DECEMBER 2004 VOLUME 17, NUMBER 12 © Copyright 2004 by the American Chemical Society
Chemical Profiles Time Course Toxicogenomic Profiles in CD-1 Mice after Nontoxic and Nonlethal Hepatotoxic Paracetamol Administration D. P. Williams,*,†,‡ C. Garcia-Allan,†,§,| G. Hanton,⊥ J. L. LeNet,⊥ J. P. Provost,⊥ P. Brain,# R. Walsh,§ G. I. Johnston,§ D. A. Smith,| and B. K. Park‡ Department of Pharmacology and Therapeutics, University of Liverpool, Sherrington Building, Ashton Street, P.O. Box 147, Liverpool, Merseyside L69 3GE, United Kingdom, Discovery Biology, Pharmacokinetics, Dynamics and Metabolism, and Development Operations, Non Clinical Statistics, Pfizer Global Research and Development, IPC 432, Ramsgate Road, Sandwich, Kent CT13 9NJ, United Kingdom, and Pfizer Amboise, Z.I. Poce´ -sur-Cisse B.P. 159, F-37401 Amboise Cedex, France Received June 9, 2004
Adverse drug reactions are a major clinical problem. Drug-induced hepatotoxicity constitutes a large percentage of these reactions. A thorough understanding of the genetic events, specifically, the early “decision-making” processes underlying biological changes caused by drugs and metabolites, is required. To assist in the understanding of these events, we have employed the model hepatotoxin, paracetamol (APAP), and GeneChip technology to investigate global genetic events seen after nontoxic and toxic doses in the mouse. Mice were dosed [vehicle, nontoxic APAP (1 mmol/ kg), and toxic APAP (3.5 mmol/kg)], and individual hepatic RNA samples were hybridized to separate chips to determine interanimal variation. Statistical analysis detected 175 CD-1 mouse genes that were significantly regulated (P < 4.1 × 10-6), and nonsignificant genes were discarded. For clarity, the significantly regulated genes were then binned into categories according to their major functions antioxidant, glutathione, metabolism, transcription, immune, and apoptosis. There was no hepatic stress observed after dosing 1 mmol/kg APAP, when measured by serum alanine aminotransferase levels. Hepatic toxicity was observed at both 4 and 24 h after a 3.5 mmol/kg dose of APAP. Time course expression profiles for selected genes have been created. These results demonstrate that most active gene expression occurs around 4 h after a toxic dose of APAP. Down-regulation of these genes is observed over 24 h, coinciding with the development of overt toxicity. These data provide a deeper understanding of the in vivo time course of physiological responses of the liver to chemical stress and provide a logical step forward for the investigation of new chemical entities demonstrated positive in chemically reactive metabolite screens. The complete data set can be viewed at http:// www.ebi.ac.uk/arrayexpress/. The accession number is E-MEXP-82.
Introduction The analgesic, paracetamol (APAP),1 causes hepatic failure in overdose through cytochrome P450 (CYP)-
mediated metabolism to N-acetyl-p-benzoquinoneimine (NAPQI) species (1-5). The precise mechanisms of initiation of a biological response are unknown. Additionally, §
* To whom correspondence should be addressed. Tel: +44(0)151 794 5791. Fax: +44(0)151 794 5540. E-mail:
[email protected]. † These authors contributed equally to the production of this manuscript. ‡ University of Liverpool.
Discovery Biology, Pfizer Global Research and Development. Pharmacokinetics, Dynamics and Metabolism, Pfizer Global Research and Development. ⊥ Pfizer Amboise. # Development Operations, Non Clinical Statistics, Pfizer Global Research and Development. |
10.1021/tx049846x CCC: $27.50 © 2004 American Chemical Society Published on Web 10/28/2004
1552
Chem. Res. Toxicol., Vol. 17, No. 12, 2004
the mechanisms enabling cellular decisions, such as defense or development of toxicity, have yet to be determined (6). The mouse is a more sensitive model of APAP-induced toxicity than the rat (7, 8). This is partly due to differences in the activity of CYP2E1; a higher activity in the murine hepatocyte leads to a greater turnover of APAP to NAPQI, the toxic metabolite (9). However, this cannot be the only factor; the mouse, similar to humans, demonstrates severe and fatal liver damage with doses of 250-375 mg/kg (10-12), while rats are able to survive doses of up to 4250 mg/kg (13). The genetic and biological defense systems to toxic metabolites, both in vivo and in vitro in rats and mice, have been a subject of interest for the past 30 years (6, 12, 14-24). More recently, however, the techniques used to define the mechanism of toxicity of APAP are more concerned with global responses, i.e., whole cell and/or animal responses (6, 15, 25). Investigations describing the initiating genomic, proteomic, and metabonomic events in NAPQI toxicity (6, 25) conclude that NAPQI causes disruptions in amino acid metabolism, β-oxidation, and the citric acid cycle leading to toxicity (25). NAPQI is ultimately responsible for the initial toxicity; however, genetic and protein regulation are responsible for the development and/or defense from toxicity (6) within this particular mouse strain. The aim of this study was to compare the time course of global hepatic gene regulation after a nontoxic and toxic dose of APAP in an animal model that is sensitive to APAP toxicity, the male CD-1 mouse. Questions being addressed are what genes are regulated after a nontoxic and toxic dose of APAP and can we disseminate the significantly expressed genes into those responsible for defense and recovery and those responsible for the development of toxicity and cell death? Genetic events were measured using Affymetrix Gene Chips (MGU74Av2) at different time points after dosing male CD-1 mice with APAP and compared to control-dosed animals. An important point concerning this study is that for each mouse a single chip was used; that is, there was no pooling of either livers or RNA samples. Microarray data have been analyzed using a novel nonparametric method that transforms the data to stabilize the variance found in microarray results (Brain et al., in preparation). For many of the genes, it is the first time that they have been associated with drug-induced hepatocellular stress.
Materials and Methods Animal Dosing Regime. The protocols described were undertaken in accordance with criteria outlined in a license granted under the Animals (Scientific Procedures) Act 1986 and approved by the University of Liverpool Animal Ethics Committee. Male CD-1 mice (n ) 3; 25-30 g; Charles River) were administered a single intraperitoneal dose of APAP (1 and 3.5 mmol/kg) dissolved in vehicle (saline 0.9%; Table 1). Animals treated with vehicle alone were used as controls. At either 1, 4, or 24 h after dosing, animals were killed using a rising concentration of carbon dioxide until cessation of inhalation was observed, after which, blood was collected by cardiac puncture. The liver was removed and rinsed in ice-cold saline (0.9%), and 1 Abbreviations: APAP, N-acetyl-p-aminophenol (paracetamol, acetaminophen); CYP, cytochrome P450; NAPQI, N-acetyl-p-benzoquinone imine; ALAT, alanine aminotransferase; HO-1, haem oxygenase1; HNF3, hepatocyte nuclear factor 3; γ-GCS, γ-glutamylcysteine synthase; HSP, heat shock protein; GSH, glutathione; D°, dose alone significant; D° × Ti, dose and time significant; Ti time alone significant; LSD, least significant difference.
Williams et al. Table 1. Significantly Regulated Murine Hepatic Genes after Dosing APAP (151 and 529 mg/kg)a Affy ID
GenbankID
99649 at 93268 at 93026 at
U85414 AI852001 AW12433
96680 at 93573 at 93875 at 96564 at 160101 101561 104598 at 96254 at 101955 at 98111 at 95057 at 99109 at 93975 at a
gene name GSH related genes glutamate-cysteine ligase, catalyticsubunit lactoylglutathione lyase GSH S-transferase µ
antioxidant-related genes AI835630 DnaJ (HSP40) homologue, subfamily B, member 9 V00835 metallothionein 1 M12571 mouse heat shock inducible (HSP70A1) X54401 heat shock 70 kDa protein 8 X56824 heme oxygenase (decycling) 1 K02236 mouse metallothionein II (MT-II) gene X61940 protein tyrosine phosphatase, nonreceptor type 16 AB028272 DnaJ (HSP40) homologue, subfamily B, member 1 AJ002387 heat shock 70 kDa protein 5 (glucose-regulated protein, 78 kDa) L40406 HSP, 105 kDa AI846938 homocysteine inducible, endoplasmicreticulum stress inducible, ubiquitin-like domain member 1 M59821 immediate earlyresponse 2 AI853531 Rho GTPase activator
All genes in the table show P < 4.1 × 10-6.
Figure 1. Average serum ALAT for mice given an ip dose of either vehicle control or 1 or 3.5 mmol/kg APAP. The vertical bar is the 95% LSD; two means that are vertically further apart than the length of this bar are significantly different at, at least, the 5% level. a portion of it was immediately frozen in liquid nitrogen and stored at -80 °C until use. Clinical Chemistry Analyses. Blood samples were stored at 4 °C and allowed to clot overnight. Serum was isolated by centrifugation at 4000g for 30 s, and serum alanine aminotransferase (ALAT) levels were determined using the Sigma GPtransaminase kit, according to the manufacturer’s instructions. Gene Expression Profiling. Information concerning measurement of gene expression profiling data management and expression analysis using Affymetrix Gene Chips (Murine genome U74Av2) in this manuscript conforms to the MIAME (Minimum Information About a Microarray Experiment) checklist (26, 27). The complete data set has been submitted to and accepted by ArrayExpress, the microarray database of the European Bioinformatics Institute (http://www.ebi.ac.uk/ arrayexpress/). The accession number is E-MEXP-82. Liver RNA was extracted by previously described methods (28). The concentration, purity, and integrity of the RNA were determined by spectrophotometry and agarose gel electrophoresis. The starting material was DNase-treated total RNA (10 µg). First and second strand cDNA was synthesized, cleaned, and washed as indicated on the gene expression technical manual (Affymetrix). Aliquots (2 µL) of cDNA were assessed for size distribution by gel electrophoresis analysis. The cDNA then underwent in vitro transcription and target hybridization to the probe arrays, which were washed, stained, and scanned according to the manufacturers instructions (Affymetrix). Data Management and Expression Analysis. The analysis of the hybridization intensity data from the probe arrays
Toxicogenomic Profiles in CD-1 Mice
Chem. Res. Toxicol., Vol. 17, No. 12, 2004 1553
Figure 2. Dose vs time graphs showing the antioxidant-related genes regulated in mice after dosing either vehicle (-[- refer to the Materials and Methods), 1 mmol/kg APAP (-9-), and 3.5 mmol/kg APAP (-2-) for 1, 4, and 24 h. The vertical bar is the 95% LSD. was done using MicroArray Suite (MAS) 5.0 software (Affymetrix). QC Analysis. Before the data from a particular chip was analyzed, a number of QC criteria had to be achieved. (i) Chip images were visually checked for any faults or debris. Debris was masked using the Affymetrix MAS5 software, but if the masking covered >200 probe cells, the hybridization would be repeated using a fresh chip. (ii) Grid alignment was checked visually, to ensure that probe cells were effectively separated. (iii) Raw Q values had to be similar to each other and range between 1.5 and 3. (iv) Control GAPDH and/or actin calls had to be present. (v) A 3′/5′ ratio of the GAPDH and actin controls had to be
