Effect of Two 5-HT6 Receptor Antagonists on the Rat Liver: a Molecular Approach

Effect of two 5-HT6 receptor antagonists on the rat liver: a molecular approach

L Suter1 ABSTRACT 1 Serotonin is involved in disorders of the central nervous system; thus, specific M Haiker 5-HT receptor antagonists have therapeutic potential. Nevertheless, 1 6 MC de Vera preclinical tests showed that Ro 65-7199 caused hepatic steatosis. Here, S Albertini1 we investigated the hepatic effects of Ro 65-7199 and Ro 66-0074 using toxicogenomics. The profiles obtained after exposure of rats to both 1F Hoffmann-La Roche Ltd, Toxicology, compounds clearly show that two pharmacologically closely related Grenzacherstrasse, Basle, Switzerland compounds with different toxicological profiles can be distinguished based Correspondence: on gene expression profiles. Moreover, side effects can be detected earlier Dr L Suter, F Hoffmann-La Roche Ltd, with toxicogenomics than with conventional end points. A possible link Toxicology (90/505a), Grenzacherstrasse, between the sterol metabolic pathway, the induction of CYP2B, and the CH 4070 Basle, Switzerland. hepatic fat accumulation was also established. Summarizing, gene expres- Tel: þ 41 61 6878586 Fax: þ 41 61 688 1929 sion profiles allow both compounds to be distinguished according to their E-mail: [email protected] toxicity and provide mechanistic insights. The results clearly show the power of toxicogenomics as a tool for obtaining characteristic fingerprints at early time-points and for generating mechanistic hypotheses. The Pharmacogenomics Journal (2003) 3, 320–334. doi:10.1038/sj.tpj.6500207 Published online 11 November 2003

Keywords: toxicogenomics; 5-HT6 receptor antagonist; liver; steatosis; drug metabolizing enzymes; CYP P450

INTRODUCTION Serotonin (5-hydroxytryptamine, 5-HT) has been shown to be involved in the pathogenesis of a number of disorders in the central nervous system.1 This neurotransmitter binds to the serotonin receptor family, which is a member of the larger family of G-protein-coupled receptors.2 Several serotonin receptors (5- 3 HT receptors) have been cloned and characterized, including the 5-HT6 receptor. Northern blot analysis has shown that the 5-HT6 receptor is expressed mainly in brain, with very weak to no expression in the periphery.3,4 This receptor appears to be involved in certain anxiety disorders, and functional studies have demonstrated that blocking the receptor enhances acetylcholine neurotransmis- sion in the rat brain.5,6 Furthermore, experiments performed with antisense

oligonucleotides and with specific 5-HT6 receptor antagonists showed that blocking this receptor type produced an increase in the retention of learning in rats.7 Thus, its functional characteristics in the behavioral studies have shown

that the 5-HT6 receptor is a potential target for the treatment of memory deficit as in patients with Alzheimer’s disease. Moreover, due to the specific localization

of the receptor in the brain, fewer side effects are expected with specific 5-HT6 receptor antagonists than with other commonly used therapies such as Received: 03 April 2003 acetylcholinesterase inhibitors or muscarinic agonists. One of the compounds Revised: 29 August 2003 Accepted: 12 September 2003 selected as a possible clinical candidate was 4-amino-N-(6-bromo-1H-indol-4-yl)- Published online 11 November 2003 benzenesulfonamide (Ro 65-7199). Unpublished toxicity data have shown that Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 321

the compound caused hepatic lipid accumulation in rats RESULTS treated with high doses. The lack of expression of 5-HT6 Clinical Chemistry and Histopathology receptor in the liver provided a strong argument against As seen in previous studies, there was a dose-related increase toxicity due to exaggerated pharmacological activity. In in fatty change in the liver. Fatty change was assessed addition, another 5-HT6 antagonist, 4-(2-bromo-6-pyrroli- microscopically for each animal using a semiquantitative din-1-yl-pyridine-4-sulfonyl)-phenylamine (Ro 66-0074), grading ranging from 0 (no finding) to 5 (severe finding). did not cause hepatic steatosis. Ro 66-0074 is closely related Group severity values were obtained by averaging the to Ro 65-7199 since it displays similar pharmacological gradings of all animals in the group. After 1 week of activity (Figure 1). administration of Ro 65-7199 (Table 1, Figure 2), the mean In the present study we sought to further investigate Ro severity increased from 1.0 in the low-dose group to 1.8 in 65-7199-induced hepatotoxicity, which in this case refers the mid-dose group to 2.8 in the high-dose group. Regarding specifically to hepatic microvesicular steatosis, using a the incidence of the findings, only one out of four animals toxicogenomics approach. In order to gain insight into the in the time-matched control group showed fatty change, possible molecular mechanisms underlying the described while all rats treated during 7 days with Ro 65-7199 were hepatic findings, the effect of acute (single dose) and affected. In addition, hepatocellular vacuolation affecting subchronic (1 week, daily dosing) administration of Ro 65- all rats in the high-dose group (mean severity 2.6) was 7199 on hepatic gene expression was assessed using noted. Hepatocellular vacuolation was characterized by Affymetrix GeneChips. In addition, the effect of acute multiple small discrete periacinar vacuoles that sometimes exposure to Ro 66-0074 was also analyzed. Thus, we have coalesced to form larger ones and stained positive for lipids generated gene expression profiles of the livers of rats (microsteatosis). treated with both 5-HT6 receptor antagonists and evaluated A single administration of Ro 65-7199 at 400 mg/kg the differences regarding the modulation of gene resulted in an increase in the severity of fatty change (mean expression. grade 2.2) at 24 h with respect to the controls (mean grade With this work we also intend to provide a set of data 1.2) and with Ro 66-0074 at 400 mg/kg (mean grade 1.4). supporting toxicogenomics as an approach to address This change was correlated with hepatocellular vacuolation toxicological issues and liability.8–12 Several investigators (microsteatosis). In addition, rats treated with Ro 65-7199 from academia and industry have vastly speculated upon showed a slight increase in the incidence (three out of five) the usefulness of gene expression analysis in toxicology, but and grade (1.7) of mitosis. relatively few studies describing the characterization of There were only negligible differences in the livers of rats already well-known hepatotoxins using transcript profiling 6 h after treatment with Ro 65-7199 or Ro 66-0074 when have been published.13–15 compared with their respective controls. Hepatic lipid accumulation after 7 days of Ro 65-7199 treatment (400 mg/kg/day) was accompanied by a slight decrease in serum cholesterol and triglycerides, as well as by a slight increase in serum 50-nucleotidase (5-ND) and g-glutathione transferase (g-GT). The latter reached detectable levels only in the animals treated with the high dose of Ro 65-7199 during 7 days and was under the detection limit for all other groups (data not shown). Liver transaminases (aspartate aminotransferase (AST) and alanine aminotrans- ferse (ALT)) and lactate dehydrogenase (LDH), typical markers of hepatic toxicity, were decreased in most treatment groups rather than increased (Table 2). Serum triglycerides showed a dose-dependent decrease after subchronic treatment, but this finding did not reach statistical significance. Conversely, a transient, significant increase in serum triglycerides was seen after acute treatment with Ro 65-7199 at 24 h. Total hepatic lipid content was also increased at high doses of Ro 65-7199 after 7 days or 24 h of treatment but this increase did not show dose dependence. In addition, an increase in hepatic lipid content was also observed with Ro 66-0074 after 24 h of treatment (Table 2).

Gene Expression Analysis Analysis of the gene expression data showed a relatively Figure 1 Chemical structures of Ro 65-7199 (4-amino-N-(6-bromo- small number of genes modulated by either compound. It 1H -indol-4-yl)-benzenesulfonamide) and Ro 66-0074 (4-(2-bromo- was interesting to note that even the livers of animals 6-pyrrolidin-1-yl-pyridine-4-sulfonyl)-phenylamine). treated during 7 days with the high dose of Ro 65-7199 and

www.nature.com/tpj Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 322

Table 1 Assessment of the histopathological findings in liver sections

Lesion Subchronic, 7 days Acute, 6 h Acute, 24 h

Control Ro 65-7199 Control Ro 65-7199 Ro 66-0074 Control Ro 65-7199 Ro 66-0074 (vehicle) (vehicle) (400 mg/kg) (400 mg/kg) (vehicle) (400 mg/kg) 400 mg/kg 30 mg/ 100 mg/ 400 mg/ kg/day kg/day kg/day

Fatty change 1.0 1.0 1.8 2.8 1.3 1.3 1.4 1.2 2.2 1.4 (1/4) (5/5) (4/4) (5/5) (4/5) (3/5) (5/5) (5/5) (5/5) (5/5) Vacuolation — — — 2.6 — — — — 2.0 — (5/5) (3/5) Mitosis — — — — — — — 1.0 1.7 — (1/5) (3/5)

Findings are expressed as average grading for the group and its incidence (number of animals affected/number of animals evaluated). Grading ranges from 0 (no finding) to 5 (severe findings).

Figure 2 Representative microphotographs of liver sections and stained with hematoxylin–eosin and with Fat-Red-O- for the visualization of lipids (stained red). Magnification: Â 20. Tissue was collected after 7 days of treatment with increasing doses of Ro 65-7199. (a) Control, (b) 30 mg/kg/day, (c) 100 mg/kg/day, and (d): 400 mg/kg/day.

displaying steatosis showed modulation (two-fold change, at 24 h (more than two-thirds of the probe sets). Very Pr0.05, signal Z100) of only 85 out of 8800 probe sets few genes were regulated by the high dose (400 mg/kg) of represented on the array. From these 85 probe sets, only 30 the nonhepatotoxic compound Ro 66-0074 at 6 or 24 h were induced while the rest were repressed. Approximately after a single administration. The expression levels of the half of the genes upregulated by Ro 65-7199 after 7 days of genes used for the cluster analysis depicted in Figure 3 treatment at 400 mg/kg showed induction at the lower dose are listed in Table 3. In addition, expression levels of levels (30 and 100 mg/kg) in a dose-dependant manner. selected genes discussed further due to mechanistic rele- Some of the genes induced by subchronic treatment with vance are included in Table 4. A complete list of all gene Ro 65-7199 were also upregulated following acute treat- expression values can be found at http://www.roche.com/ ment at 6 h (approximately one-third of the probe sets) and science-download.htm.

The Pharmacogenomics Journal Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 323

Table 2 Effect of the compounds on clinical chemistry parameters in serum and hepatic tissue

Treatment group Trigly # T-Chol # 5-ND # LDH # AST # ALT # T-Lipida Triglya PhLipa

Ro 65-7199 (400 mg/kg, 6 h) 79769 93742 105714 ND 142741* 113732 114729 1767178 108723 Ro 66-0074 (400 mg/kg, 6 h) 109774 108738 105718 ND 116729 99723 106739 1807206 98735 Ro 65-7199 (400 mg/kg, 24 h) 2447114* 60730* 88711* 56731* 99732 86715* 155760* 70763 126734* Ro 66-0074 (400 mg/kg, 24 h) 104745 117748 96716 82741 102722 110712* 135752 80798 115733 Ro 65-7199 (30 mg/kg, 7 days) 123778 93719 100713 51714* 77711* 109719 118753 1847336 116725 Ro 65-7199 (100 mg/kg, 7 days) 104767 87721 113722 46712* 78712* 114726 111751 1697177 107739 Ro 65-7199 (400 mg/kg, 7 days) 68735 77731 142770 49719* 7379* 108722 143756* 1627221 142746*

Values are expressed as percentage of the matched time controls7SD. Trigly: triglycerides; T-Chol: total cholesterol, T-Lipid: total lipids; PhLip: phospholipids; ND: not determined; #: lipid levels and enzyme activities measured in serum. *Pp5% after t-test. aLevels measured in hepatic tissue extracts.

Further analysis using hierarchical clustering (Figure 3a In particular, the induction of CYP2B2 after 7 days of and b) showed a characteristic pattern for Ro 65-7199 across repeated treatment with Ro 65-7199 (400 mg/kg/day) re- doses and time points that allowed its differentiation from sulted in a weaker induction, as expected from the dose Ro 66-0074. Notably, genes modulated after subchronic dependence observed using microarrays. This weaker induc- treatment with Ro 65-7199 could be employed to differ- tion might be due to a technical artifact, as indicated by the entiate both compounds after acute administration. Among dose-dependent induction detected with the primer pair the genes induced by Ro 65-7199 were several probe sets directed to a different portion of the sequence (CYP2B, exon specific to cytochromes P450 CYP2B1/2 and CYP3A1, while 9). For the weakly expressed glutathione reductase, the the repressed genes included ornithine aminotransferase slight induction (2.1-fold at the high dose after 7 days) (AA893325), alcohol sulfotransferase (D14987), hepatocyte detected with the microarray after 7 days of Ro-65-7199 nuclear factor 3 gamma (HNF-3g, AB017044), and serine administration could not be confirmed (Table 5). The results dehydratase (J03863). In addition, some genes related to obtained by PCR analysis show that the expression levels of lipid and cholesterol homeostasis, such as hydroxymethyl- these genes after acute administration (6 and 24 h) would glutaryl-coA synthase (HMGCoA Synthase, X52625) and suffice to differentiate both compounds, even though the CYP7B1 (U36992), were deregulated by the treatment with genes were selected based on the results obtained after the steatotic compound Ro 65-7199. Although the extent of subchronic exposures to Ro 65-7199. the change did not always reach the arbitrarily set threshold of two-fold, some genes appeared consistently regulated and Western Blots possibly bear biological relevance. In particular, senescence Among the upregulated probe sets, several were specific to marker protein (D31662), hepatocyte nuclear factor 1-alpha cytochrome P450. In particular, an induction of CYP2B was (HNF-1a, J03170) and the epidermal and the hepatic forms detected by several probe sets and confirmed by RT-PCR. of fatty acid binding protein (S69874 and V01235) were Western blots were performed using specific antibodies significantly repressed, while malic enzyme (AI171506) and against the CYP2B family in order to evaluate if the clear epoxide hydrolase (M26125) were induced (Table 4). induction of messenger RNA also led to an increase in the Results obtained using the Affymetrix arrays confirmed hepatic protein levels. In addition, Western blots using an that expression levels of the pharmacological target of both antibody against CYP2C11, whose mRNA was not induced test compounds, the 5-HT6 receptor (S62043), is not by the treatment, was also assessed (Figure 4). The results of detectable in the liver either in controls or treated animals. the protein levels of both cytochromes closely paralleled the amounts of mRNA at 24 h and at 7 days after Ro 65-7199 RT-PCR treatment. However, the induction of CYP2B could not be The cluster analysis and the data depicted in Figure 3 and detected 6 h after administration of Ro 65-7199, in spite of Table 3 show several genes that allow the discrimination of the increased levels of messenger RNA (Figure 5). As both compounds after acute administration. A group of expected, none of the treatments resulted in an induction genes induced by Ro 65-7199 but not by Ro 66-0074 was of CYP2C11. chosen for confirmation of the gene expression results using RT-PCR monitored by Sybr-green. Generally, the induction DISCUSSION obtained with the microarrays could be confirmed by RT- We investigated the effects of two 5-HT6 receptor antago- PCR, although the fold changes obtained with both plat- nists on the gene expression patterns in the liver of male forms often differed. This is particularly true for the rats. Both compounds share the same pharmacological cytochrome P450 CYP2B, where induction could be clearly target, but results from prior unpublished studies uncovered confirmed using two different primer pairs. Nevertheless, it strikingly different toxicity profiles: while Ro 65-7199 is needs to be pointed out that the transcriptional induction of known to cause hepatic steatosis, Ro 66-0074 has never been this gene as assessed by RT-PCR shows some inconsistencies. shown to induce any hepatotoxicity. In the present study,

www.nature.com/tpj Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 324

animals.3,4 Since the pharmacological target is absent from hepatic tissue, it is expected that these compounds will cause few, if any direct pharmacological reactions in the liver. Indeed, this was confirmed by the relatively small number of genes that appeared modulated even after treatment with a high dose of the toxic compound Ro 65- 7199 during 7 consecutive days. In previous toxicogenomics studies using microarrays evaluating a similar number of genes, the investigated model hepatotoxins usually affected hundreds of genes, especially when administered at high doses.14,16 This is due to the direct effect of the compounds on the liver, as well as to indirect effects due to widespread hepatocellular damage. In the present study, the compounds did not elicit a pharmacological effect in the liver, not even at the high doses selected for toxicity studies, which probably surpass the expected human exposure. Thus, the changes in gene expression may be chiefly attributed to the mechanisms leading to or caused by the specific toxic event triggered by Ro 65-7199 or to unspecific signals. The gene expression profiles obtained after exposure to both compounds clearly show that they can be distinguished based on their effect on gene expression patterns in the liver after acute exposure. As shown in the cluster analysis (Figure 3), this was true at 6 and 24 h after single administration, while conventional toxicology end points were unspecific. These results demonstrate that two very closely related compounds can be differentiated using toxicogenomics. More- over, potential side effects can be detected earlier with gene expression analysis than with conventional toxicology measurements. This finding is vital in order to further assess the usefulness of toxicogenomics approaches to improve the process of detecting the toxic potential of compounds quicker and more accurately than with conventional toxicity studies. In addition, gene expression results after subchronic exposure to Ro 65-7199 showed that the modulation of gene expression also reflects the histological findings. As can be extracted from Table 4, a group of the Figure 3 Hierarchical clustering using 85 probe sets significantly modulated genes shows a dose dependence in the intensity (two-fold change, P-value r0.05) regulated after 7 days of of the modulation that parallels the severity of the treatment with Ro 65-7199 (400 mg/kg/day). Cluster analysis histological findings. (either by groups or by genes) was performed using a correlation From a mechanistic point of view, analysis of complex algorithm to calculate the distance function and the average gene expression profiles has allowed us to identify genes linkage method. (a) Cluster by treatment group. The nonsteatotic that may be associated with the toxic event under Ro 66-0074 clusters together with the vehicle-treated animals, while Ro 65-7199 forms a separate branch. (b) Heat map showing the investigation. In this case, it was surprising that some induction (red) and repression (blue) of the 85 regulated genes in cytochromes P450 (CYP2B and CYP3A1) were consistently each treated group with respect to the matched controls. A list of induced by Ro 65-7199. In particular, the induction of these genes including their names, accession numbers, and CYP2B2 is a gene expression change that may be associated numerical values for fold induction can be found in Table 3. with the toxicity elicited by Ro 65-7199 (Figure 5). Our results show a dose-dependent induction of CYP2B2 mRNA after 7 days of treatment as well as at 6 and 24 h after a single we have reproduced the findings of the above-mentioned dose administration of Ro 65-7199. The nonhepatotoxic Ro unpublished toxicology studies with respect to the hepato- 66-0074 did not cause the induction of this gene. Moreover, toxicity (steatosis) caused by Ro 65-7199 after subchronic the increase in mRNA was paralleled by an increase in exposure. Moreover, we have gained insight into the immunoreactive protein as detected by Western blot underlying mechanisms leading to or at least indicative of analysis at 24 h and at 7 days. The lack of increase in the such toxic events. Gene expression results obtained using protein content at 6 h is due to the time lag between Affymetrix GeneChips confirmed the lack of expression of transcription and translation processes described else- 16,17 the 5-HT6 receptor in the liver of control and treated where. This induction is specific for certain cytochromes

The Pharmacogenomics Journal Table 3 Gene expression values of genes that are significantly modulated by 400 mg/kg/day of Ro 65-7199 during 7 days, as depicted in the heat map in Figure 3b

Affymetrix probe set ID Acc. no. Gene Short name Max. signal Ro 65-7199 Ro 65-7199 Ro 65-7199 Ro 65-7199 Ro 66-0074 Ro 65-7199 Ro 66-0074 symbol (30 mg/kg/day, (100 mg/kg/day, (400 mg/kg/day, (400 mg/kg, 6 h) (400 mg/kg, 6 h) (400 mg/kg, (400 mg/kg, 7 days) 7 days) 7 days) 24 h) 24 h)

FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value

Cytokine-inducible AF065161_at AF065161 Cish SOCS 182 1.3 0.474 1.7 0.337 2.7 0.011 0.5 0.184 0.9 0.847 0.9 0.860 1.5 0.497 V01216_at V01216 Orm Alpha-1-acid 12427 1.4 0.129 2.0 0.032 2.5 0.007 0.9 0.377 0.8 0.311 1.8 0.053 0.8 0.051 glycoprotein Energy rc_AI171506_g_at AI171506 Me1 Malic enzyme 996 2.0 0.025 2.1 0.021 2.9 0.045 1.6 0.063 2.3 0.070 2.2 0.077 1.2 0.300 rc_AA893325_at AA893325 Oat Ornithine 315 1.0 0.911 0.8 0.548 0.4 0.011 1.2 0.550 1.0 0.728 0.9 0.819 1.1 0.788 aminotransferase M93297cds_at M93297 NA Ornithine 368 1.1 0.631 0.8 0.394 0.4 0.001 1.1 0.526 1.2 0.108 0.6 0.128 1.0 0.907 aminotransferase J03863_at J03863 Sdh Serine dehydratase 1671 0.9 0.693 0.6 0.067 0.5 0.032 1.6 0.296 1.1 0.828 0.3 0.034 0.4 0.068 J03865mRNA_f_at J03865 NA Serine dehydratase 437 0.7 0.204 0.5 0.059 0.4 0.040 1.4 0.047 1.0 0.984 0.8 0.298 0.7 0.302 X13119cds_s_at X13119 NA Serine dehydratase 307 0.8 0.395 0.6 0.113 0.4 0.036 2.6 0.217 0.8 0.708 0.4 0.046 0.4 0.040 M35270complete_seq_i_at M35270 Spat Serine-pyruvate 1198 0.9 0.413 0.7 0.020 0.5 0.007 1.0 0.700 1.0 0.724 1.2 0.282 1.4 0.065 aminotransferase M35270complete_seq_r_at M35270 Spat Serine-pyruvate 337 0.7 0.066 0.6 0.023 0.4 0.010 0.9 0.546 1.0 0.875 1.2 0.524 1.1 0.757 aminotransferase Lipid metabolism and transport S69874_s_at S69874 NA FABP (epidermal) 792 0.9 0.730 0.5 0.045 0.4 0.018 1.3 0.177 1.2 0.484 0.9 0.590 1.1 0.693 X52625_at X52625 Hmgcs1 HMG-CoA synthase 1469 1.6 0.008 1.4 0.320 2.0 0.023 0.7 0.138 0.9 0.764 0.7 0.055 1.2 0.215 Metabolism

D14987_f_at D14987 Sth2 Alcohol 3292 0.5 0.045 0.5 0.046 0.4 0.038 1.0 0.895 1.5 0.068 1.0 0.844 1.0 0.792 liver rat the on Suter antagonists L receptor 5-HT6 of Effect sulfonotransferase D14988_f_at D14988 Sth2 Alcohol 6654 0.5 0.072 0.6 0.078 0.5 0.049 1.1 0.618 1.4 0.108 1.3 0.282 1.2 0.370 tal et sulfonotransferase rc_AA818122_f_at AA818122 Sth2 Alcohol 3948 0.5 0.067 0.5 0.065 0.3 0.031 1.4 0.271 1.6 0.046 1.5 0.106 0.8 0.469 sulfonotransferase AF001898_at AF001898 Aldh1a1 Aldehyde 4113 1.3 0.335 1.8 0.105 2.5 0.010 0.7 0.111 0.9 0.599 3.7 0.048 1.8 0.093 dehydrogenase D87336_at D87336 NA Bleomycin hydrolase 146 0.8 0.066 0.8 0.302 0.5 0.000 1.8 0.060 1.6 0.122 1.2 0.502 0.7 0.078 AB010635_s_at AB010635 Ces2 Carboxylesterase 2 2177 2.0 0.085 4.1 0.027 6.1 0.017 1.2 0.426 1.0 0.801 2.8 0.068 0.8 0.293 U73174_g_at U73174 Gsr Glutathione reductase 248 1.5 0.175 1.4 0.116 2.1 0.007 1.1 0.837 1.3 0.238 1.8 0.160 0.8 0.240 K00136mRNA_at K00136 Gsta2 Glutathione S- 7805 1.5 0.014 2.1 0.004 2.5 0.010 1.4 0.061 1.5 0.097 1.5 0.130 0.8 0.285 transferase ya M13506_at M13506 UDPGTR-2 UDPGT 5809 3.1 0.001 4.3 0.003 5.6 0.002 1.5 0.181 1.3 0.204 3.0 0.003 0.9 0.757 Metabolism/cytochromes P450 K03241cds_s_at K03241 NA CYP1A2 3384 1.2 0.494 1.9 0.000 2.0 0.029 1.2 0.413 1.3 0.304 1.8 0.119 0.8 0.261 M11251cds_f_at M11251 NA CYPB exon1 17769 3.6 0.002 9.6 0.028 13.3 0.000 10.7 0.002 1.1 0.775 9.8 0.001 0.8 0.531 D17349cds_f_at D17349 NA CYP2B15 4160 3.1 0.003 5.0 0.028 6.7 0.002 4.7 0.001 1.2 0.245 5.2 0.001 1.0 0.987 K01721mRNA_s_at K01721 Cyp2b15 CYP2B15 27043 4.0 0.000 4.3 0.014 3.2 0.025 9.6 0.000 1.5 0.044 3.5 0.031 1.4 0.233 J00728cds_f_at J00728 NA CYP2B2 27994 3.4 0.000 4.9 0.002 5.4 0.000 5.8 0.002 1.1 0.782 4.7 0.002 1.2 0.287 K00996mRNA_s_at K00996 NA CYP2B2 18428 6.5 0.001 15.8 0.028 20.8 0.002 11.3 0.006 1.2 0.193 17.5 0.001 1.3 0.268 M13234cds_f_at M13234 NA CYP2B2 26127 4.2 0.002 7.3 0.007 9.2 0.000 8.2 0.002 1.4 0.166 8.3 0.000 1.2 0.558 M14776_f_at M14776 NA CYP2C6 11687 2.1 0.008 2.4 0.009 2.5 0.007 1.2 0.560 1.0 0.889 1.7 0.005 0.9 0.643 www.nature.com/tpj rc_AA945571_s_at AA945571 Cyp2c6 CYP2C6 20736 2.2 0.000 2.4 0.000 2.7 0.000 1.1 0.159 0.8 0.018 1.6 0.006 0.9 0.510 D13912_s_at D13912 Cyp3a1 CYP3A1 11790 1.5 0.017 3.6 0.048 5.7 0.009 1.5 0.035 1.2 0.077 3.6 0.026 0.8 0.320 L24207_i_at L24207 Cyp3a1 CYP3A1 716 1.4 0.056 1.5 0.085 2.6 0.048 1.1 0.181 1.2 0.026 3.1 0.078 1.2 0.237 L24207_r_at L24207 Cyp3a1 CYP3A1 601 1.7 0.162 2.9 0.010 4.2 0.029 1.3 0.216 1.6 0.035 4.3 0.035 1.2 0.394 X64401cds_s_at X64401 Cyp3a1 CYP3A1 27871 1.8 0.002 3.0 0.024 3.1 0.005 1.2 0.210 1.0 0.982 1.6 0.029 0.8 0.384 U36992_at U36992 Cyp7b1 CYP7B1 189 0.8 0.227 0.7 0.097 0.5 0.004 1.3 0.284 1.0 0.833 1.6 0.221 1.3 0.338

L00320cds_f_at L00320 NA CYPB exon9 16700 5.1 0.005 12.6 0.027 17.2 0.002 13.5 0.002 1.4 0.235 11.0 0.001 1.1 0.749 325 M24239cds#2_f_at M24239 NA CYP 23856 2.0 0.000 2.3 0.000 2.5 0.000 1.1 0.357 1.0 0.772 1.8 0.000 0.9 0.510 h hraoeoisJournal Pharmacogenomics The Table 3 (continued) 326

FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value

Nuclear receptors/receptors AB017044exon_at AB017044 NA HNF3-g 210 0.8 0.022 0.8 0.468 0.5 0.000 0.9 0.745 1.0 0.986 0.8 0.282 0.9 0.622 L48060_s_at L48060 Prlr Prolactin receptor 237 0.3 0.063 0.8 0.408 0.3 0.048 1.3 0.461 1.5 0.025 4.0 0.031 2.6 0.036

U56839_at U56839 P2ry2 Purinergic receptor 312 1.9 0.083 1.7 0.007 2.6 0.048 2.2 0.044 1.4 0.193 1.2 0.576 1.2 0.383 liver rat the on antagonists receptor 5-HT6 of Effect P2Y AF091566_f_at AF091566 NA htf-sp1 olfactory 228 0.7 0.157 0.5 0.011 0.4 0.003 0.9 0.530 1.4 0.070 0.9 0.560 1.0 0.883 receptor Tissue damage/differentiation (NGF-inducible)/cell cycle regulation X07648cds_at X07648 NA Amyloidogenic 330 1.3 0.050 1.4 0.055 2.0 0.009 0.9 0.653 1.0 0.943 1.8 0.104 1.4 0.025 glycoprotein U52102_at U52102 Crmp1 Dihydropyrimidinase- 165 0.7 0.085 0.6 0.064 0.3 0.000 0.7 0.002 0.6 0.001 0.8 0.351 0.9 0.698 related protein-1 U31777_g_at U31777 Drpla Dentatorubral 133 0.6 0.132 0.7 0.257 0.4 0.044 1.1 0.821 1.5 0.224 4.2 0.000 2.6 0.014 pallidoluysian atrophy AF050659UTR#1_at AF050659 NA ANIA-7 115 0.6 0.034 0.5 0.003 0.5 0.001 1.6 0.061 0.6 0.224 1.2 0.299 1.1 0.688 E12286cds_at E12286 NA gm2 activator protein 524 0.6 0.059 0.8 0.468 0.5 0.034 1.0 0.771 0.9 0.745 0.8 0.503 0.8 0.582 U95178_s_at U95178 Dab2 Mitogen-responsive 182 0.6 0.049 0.8 0.195 0.5 0.032 0.9 0.673 1.0 0.924 1.1 0.715 0.8 0.485 phosphoprotein (Drosophila) Suter L Cell–cell junction, structural proteins

AF090134_at AF090134 Veli1 lin-7-Ba 281 0.7 0.133 0.7 0.144 0.5 0.031 0.8 0.152 1.1 0.495 1.0 0.929 1.1 0.298 al et M28259cds_at M28259 NA Fibronectin 438 0.6 0.048 0.9 0.663 0.5 0.016 1.1 0.522 1.1 0.448 0.7 0.108 0.8 0.388 M19357cds_f_at M19357 NA Gamma crystallin f 107 2.2 0.007 2.5 0.054 2.9 0.022 1.1 0.658 0.7 0.206 2.2 0.102 1.9 0.030 AF016252_at AF016252 LOC84686 Spinophilin 284 0.7 0.110 0.6 0.061 0.5 0.027 1.1 0.612 1.2 0.129 0.9 0.517 1.0 0.762 S78154_at S78154 ATP-regulated k+ 130 0.9 0.388 0.6 0.062 0.4 0.005 1.0 0.911 0.7 0.142 1.1 0.509 1.4 0.003 channel rc_AI044985_g_at AI044985 Gjb5 Gap junction 279 0.7 0.095 0.7 0.084 0.4 0.007 0.8 0.335 1.4 0.020 1.1 0.913 1.1 0.788 membrane channel protein beta 5 Transcription factors/DNA interacting proteins D13417_at D13417 Hes1 Hes 1 186 0.4 0.001 0.4 0.003 0.4 0.002 1.4 0.419 1.0 0.990 0.9 0.520 0.9 0.590 rc_AI137583_at AI137583 Id2 Inhibitor of DNA 1029 0.8 0.406 0.4 0.027 0.4 0.025 1.4 0.100 1.4 0.092 2.0 0.034 1.5 0.078 binding 2 rc_AI112516_at AI112516 Zfp36l1 Zinc-finger protein 36 1251 0.7 0.087 0.6 0.040 0.5 0.013 1.4 0.402 1.2 0.273 1.1 0.613 0.9 0.637 Kinases/phosphatases L27061_g_at L27061 NA cAMP-dependent 125 0.5 0.020 0.8 0.054 0.4 0.008 1.0 0.937 1.1 0.688 0.9 0.640 0.7 0.370 PDE4c L01624_at L01624 Sgk Serum/glucocorticoid- 145 0.7 0.109 0.5 0.048 0.5 0.037 1.1 0.495 1.1 0.186 1.2 0.472 2.3 0.032 regulated kinase rc_AA955477_g_at AA955477 Map3k1 MAPK3 kinase 139 0.6 0.069 0.8 0.213 0.4 0.012 1.3 0.375 1.4 0.070 1.0 0.946 1.1 0.695 U27319exon_at U27319 NA Hexokinase 1 110 0.6 0.112 0.6 0.076 0.5 0.033 0.8 0.469 0.6 0.075 0.9 0.797 0.8 0.353 U48596_g_at U48596 Map3k1 MEKK1 370 0.7 0.005 0.6 0.041 0.5 0.000 1.3 0.318 1.1 0.462 0.9 0.446 0.9 0.329 U50412_at U50412 Pik3r1 Phosphoinositide 3- 114 0.5 0.052 0.6 0.145 0.4 0.032 0.6 0.099 0.7 0.191 0.8 0.565 1.1 0.808 kinase X95577_at X95577 Prkab1 AMPK-beta chain 317 0.6 0.036 0.6 0.065 0.3 0.012 1.2 0.502 1.0 0.854 1.4 0.174 1.7 0.016 Miscellaneous AF037072_at AF037072 Ca3 Carbonic anhydrase 3 5707 1.1 0.675 0.6 0.020 0.5 0.005 0.8 0.182 0.9 0.226 0.3 0.002 0.9 0.429 AF069782_at AF069782 NA Nucleolar protein 120 0.7 0.204 0.6 0.098 0.5 0.043 1.4 0.255 1.4 0.301 1.9 0.223 1.1 0.654 nop5 C07012_f_at C07012 Ppicap Peptidylprolyl 228 0.8 0.494 0.4 0.028 0.4 0.031 1.4 0.289 1.7 0.197 1.9 0.104 1.5 0.375 isomerase C-associated protein L07925_g_at L07925 Ralgds Ral guanine nucleotide 113 0.9 0.609 0.9 0.506 0.5 0.024 1.4 0.410 0.9 0.810 1.4 0.095 1.5 0.025 dissociation stimulator M12822cds_f_at M12822 NA Ig kappa chain 118 0.5 0.037 0.7 0.014 0.4 0.031 2.6 0.058 3.3 0.018 0.8 0.609 0.8 0.565 Table 3 (continued)

FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value

M58308_at M58308 Hal Histidine ammonia 1271 0.9 0.201 0.6 0.013 0.5 0.000 0.9 0.440 0.8 0.188 1.0 0.670 1.2 0.327 lyase M75281_at M75281 NA Cystatin s precursor 464 0.5 0.083 0.8 0.508 0.4 0.048 1.3 0.565 1.1 0.608 1.3 0.483 1.3 0.556 M92042_at M92042 Ndst1 N-deacetylase/N- 118 0.8 0.423 1.0 0.808 0.5 0.012 1.1 0.626 1.1 0.641 1.0 0.794 0.8 0.448 sulfotransferase 1 rc_AA859740_at AA859740 NA N-deacetylase/N- 117 0.6 0.040 0.7 0.000 0.5 0.001 0.8 0.450 1.0 0.931 0.7 0.256 0.9 0.573 sulfotransferase 1 rc_AA851379_at AA851379 NA NADH-ubiquinone 141 3.2 0.000 3.1 0.081 3.1 0.024 1.0 0.991 0.6 0.206 1.6 0.256 1.8 0.212 oxidoreductase rc_AA875054_at AA875054 Tcp1 T-complex 1 186 1.2 0.735 0.8 0.617 2.2 0.023 1.5 0.297 1.5 0.367 2.2 0.112 2.6 0.131 rc_AI232087_at AI232087 Hao3 Hydroxyacid oxidase 3 1510 1.1 0.686 0.6 0.039 0.5 0.015 0.8 0.116 0.9 0.398 0.7 0.053 1.2 0.280 S81025_at S81025 NA NAG-galactosyl 125 0.8 0.469 0.6 0.088 0.4 0.006 1.2 0.613 1.3 0.510 0.8 0.386 0.9 0.783

transferase liver rat the on Suter antagonists L receptor 5-HT6 of Effect X06357cds_s_at X06357 NA Serine-pyruvate 2583 0.6 0.058 0.6 0.041 0.5 0.020 1.2 0.042 1.2 0.253 1.1 0.775 1.1 0.522

aminotransferase al et rc_AA893011_at AA893011 NA Na-dependent BA 243 1.3 0.171 1.9 0.116 2.2 0.026 0.8 0.360 0.9 0.752 1.2 0.321 1.3 0.315 transporter U03388_s_at U03388 Ptgs1 COX-1 113 0.8 0.331 0.8 0.343 0.4 0.024 1.1 0.837 1.0 0.977 0.8 0.349 1.3 0.032 U17565_at U17565 Mcmd6 Mini-chromosome 151 0.8 0.467 0.5 0.077 0.4 0.035 1.1 0.897 1.1 0.770 2.3 0.025 1.6 0.166 maintenance deficient 6(S. cerevisiae) X06150cds_g_at X06150 Gnmt Glycine 1817 0.8 0.024 0.8 0.226 0.5 0.000 1.5 0.058 0.9 0.634 0.5 0.007 0.9 0.593 methyltransferase Unknown S74257_at S74257 NA Unknown 126 1.0 0.584 1.0 0.923 0.4 0.008 1.5 0.152 1.4 0.197 1.2 0.596 1.2 0.323 rc_AA892317_at AA892317 NA EST 259 1.1 0.673 2.4 0.078 2.2 0.011 1.2 0.694 1.4 0.014 3.8 0.028 2.4 0.024 rc_AA859722_at AA859722 NA EST 133 0.9 0.274 0.4 0.110 0.5 0.014 1.8 0.257 1.4 0.451 1.3 0.386 1.2 0.715 rc_AA893980_at AA893980 NA EST 136 0.6 0.031 0.4 0.012 0.4 0.010 1.6 0.100 0.9 0.869 2.3 0.035 1.9 0.116

# Genebank acc. no., gene symbol, and short name: extracted and adapted from Affymetrix descriptions (www.affymetrix.com). NA: not available. Max. signal: the average intensity of the treatment group with the highest signal. This parameter indicates whether a gene is high or low expressed. Probe sets with intensity lower than 100 are considered to be nondetectable. FChg: induction/repression expressed as fold change with respect to the time-matched control. P-value: significance level obtained from a t-test. www.nature.com/tpj 327 h hraoeoisJournal Pharmacogenomics The 328

Table 4 Gene expression values of genes that have been discussed due to their function and/or to the degree of modulation by the test compounds

FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value

Pharmacological target S62043_s_at S62043 NA 5-HT6 receptor 67 0.6 0.061 0.6 0.124 0.7 0.131 1.4 0.330 1.6 0.267 1.1 0.793 1.5 0.314 feto -T eetratgnsso h a liver rat the on antagonists receptor 5-HT6 of Effect Cytokine-inducible M60921_g_at M60921 Btg2 PC3 (NGF-inducible) 175 1.1 0.804 0.8 0.533 1.0 0.905 0.6 0.246 0.7 0.405 1.9 0.076 0.5 0.062 Energy D10874_at D10874 Atp6l ATP synthase 2099 0.8 0.011 0.8 0.149 0.7 0.004 0.8 0.316 0.8 0.252 1.2 0.220 1.1 0.455 proteolipid D13124_s_at D13124 Atp5g2 ATP synthase 2927 0.9 0.026 0.8 0.018 0.8 0.012 1.1 0.285 1.1 0.307 1.1 0.476 1.1 0.201 proteolipid Lipid metabolism and transport rc_AA799541_at AA799541 Dci 3-2-Trans-enoyl-CoA 127 0.5 0.005 0.6 0.000 0.5 0.005 1.4 0.303 1.4 0.275 1.1 0.588 1.0 0.869 isomerase rc_AI170568_s_at AI170568 Dci 3-2-Trans-enoyl-CoA 598 0.6 0.023 0.6 0.021 0.7 0.046 1.0 0.979 1.0 0.846 1.6 0.224 1.3 0.252 isomerase D00512_g_at D00512 NA Acetoacetyl-CoA 544 0.9 0.333 1.0 0.737 0.8 0.018 1.0 0.734 1.1 0.374 0.9 0.762 1.1 0.460 thiolase

X05341_at X05341 Acaa2 Acetyl-CoA 7193 1.3 0.425 1.3 0.168 1.4 0.028 0.8 0.091 0.9 0.653 1.5 0.154 2.0 0.051 Suter L acetyltransferase

M00002_at M00002 Apoa4 Apo AIV 496 0.8 0.182 0.8 0.294 0.6 0.006 1.0 0.976 1.0 0.992 0.9 0.310 1.1 0.584 al et L07736_at L07736 Cpt1a CPT-1 1222 0.7 0.095 0.7 0.051 0.5 0.019 1.4 0.260 1.1 0.285 0.8 0.224 0.8 0.366 M26125_at M26125 Ephx1 Epoxide hydrolase 10075 1.4 0.020 1.5 0.008 1.9 0.003 1.2 0.134 1.1 0.095 1.9 0.021 1.0 0.809 rc_AI104882_s_at AI104882 Ephx2 Epoxide hydrolase 258 1.4 0.077 1.3 0.208 1.3 0.049 0.9 0.829 1.1 0.692 1.0 0.925 0.9 0.839 X60328_g_at X60328 Ephx2 Epoxide hydrolase 199 1.7 0.052 1.4 0.266 1.4 0.034 0.8 0.454 0.7 0.046 0.8 0.027 1.0 0.930 V01235_at V01235 Fabp1 FABP (liver) 8695 0.8 0.038 0.7 0.010 0.7 0.016 0.8 0.209 0.8 0.126 0.9 0.438 1.1 0.410 S62097_s_at S62097 NA Geranylgeranyl 122 0.8 0.009 0.7 0.009 0.6 0.001 1.1 0.650 1.3 0.185 1.4 0.192 1.1 0.635 transferase rc_AI171090_g_at AI171090 Hmgcl HMG-CoA lyase 459 0.9 0.430 0.7 0.003 0.7 0.002 0.9 0.323 0.9 0.562 1.3 0.452 1.4 0.076 M33648_at M33648 Hmgcs2 HMG-CoA synthase 9080 0.9 0.111 0.8 0.018 0.8 0.014 0.9 0.136 0.8 0.087 0.9 0.411 1.1 0.390 L22788_i_at L22788 Fabp6 Ileal lipid binding 6928 1.4 0.067 1.4 0.056 1.6 0.012 1.1 0.483 1.2 0.144 1.6 0.014 1.0 0.918 protein D90109_at D90109 Facl2 Long-chain acyl-CoA 2641 1.1 0.639 0.7 0.024 0.8 0.048 0.9 0.390 0.9 0.247 0.7 0.063 1.1 0.623 synthetase rc_AI044900_s_at AI044900 Facl2 Long-chain acyl-CoA 7158 0.8 0.069 0.8 0.009 0.8 0.024 0.8 0.206 0.8 0.105 0.8 0.310 1.4 0.016 synthetase S70011_g_at S70011 NA Tricarboxylate carrier 399 1.0 0.711 0.7 0.001 0.6 0.000 1.1 0.474 1.1 0.334 1.1 0.606 1.0 0.950 Metabolism X72792cds_s_at X72792 NA Alcohol 3654 0.9 0.453 0.7 0.093 0.6 0.019 0.7 0.084 0.9 0.249 0.8 0.246 1.0 0.943 dehydrogenase D50580_at D50580 LOC192257 Carboxylesterase 2 2154 0.9 0.797 1.4 0.485 2.3 0.053 1.2 0.463 0.9 0.718 2.6 0.002 1.4 0.189 L24896_s_at L24896 Gpx4 Glutathione 1100 0.6 0.004 0.8 0.147 0.7 0.000 0.9 0.752 1.0 0.850 1.0 0.840 0.9 0.750 peroxidase U73174_at U73174 Gsr Glutathione reductase 118 1.0 0.915 1.2 0.586 1.6 0.136 0.9 0.759 1.1 0.652 2.5 0.054 1.4 0.159 Metabolism/cytochromes P450 J02657_s_at J02657 Cyp2c CYP2C11 29188 1.1 0.310 1.2 0.105 1.1 0.370 1.0 0.989 1.1 0.305 0.7 0.108 0.7 0.111 M18363cds_s_at M18363 NA CYP2C11 11716 1.5 0.001 1.4 0.036 1.3 0.055 1.0 0.961 1.1 0.595 0.7 0.129 0.7 0.097 X79081mRNA_f_at X79081 NA CYP2C11 6716 1.4 0.221 1.4 0.132 1.3 0.312 1.2 0.179 1.2 0.341 0.8 0.211 0.8 0.127 Nuclear receptors/receptors AF082124_s_at AF082124 Ahr AhR 98 2.1 0.112 1.6 0.367 2.3 0.071 2.2 0.139 0.7 0.614 1.1 0.894 0.3 0.038 AF082125_s_at AF082125 Ahr AhR 244 3.0 0.174 0.8 0.299 5.1 0.106 2.3 0.211 0.3 0.089 2.5 0.353 0.3 0.215 U61184_at U61184 Arnt1 AhR nuclear 163 0.9 0.666 0.8 0.051 0.7 0.024 1.2 0.019 1.6 0.000 0.9 0.616 1.0 0.737 translocator (ARNT) J03170_at J03170 Tcf1 HNF1-a 268 0.8 0.077 0.7 0.050 0.7 0.038 1.3 0.024 1.3 0.079 1.3 0.028 1.4 0.109 Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 329

P450, such as CYP2B, CYP2C6, and CYP3A1, while no induction was observed for CYP2C11 or other cytochromes

24 h) such as the CYP1A and the CYP4A families. The CYP2B Ro 66-0074 (400 mg/kg, family, and to a lesser extent CYP3A1 and CYP2C6, is known to be inducible by phenobarbital, a rodent-specific non- genotoxic carcinogen with so far little relevance to human hepatotoxicity.18–21 Several probe sets detected the induc- 24 h) tion of genes from the CYP2B family, including CYP2B2 Ro 65-7199 (400 mg/kg, (exons 1 and 9) and CYP2B15, which are all inducible by phenobarbital.22,23 In the present study, a novel link between the induction of CYP2B2 and other phenobarbital-inducible cytochromes P450 in the rat, and the poten-

Ro 66-0074 tially human-relevant microvesicular steatosis could be (400 mg/kg, 6 h) established. Whether the induction of this gene is a cause or an effect of the underlying mechanisms or a possible etectable. FChg: induction/repression expressed as fold

signal: the average intensity of the treatment group with adaptive response is not clear. In any case, this phenobarbital-like response occurs prior to the appearance of major

Ro 65-7199 histological findings, already at the low dose (30 mg/kg, 7 (400 mg/kg, 6 h) days) administered subchronically and at the earliest tested time point (400 mg/kg, 6 h) after acute exposure. At this time, it is difficult to define the mechanistic link between the observed modulation of gene expression and 7 days)

Ro 65-7199 the subsequent liver damage, but the obtained results allow (400 mg/kg/day, for some hypothesis generation. It has been previously published that the effect of phenobarbital on CYP2B induction is gender dependent and that this induction is partially inhibited either at the transcriptional or at the 7 days) 24 Ro 65-7199 post-translational level in females and in younger animals. (100 mg/kg/day, In addition, Kawamoto et al25 have shown that estrogens are activators of the constitutive androstane receptor (CAR), a nuclear receptor that induces the transcription of the CYP2B -test. t

7 days) family after exposure to xenobiotics such as phenobarbital. FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value FChg P-value

(30 mg/kg/day, Thus, there is a link between gender and CYP2B induction, as there is a link between steatosis and gender, women being more prone to hepatic steatosis than men.26 However, the 193 1.3 0.005 1.3 0.029 1.7 0.037 1.2 0.750gender-specific 1.1 0.763 1.8 0.295 1.2 response 0.597 was not addressed in this study, since only male animals were analyzed in order to minimize variability due to asynchrony in the estrus cycles. Addition- ally, it has recently been published that an endogenous molecule related to cholesterol homeostasis interacts with the phenobarbital-response enhancer sequences of some phenobarbital-induced cytopchromes P450,27 and induc- Short name Max. signal Ro 65-7199 glycoprotein tion of rat CYP2B was observed after inhibition of cholester-

-value: significance level obtained from a 28 P ol synthesis. These reports, taken together with the observed induction of CYPs and reduction in serum

symbol cholesterol, suggest a possible causal relation between an impairment of the sterol metabolic pathway (possibly an inhibition of the cholesterol synthesis), the induction of CYP2B, and the fat accumulation in the hepatocytes. Additional results supporting the relationship of cholesterol homeostasis and induction of CYP2B caused by CAR activation as a mechanism of hepatic defense toward bile acid toxicity have been published.29,30 These reports under- line the complex relationships between bile acid, cholester- Continued ol, and lipid homeostasis and activation of nuclear receptors, including CAR.30 Furthermore, our results show a significant decrease in circulating cholesterol and tran- Table 4 Affymetrix probe set ID Acc. no. Gene Tissue damage/differentiation (NGF-inducible)/cell cycleD31662exon#4_s_at regulation X07648cds_g_at# Genebank acc. D31662 no., genethe symbol, highest and signal. short name: This extracted parameterchange NA and indicates with adapted whether respect from a X07648 to Affymetrix gene descriptions the is (www.affymetrix.com). time-matched high NA: control. or not low available. Max. expressed. NA Probe sets with SMP-30 intensity lower than 100 are considered to be nond Amyloidogenic 2116scriptional 1.1 0.561 0.8 regulation 0.087 0.7 0.032 of 0.8 enzymes 0.391 1.0 0.784 related 0.7 0.028 1.2 to 0.234 the sterol

www.nature.com/tpj Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 330 0.106 0.076 (t-test) P-value 5.1 0.017 FChg Max. 9 16.2 0.002 Signal t 26126 8.2 0.000 Affy ID Max. s achieved in the animals treated with 400 mg/kg Ro 66-0074 (400 mg/kg)

Figure 4 Western blots of liver extracts with antibodies specific for CYP2B (upper panels (a, b)) and CYP2C11 (lower panels (c, d)). Two representative animals from each treatment group were analyzed. In all, 10 lg of total hepatic protein was loaded in each Ro 65-7199 (400 mg/kg) lane. Left panels (a, c): lane 1: MW markers; lanes 2 and 3: control (6 h); lanes 4 and 5: Ro 65-7199 (6 h); lanes 6 and 7: Ro 66-0074 -values in italic). P (6 h); lanes 8 and 9: controls (24 h); lanes 10 and 11: Ro 65-7199 (24 h); lanes 12 and 13: Ro 66-0074 (24 h). Right panels (b, d): lane 1: MW markers; lanes 2 and 3: controls (7 days); lanes 4 and 5: 30 Ro 66-0074 (400 mg/kg) mg/kg/day Ro 65-7199 (7 days); lanes 6 and 7: 100 mg/kg/day Ro 65-7199 (7 days); lanes 8 and 9: 400 mg/kg/day Ro 65-7199 (7 days); lane 11: positive control (phenobarbital-induced rat microsomal extract). Inlet bargraphs represent the densitometric quantification of each gel. Ro 65-7199 (400 mg/kg) Ro 65-7199 (400 mg/kg) Ro 65-7199 (100 mg/kg) (30 mg/kg) Ro 65-7199 3.4 (2.2–5.3) 3.0 (1.9–4.5) 5.9 (4.0–8.6) 4.1 1.4–11.6) 0.1 (0.0–0.1) 5.0 (1.0–26.2) 1.8 (0.5–5.8) M13506_at 5809 4.6 0.002

Figure 5 Graphic representation of CYP2B protein levels (line) and gene expression induction (bars) after treatment with Ro 65-7199 or Ro 66-0074 at several times and doses. X07648AF082124 ND ND ND ND 2.0 (1.3–3.0) 2.0 (0.9–4.5) 1.0 (0.5–1.9) 2.7 (0.8–8.8) 1.0 (0.6–1.7) 1.7 (0.6–4.8) 1.7 (1.3–2.2) 1.1 (0.3–3.7) 1.4 (0.8–2.2) 1.2 X07648cds_at (0.5–3.0) AF082125_s_at 329 244 2.0 4.1 0.009 M60921L00320 NDRNUD2A10 M35086; J05482 20.5 (6.6–63.6) 87.9 (35.4–218.6)210.5 ND (74.3–596.3) 90.6 (41.8–196.3) 0.7 (0.3–2.0) 1.1 (0.3–4.5) 95.1 (53.4–169.0) 0.9 (0.5–1.6) L00320cds_f_at 0.5 (0.2–0.9) 1669 0.4 (0.2–1.0) 2.6 (1.5–4.6) 0.6 (0.4–0.9) M60921_g_at 175 0.9

metabolic pathway (downregulation of cytoplasmic HMGCoA synthase and CYP7B1), and of genes coding for proteins involved in cellular lipid homeostasis (ie malic 450 CYP2B2450 J00728 CYP3A1450 D13912 CYP2B 20.2 (9.4–43.2) 27.5 (13.9–218.6) 3.6 (0.7–18.3) ND 31.4 (16.7–59.2) 3.1 (1.9–5.2) 72.0 (10.1–514.2) ND 4.4 (0.7–27.1) M13234cds_f_a 3.5 (2.6–4.8) 1.1 (0.8–1.5) 0.9 (0.6–1.3) 2.3 (1.3–4.3) 0.8 (0.5–1.1) D13912_s_at 11790 4.7 0.009 Results of QRT-PCR analysis of a subset of genes and comparison with the results obtained using the Affymetrix platform P P P enzyme and epoxide hydrolase). Thus, the link between hepatic lipid accumulation and CYP2B induction probably

a occurs due to the cross talk between several nuclear For btg2, maximal induction (measured by microarrays and RT-PCR) occurred 24 h after acute administration. For all other genes, maximal induction wa Alzheimer’s disease amyloid a4 protein (homolog) Aryl hydrocarbon receptor (AhR) btg2 protein (ngf-inducible antiproliferative protein pc3) CarboxylesteraseCytochrome Cytochrome Cytochrome exon 9 AB010635Glutathione reductaseUDP-glucuronosyltransferase 2b1 U73174 2.9 (1.0–8.3) 6.3 (2.8–14.4) 12.3 (5.8–26.1) 1.1 ND (0.6–2.0) 1.3 (0.8–2.0) 5.0 (1.6–16) ND 2.8 (1.7–4.6) AB010635_s_at 1.4 (1.1–1.9) 2177 1.2 (0.8–1.8) 1.1 (0.7–1.8) 1.9 (1.3–2.8) 1.0 (0.8–1.4) U73174_g_at 248 2.1 0.007 Ro 65-7199 during 7 days. The induction observed for btg2 and AhR using microarrays was considered statistically nonsignificant ( Table 5 Description Genebank acc. no. 7 daysRT-PCR data area expressed as mean induction values of five animals per dose group; numbers in parentheses indicate 6 minimal h andreceptors maximal fold changes. and the impairement 24 h of the cholesterol Affymetric results

The Pharmacogenomics Journal Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 331

synthesis.31 In fact the slight, yet significant transcriptional causes leading to the observed liabilities and provide targets regulation of aryl hydrocarbon receptor (AhR) and hepatic amenable to mechanistic in vitro screening tools. nuclear factors (HNF-1a and HNF-3g) might be additional indicators of such a mechanism. MATERIALS AND METHODS Alternatively, it can also be argued that the induction of Animals and Treatment CYPs by the steatotic 5-HT6 receptor antagonist Ro65-7199 This study was designed according to previous unpublished is a reaction to fatty change in the liver rather than part of routine toxicity studies. All animals received humane care as its cause. However, the fact that the upregulation of CYP2B specified by Swiss law and in accordance with the ‘Guide for occurs prior to the onset of steatosis and that an excess of fat the care and use of laboratory animals’ published by the storage in the liver is known to decrease protein concentra- NIH. Male Wistar rats were purchased from BRL (Fu¨ll- 32 tion and enzymatic activities of several CYPs make this ingsdorf, Switzerland) and housed individually. For the unlikely. In any case, there is no evidence that the induced acute exposures, rats (five males/group) were dosed orally CYP2B family is involved in the metabolism of Ro 65-7199. (gavage) once with a high dose (400 mg/kg) of Ro 65-7199 or Metabolic studies demonstrated that the main metabolite of Ro 66-0074 and killed by CO2 asphyxiation 6 or 24 h after this compound is an N-glucoronide, while phase I metabo- dosing. For the subchronic exposures, animals were dosed lism plays a minor role in the rat. However, studies orally (gavage) for 7 days with Ro 65-7199 at 30, 100, or performed with human metabolizing enzymes showed an 400 mg/kg/day and killed 24 h after the last administration. inhibition of human CYP3A4 by Ro 65-7199 (data not For each time point, a time-matched control group (five shown). animals/group) was dosed with the vehicle (distilled water). Another finding that might bear mechanistic relevance to The livers of all animals were excised immediately after the steatosis caused by Ro 65-7199 is the transcriptional killing and placed either in RNALater (Ambion, TX, USA) regulation of several proteins involved in cellular lipid and stored at À201C for gene expression analysis or in liquid homeostasis. Some of these mRNAs were sometimes only nitrogen and kept at À801C for subsequent protein (Western slightly regulated (50% up- or downregulation) by Ro 65- blot) and lipid content measurement. Blood samples for 7199. Among these, an upregulation of microsomal epoxide clinical chemistry were obtained shortly before killing. hydrolase (M26125) and a downregulation of the cutaneous Additional liver samples obtained during necropsy were and hepatic isoforms of fatty acid-binding protein (S69874, fixed for routine histopathological evaluation. V01235) were observed. The latter downregulation could have been a consequence of reduced levels of HNF-1a Clinical Pathology and Histopathology observed. Indeed, the dose-dependent downregulation of Lipid levels and the enzymatic activities of AST, ALT, LDH, HNF-1a, followed by lower levels of FABP are in line with g-GT, and 5-ND were measured in serum samples. Auto- previous reports describing decreased expression of the mated analysis was performed using commercially available hepatic form of FABP in null-HNF1a mice displaying fat test kits (Roche Diagnostics, Mannheim, Germany) on a 33 accumulation in the liver. In our model, this gene Cobas Fara autoanalyzer (Roche, Rotkreuz, Switzerland). expression finding seems to be a late manifestation of Lipids were extracted from liver homogenates as described toxicity, since it only becomes apparent after subchronic by Freneaux et al,34 and the levels of triglycerides, phos- administration of the high dose of Ro 65-7199. pholipids, and total lipids were measured on a Cobas Fara Summarizing, in this study, toxicogenomics allowed an autoanalyzer. For histological evaluation, samples from the in-depth analysis of dose- (several doses during 7 consecu- liver (caudate process, left lateral lobe) were fixed in 10% tive days) and time- (6 and 24 h and 7 days) related gene buffered formalin for at least 24 h, and then embedded in expression events related to the steatotic compound Ro 65- Paraplast. Sections approximately 2–3 mm thick were cut and 7199. Moreover, the inclusion of animals treated acutely stained with hematoxylin–eosin and with Fat-Red-O- for the with the nontoxic Ro 66-0074 enabled us to define groups of visualization of lipids. genes that allowed differentiation of two structurally and pharmacologically closely related compounds with different Gene Expression Analysis toxicological profiles. The gene expression profiles obtained Total RNA was extracted from the livers of all animals not only allowed us to clearly distinguish both compounds following standard laboratory protocols. Briefly, total RNA according to their toxicity before hepatotoxicity was was extracted from approximately 150 mg of liver tissue apparent by conventional end points, but also allowed us using RNAzol (Tel-Tex Inc., TX, USA) and a commercially to identify some genes that might be directly or indirectly available kit (Bio 101, CA, USA). Total RNA was purified involved in the underlying mechanisms of toxicity. The using RNeasy columns (Qiagen, Hilden, Germany) before results presented in this paper clearly show the power of quantification and assessment of ribosomal RNA integrity toxicogenomics as a tool for generating characteristic on agarose gels. fingerprints at early time-points as well as a tool to generate knowledge-based mechanistic hypotheses. Once validated, GeneChip Analysis toxicogenomics could significantly shorten the duration of Double-stranded cDNA was synthesized from 20 mg of total pilot toxicity studies, thus reducing the amount of com- RNA using a commercially available cDNA-synthesis kit pound needed, and limiting the cost. It will also indicate with an oligo dT-T7 promoter primer (Roche Molecular

www.nature.com/tpj Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 332

Biochemicals, Mannheim, Germany). The obtained cDNA was used as a template for in vitro transcription using the Megascript kit purchased from Ambion (TX, USA) supple- mented with biotinylated nucleotides (Bio-11-CTP and Bio- 16-UTP) provided by Roche Molecular Biochemicals (Man- nheim, Germany). Fragmented in vitro transcripts (cRNAs) were hybridized overnight onto commercially available rat microarrays containing 8799 rat-specific probe sets (RG- U34A, Affymetrix, CA, USA). The hybridized samples were stained with streptavidin-R-phycoerythrin (SAPE, Molecular Probes, CA, USA) and the signal amplified using a biotinylated goat anti-streptavidin antibody (Vector Laboratories, CA, USA) followed by a final staining with SAPE. Washing, staining, and amplification were carried out in a fluidics station provided by Affymetrix. Microarrays were scanned in an Affymetrix GeneArray scanner (gain setting: 18 000). The obtained image files were analyzed with the software Microarray Suite 5.0 (Affymetrix), while pairwise compar- isons between treated and control group and statistical analysis were performed with in-house developed software (RACE-A, Roche, Basel, Switzerland).

RT-PCR After digestion of genomic DNA using ‘Absolutely RNA’- Miniprep kit (Stratagene Europe, Amsterdam, Netherlands), 10 mg of total RNA was used for synthesis of single-stranded cDNA using a commercially available cDNA-synthesis kit with an oligo dT-T7 promoter primer (Superscript II, Gibco BRL Paisley, Scotland). The obtained cDNA was used as a template for several PCR reactions using appropriate specific primers (Table 6) and the SYBR Green PCR Master Mix (Applied Biosystems, Rotkreuz, Switzerland). The produc- tion of double-stranded amplification products binding SYBR Green was measured using the ABI-PRISM 7700 sequence detection system (Applied Biosystems). The speci- ficity of the amplified products was assessed by recording the melting curve and by evaluation of the product (size and number of bands) on agarose gels.

Western Blot Analysis Protein amount in hepatic extracts was quantified using Bio Rad protein assay. 10 mg of total liver protein was denatured in load buffer and separated using SDS-PAGE on a Novex gradient gel (4–20% Tris-glycine, Invitrogen, Basel, Switzer- land). After migration, proteins were transferred onto nitrocellulose membranes. Effectiveness of the transfer was assessed by staining the nitrocellulose membrane with Ponceau S red and by staining the gels with Simply Blue Safestain. Membranes were incubated with a specific first antibody (either anti-cytochrome P450 2B1 or anti-cytochrome P450 2C11, raised in goat, purchased at GenTest, Massachusetts, USA) diluted 1 : 1000. Incubation with the

second antibody (donkey anti-sheep/goat immunoglobulin 450 CYP3A1450 CYP2B2450 CYP2B exon 9 D13912 L00320 RNUD2A10, M35086, J05482 CTCTGCCACCACCTGAAACAA GGAACCCAATGACATCCAGAGA CTTTCCTTTGTCCTGCATTCCC CAACCCTTGATGACCGCAGTA TCAATGCTGCCCTTGTTCTCC CGCCAAGACAAATGTGCTTTC horseradish peroxidase conjugated, from Chemikon), di- P P P luted 1 : 1000, was subsequently performed under the Sequences of the primers used for QRT-PCR same incubation conditions. Chemoluminiscence was quantified by densitometry in a Multimage Light Cabinet DescriptionAryl hydrocarbon receptor (AhR)Alzheimer’s disease amyloid a4btg2 protein protein (homolog) (ngf-inducible antiproliferativeCarboxylesterase protein PC3).Cytochrome Cytochrome M60921Cytochrome Glutathione reductase X07648UDP-glucuronosyltransferase 2b1 AF082124 TTGGCCTAGCCAAGGTAAAAGG Genebank ACACATGGCCAGAGTTGAAGCC acc. no. ATAGCCCACCCTCCAAAAACG RNUD2A10, M35086, TCTTGAATCTCCTCAGCCACGG J05482 AB010635 AAGTCAATGGCATGTTCGCC CTCTGCCACCACCTGAAACAA U73174 Forward primer TGCTCGGACTCTGAAACATGG GGAACCCAATGACATCCAGAGA CAACATGCACCCAGCTATTTCA AGTCTTGGTCCAGAACTGCAGC CATGATCACGTGGATTACGGC Reverse primer GAACCCATCACTGGTTATCCCC (Alpha Inotech Corporation, San Leandro, CA, USA) using Table 6

The Pharmacogenomics Journal Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 333

Lumilight (Roche Diagnostics AG, Rotkreuz, Switzerland) 5 Sleight AJ, Monsma Jr FJ, Borroni E, Austin RH, Bourson A. Effects of solution. altered 5-ht6 expression in the rat: functional studies using antisense oligonucleotides. Behav Brain Res 1996; 73: 245–248. 6 Bourson A, Borroni E, Austin RH, Monsma Jr FJ, Sleight AJ. Determina- Statistical Data Analysis tion of the role of the 5-ht6 receptor in the rat brain: a study All data are typically expressed as mean values with their using antisense oligonucleotides. J Pharmacol Exp Ther 1995; 274: respective SD. For each measured parameter, the statistically 173–180. significant differences with respect to the controls were 7 Woolley ML, Bentley JC, Sleight AJ, Marsden CA, Fone KC. A role for 5- calculated using the two-tailed, unpaired Student’s t-test ht6 receptors in retention of spatial learning in the Morris water maze. Neuropharmacology 2001; 41: 210–219. (Pr0.05). For the evaluation of gene expression data using 8 Pennie WD, Kimber I. Toxicogenomics; transcript profiling and potential microarrays, output files from the Affymetrix software (MAS application to chemical allergy. Toxicol In Vitro 2002; 16: 319–326. 5.0) were further analyzed using in-house developed 9 Simmons PT, Portier CJ. Toxicogenomics: the new frontier in risk analysis tools (RACE-A, Roche, Basle, Switzerland). The analysis. Carcinogenesis 2002; 23: 903–905. 10 Marchant GE. Toxicogenomics and toxic torts. Trends Biotechnol 2002; expression level for each gene is expressed as the mean 20: 329–332. fluorescence intensity (average difference) and SD of five 11 Novatchkova M, Eisenhaber F. Can molecular mechanisms of biological replicates after Nalimov outlier removal (99% confidence). processes be extracted from expression profiles? Case study: endothelial Cluster analysis was performed using the correlation func- contribution to tumor-induced angiogenesis. BioEssays 2001; 23: 1159– 1175. tion to determine the distance matrix and the average 12 de Longueville F, Surry D, Meneses-Lorente G, Bertholet V, Talbot V, linkage method. Gene expression measured by RT-PCR was Evrard S et al. Gene expression profiling of drug metabolism and evaluated using mean CT values of triplicates, normalized toxicology markers using a low-density DNA microarray. Biochem against the CT values for the housekeeping gene GAPDH. Pharmacol 2002; 64: 137–149. 13 Waring JF, Ciurlionis R, Jolly RA, Heindel M, Ulrich RG. Microarray Results are expressed as mean fold changes with respect to analysis of hepatotoxins in vitro reveals a correlation between gene the time-matched control group. The threshold for gene expression profiles and mechanisms of toxicity. Toxicol Lett 2001; 120: induction considered relevant was usually set as 100% up-or 359–368. downregulation (2- or 0.5-fold change). Protein levels 14 Bulera SJ, Eddy SM, Ferguson E, Jatkoe TA, Reindel JF, Bleavins MR et al. RNA expression in the early characterization of hepatotoxicants in evaluated by Western blots were measured semiquantita- Wistar rats by high-density DNA microarrays. Hepatology 2001; 33: tively by densitometry and are expressed in arbitrary units; 1239–1258. the values are given as the mean of two representative 15 Hamadeh HK, Bushel PR, Jayadev S, DiSorbo O, Bennett L, Li L et al. animals per dose group. Prediction of compound signature using high density gene expression profiling. Toxicol Sci 2002; 67: 232–240. 16 Fountoulakis M, de Vera M, Crameri F, Boess F, Gasser R, Albertini S et ACKNOWLEDGEMENTS al. Modulation of gene and protein expression by carbon tetrachloride We would like to thank Ms Simona Romer for her excellent in the rat liver. Toxicol Appl Pharmacol 2002; 183:71. technical assistance. Special thanks go to Dr Rodolfo Gasser, Prof. 17 Lian Z, Kluger Y, Greenbaum DS, Tuck D, Gerstein M, Berliner N et al. Ulrich Certa, and Dr Franziska Boess for the many fruitful scientific Genomic and proteomic analysis of the myeloid differentiation discussions and their constant support, as well as for their critical program: global analysis of gene expression during induced differentia- reading of the manuscript. We are also deeply indebted to Dr Martin tion in the MPRO cell line. Blood 2002; 100: 3209–3220. 18 Whysner J, Ross PM, Williams GM. Phenobarbital mechanistic data and Neeb, Detlef Wolf, Clemens Broger, and the Bioinformatics group at risk assessment: enzyme induction, enhanced cell proliferation, and F Hoffmann-La Roche AG for providing the tools and the assistance tumor promotion. Pharmacol Ther 1996; 71: 153–191. that made possible the data analysis. 19 Omiecinski CJ, Hassett C, Costa P. Developmental expression and in situ localization of the phenobarbital-inducible rat hepatic mRNAs for cytochromes CYP2B1, CYP2B2, CYP2C6, and CYP3A1. Mol Pharmacol ABBREVIATIONS 1990; 38: 462–470. ALT alanine aminotransferase 20 Hoen PA, Bijsterbosch MK, van Berkel TJ, Vermeulen NP, Commandeur AST aspartate aminotransferase JN. Midazolam is a phenobarbital-like cytochrome P450 inducer in rats. c-GT c-glutathione transferase J Pharmacol Exp Ther 2001; 299: 921–927. 5-HT serotonin 21 Waxman DJ, Walsh C. Cytochrome P-450 isozyme 1 from phenobarbi- LDH lactate dehydrogenase tal-induced rat liver: purification, characterization, and interactions 5-ND 50-nucleotidase with metyrapone and cytochrome b5. Biochemistry 1983; 22: 4846–4855. 22 Nakayama K, Suwa Y, Mizukami Y, Sogawa K, Fujii-Kuriyama Y. Cloning and sequencing of a novel rat cytochrome P450 2B-encoding gene. Gene 1993; 136: 333–336. REFERENCES 23 Suwa Y, Mizukami Y, Sogawa K, Fujii-Kuriyama Y. Gene structure of a 1 Mossner R, Schmitt A, Syagailo Y, Gerlach M, Riederer P, Lesch KP. The major form of phenobarbital-inducible cytochrome P- 450 in rat liver. serotonin transporter in Alzheimer’s and Parkinson’s disease. J Neural J Biol Chem 1985; 260: 7980–7984. Transm Suppl 2000; 60: 345–350. 24 Agrawal AK, Shapiro BH. Phenobarbital induction of hepatic CYP2B1 2 Saudou F, Hen R. 5-Hydroxytryptamine receptor subtypes in vertebrates and CYP2B2: pretranscriptional and post-transcriptional effects of and invertebrates. Neurochem Int 1994; 25: 503–532. gender, adult age, and phenobarbital dose. Mol Pharmacol 1996; 49: 3 Ruat M, Traiffort E, Arrang JM, Tardivel-Lacombe J, Diaz J, Leurs R et al. A 523–531. novel rat serotonin (5-HT6) receptor: molecular cloning, localization 25 Kawamoto T, Kakizaki S, Yoshinari K, Negishi M. Estrogen activation of and stimulation of cAMP accumulation. Biochem Biophys Res Commun the nuclear orphan receptor CAR (constitutive active receptor) in 1993; 193: 268–276. induction of the mouse Cyp2b10 gene. Mol Endocrinol 2000; 14: 4 Yoshioka M, Matsumoto M, Togashi H, Mori K, Saito H. Central 1897–1905. distribution and function of 5-HT6 receptor subtype in the rat brain. Life 26 George DK, Crawford DH. Antibacterial-induced hepatotoxicity. In- Sci 1998; 62: 1473–1477. cidence, prevention and management. Drug Saf 1996; 15: 79–85.

www.nature.com/tpj Effect of 5-HT6 receptor antagonists on the rat liver L Suter et al 334

27 Ourlin JC, Handschin C, Kaufmann M, Meyer UA. A link between 31 Honkakoski P, Negishi M. Regulation of cytochrome P450 (CYP) genes cholesterol levels and phenobarbital induction of cytochromes P450. by nuclear receptors. Biochem J 2000; 347(Part 2): 321–337. Biochem Biophys Res Commun 2002; 291: 378–384. 32 Leclercq I, Horsmans Y, Desager JP, Delzenne N, Geubel AP. Reduction 28 Kocarek TA, Kraniak JM, Reddy AB. Regulation of rat hepatic cytochrome in hepatic cytochrome P-450 is correlated to the degree of liver fat P450 expression by sterol biosynthesis inhibition: inhibitors of squalene content in animal models of steatosis in the absence of inflammation. synthase are potent inducers of CYP2B expression in primary cultured rat J Hepatol 1998; 28: 410–416. hepatocytes and rat liver. Mol Pharmacol 1998; 54: 474–484. 33 Akiyama TE, Ward JM, Gonzalez FJ. Regulation of the liver fatty acid- 29 Yamada H, Matsuki Y, Yamaguchi T, Oguri K. Effect of a ligand selective binding protein gene by hepatocyte nuclear factor 1alpha (HNF1alpha). for peripheral benzodiazepine receptors on the expression of rat hepatic P-450 cytochromes: assessment of the effect in vivo and in a hepatocyte Alterations in fatty acid homeostasis in HNF1alpha-deficient mice. J Biol culture system. Drug Metab Dispos 1999; 27: 1242–1247. Chem 2000; 275: 27117–27122. 30 Guo GL, Lambert G, Negishi M, Ward JM, Brewer Jr HB, Kliewer SA et al. 34 Freneaux E, Labbe G, Letteron P, The Le D, Degott C, Geneve J et al. Complementary roles of Farnesoid X receptor, pregnane X receptor, Inhibition of the mitochondrial oxidation of fatty acids by tetracycline in and constitutive Androstane receptor in protection against bile acid mice and in man: possible role in microvesicular steatosis induced by toxicity. J Biol Chem 2003. this antibiotic. Hepatology 1988; 8: 1056–1062.

The Pharmacogenomics Journal