDOR and PART algorithm discovered the true case-control rules in more than 83%, 83% and 44% of the datasets, respectively. All three algorithms were able to detect the attributes utilized in the respective case-control models in most datasets. Conclusions The current analyses substantiate the utility of rule based classifiers such as RIPPER, RIDOR and PART for the detection of gene-gene/gene-environment interactions in genetic association studies. These classifiers could provide a valuable new method, complementing existing approaches, in the analysis of genetic association studies. The methods provide an advantage in being able to handle both categorical and continuous variable types. Further, because the outputs of the analyses are easy to interpret, the rule based classifier approach could quickly generate testable hypotheses for additional evaluation. Since the algorithms are computationally inexpensive, they may serve as valuable tools for preselection of attributes to be used in more complex, computationally intensive approaches. Whether used in isolation or in conjunction with other tools, rule based classifiers are an important addition to the armamentarium of tools available for analyses of complex genetic association studies.

Objective: Nevirapine is widely prescribed for HIV-1 infection. We characterized relationships between nevirapine-associated cutaneous and hepatic adverse events and genetic variants among HIV-infected adults. Design: We retrospectively identified cases and controls. Cases experienced symptomatic nevirapine-associated severe (grade III/IV) cutaneous and/or hepatic adverse events within 8 weeks of initiating nevirapine. Controls did not experience adverse events during more than 18 weeks of nevirapine therapy. Methods: Cases and controls were matched 1: 2 on baseline CD4 T-cell count, sex, and race. Individuals with 150 or less CD4 T cells/μl at baseline were excluded. We characterized 123 human leukocyte antigen (HLA) alleles and 2744 single-nucleotide polymorphisms in major histocompatibility complex (MHC) and drug metabolism and transport genes. Results: We studied 276 evaluable cases (175 cutaneous adverse events, 101 hepatic adverse events) and 587 controls. Cutaneous adverse events were associated with CYP2B6 516G→T (OR 1.66, all), HLA-Cw*04 (OR 2.51, all), and HLA-B*35 (OR 3.47, Asians; 5.65, Thais). Risk for cutaneous adverse events was particularly high among Blacks with CYP2B6 516TT and HLA-Cw*04 (OR 18.90) and Asians with HLA-B*35 and HLA-Cw*04 (OR 18.34). Hepatic adverse events were associated with HLA-DRB*01 (OR 3.02, Whites), but not CYP2B6 genotypes. Associations differed by population, at least in part reflecting allele frequencies. Conclusion: Among patients with at least 150 CD4 T cells/μl, polymorphisms in drug metabolism and immune response pathways were associated with greater likelihood of risk for nevirapine-related adverse events. Results suggest fundamentally different mechanisms of adverse events: cutaneous, most likely MHC class I-mediated, influenced by nevirapine CYP2B6 metabolism; hepatic, most likely MHC class II-mediated and unaffected by such metabolism. These risk variants are insensitive for routine clinical screening.

The first formal qualification of safety biomarkers for regulatory decision making marks a milestone in the application of biomarkers to drug development. Following submission of drug toxicity studies and analyses of biomarker performance to the Food and Drug Administration (FDA) and European Medicines Agency (EMEA) by the Predictive Safety Testing Consortium's (PSTC) Nephrotoxicity Working Group, seven renal safety biomarkers have been qualified for limited use in nonclinical and clinical drug development to help guide safety assessments. This was a pilot process, and the experience gained will both facilitate better understanding of how the qualification process will probably evolve and clarify the minimal requirements necessary to evaluate the performance of biomarkers of organ injury within specific contexts.

Long-term administration of methotrexate (MTX) for management of chronic inflammatory diseases is associated with risk of liver damage. In this study, we examined the transcriptional profiles of livers from patients treated with MTX. The possibility that expression signatures correlate with grade of fibrosis or underlying rheumatic disease was evaluated. Twenty-seven patients taking MTX were accrued for this study. Ten non-MTX-exposed normal liver specimens were used as controls. Global mRNA expression was assayed using oligonucleotide arrays. A total of 205 genes were significantly altered in MTX-exposed livers. Six of these genes were validated by qPCR. Two genes, CLN8 and ANKH that map to chromosomal locations previously associated with rheumatoid arthritis, were found to be elevated in MTX-exposed samples. Subsequent pathway analysis indicates that MTX exposure is associated with the following key alterations: (1) upregulation of lipid biosynthetic genes, consistent with MTX-induced steatosis, (2) downregulation of proinflammatory chemokines, consistent with the anti-inflammatory effects of MTX, and (3) elevation of complement pathway gene expression. Complement 5, shown earlier to be correlated with liver fibrosis in mice, was found to be elevated (twofold) in MTX-exposed livers. In conclusion, we have found the expression of a number of genes associated with rheumatic disease and/or MTX exposure to be significantly different. Differences in complement expression provide the rationale for future correlative studies between MTX-induced liver fibrosis and C5 alleles in order to identify patients with increased risk for fibrosis.

BACKGROUND: Toxicogenomics experiments often reveal thousands of transcript alterations that are related to multiple processes, making it difficult to identify key gene changes that are related to the toxicity of interest.OBJECTIVES: The objective of this study was to compare gene expression changes in a nontarget tissue to the target tissue for toxicity to help identify toxicity-related genes.METHODS: Male rats were given the hepatotoxicant methapyrilene at two dose levels, with livers and kidneys removed 24 hr after one, three, and seven doses for gene expression analysis. To identify gene changes likely to be related to toxicity, we analyzed genes on the basis of their temporal pattern of change using a program developed at the National Institute of Environmental Health Sciences, termed "EPIG" (extracting gene expression patterns and identifying co-expressed genes).RESULTS: High-dose methapyrilene elicited hepatic damage that increased in severity with the number of doses, whereas no treatment-related lesions were observed in the kidney. High-dose methapyrilene elicited thousands of gene changes in the liver at each time point, whereas many fewer gene changes were observed in the kidney. EPIG analysis identified patterns of gene expression correlated to the observed toxicity, including genes associated with endoplasmic reticulum stress and the unfolded protein response.CONCLUSIONS: By factoring in dose level, number of doses, and tissue into the analysis of gene expression elicited by methapyrilene, we were able to identify genes likely to not be implicated in toxicity, thereby allowing us to focus on a subset of genes to identify toxicity-related processes.

Molecular techniques, such as cDNA microarrays, are being used to aid in the elucidation of the mechanisms of toxicity of a variety of compounds. In this study, we evaluate the molecular effects of furan in the rat liver. Sprague–Dawley rats were exposed to 4 or 40mg/kg furan for up to 14 days. Furan induced an initial degenerative and necrotic phenotype that was followed by inflammation and fibrosis, consistent with previous observations for this compound. RNA was harvested from each lobe of the liver at several time points to observe whether lobe-specific gene expression effects occurred. Similar gene expression changes were observed in all lobes, however the magnitude of gene expression change was more pronounced in the right lobe. Finally, to help determine the correlation between gene expression changes and liver pathology, we applied traditional microarray visualization tools to the assessment of clinical chemistry and pathology parameters.

Microarray technology continues to gain increased acceptance in the drug development process, particularly at the stage of toxicology and safety assessment. In the current study, microarrays were used to investigate gene expression changes associated with hepatotoxicity, the most commonly reported clinical liability with pharmaceutical agents. Acetaminophen, methotrexate, methapyrilene, furan and phenytoin were used as benchmark compounds capable of inducing specific but different types of hepatotoxicity. The goal of the work was to define gene expression profiles capable of distinguishing the different subtypes of hepatotoxicity. Sprague-Dawley rats were orally dosed with acetaminophen (single dose, 4500 mg/kg for 6, 24 and 72 h), methotrexate (1mg/kg per day for 1, 7 and 14 days), methapyrilene (100mg/kg per day for 3 and 7 days), furan (40 mg/kg per day for 1, 3, 7 and 14 days) or phenytoin (300 mg/kg per day for 14 days). Hepatic gene expression was assessed using toxicology-specific gene arrays containing 684 target genes or expressed sequence tags (ESTs). Principal component analysis (PCA) of gene expression data was able to provide a clear distinction of each compound, suggesting that gene expression data can be used to discern different hepatotoxic agents and toxicity endpoints. Gene expression data were applied to the multiplicity-adjusted permutation test and significantly changed genes were categorized and correlated to hepatotoxic endpoints. Repression of enzymes involved in lipid oxidation (acyl-CoA dehydrogenase, medium chain, enoyl CoA hydratase, very long-chain acyl-CoA synthetase) were associated with microvesicular lipidosis. Likewise, subsets of genes associated with hepatotocellular necrosis, inflammation, hepatitis, bile duct hyperplasia and fibrosis have been identified. The current study illustrates that expression profiling can be used to: (1) distinguish different hepatotoxic endpoints; (2) predict the development of toxic endpoints; and (3) develop hypotheses regarding mechanisms of toxicity.

DNA microarrays, used to measure the gene expression of thousands of genes simultaneously, hold promise for future application in efficient screening of therapeutic drugs. This will be aided by the development and population of a database with gene expression profiles corresponding to biological responses to exposures to known compounds whose toxicological and pathological endpoints are well characterized. Such databases could then be interrogated, using profiles corresponding to biological responses to drugs after developmental or environmental exposures. A positive correlation with an archived profile could lead to some knowledge regarding the potential effects of the tested compound or exposure. We have previously shown that cDNA microarrays can be used to generate chemical-specific gene expression profiles that can be distinguished across and within compound classes, using clustering, simple correlation, or principal component analyses. In this report, we test the hypothesis that knowledge can be gained regarding the nature of blinded samples, using an initial training set comprised of gene expression profiles derived from rat liver exposed to clofibrate, Wyeth 14,643, gemfibrozil, or phenobarbital for 24 h or 2 weeks of exposure. Highly discriminant genes were derived from our database training set using approaches including linear discriminant analysis (LDA) and genetic algorithm/K-nearest neighbors (GA/KNN). Using these genes in the analysis of coded liver RNA samples derived from 24-h, 3-day, or 2-week exposures to phenytoin, diethylhexylpthalate, or hexobarbital led to successful prediction of whether these samples were derived from livers of rats exposed to enzyme inducers or to peroxisome proliferators. This validates our initial hypothesis and lends credibility to the concept that the further development of a gene expression database for chemical effects will greatly enhance the hazard identification processes.

Methapyrilene (MP) exposure of animals can result in an array of adverse pathological responses including hepatotoxicity. This study investigates gene expression and histopathological alterations in response to MP treatment in order to 1) utilize computational approaches to classify samples derived from livers of MP treated rats based on severity of toxicity incurred in the corresponding tissue, 2) to phenotypically anchor gene expression patterns, and 3) to gain insight into mechanism(s) of methapyrilene hepatotoxicity. Large-scale differential gene expression levels associated with the exposure of male Sprague—Dawley rats to the rodent hepatic carcinogen MP for 1, 3, or 7 days after daily dosage with 10 or 100 mg/kg/day were monitored. Hierarchical clustering and principal component analysis were successful in classifying samples in agreement with microscopic observations and revealed low-dose effects that were not observed histopathologically. Data from cDNA microarray analysis corroborated observed histopathological alterations such as hepatocellular necrosis, bile duct hyperplasia, microvesicular vacuolization, and portal inflammation observed in the livers of MP exposed rats and provided insight into the role of specific genes in the studied toxicological processes.

Microarrays are a new technology used to study global gene expression and to decipher biological pathways. In the current study, microarrays were used to examine gene expression patterns associated with cisplatin-mediated nephrotoxicity. Sprague-Dawley rats received either single or seven daily ip doses of cisplatin (0.5 or 1 mg/kg/day) or the inactive isomer transplatin (1 or 3 mg/kg/day). Histopathological evaluation revealed renal proximal tubular necrosis in animals that received cisplatin for 7 days, but no hepatotoxic findings. Microarray analyses were performed using rat specific arrays containing 250 toxicity-related genes. Prominent gene expression changes were observed only in the kidneys of rats that received cisplatin for 7 days. Mechanistically, the gene expression pattern elicited by cisplatin (e.g., Bax upward arrow and SMP-30 downward arrow) suggested the occurrence of apoptosis and the perturbation of intracellular calcium homeostasis. The induction of multidrug resistance genes (MDR1 upward arrow, P-gp upward arrow) and tissue remodeling proteins (clusterin upward arrow, IGFBP-1 upward arrow, and TIMP-1 upward arrow) indicated the development of cisplatin resistance and tissue regeneration. Select gene expression changes were further confirmed by TaqMan analyses. Gene expression changes were not observed in the liver following cisplatin administration. In contrast to these in vivo findings, studies using NRK-52E kidney epithelial cells and clone-9 liver cells suggested that liver cells were more sensitive to cisplatin treatment. The discrepancies between the in vivo and in vitro results suggest that caution should be taken when extrapolating data from in vivo to in vitro systems. Nonetheless, the current study elucidates the biochemical pathways involved in cisplatin toxicity and demonstrates the utility of microarrays in toxicological studies.

Two immortalized cell lines, sup (+) and sup (-), derived from mutagenized Syrian hamster embryo cells, were used to study the relationship and temporal order between calcium and ceramide signals during apoptosis. The early preneoplastic cells, termed sup (+), suppress tumorigenicity when hybridized with tumor cells, whereas later-stage sup (-) cells do not. In reduced serum conditions, sup (+) cells cease proliferating and undergo apoptosis; in contrast, sup (-) cells continue slow growth and undergo necrosis. In sup (+) cells, decreased endoplasmic reticulum (ER) calcium occurs 4 h after low serum treatment and precedes apoptosis. Significant elevations in ceramide are observed 16 h after reduced serum treatment of sup (+) cells but are not found in sup (-) cells. Inhibiting ER calcium depletion in low serum-treated sup (+) cells by treating with high levels of calcium prevents both ceramide generation and apoptosis. Conversely, inducing ER calcium depletion in sup (-) cells by treating with low serum plus thapsigargin results in elevated ceramide levels and apoptosis. Furthermore, C(6)-ceramide treatment induced apoptosis of sup (-) cells in low serum, a condition that does not normally cause apoptosis. C(6)-ceramide treatment did not induce apoptosis in either sup (+) or sup (-) cells in 10% serum but did cause G(2)/M arrest. These studies show that ceramide production is downstream of ER calcium release.