* Drug Metabolism and Toxicology, Division of Pharmaceutical Sciences, Kanazawa University, Kanazawa, Japan, and Life science group, Hitachi Ltd., Saitama, Japan
1 To whom correspondence should be addressed at Drug Metabolism and Toxicology, Division of Pharmaceutical Sciences, Kanazawa University, Kakuma-machi, Kanazawa 920-1192, Japan. Fax: +81-762344407. E-mail: tyokoi{at}kenroku.kanazawa-u.ac.jp.
Received May 27, 2005; accepted June 2, 2005
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: gene expression profiles; hepatotoxicity; DNA microarray.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The liver is one of the first organs to be exposed to peroral-administered chemicals via the portal vein. Chemical concentrations in the liver are often much higher than the peak plasma concentration. The liver is also the major site for xenobiotic metabolism, and various chemicals can lead to the formation of active metabolites with toxic effects. The high concentration exposure and metabolic activity make the liver one of the primary targets for various types of chemical-induced toxicity.
The five typical hepatotoxicants chosen in this study were acetaminophen (APAP, p-acetamidophenol), bromobenzene (BB), carbon tetrachloride (CT), dimethylnitrosamine (DMN), and thioacetamide (TA). APAP is known as a mild analgesic drug, but it is a potent hepatotoxicant at high doses and in persons with enhanced susceptibility. APAP is largely (apparently more than 80%) converted to conjugates of glucuronate and sulfate. A minor amount, less than 5%, is metabolized to an active metabolite, mainly N-acetyl-p-benzoquinone imine (NAPQI) by cytochrome P450 (CYP) 2E1, which binds promptly to glutathione (GSH). Other metabolites (5 to 15%) appear to have no toxicity (Zimmerman, 1999). When an active metabolite exceeds the GSH content, excess metabolite binds to tissue molecules and manifests toxicity such as necrosis. BB, a traditional hepatotoxicant, is subjected to cytochrome P450-mediated epoxidation, and a major metabolite is 3,4-epoxide. Detoxication of the active metabolite is by GSH conjugation. At high BB doses, due to the conjugation to the epoxides, liver GSH shortage and secondary reactions such as lipid peroxidation, intracellular calcium alteration, and mitochondrial dysfunction finally lead to cell death (Heijne et al., 2003
, Zimmerman, 1999
). CT is a potent hepatotoxicant, and a single dose leads promptly to severe necrosis and steatosis. CT liver necrosis is caused by trichloromethyl free radicals from a CYP2E1-mediated pathway. The covalent binding of trichloromethyl to cell protein is considered the initial step of sequential events leading to membrane lipid peroxidation and, finally, to cell necrosis (Jeong, 1999
; Zimmerman, 1999
). DMN is the most potent toxicant of all dialkylnitrosamines and leads to hemorrhagic necrosis and steatosis. The DMN toxicity process is as follows: first, demethylation of CYP2E1 to monomethylnitrosamine; second, spontaneous change of diasomethane; finally, methylation of cell components (Zimmerman, 1999
). Thioacetamide (TA) is a potent hepatotoxicant that requires metabolic activation by mixed-function oxidases. Generally, CYP2B, CYP2E1, and FMOs metabolize TA to its toxic metabolites (Hunter et al., 1977
; Wang et al., 2000
), and these intermediate metabolites might bind to cellular proteins by the formation of acetylimidolysine derivatives (Dyroff and Neal, 1981
). TA is apparently converted to thioacetamide-S-oxide and is presumably converted to an active toxic metabolite that binds covalently to tissue molecules, provoking necrosis (Zimmerman, 1999
).
The aims of this study were to find available toxicity marker genes, to investigate the correlation between biochemical markers and gene expression profiles, and to create a new evaluation method using DNA microarray. In this study, we attempted to apply our microarray of drug-response gene expressions for the evaluation of chemical-induced hepatotoxicity in rats.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Administration of chemicals and assessment of liver injury.
Eighty-eight rats were assigned to 22 groups (four rats/group). The dosing solutions were prepared as follows with each vehicle. APAP: 500 mg/kg in corn oil; BB: 2.5 mmol/kg in corn oil; CT: 1 ml/kg in corn oil; DMN: 20 mg/kg in saline; TA: 400 mg/kg in saline; control: vehicle for saline or corn oil group. The chemicals were intraperitoneally injected in a single bolus at a volume of 2 ml/kg. At the indicated time (6, 12, 24, 48 h after administration), the rats were sacrificed, and the liver and serum samples were collected. Four typical biochemical markers for hepatotoxicity (aspartate aminotransferase, AST; alanine aminotransferase, ALT; lactate dehydrogenase, LDH; alkaline phosphatase, ALP) were measured by SRL, Inc. (Tokyo, Japan).
RNA isolation.
Total hepatic RNA was isolated using ISOGEN. Approximately 100 mg of whole liver were lysed with 1.0 ml of the lysis solution. Chloroform (200 µl) was added and vortexed vigorously for 15 s. The mixture was centrifuged at 15,000 x g for 15 min at 4°C. The aqueous phase was transferred carefully to a new tube, and the RNA was precipitated with 0.5 ml of isopropyl alcohol for 10 min at room temperature. The mixture was centrifuged at 15,000 x g for 10 min. After washing with 75% ethanol, the pellet was dissolved in diethylpyrocarbonate-treated water. Equal amounts of total mRNA from each hepatotoxicant-administered sample were pooled and used for the microarray analysis and real-time reverse transcriptase (RT)-PCR.
In vitro amplification and DNA microarray.
cDNA targets were prepared from pooled total RNA by in vitro transcription reaction as described previously (Luo et al., 1999). Amplified RNA (6 µg) was reverse transcribed by random hexamer and aminoallyl-dUTP. The synthesized cDNA was labeled with NHS-ester Cy3 or Cy5 (Hughes et al., 2001
). The labeled cDNA was applied to the cDNA microarray (Rat Drug Response Chip containing 1,097 genes, Hitachi, Tokyo, Japan). In order to confirm the microarray data in the APAP group, a Rat cDNA Microarray kit G4105A containing 14,815 genes (Agilent Technologies, Palo Alto, CA) was used. Hybridization was performed at 62°C for 14 h. After washing, the microarray was scanned on a ScanArray 5000 (Packard BioChip Technologies, Billerica, MA), and the image was analyzed using QuantArray software (Packard BioChip Technologies). The signal intensity of each spot was calibrated by subtraction of the intensity of the control.
Real-time RT-PCR.
Rat Cathepsin L (CtsL), Diazepam binding inhibitor (Dbi), Heme oxygenase 1 (Hmox1), Sulfotransferase 1a2 (Sult1a2), T-cell death associated gene (Tdag), and GAPDH were quantified by real-time RT-PCR. Primer sequences using in this study were as follows: CtsL, 5'-TCT ACT ATG AAC CCA ACT G-3' and 5'-GAT TCA AGT ACC ATG GTC T-3'; Dbi, 5'-CCA ACT GAT GAA GAG ATG CTG T-3' and 5'-CCC TAA CAT ATC AGA GCC ATG T-3'; Hmox1, 5'-ATA GAG CGA AAC AAG CAG A-3' and 5'-TAG AGC TGT TTG AAC TTG G-3'; Sult1a2, 5'-TCA TTG AGT GGA CTT TGC CTT-3' and 5'-CAC TTT TCC AGC TTT GAA CTG-3'; Tdag, 5'-CCA AGC AGG TAC AAC ATC AG-3' and 5'-TTC TGC CTC GTA GAC TTG AC-3'. For RT process, total RNA (4 µg) and 150 ng random hexamer were mixed and incubated at 70°C for 10 min. RNA solution was added to a reaction mixture containing 100 units of ReverTra Ace, reaction buffer, and 0.5 mM dNTPs in a final volume of 40 µl. The reaction mixture was incubated at 30°C for 10 min, 42°C for 1 h, and heated at 98°C for 10 min to inactivate the enzyme. Real-time PCR was performed using the Smart Cycler® (Cepheid, Sunnyvale, CA) with Smart Cycler® software (Ver. 1.2b). PCR mixture contained 1 µl of template cDNA, SYBR® Premix Ex TaqTM solution and 10 pmol of sense and antisense primers. The PCR condition for GAPDH and Sult1a2 was as follows: after an initial denaturation at 95°C for 30 s, the amplification was performed by denaturation at 94°C for 4 s, annealing and extension at 64°C for 20 s for 45 cycles. The PCR condition for other genes was as follows: after an initial denaturation at 95°C for 20 s, the amplification was performed by denaturation at 95°C for 5 s, annealing at 55°C for 10 s, and extension at 72°C for 15 s for 45 cycles. Amplified products were monitored directly by measuring the increase of the dye intensity of the SYBR Green I (Molecular Probes, Eugene, OR) that binds to double-strand DNA amplified by PCR.
Data management.
Fold-change determination, experiment normalization, and clustering analysis were performed with GeneSpring software (Agilent Technologies). Gene expression values for each chip were normalized to the intensity-dependent (LOWESS) normalization built in GeneSpring. In the Quality-Threshold (QT) clustering analyses, a standard correlation method was used in this study. Fold-change filters included the requirement that the genes be present in at least 200% of the administered samples for up-regulated genes, and 50% of controls for down-regulated genes.
Statistics.
The t-test was used to detect significant differences between the means of two groups relative to the observed variance within groups.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
Genes Appeared in All Five of the Chemical-Administered Groups
For further analysis, the gene expression profiles identified in the gene clusters of all chemicals in Figure 4A were displayed. Three genes were contained in all the up-regulated type clusters and 17 genes in the down-regulated type clusters of all the chemical-administered groups as shown in Figure 4. The expression profiles of 17 down-regulated genes are shown in Figure 4A. The blue lines indicate the expression profiles of the genes that appeared not only in Figure 2 but also in Figure 4. The expression profiles of the genes that appeared in all chemical groups but only in Figure 4 are indicated by black lines. The three common genes in the up-regulated type cluster were not listed in Figure 2 (data not shown). In the down-regulated type cluster, eight genes of all the regulated groups were also listed in Figure 2. The 3 up-regulated and 17 down-regulated genes were analyzed in the following QT clustering.
QT Clustering for the Genes That Appeared in at Least Four of Five Chemical-Administered Groups
For further analysis, the genes that appeared in at least four of the five chemical-administered groups were analyzed. Thirty-seven genes were identified in the up-regulated type cluster and 60 genes in the down-regulated type cluster, as shown in Figure 4. In the up-regulated type cluster, 7 of 37 genes were overlapped with the genes of the up-regulated groups shown in Figure 2. In the down-regulated type cluster, all genes of the down-regulated groups in Figure 2 were overlapped. Further QT clustering was performed using the 37 up-regulated or 60 down-regulated genes identified in Figure 4A. The analysis setting for the minimal cluster size was 10 genes and the minimal correlation coefficient was 0.65. Twenty-two of 37 genes were identified as the up-regulated type cluster, and 44 of 60 genes as the down-regulated type cluster. The profiles of the 22 and 44 identified genes and the average profile are shown in Figures 5B and 5C, respectively. The expression profiles of the genes that also appeared in Figure 2 are indicated by blue lines. The expression profiles of the clustered gene that did not appear in Figure 2 are indicated by black lines. Seven of 10 genes in the up-regulated group in Figure 2 and all 10 genes in the down-regulated group in Figure 2 were overlapped as a result of the clustering (Figs. 5B and 5C). As shown in Figures 5A, 5B, and 5C, the average up-regulated or down-regulated peaks were correlated with the maximal toxic times.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In this study, we evaluated five typical hepatotoxic chemicals these were thought to cause zone-3 necrosis (Zimmerman, 1999). The dose levels of APAP (Price and Jollow, 1982
; Sato and Izumi, 1989
), BB (Chakrabarti and Brodeur, 1984
), CT (Theocharis et al., 2001
), DMN (Asakura et al., 1998
), and TA (Wang et al., 2000
; Zaragoza et al., 2000
) were selected based on their association with detectable hepatotoxicity as previously reported. These data confirmed that the hepatotoxicity models of all the chemical-administered groups were successfully conducted, and the toxic time points of APAP, BB, CT, DMN, and TA were estimated as 12, 24, 6, 48, and 24 h, respectively. The maximal toxic times of BB (Chakrabarti and Brodeur, 1984
), DMN (Asakura et al., 1998
), and TA (Wang et al., 2000
; Zaragoza et al., 2000
, respectively) in rats were the same as previously reported. In the CT-administered rats, AST and ALT elevated significantly at 6 and 48 h in the present study, but the CT toxicity assessed by serum AST and ALT increased at 6 to 24 h (AST) or 6 to 36 h (ALT) in a time-dependent manner, respectively (Zimmerman, 1999
). In APAP-administered rats, AST and ALT elevated significantly at 6 and 12 h, but the serum AST and ALT were previously reported to be evaluated at 24 h by APAP administration (Hong et al., 1992
; Wang et al., 1999
). In the present study, the biochemical markers reflected the major gene expression profiles (Figs. 2, 3, 4, and 5).
We performed hierarchical clustering using gene groups whose expression levels were distinctively changed at the toxic time points (Fig. 1). At the maximal toxic time, CT and TA were sorted in a relatively close cluster. At the maximal toxic time, BB and DMN were sorted in a similar cluster. However, all APAP-administered groups were sorted in a different cluster. We performed many other types of hierarchical clustering by using other gene groups such as enzymes, signal transductions, and so on, and the results were almost the same as shown in Figure 1 (data not shown). Although studies concerning many hepatotoxicants including APAP, BB, CT, DMN, and TA administered to Sprague-Dawley rats have been reported (Kier et al., 2004, McMillian et al., 2004a
,b
), there has been no attempt of such a hierarchical clustering analysis using five chemicals. Thus, that the gene expression profiles of APAP administration were different from other those of the four chemicals constitutes new information.
In handling microarray data, it is necessary to consider what kinds of effects are relevant to the purpose of the experiments. We identified 20 representative genes whose up- or down-regulation peaks overlapped with the maximal toxic time (Fig. 2). In each of the chemical-administered rats, almost all genes identified in the present study showed similar expression profiles. The expression profile was confirmed in five representative genes, resulting in the overlapped profile with that of DNA microarray. Moreover, 17 of 20 genes were also identified by QT clustering analysis (Fig. 5). Data from QT clustering are independent of the hepatotoxicity estimated by serum biochemical markers. The present results showed the potential usefulness of 17 identified genes as toxicity markers. Most of the identified genes were not described previously as having a relationship with hepatotoxic chemicals. For example, TA administration up-regulated rat aldorase A mRNA (Bulera et al., 2001). CtsL was up-regulated at the hepatic mRNA level after 24 h in BB- and TA-administered rats (Heijne et al., 2003
; Bulera et al., 2001
, respectively). Hmox1 was reported to be up-regulated by four chemicals, APAP (Chiu et al., 2002
), BB (Heijne et al., 2003
), CT (Montosi et al., 1998
), and TA (Bulera et al., 2001
; Matsuura et al., 1988
), in rats in vivo. In the present study, the expression of CYP2E1, which possibly catalyzes the induced toxicity of the five chemicals (Jeong, 1999
; Lauriault et al., 1992
; Wang et al., 2000
; Zimmerman, 1999
) was only slightly changed (data not shown), suggesting that the induction of CYP2E1 would not be involved.
QT clustering analysis is usually performed to determine the specific gene expression patterns. The resulting clusters gave us a good indication of the types of gene expression patterns that existed in the data (Heyer et al., 1999). The gene expression patterns obtained from QT clustering analysis were major patterns expressed with each type of chemical administration (Fig. 4A). However, the extent of toxicity estimated by serum biochemical markers was different for each chemical. The average of each gene expression profile from QT clustering was overlapped with the changes of the serum biochemical markers. Furthermore, we performed QT clustering using Agilent Rat cDNA microarray kit G4105A in APAP-administered rat samples, the results of which showed almost the same expression profile (Fig. 4B). From this cDNA microarray, we confirmed the reproducibility of the new QT clustering data.
In conclusion, we identified 17 potential toxicity markers. It was not clarified whether all of the genes were related to the development of toxicity or whether these genes were related to each other. In the present study, we found that the expression profile analysis of the chemical administration could be used to estimate the maximal toxic time independently of the toxicity grade. This expression profile analysis could also be a tool for identifying potential hepatotoxicants. This would be a new approach for determining hepatotoxicity by microarray analysis. We demonstrated that these two approaches, serum biochemical markers and two different QT clustering analyses, yielded the same results. For further studies, detailed time courses, multidose levels, and evaluation of other hepatotoxicants will be performed.
![]() |
SUPPLEMENTARY DATA |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bulera, S. J., Eddy, S. M., Ferguson, E., Jatkoe, T. A., Reindel, J. F., Bleavins, M. R., and De La Iglesia, F. A. (2001). RNA expression in the early characterization of hepatotoxicants in Wistar rats by high-density DNA microarrays. Hepatology 33, 12391258.[CrossRef][ISI][Medline]
Chakrabarti, S., and Brodeur, J. (1984). Dose-dependent metabolic excretion of bromobenzene and its possible relationship to hepatotoxicity in rats. J. Toxicol. Environ. Health 14, 379391.[ISI][Medline]
Chiu, H., Brittingham, J. A., and Laskin, D. L. (2002). Differential induction of heme oxygenase-1 in macrophages and hepatocytes during acetaminophen-induced hepatotoxicity in the rat: Effects of hemin and biliverdin. Toxicol. Appl. Pharmacol. 181, 106115.[CrossRef][ISI][Medline]
Dyroff, M. C., and Neal, R. A. (1981). Identification of the major protein adduct formed in rat liver after thioacetamide administration. Cancer Res. 41, 34303435.[Abstract]
Heijne, W. H., Stierum, R. H., Slijper, M., van Bladeren, P. J., and van Ommen, B. (2003). Toxicogenomics of bromobenzene hepatotoxicity: A combined transcriptomics and proteomics approach. Biochem. Pharmacol. 65, 857875.[CrossRef][ISI][Medline]
Heyer, L. J., Kruglyak, S., and Yooseph, S. (1999). Exploring expression data: Identification and analysis of coexpressed genes. Genome Res. 11, 11061115.[CrossRef]
Hong, R. W., Rounds, J. D., Helton, W. S., Robinson, M. K., and Wilmore, D. W. (1992). Glutamine preserves liver glutathione after lethal hepatic injury. Ann. Surg. 215, 114119.[ISI][Medline]
Hughes, T. R., Mao, M., Jones, A. R., Burchard, J., Marton, M. J., Shannon, K. W., Lefkowitz, S. M., Ziman, M., Schelter, J. M., Meyer, M. R., et al. (2001). Expression profiling using microarrays fabricated by an ink-jet oligonucleotide synthesizer. Nat. Biotech. 19, 342347.[CrossRef][ISI][Medline]
Hunter, A. L., Holscher, M. A., and Neal, R. A. (1977). Thioacetamide-induced hepatic necrosis. I. Involvement of the mixed-function oxidase enzyme system. J. Pharmacol. Exp. Ther. 200, 439448.[Abstract]
Jeong, H. G. (1999). Inhibition of cytochrome P450 2E1 expression by oleanolic acid: Hepatoprotective effects against carbon tetrachloride-induced hepatic injury. Toxicol. Lett. 105, 215222.[CrossRef][ISI][Medline]
Kier, L. D., Neft, R., Tang, L., Suizu, R., Cook, T., Onsurez, K., Tiegler, K., Sakai, Y., Ortiz, M., Nolan, T., et al. (2004). Applications of microarrays with toxicologically relevant genes (tox genes) for the evaluation of chemical toxicants in Sprague Dawley rats in vivo and human hepatocytes in vitro. Mutat. Res. 549, 101113.[ISI][Medline]
Lauriault, V. V., Khan, S., and O'Brien, P. J. (1992). Hepatocyte cytotoxicity induced by various hepatotoxins mediated by cytochrome P-450IIE1: Protection with diethyldithiocarbamate administration. Chem. Biol. Interact. 81, 271289.[CrossRef][ISI][Medline]
Luo, L., Salunga, R. C., Guo, H., Bittner, A., Joy, K. C., Galindo, J. E., Xiao, H., Rogers, K. E., Wan, J. S., Jackson, M. R., et al. (1999). Gene expression profiles of laser-captured adjacent neuronal subtypes. Nat. Med. 5, 117122.[CrossRef][ISI][Medline]
Matsuura, Y., Fukuda, T., Yoshida, T., and Kuroiwa, Y. (1988). Inhibitory effect of zinc-protoporphyrin on the induction of heme oxygenase and the associated decrease in cytochrome P-450 content in rats. Toxicology 50, 169180.[CrossRef][ISI][Medline]
McMillian, M., Nie, A. Y., Parker, J. B., Leone, A., Bryant, S., Herlich, J., Yieh, L., Bittner, A., Liu, X., Wan, J., et al. (2004a). Inverse gene expression patterns for macrophage activating hepatotoxicants and peroxisome proliferators in rat liver. Biochem. Pharmacol. 67, 21412165.[CrossRef][ISI][Medline]
McMillian, M., Nie, A. Y., Parker, J. B., Leone, A., Bryant, S., Kemmerer, M., Herlich, J., Liu, Y., Yieh, L., Bittner, A., et al. (2004b). A gene expression signature for oxidant stress/reactive metabolites in rat liver. Biochem Pharmacol. 68, 22492261.[CrossRef][ISI][Medline]
Montosi, G., Garuti, C., Iannone, A., and Pietrangelo, A. (1998). Spatial and temporal dynamics of hepatic stellate cell activation during oxidant-stress-induced fibrogenesis. Am. J. Pathol. 152, 13191326.[Abstract]
Price, V. F., and Jollow, D. J. (1982). Increased resistance of diabetic rats to acetaminophen-induced hepatotoxicity. J. Pharmacol. Exp. Ther. 220, 504513.[Abstract]
Sato, C., and Izumi, N. (1989). Mechanism of increased hepatotoxicity of acetaminophen by the simultaneous administration of caffeine in the rat. J. Pharmacol. Exp. Ther. 248, 12431247.[Abstract]
Theocharis, S. E., Margeli, A. P., Skaltsas, S. D., Spiliopoulou, C. A., and Koutselinis, A. S. (2001). Induction of metallothionein in the liver of carbon tetrachloride intoxicated rats: An immunohistochemical study. Toxicology 161, 129138.[CrossRef][ISI][Medline]
Wang, T., Shankar, K., Ronis, M. J., and Mehendale, H. M. (2000). Potentiation of thioacetamide liver injury in diabetic rats is due to induced CYP2E1. J. Pharmacol. Exp. Ther. 294, 473479.
Wang, P. Y., Kaneko, T., Wang, Y., and Sato, A. (1999). Acarbose alone or in combination with ethanol potentiates the hepatotoxicity of carbon tetrachloride and acetaminophen in rats. Hepatology 29, 161165.[CrossRef][ISI][Medline]
Zaragoza, A., Andres, D., Sarrion, D., and Cascales, M. (2000). Potentiation of thioacetamide hepatotoxicity by phenobarbital pretreatment in rats. Inducibility of FAD monooxygenase system and age effect. Chem. Biol. Interact. 124, 87101.[CrossRef][ISI][Medline]
Zimmerman, H. J. (1999). Hepatotoxicity: The Adverse Effects of Drugs and Other Chemicals on the Liver, 2nd ed., pp. 229274. Lippncott Williams & Wilkins, Philadelphia.
|