QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Data Supplements All Versions of this Article: jpet.104.068908v1 jpet.104.068908v2 311/2/728    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Williams, E. T. Articles by Strobel, H. W. Search for Related Content PubMed PubMed Citation Articles by Williams, E. T. Articles by Strobel, H. W.

ABSORPTION, DISTRIBUTION, METABOLISM, AND EXCRETION

Estrogen Regulation of the Cytochrome P450 3A Subfamily in Humans

Departments of Biochemistry and Molecular Biology (E.T.W., H.W.S.) and Integrative Biology and Pharmacology (P.J.A.D., D.S.L., G.L.S.), Medical School, University of Texas Health Science Center at Houston, Houston, Texas; Department of Statistics, Texas A&M University, College Station, Texas (M.L.); and Department of Drug Disposition, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana (S.A.W.)

Received April 27, 2004; accepted July 21, 2004.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
This study examines the possible role of estrogen in regulating the expression of the human CYP3A subfamily: CYP3A4, CYP3A5, CYP3A7, and CYP3A43. To accomplish this goal, mRNA was quantified from human livers and endometrial samples, and total CYP3A protein levels were evaluated by Western immunoblot analysis of the liver samples. The human endometrial samples were from premenopausal and postmenopausal women. The premenopausal endometrium was either in the proliferative or secretory phase, whereas for the postmenopausal endometrium samples, the women had been treated with either a placebo or estropipate, an estrogen substitute. After analyses, CYP3A4 mRNA was shown to have lower hepatic expression in females than in males. In the endometrium, CYP3A4 and CYP3A43 are down-regulated by estrogen, whereas CYP3A5 is expressed at higher levels during the secretory phase. CYP3A7 was not detected in the endometrium. In addition, the CYP3A subfamily showed increased mRNA expression in the liver as age increased. The expression levels of total CYP3A protein and total CYP3A mRNA showed good correlation. Despite apparent regulation of CYP3A4 mRNA expression by estrogen, the effects of estrogen may be overshadowed by additional regulators of gene expression.

In humans, four CYP3As are known to be expressed: CYP3A4, CYP3A5, CYP3A7, and CYP3A43. CYP3A4 is the most abundant P450 (Guengerich, 1995) and contributes to the metabolism of the largest percentage of clinically used drugs (Evans and Relling, 1999). CYP3A5 is polymorphically expressed, and in some individuals, CYP3A5 expression can equal that of CYP3A4 (Kuehl et al., 2001). CYP3A7 is known for its expression in fetal liver (Komori et al., 1990); however, expression has also been shown in some adult tissues (Burk et al., 2002). The latest human CYP3A to be discovered, CYP3A43, is thought to have the strongest mRNA expression in the liver and testis (Westlind et al., 2001) or the prostate (Gellner et al., 2001), but the expression of CYP3A43 is significantly lower than that of CYP3A4 (Gellner et al., 2001) and CYP3A5 (Westlind et al., 2001). Some of the substrates known to be metabolized by the CYP3As include erythromycin (Brian et al., 1990), cyclosporine (Kronbach et al., 1988), warfarin (Kaminsky and Zhang, 1997), and 17-estradiol (Lee et al., 2001).

Conducting in vivo research based on humans is extremely difficult, especially when examining estrogen regulation. Most of the techniques used in animal models are not acceptable for use on humans. Therefore, many researchers have relied on available tissue resources.

Recent studies of gender-based CYP3A expression in the liver include those of Westlind-Johnsson et al. (2003) and Wolbold et al. (2003). Westlind-Johnsson et al. did not uncover a significant difference between men and women for CYP3A4, CYP3A5, or CYP3A43 mRNA expression; however, Wolbold et al. reported that women had 2-fold higher expression of CYP3A4 mRNA and 3-fold higher protein expression than men.

The classical tissue for assaying P450 is the liver, since it is the major site of detoxification for the body and the principal location of drug metabolism. Comparing male and female livers will not directly identify genes that are regulated in part by estrogen; however, knowing gender differences in mRNA expression may suggest possible candidates for further investigation.

The endometrium may be a better tissue to investigate in terms of responsiveness to estrogen. It is not a primary site for drug metabolism; nevertheless, the CYP3As are present in the endometrium and may play an endogenous role. For premenopausal endometrium, studies of CYP3A mRNA expression include studies by Schuetz et al. (1993), Hukkanen et al. (1998), and Sarkar et al. (2003). Schuetz et al. reported that CYP3A7 was expressed higher in the secretory phase than the proliferative phase. Hukkanen et al. detected CYP3A4 and CYP3A5 but did not explore differences between phases, and CYP3A7 was not detected. Sarkar et al. found a significant difference between the two phases for CYP3A7, but not CYP3A4.

Similar to the results reported for CYP3A expression in liver, the literature discussing CYP3A expression in premenopausal endometrium is equally contradictory. Hukkanen et al. (1998) did not detect CYP3A7 expression; however, Schuetz et al. (1993) and Sarkar et al. (2003) did detect CYP3A7 expression, but they contradict each other in terms of the phase in which CYP3A7 is expressed at higher levels. Sarkar et al. showed that CYP3A7 was expressed at higher levels in the proliferative phase, whereas Schuetz et al. found higher expression during the secretory phase than during the proliferative phase.

The current study provides a more inclusive exploration into the relationship between estrogen and the human CYP3A subfamily by examining the expression of the CYP3A forms utilizing samples from the liver, premenopausal endometrium, and postmenopausal endometrium with and without exogenous estrogen treatment. Additionally, in the current study, the effect of age on CYP3A expression has also been analyzed.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Human Liver and Endometrial Samples. A total of 27 liver samples, 20 postmenopausal endometrial samples, and 13 premenopausal endometrial samples were obtained from various sources.

Of the human liver samples, 20 (HL1-HL20) were obtained under approved protocols from the Medical College of Wisconsin or the Indiana University School of Medicine, six (HL21-HL26) from Steve Strom at the University of Pittsburgh, and one (HL27) from the International Institute for the Advancement of Medicine (IIAM; Exton, PA). Most of the livers originated from transplant sources, although four originated from biopsy. All liver samples were from persons that died of accidental causes.

Thirteen human premenopausal endometrial samples were collected from uteri at the time of hysterectomy. Of the 13 hysterectomies, seven were performed for cervical cancer, four for leiomyomas, and two for ovarian neoplasms. All 13 samples were histologically normal. Determination of the phase of the menstrual cycle, whether proliferative or secretory, was made by histological staining.

The 20 human postmenopausal endometrial samples consisted of two groups of 10. One group was obtained from women treated with a placebo, whereas the other was obtained from women treated with estropipate (Wyeth Research, Philadelphia, PA), formerly identified as piperazine estrone sulfate. Each group was randomly selected from a larger group of samples. The estropipate-treated women were administered 0.625 mg of estropipate daily for 6 months prior to sample collection or biopsy. Please refer to Table 1 for clinical information regarding the postmenopausal endometrial samples.

RNA Isolation from Human Samples. For each human liver sample, 100 mg of whole liver was added to 1 ml of RNA STAT-60 reagent (CS-111; Tel-Test Inc., Friendswood, TX). The samples were homogenized until no debris was visible and allowed to sit at room temperature for 5 min. After adding 200 µl of chloroform, the mixtures were shaken for 15 s and allowed to sit at room temperature for 2 to 3 min. Then, the samples were centrifuged at 4°C for 15 min. The clear supernatant fractions were transferred to clean tubes, mixed with 500 µl of isopropanol, and allowed to sit at room temperature for 5 to 10 min. Next, the mixtures were centrifuged at 4°C for 10 min, and the supernatant fractions were discarded. The remaining pellets were washed with 1 ml of 75% ethanol and centrifuged at 4°C for 5 min. The supernatant fractions were discarded, and the pellet was dried and resuspended in water treated with diethyl pyrocarbonate.

The endometrial samples were processed as specified by Deng et al. (2003). All samples were DNase I-treated and stored at -80°C until analysis.

Quantitative Real-Time PCR. All liver and endometrial RNA samples were quantified using quantitative real-time PCR with an ABI Prism 7700 sequence detection system (Applied Biosystems, Foster City, CA).

The primer and probe sets were designed to overlap the exon/exon junctions of mRNA using the Primer Express software (Applied Biosystems). Regions of high homology between the human CYP3As were excluded during the design. The probes contain a 5' 6-FAM (5-carboxyfluorescein) and a 3' TAMRA (5-carboxytetramethylrhodamine). The standards were oligonucleotides spanning the region of the amplimer, three bases upstream of the forward primer to three bases downstream of the reverse primer. The primers, probes, and standards utilized in this study are listed in Supplemental Material Table 1 (Supplemental tables are available at http://jpet.aspet.journals.org) and were synthesized by SeqWright (Houston, TX), Integrated DNA Technologies (Coralville, IA), or BioSource International (Camarillo, CA).

For each sequence quantitated, four parameters were measured: 5-log dilution series of the oligonucleotide standards, no template controls (NTC), unknown samples, and no amplification controls (NAC) per unknown sample. The NTC omits RNA to ensure contaminant-free reagents, whereas the NAC omits the reverse transcriptase to ensure the lack of DNA contamination. All standards, NTCs, and unknown samples were conducted in triplicate with one NAC per unknown sample. The assays used reagents provided in the Superscript One-Step RT-PCR with platinum Taq kit (catalog no. 10928-042; Invitrogen, Carlsbad, CA). A total of 50 µl was aliquoted to each well and consisted of 1x reaction mix (a buffer with 0.4 mM each deoxynucleoside-5'-triphosphate and 2.4 mM magnesium sulfate), an additional 1.8 mM magnesium sulfate, 800 nM each of forward and reverse primers, and 200 nM probe. The standards, samples, and NTCs contained 1 µl of RT/Taq mixture, whereas the NACs contained 1 U of platinum Taq DNA polymerase (catalog no. 10966-018; Invitrogen).

The template used for the standards was a serial 10-fold dilution of the oligonucleotide standard ranging from 100 aM to 1 pM (quantity calculation described below). The NTC contained water instead of a template. For the liver samples, a template of 100 ng of RNA was used for each well containing samples and NAC. On the other hand, the endometrial samples had a template of 10 ng of RNA for the -actin assay and 40 ng of RNA for the CYP3A transcripts quantitated.

Each assay was tested for cross-reactivity using the assay mixture outlined above. The template mixture for the CYP3A4 assay contained the full-length cDNA for CYP3A5, CYP3A7, and CYP3A43. As a positive control, 1 ng of the full-length CYP3A4 cDNA was used as the template, whereas the negative control had no template. Likewise, the other CYP3A assays were tested for cross-reactivity. Only in the positive control reactions did amplification occur, suggesting no cross-reactivity among assays.

The results are reported in quantity of transcripts. By using the molecular weight of the standards, the number of grams to achieve the desired number of standard templates was calculated. ABI Prism's software then used the quantity of the standards to calculate the quantity of the samples.

Western Immunoblot Analysis. From the human liver samples, microsomes were prepared as previously described (Kalsotra et al., 2002). The protein concentration of the microsomes was determined using the BCA protein assay kit from Pierce (catalog no. 23225; Rockford, IL). A total of 50 µg of each sample was loaded onto 4 to 20% gradient gels (catalog no. 161-1105EDU; Bio-Rad, Hercules, CA) and electrophoresed at 100 V until the dye-front reached the end of the gel. The gels were transferred to pure nitrocellulose membranes using the suggested protocol supplied by the semidry electrophoretic transfer cell from Bio-Rad (catalog no. 170-3940). The membranes were blocked overnight using 5% (w/v) evaporated milk in Tris-buffered saline containing 0.0005% (v/v) Tween 20. The membranes were probed at a 1:1000 dilution using the CYP3A4 polyclonal antibody from Research Diagnostics (RDI-CYP3A4abr; Flanders, NJ) as the primary antibody for 2 h and at 1:1250 using a goat anti-rabbit horseradish peroxidase conjugate (catalog no. 170-5046; Bio-Rad) as the secondary antibody for 1 h. SuperSignal West Pico chemiluminescence (catalog no. 34080; Pierce) was used for detection in the ChemiGenius2 (Syngene, Frederick, MD). GeneTools 3.04b by Syngene quantitated the bands.

For separation of the individual CYP3A isoforms, CYP3A4 (catalog no. 456207; BD Gentest, Bedford, MA), CYP3A5 (catalog no. 456235; BD Gentest), and CYP3A7 (catalog no. 456237; BD Gentest) were purchased. CYP3A43 was excluded since it is not commercially available. All enzymes were loaded either individually or mixed.

Statistical Analyses. In the liver samples, gender and age were examined for effects. The data were transformed to normality using the Box-Cox transformation. Power transformations of -0.25, logarithm, -0.25, and logarithm were used in the case of CYP3A4, CYP3A5, CYP3A7, and CYP3A43, respectively.

A previous study using human liver samples was reported by Westlind-Johnsson et al. (2003); therefore, these data were also examined for gender and age effects utilizing the same statistical methods used to analyze the new data presented in this study. Power transformations of 0.25, 0.25, 0.25, and 0.50 were used in the case of CYP3A4, CYP3A5, CYP3A43, and pregnane X receptor (PXR), respectively.

For any CYP3A mRNA expression in the human liver that showed both an age and gender effect, a nonparametric test (Kruskal-Wallis) was conducted to confirm the parametric results. This analysis divided the human liver samples into two groups: females 55 years of age and under versus males of all ages and females over the age of 55. The age of 55 years was chosen as a conservative approximation of the median age for menopause.

In the endometrial samples, differences in mRNA expression were tested between proliferative and secretory stages in premenopausal individuals, as well as the placebo and estropipate treatments in the postmenopausal individuals. For CYP3A5, only data from premenopausal individuals were available due to the lack of available postmenopausal samples. Power transformations of 0.5, logarithm, and -0.25 were used in the case of CYP3A4, CYP3A5, and CYP3A43, respectively. Differences between different stages and treatments were tested using Scheffe's test for multiple comparisons.

For all the above analyses, in the SAS PROC MIXED procedure (SAS Institute, Cary, NC), compound symmetry covariance structure was used on observations from the same individual. We also standardized each observed level of RNA by the average level of -actin in each individual.

The mRNA expression levels of all four enzymes collected in this study were combined for each sample and analyzed using a power transformation of -0.25. The Kruskal-Wallis analysis described above was also conducted on the combined data. The total protein expression levels were analyzed using a power transformation of -0.25 and analyzed by the Kruskal-Wallis analysis described above. Spearman's correlation was conducted between the total mRNA expression and the total protein expression. A p value less than 0.05 is considered to be statistically significant.

	Results

Top Abstract Materials and Methods Results Discussion References

 
In this study, RNA samples from two human tissues were analyzed as well as protein expression from the liver. The first tissue analyzed was a comparison of human liver samples from males and females for CYP3A mRNA expression levels. The second tissue analyzed was the human endometrium.

A total of 27 human liver samples were collected. For each sample, the mRNA was quantified by quantitative real-time PCR for -actin, CYP3A4, CYP3A5, CYP3A7, and CYP3A43 mRNA expression. Each CYP3A was normalized against -actin, and the results are listed in Supplemental Material Table 2.

Statistical analyses were conducted on the CYP3A4 mRNA levels of the liver samples for age and gender effects. For CYP3A4 mRNA, age and gender effects were found to be statistically significant covariates, as shown in Table 2, with younger females (Fig. 1A) expressing lower mRNA levels than younger males (Fig. 1B). For CYP3A7 and CYP3A43, only an age effect was found to be statistically significant. For these three genes, mRNA expression was found to increase with age, as seen in Fig. 1, A, B, D, and E. The estimated model for each is given in Supplemental Material Table 3 and shown in Fig. 1. On the other hand, no significant age or gender effect was found for CYP3A5, as shown in Fig. 1C.

View larger version (16K): [in this window] [in a new window]   Fig. 1. Data from Supplemental Material Table 2 plotted for females () and males () with the estimated model (solid line) from Supplemental Material Table 3 and 95% confidence interval (dashed lines) for CYP3A4 (A and B), CYP3A7 (D), and CYP3A43 (E). CYP3A5 (C) only shows data plotted due to the absence of statistically significant effects. p values are listed in Table 2.

As confirmation of the results obtained from the CYP3A4 regression analyses in this study, the Kruskal-Wallis nonparametric test was used on the mRNA expression levels of hepatic CYP3A4. The data divided into two groups, as described under Materials and Methods, reveals a significant result (p = 0.0190) that women 55 years and younger had a much lower expression of CYP3A4 mRNA. More samples, particularly from younger females, would allow for a refinement of the statistical model.

Recently, a similar study was reported by Westlind-Johnsson et al. (2003). In their study, the mRNA levels for CYP3A4, CYP3A5, CYP3A43, and PXR were quantified in the human liver after normalization against human acidic ribosomal phosphoprotein. The results of the Westlind-Johnsson et al. study were analyzed in a similar manner to the data generated in this study. An age effect was found to be statistically significant for CYP3A4, CYP3A5, and CYP3A43, as shown in Table 2. CYP3A7 was not quantified in the Westlind-Johnsson et al. study. Similar to the results reported here, the data from Westlind-Johnsson et al. show that the level of mRNA increases with age. The estimated models are given in Supplemental Material Table 3 and shown in Fig. 2. No statistically significant age or gender effect was found for PXR.

View larger version (17K): [in this window] [in a new window]   Fig. 2. Data from Westlind-Johnsson et al. (2003) plotted for females () and males () with the estimated model (solid line) from Supplemental Material Table 3 and 95% confidence interval (dashed lines) for CYP3A4 (A), CYP3A5 (B), and CYP3A43 (C). PXR (D) only shows data plotted due to the absence of statistically significant effects. p values are listed in Table 2.

A total of 27 human livers were obtained and assayed for the mRNA levels of all four human CYP3A forms (Supplemental Material Table 2) with -actin as the normalizer. In the study conducted by Westlind-Johnsson et al. (2003), 46 human livers were assayed for CYP3A4, CYP3A5, CYP3A43, and PXR mRNA with human acidic ribosomal phosphoprotein as the normalizer. Since these two studies used different genes for normalization, the results were not combined.

In addition to the human liver samples, CYP3A mRNA levels from human endometrial samples were analyzed. Two sets of samples from human endometria were obtained, premenopausal and postmenopausal. The premenopausal samples were separated into proliferative and secretory phases. A few samples were found to be inactive and atrophic, suggesting the possibility that the women were undergoing menopause. Other samples were found to have progestational effects, suggesting that this group of women were using oral contraceptives. The endometrial samples that were inactive or showed progestational effects were excluded from analyses. Supplemental Material Table 4 shows the levels of CYP3A gene expression, normalized to -actin, for the samples that were included and excluded from statistical analyses.

The postmenopausal samples were divided into two treatment groups, either placebo or estropipate. To determine which human CYP3A genes should be quantified, a pool of 10 randomized samples for initial testing was generated for both groups; the results are shown in Supplemental Material Table 5. After quantitating the pool samples, a new group of 10 randomized samples were analyzed for both groups and normalized to -actin; the results are shown in Supplemental Material Table 6. Clinical information regarding each group is compiled in Table 1. For expression levels for CYP3A4 mRNA, the placebo-treated endometrium was higher than the estropipate-treated endometrium and premenopausal (both phases) endometrium, as shown in Fig. 3A. The multiple comparison adjusted p values were determined for placebo- versus estropipate-treated (p = 0.0213), placebo-treated versus proliferative phase (p < 0.0001), and placebo-treated versus secretory phase (p = 0.0019). Verification of the results was obtained using the Kruskal-Wallis nonparametric test, which gave a statistically significant p value (p = 0.0003). For CYP3A5 (Fig. 3B), the proliferative phase was significantly lower than the secretory phase (p = 0.0058). The Kruskal-Wallis nonparametric test that was used as confirmation of CYP3A5 had a statistically significant p value (p = 0.0087). No statistical differences were found for CYP3A43 (Fig. 3C). Table 3 provides the estimated means and standard errors for these determinations.

View larger version (8K): [in this window] [in a new window]   Fig. 3. Comparison of premenopausal phases (Supplemental Material Table 4) and postmenopausal treatments (Supplemental Material Table 6) for CYP3A4 (A), CYP3A5 (B), and CYP3A43 (C). The error bars represent the standard error (Table 3). A single asterisk (*) denotes p < 0.05, and a double asterisk (**) denotes p < 0.01. Estimated means and standard errors are listed in Table 3.

From the human liver samples gathered in this study, all mRNA expression levels were combined for each sample and analyzed in a fashion similar to the analysis for CYP3A4. Both age (p = 0.0051) and gender (p = 0.0498) effects were found to be significant, but the interaction between the two was not statistically significant. The Kruskal-Wallis analysis for two groups was significant (p = 0.0290).

The results presented thus far reveal associations between age and/or gender effects and mRNA expression levels for human CYP3A subfamily members as determined by quantitative real-time PCR. To extend the results seen with mRNA expression to protein expression, the separation of individual CYP3A isoforms was attempted, as seen in Fig. 4. The CYP3A isoforms were not separated by electrophoresis, which is consistent with previous results reported (Domanski et al., 2001) using different conditions. Therefore, the total CYP3A protein expression was examined for associations with age and/or gender effects. Using similar statistical analyses for total protein expression, as had been conducted for total mRNA expression, yielded no statistically significant relationships with age and/or gender. However, the Spearman's correlation coefficient (0.62149), which is a measure of the degree of relatedness between variables, was significant (p = 0.0005) for the relationship between total protein and total mRNA expression. The human endometrium samples were not analyzed due to limited quantity of samples obtainable at collection.

View larger version (10K): [in this window] [in a new window]   Fig. 4. Separation of commercially-available human CYP3A proteins. The proteins were loaded onto the gel either individually or mixed. The bands shown are as follows: CYP3A4, CYP3A5, CYP3A7, CYP3A4/CYP3A5, CYP3A5/CYP3A7, CYP3A4/CYP3A7, and CYP3A4/CYP3A5/CYP3A7.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The focus of this study was to determine whether estrogen regulates the expression of the human CYP3A subfamily. To accomplish this goal, liver and endometrial samples were assayed for the mRNA levels for CYP3A4, CYP3A5, CYP3A7, and CYP3A43 and normalized against -actin. The mRNA of the liver was then compared with liver protein expression.

The increase in liver mRNA expression with age is consistent across all human CYP3A isoforms. The reason for this observation is not obvious, but one possibility is a tie to a common function of the CYP3A isoforms, the ability to metabolize exogenous compounds. Some of the exogenous substrates are known to up-regulate the expression of CYP3A4 through activation of PXR (Goodwin et al., 1999). The possibility for increased PXR activation with age exists despite the fact that PXR mRNA expression was not found to increase with age, as shown by the data from the Westlind-Johnsson et al. study (Fig. 2D). The increase in PXR activation may be the result of an increase in medications as age increases; thus, no increase in PXR mRNA expression would be observed. This increase in medications taken as one ages would be expected in the general population. However, according to the known medical histories of the liver donors used in the current study, no such difference was evident. Thus, PXR activation causing the increase in CYP3A expression as age increases cannot be confirmed.

A reason the CYP3As increase with age may lie in the source of the liver samples obtained. Most of the livers were obtained from transplantation, others from biopsy, but the source is not known for all samples. Therefore, the liver samples may have been handled differently depending upon the source. Of the samples with known sources, the samples of people aged 58 and younger were from transplantation, and the samples of people aged 59 and older were from biopsy. Our observations suggest a difference in mRNA expression quantified for each CYP3A versus the source of the sample; however, upon analysis, the difference was not statistically significant. Therefore, it is unlikely that the source of the samples had any significant effects on the expression levels of the CYP3As as a function of age.

In liver, CYP3A4 showed both gender and age dependencies. The results show that postmenopausal women express CYP3A4 mRNA at equal levels with men, which supports the notion that estrogen down-regulates CYP3A4 mRNA expression and is consistent with the results of the human endometrial samples in this study. Another study (Wolbold et al., 2003) reports differing findings, but did not examine age. In their study, the mean age for their female control group is about 50 (premenopausal), whereas the mean age for females exposed to drugs is about 65 (postmenopausal). Using the results of our study, the findings by Wolbold et al. would be expected since their female control group is premenopausal (suppressed CYP3A4 expression) and their female test group is postmenopausal (unsuppressed CYP3A4 expression) versus men of similar ages (unsuppressed CYP3A4 expression).

Premenopausal and postmenopausal endometrial samples were also collected and the mRNA levels of the CYP3A isoforms were quantified. To conserve the postmenopausal samples, pools of 10 samples each were generated from placebo- and estropipate-treated. The results of the pools suggested that CYP3A7 could not be quantified but that the other three CYP3A forms would be good candidates for quantification. CYP3A4 was chosen to be quantified since it is the dominant CYP3A subfamily member expressed in the human liver and has been reported to show gender bias in the liver (Wolbold et al., 2003). Since CYP3A43 is the latest CYP3A discovered in humans, it was also examined. CYP3A5 was not analyzed due to limited sample quantity.

For the premenopausal endometrial samples, the results of this study are in agreement with the results of Sarkar et al. (2003) for CYP3A4. Specifically, CYP3A4 remains constant during the proliferative and secretory phases (Fig. 3A). Postmenopausal endometrial samples, treated with a placebo, show a drastic increase in CYP3A4 mRNA expression; however, those treated with an exogenous estrogen express lower levels of CYP3A4, although not as low as premenopausal samples. This suggests that CYP3A4 is regulated by estrogen. At least two possibilities exist as to why the premenopausal endometrium expresses lower levels of CYP3A4 than the estrogen-treated postmenopausal endometrium. One possibility is that the cycling endometrium has additional regulators of CYP3A4 expression that are inactive in the postmenopausal state. A second possibility refers to a result of this study in the human liver, that is CYP3A4 mRNA expression increases as age increases. On average, the postmenopausal women are older than the premenopausal women. Also, in conjunction, these explanations are plausible.

According to this study, CYP3A5 expression is significantly higher in the secretory phase than the proliferative phase of the endometrium (Fig. 3B). Another study reported CYP3A5 expression in premenopausal endometrium without exploring differences between phases (Hukkanen et al., 1998).

The current study did not detect CYP3A7 mRNA in the vast majority of the endometrial samples in agreement with a previous study (Hukkanen et al., 1998), but in disagreement with two other studies (Schuetz et al., 1993; Sarkar et al., 2003). The findings of Schuetz et al. for CYP3A7 are similar to the findings for CYP3A5 in our study. For Schuetz et al., the possibility exists for cross-reactivity of the CYP3A7 probe with CYP3A5. At the time of the Schuetz et al. publication, the degree of similarity for a large portion of the probe between CYP3A5 and CYP3A7 was not known. Therefore, the Schuetz et al. findings could be reporting CYP3A5 mRNA expression in the endometrium as well as CYP3A7.

CYP3A43 appears to be down-regulated by estrogen (Fig. 3C) since the endometrial samples from the placebo-treated, postmenopausal women appear to have higher mRNA expression than those of the estrogen-treated postmenopausal and premenopausal women, although the changes are not significant. CYP3A43 could be expressed in higher levels in the proliferative rather than the secretory phase, although additional studies are necessary for a more definitive answer.

The results of the current study are consistent with the notion of estrogen decreasing the mRNA levels of CYP3A4 in the human postmenopausal endometrium and may do the same to CYP3A43 (based on our data). With this information, and setting aside the effects of other possible regulators of transcription, the mRNA levels of CYP3A4 and CYP3A43 would be expected to be lower during the proliferative phase in human premenopausal endometrium compared with the secretory phase. This difference is seen with CYP3A5 but not with CYP3A4 and CYP3A43. In fact, the results of CYP3A43 are opposite to the prediction; the proliferative phase is higher than the secretory phase. For CYP3A4 and CYP3A43, additional regulators of transcription may cause the deviation from the model described above. The addition of estrogen to the postmenopausal endometrium may be causing the atrophic endometrium to mimic the premenopausal endometrium (Nilsson et al., 1980). Nilsson et al. demonstrated that morphologically atrophic endometrium subjected to estrogen resembles premenopausal endometrium. If the postmenopausal endometrium does resemble premenopausal endometrium as a result of estrogen treatment, then the down-regulation observed for CYP3A4, and possibly CYP3A43, may be a result of other factors found in premenopausal endometrium and estrogen may or may not be regulating the CYP3A forms. In addition, the CYP3A mRNA levels of the liver samples, except for CYP3A4, showed no difference between genders. Since additional regulators of transcription are most likely involved, it is not surprising to observe a difference between the results obtained and what was predicted.

In contrast to the mRNA levels observed, reports have been published indicating no clinical difference of midazolam (Gorski et al., 2000) and erythromycin (Harris et al., 1996) metabolism between untreated, postmenopausal women and postmenopausal women taking hormone replacement therapy. Therefore, despite CYP3A4 mRNA levels being suppressed by estrogen, this form of regulation appears to play a small role, if any, in regard to clinical significance for the individual possibly due to an increase in the metabolic role played by other P450 forms.

The results of the mRNA expression are consistent with the results of the total protein expression reported in this study. Although total protein did not significantly increase with age or gender as total mRNA did, a significant positive relationship exists between the two, based on the Spearman's correlation coefficient. This means that as the expression of total mRNA or total protein increases, the other will also tend to have a higher expression value. However, the relationship is not very strong, indicating that factors other than mRNA expression may influence protein expression level.

On the other hand, CYP3A4 may be regulated by estrogen only in certain tissues similar to rat CYP3A9 (Anakk et al., 2003). Anakk et al. showed that CYP3A9 was regulated by estrogen differently in the liver and brain. The liver predominantly expressed estrogen receptor , whereas the brain expressed estrogen receptor .

In conclusion, estrogen may be important for tissue-specific expression of the CYP3As. The literature is divided on the issue of gender differences in the liver expression of CYP3A4. Perhaps this reflects, among many other things, the difficulty of doing population studies when in actuality other individual differences among women and among men make strong contributions to overall expression of the drug metabolizing enzymes.

	Acknowledgements

 
We thank Cheri Turman, Sayeepriyadarshini Anakk, Auinash Kalsotra, and Mikhail Bogdanov for thoughtful discussion of this work.

	Footnotes

 
This study was supported by the Women's Fund for Health Education and Research, 2002; Houston, Texas.

doi:10.1124/jpet.104.068908.

ABBREVIATIONS: P450, cytochrome P450; PCR, polymerase chain reaction; NTC, no template control; NAC, no amplification control; PXR, pregnane X receptor.

Address correspondence to: Henry W. Strobel, 6431 Fannin, MSB 6.200, Houston, TX 77030. E-mail: henry.w.strobel{at}uth.tmc.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Anakk S, Ku CY, Vore M, and Strobel HW (2003) Insights into gender bias: rat cytochrome P450 3A9. J Pharmacol Exp Ther 305: 703-709.[Abstract/Free Full Text]

Brian WR, Sari MA, Iwasaki M, Shimada T, Kaminsky LS, and Guengerich FP (1990) Catalytic activities of human liver cytochrome P-450 IIIA4 expressed in Saccharomyces cerevisiae. Biochemistry 29: 11280-11292.[CrossRef][Medline]

Burk O, Tegude H, Koch I, Hustert E, Wolbold R, Glaeser H, Klein K, Fromm MF, Nuessler AK, Neuhaus P, et al. (2002) Molecular mechanisms of polymorphic CYP3A7 expression in adult human liver and intestine. J Biol Chem 277: 24280-24288.[Abstract/Free Full Text]

Deng L, Shipley GL, Loose-Mitchell DS, Stancel GM, Broaddus R, Pickar JH, and Davies PJ (2003) Coordinate regulation of the production and signaling of retinoic acid by estrogen in the human endometrium. J Clin Endocrinol Metab 88: 2157-2163.[Abstract/Free Full Text]

Domanski TL, Finta C, Halpert JR, and Zaphiropoulos PG (2001) cDNA cloning and initial characterization of CYP3A43, a novel human cytochrome P450. Mol Pharmacol 59: 386-392.[Abstract/Free Full Text]

Evans WE and Relling MV (1999) Pharmacogenomics: translating functional genomics into rational therapeutics. Science (Wash DC) 286: 487-491.[Abstract/Free Full Text]

Gellner K, Eiselt R, Hustert E, Arnold H, Koch I, Haberl M, Deglmann CJ, Burk O, Buntefuss D, Escher S, et al. (2001) Genomic organization of the human CYP3A locus: identification of a new, inducible CYP3A gene. Pharmacogenetics 11: 111-121.[CrossRef][Medline]

Goodwin B, Hodgson E, and Liddle C (1999) The orphan human pregnane X receptor mediates the transcriptional activation of CYP3A4 by rifampicin through a distal enhancer module. Mol Pharmacol 56: 1329-1339.[Abstract/Free Full Text]

Gorski JC, Wang Z, Haehner-Daniels BD, Wrighton SA, and Hall SD (2000) The effect of hormone replacement therapy on CYP3A activity. Clin Pharmacol Ther 68: 412-417.[CrossRef][Medline]

Guengerich FP (1995) Human cytochrome P450 enzymes, in Cytochrome P450: Structure, Mechanism and Biochemistry (Ortiz de Montellano PR ed) pp 473-536, Plenum Press, New York and London.

Guengerich FP (1999) Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu Rev Pharmacol Toxicol 39: 1-17.[CrossRef][Medline]

Harris RZ, Benet LZ, and Schwartz JB (1995) Gender effects in pharmacokinetics and pharmacodynamics. Drugs 50: 222-239.[Medline]

Harris RZ, Tsunoda SM, Mroczkowski P, Wong H, and Benet LZ (1996) The effects of menopause and hormone replacement therapies on prednisolone and erythromycin pharmacokinetics. Clin Pharmacol Ther 59: 429-435.[Medline]

Hukkanen J, Mantyla M, Kangas L, Wirta P, Hakkola J, Paakki P, Evisalmi S, Pelkonen O, and Raunio H (1998) Expression of cytochrome P450 genes encoding enzymes active in the metabolism of tamoxifen in human uterine endometrium. Pharmacol Toxicol 82: 93-97.[Medline]

Kalsotra A, Anakk S, Boehme CL, and Strobel HW (2002) Sexual dimorphism and tissue specificity in the expression of CYP4F forms in Sprague Dawley rats. Drug Metab Dispos 30: 1022-1028.[Abstract/Free Full Text]

Kaminsky LS and Zhang ZY (1997) Human P450 metabolism of warfarin. Pharmacol Ther 73: 67-74.[CrossRef][Medline]

Komori M, Nishio K, Kitada M, Shiramatsu K, Muroya K, Soma M, Nagashima K, and Kamataki T (1990) Fetus-specific expression of a form of cytochrome P-450 in human livers. Biochemistry 29: 4430-4443.[CrossRef][Medline]

Kronbach T, Fischer V, and Meyer UA (1988) Cyclosporine metabolism in human liver: identification of a cytochrome P-450III gene family as the major cyclosporinemetabolizing enzyme explains interactions of cyclosporine with other drugs. Clin Pharmacol Ther 43: 630-635.[Medline]

Kuehl P, Zhang J, Lin Y, Lamba J, Assem M, Schuetz J, Watkins PB, Daly A, Wrighton SA, Hall SD, et al. (2001) Sequence diversity in CYP3A promoters and characterization of the genetic basis of polymorphic CYP3A5 expression. Nat Genet 27: 383-391.[CrossRef][Medline]

Lee AJ, Kosh JW, Conney AH, and Zhu BT (2001) Characterization of the NADPH-dependent metabolism of 17beta-estradiol to multiple metabolites by human liver microsomes and selectively expressed human cytochrome p450 3A4 and 3A5. J Pharmacol Exp Ther 298: 420-432.[Abstract/Free Full Text]

Nilsson O, Englund D, Weiner E, and Victor A (1980) Endometrial effects of levonorgestrel and estradiol: a scanning electron microscopic study of the luminal epithelium. Contraception 22: 71-83.[Medline]

Sarkar MA, Vadlamuri V, Ghosh S, and Glover DD (2003) Expression and cyclic variability of CYP3A4 and CYP3A7 isoforms in human endometrium and cervix during the menstrual cycle. Drug Metab Dispos 31: 1-6.[Abstract/Free Full Text]

Schuetz JD, Kauma S, and Guzelian PS (1993) Identification of the fetal liver cytochrome CYP3A7 in human endometrium and placenta. J Clin Investig 92: 1018-1024.

Westlind A, Malmebo S, Johansson I, Otter C, Andersson TB, Ingelman-Sundberg M, and Oscarson M (2001) Cloning and tissue distribution of a novel human cytochrome p450 of the CYP3A subfamily, CYP3A43. Biochem Biophys Res Commun 281: 1349-1355.[CrossRef][Medline]

Westlind-Johnsson A, Malmebo S, Johansson A, Otter C, Andersson TB, Johansson I, Edwards RJ, Boobis AR, and Ingelman-Sundberg M (2003) Comparative analysis of CYP3A expression in human liver suggests only a minor role for CYP3A5 in drug metabolism. Drug Metab Dispos 31: 755-761.[Abstract/Free Full Text]

Wolbold R, Klein K, Burk O, Nussler AK, Neuhaus P, Eichelbaum M, Schwab M, and Zanger UM (2003) Sex is a major determinant of CYP3A4 expression in human liver. Hepatology 389: 978-988.[CrossRef]

This article has been cited by other articles:

				P. Hanifi-Moghaddam, B. Boers-Sijmons, A. H.A. Klaassens, F. H. van Wijk, W. F. Van Ijcken, P. Van der Spek, H. A.M. Verheul, H. J. Kloosterboer, C. W. Burger, and L. J. Blok Difference in signalling between various hormone therapies in endometrium, myometrium and upper part of the vagina Hum. Reprod., February 1, 2008; 23(2): 298 - 305. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplements All Versions of this Article: jpet.104.068908v1 jpet.104.068908v2 311/2/728    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Williams, E. T. Articles by Strobel, H. W. Search for Related Content PubMed PubMed Citation Articles by Williams, E. T. Articles by Strobel, H. W.