Abstract
Some members of the CYP3A subfamily show gender-dependent expression. Using quantitative real-time polymerase chain reaction, we report that female rats have 28-fold higher CYP3A9 mRNA levels than males in liver and 3.8-fold higher in lung. Furthermore, the CYP3A9 expression profile in kidney exhibits a regio-specific distribution, i.e., a 10-fold higher expression in cortex compared with medulla. Also, we observed tissue-specific estrogen regulation of the CYP3A9 message. Estrogen treatment caused a significant up-regulation in liver and a marked down-regulation both in the cortex and medulla of the kidney. Upon ovariectomy, hepatic and brain CYP3A9 expression were reduced significantly, but a modest increase in kidney expression was observed. The effects of ovariectomy on CYP3A9 gene expression were reversed upon exogenous estrogen treatment. CYP3A protein levels and hepatic microsomal activity toward benzphetamine after various treatments showed changes parallel to CYP3A9 mRNA levels. We report for the first time that CYP3A9 levels change dramatically during the course of pregnancy.
Cytochromes P450 (P450) comprise a superfamily of microsomal heme proteins that catalyze the oxidative metabolism of various endobiotics and xenobiotics (Wrighton et al., 1990). The CYP3A subfamily, the most abundantly expressed among the P450 superfamily, is responsible for aspects of the metabolism of endogenous steroids such as testosterone, progesterone, lithocholic acid, and for the primary metabolism of more than 50% of clinically utilized drugs (Gillam et al., 1993; Yamazaki and Shimada, 1997; Wang et al., 2000). Moreover, CYP3As play an important role in presystemic elimination of drugs and provide a mechanism for clinically observed drug-drug interactions (Moore et al., 2000).
In rats, the CYP3A subfamily consists of five related genes, CYP3A1, CYP3A2, CYP3A9, CYP3A18, and CYP3A23. Although these isoforms are predominantly expressed in liver (Komori and Oda, 1994), extrahepatic expression in brain (Wang et al., 1996), intestine (Watkins, 1992), kidneys (Debri et al., 1995), leukocytes (Mahnke et al., 1996), and lungs has also been reported. In addition, CYP3As in rats show sex-, tissue-, species-, and age-dependent expression patterns (Waxman et al., 1995). For instance, CYP3A2 expression is observed only in neonatal female rats, while it is demonstrable throughout the life-cycle of male rats (Gonzalez et al., 1986). Likewise, CYP3A1 and CYP3A18 have been shown to be male-dominant isoforms (Strotkamp et al., 1995).
Using Northern blot analysis, our laboratory previously reported that hepatic CYP3A9 expression was 10-fold higher in females than in male rats (Wang and Strobel, 1997). Subsequently, several other studies, using RTPCR, reported a 2- to 6-fold higher expression in female rats (Mahnke et al., 1997; Robertson et al., 1998). These mRNA expression profiles would classify CYP3A9 distribution as female-dominant rather than female-specific according to the definition of Kato and Yamazoe (1993). To clarify this discrepancy, we developed a quantitative real-time PCR (QRTPCR) assay that would enable a more precise quantitation of CYP3A9 levels. QRTPCR has been used recently to distinguish among various closely related P450 isoforms (Burczynski et al., 2001; Kalsotra et al., 2002).
In this study we defined the extrahepatic expression profile of CYP3A9 and determined more precisely the difference between male and female CYP3A9 gene expression. The regulation of CYP3A9 expression by progesterone (PROG) and estrogen (E2) was also evaluated. The catalytic capacity of microsomes from control, ovariectomized (OVEX), and OVEX plus E2-treated rats was analyzed to examine further the correlation between CYP3A9 expression and CYP3A9 function. Finally, to investigate whether CYP3A9 showed altered levels in pregnant rats, the CYP3A9 mRNA expression pattern was evaluated at various stages of pregnancy.
Materials and Methods
Materials.
Estradiol benzoate (E2) and PROG were obtained from Sigma-Aldrich (St. Louis, MO). Anti-CYP3A2 antibody was purchased from Research Diagnostics Inc. (Flanders, NJ). Benzphetamine was a gift from The Upjohn Company (Kalamazoo, MI).
Animal Treatment.
Sprague-Dawley rats of either sex (175–200 g body weight) were purchased from Harlan Laboratories (Indianapolis, IN). Animals were allowed food and water ad libitum and were subjected to a 12-h light/dark cycle. To evaluate the role of E2 on CYP3A9 expression, female rats were ovariectomized, allowed to recover for 4 days, and treated with E2 (250 μg/kg), PROG (50 mg/kg), E2 and PROG, or sesame oil s.c. for 2 days. To investigate the dose-response relation of CYP3A9 expression to E2, female rats were treated with E2 250 μg or 500 μg/kg for one week. To determine the CYP3A9 expression pattern during pregnancy, day 13-, day 19-, and day 21-pregnant rats were analyzed for CYP3A9 mRNA expression. At the end of the experiments all rats, including controls, were sacrificed, tissues excised, immediately frozen in liquid nitrogen, and stored at −80°C until used for analysis.
RNA Isolation and Northern Blotting.
Frozen tissues were thawed on ice, and total RNA was isolated using RNA STAT-60 (Tel Test, Inc., Friendswood, TX) according to the manufacturer's instructions. All samples were DNase-treated using RQ1 DNase (Promega, Madison, WI). The quality of the isolated RNA was assessed by electrophoresis on a 1% agarose gel based on the integrity of 28S and 18S bands after ethidium bromide staining. Thirty-five μg of total RNA for each of the samples were denatured and electrophoresed on a 1% agarose gel containing 2.2 M formaldehyde. RNA was transferred onto a Zeta probe nitrocellulose membrane (Bio-Rad, Hercules, CA), prehybridized overnight, and hybridized with 32P-labeled DNA probe specific for CYP3A9. A 302-bp NotI-NcoI fragment representing the 5′-end of the CYP3A9 was used as the Northern probe (Wang and Strobel, 1997). The Northern blot was hybridized at 42°C and washed at 65°C. After washing the membranes twice for 15 min with 0.2× SSC and once for 15 min with 0.2× SSC and 0.1% SDS solution, they were exposed to X-ray film.
Design of Real-Time Quantitative PCR Primers and Probes.
PCR primers and fluorescent probe sequences were designed at the 3′-untranslated region (region of maximal specificity) of the CYP3A genes using Primer Express software (Applied Biosystems, Foster City, CA) and custom synthesized by IDT DNA Technology, Inc. (Coralville, IA). The sequences of primers and probes for CYP3A9 and rat cyclophilin are shown in Table 1. The specificity of primers and probes was confirmed by NCBI BLAST and by sequencing the amplified products.
Reverse Transcription and Quantitative Polymerase Chain Reaction.
Aliquots (200 ng) of total RNA to be analyzed were reverse transcribed in quadruplicate [including an RT (−) blank to account for amplification of contaminating genomic DNA] for each sample with 1× SSII buffer, 200 nM reverse primer, 500 μM dNTPs, and 10 U/10 μl Superscript II (Invitrogen, Carlsbad, CA) at 50°C for 30 min, followed by 72°C for 10 min. Forty microliters of PCR mix (containing 1× PCR buffer, 200 nM forward primer, 200 nM reverse primer, 4 mM MgCl2, 1.25 U/50 μl Taqpolymerase, and 150 nM fluorogenic probe) were added to 10 μl of the RT reaction. Amplification was performed using an ABI Prism 7700 (Applied Biosystems) at 95°C for 1 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Standard curves were generated by plotting Ct versus the log of the amount of amplicon (custom-made from IDT Inc.) specific for CYP3A9 (20 attograms–2 pg), and were used to compare the relative amount of CYP3A9 mRNA in the samples. Data were analyzed by the use of the Sequence Detection Application and relative values of mRNA also were generated by normalizing copy number values of CYP3A9 to the copy number values of rat cyclophilin.
Preparation of Microsomes.
Microsomes from different tissues were prepared as previously described by Saito and Strobel (1981). Briefly, tissues were homogenized (20 strokes) in 6 volumes of potassium phosphate buffer (pH 7.4) containing 20 mM KPi, 0.25 M sucrose, 1 mM EDTA, and a cocktail of protease inhibitors (1 mM PMSF, 1 μg/ml leupeptin, and 0.7 μg/ml pepstatin). Each homogenate was centrifuged at 8500 rpm for 20 min and the supernatant fraction was collected. The supernatant fraction was further centrifuged at 100,000g for 45 min; the pellet washed in fresh buffer, and again centrifuged at 100,000g for 45 min. The washed pellets were resuspended in the homogenate buffer and stored at −80°C until analysis.
Western Blotting.
The total protein concentration in microsomes was determined by BCA assay using bovine serum albumin (BSA) as the standard. Samples were boiled in Laemmli buffer and resolved on 4 to 15% gradient Tris-glycine sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS PAGE) gels. Proteins were transferred to nitrocellulose membranes using a semidry transfer apparatus. Membranes were blocked for 4 h with 5% dried nonfat milk in TTBS (0.05% Tween 20 in Tris-buffered saline) followed by an overnight incubation at 4°C in 1:1000 dilution of polyclonal rabbit anti-rat primary antibody raised against CYP3A2, which recognizes most CYP3A forms. Membranes were then washed and incubated at room temperature with HRP-conjugated goat anti-rabbit secondary antibody (1:1000 dilution) for 2 h. Immunoreactivity was detected using an HRP chemiluminescence system (Pierce, Rockford, IL).
Activity Assays.
The N-demethylation assay for benzphetamine was performed in a final volume of 2 ml with 100 mM potassium phosphate buffer (pH 7.7), containing an aliquot of microsomal protein, and 1 mM benzphetamine HCl. The reaction was initiated by the addition of NADPH to a final concentration of 1 mM (Sigma-Aldrich) and incubated at 37°C for 10 min and terminated by addition of 17.5% perchloric acid. The fluorimetric measurement of the formaldehyde product formation was carried out as described previously (Wang and Strobel, 1997).
Statistical Analysis.
Data are presented as mean ± S.E. Statistical significance for differences in CYP3A9 expression between male and female groups was determined using a two-tailed unpaired Student's t test. Multiple groups in the OVEX, hormonal, or pregnancy study were compared using one-way analysis of variance (ANOVA). Statistical differences were considered significant if P < 0.05.
Results
Tissue Distribution of CYP3A9.
QRTPCR analysis revealed 28-fold higher CYP3A9 expression in female than in male rat livers (Fig. 1a), thus making CYP3A9 female-specific according to Kato and Yamazoe (1993). CYP3A9 was predominantly expressed in liver followed by lung, which showed 3.8-fold higher expression in females (P < 0.05), while renal CYP3A9 levels were equally distributed in rats of either sex. CYP3A9 is also expressed in male and female brain. We can summarize the tissue distribution as liver ≫ lung > kidney ∼ brain (Table 2). Furthermore, within the kidney we observed a regio-selective distribution of CYP3A9 with a 10-fold higher expression in cortex compared with medulla (Fig. 1b).
Effects of E2 on CYP3A9 Expression.
Upon treatment of female rats with 250 μg/kg E2 we observed 20-fold induction of hepatic CYP3A9 mRNA expression compared with controls. Doubling the dose of E2 to 500 μg/kg did not increase CYP3A9 expression any further compared with low-dosed rats (Fig. 2). Conversely, 250 μg/kg E2 treatment led to a 10-fold decrease of CYP3A9 message in renal cortex. CYP3A9 levels in cortex could not be detected at a 500 μg/kg dose of estradiol. Medullary CYP3A9 expression in the kidney was completely eliminated even at the 250 μg/kg E2 treatment. Thus, we observed opposite effects on hepatic and renal CYP3A9 levels upon exogenous E2 treatments. In brain, no significant change in CYP3A9 message level was observed following E2treatment (Fig. 2).
Effects of Ovariectomy and Hormone Treatment.
To examine further the effects of E2 on CYP3A9 gene expression, animals were OVEX, and treated with PROG (50 mg/kg), E2 (250 μg/kg), or both. We observed a marked decrease in CYP3A9 mRNA levels in liver and brain of OVEX animals. In kidney, however, we found a modest increase in CYP3A9 expression after OVEX (Table 2). Exogenous E2 treatment reversed the OVEX effects in liver such that CYP3A9 levels were significantly higher than controls. E2 treatment, however, could not restore CYP3A9 expression in brain after OVEX. Again, renal CYP3A9 exhibited a dramatic down-regulation upon E2 treatment in OVEX animals. PROG, however, could partially restore CYP3A9 levels in liver. Brain CYP3A9 expression was not altered after PROG treatment. The combined dosing of E2 and PROG averaged the effects of individual treatments in all the tissues used in this study (Table 2).
Northern Hybridization Analysis.
To validate the QRTPCR results, we performed Northern blot analysis using rat liver RNA from various treatment groups. We observed similar patterns of expression as reported earlier using QRTPCR assays. The Northern blot shown in Fig.3 depicts diminished CYP3A9 mRNA expression in OVEX animals compared with controls, which can be returned to control levels by E2 treatment. PROG, however, does not show such a drastic increase in expression, and the dual hormone treatment appears to average their individual effects on CYP3A9 mRNA levels.
CYP3A9 Expression Pattern during Pregnancy.
To determine CYP3A9 mRNA levels during the course of pregnancy, we analyzed day 13-, day 19-, and day 21-pregnant rats. The data suggest that hepatic and brain CYP3A9 message levels vary as pregnancy progresses, reaching the highest level of expression during late pregnancy. Also, it is interesting to note that hepatic CYP3A9 mRNA levels significantly decreased during early stages of pregnancy when compared with nonpregnant controls. The CYP3A9 level of day 13-pregnant animals was 0.94 ± 0.34 (% R-cyclophilin) while the nonpregnant animals showed 67.5 ± 24.9 (%R-cyclophilin) values. On day 19, CYP3A9 expression in liver was increased 2-fold when compared with day 13. On day 21, CYP3A9 mRNA expression levels in liver showed a significant 28-fold increase when compared with day 13. On the contrary, renal CYP3A9 expression was decreased significantly on day 19 and day 21 (Fig.4). In brain, we observed a significant increase in CYP3A9 mRNA levels on day 21 when compared with day 19 (Fig. 4).
CYP3A Protein Levels after Hormone Treatments.
The data clearly show that E2 affects the regulation of CYP3A9 at the mRNA level. To examine whether the message translates to consequent fluctuations in CYP3A9 protein levels, we performed Western blot analysis using hepatic and renal microsomes. In the absence of a specific CYP3A9 antibody, we used a commercial polyclonal anti-CYP3A2 antibody, which cross-reacts with purified CYP3A9 protein. From the data shown in Fig. 5, it is evident that CYP3A protein levels in liver reduced significantly after OVEX and was restored upon E2 treatment. The degree of this change in protein levels corresponds to the change in mRNA expression and is consistent with the notion that in female rat liver, CYP3A9 constitutes a major portion of CYP3A protein expressed since all other rat CYP3A forms are male-dominant (Mahnke et al., 1997; Robertson et al., 1998). On the contrary, renal microsomes exhibited a significant increase in CYP3A protein levels upon OVEX, which was reduced to control amounts after E2 treatment. Thus, the protein changes observed were consistent with the changes observed in CYP3A9 mRNA levels.
CYP3A Catalytic Activity.
N-Demethylation activity toward benzphetamine was measured to assess functional endpoints of the changes observed in the CYP3A protein levels. Hepatic microsomes were incubated with the substrate and the catalytic capacity was determined as described under Materials and Methods. The specific activity toward benzphetamine in microsomes from OVEX rats was significantly reduced in comparison to control, and was partially restored in E2-treated OVEX rats (Fig. 6). Similar results were seen when ethylmorphine was used as a substrate (data not shown). It should be noted, however, that changes in microsomal N-demethylation activities do not follow as closely to the changes in either CYP3A protein levels or CYP3A9 mRNA expression after various treatments. This is consistent with expectation since apart from CYP3As, benzphetamineN-demethylation is also catalyzed by the CYP1, CYP2, and CYP4 families to varying extents. Therefore, changes in total microsomal benzphetamine demethylation reflect the sum of all contributing families.
Discussion
QRTPCR has been used recently by many groups to study P450 induction and expression patterns in cultured rat hepatocytes and in vivo animal models (Bowen et al., 2000; Burczynski et al., 2001;Kalsotra et al., 2002). QRTPCR assays were standardized and utilized to quantitate CYP3A9 message in rats. In terms of tissue distribution, maximum expression of CYP3A9 was observed in liver, which is the primary site for all phase I drug metabolism, followed by lung and equivalent levels of expression in kidney and brain. Furthermore, our results indicate a regio-selective CYP3A9 expression in kidney in favor of cortex over medulla. We have seen such regioselectivity with CYP4F family members, also. It seems reasonable that P450 levels are higher in cortex where glomerular filtration and resorption occurs, accounting for a protective role against drug-induced renal toxicity.
Sex-dependent expression of cytochromes P450 and their influence on drug metabolism and drug toxicity have been previously observed. Animal experiments, principally in rats, have revealed that expression of a number of P450 enzymes is sex-dependent (Waxman et al., 1995; Wang and Strobel, 1997; Kalsotra et al., 2002). The present study was performed in part to verify whether CYP3A9 truly exhibits sexually dimorphic expression in favor of females. We observed a 28-fold higher hepatic CYP3A9 expression in female rats than males, consistent with a 10-fold higher response published earlier using Northern blot techniques (Wang and Strobel, 1997). Also, CYP3A9 showed a 3.8-fold higher expression in lung. On the contrary, rat CYP3A1, CYP3A18 (Robertson et al., 1998), and CYP3A2 (Mahnke et al., 1997) are expressed in a male-dominant manner. Several players, including growth hormone, have been implicated in gender-specific expression of cytochrome P450 3As, which in turn are influenced by steroid hormones such as E2 and testosterone (Kawai et al., 2000). More recently, involvement of Signal Transducer and Activators of Transcription (STAT) family members, namely STAT 5a and STAT 5b, in growth hormone-mediated sexually dimorphic expression of genes has been reported (Park et al., 1999).
To understand the mechanism of gender specific CYP3A9 regulation we treated OVEX rats with female sex hormones E2 and PROG. The data suggest that E2 can differentially regulate hepatic and renal CYP3A9 in a tissue-specific manner. After OVEX, the absence of E2 leads to significant down-regulation of CYP3A9 mRNA levels in liver and brain, but a marked increase in kidney. These responses could be reversed by exogenous E2 treatment in liver but not to the same extent with PROG. It is known that E2 can cause both positive and negative regulation and tissue-selective responses through its receptors (Pettersson et al., 2000; Hall et al., 2001). The predominant biological effects of E2 are mediated through its receptors ERα and ERβ. In rats, ERα shows high expression in liver, kidney, testis, and ovaries, whereas ERβ is predominant in brain, lung, bladder, and epididymis (Kuiper et al., 1996). The distribution of estrogen receptors α and β in different tissues could offer one possible explanation for tissue-specific E2 response (Diel, 2002).
Unlike the reduced expression of hepatic CYP3A9 after OVEX, the reduced CYP3A9 levels in the brain could not be restored with exogenous E2. This result is in contrast to the previous report from our laboratory which showed using a PCR southern analysis, that brain CYP3A9 was partially restored upon E2treatment (Wang and Strobel, 1997). The QRTPCR technique used in the present study is more sensitive than PCR southern analysis and provides a better quantitation of CYP3A9 mRNA levels. Also, while E2 readily penetrates the blood-brain barrier, it is not retained by this organ, necessitating constant intracranial dosing to achieve sustained and pharmacologically relevant brain concentrations (Bodor and Buchwald, 1999). Furthermore, transporters such as P-glycoprotein (Rao et al., 1994), organic anion transporter peptide-2, and organic anion transporters (Sugiyama et al., 2001) can effectively efflux steroid hormones across the blood-brain barrier. This could partially explain the inability of the exogenous E2 regimen to restore the decreased CYP3A9 levels in brain upon OVEX.
We also examined the response of different concentrations of E2 on CYP3A9 mRNA levels in different tissues and found no linear correlation between the two doses used and CYP3A9 expression. This may imply that the E2 saturates at concentrations near a 500 μg/kg daily dose.
Another dramatic finding of this study is that CYP3A9 expression changed drastically in a tissue-specific manner throughout the course of pregnancy. The hepatic mRNA levels of CYP3A9 decreased on day 13 and day 19 of pregnancy when compared with the nonpregnant rats, but increased significantly on day 21 compared with day 13. This decrease in CYP3A9 levels could be due to suppression of CYP3A9 message by high plasma PROG levels reached during early pregnancy (Gaspar et al., 2001). This is also true with our data seen in OVEX animals treated with E2 and PROG, where the up-regulation exhibited with E2 treatment is reduced significantly in liver when both PROG and E2 are used. An analogous antagonistic role for PROG is observed in cytochrome P450 2D mRNA expression (Baum and Strobel, 1997). These data are also consistent with the recent finding that demonstrates a 4-fold increase in ERα expression in liver from day 15 to day 21 while ERβ remains fairly constant (Okada et al., 2002). Moreover, plasma E2 levels increase significantly to 500 pmol on day 21 as compared with 250 pmol on day 14 and day 19 of pregnancy (Gaspar et al., 2001). This increase provides a putative basis for the increased hepatic CYP3A9 expression seen at day 21 compared with day 14 or day 19.
Based on the above results, the ability of E2 to modulate CYP3A protein and its catalytic activity was investigated. In the absence of a monoclonal CYP3A9 antibody a polyclonal anti-CYP3A2 antibody was used. The CYP3A protein levels from liver and kidney changed in an analogous fashion to that of CYP3A9 message after OVEX followed by the different hormone treatments. These observations demonstrate that CYP3A9 mRNA regulation by E2 is also reflected downstream at the protein level. In addition, the activity data correlated well with the observed changes in mRNA and protein levels upon OVEX followed by hormonal regimen, suggesting that these changes have metabolic consequences for drug treatments as well as for steroid metabolism.
The existence of sex-dependent differences in drug metabolism is not unique to rats. Such differences have been reported in humans (Hunt et al., 1992). Therefore, elucidating the underlying mechanism of E2 regulation of CYP3A9 and its differential tissue-specific responses may help us gain insights into the mechanisms that lead to gender-associated differences in human therapeutic responses seen after treatment of men and women with the same drug (Rademaker, 2001). Since CYP3As are clinically relevant drug metabolizing enzymes, and are responsive to hormone levels, it is important to consider CYP3A status during different stages of pregnancy because the plasma concentrations of drugs administered during this period can significantly oscillate with the varying CYP3A levels.
Acknowledgments
We express thanks to Auinash Kalsotra and Cheri Turman for careful review of the manuscript. Our sincere thanks to Dr. Barbara Sanborn for help.
Footnotes
-
This work was supported by Grant MH58297 from the National Institute of Mental Health (Department of Health and Human Welfare) and by a Shell Toxicology Fellowship awarded to S.A. The data presented here form part of the dissertation of Sayeepriyadarshini Anakk submitted to the faculty of the University of Texas Graduate School of Biomedical Sciences in partial fulfillment of the requirements for the Doctor of Philosophy degree.
-
DOI: 10.1124/jpet.102.048090
- Abbreviations:
- P450
- cytochrome P450
- QRTPCR
- quantitative real-time polymerase chain reaction
- ER
- estrogen receptor
- OVEX
- ovariectomy
- E2
- estradiol benzoate
- PROG
- progesterone
- Received December 19, 2002.
- Accepted January 29, 2003.
- The American Society for Pharmacology and Experimental Therapeutics