Age-related Differences in CYP3A Abundance and Activity in the Liver of the Göttingen Minipig
Abstract
In view of paediatric drug development, regulatory authorities often request safety studies in juvenile animals, including minipigs. Unfortunately, knowledge on the ontogeny of the biotransformation processes in animal models remains scarce and impedes a correct interpretation of the toxicity findings. CYP3A4 is one of the most important drug-metabolizing enzymes in human beings and shows important similarities with CYP3A in the minipig. Therefore, the aim of this study was to assess the abundance and activity of CYP3A in liver microsomes from foetal, juvenile (days 1, 3, 7 and 28) and adult male and female Göttingen minipigs. CYP3A abundance was studied by an indirect enzyme-linked immunosorbent assay (ELISA), whereas CYP3A activity was assessed by a biotransformation assay with Luciferin-IPA. CYP3A abundance could not be detected until day 3. From day 7 onwards, a gradual increase in expression was noted, leading to the highest abundance in adult animals. CYP3A activity was not detectable in foetuses and 1-day-old animals. The CYP3A activity was detectable, but below the LLOQ in day 3 animals and increased gradually with age to reach the highest level in adults. The CYP3cide and ketoconazole inhibition, and testosterone and midazolam reduction of Luciferin-IPA metabolism in minipig liver microsomes substantiate that Luciferin-IPA is metabolized by CYP3A in minipigs. A positive correlation was found between CYP3A abundance and biotransformation of Luciferin-IPA (Pearson r = 0.863; p < 0.0001). In conclusion, both abundance and activity of CYP3A increased gradually in juvenile minipigs, but remained below the levels observed in adult animals.
Children form a substantial and vulnerable group in the human population with regard to the use of medicines. Unfortunately, the majority of drugs on the market have not been tested nor evaluated in the paediatric age group, despite known differences in drug safety between mature and immature bodies 1, 2. The introduction of the Paediatric Regulation (EC) No 1901/2006 in December 2006 has drastically changed the efforts of pharmaceutical companies concerning paediatric drug development. This has also resulted in regulatory requests for preclinical studies in juvenile animals prior to the start of clinical trials in young patients. Unfortunately, the neonatal and young age groups remain a challenging population as important developmental changes occur in these individuals 3. Maturation of metabolizing enzymes (Phase I and II), changes in body composition (water/lipid partition), liver development and maturation of kidneys are some of the factors that impede the insight into the absorption, distribution, metabolism and excretion of drugs in children 4.
Cytochrome P450 enzymes (CYP) are important drug-metabolizing enzymes, which facilitate the excretion of drugs from the body. Unfortunately, CYP enzymes can also be responsible for adverse drug reactions and drug–drug interactions. Maturational differences in CYP activity can even increase the risk of undesired effects and under- or overdosing of drugs in the paediatric age group 2. From one year of age onwards, CYP3A4 is the most abundant member of the CYP3A subfamily in the human liver 2, 5, 6 and is responsible for the oxidative metabolism of 30–50% of human drugs on the market 7-9. CYP3A5 and CYP3A7 are also members of the human CYP3A subfamily. CYP3A7 is considered the foetal isoform, although it can be present in adult livers too. CYP3A5 shows no clear age-related expression pattern, is polymorphically expressed and has overlapping substrate specificities with CYP3A4 6, 9-11. CYP3A7 and CYP3A5 show 88% and 83% identity of amino acids with CYP3A4, respectively 5. In view of the use of minipigs in juvenile toxicity studies, it is also important to know the developmental expression pattern of CYP3A in this species. Four CYP3A isoforms have been identified in the porcine species, showing at least 75% identity of amino acid sequences compared to human CYP3A4, that is CYP3A22, CYP3A29, CYP3A39 and CYP3A46. CYP3A22 and CYP3A29 are considered CYP3A4 homologues in Göttingen minipigs, although the presence of CYP3A39 and CYP3A46 in the Göttingen minipig cannot be excluded 12. The CYP3A enzymes in minipigs present even similar biotransformation properties as the human homologues, for example nifedipine oxidation and testosterone 6β-hydroxylation 13-16. In addition, pigs and human beings share anatomical and physiological characteristics, especially regarding the digestive and cardiovascular systems, which explains the high interest for this species as pharmacological and toxicological model 17.
The aim of this study was to assess the abundance and activity of CYP3A in liver microsomes from foetal, neonatal, juvenile and adult Göttingen minipigs. Abundance of CYP3A was evaluated with an indirect enzyme-linked immunosorbent assay (ELISA) for quantification of human CYP3A4, whereas activity was estimated by incubation of minipig liver microsomes with Luciferin-IPA, a highly specific substrate for human CYP3A4 18. We evaluated the inhibitory effect of CYP3cide and ketoconazole and the interaction of the CYP3A substrates testosterone and midazolam on the metabolism of Luciferin-IPA.
Materials and Methods
Liver samples
Livers were obtained from healthy Göttingen minipigs. No CYP induction was performed. Ten pregnant sows were a kind gift from Ellegaard Göttingen minipig A/S (Dalmose, Denmark). Janssen Research (Beerse, Belgium) kindly provided liver samples from four adult, male Göttingen minipigs.
The following age groups were investigated: 84–86 days of gestation (n = 8), 108 days of gestation (n = 8), day 1 (within 24 hr after birth) (n = 8), day 3 (n = 8), day 7 (n = 9), day 28 (n = 10) and adult (n = 9). As normal gestation length in the minipig is 112 to 115 days, the evaluation of the foetal age groups is limited to the third trimester of foetal development. Post-natal day 28 is considered the weaning age in the Göttingen minipig. This age range was chosen to cover the first year of life in children, as important changes in CYP3A4 expression and activity occur in this period 5, 11. Both genders were equally represented in each age group, except in group day 7 (males: n = 5, females: n = 4) and the adult age group (males: n = 4, females: n = 5). Liver samples from five of ten adult sows were randomly selected in this study to obtain similar group sizes. The age of the adult males and females ranged between 18–24 months and 14–33 months, respectively. The adult sows were killed by electrical stunning, followed by exsanguination either before or after delivery, according to the desired age of their offspring. The foetuses were harvested and placed immediately on ice until further processing. The neonatal and juvenile piglets were naturally delivered and housed with the sow until euthanasia. The piglets were randomly allocated to a specific post-natal age group [day 1 (PND1), day 3 (PND3), day 7 (PND7), day 28 (PND28)]. Due to practical reasons, the piglets were not killed by electrical stunning, but they were anaesthetized by an intraperitoneal injection of sodium pentobarbital 20% (Kela NV, Hoogstraten, Belgium) (90 mg/kg), followed by exsanguination. The liver was dissected and rinsed with ice-cold 0.01 M phosphate-buffered saline (pH 7.4). Samples were taken from the lateral liver lobes and immediately snap-frozen in liquid nitrogen. These samples were stored at −80°C until the isolation of liver microsomes. The Ethical Committee of Animal Experimentation from the University of Antwerp (Belgium) approved the protocol and use of the animals.
Isolation of liver microsomes
Liver tissue was thawed on ice and washed with ice-cold homogenizing buffer (0.01 M potassium phosphate (KPO4) buffer (451201; Corning Incorporated, Corning, NY, USA) containing 1.15% potassium chloride). Excess of moisture was removed by blotting the tissue on paper towels. The liver tissue was minced into small pieces by means of surgical scissors and weighed. For each gram of tissue, a threefold volume in millilitre of ice-cold homogenizing buffer was added. The tissue was homogenized with the Polytron® System PT 1200 E (230 V, 50 Hz) on ice for maximum 10 sec. As a final homogenization step, a motor-driven Potter Elvehjem with Teflon pestle was used (15 g, 5 to 10 up-and-down strokes). All homogenization steps were performed on ice. The homogenate was centrifuged at 12 000 × g for 20 min. at 4°C. The resulting supernatant (S9-fraction) was centrifuged at 100 000 × g for 60 min. at 4°C. The resulting pellet was resuspended with homogenizing buffer and centrifuged at 100 000 × g for 40 min. at 4°C. The resulting microsomal pellet was resuspended in storage buffer [0.1 M KPO4 buffer containing 250 mM sucrose and Halt™ Protease Inhibitor Single-Use Cocktail (Thermo Fisher Scientific, Rockford, IL, USA)] and stored at −80°C until use. For each mL S9-fraction that was centrifuged, 150 μL of storage buffer was added to the microsomal pellet and homogenized. Total protein concentration was determined by the Pierce® BCA Protein Assay Kit with bovine serum albumin as a standard (Thermo Fisher Scientific, Rockford).
ELISA
An indirect ELISA using commercially available antibodies was used to determine CYP3A abundance in the liver microsomes from the Göttingen minipigs. De Bock et al. 19 developed this ELISA for the detection of CYP3A4 in HLM. The ELISA was performed according to the instructions described by De Bock et al. (2012). Microsomal samples, calibrators and validation samples were diluted to a final concentration of 10 μg of microsomal protein/mL using a carbonate–bicarbonate buffer (pH 9.4). In addition to the validation samples, CYP3A4 Baculosomes® Plus Reagent, rHuman (P2377; Life Technologies, Thermo Fisher Scientific, Rockford) and HLM (HMMC-H3A4-PL040; Life Technologies, Thermo Fisher Scientific, Rockford) were evaluated as they were also included in the biotransformation assays with Luciferin-IPA. From each diluted sample and calibrator, 100 μL was loaded on a black 96-well MaxiSorp® microtiter plate (Nunc, Roskilde, Denmark), resulting in 1 μg of microsomal protein per well. The primary antibody, which binds to the antigen to be detected, was a CYP3A4-purified MaxPab rabbit polyclonal antibody, raised against the full-length human CYP3A4 protein (H00001576-D01P; Abnova, Taipei, Taiwan). The Pierce®Goat anti-rabbit horseradish peroxidase conjugated IgG (31460; Pierce Biotechnology, Rockford) was used as secondary antibody. Peroxidase activity was detected with the QuantaBlu™ Fluorogenic peroxidase substrate kit (Thermo Scientific, Rockford), which is a fluorogenic substrate. Fluorescence was determined with the Ascent Fluoroscan (Thermo Scientific, Rockford) at excitation and emission wavelengths of 320 and 405 nm, respectively. This ELISA was developed to quantify amounts between 2 and 300 pmol CYP3A4/mg microsomal protein (MP) in HLM with a 5-parameter logistics function with 1/x weighting factor. The ELISA was validated for sensitivity, accuracy, precision, working range and calibration 19. The data represent the mean value of two technical replicates for each sample.
Incubation conditions for Luciferin-IPA in minipig liver microsomes
The kinetic studies were performed in non-treated Nunc™ F96 Microwell™ white Polystyrene plates (236205; Thermo Fisher Scientific, Rockford). A range of seven protein concentrations of liver microsomes from an adult female Göttingen minipig was tested for linearity (0.125–8 μg/50 μL). The kinetic profile for Luciferin-IPA (P450-Glo™ CYP3A4 Assay, V9001; Promega Corporation, Madison, WI, USA) was generated from eight substrate concentration points (0.3125–40 μM). The estimated Km was 3.66 ± 0.71 μM. To prevent substrate inhibition in the lowest age groups and to favour biotransformation of Luciferin-IPA by the porcine ‘CYP3A4 homologue’, a concentration of 1 μM Luciferin-IPA, which is below the estimated Km, was chosen as final substrate concentration in the incubations. The incubation time (10 min.) and microsomal protein concentration (20 μg/mL) were within the linear range for Luciferin-IPA.
In each incubation well, 1 μg of hepatic microsomal protein, 0.1 M KPO4 buffer (pH 7.4), 1.3 mM NADP+, 3.3 mM glucose-6-phosphate, 0.4 U/mL glucose-6-phosphate dehydrogenase, 3.3 mM magnesium chloride (451 220 and 451 200; Corning Incorporated) and 1 μM Luciferin-IPA were co-incubated. The total incubation volume was 50 μL per well. Microsomal dilutions in 0.1 M KPO4 buffer were pre-incubated at room temperature (RT) for 10 min. The Luciferin-IPA and the NADPH-regenerating system in KPO4 buffer were pre-incubated at RT for 10 min. The reaction was initiated by the addition of the microsomal protein to the remainder of the incubation mixture. The plate was first incubated for 10 min. at RT. Subsequently, 50 μL of Luciferin detection reagent with esterase (V859A and V144A; Promega Corporation) was added to each well, mixed and incubated for 20 min. at RT to stabilize the luminescent signal. Luminescence was measured with a Tecan Genios (Tecan Group Ltd., Männedorf, Switzerland). The concentration of the metabolite D-Luciferin, generated by CYP3A from Luciferin-IPA, was quantified by comparing luminescence from the incubation mixtures to that from a D-Luciferin standard curve (Beetle Luciferin, Potassium Salt, E1601; Promega Corporation). Reaction velocities were calculated in units of picomoles of D-Luciferin formed per minute per milligram of microsomal protein (pmol/min/mg MP). CYP3A4 Baculosomes® Plus Reagent, rHuman (P2377; Life Technologies, Thermo Fisher Scientific, MA) and HLM (HMMC-H3A4-PL040; Life Technologies, Thermo Fisher Scientific, MA) were used as a positive control. Insect cell control supersomes (456201; Corning Incorporated), lacking CYP450 enzymes, were used as a minus-P450 control. Positive and negative controls were included in each well plate and similarly treated to the minipig liver microsomes. Results from the insect cell control supersomes were subtracted from the values obtained for the minipig liver microsomes, CYP3A4 Baculosomes® and human liver microsomes. The lower limit of detection (LLOD) and the lower limit of quantification (LLOQ) were 1.18 and 3.91 nM, respectively. The data represent the mean value for each sample obtained in three separate assays, with two technical replicates within each assay.
Inhibition of biotransformation of Luciferin-IPA with CYP3cide
Ten micrograms of microsomal protein from an adult Göttingen minipig sow and from HLM (HMMC-H3A4-PL040) was pre-incubated with a range of 8 CYP3cide 20 concentrations (0–1 μM) (PZ0195; Sigma-Aldrich, St. Louis, MO, USA) and with 1.3 mM NADP+, 3.3 mM glucose-6-phosphate, 0.4 U/mL glucose-6-phosphate dehydrogenase, 3.3 mM magnesium chloride in 0.1 M KPO4 buffer at 37°C for 10 min. The total incubation volume was 500 μL. Then, 10 μL of a 51 μM Luciferin-IPA solution was added to the incubation mixture, resulting in a 1 μM Luciferin-IPA concentration in the final incubation mixture. After 10 min. of incubation at 37°C, 50 μL of the final incubation mixture was added to 50 μL of Luciferin detection reagent with esterase in a non-treated Nunc™ F96 Microwell™ white Polystyrene plate to stabilize the signal for 20 min at RT. The continuation of the procedure was similar to the description in the previous paragraph. The reaction velocities with co-addition of CYP3cide were expressed as a percentage ratio of the control velocity with no inhibitor present.
Co-incubation of Luciferin-IPA with testosterone, midazolam or ketoconazole
Five micrograms of microsomal protein from an adult Göttingen minipig sow and from HLM (HMMC-H3A4-PL040) was co-incubated with a range of 7 testosterone concentrations (0–50 μM) (T037; Sigma-Aldrich), a range of 10 midazolam concentrations (0–50 μM) (Dormicum; Roche Holding AG, Basel, Switzerland) or a range of 6 ketoconazole concentrations (0–2 μM) (K1003; Sigma-Aldrich) together with 1.3 mM NADP+, 3.3 mM glucose-6-phosphate, 0.4 U/mL glucose-6-phosphate dehydrogenase, 3.3 mM magnesium chloride and 1 μM Luciferin-IPA in 0.1 M KPO4 buffer at 37°C for 10 min. Reactions were initiated by the addition of the microsomes. The total incubation volume was 250 μL. After 10 min. of incubation at 37°C, 50 μL of the final incubation mixture was added to 50 μL of Luciferin detection reagent with esterase in a non-treated Nunc™ F96 Microwell™ white Polystyrene plate to stabilize the signal for 20 min. at RT. The continuation of the procedure was similar to the description in the previous paragraph. The reaction velocities with co-addition of TST, MDZ or KCZ were expressed as a percentage ratio of the control velocity with no additional substrate or inhibitor present.
Mathematical and statistical analyses
For the ELISA test, calibration curves were fitted and data were analysed using the Masterplex® Readerfit 2010 software (Hitachi, San Francisco, CA, USA). The Kruskal–Wallis (mean rank) test was used to detect age- and gender-related differences in CYP3A abundance for the PND28 and adult animals. A p-value < 0.05 was considered statistically significant. Estimation of Km and Vmax for CYP3A and Luciferin-IPA in the minipig liver microsomes was performed by a nonlinear regression analysis in GraphPad Prism version 6.0 f (GraphPad Software, Inc., La Jolla, CA, USA). Estimation of IC50 values was performed by a nonlinear regression analysis with a four-parameter logistic curve in GraphPad Prism version 6.0 f (GraphPad Software, Inc.). Calculation of velocities of formation of D-Luciferin was performed in Microsoft Excel® (version 14.3.1; Microsoft Corporation, Redmond, WA, USA). The Kruskal–Wallis (mean rank) test was used to statistically analyse the results from the biotransformation assay with Luciferin-IPA and differences between age groups (PND7, PND28 and adult) and genders. Bonferroni correction for pairwise comparisons between age groups adjusted the p-value to 0.025. Consequently, a p-value < 0.025 was considered statistically significant. Prior to the Kruskal–Wallis test, homogeneity of variances was tested with a nonparametric Levene's test. Statistical analyses were performed with IBM SPSS statistics (version 20; IBM, Armonk, NY, USA). Correlation between CYP3A abundance and formation of D-Luciferin was calculated with a Pearson correlation coefficient on the Ln-transformed data from PND 7, PND 28 and adult animals in Graphpad Prism version 6.0 f (GraphPad Software, Inc.).
Results
ELISA
The results represent a semiquantitative measurement of CYP3A in the liver of the Göttingen minipig and should be considered relative values based on the signal to human CYP3A4. All foetal, PND1 and PND3 samples were below the LLOQ of 2 pmol CYP3A/mg of MP. In samples of PND7 piglets, six of nine samples were still below the LLOQ. The abundance of the other three samples ranged between 3.53 and 5.11 pmol CYP3A/mg of MP. At PND28, all liver samples had levels ranging between 7.19 and 35.7 pmol CYP3A/mg of microsomal MP. For the adult animals, levels ranged between 25 and 101 pmol CYP3A/mg of MP (table 1), being significantly higher than in PND28 animals (p = 0.001). No gender-related differences were observed among PND 28 and adult animals (p = 0.935) (fig. 1). The HLM showed an abundance of 87.4 pmol CYP3A4/mg of MP. The CYP3A4 Baculosomes® showed an abundance of 140 pmol CYP3A4/mg of MP.
| Age | Gender | Number of animals | CYP3A abundance Mean ± S.D. |
|---|---|---|---|
| Day 7 | Female | N = 4 | Below LLOQ |
| Male | N = 3 | 4.07 ± 0.9 | |
| N = 2 | Below LLOQ | ||
| Day 28 | Female | N = 5 | 19 ± 12.1 |
| Male | N = 5 | 20.8 ± 11.3 | |
| Total | N = 10 | 19.9 ± 11.1 | |
| Adult | Female | N = 5 | 74.1 ± 29.5 |
| Male | N = 4 | 66.5 ± 17.3 | |
| Total | N = 9 | 70.7 ± 23.8 |
- Results from ELISA are presented as mean ± standard deviation (±S.D.) and are expressed in ‘pmol/mg of microsomal protein’. These results are relative values, based on the signal to human CYP3A4. All foetal (84–86 and 108 days of gestation), PND1 and PND3 samples were below the LLOQ with four females and four males in each age group.
- LLOQ, lower limit of quantification.
Incubations with Luciferin-IPA
A statistically significant difference in CYP activity was observed between the different age groups by the Kruskal–Wallis test (p < 0.0001). No appreciable metabolism of Luciferin-IPA was observed at 84–86 days and 108 days of gestation (fig. 2). These foetal piglets had reached about 75% and 95% of gestation, respectively. Also for PND1 animals, no formation of D-Luciferin was detected. At PND3, D-Luciferin was formed at a level above the LLOD for two of eight animals, but levels were still below the LLOQ. With one exception, all PND7 animals showed levels above the LLOD, and five animals showed levels above the LLOQ (mean ± S.D.: 31.9 ± 9.38 pmol/min/mg MP). At PND28, the rate of metabolism (93.7 ± 30.6 pmol/min/mg MP) was increased compared to the younger groups, with the exception of one animal that showed a value between the LLOD and LLOQ The highest velocity was noted in the adult animals (264 ± 76.3 pmol/min/mg MP). No gender-related differences among age groups (PND7, PND28 and adult) were observed (p = 0.356). The HLM and the CYP3A4 Baculosomes® showed a velocity of 26.3 and 178 pmol/min/mg MP, respectively.
Inhibition of biotransformation of Luciferin-IPA with CYP3cide
CYP3cide inhibited biotransformation of Luciferin-IPA in both HLM and minipig liver microsomes from an adult sow. Metabolism of Luciferin-IPA in human liver microsomes was already inhibited for 90% at 0.03125 μM CYP3cide, whereas this concentration only caused 7.4% inhibition in minipig liver microsomes. However, 0.125 and 1 μM CYP3cide caused 47% and 92% inhibition in minipig liver microsomes, respectively (fig. 3).
Co-incubation of Luciferin-IPA with testosterone, midazolam or ketoconazole
Testosterone clearly showed a dose-related inhibition of D-Luciferin formation in the minipig liver microsomes, which was less pronounced in the HLM, whereas midazolam and ketoconazole resulted in an inhibition of the formation of D-Luciferin in both minipig liver microsomes and HLM (fig. 4).
Correlation between CYP3A abundance and activity
A positive correlation was found between CYP3A abundance and formation of D-Luciferin for the PND7, PND28 and adult minipigs (Pearson r = 0.863, p < 0.0001) (fig. 5).
Discussion
To gain insight into the ontogeny of CYP3A in the liver of the Göttingen minipig, we investigated the protein levels of CYP3A by means of an ELISA, which was originally developed to quantify human CYP3A4 19. As cross-reactivity of the primary antibody with CYP3A5 and CYP3A7 was not evaluated, we consider these results as abundances of CYP3A isoforms in general. No evidence exists that porcine CYP3A and human CYP3A4 bind in an identical way with the primary antibody in this ELISA. Therefore, the obtained data are semiquantitative and show abundances relative to the abundance of CYP3A in human beings. Nevertheless, the amino acid sequences of CYP3A22 and CYP3A29 in Göttingen minipigs show 75% sequence identity compared to human CYP3A4, so binding of the primary antibody with porcine CYP3A can be expected 12. This ELISA revealed a very low or absent CYP3A expression until 7 days of age. From day 7 onwards, a gradual increase in CYP3A abundance was noted with a significantly higher CYP3A expression in the adult animals compared to the younger animals. This post-natal increase in CYP3A expression in Göttingen minipig livers is in agreement with data from a previous immunohistochemical study performed by our group 21 and with data on CYP3A29 mRNA expression in the liver from Bama miniature pigs, showing very low expression in neonatal animals and increasing expression with age 22. Additionally, it corresponds to human data for CYP3A4. Stevens et al. noted low CYP3A4 expression levels in HLM, with a gradually increasing expression during the first six months of life. In the age group of 5 to 15 years, they found still lower CYP3A4 levels compared to these in adults 6. This age-related onset of CYP3A4 protein is in accordance with similar observations for CYP3A4 mRNA expression 5, 23, 24. Indeed, CYP3A4 levels start to increase after birth and reach adult levels from one year of age onwards, whereas several studies suggest that the ‘foetal’ CYP3A7 isoform remains the dominant CYP3A enzyme until one year of age 2. An age-related pattern, similar to human CYP3A4, was observed in our study, which suggests that our polyclonal antibody did not cross-react with a CYP3A7 homologue. The absence of a CYP3A7 homologue in the Göttingen minipig is unlikely as Hermann et al. detected some CYP3A7-like enzyme in foetuses at 100 days of gestation by Western blotting. They also found a nine times higher CYP3A29 mRNA expression in the day 1 animals compared to the foetuses 25.
We assessed the CYP3A activity by the incubation of minipig liver microsomes with Luciferin-IPA. In human beings, Luciferin-IPA is considered a highly specific substrate for CYP3A, with a 102 times higher enzymatic efficiency for CYP3A4 compared with CYP3A7 and a nine times higher enzymatic efficiency for CYP3A4 compared with CYP3A5 18, 26. To promote binding of Luciferin-IPA to the minipig ‘CYP3A4 homologue’ and to prevent substrate inhibition in the lowest age groups, the concentration of Luciferin-IPA in the incubation mixtures (1 μM) was chosen below the estimated Km (3.659 μM). We showed that Luciferin-IPA was extensively metabolized by the minipig liver microsomes, supporting CYP3A activity. Additionally, pre-incubation of minipig liver microsomes with CYP3cide, which is a mechanism-based inactivator of CYP3A4, inhibited the metabolism of Luciferin-IPA. This result was further substantiated by the inhibition of D-Luciferin formation by testosterone, midazolam and ketoconazole. Ketoconazole is a typical CYP3A inhibitor and is a potent inhibitor of Luciferin-IPA biotransformation in human hepatocytes 26. In pigs, ketoconazole is a potent inhibitor of CYP3A activity, too 16, 27, 28. Also in our study, ketoconazole clearly inhibited the metabolism of Luciferin-IPA in both HLM and minipig liver microsomes. Testosterone and midazolam are two commonly used CYP3A4 substrates in human beings, and also in pigs, they are metabolized by CYP3A 13-16, 28-30. Co-incubation of Luciferin-IPA with testosterone or midazolam resulted in a reduction of the biotransformation of Luciferin-IPA in the minipig liver microsomes. In contrast, the tested range of concentrations of testosterone did not extensively inhibit the metabolism of Luciferin-IPA in the HLM. Wang et al. studied drug–drug interactions of CYP3A substrates with CYP3A4 in HLM. In their study, testosterone was not able to inhibit nifedipine oxidation and inhibited only partially midazolam 1′-hydroxylation. Conversely, nifedipine and midazolam were able to inhibit the 6β-hydroxylation of testosterone 31. Their study showed that the effect of interaction of two CYP3A substrates with CYP3A is substrate dependent. Our results indicate that the effect of testosterone on Luciferin-IPA metabolism is less pronounced in the HLM than in the minipig liver microsomes, although testosterone is a well-characterized CYP3A4 substrate in human beings.
The results from the biotransformation of Luciferin-IPA among different age groups reflected the results obtained by ELISA. This age-related pattern was observed for the metabolism of midazolam in male Camborough-29 pigs as well. The midazolam 1′-hydroxylation was very low at day 1, was significantly raised at two weeks of age and remained at a similar lever until five weeks of age. From five weeks onwards, a linear increase was present till 20 weeks of age 30. These activity data are also in agreement with the age-related trend in human beings. Lacroix et al. found extremely weak levels of CYP3A4 activity (6β-hydroxylation of testosterone) in the foetal liver, which increased to 30–40% and 100% of adult levels at one month and one year of age, respectively 5. Blanco et al. 32 studied midazolam as a substrate for CYP3A4/3A5 in hepatic liver microsomes in age groups ranging from 0.5 to 93 years of age, but could not detect age-related differences. These data from Blanco et al. and Lacroix et al. suggest that CYP3A4 approximates adult levels already by one year of age. This contrasts, however, to data from Stevens et al., who showed that CYP3A4 activity in children between 5 and 15 years was not yet at adult levels 6. High interindividual variation in CYP3A4 activity, as described for adults, may also be responsible for differences in the paediatric population 8, 10, 33.
Regarding gender differences, the mean velocity of Luciferin-IPA metabolism was 42% higher in female than in male adult minipigs, although not statistically significant (p = 0.05). As the number of animals was limited to five and four animals per gender, a larger sample size may elucidate whether a gender difference is present or not. Skaanild and Friis 34 found a weak but statistically significantly higher CYP3A expression in female Göttingen minipigs than in males at 3.5 to 4 months of age. In man, gender differences in CYP3A4 activity have also been described in vitro and in vivo, with a higher CYP3A4 activity in females compared to males 35-39. Whether this difference in CYP3A activity is clinically relevant is questionable, as Chen et al. 38 found only a minor difference in AUC in contrast to significantly higher CYP3A activity in women. A clear explanation for these gender-related differences has not yet been found. A differential effect of sex steroid hormones may be possible 37-39. However, data are conflicting in human beings. Shimada et al. found no gender-related differences in CYP3A abundance and activity in the liver microsomes from 30 Japanese and Caucasians, neither did Snawder et al. in 40 human liver samples 7, 8.
We found a positive correlation between CYP3A abundance and biotransformation of Luciferin-IPA for the PND7, PND28 and the adult animals (r = 0.863; p < 0.0001). Skaanild and Friis also found a positive correlation between CYP3A abundance and activity in conventional pigs (r2 = 0.63; p < 0.0001) 34. A positive correlation between CYP3A4 abundance and activity has also been described in human beings for testosterone 6β-hydroxylation (r2 = 0.724) and for bufalin 5β-hydroxylation (r = 0.943) 8, 9.
Comparison of the results for CYP3A abundance and activity between the HLM and the adult minipig liver microsomes in this study showed a much higher activity in the adult minipig liver microsomes than in the HLM, although CYP3A abundance was at a comparable level. As the primary antibody in the ELISA was directed against the full length of the human CYP3A4, the abundance of porcine CYP3A may have been underestimated compared to the HLM, which may explain the higher activity in the adult minipig liver microsomes despite similar abundances. However, the biotransformation of Luciferin-IPA, the inhibitory potential of CYP3cide and ketoconazole and the reduction of Luciferin-IPA biotransformation by testosterone or midazolam in this study support the hypothesis that porcine CYP3A homologues are responsible for the metabolism of Luciferin-IPA.
To summarize, CYP3A abundance and activity in the liver of the Göttingen minipig clearly show a post-natal maturational pattern, which is in accordance with data for CYP3A4 in human beings. Based on these results, we consider the juvenile Göttingen minipig to be a good model for studies involving CYP3A substrates.
Acknowledgements
The authors would like to thank Ellegaard Göttingen Minipig A/S and Janssen Research for the kind donation of animals and tissue samples. The authors are members of COST Action BM1308 ‘Sharing Advances on Large Animal Models (SALAAM)’.
Conflict of Interest
The authors declare no conflict of interest.
