- Top of page
- Materials and Methods
Abstract: The expression of drug metabolizing cytochrome P4502A (CYP2A) is highly gender-dependent in minipigs with the highest activity in females. In other species, orthologs of CYP2A have been shown to be under the regulation of nuclear receptor constitutive androstane receptor, whereas little is known about regulation in pigs. To investigate the effect of sex hormones on porcine cytochrome P450 CYP2A and CYP3A expression was assessed in liver samples taken before and after castration of sexually mature minipig boars. Removal of the primary androgen source resulted in significant increases of CYP2A mRNA, protein and enzyme activity levels. Likewise, expression of CYP3A was increased, although to a lesser extent. To examine the involvement of constitutive androstane receptor in the regulation of CYP2A, primary porcine hepatocytes were exposed to modulators of murine constitutive androstane receptor and human constitutive androstane receptor activity. The CYP2A activity was significantly increased by exposure to phenobarbital, an indirect activator of constitutive androstane receptor, and the human constitutive androstane receptor-ligand CITCO. In contrast, no effect was seen following exposure to the potent murine constitutive androstane receptor-ligand TCPOBOP and the hormonal murine constitutive androstane receptor-ligands androstenol and oestrone. Thus, the results support that 1) porcine CYP2A is reversibly inhibited by androgens on a transcriptional basis in vivo; 2) the induction profile of CYP2A in vitro shares similarity with that of human constitutive androstane receptor-regulated CYPs, indicating an involvement of a porcine constitutive androstane receptor in the regulation of CYP2A.
Materials and Methods
- Top of page
- Materials and Methods
Chemicals. Rapinovet® (Propofol 10 mg/ml, Schering-Plough Animal Health, Denmark), Anorfin® (buprenorphine 0.3 mg/ml, GEA, Denmark), Streptocillin® vet. (dihydrostreptomycin 250 mg/ml, benzylpenicillinprocaine 200,000 IU/ml, Boehringer Ingelheim, Germany), bovine serum albumin fraction V and trypan blue (Applichem, Germany), 5α-androsten-3α-ol (androstenol), 5-β-pregnane-3,20-dione (5β-PD) and 1,3,5(10)-oestratrien-3-ol-17-one (oestrone) from Steraloids, USA, 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP) and 6-(4-chlorophenyl)imidazo[2,1-b][1,3]-thiazole-5-carbaldehyde O-(3,4-dichlorobenzyl)oxime (CITCO) from Calbiochem, USA, RNeasy RNA isolation kit (Qiagen, USA). Gels for electrophoreses, First Strand cDNA Synthesis Kit, Taq DNA polymerase, silver staining kit and ECLTM Western blotting detection reagents were purchased from Amersham Biosciences, UK. Primers for PCR reactions were from Gibco, England, and Tag Copenhagen, Denmark. Primary antibodies were obtained from Gentest, USA, and secondary antibodies from Dako, Denmark. Heparin Leo 5000 IU/ml from Leo Pharma, Denmark. Collagenase, hydrocortisone, Hanks' Balanced Salts (HBSS) with and without phenol red, Williams' Medium E, hepes, nifedipine, coumarin, 7-OH-coumarin, sulfatase (19.1 U/ml) and β-glucuronidase (140 U/ml) were from Sigma-Aldrich, USA. All other chemicals were of analytical grade.
Protocol 1: Castration of minipigs. To detect changes in CYP2A and CYP3A expression and activity, liver samples were taken before and after castration of intact boars. Expression was measured on mRNA and protein levels, and activity levels of coumarin 7-hydroxylase (COH) and nifedipine 6-hydroxylase (NIF) were measured, reactions catalyzed by CYP2A and CYP3A respectively. To rule out possible local expression changes caused by inflammation following surgery, samples taken after castration were from both the previously resected lobe (left medial) as well as the right medial lobe.
Animals and surgical procedure. The experiment was performed using 6 Göttingen minipigs (Ellegaard, Denmark), all sexually mature intact males (age 5–18 months). The pigs were fed commercial chow and provided water ad libitum. They were fasted for 12 hr and anaesthetized with an initial intravenous dose of 5 mg propofol/kg body wt. General anaesthesia was maintained with intermittent intravenous injections of 0.5–1 mg propofol/kg body wt. to keep sufficient depth of anaesthesia. A cranial midline laparotomy was performed and the left medial liverlobe was localized. Hepatic resection was carried out using an electric scalpel, and haemostasis was achieved by applying 4 to 6 Guillotine sutures near the cut surface. Approximately 3 g of liver were removed and placed in liquid nitrogen. After closing the laparotomy, bilateral orchidectomy was performed, leaving the scrotal incisions open. Before finishing anaesthesia the pigs received an intramuscular analgesic (buprenorphine, 0.01 mg/kg body wt.) and antibiotics (streptocillin, 1 ml/10 kg body wt.). Streptocillin injection was repeated after 24 hr. The pigs were kept analgetized with repeated injections of buprenorphine for the first 24–48 hr. At day 10 after surgery the sutures were removed. No complications were observed during recovery. Approximately one month after surgery the pigs were anaesthetized using propofol in the dosage regimen described above and euthanized using a captive bolt pistol, and pieces of the previously operated liverlobe (left medial lobe) and the right medial lobe were removed and placed in liquid nitrogen.
Isolation of liver microsomes. Liver microsomes were isolated as earlier described (Skaanild & Friis 1999). Microsomal protein content was determined by a modified Lowry method (Peterson 1977) with bovine serum albumin as a standard.
P450 concentration. The microsome suspension (100 μl) was mixed with 1.9 ml of P450 buffer (0.1 M KH2PO4/Na2HPO4, pH 7.4, 0.5% Triton X-100, 20% glycerol, 1 nM EDTA, 1 mM dithiothreitol) and kept on ice. The sample was aerated with carbon monoxide for 30 sec., and baseline spectrum was measured spectrophotometrically from 400 to 500 nm. To the sample was added 50 μl sodium-dithionite solution (200 mg/ml), it rested for 2 min. and was measured again. The difference spectrum was used to calculate the P450 concentration according to Omura & Sato (1964).
Semiquantitative PCR. Isolation of mRNA, RT-PCR and semi-quantitative PCR were performed as described previously (Skaanild & Friis 1999). Duplex PCR reactions were set up as follows: CYP2A) CYP2A primers (25 pmol of each) and β-actin primers (7 pmol of each), CYP3A) CYP3A primers (30 pmol of each) and β-actin (20 pmol of each). The PCR products were quantified using Quantiscan software and the density of the specific products was divided by the density of the β-actin bands. The primer sequences were of human origin and are listed in table 1. Primers for CYP3A were designed from the human cDNA sequence for CYP3A4 (GeneBank no. NM 017460).
Table 1. Primers used for PCR analysis and product sizes.
|Gene||Primers||Product size (bp)||Ref.|
|β-actin||sense||234||(Hakkola et al. 1996)|
|CYP2A||sense||350||(Hakkola et al. 1996)|
| ||antisense |
|CYP3A||sense||349||GeneBank no. NM 017460|
Immunoblotting, coumarin 7-hydroxylase and nifedipine 6-hydroxylase assays were performed as described previously (Skaanild & Friis 1999).
Protocol 2: Primary porcine hepatocytes. To determine the effects of ligands and activators of murine constitutive androstane receptor and human constitutive androstane receptor on the expression of CYP2A and CYP3A, primary hepatocytes were isolated and in one series exposed to androstenol, TCPOBOP or phenobarbital and in another to CITCO, 5-β-pregnane-3,20-dione (5β-PD) or oestrone. Enzymatic activities of CYP2A and CYP3A were determined.
Isolation of primary hepatocytes and exposure to substances. Three female cross-bred pigs (Danish Landrace×Yorkshire×Duroc), body weight 19–31 kg, were used in each series of exposure. Pigs were fed commercial chow and provided water ad libitum. The pigs received 1 ml of heparin IV and were euthanized using a captive bolt pistol and exsanguinated. Hepatocytes were isolated as described by Olsen et al. (1997) using a two step collagenase perfusion modified from Monshouwer et al. (1996). The isolated hepatocytes were suspended in Williams' E supplemented with insulin (1 μM), hydrocortisone (1 μM), penicillin (1 μg/ml), gentamycin (50 μg/ml), glutamine (1.66 mM) and 4% fetal calf serum to a final concentration of 0.8 · 106 viable cells/ml (assessed by trypan blue exclusion). Hepatocytes were plated in volumes of 3 ml on 60.0/15 mm Petri dishes. The cells were allowed to attach for 3 hr, whereafter medium was changed and foetal calf serum was omitted. Exposure of the cells was initiated at this time by adding androstenol (1 or 10 μM), TCPOBOP (50 or 500 nM), or PB (0.15 or 1.5 mM) to the first series of cells and CITCO (1, 3 or 10 μM), 5-β-PD (10, 30 or 100 μM) or oestrone (1, 3 or 10 μM) to the second. In the first series, all hepatocytes received 0.1% (v/v) methanol as this was the solvent for androstenol and TCPOBOP. In the second series, all hepatocytes received the solvent dimethylsulphoxide to a concentration of 0.1% (v/v). Medium was changed daily.
At the start of exposure (0 hr) lactate dehydrogenase (LDH) leakage, coumarin 7-hydroxylase and nifedipine 6-hydroxylase activities were measured. Coumarin 7-hydroxylase and nifedipine 6-hydroxylase were determined again in all groups after 3, 24 and 72 hr of exposure.
Protein content. Dishes were washed twice with 37 ° phosphate buffer (0.125 M, pH 7.4), and harvested in 2 ml ice-cold phosphate buffer. The suspension was sonicated, and protein content was determined by a modified Lowry method (Peterson 1977) using bovine serum albumin as a standard.
Coumarin 7-hydroxylase activity. Dishes were washed with 37 ° HBSS, and the reaction was initiated by adding 3 ml coumarin/HBSS (100 μM) to dishes. After 30 min. of incubation the reaction was stopped by adding 3 ml methanol to the dishes, and cells and medium were harvested. The 7-OH coumarin was measured after enzymatic treatment with glucuronidase and sulfatase as the majority of this metabolite is present in the cells and medium in conjungated forms: The methanol was evaporated and 100 μl of the remainder was incubated with 100 μl buffer (0.1 M KPO4, pH 7), 20 μl glucuronidase and 5 μl sulfatase for 30 min. at 37 °. Samples were analyzed under the following HPLC-conditions: Column: Waters Nova-Pak, C18, 3.9 · 150 mm, particle size 4 μm; mobile phase: Solvent A: 5% CH3CN/1% concentrated CH3COOH/water containing 5 mM tetrabutylammonium phosphate (TBA), Solvent B: 50% CH3CN/1% concentrated CH3COOH/water containing 5 mM TBA. Flow rate 1.2 ml/min. The gradient program used was as follows: Initially 100% solvent A, ramp to 100% solvent B over 5 min., return to 100% solvent A at 7.5 min., total program time: 13 min. Injection volume: 100 μl. Detection: Ex.: 360 nm, Em.: 470.
Nifedipine 6-hydroxylase activity. Dishes were washed with 37 ° HBSS, and the reaction was initiated by adding 3 ml nifedipine/HBSS (200 μM) to dishes. After 10 min. of incubation the reaction was stopped by harvesting cells and transferring cells plus medium to 4 ml dichloromethane with 200 μl 1 M Na2CO3, 2 M NaCl (pH 10.5). Samples were extracted and analyzed under the HPLC-conditions referred in protocol 1.
Statistics. Statistics were performed in SAS® using the Mixed Procedure to detect effects of castration (protocol 1) or treatment (protocol 2). Pig was designated random effect. P-values ≤0.05 were considered statistically significant.
- Top of page
- Materials and Methods
To examine the effects of testicular products on CYP2A expression in minipigs, six mature minipig boars were castrated. Following castration, CYP2A mRNA, protein and enzyme activity increased significantly, indicating an inhibitory effect of androgens. Likewise CYP3A parameters were increased significantly, although not to the same extent as CYP2A. Regulation was transcriptional or at least pretranslational, as all levels of expression were increased in castrates. The values of CYP2A activity after castration (mean 213 pmol/mg prot/min.) did not reach the values previously reported in female minipigs (approximately 360 pmol/mg prot/min.) (Skaanild & Friis 1999). It is possible, that either imprinting events have taken place in the males permanently suppressing CYP2A expression to some extent, or that the presence of female sex steroids enhances CYP2A expression even further in females. This agrees with a possible involvement of constitutive androstane receptor in the regulation of CYP2A, as constitutive androstane receptor, at least in mice, is activated by oestrogens (Kawamoto et al. 2000) and deactivated by androgens (Forman et al. 1998).
Castration slightly increased the expression of CYP3A, which exhibits only minor gender dependence in minipigs with females possessing higher levels (Skaanild & Friis 1999). The mean nifedipine 6-hydroxylase activity in the males of 1.9 nmol/mg prot/min. corresponds well to the 2.2 nmol/mg prot/min. reported by Anzenbacher et al. (1998).
Total P450 levels ranged from 371 to 843 pmol/mg protein; a relatively large span although not unusually large for minipigs (Skaanild & Friis 1999).
In vitro CYP2A activity was clearly increased upon exposure to human constitutive androstane receptor-ligand CITCO and the general constitutive androstane receptor-activator phenobarbital after 72 hr, further supporting an involvement of a constitutive androstane receptor-ortholog in the regulation of CYP2A in pigs. Despite a 14 times increase in CYP2A activity upon phenobarbital-exposure after 72 hr, the exposed hepatocytes still exhibited less than 30% of the initial activity of controls at 0 hr. Thus even the strongest inducer tested was not able to restore CYP2A levels to starting level or above. Nevertheless, the 14 times induction is considerably stronger than reported double inductions of CYP2A6 by phenobarbital in primary human hepatocytes (Donato et al. 2000; Madan et al. 2003). In murine hepatocytes exposure to phenobarbital in similar concentrations induces coumarin 7-hydroxylase between 4 and 13 times after 48 hours of exposure (Salonpääet al. 1997; Posti et al. 1999; Donato et al. 2000). The human constitutive androstane receptor-ligands CITCO and 5β-pregnane-3,20-dione increased CYP2A activity slightly (1.3–1.4 times) regardless of concentration after 24 hr, whereas after 72 hr a more robust and concentration-dependent induction (2–3 times) was seen in CITCO-treated cells. For comparison, Maglich and coworkers (2003) reported 2- to 9 times increases in CYP2A6 mRNA levels in human hepatocytes exposed to 1 μM CITCO for 48 hr. Finally, potent murine constitutive androstane receptor-ligands androstenol, TCPOBOP and oestrone had no effect on porcine CYP2A activity, hence it appears that the porcine CYP2A-induction profile qualitatively shares more similarity with that of human CYP2A6 than that of murine Cyp2a's.
The observed decrease in CYP2A activity during the time of culture is in accordance with a previous study (Skaanild & Friis 2000). Although CYP2A expression is subject to some decline over time in primary hepatocytes from other species (Salonpääet al. 1994; Posti et al. 1999; Donato et al. 2000; Meunier et al. 2000), the reported decreases are not as pronounced as those observed in porcine hepatocytes. In primary murine hepatocytes, coumarin 7-hydroxylase has actually been reported to increase from 4 pmol/mg protein/min. to 7 pmol/mg protein/min. during the first 24 hr of culture (Salonpääet al. 1994), whereas others have observed a decrease from 4.9 pmol/mg protein/min. at 24 hr to 3.1 pmol/mg protein/min. at 48 hr (Donato et al. 2000). In primary human hepatocytes, a decrease in coumarin 7-hydroxylase from 0.6 nmol/mg microsomal protein/min. to approximately 0.2 nmol/mg microsomal protein/min. during 72 hr of incubation has been reported (Meunier et al. 2000); a decrease to one third. In contrast coumarin 7-hydroxylase activity of the porcine hepatocytes of the present study was reduced to less than half during the first 24 hr and after 72 hr less than 2% of the initial activity was left. Attempts to maintain the activity of CYP2A in porcine hepatocytes with growth hormone and autologous serum in the culture medium have proven without effect (Skaanild & Friis 2000). It thus appears that porcine hepatocytes have distinct requirements for constitutive expression of CYP2A in culture differing from primary hepatocytes from other traditionally used species. The decrease in CYP2A expression in culture obviously puts limitations to the use of this in vitro system and may also explain why androstenol had no effect, as it seems likely that the inhibitive action of this androgen would be negligible on an enzyme with expression rapidly decreasing towards zero anyway.
Several of the tested substances were expected to induce CYP3A activity, as they have been described to activate murine pregnane X receptor or human pregnane X receptor or both. Examples include androstenol, an activator of murine pregnane X receptor and human pregnane X receptor, and TCPOBOP, which activates hPXR, but not the murine ortholog (Moore et al. 2000). CYP3A was not induced by androstenol nor TCPOBOP, but induction of CYP3A was possible as demonstrated by the exposure to phenobarbital. The degree of induction of CYP3A by phenobarbital was between 3 to 4.5 times at 72 hr of exposure, which is in accordance with the results of Olsen et al. (1997), where phenobarbital-exposure resulted in an approximate 5 times induction of CYP3A activity. CYP3A was also slightly induced by 5β-pregnane-3,20-dione, an activator of both murine pregnane X receptor and human pregnane X receptorXR (Moore et al. 2000).
In conclusion 1) Testicular products inhibit CYP2A expression on a transcriptional basis in male minipigs, and castration partly reverses the inhibition. Furthermore, CYP3A expression is increased after castration. 2) CYP2A activity is increased in primary porcine hepatocytes exposed to activators of human constitutive androstane receptor phenobarbital and CITCO, indicating that constitutive androstane receptor may be a regulator of CYP2A and suggesting similarities between the regulation of pig and human CYP2As in vitro.