Introduction

Top Abstract Introduction Results Discussion References

Rat liver contains at least a dozen sex-dependent¹ isoforms of CYP² that are regulated by the sex-dependent profiles of circulating growth hormone (Legraverend et al., 1992a; Waxman, 1992). Male rats secrete growth hormone in episodic bursts (~200-300 ng/ml) every 3.5 to 4 h. Between the peaks, growth hormone levels are undetectable. In females, the hormone pulses are more frequent and irregular and are of lower magnitude than those in males, whereas the interpulse concentrations of growth hormone are always measurable (Shapiro et al., 1995).

In the rat, the sexually dimorphic patterns of growth hormone secretion determine the expression levels of sex-dependent CYP isoforms. In the case of CYP2C12, the major female-specific form is dependent on the feminine profile of continuous growth hormone secretion. Exposure to the masculine profile of episodic hormone release, as well as the absence of the hormone from the circulation (e.g., hypophysectomy), completely prevents expression of CYP2C12 (Ram and Waxman, 1990; Legraverend et al., 1992b). Female-predominant CYP2C7 expression is also dependent on the feminine growth hormone profile and is completely suppressed in the hypophysectomized rat. However, exposure to the masculine profile allows expression of CYP2C7 at 25 to 40% of normal female levels (Ram and Waxman, 1990; Westin et al., 1990). After hypophysectomy, female-predominant CYP2A1 (male/female: ~1:3) concentrations decline to around male levels and are restored to intact female-like levels with continuously administered growth hormone (Pampori and Shapiro, 1996). Thus, continuous exposure of hepatocytes to growth hormone is the major factor regulating expression of female-dependent isoforms.

Although there are additional examples, including male-dependent isoforms, it has become evident that the expression or suppression of each isoform of CYP is likely to be regulated by a different "signal", or perhaps a differential sensitivity to the signal, in the sexually dimorphic growth hormone profile (Pampori and Shapiro, 1996, 1999; Agrawal and Shapiro, 2000a). These signals may be recognized by the hepatocyte in the frequencies and/or durations of the pulse and interpulse periods. Alternatively, the hepatocyte can monitor the mean plasma concentration of the hormone. The use of isolated hepatocytes in culture has become a very important model for investigating the oxidative metabolism and toxicity of a variety of endogenous and exogenous compounds as well as mechanisms regulating CYP expression (Schuetz et al., 1988; Liddle et al., 1992; Mitaka, 1998). The in vitro model has the advantage of using defined conditions without the confounding influence of the nutritional and hormonal status observed in vivo. However, the activities of various xenobiotic-metabolizing enzymes rapidly decline in culture (Bissell and Guzelian, 1980), especially the constitutive isoforms of hepatic CYP (Wright and Paine, 1992; Niwa et al., 1996). The loss of differentiated gene expression, including CYPs, in cultured hepatocytes has been attributed to such factors as the collagenase isolation procedure, loss of cell-cell interactions, absence of an extracellular biomatrix, and deficiencies in the culture media (Wright and Paine, 1992; Rana et al., 1994; Tomita et al., 1995). In recent years, several studies have reported a variety of reproducible alterations to the culture to minimize the decrease in CYPs, particularly the inducible forms, which includes the use of an extracellular matrix, chemically and hormonally defined culture media conditions, and coculture of hepatocytes with nonparenchymal cells (Waxman et al., 1990; Donato et al., 1994; Sidhu et al., 1994; Niwa et al., 1996). Regarding the female-dependent CYPs, CYP2C12 has been expressed for several days in primary culture (Guzelian et al., 1988; Legraverend et al., 1992b), as has CYP2C7 (Westin et al., 1997). Depending on culture conditions, expression of CYP2A1 has been substantially maintained for 10 to 20 days (Waxman et al., 1990) or suppressed after 5 days in culture (Kocarek et al., 1993). Since constituent CYP isoforms in primary hepatocyte cultures are usually expressed at concentrations below physiologic levels, it is difficult to assess whether the isoform is actually being induced, is in a steady-state condition, or is in slow decline (Sasaki et al., 1999). In the present article, we report that media supplementation (e.g., amino acids, hormones) and the presence of epidermal growth factor play an important role in the expression as well as suppression of female-dependent isoforms CYP2A1 and CYP2C12 in hepatocytes maintained in primary culture for 7 days.

Experimental Procedures

Materials. Hepatocyte cultures were prepared from adult female Sprague-Dawley rats (~200 g) purchased from Charles River Laboratories (Wilmington, MA). Penicillin/streptomycin, Williams' E medium, Dulbecco's modified Eagle's medium (DMEM), DMEM-Ham's F12 (1:1) and Ham's F12K media and other cell culture buffers were obtained from Biofluids, Inc. (Rockville, MD). Collagenase P and Complete protease inhibitor cocktail tablets were from Roche Molecular Biochemicals (Mannheim, Germany). TRIzol, HepatoZYME-serum-free medium, fetal bovine serum, bovine transferrine and epidermal growth factor were purchased from Life Technologies, Inc. (Grand Island, NY). Percoll was obtained from Amersham Pharmacia Biotech AB (Uppsala, Sweden). Matrigel and collagen type I rat tail were from Collaborative Research (Bedford, MA). Rat growth hormone (1.8 IU/mg) was obtained through the National Hormone and Pituitary Program and Dr. A. F. Parlow (UCLA Medical Center, Torrance, CA). Oligonucleotides were synthesized at the core facility of the School of Veterinary Medicine, University of Pennsylvania (Philadelphia, PA). The polyclonal anti-rat CYP2A1 antibody was a gift from Dr. Susumu Imaoka (Osaka City University Medical School, Osaka, Japan). The monoclonal anti-rat CYP2C12 antibody was kindly provided by Dr. Marika Rönnholm (Huddinge University Hospital, Huddinge, Sweden). Antibodies to C/EBP (14AA) and p21 (C-19) were from Santa Cruz Biotechnology, Inc. (St. Cruz, CA). Insulin, aminolevulinic acid, dexamethasone, and linoleic-albumin were from Sigma Chemical Co. (St. Louis, MO). The CellTiter 96 Aqueous One solution (MTS) cell proliferation assay was from Promega Corporation (Madison, WI). All other materials were of high purity and purchased from commercial sources.

Hepatocyte Isolation and Culture. Isolation of rat hepatocytes was performed by in situ perfusion of collagenase through the portal vein of anesthetized rats (Seglen, 1976). In brief, initial perfusion with a calcium-free buffer was followed by a solution of collagenase (0.05% w/v). The "softened" liver was excised, and the hepatocytes were separated from connective tissue by filtering through 100-µm macroporous filters (Spectrum Co., Laguna Hills, CA) and from nonparenchymal cells by repeated low speed centrifugation in wash medium: high glucose DMEM (4.5 g/l) containing streptomycin (100 µg/ml), penicillin (100 U/ml), Hepes (15 mM), pyruvate (5 mM), and fetal bovine serum (5%). The cell pellet was suspended in wash medium and mixed with an equal volume of Percoll density media, buffered with phosphate-buffered saline, and centrifuged for 10 min at 50g. The pellet was washed three times with wash media before cell counting. The viability of the initial cell suspension of hepatocytes was typically between 80 and 90% (trypan blue). After allowing 2 to 3 h for cell attachment, serum-containing medium was discarded and replaced by serum-free medium. The hepatocytes were plated at a density of 1.8 to 2.4 × 10⁵ viable cells/cm² in DMEM-Ham's F12 (DMEM-F12), DMEM-Ham's F12K (1:1) (DMEM-F12K), or HepatoZYME-serum-free medium (HepatoZYME) on 35-, 60-, or 100-mm plastic tissue culture dishes coated with either collagen type I at 7.27 ug/cm² or matrigel (274 ug/cm²). Both DMEM-F12 and DMEM-F12K media were supplemented with streptomycin (100 µg/ml), penicillin (100 U/ml), glutamine (2 mM), Hepes (15 mM), insulin (10 µg/ml), bovine transferrine (10 µg/ml), Na₂SeO₃ (10 ng/ml), aminolevulinic acid (2 µg/ml), dexamethasone (10 nM), glucose (25 mM), linoleic acid-albumin (0.5 mg/ml), pyruvate (5 mM), and rat growth hormone (100 ng/ml). HepatoZYME medium was supplemented with glutamine (2 mM), streptomycin (100 µg/ml), penicillin (100 U/ml), and rat growth hormone (100 ng/ml). The cultures were also supplemented with the antimycotic fungizone (0.25 µg/ml) for the initial 48 h only. Cultures were maintained in a humidified incubator at 37°C under an atmosphere of 5% CO₂/95% air. The medium was changed daily and cells were harvested at the time points indicated.

Isolation of Total RNA and Northern Blots. Hepatocyte cultures were washed with ice-cold phosphate-buffered saline containing 5 mM EDTA. Cells and gel were then removed from the culture dishes with a cell scraper, transferred to tubes, and placed on ice for approximately 1 h to dissolve the matrigel. After sedimentation, the cell pellets were kept at 70°C until extraction of RNA. Cells were lysed in TRIzol by several passages through a Pasteur pipette. Chloroform was added and mixed vigorously, followed by centrifugation. RNA in the aqueous phase was precipitated with isopropanol and washed with 75% ethanol. RNA pellets were dissolved in Tris-EDTA buffer. RNA yield and purity were assessed by absorbance at 260 and 260/280 nm, respectively. RNA samples (10-20 µg) were resolved on denaturing 1% agarose gels and transferred onto Nytran N filters from Schleicher & Schuell (Keene, NH). Specific oligonucleotides (Pampori and Shapiro, 1996) were radiolabeled with -³²P. Prehybridization and hybridization conditions have been described previously (Pampori and Shapiro, 1996). RNA bands were visualized by autoradiography. To verify RNA integrity and equal loading of RNA, membranes hybridized with CYP2C12 and CYP2A1 oligonucleotides were stripped off and reprobed with 18S rRNA.

Preparation of Cell and Nuclear Extracts for Immunoblots. To isolate protein for immunoblots, cultured hepatocytes were harvested as described above. Frozen cell pellets were then resuspended in lysis buffer containing 50 mmol Tris-HCl (pH 7.5), 0.3 M NaCl, 1% Triton X-100, 5 mM EDTA, 0.5% Nonidet P-40, and 10 µg/ml of leupeptin and aprotinin. The crude extract was passed through a 22-gauge needle 10 times. The solution was then gently mixed at 4°C for 20 min and centrifuged at 12,000g for 20 min. The supernatant (whole cell lysate) was then removed and stored at 70°C until analyses. Protein concentration was determined by the Bio-Rad (Hercules, CA) protein assay with bovine serum albumin as standard. Fifty micrograms of protein was resolved on 7.5% SDS-polyacrylamide gel electrophoresis and transferred electrophoretically onto nitrocellulose filters with a Trans-Blot semidry transfer cell (Bio-Rad). Membranes were stained with Ponceau Red to ensure equivalent amounts of protein loading and electrophoretic transfer among samples. The membranes were then blocked in Tris-buffered saline (10 mM Tris-HCl, 150 mM NaCl) with 5% (w/v) dry milk and 0.1% (v/v) Tween 20 and incubated with monoclonal antibody against rat CYP2C12 (1:500) or rabbit polyclonal antibody against rat CYP2A1 (1:1000) in blocking buffer for 2 h. Protein from whole cell extracts was also used for immunodetection with the p21 antibody (1:200) with overnight incubation at 4°C in blocking buffer. Following washes and incubation with the appropriate secondary antibodies, the antibody binding was visualized using an enhanced chemiluminescence detection system.

MTS Assay. Cell viability was determined by a modification of the CellTiter 96 Aqueous One solution cell proliferation assay from Promega Corporation. In brief, cells at different densities were incubated in 96-well plates with the serum-free medium conditions described above. The medium was changed daily, and 30 min or 1 h before the end of the incubation, a tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt], MTS, and an electron-coupling reagent (phenazine methosulfate) were added to the medium of the cultured cells. The conversion of MTS into the aqueous soluble product formazan is presumably accomplished by functional mitochondrial enzymes of metabolically active cells. The process is indicative of cell viability and correlated to cell number. Because primary cultured hepatocytes at the cell densities used here may produce more formazan than other cells, we used a more diluted ratio of the reagent (10 µl) and shorter incubation times (30 min or 1 h) than the original protocol. The absorbance at 490 nm of the cells and medium in the wells was measured.

	Results

Top Abstract Introduction Results Discussion References

Hepatocytes were isolated from the livers of adult female rats and plated on collagen or matrigel-coated dishes and cultured with Williams' E medium. This medium was initially chosen because it has been widely reported to improve the expression of several CYP mRNAs in cultured human (George et al., 1997) and rat (Sidhu and Omiecinski, 1995) hepatocytes. As shown in Fig. 1, hepatocytes cultured for 3 days in the absence of growth hormone, in either collagen or matrigel, did not sustain a significant expression of the female-specific constitutive isoform CYP2C12. The decline in CYP2C12 mRNA levels did not result from a loss in competency since hepatocytes cultured on matrigel-coated dishes and supplemented with growth hormone (100 ng/ml) exhibited normal-like CYP2C12 expression during the 3 days of the study.³ Unlike CYP2C12, mRNA expression of female-dependent CYP2A1 was substantial in hepatocytes cultured on either collagen or matrigel, the latter being more effective. The addition of growth hormone to the cells cultured on matrigel had no additional effects on CYP2A1 mRNA expression levels.

View larger version (74K): [in this window] [in a new window]   Fig. 1.   The effect of growth hormone on the expression of female-dependent hepatic isoforms in culture. Time course of female-specific CYP2C12 and female-predominant CYP2A1 gene expression in primary cultures of female rat hepatocytes cultured in either collagen or matrigel-coated dishes in the absence or presence of rat growth hormone (rGH). Hepatocytes were cultured in Williams' E medium and harvested 1, 2, and 3 days after plating. Equal amounts of total RNA were analyzed (20 µg) by Northern blotting with specific 32P-labeled oligonucleotide probes as described under Experimental Procedures. Total RNA obtained from the livers of normal female rats and precultured isolated hepatocytes are included for comparison. The blots were stripped of the CYP probes and then rehybridized for detection of ribosomal 18S mRNA. This figure is representative of two independent experiments.

Although it was our intention to investigate the mechanism(s) regulating long-term expression of hepatic CYPs in culture, the hepatocytes cultured in Williams' E Medium for periods longer than 3 days exhibited poor viability, as indicated by alterations in morphology. Accordingly, since it had been reported that the complex medium HepatoZYME could sustain hepatocyte viability and phenotype close to normal (Price et al., 1997), we sought to improve the culture conditions to favor the long-term survival of rat female hepatocytes by using the HepatoZYME medium. In addition, we used two other serum-free media: DMEM-F12 (Buggs et al., 1998) and a modification we called DMEM-F12K (DMEM/F12K, 1:1 v/v) that contained a significant increase in selective amino acids e.g., alanine, asparagine, aspartic acid, cysteine, glutamic acid, and proline when compared with DMEM-F12. As shown in Fig. 2, the mRNA for CYP2C12 was present in isolated hepatocytes, and the amount of the transcript was maintained at near normal levels for 7 days when the cells were cultured in either DMEM-F12 or DMEM-F12K media. Whereas CYP2C12 mRNA was consistently high on day 1 in cells cultured with HepatoZYME, the transcript of the isoform after 5 days in the media was dramatically decreased to undetectable levels. Whereas the expression of CYP2A1 mRNA was maintained for as long as 7 days, unlike CYP2C12, it was not significantly affected by the type of culture medium. Parenthetically, it should be mentioned that in the presence of growth hormone, CYP2C12 mRNA levels were similarly expressed when hepatocytes were cultured on a substratum of either collagen (Fig. 2) or matrigel (Fig. 1). We also found that DMEM-F12K plus growth hormone also sustained expression of female-dependent CYP2C7 (data not show). Lastly, there was no detectable expression of the major male-specific isoform CYP2C11 under any of the culture conditions (data not shown).

View larger version (72K): [in this window] [in a new window]   Fig. 2.   Comparison of medium formulations on CYP2C12 and CYP2A1 gene expression in primary cultures of female rat hepatocytes. Hepatocytes were plated on collagen-coated dishes, cultured in either DMEM-F12, DMEM-F12K, or HepatoZYME, all in the presence of rat growth hormone. Total RNA (15 µg) was extracted from normal liver and cultured hepatocytes and analyzed using Northern blot technology as described under Experimental Procedures. Although ribosomal 18S indicates a lower loading in the last lane for HepatoZYME medium on day 7, the data still indicate a significant absence of CYP2C12 expression, in agreement with HepatoZYME medium for days 3 and 5, but little change in CYP2A1 mRNA. Results are representative of two separate sets of cultures from different rat livers.

The above results indicate that media formulation DMEM-F12 or DMEM-F12K supplemented with growth hormone substantially sustained the expression of CYP2C12 mRNA longer than 3 days in hepatocytes cultured on collagen-coated dishes. Next, we examined which substratum/medium combination could maximally sustain expression of CYP2C12 mRNA. It has been widely reported that the culture of hepatocytes in a complex substratum (i.e., matrigel) improves liver-specific gene expression, including induction of CYP genes (Schuetz et al., 1988; Kocarek et al., 1993). When we compared the effects of collagen and matrigel on hepatocytes cultured for 5 days in DMEM-F12K medium, we observed a greater expression of CYP2C12 (Fig. 3) in cells cultured on the matrigel-coated plates. Similar to the results presented above (Fig. 2), the expression of CYP2C12 mRNA was not detected in cells after being cultured for 5 days in the HepatoZYME formulation on either collagen or matrigel-coated plates. Thus, regardless of the substratum used, cells cultured with the HepatoZYME medium failed to sustain CYP2C12 expression. Again, the expression of CYP2A1 mRNA did not vary significantly with the choice of substratum or media (Figs. 1-3). Because we found optimal expression of CYP2C12 on the matrigel substrate, we chose to perform all subsequent experiments on matrigel-coated plates.

View larger version (62K): [in this window] [in a new window]   Fig. 3.   Effects of culture medium and substratum on CYP2C12 and CYP2A1 gene expression. Hepatocytes were plated on collagen or matrigel-coated dishes and cultured in serum-free DMEM-F12K or HepatoZYME medium, both supplemented with rat growth hormone for 1 and 5 days. Total RNA (10 µg) was extracted and analyzed using Northern blot technology and radiolabeled specific oligonucleotides as described under Experimental Procedures. Total RNA (10 µg) from the livers of normal male and female rats as well as precultured isolated hepatocytes are included for comparative purposes. Data are from a single representative experiment.

We next investigated whether the suppressed expression of CYP2C12 in HepatoZYME could be prevented by switching the medium conditions of the cells after 48 h in culture. As shown in Fig. 4, CYP2C12 mRNA was detectable in cells cultured on matrigel-coated plates with DMEM-F12K medium for as long as 7 days, with some decline in the last day. CYP2C12 mRNA almost disappeared by day 3 in cells cultured in either DMEM-F12K or HepatoZYME medium but increased by day 5 only in cells cultured with DMEM-F12K. In agreement with previous reports measuring other isoforms (Kocarek et al., 1993), we observed this pattern of "crash-rise" only in cultures using matrigel as substratum, possibly resulting from a transient accumulation of nitric oxide (Lopez-Garcia, 1998). The suppression of CYP2C12 mRNA after day 1 in cells cultured in HepatoZYME persisted through day 7. When the cells were cultured for 48 h with DMEM-F12K and then switched to HepatoZYME medium for the following 3 days, expression of CYP2C12 mRNA was lost. Conversely, cells cultured for 48 h in HepatoZYME medium and then switched to DMEM-F12K medium exhibited restored CYP2C12 mRNA expression by day 5, which was still evident by day 7. CYP2A1 was expressed under all conditions, although at somewhat greater levels in DMEM-F12K. Although less dramatically than CYP2C12, CYP2A1 mRNA slightly decreased on day 3 of culture but consistently restored expression thereafter.

View larger version (72K): [in this window] [in a new window]   Fig. 4.   Reinduction of CYP2C12 mRNA by altering culture media. Female rat hepatocytes were plated on matrigel-coated dishes and cultured in serum-free DMEM-F12K or HepatoZYME medium supplemented with rat growth hormone for the indicated times. Initially, some hepatocytes were cultured for 48 h in DMEM-F12K or HepatoZYME medium and subsequently changed to HepatoZYME or DMEM-F12K, respectively, for the remaining time. Total RNA (15 µg) was extracted and analyzed as described under Experimental Procedures. Total RNA (15 µg) from the livers of normal male and female rats are included for comparison. The results are representative of two separate sets of cultures from different rat livers.

Western blot analyses (Fig. 5) of female hepatocytes cultured solely with DMEM-F12K for 7 days exhibited significant CYP2C12 protein levels, reflecting results observed at the mRNA level. As expected from mRNA results (Fig. 4), there was a complete loss of CYP2C12 protein in cells cultured either continuously or for the last 5 days with HepatoZYME medium. In contrast to results at the mRNA level, there was no observable reinduction of CYP2C12 protein following the change of medium from HepatoZYME at 48 h to DMEM-F12K. Whereas CYP2A1 mRNA was unaffected by media conditions, the protein levels showed a notable decline when cells were continuously cultured or switched to HepatoZYME medium. In contrast, cells cultured continuously in DMEM-F12K or switched from the HepatoZYME to DMEM-F12K medium exhibited similar elevated levels of CYP2A1 protein as mRNA.

View larger version (33K): [in this window] [in a new window]   Fig. 5.   Effects of culture conditions on expression and reinduction of CYP2C12 and CYP2A1 proteins. Female rat hepatocytes were plated on matrigel-coated dishes and cultured in serum-free DMEM-F12K or HepatoZYME medium supplemented with rat growth hormone. Initially, some hepatocytes were cultured for 48 h in DMEM-F12K or HepatoZYME medium and subsequently changed to HepatoZYME or DMEM-F12K, respectively, for the remaining time. Whole cell lysates (50 µg) were prepared and analyzed in immunoblots using CYP2C12 and CYP2A1 antibodies as indicated under Experimental Procedures. Microsomal protein (5 µg) from female rat liver is included as a positive control. The results are representative of two separate sets of cultures from different rat livers.

Since HepatoZYME medium contains EGF, a reported suppressor of growth hormone-independent male-specific CYP2C11 levels (Ching et al., 1996), it seemed possible that the presence of the growth factor in HepatoZYME medium could have been responsible for the suppression of CYP2C12 and CYP2A1 expression. Accordingly, we cultured hepatocytes in DMEM-F12K medium with or without EGF (20 ng/ml). Figure 6 shows a dramatic suppression of CYP2C12 protein levels in hepatocytes. Moreover, cultures in DMEM-F12K plus EGF for 48 h and then switched to DMEM-F12K medium without EGF re-expressed normal levels of CYP2C12 protein not observed when hepatocytes were cultured in HepatoZYME and then switched to DMEM-F12K (Fig. 5). CYP2A1 protein content of hepatocytes cultured with DMEM-F12K plus EGF showed a decline consistent with that observed when the cells were cultured in the HepatoZYME medium.

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Effects of EGF supplementation on CYP2C12 and CYP2A1 protein expression in culture. Female rat hepatocytes were plated on matrigel-coated dishes and serum-free DMEM-F12K medium with or without EGF (20 ng/ml) for the indicated times. Rat growth hormone was also added to all treatments. Initially, some hepatocytes were cultured for 48 h in medium without (or with) EGF and subsequently changed to medium with (or without) EGF for the remaining time. Whole cell lysates (50 µg) were prepared and analyzed in immunoblots using CYP2C12 and CYP2A1 antibodies as indicated under Experimental Procedures. Microsomal protein (5 µg) from normal female liver is included for comparison. The results are representative of two separate sets of cultures from different rat livers.

Next, we examined mediating factors involved in the decreased levels of both CYP2C12 and CYP2A1 protein in hepatocytes cultured with EGF. The presence of CYP is associated with nonproliferating terminally differentiated adult rat hepatocytes. (Greuet et al., 1997). A characteristic of hepatocytes in growth arrest is the presence of the cyclin-dependent kinase inhibitor p21, which can complex with the cell cycle-related proteins cyclin-CDK (Wu et al., 1996). Cells cultured in a confluent state in the absence of serum or growth factors similar to conditions in our study should be in growth arrest or in a differentiated state. Although p21 is present in the nucleus, it is more commonly localized in the cytoplasm (Wu et al., 1996). Figure 7 shows that after 24 h of culture, p21 protein levels present in the whole cell lysate were induced (not being present in the freshly prepared hepatocytes before culturing) at comparable concentrations in cell extracts from hepatocytes cultured with or without EGF. On day 3 of culture, p21 protein was higher in cells cultured in the presence of EGF, and by day 5 it was detected only in hepatocytes cultured with EGF, albeit at barely measurable levels. Regardless of the presence or absence of EGF, p21 was undetectable in cells cultured at 7 days. At no time, under any of our conditions, was p21, with the same molecular weight identified by the p21 antibody, detected in the nuclear fraction (data not presented). The observation that p21 protein was only detected in the whole cell extracts suggests that the use of hypertonic/detergent buffer during the cell lysate preparation may have resulted in a significant loss of certain nuclear proteins, particularly p21, into the cell extracts. We note, however, that other nuclear proteins (see below) were more resistant and need further high salt extraction.

View larger version (30K): [in this window] [in a new window]   Fig. 7.   Effect of EGF supplementation on p21 protein expression. Female rat hepatocytes were plated on matrigel-coated dishes and serum-free DMEM-F12K medium with or without EGF (20 ng/ml) for the indicated times. Rat growth hormone was also added to all treatments. Initially, some hepatocytes were cultured for 48 h in medium without (or with) EGF and subsequently changed to medium with (or without) EGF for the remaining time. Whole cell lysates (100 µg) were prepared and analyzed in immunoblots using p21 antibody as indicated under Experimental Procedures. The nuclear extract (20 µg) of K-ras transformed rat kidney cells (KNRK) is included as a positive control. The results are representative of two separate sets of cultures from different rat livers.

It has been reported that growth hormone-dependent regulation of CYP2C12 is accomplished only in the presence of the C/EBP (Tollet et al., 1995). In an attempt to elucidate the mechanism(s) involved in the EGF-mediated reduction of CYP2C12 protein in cultured hepatocytes, we measured levels of the transcription factor C/EBP (Fig. 8). Basically, we observed no direct relationship between nuclear C/EBP levels and CYP2C12 concentrations. That is, an EGF-dependent suppression of CYP2C12 was not associated with a decline in C/EBP levels.

View larger version (35K): [in this window] [in a new window]   Fig. 8.   Effect of EGF supplementation on C/EBP protein expression. Female rat hepatocytes were plated on matrigel-coated dishes and serum-free DMEM-F12K medium with or without EGF (20 ng/ml) for the indicated times. Rat growth hormone was also added to all treatments. Initially, some hepatocytes were cultured for 48 h in medium without (or with) EGF and subsequently changed to medium with (or without) EGF for the remaining time. Nuclear extracts (25 µg) were prepared and analyzed in immunoblots using C/EBP antibody as indicated under Experimental Procedures. Rat liver nuclear extract is included as positive control (20 µg). The results are representative of two separate sets of cultures from different rat livers.

Metabolism of tetrazolium salts, such as MTS, is a very sensitive index for measuring hepatocyte cell growth and viability (Fujii et al., 1995). Apparently, a decrease in formazan formation occurs before membrane damage can be detected. Optimizing the MTS assay for our cell culture conditions (matrigel-coated plates, cell density, etc.), we found a linear response between the cell number and the absorbance at 490 nm (Fig. 9A). Using the number of cells per well comparable with the cell density used in our studies (77,000 cells/well = 2.34 × 10⁵ cells/cm²; Fig. 9B), we observed, regardless of the presence of EGF in the medium, a consistent and similar viability of hepatocytes at the indicated incubation times (Fig. 9B). Of course, the daily change in medium with potential disruption of the matrigel substratum likely contributed to the declining values observed during the first days of culture. Although we do not know if EGF alters cell turnover, the data show that at each time point assayed EGF had no effect on the number of viable cells in culture. In this regard, viability was unaffected or somewhat enhanced in hepatocyte cultures grown with the HepatoZYME medium where both CYP2C12 mRNA and protein expression were suppressed (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 9.   Effects of incubation length and media supplementation on the viability of primary rat hepatocyte cultures. A, effects of cell number on viability as measured at an absorbance of 490 nm using the MTS. Increasing numbers of hepatocytes were plated on a 96-well plate with DMEM-F12K medium, increasing the density from 1.5 × 104 cells/cm2 (5,000 cells) to 3.0 × 105 cells/cm2 (100,000 cells). There are four replicates for each point. The background absorbance at 0 cells/well was subtracted from the data. B, effects of EGF on the viability of rat hepatocytes in culture. Cells (77,000 cells/well; 2.34 × 105 cells/cm2) were plated in a 96-well plate with DMEM-F12K medium with or without EGF (20 ng/ml), and in some cultures the medium was reversed after 48 h. The medium was replaced daily for 7 days. At the end of each time point, the MTS assay was used to calculate the percentage of viability (see Experimental Procedures). The readings taken from cells cultured 2 h after the isolation (day 0) were considered 100% viable. Each point represents the mean ± S.D. of triplicates.

	Discussion

Top Abstract Introduction Results Discussion References

An aim of this study was to identify suitable culture conditions of hepatocytes capable of sustaining long-term expression of the CYP genes present in the liver of adult female rats. By using a combination of matrigel (or collagen), high cell-to-cell contact, and growth factor-free media, we were able to maintain considerable and often physiologic-like expression levels (mRNA and protein) of the constitutive isoforms CYP2C12 and CYP2A1 for as long as 7 days. The cultures displayed good long-term viability. The data presented in this report also show that the presence of growth factors in the culture medium, such as EGF, significantly decreases the expression of both CYP2C12 and CYP2A1.

High concentrations of amino acids are especially effective in long-term cultures of hepatocytes to stimulate protein synthesis as well as to decrease protein degradation (Bissell and Guzelian, 1980; Mitaka, 1998). HepatoZYME, is a serum-free medium for the long-term maintenance of hepatocyte phenotypic expression, including several forms of CYP (Price et al., 1997). The HepatoZYME formulation is based on a modified Chee's essential medium, which has a high concentration of amino acids (Waxman et al., 1990; Liu et al., 1996), vitamins, and nutrients in addition to dexamethasone and insulin (Price et al., 1997). The above reports led us to choose HepatoZYME as a medium to culture hepatocytes for the long-term expression of CYP2C12 and CYP2A1. We also used two more serum-free formulation media: DMEM-F12 medium (Buggs et al., 1998) and a similar combination, DMEM-F12K (1:1, v/v). Unexpectedly, female hepatocytes cultured with HepatoZYME showed a dramatic loss of female-specific, growth hormone-dependent CYP2C12 mRNA and a complete suppression of its protein. In contrast, hepatocytes cultured using the other two media formulations exhibited sustained, normal-like levels of CYP2C12 and CYP2A1 mRNA and proteins. Although hepatocytes maintained in HepatoZYME expressed considerable CYP2A1 mRNA, protein levels of the isoform were markedly suppressed, suggesting a dysfunction in translation capability of the cell. Moreover, the apparent uncoupling of transcription from translation⁴ in HepatoZYME-cultured hepatocytes cautions against assumptions solely based upon mRNA levels. The differential effects of CYP2C12 and CYP2A1 to HepatoZYME may be related to the fact that CYP2C12 expression is completely dependent on growth hormone, whereas CYP2A1 can be expressed, albeit at reduced levels in the absence of the hormone (Pampori and Shapiro, 1996, 1999).

The suppression of CYP2C12 and CYP2A1 in hepatocytes cultured with the HepatoZYME medium was observed in cultures using rat tail collagen or matrigel as a substratum. Similar to previous reports (Liu et al., 1996), cells grown in collagen substratum spread and adopted a flattened morphology, while cells grown on matrigel were spherical and clustered into colonies. Hepatocytes in culture using matrigel have been associated with optimal preservation of liver CYP (Guzelian et al., 1988; Liddle et al., 1992; Kocarek et al., 1993), and in agreement, we observed greater expression of CYP2C12 and to a lesser extent CYP2A1 in hepatocytes plated on matrigel rather than collagen.

Clearly, the suppression of CYP2C12 and CYP2A1 by factor(s) in the HepatoZYME was reversible, as replacing the media with DMEM-F12K restored the isoforms. We supplemented both DMEM media with basic components for hepatocyte function also present in the HepatoZYME formulation (Price et al., 1997): insulin (Tollet et al., 1995; Buggs et al., 1998), dexamethasone (Waxman et al., 1990; Liu et al., 1996), aminolevulinic acid, linoleic acid/bovine serum albumin (Bissell and Guzelian, 1980), transferrine, and sodium selenite (Waxman et al., 1990; Buggs et al., 1998). However, HepatoZYME contains EGF, a mitogen-regulating hepatic regeneration and proliferation (Kimura and Ogihara, 1997; Grunnet et al., 1999) not found in DMEM media. The expression of several liver-specific functions appears to be inversely related to the degree of proliferation. For example, the male-specific isoforms of CYP, CYP2C11 and CYP3A2, are markedly suppressed in the regenerating liver (Liddle et al., 1989; Ronis et al., 1992). In the adult animal, hepatocytes are highly differentiated cells supporting a population of P450s different from that observed in differentiating liver (Agrawal and Shapiro, 2000b). Hepatic secretion of EGF is correlated with the progression of hepatocytes from the quiescent, mature state (G₀) to the G₁ state of the cell cycle and further progression to the formation of new, immature cells. (Loyer et al., 1996; Grunnet et al., 1999). In this regard, hepatocytes cultured continuously in HepatoZYME medium or switched after 48 h from DMEM-F12K to HepatoZYME containing epidermal growth factor may have entered a dedifferentiated state unable to support expression of CYP2C12 and CYP2A1 characteristic of adult female liver. Regardless of the effect of EGF on hepatocyte differentiation, suppression of CYP2C12 and CYP2A1 could not be explained by a general decline in cell viability as determined by mitochondrial oxidative activity.

The fact that hepatocytes cultured in EGF exhibited more p21 protein after 5 days than hepatocytes cultured in its absence reflects an expected countereffect by the cells in our culture conditions to retard the hepatocytes' progress in the cell cycle. p21 is a potent inhibitor of cyclin-dependent kinases acting possibly at the G₁-S cell cycle checkpoint (Albrecht and Hansen, 1999).

The preservation of hepatic expression of CYP2C12 and CYP2A1 in culture requires not only the inhibition of proliferation or DNA synthesis, due, at least in part, to the high cell density and resulting cell-cell contact inhibition of cell growth (Kimura and Ogihara, 1997) in our cultures, but also to the maintenance of appropriate transcription factors. Accordingly, a possible explanation for the EGF suppression of CYP2C12 could be a decline in P450-required transcription factors. It has been reported that C/EBP is one of several requisite transcription factors mediating activation of CYP2C12 gene expression (Tollet et al., 1995). Thus, we examined the consequences of EGF in culture on C/EBP levels. We did not, however, find differences in the levels of the 42-kDa C/EBP between cells cultured with or without EGF, suggesting that the presence of the transcription factor, alone, is not sufficient to assure CYP2C12 expression (Sasaki et al., 1999). Our findings are consistent with those reporting that growth hormone at the concentration used in our studies (100 ng/ml) induces both the 42- and 30-kDa isoforms of C/EBP protein in cultured rat hepatocytes (Strand et al., 2000).

In conclusion, we present a primary hepatocyte culture system plated at high density in the presence of insulin, dexamethasone, and growth hormone, but without growth factors, that sustains normal-like expression levels (mRNA and protein) of key female-dependent CYP2C12 and CYP2A1 (as well as CYP2C7) for at least 7 days. The mechanism by which EGF suppresses CYP2C12 and CY2A1 as well as growth hormone-independent CYP2C11 [reflecting ~25% of intact levels (Ching et al., 1996)] is unknown, but it probably does not involve alterations in C/EBP mediation of CYP2C12 or CYP2A1 levels. Lastly, it should be noted that culture conditions can differentially affect P450 transcription and translation and should be appropriately evaluated before use.