Abstract
The constitutive androstane receptor (CAR) and pregnane X receptor (PXR) mediate the expression of mammalian cytochrome P450 (P450) 2B genes, including CYP2B6 in humans. Large interindividual differences exist in hepatic CYP2B6 expression, but the molecular basis for this variability is not well understood. In the present study, we developed real-time polymerase chain reaction methods to measure CYP2B6, CAR, and PXR mRNA expression and compared the levels in a panel of 12 individual human liver samples. The transcripts of CAR and CYP2B6 were present in all the samples analyzed, whereas those of PXR were detectable in all but one sample. A striking finding was the 240-fold interindividual variability in hepatic CAR mRNA levels, which was similar to the variability (278-fold) in CYP2B6 mRNA levels but greater than the 27-fold variability in PXR mRNA expression. Additional analysis revealed positive and statistically significant correlations between the mRNA levels of CAR and CYP2B6 (r2 = 0.63, p = 0.002), PXR and CYP2B6 (r2 = 0.75. p < 0.001), and CAR and PXR (r2 = 0.86,p < 0.001). In summary, substantial interindividual differences exist in hepatic CAR and, to a lesser extent, PXR gene expression. The variability in the abundance of these transcription factors may contribute to the large interindividual differences in CYP2B6 gene expression in human liver.
Nuclear receptors (NR1) are transcription factors that regulate the expression of genes involved in a broad range of biological processes, such as differentiation, metabolism, and reproduction (Aranda and Pascual, 2001). Among the members in the NR1I subfamily are the constitutive androstane receptor (CAR; NR1I3) (Baes et al., 1994) and the pregnane X receptor (PXR; NR1I2) (Kliewer et al., 1998). PXR has also been termed the steroid and xenobiotic receptor (Blumberg et al., 1998) and pregnane-activated receptor (Bertilsson et al., 1998).
Seminal studies by Negishi and coworkers led to the discovery of CAR as a key regulator in the expression of rodent and human cytochrome P450 (P450) 2B genes (Sueyoshi and Negishi, 2001). The experimental evidence obtained to date indicates that induction of these genes is a receptor-mediated event in which the presence of a CYP2B inducer triggers nuclear translocation of CAR, possibly via receptor dephosphorylation. Subsequently, CAR forms a heterodimer with retinoid X receptor α (RXRα). The binding of the CAR-RXRα heterodimer to the NR-binding sites (direct repeat-4 motifs) in the 5′-flanking sequence of CYP2B genes results in the activation of a 51-base pair phenobarbital-responsive enhancer module. Recently, experiments with primary cultures of human hepatocytes have shown that PXR also mediates CYP2B6 expression as a consequence of its recognition of the phenobarbital-responsive enhancer module (Goodwin et al., 2001). Similar to CAR, PXR forms a heterodimer with RXRα, but in contrast to CAR, PXR is constitutively inactive and is present only in the nucleus (Goodwin et al., 2002).
CYP2B6 catalyzes the biotransformation of clinically useful drugs, including cyclophosphamide (Chang et al., 1993). In addition, it metabolizes methoxychlor, a broad-spectrum pesticide, to its estrogenic metabolites (Dehal and Kupfer, 1994) and activates promutagens such as aflatoxin B1 and the tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (Code et al., 1997). CYP2B6 is subject to induction by various drugs, including phenobarbital, rifampin, and dexamethasone, as demonstrated in experiments with primary cultures of human hepatocytes (Chang et al., 1997). A clinically relevant finding is the considerable interindividual variability in hepatic CYP2B6 expression. This has been shown at the level of mRNA (Yamano et al., 1989; Rodriguez-Antona et al., 2001), protein (Code et al., 1997; Ekins et al., 1998; Yang et al., 1998; Lang et al., 2001), and catalytic activity (Ekins et al., 1997, 1998). Such differences may contribute to interindividual variability in response to drugs and susceptibility to xenobiotic toxicity.
The present study was conducted to determine whether the interindividual variability in CYP2B6 gene expression is associated with a similar degree of variability in the levels of CAR and PXR, two receptors that are known to mediate CYP2B6 expression (Goodwin et al., 2001; Sueyoshi and Negishi, 2001). We developed real-time, rapid-cycle polymerase chain reaction (PCR) methods to measure CYP2B6, CAR, and PXR mRNA expression and compared the levels of CYP2B6 to those of CAR and PXR in a panel of 12 individual human liver samples.
Materials and Methods
Chemicals and Reagents.
TriZol, Superscript II reverse transcriptase, Platinum TaqDNA polymerase, oligo(dT)12–16 primer, dithiothreitol, deoxynucleoside-5′-triphosphate mix, magnesium chloride, and deoxyribonuclease I were purchased from Invitrogen Canada Inc. (Burlington, Ontario, Canada). SYBR Green I and bovine serum albumin were bought from Sigma-Aldrich (St. Louis, MO). RiboGreen RNA Quantitation kit and PicoGreen dsDNA Quantitation kit were purchased from Molecular Probes, Inc. (Eugene, OR). Forward and reverse PCR primers were synthesized and reverse-phase purified at the Nucleic Acid and Protein Service Unit, University of British Columbia (Vancouver, British Columbia, Canada).
Source of Human Liver Samples.
Liver tissue samples were provided by Dr. James R. Olson (Department of Pharmacology and Toxicology, State University of New York, Buffalo, NY). The information on the donors has been described elsewhere (Chang et al., 2003).
Isolation and Quantification of Total RNA.
Total liver RNA was isolated using TriZol. RNA concentration was quantified using the RiboGreen RNA Quantitation kit (Molecular Probes, Inc.), according to the manufacturer's protocol.
Reverse Transcription and Quantification of Total cDNA.
RNA was transcribed using SuperScript II reverse transcriptase. Concentrations of the synthesized cDNA samples were quantified using the PicoGreen dsDNA Quantitation kit (Molecular Probes, Inc.), according to the manufacturer's protocol.
Design of PCR Primers.
Sequences for the forward (5′-CCA-GCT-CAT-CTG-TTC-ATC-CA-3′) and reverse (5′-GGT-AAC-TCC-AGG-TCG-GTC-AG-3′) primers for CAR (GenBank accession no. Z30425; Baes et al., 1994), forward (5′-CAA-GCG-GAA-GAA-AAG-TGA-ACG-3′) and reverse (5′-CAC-AGA-TCT-TTC-CGG-ACC-TG-3′) primers for PXR (GenBank accession no. AF061056, Lehmann et al., 1998), and forward (5′-GCG-TGT-GGT-TCA-TTC-ACA-AA-3′) and reverse (5′-AAT-TTA-GCC-AGG-CGT-GGT-G-3′) primers for CYP2B6 (GenBank accession no. M29874; Yamano et al., 1989) were designed using the Primer3 software program (version 0.2, www-genome.wi.mit.edu). The CAR primers were designed to amplify CAR1 (Pascussi et al., 2000b) and not the human homolog of the spliced mouse CAR2 isoform (Choi et al., 1997). The PXR primers were designed to amplify PXR.1 and not PXR.2, which is a splice variant of PXR.1 and lacks amino acid residues 174–210 in the putative ligand-binding domain (Dotzlaw et al., 1999). The CYP2B6 primers were designed to amplify CYP2B6 but not the CYP2B7 pseudogene (Yamano et al., 1989). However, they will amplify the CYP2B6 alleles identified to date (Lang et al., 2001).
Real-Time PCR Analysis.
CAR, PXR, and CYP2B6 cDNA samples were amplified by a real-time DNA thermal cycler (LightCycler; Roche Diagnostics, Laval, Quebec, Canada). Each 20-μl reaction mixture contained 0.2 unit PlatinumTaq DNA polymerase in 1× PCR reaction buffer [20 mM Tris-HCl (pH 8.4) and 50 mM KCl], 2 mM magnesium chloride (except for CYP2B6 in which the concentration was 4 mM), 1 ng cDNA, 200 μM deoxynucleoside-5′-triphosphate mix, 0.2 μM of forward and reverse primers, 0.25 mg/ml bovine serum albumin, and 2 μl of a 3.3× SYBR Green I solution. The PCR conditions were as follows: initial denaturation was at 95°C for 5 min followed by cycles of denaturation at 95°C (5 s), annealing at 65°C (25 s for CAR and 15 s for PXR and CYP2B6), and extension at 72°C (10 s for CAR and PXR and 5 s for CYP2B6). Fluorescence readings were recorded at a temperature several degrees less than the melting temperature of the amplicon (85°C for CAR, 86°C for CYP2B6, and 87°C for PXR). Calibration curves were constructed by plotting the cross point against known amounts of the amplicon, which was quantified using the PicoGreen dsDNA Quantitation kit (Molecular Probes, Inc.).
Results and Discussion
CYP2B6 (Fig. 1A) and CAR (Fig. 1B) mRNA were expressed in each of the 12 human liver samples analyzed, whereas PXR mRNA (Fig. 1C) was detected in 11 of these samples. A novel finding is the large interindividual differences (240-fold) in CAR mRNA expression (Fig. 1B), which was similar to the variability (278-fold) in CYP2B6 mRNA levels, but greater than the variability (27-fold) in PXR mRNA expression (Fig. 1C). There were two liver samples that had undetectable or very low levels of CYP2B6, CAR, or PXR gene expression, but this was unlikely due to RNA degradation because sample 7, which had no detectable mRNA levels of PXR (Fig. 1C), was found to express CYP2B6 (Fig. 1A) and CAR (Fig. 1B) in addition to CYP1B1 mRNA (Chang et al., 2003). Similarly, sample 6, which had the lowest level of CYP2B6 mRNA (Fig. 1A), was found to have quantifiable levels of CAR (Fig. 1B) and PXR (Fig. 1C) as well as CYP1B1 and CYP1A2 mRNA (Chang et al., 2003).
Real-time PCR analysis of CYP2B6, CAR, and PXR gene expression in human liver.
Total RNA was isolated from a panel of 12 individual liver samples and reverse transcription was performed. CYP2B6 (panel A), CAR (panel B), and PXR cDNA (panel C) were amplified in duplicates by real-time PCR as described under Materials and Methods. Results are expressed as relative mRNA expression (the levels were normalized to the sample that had the lowest quantifiable level, which was assigned a nominal value of 1).
A positive and statistically significant correlation (r2 = 0.63, p = 0.002) was obtained between CYP2B6 and CAR mRNA levels (Fig.2A). CYP2B6 and PXR mRNA levels were also highly correlated (r2 = 0.75,p < 0.001). The strongest correlation (r2 = 0.86, p < 0.001) was obtained between CAR and PXR gene expression (Fig. 2C). Only one other study has performed correlational analysis on the expression of CAR, PXR, and a P450 gene in human liver samples (Pascussi et al., 2001). In that study, the mRNA levels of CYP3A4 were highly correlated with those of CAR (r2 = 0.89) and PXR (r2 = 0.68).
Correlational analyses of human hepatic CYP2B6, CAR, and PXR mRNA levels.
Shown are correlational analysis of CYP2B6 and CAR mRNA levels (panel A), CYP2B6 and PXR mRNA levels (panel B), and PXR and CAR mRNA levels (panel C) in a panel of 12 individual human liver samples. The data are from Fig. 1, A to C.
The basis for the observed interindividual variability in hepatic CAR and PXR mRNA levels is not known. The expression of these receptors is subject to modulation by drugs and other factors. For example, treatment of primary cultures of human hepatocytes with low micromolar concentrations of dexamethasone, prednisolone, or hydrocortisone transcriptionally increases CAR (Pascussi et al., 2000b) and PXR (Pascussi et al., 2000a) expression, suggesting a role for the glucocorticoid receptor. In contrast, the addition of interleukin-6 to primary cultures of human hepatocytes results in down-regulation of both CAR and PXR mRNA expression (Pascussi et al., 2000c). Similarly, bacterial lipopolysaccharide-induced acute inflammation in mice results in a decrease in the expression of these two genes (Beigneux et al., 2002). Currently, it is not known if genetic factors play a role in the interindividual variability in CAR and PXR gene expression. Allelic variants of CAR (Pascussi et al., 2000b) and PXR (Zhang et al., 2001) have been identified, but specific mutations associated with altered gene expression have not been reported.
Interindividual differences in CYP2B6 expression are usually attributed to environmental factors, such as enzyme induction by drugs and other xenobiotics (Ekins and Wrighton, 1999). However, genetic factors may also play a role. For example, individuals with the CYP2B6*5 or CYP2B6*7 allele have reduced hepatic microsomal CYP2B6 protein content and decreased CYP2B6-mediated S-mephenytoinN-demethylase activity (Lang et al., 2001). In the present study, significant correlations existed between CAR and CYP2B6 and between PXR and CYP2B6 mRNA levels, suggesting that the interindividual differences in CYP2B6 gene expression may also be influenced by the interindividual variability in the abundance of the transcription factors involved in the regulation of CYP2B6 expression.
In summary, the major findings from the present study are as follows: 1) large interindividual differences (240-fold) in CAR mRNA levels were found in a panel (N = 12) of human liver samples; 2) the extent of the variability was similar to that obtained for CYP2B6 mRNA (278-fold), but considerably greater than that for PXR mRNA (27-fold); and 3) positive and statistically significant correlations were obtained between CAR and CYP2B6, PXR and CYP2B6, and CAR and PXR gene expression.
Acknowledgments
The authors thank Dr. James R. Olson (State University of New York, Buffalo, NY) for the generous provision of the human liver samples.
Footnotes
-
This research was supported by Grant MOP-42385 (to T.K.H.C.) from the Canadian Institutes of Health Research (CIHR) and a major equipment grant (to T.K.H.C. and S.M.B.) from the Dawson Endowment Fund in Pharmaceutical Sciences. T.K.H.C. received a Research Career Award in the Health Sciences from CIHR and Rx&D Health Research Foundation.
- Abbreviations used are::
- NR
- nuclear receptors
- CAR
- constitutive androstane receptor
- PXR
- pregnane X receptor
- P450
- cytochrome P450
- RXRα
- retinoid X receptor α
- PCR
- polymerase chain reaction
- Received August 15, 2002.
- Accepted September 30, 2002.
- The American Society for Pharmacology and Experimental Therapeutics