Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201 www.elsevier.com/locate/cbpc

Tissue distribution of 3A (CYP3A) in brushtail possums (Trichosurus vulpecula) exposed to Eucalyptus terpenes ⁎ Jennifer S. Sorensen a, Kristian C. Forbey b, Robert L. Tanquay c, Bernie McLeod d,

a NPS Pharmaceuticals, 383 Colorow Dr., Salt Lake City, UT 84108, USA b Myriad Genetics, 320 Wakara Way, Salt Lake City, UT 84108, USA c Department of Environmental and Molecular Toxicology, Oregon State University, Corvallis, OR 97331, USA d AgResearch Invermay, PB 50034, Mosgiel, New Zealand Received 27 August 2006; received in revised form 17 November 2006; accepted 30 November 2006 Available online 10 January 2007

Abstract

We evaluated the distribution pattern of a specific xenobiotic metabolizing , cytochrome P450 3A (CYP3A) in the common brushtail possum (Trichosurus vulpecula). Western blot studies using CYP3A antibodies were used to compare CYP3A levels in the intestine, liver, kidney, brain, testes and adrenal gland in possums fed diets with and without a mixture of terpenes. Possums appear to produce at least 3 different CYP3A- like isoforms that are differentially expressed in various tissues. The liver and duodenum produce all three isoforms (CYP3A P1, P2, P3), the jejunum only produces CYP3A P1, the ileum, kidney, testes and adrenal only produce CYP3A P2 and the brain only produces CYP3A P3. Terpene treatment did not alter relative levels of isoforms present in any tissue type. This study is the first to identify the presence and differential expression of several CYP3A-like isoforms in a variety of tissues of a wild mammalian herbivore. Data suggest that CYP3A-like are not induced by terpenes. However, the wide distribution of CYP3A-like isoforms in a variety of tissues, suggests that these enzymes are an important mechanism for metabolism in possums and may contribute to the high tolerance possums have to a wide range of xenobiotics. © 2007 Elsevier Inc. All rights reserved.

Keywords: Brushtail possum (Trichosurus vulpecula); Cytochrome P450 3A (CYP3A); Terpenes; Tissue distribution

1. Introduction CYP3A is an important metabolizer of PSMs in a variety of species. Sheep and hamsters rely on CYP3A for The cytochrome P450s are a group of enzymes responsible the metabolism of the pyrrolizidine alkaloid, senecionine for the oxidative metabolism of a wide range of xenobiotics in (Huan et al., 1998). The terpenes 1–8 cineole, 1–4 cineole, α mammals. Of the four CYP families (CYP1, CYP2, CYP3 and β-thujones, β-myrcene and citral found in a variety of plant and CYP4), the CYP3A subfamily is the most abundant and oils are metabolized by CYP3A in human and rat liver considered to be the most important xenobiotic metabolizing microsomes (De-Oliveira et al., 1997; Hiroi et al., 1995; enzyme in mammals. CYP3A is often directly responsible for Miyazawa et al., 2001a,b; Roffey et al., 1990). variation in drug tolerance in humans and laboratory species Several studies suggest that CYP3A is also a plausible (Lamda et al., 2002). Likewise, CYP3A may explain variability mechanism for the metabolism of terpenes in common brushtail in tolerance to plant secondary metabolites (PSM) in herbivores. possums (Trichosurus vulpecula). For example, inhibition of 1– The CYPs are a well-supported counter-defense mechanism 8 cineole hydroxylation by the CYP3A inhibitor, ketoconozole, in employed by herbivores to metabolize, and subsequently brushtail possum liver microsomes, suggests that CYP3A is likely eliminate, ingested plant secondary metabolites. Specifically, involved in 1–8 cineole metabolism in possums (Pass et al., 2001, 2002). Moreover, CYP3A enzymes have been identified in the liver of possums (Ho et al., 1998). Although CYP3A may play a ⁎ Corresponding author. role in PSM metabolism, several studies suggest that possums may E-mail address: [email protected] (B. McLeod). have lower reliance on CYP3A than other species. For example,

1532-0456/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpc.2006.11.016 J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201 195 the metabolism of midazolam, a major CYP3A substrate, was primary antibodies, rabbit anti-human cytochrome P450 3A4 lower in possum livers than human liver (Ho et al., 1998) and the polyclonal antibody and mouse anti-human β-actin (internal activity of androstenedoine 6 β-hydroxylase, a marker for control) were purchased from GeneTex, Inc, San Antonio, TX, CYP3A, was lower in possum liver than in rat liver (Pass et al., USA. The secondary antibodies, goat anti-rabbit IgG-HRP for 1999). CYP3A and goat anti-mouse for β-actin were purchased from To better understand the role of CYP3A in the metabolism of Southern Biotechnology Associates, Inc., Birmingham, AL, PSMs in the common brushtail possum, we investigated the USA. Protease Inhibitor Cocktail Set III and IV were purchased distribution and relative levels of CYP3A isoforms in the liver from CalBiochem, San Diego, CA, USA. Mouse liver and in non-hepatic tissues, including the intestine, kidney, brain, was a generous gift from Dr. David Williams. RNAlater testes and adrenal of this mammalian herbivore. Specifically, we Solution for tissue preservation and PARIS kit for parallel compared levels of CYP3A-like isoforms in the possum with isolation of and RNA were purchased from Ambion, and without oral exposure to a mixture of naturally consumed Austin, TX, USA. Proteins were separated by electrophoresis terpenes, namely: 1,8-cineole, p-cymene, α-pinene, and limo- on 7.5% Tris–HCl-polyacrylamide Ready Precast Gels from nene. The extra-hepatic tissues were investigated to assess their Bio-Rad Laboratories, Hercules, CA, USA. An ECL-plus importance in the metabolism and elimination of terpene and/or chemiluminescent detection kit from Pierce Biotechnology, other xenobiotics in possums. For example, the intestine is the Inc., Rockford, IL, USA, facilitated visualization of proteins. first barrier tissue for most ingested xenobiotics and harbors a Halothane (Halothane-VCA) for possum anesthesia was variety of xenobiotic metabolizing enzymes (Ilett et al., 1993; purchased from Phoenix Pharmaceuticals, Belmont, CA, USA Watkins, 1997; Von Richter et al., 2004). Moreover, given that and Pentobarb 500 for possum euthanasia was purchased from the proximal end of the intestine (e.g. duodenum) will encounter the National Veterinary Supplies, Christchurch, New Zealand. the majority of ingested PSMs prior to absorption, whereas the Cereal-based pellets for feeding studies (Opossum pellets) were distal end of the intestine will receive a limited number of non- purchased from Western Animal Nutrition, Rangiora, New absorbed PSMs, we predict that the proximal intestine will Zealand. harbor higher levels of CYP3A than the distal end. Given that the cortex of the kidney functions as a filtration system for blood 2.2. Animals containing parent PSMs and metabolites, whereas the medulla is primarily responsible for excreting metabolites, we predict the A total of 12 adult, male common brushtail possums were cortex will possess higher levels of CYP3A isoforms than the captured near Invermay Agricultural Centre (Mosgiel, Otago, medulla. The brain, testes and adrenal are important functional New Zealand, 45° 52′ S 170° 31′ E) using box traps baited with organs that we predict will be protected from xenobiotic apple. At the time of capture, each animal was screened for exposure by CYP3A isoforms. health status and only apparently healthy animals were in- The specific objectives of this study were: (1) to characterize cluded. Prior to use, animals were group housed in environ- relative abundance and distribution of CYP3A-like proteins in mentally-enriched pens (approximate 6×4 m2,8–10 possums/ various tissues of the possum and (2) to determine whether pen) under conditions of natural daylight and temperature CYP3A-like proteins are upregulated in response to terpenes in and fed a mixed diet of fresh fruit and specially-formulated the diet by comparing levels of CYP3A between control and cereal-based pellets (Opossum pellets), with fresh water terpene-treated possums. available ad libitum (Mcleod et al., 1997). Capture and housing of the animals and all experimental procedures performed, 2. Materials and methods had been given prior approval by the AgResearch Invermay Animal Ethics Committee according to the Animal Welfare We identified and semi-quantified CYP3A-like proteins in Act, 1999. various tissues with respect to ingestion of plant toxins in the generalist mammalian herbivore, the common brushtail pos- 2.3. Diet sum. In general, we compared levels of CYP3A-like proteins in the liver, various segments of the intestine (duodenum, jejunum, Control, acclimation and terpene diets consisted of cereal- ileum, caecum and colon), kidney (medulla and cortex), brain, based pellets. The cereal-based pellets contain approximately testes, and adrenal of possums under two dietary treatments: 1) 13% protein, 12% digestible energy and 8% fiber. For all diets, control treatment (i.e. without toxins) and 2) 4.3% terpene the pellets were ground and mixed with 50% water (by weight) mixture treatment (by wet weight of diet). At the end of each to ensure that all diets were of a similar consistency. The control treatment, tissues were collected from each animal and prepared diet included water alone, the acclimation diet contained a for semi-quantitative analysis of CYP3A levels using Western homogenous mix of 2.1% terpenes (by wet weight) and the Blot analysis. terpene treatment diet contained 4.3% terpenes (by wet weight). The terpene mixture included in the diets (acclimation and 2.1. Chemicals terpene) contained 1,8 cineole, p-cymene, α-pinene, and (S)- (−)-limonene in quantities of: 566, 16, 214 and 69 mg/kg body 1,8-Cineole, p-cymene, α-pinene, and limonene were mass, respectively. These quantities represent a proportion of purchased from Sigma-Aldrich, St. Louis, MO, USA. The terpenes in the diet that are similar to those detected in 196 J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201

Eucalyptus melliodora, a natural food of possums in the wild the control group (n =6)werefedthecontroldietforafurther (Boland et al., 1991). 5 d. Those in the terpene-treatment group (n=6) were fed the acclimation diet for 2 d and then the terpene diet for an 2.4. Experimental additional 3 d. One animal in the terpene-treatment group was removed from the terpene group prior to study completion Possums were randomly allocated to two groups and duetolackofintakeoftheterpene diet resulting in a sig- housed in individual cages (56×50×37 cm) with feeders and a nificant reduction in body mass (N5% over 2 d). All diets were nesting box, to allow for diet intake to be monitored. All freshly prepared daily and fed in excess of the intake require- possums were maintained on a pre-treatment diet of cereal- ments to maintain body mass. Diets were offered at approxi- based pellets for 8 d prior to treatments. The ground control mately1600heachday,fora24hperiod.Freshwaterwas treatment diet was then fed to all possums for 2 d. Animals in provided ad libitum throughout the experiment. Body mass

Fig. 1. Immunoblot analysis of CYP3A isoforms (50 μg protein/lane) in the liver, duodenum (Duod), jejunum (Jeju), testes (Test), adrenal gland (Adren), kidney medulla (K medu), kidney cortex (K cort), ileum and brain of brushtail possums, using rabbit anti-human cytochrome P450 3A4 polyclonal antibody. P1, P2 and P3 refer to each of the three CYP3A-like isoforms detected in tissues. The bottom X axis indicates individuals: M is the positive control mouse liver sample; C represents animals fed the control diet for the entire experiment (numbers 1, 4, 5, 8, 10, 11); and T represents animals fed the terpene diet (numbers 2, 3, 6, 7, 9). The asterisk (⁎) represents the possum that was removed from the terpene treatment prior to completion of the study due to lack of intake and significant body mass loss and was not included in statistical analysis.aindicates poor protein loading resulting in aberrantly low band intensity for β-actin (e.g. outliers) and represents values that were not included in statistical analysis. J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201 197 was measured on the day of allocation to treatment groups and in 1% nonfat milk in TBS buffer incubated for 1 h. CYP3A and again at the time of tissue collection. β-actin were immunostained with ECL-plus chemiluminescent detection reagent and exposed to x-ray film (Fig. 1). All pos- 2.5. Tissue collection sums, including the possum that did not complete the terpene treatment were included in the Western Blot (Fig. 1). At completion of the study, tissues were collected immedi- ately after euthanasia. Animals were lightly anaesthetized with 2.8. Data analysis halothane by inhalation (Halothane-VCA) and then euthanized by intra-cardiac injection of barbiturate (0.3–0.4 mL, Pentobarb Band density on x-ray film was integrated by ImageJ 1.34s 500). Immediately after death, a midline incision was made on (version 10.2, Wayne Rasband, National Institutes of Health, the ventral surface of the possum to expose the abdominal USA) with β-actin blotting used as an internal control to cavity. The gastrointestinal tract was removed from the account for variation in protein loading and as a normalizing esophagus (1 cm anterior to the stomach) to the rectum and agent for semi-quantitative analysis. All statistical analyses dissected into its component parts. The anterior 10 cm of the were performed using JMP software for Macintosh (SAS small intestine was deemed to be duodenum, the posterior 50% Institute, 2003). The hypothesis that CYP3A mediates tolerance to be ileum and the remainder to be jejunum. The border to terpenes was tested by comparing protein levels of CYP3A in between the proximal and distal colon was deemed to be the possum tissues, between the control and terpene groups, using region where fecal pellet formation was first apparent. Whole separate ANOVAs for each variable. The mean differences tissues were excised and weighed and/or measured for length. between control (n=6) and terpene-treated (n=5) groups were For larger tissues (liver, intestine segments, kidney, brain), a compared using separate one-way ANOVA for each tissue type. small section of approximately 5 mm3 was dissected out of the Data was tested for homogeneity of variance using the Levene whole tissue from approximately the same location for each test for Equal Variance. On occasions where variances were not animal. For intestinal sections, tissue was taken from the mid- homogenous, the Welch ANOVA testing Means Equal, point of each region. Tissue segments or whole tissues (testes allowing for unequal standard deviations, was used. and adrenal) were placed immediately in approximately 10 The band intensity for P3 in the liver and duodenum was not times the volume of RNAlater solution and stored at −80 °C intense enough to perform a separate analysis of this possible until analyzed. CYP isoform and is therefore not included in analyses. Some of the banding areas were not included in analysis due to apparent 2.6. Tissue preparation poor protein loading into the gel or poor electrophoretic transfer from gel to membrane resulting in aberrantly low band intensity Each tissue section was prepared for semi-quantification via of β-actin (i.e. outliers). The removal of each animal for Western Blot. Tissue segments were removed from RNAlater individual tissues is reported in Fig. 1. The possum that did not solution, blotted dry, weighed and placed in 600 mL of cell complete the terpene treatment due to lack of intake and loss of disruption buffer with added protease inhibitors (1:1000 dilution; Protease Inhibition Cocktail Set IV. Samples were homogenized with a motorized hand-held micropestle on ice until tissues were broken up (1–2 min). Samples were then centrifuged at 15,000 ×g for 2 min, vortexed to break up suspension and then re-spun at maximum speed for 1 min using a Beckman benchtop centrifuge. One half of the supernatant was removed and stored at −20 °C for future RNA studies. The remaining supernatant was flash frozen in liquid nitrogen and then stored at −80 °C for subsequent Western Blot analyses.

2.7. Western blot analysis

The concentration of protein present in the supernatant was determined by the Bradford method (Bradford, 1976). Fifty micrograms of protein from each homogenized tissue section wasresolvedin7.5%Tris–HCl-polyacrylamide gel and electrotransferred at 200 mA for 2.5 h to a nitrocellulose membrane. The membrane was incubated with rabbit anti- human CYP3A4 polyclonal antibody at a 1:1000 dilution and mouse anti-human β-actin antibody at 1:5000 dilution in 1% nonfat milk in TBS buffer for 2 h. The secondary antibodies Fig. 2. Mean daily dry matter (DM) intake (g/d;±SE) of control (hashed bar) and were a peroxidase-conjugated goat anti-rabbit IgG-HRP for terpene-treated (open bar) brushtail possums during the pre-treatment period and CYP3A, and goat anti-mouse for β-actin, in a 1:5000 dilution at the end of the treatment period. 198 J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201

periods (Fig. 3). Terpene intake reached a maximum of approximately 3.4 g terpenes/d by the end of the terpene- treatment period. There was no significant difference in body mass between the two treatment groups, either prior to the start of the pretreatment period (F(1, 10) =0.12,P=0.74; control=2.6± 0.12 kg, terpene=2.5±0.12 kg) or at the end of the study (F(1, 10) = 0.40, P=0.54; control=2.7±0.062 kg, terpene=2.7±0.14 kg).

3.2. Tissue mass and length

There was no difference in the mass or length of the majority of tissues (liver, kidney, testes, adrenal, duodenum, jejunum, ileum, caecum, proximal and distal colon between control and treatment groups, Table 1). The duodenum was significantly longer in the treatment group (46.4±9.7 cm) than in the control group (35.7±3.9 cm), despite similar duodenal mass in the two groups.

3.3. Western Blots Fig. 3. Mean daily terpene intake (g/d;±SE) by brushtail possums in the terpene- treated group on the 2 d of acclimation diet (A2, A3; 2.1% terpene by wet weight) and 3 d of the terpene diet (T1, T2, T3; 4.3% terpene by wet weight). Immunochemical detection of CYP3A-like enzymes in possums revealed three separate bands, referred to as P1, P2, and P3 (Fig. 1). This banding pattern indicates an incomplete body mass (asterisk in Fig. 1) were not included in the statistical specificity of the antibody in possums and the presence of at analysis. least three immunoreactive proteins. We assume these proteins are isoforms of CYP3A due to the strong reactivity to the 3. Results CYP3A antibody and the similarity to the mouse liver positive control (lane M, Fig. 1) for CYP3A4 (at 53 and 43 kD) and β-actin 3.1. Intake and body mass (at 41 kD). Possum CYP3A isoform 1 (P1), is approximately 53 kD, possum isoform 2 (P2), is approximately 48 kD and There was no significant difference in mean food intake (dry possum isoform 3 (P3), is approximately 44 kD. The mouse liver matter (DM)/d) between control and terpene-treatment groups, positive control also exhibited multiple banding patterns for either at the end of the pre-treatment period when both were CYP3A. consuming the control diet (F(1, 10) =0.81, P=0.39; control= There was preferential distribution of the isoforms between 76.7±9.11 g DM/d, terpene =89.9±11.9 g DM/d) or at the end specific tissues. Only the liver and duodenum had all three of the treatment period (F(1, 10) =0.31, P=0.59; control=89.9± isoforms (P1, P2, P3, Fig. 1). The jejunum only had P1. The 8.28 11 g DM/d, terpene=80.7±15.4 11 g DM/d; Fig. 2). testes, adrenal, kidney (medulla and cortex) and ileum only had Possums on the terpene treatment increased their intake of P2. The brain only had P3. The caecum and proximal and distal terpenes throughout the acclimation and terpene-treatment colon tissues did not express detectable levels of any CYP3A

Table 1 ANOVA results for mean differences between tissue size, β-actin area and CYP/β-actin ratio of brushtail possum tissues in control versus terpene-treated groups Tissue Tissue size β-actin area CYP/β-actin ratio a Liver F1,10 =0.31; P=0.59 F1,10 =1.72; P=0.22 P1+P2: F1,10 =0.84; P=0.38 P1: F1,10 =0.09; P=0.77 P2: F1,10 =1.00; P=0.34 a Duodenum Mass: F1,10 =0.67; P=0.43 F1,10 =0.81; P=0.39 P1+P2: F1,10 =0.10; P=0.76 Length: F1,10 =6.27; P=0.03 P1: F1,10 =0.03; P=0.87 P2: F1,10 =0.16; P=0.70 Jejunum Mass: F1,10 =0.55; P=0.48 F1,8 =0.12; P=0.73 P1: F1,8 =0.27; P=0.62 Length: F1,10 =0.21; P=0.66 Testes F1,10 =0.90; P=0.37 F1,10 =2.49; P=0.15 P2: F1,10 =0.01; P=0.91 Adrenal F1,10 =1.50; P=0.25 F1,10 =0.01; P=0.91 P2: F1,10 =0.18; P=0.68 Kidney medulla Whole kidney F1,10 =0.0009; P=0.98 F1,10 =2.60; P=0.14 P2: F1,10 =0.13; P=0.73 Kidney cortex F1,10 =0.13; P=0.73 P2: F1,10 =0.36; P=0.56 Ileum Mass: F1,10 =0.45; P=0.52 F1,8 =0.06; P=0.81 P2: F1,8 =0.51; P=0.56 Length: F1,10 =0.81; P=0.39 Brain F1,10 =0.005; P=0.95 P3: F1,10 =2.9; P=0.12 aP3 banding was not dense enough to quantify. J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201 199 isoform in these possums. There was a consistent pattern of the testes is a known substrate for CYP3A4 (Choi et al., 2005). CYP3A isoform distribution between tissues, regardless of Although comparisons between tissue types are limited with our treatment. There was no difference in relative levels of CYP3A- method of analysis, the increased frequency of detection of like isoforms between control and terpene-treated possums, in CYP3A-like isoforms in the cortex of the kidney compared to any tissue type (Table 1). the medulla demonstrated in this study has been observed in rats (Anakk et al., 2003) and is consistent with the finding that the 4. Discussion kidney cortex contains the cytochrome P450 system as a barrier to xenobiotics, while the medulla does not (Rush et al., 1984). The hypothesis that tolerance to PSMs is regulated by the Finally, the brain only contains isoform P3 and thus may be CYP3A subfamily of enzymes in common brushtail possums responsible for peripherally restricting certain xenobiotics from was not supported by this study. Possums fed plant toxins (i.e. the brain. the terpene mix) did not have higher CYP3A protein levels than Detection of multiple bands may indicate incomplete possums on control diet. However, despite the lack of specificity of CYP3A antibodies. The rabbit anti-human differential expression of CYP3A between terpene and control CYP3A4 polyclonal antibody does detect other P450 isoforms diets, the distribution of CYP3A in a variety of tissues suggests to a lesser extent (www.genetex.com). Incomplete specificity that CYP3A does play an important role in the metabolism of has been found in sheep and hamsters (Huan et al., 1998), xenobiotics in possums. humans (Mei et al., 1999) and in the mouse liver positive control used in this experiment (Fig. 1). Alternatively, multiple 4.1. Tissue distribution bands may represent post-translational modified products of a CYP3A isoforms (Oesch-Bartlomowicz and Oesch, 2005). Full Although the liver is generally considered to be the major site identification and characterization of each isoform requires of xenobiotic metabolism, other tissues can contribute (Nishimura cloning and sequencing the CYP3A enzyme in possums. et al., 2004). In humans and other eutherian species, CYP3A is an CYP3A-like isoform variability found between tissues of an important first line of defense in the intestine (Watkins, 1997)and individual possum and between possums for an individual tissue plays a role in minimizing xenobiotic accumulation in important is consistent to the variability found in humans (Koch et al., tissues such as the brain and testes. Previous studies in 2002). For example, in humans, CYP3A4, CYP3A5, CYP3A7 mammalian herbivores may have underestimated the importance were all expressed in the intestine, whereas only CYP3A5 and of CYP3A by focusing only on hepatic CYP3A levels. This is the CYP3A7, but not CYP3A4 were detected in the adrenal. In the first study to investigate the distribution of these enzymes in testes, CYP3A7 was expressed to a lesser extent than other various tissues and highlights the potential importance of tissues, CYP3A5 was barely detectable and CYP3A4 was not metabolism in non-hepatic tissues. Given the semi-quantitative present. As in humans, the CYP3A variability between tissues nature of Western Blots, this study was unable to directly compare and between individual possums may affect the predisposition of CYP3A levels between tissue types. Future studies should these animals to xenobiotics that are bioactivated by specific investigate the relative role of CYP3A and other xenobiotic CYP3A isozymes (Kirby et al., 1993) and should be explored metabolizing enzymes by quantifying the amount of these further. enzymes present in various tissues, using quantitative PCR and/ There was variability in band intensity for both β-actin and or more quantitative immunochemical techniques. the CYP3A-like isoforms in certain tissues. Specifically, the One interesting feature of CYP3A in possums is that three variability in β-actin band intensity (e.g. jejunum, ileum) is isoforms were detected. Although we were unable to conclude most likely attributed to poor protein loading resulting in faint which specific CYP3A isoforms are preferentially expressed in banding, overloading of protein resulting in band smearing, possum tissues, this study suggests that particular isoforms may and/or inefficient transfer of proteins from the gel to be more important in certain tissues. For example, that the liver nitrocellulose membrane. Although the loading variability can and duodenum produce all three isoforms (Fig. 1) suggests significantly affect the visual observation of banding intensity that these tissues may be exposed to the highest diversity for CYP3A-like isoforms, this variability is controlled for and concentrations of xenobiotics, including ingested PSMs in analysis between treatments by using the ratio of CYP3A to and therefore, may require a more diverse set of xenobiotic β-actin intensity and by removing extreme loading variants as metabolizing enzymes. Conversely, as the caecum and proximal indicated in Fig. 1. The faint CYP3A-like isoform signals and distal colon did not express detectable levels of any CYP3A relative to intense β-actin signals (e.g. animal 4, 7, 8 and 10 isoform, this may suggest that these tissues are exposed to liver samples; animal 10 kidney cortex sample) and lack of minimal levels of PSMs or other xenobiotics or do not play a role signal in the adrenal, kidney medulla, ileum of certain in PSM metabolism. Additional studies are needed to confirm individuals describes the variability in the CYP3A metabolizing whether CYP3A-like isoforms are completely absent from the enzymes in possums. The variability in CYP3A and β-actin caecum and colon or simply present in low levels. Isoform P2, banding intensity in the mouse liver positive control (M, Fig. 1) found in the testes, adrenal and kidney (medulla and cortex), is is likely attributed to the degradation of protein in this tissue possibly responsible for metabolism of xenobiotics that are not sample over time due to repeated freeze-thaw cycles for the eliminated in first pass metabolism or are exogenously produced single mouse liver sample. The mouse liver variability does not in these specific tissues. For example, testosterone produced in influence results of this study as this sample was only used to 200 J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201 verify experimental conditions and confirm band location for terpenes over 5 d induce CYP3A-like isoform compared to possum samples. animals not exposed to terpenes. Although the functional role of CYP3A in possums was not investigated, this study has 4.2. Control versus terpene diets provided empirical data that identifies the prevalence of CYP3A in tissues other than the liver in possums. Future studies that The lack of difference in expression between the control and identify the distribution and induction of additional xenobiotic terpene-treated groups may suggest that possums have consti- metabolizing enzymes in various tissues would provide insight tutively high CYP3A levels. Possums are well known for their into the mechanisms that regulate toxin absorption and varied tolerance to both toxic plants (Marsh et al., 2003; Stupans metabolism in possums. et al., 2001) and biocontrol poisons (Eason et al., 1993, 1999). Therefore, their ability to metabolize and tolerate such diverse Acknowledgments xenobiotics may be attributed to constitutively high levels of the CYP3A subfamily of enzymes. It may be possible that CYP3A We thank the Euan Thompson (AgResearch Invermay) for enzyme levels were elevated by pre-treatment exposure to collection and maintenance of possums and the Invermay terpenes. However, possums did not have access to any foods possum research team for assistance with tissue collection. We containing terpenes for at least 8 d prior to the start of the control thank Dr. Suong Ngo for her expertise on in vitro assays in diet. This is expected to be the sufficient time for induced possums and Dr. David Williams for technical assistance with enzyme concentrations to return to baseline levels. assays and laboratory space for experiments. Mark Reimers, Lack of differences between control and terpene-treated Jane LaDu, Beth Siddens and Eric Andreasen all provided possums may indicate that the exposure to terpene in the present invaluable assistance with assays. Research was supported by study (3 d of 4.3% terpene diet) was not enough to induce NSF International Research Fellowship INT-0301898 and an up-regulation of CYP3A enzymes. However, xenobiotic metab- Oregon State University General Research Fund ADG056 to JS olizing enzymes are thought to be fully induced after three con- Sorensen. secutive days of xenobiotic exposure (Sipes and Gandolfi, 1986). The plateau in terpene intake during the 3 d of the 4.3% terpene References diet (Fig. 2) further suggests that the xenobiotic metabolizing enzymes required for terpene metabolism are maximally induced Anakk, S., Ku, C.Y., Vore, M., Strobel, H.W., 2003. Insights into gender bias: rat in these possums. However, it is possible that increased tolerance cytochrome P450 3A9. J. Pharmacol. Exp. Ther. 305, 703–709. (i.e. intake) to terpenes by possums may not be associated with Boland, D.J., Brophy, J.J., House, A.P.N., 1991. Eucalyptus Oils: Use, Chemistry, Distillation and Marketing. Inkata Press, Melbourne. CYP3A induction, but rather the induction of other xenobiotic Boyle, R., Mclean, S., Foley, W.J., Davies, N.W., 1999. Comparative metabolizing enzymes. Research measuring a diverse array of metabolism of dietary terpene, p-cymene, in generalist and specialist CYP isozymes following terpene exposure is warranted. folivorous marsupials. J. Chem. Ecol. 25, 2109–2127. A lack of CYP3A induction after exposure to terpenes in Boyle, R., Mclean, S., Davies, N.W., 2000. Biotransformation of 1,8-cineole in – these possums, would be in contrast to previous investigations in the brushtail possum (Trichosurus vulpecula). Xenobiotica 30, 915 932. – – Bradford, M.A., 1976. A rapid and sensitive method for the quantification of other mammals. The terpenes 1 8 cineole and 1 4 cineole are microgram quantities of protein utilizing the principle of protein dye metabolized by CYP3A in human and rat liver microsomes binding. Anal. Biochem. 72, 248–254. (De-Oliveira et al., 1997; Miyazawa et al., 2001a,b; Roffey et al., Choi, M.H., Skipper, P.L., Wishnok, J.S., Tannenbaum, S.R., 2005. Character- 1990). Although this may indicate that terpenes are not sub- ization of testosterone 11 beta-hydroxylation catalyzed by human liver – strates for CYP3A, it is more plausible to suggest that possums microsomal cytochromes P450. Drug Metab. Dispos. 33, 714 718. De-Oliveira, A.C.A.X., Ribeiro-Pinto, L., Otto, S.S., Goncalves, A., Paumgartten, do use CYP3A, but preferentially utilize different pathways for F.J.R., 1997. Induction of liver monoxygenases by β-myrcene. Toxicology terpene metabolism. For example, CYP2E, lauric acid hydrox- 124, 135–140. ylase, acyl CoA oxidase and peroxisomal cyanide-intensive Eason, C.T., Frampton, C.M., Henderson, R., Thomas, M.D., Morgan, D.R., palmitoyl coenzyme are elevated in possum liver following 1993. Sodium mono-fluoroacetate and alternative toxins for possum control. – terpene exposure (Ngo et al., 2003). These enzyme systems have N.Z. J. Zool. 20, 329 334. Eason, C.T., Wright, G.R., Gooneratne, R., 1999. Pharmacokinetics of antipyrine, received relatively little attention compared to CYP3A by those warfarin and paracetamol in the brushtail possum. J. Appl. Toxicol. 19, 157–161. studying plant–herbivore interactions (Boyle et al., 2000, 1999; Hiroi, T., Miyazaki, Y., Kobayashi, Y., Imaoka, S., Funae, Y., 1995. Induction of Pass et al., 1999, 2001, 2002), and their function in tissues other hepatic P450s in rat by essential wood and leaf oils. Xenobiotica 25, 457–467. than the liver has not been investigated. Given the results of this Ho, P., Luo, X., Macauley, J.S., Grigor, M.R., Wanwimolruk, S., 1998. In vitro study, their contribution to terpene metabolism in various tissues hepatic metabolism of CYP3A-mediated drugs quinine and midazolam in the common brush-tailed possum (Trichosurus vulpecula). Environ. Toxicol. should be investigated. Chem. 17, 317–324. Huan, J.Y., Miranda, C.L., Buhler, D.R., Cheeke, P.R., 1998. Species differences 5. Conclusions in the hepatic microsomal enzyme metabolism of the pyrrolizidine alkaloids. Toxicol. Lett. 99, 127–137. The research compared the relative importance of the Ilett, K., Tee, L., Reeves, P., Minchin, R., 1993. Metabolism of drugs and other xenobiotics in the gut lumen and wall. Pharmacol. Ther. 46, 67–93. xenobiotic metabolizing enzyme CYP3A in the foraging Kirby, G.M., Wolf, C.R., Neal, G.E., Judah, D.J., Henderson, C.J., Srivatanakul, ecology of a mammalian herbivore with specific insight into P., Wild, C.P., 1993. In vitro metabolism of aflatoxin B1 by normal and toxin tolerance. We found no evidence that possums exposed to tumorous liver tissue from Thailand. Carcinogenesis 14, 2613–2620. J.S. Sorensen et al. / Comparative Biochemistry and Physiology, Part C 145 (2007) 194–201 201

Koch, I., Weil, R., Wolbold, R., Brockmoller, J., Hustert, E., Burk, O., Neussler, Oesch-Bartlomowicz, B., Oesch, F., 2005. Phosphorylation of cytochromes A., Neuhaus, P., Eichelbaum, M., Zanger, U., Wojnowski, L., 2002. P450: first discovery of a posttranslational modification of a drug- Interindividual variability and tissue-specificity in the expression of metabolizing enzymes. Biochem. Biophys. Res. Commun. 338, 446–449. cytochrome P450 3A mRNA. Drug Metab. Dispos. 30, 1108–1114. Pass, G.J., Mclean, S., Stupans, I., 1999. Induction of xenobiotic metabolising Lamda, J.K., Lin, Y.S., Schuetz, E.G., Thummel, K.E., 2002. Genetic enzymes in the common brushtail possum, Trichosurus vulpecula,by contribution to variable human CYP3A-mediated metabolism. Adv. Drug Eucalyptus terpenes. Comp. Biochem. Physiol. C 124, 239–246. Deliv. Rev. 54, 1271–1294. Pass, G.J., Mclean, S., Stupans, I., Davies, N., 2001. Microsomal metabolism of Marsh, K.J., Foley, W.J., Cowling, A., Wallis, I.R., 2003. Differential the terpene 1,8-cineole in the common brushtail possum (Trichosurus susceptibility to Eucalyptus secondary compounds explains feeding by the velpecula), koala (Phascolarctos cinereus), rat and human. Xenobiotica 31, common ringtail (Pseudocheirus peregrinus) and common brushtail possum 205–221. (Trichosurus vulpecula). J. Comp. Physiol. B 173, 69–79. Pass, G.J., Mclean, S., Stupans, I., Davies, N., 2002. Microsomal metabolism Mcleod, B.J., Thompson, E.G., Crawford, J.L., Shackell, G.H., 1997. Successful and of the terpene p-cymene in the common brushtail group housing of wild-caught brushtail possums (Trichosurus vulpecula). possum (Trichosurus velpecula), koala (Phascolarctos cinereus) and rat. Anim. Welf. 6, 67–76. Xenobiotica 32, 383–397. Mei, Q., Tang, C., Assang, C., Lin, Y.,Slaughter, D., Rodrigues, A.D., Baillie, T.A., Roffey, S.J., Walker, R., Gibson, G.G., 1990. Hepatic peroxisomal and Rushmore, T.H., Shou, M., 1999. Role of a potent inhibitory monoclonal microsomal enzyme induction by citral and linalool in rats. Food Chem. antibody to cytochrome P-450 3A4 in assessment of human drug metabolism. Toxicol. 28, 403–408. J. Pharmacol. Exp. Ther. 29, 749–759. Rush, G.F., Smith, J.H., Newton, J.F., Hook, J.B., 1984. Chemically induced Miyazawa, M., Shindo, M., Shimada, T., 2001a. Oxidation of 1,8-cineole, the nephrotoxicity: role of metabolic activation. Crit. Rev. Toxicol. 13, 99–160. monoterpene cyclic ether originated from eucalyptus polybractea, by SAS Institute, S., 2003. JMP Statistics and Graphics Guide. Version 5.0.1.2. cytochrome P450 3A enzymes in rat and human liver microsomes. Drug SAS Institute, Inc., Gary, North Carolina. Metab. Dispos. 29, 200–205. Sipes, I.G., Gandolfi, A.J., 1986. Biotransformation of toxicants, In: Klaasen, C.D., Miyazawa, M., Shindo, M., Shimada, T., 2001b. Roles of cytochrome P450 3A Amdurand, M.O., Doull, J. (Eds.), Casarett and Doull's Toxicology The Basic enzymes in the 2-hydroxylation of 1,4-cineole, a monoterpene cyclic ether, Science of Poisons, 3rd ed. McMillan, New York, pp. 64–98. by rat and human liver microsomes. Xenobiotica 31, 713–723. Stupans, I., Jones, B., Mckinnon, R.A., 2001. Xenobiotic metabolism in Ngo, S.N.T., Mckinnon, R.A., Stupans, I., 2003. The effects of Eucalyptus Australian marsupials. Comp. Biochem. Physiol. C 128, 367–376. terpenes on hepatic cytochrome P450 CYP4A, peroxisomal Acyl CoA Von Richter, O., Burk, O., Fromm, M., Thon, K., Eichelbaum, M., Kivisto, K., oxidase (AOX) and peroxisome proliferator activated receptor alpha 2004. Cytochrome P450 3A4 and P-glycoprotein expression in human small (PPARalpha) in the common brush tail possum (Trichosurus vulpecula). intestinal enterocytes and hepatocytes: a comparative analysis in paired Comp. Biochem. Physiol. C 136, 165–173. tissue specimens. Clin. Pharmacol. Ther. 75, 172–183. Nishimura, M., Naito, S., Yokoi, T., 2004. Tissue-specific mRNA expression Watkins, P.B., 1997. The barrier function of CYP3A4 and P-glycoprotein in the profiles of human nuclear receptor subfamilies. Drug Metab. Pharmacokinet. small bowel. Adv. Drug Deliv. Rev. 27, 161–170. 19, 135–149.