Recombinant adenovirus (Ad) serotype 5 is a vector commonly used for gene delivery. Although this vector has a natural tropism for the liver, there is a limited understanding of how Ad administration affects one of the primary hepatic processes, drug metabolism. The effects of systemic administration of a model recombinant adenoviral vector on two hepatic cytochrome P450 (P450) enzymes, CYP3A2 and 2C11, were investigated. Sprague-Dawley rats were treated with one of six vector doses, ranging from 5.7 × 106 to 5.7 × 1012 virus particles (vp)/kg. Hepatic P450 protein expression, catalytic activity, and mRNA levels were measured over 14 days. Ad administration (5.7 × 1010-5.7 × 1012 vp/kg) reduced CYP3A2 over the duration of the study. Six hours after administration of 5.7 × 1012 vp/kg, CYP3A2 activity and mRNA levels were suppressed by 45 and 65%, respectively (P ≤ 0.01). This continued throughout the study with levels dropping to 36 and 45% of controls by 14 days, respectively (P ≤ 0.01). A similar trend was detected for CYP2C11 within this dosing range. Administration of 5.7 × 106, 5.7 × 108, and 5.7 × 109 vp/kg of Ad significantly increased both CYP2C11 protein expression by 86, 71, and 107% and activity 110, 118, and 53%, respectively, above those of animals treated with saline (P ≤ 0.01). These results clearly indicate that a single dose of adenovirus significantly alters key drug metabolizing enzymes for an extended period of time and should be investigated further in the context of the design and implementation of clinical trial protocols.
Recombinant human adenovirus (Ad) serotype 5 is routinely used for gene delivery in both preclinical and clinical studies (Imperiale and Kochanek, 2004). The ability of the virus to infect a wide variety of cells and induce rapid production of high levels of a therapeutic transgene has made it an attractive candidate for liver-directed gene transfer. This was confirmed by several studies in the early 1990s in which Ad vectors induced hepatocytes to rapidly produce large amounts of therapeutic proteins (Gomez-Foix et al., 1992; Jaffe et al., 1992). Subsequent studies have shown that moderate doses of these vectors are capable of transducing more than 90% of all hepatocytes when given by intravenous injection (Nunes et al., 1999; Christ et al., 2000). At certain doses, hepatotoxicity was observed (Muruve et al., 1999). In addition, significant amounts of virus have been detected in the liver after direct injection into cancerous tumors due to vector dissemination into the circulation (Sauthoff et al., 2003). Taken together, these statements strongly suggest that Ad, when used as a medicinal agent, will accumulate to some degree in the liver. However, it is currently not clear how hepatic accumulation of the virus at various concentrations will impact drug metabolism.
The cytochrome P450 (P450) enzyme system is responsible for the majority of drug metabolism that takes place in the liver. Not only are P450 enzymes responsible for the inactivation and circulatory clearance of numerous substrates, they also convert various compounds to their pharmacologically active metabolite(s) (Nebert and Russell, 2002). In humans, isoforms from the subfamilies 3A and 2C are known to metabolize approximately 50 and 20%, respectively, of currently marketed medications (Yan and Caldwell, 2001). Because preclinical evaluations of drug-metabolizing enzymes routinely encompass the isoforms 3A2 and 2C11 when the male rat model is used, they were selected for this study. Various factors are known to cause alterations in these enzymes. It is well documented that infection and the presence of inflammatory stimuli result in aberrations of P450 enzyme levels and activities (Renton, 2000; Morgan, 2001). Thus, understanding the effects of recombinant viral vectors on P450 enzymes is important in regard to their clinical use. Patients involved in gene therapy trials are likely to be given concomitant study medications or medications to treat adverse effects. P450 alterations may delay an otherwise predictable pattern of drug metabolism resulting in higher than normal systemic blood levels or, conversely, the efficacy of a drug may be obstructed resulting in little or no therapeutic effect.
Systemic administration of adenoviral vectors activates both the innate and adaptive immune responses in animal models and humans (Trapnell and Shanley, 2002). The first of three phases occurs as early as 1 h after administration and peaks between 6 and 24 h. This acute phase is related to the innate immune response and is characterized by thrombocytopenia, intense periportal polymorphonuclear leukocyte infiltration, and elevated liver enzymes (Christ et al., 2000). These effects are largely due to the release of several cytokines and chemokines into the general circulation (Christ et al., 2000; Tarpnell and Shanley, 2002). The second phase begins 1 to 4 days after administration and is characterized by removal of transduced cells by activated lymphocytes (Yang et al., 1994). At high viral loads, this can progress to severe liver dysfunction (Morral et al., 2002). The late phase continues for several weeks after vector administration. During this time, cells initially expressing low levels of viral genes and transgene products are recognized and eliminated by cytotoxic T cells (Yang et al., 1994; Tarpnell and Shanley, 2002).
Although the toxic profile and immune response following Ad administration is well documented, there is a limited understanding of how these physiological aberrations affect other hepatic processes, such as drug metabolism. This is of particular importance as the recent approval of Gendicine by China's State Food and Drug Administration has made an adenoviral-based product the first gene medicine available for clinical use worldwide (Pearson et al., 2004). Fast-track approval granted by the United States Food and Drug Administration for a second adenoviral-based vector, Advexin, further suggests that these vectors will soon be used routinely in the clinic (Anonymous, 2003). This report summarizes our initial efforts to investigate the effects of systemic administration of a model recombinant adenoviral vector expressing Escherichia coli β-galactosidase on hepatic P450 expression and function at 0.25, 1, 4, and 14 days after treatment. Time points were based upon documented clinical sequelae associated with use of adenoviral vectors (Trapnell and Shanley, 2002). Doses ranging from 5.7 × 106 to 5.7 × 1012 virus particles (vp)/kg were selected to encompass or include those reportedly used in the clinic (Reid et al., 2002).
Materials and Methods
Materials. Unless otherwise noted, all chemicals were purchased from EMD Chemicals Inc. (Gibbstown, NJ) in the highest purity available. Phosphate-buffered saline, sorbitan mono-9octadecenoate poly (oxy-1,1-ethanedlyl) (Tween 20), diethylpyrocarbonate (DEPC), EDTA, formaldehyde, isopropanol, glucose 6-phosphate, glucose-6-phosphate dehydrogenase, NADP+, 11α-hydroxyprogesterone, testosterone, ketamine, and xylazine were purchased from Sigma-Aldrich (St. Louis, MO). Acepromazine was purchased from Fort Dodge Laboratories, Inc. (Fort Dodge, IA). Protogel acrylamide was purchased from National Diagnostics (Atlanta, GA). 5-Bromo-4-chloro-3-indolyl-β-d-galactoside (X-gal) was purchased from Gold Biotechnology Inc. (St. Louis, MO). Tissue-tek optimal cutting temperature embedding compound was purchased from Fisher Scientific Co. (Pittsburgh, PA). Dulbecco's modified Eagle's medium was purchased from Mediatech (Herndon, VA). Polyclonal rat CYP3A2 and CYP2C11 primary antibodies were purchased from BD Gentest (Woburn, MA). Corresponding horseradish peroxidase-conjugated secondary antibodies were purchased from ICN Pharmaceuticals Biochemicals Division (Aurora, OH). Isoform-specific P450 protein standards, used for relative quantification, were purchased from XenoTech, LLC (Lenexa, KS). Primers for CYP3A2 and 2C11 were purchased from Sigma-Genosys (The Woodlands, TX).
Virus Preparation. A replication defective recombinant adenoviral vector expressing the E. coli β-galactosidase transgene under the control of a cytomegalovirus promoter (AdlacZ) was amplified in 293 cells according to established protocols (Croyle et al., 2002). The virus was purified by banding cell lysates twice on cesium chloride gradients and desalting on an Econo-Pac 10DG disposable chromatography column (Bio-Rad, Hercules, CA) equilibrated with phosphate-buffered saline, pH 7.4. Viral fractions were collected, and the concentration was determined by UV spectrophotometric analysis at 260 nm (Beckman Du 530 UV/Vis; Beckman Coulter, Fullerton, CA). The total viral particle number (active and inactive) in a given preparation was determined using the method of Maizel et al. (1968) using the following formula: virus particles/ml = (absorbance at 260 nm) × (dilution factor) × 1.1 × 1012. All animals were treated with freshly purified virus.
Plaque Assay. To determine the amount of active virus in a given preparation, a plaque assay was performed on all viral preparations employed in these studies according to established protocols (Graham and van der Eb, 1973). In brief, serial dilutions of virus were made in Dulbecco's modified Eagle's medium supplemented with 2% fetal bovine serum (Cambrex Bio Science Walkersville, Inc., Walkersville, MD). An aliquot of 250 μl of each dilution was added to a monolayer of 293 cells and incubated at 37°C for 2 h. Cells were then overlaid with 0.8% agarose (SeaPlaque; Cambrex Bio Science Walkersville, Inc.) in modified Eagle's medium (without phenol red; Invitrogen, Carlsbad, CA), 2% fetal bovine serum, and 10 mM MgCl2. Plaque-forming units (pfu) were calculated according to the following formula: pfu/ml = average number of plaques × dilution factor × 4.
The particle/pfu ratio was then calculated by dividing the number of particles obtained from the absorbance reading of a preparation at 260 nm by the number of active particles (plaque-forming units/milliliter) detected by the plaque assay. The average virus particle/pfu ratio for the virus preparations used in these studies was 690:1.
Administration of Adenoviral Vectors. All procedures were approved by the Institutional Animal Care and Use Committee of The University of Texas at Austin and are in accordance with the guidelines established by the National Institutes of Health for the humane treatment of animals. Adult male Sprague-Dawley rats purchased from Harlan (Indianapolis, IN) were housed in individual cages and allowed unrestrained access to standard rodent chow (Harlan) and tap water. Following a 7-day acclimation period, all animals underwent surgery to implant a catheter into the right jugular vein. Animals received a single intramuscular administration of anesthetic mixture consisting of a 1:1:1 (v/v/v) ratio of ketamine (100 mg/ml), xylazine (20 mg/ml), and acepromazine (10 mg/ml). Rats were allowed to recover from surgery for 24 h prior to treatment. Rats were then given a single intravenous dose of AdlacZ or vehicle, phosphate-buffered saline, via the surgically implanted catheter. The six escalating dose concentrations were: 5.7 × 106, 5.7 × 108, 5.7 × 109, 5.7 × 1010, 5.7 × 1011, and 5.7 × 1012 vp/kg. Upon sacrifice, serum was collected for assessment of transaminase levels. The liver was excised, rinsed in 0.9% saline, snap frozen in liquid nitrogen, and stored at -80°C for microsome preparation and RNA analysis. A section of liver was placed in Tissue-tek embedding medium for X-gal histochemistry.
Microsome Isolation. Hepatic microsomal proteins were isolated by homogenizing tissue in 3 volumes of a 0.1 M Tris chloride buffer, pH 7.4, consisting of 0.1 M EDTA and 0.15 M potassium chloride using a PowerGen 700 homogenizer (Fisher Scientific Co.). The homogenate was centrifuged at 9000g for 20 min at 4°C. The supernatant was collected and immediately centrifuged at 377,000g for 17 min at 4°C. Following the second centrifugation, the pellet was resuspended in 0.1 M sodium pyrophosphate buffer containing 0.1 M EDTA at pH 7.4. The sample was then homogenized and centrifuged at 377,000g for an additional 17 min at 4°C. The supernatant was discarded; the pellet was resuspended and homogenized in 0.01 M Tris buffer containing 20% glycerol and stored at -80°C until analysis.
In Vitro Metabolism Studies. CYP3A2 and 2C11 metabolic activity was determined by in vitro analysis of testosterone hydroxylation as previously described (Lu et al., 2003). Reaction mixtures consisted of 1-ml volumes containing 0.1 M potassium phosphate, 250 μM testosterone in methanol, 200 μg of hepatic microsomal protein, and a NADPH regenerating system consisting of 5 mM NADP+, 100 mM glucose 6-phosphate, and 100 mM magnesium chloride. The initial mixture was incubated for 3 min at 37°C. The reaction was initiated by the addition of 5 units of glucose-6-phosphate dehydrogenase. Following a 15-min incubation, the addition of 5 ml of dichloromethane to the mixture served to quench the reaction. An internal standard, 3.6 nmol of 11α-hydroxyprogesterone, was added to each sample and mixed thoroughly. The organic layer was removed and dried under a constant stream of air. Dried extracts were then dissolved in 200 μl of methanol and stored at 4°C until further analysis. Testosterone metabolites were separated and quantified by high-pressure liquid chromatography according to a previously described method (Lu et al., 2003).
Gel Electrophoresis and Immunoblot Analysis. Twenty micrograms of protein samples were separated by size on an 8% sodium dodecyl sulfate polyacrylamide gel. Proteins were transferred to a nitrocellulose membrane and the membrane blocked with 3% nonfat dry milk (NFDM) in Tris-buffered saline, pH 7.4, for 1 h at room temperature. The nitrocellulose sheets were then incubated for 1 h at room temperature in a 1:3000 dilution of the specific primary P450 antibody in 3% NFDM. The nitrocellulose membranes were exposed to the corresponding horseradish peroxidase conjugated secondary antibody (1:3000 dilution) in 3% NFDM for 1 h. Immunoreactive proteins were detected by chemiluminescence using a Western Lighting detection kit according to the manufacturer's protocol (PerkinElmer Life and Analytical Sciences, Boston, MA). Band density was determined by exposure of the nitrocellulose membrane to Kodak Biomax film (Eastman Kodak Co., Rochester, NY). Blot densities were measured using a flatbed scanner (Microtek, Carson, CA) and analyzed using Kodak 1D image analysis software (Eastman Kodak Co.). We have established linearity for both CYP3A1/2 and 2C11 for up to 80 μg of protein (Lu, 2004). It should be noted that CYP3A1 and CYP3A2 comigrate during electrophoresis. The antibody used to detect CYP3A2 protein expression was a polyclonal antibody with cross reactivity to CYP3A1, therefore all protein expression results for CYP3A2 are reported as CYP3A1/2.
RNA Isolation. Hepatic RNA was isolated with TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. The resulting RNA was reconstituted in DEPC-treated water, quantitated spectrophotometrically at 260 nm, and adjusted to a concentration of 100 μg/ml. Samples were stored in DEPC-treated water at -80°C until analysis.
RT-PCR Analysis. One microgram of isolated hepatic RNA was reverse-transcribed with random primers using the reverse transcription kit RETROscript (Ambion, Austin, TX) following the manufacturer's instructions. PCR was then performed using ReadyMix PCR reaction mix (Sigma-Aldrich) in a final volume of 12.5 μl containing 0.5 μl of reverse-transcription product and 0.1 mM each primer. Fifteen cycles of a standard PCR cycle were carried out including a 30-s denaturation at 94°C, a 30-s annealing at 55°C, and a 30-s extension at 72°C. Cycling was initiated with a 4-min denaturation at 94°C and terminated with a 10-min incubation at 72°C. All PCR reactions were performed in a PTC-100 thermal cycler (MJ Research, Watertown, MA). The primer sequences for CYP3A2 were 5′-TTG ATC CGT TGT TCT TGT CA-3′ (sense) and 5′-GGC CAG GAA ATA CAA GAC AA-3′ (antisense), product size of 323 base pairs. CYP2C11 primer sequences were 5′-CTG CTG CTG CTG AAA CAC GTC-3′ (sense) and 5′-GGA TGA CAG CGA TAC TAT CAC-3′ (antisense), product size of 249 base pairs. QuantumRNA 18S internal standards (Ambion) using Ambion's competimer technology were coamplified with CYP3A2 or 2C11 in individual reaction tubes. The ratio of 18S primer to competimer for all CYP3A2 and 2C11 duplex PCR reactions was 4:6, as proposed by the manufacturer. Amplicons were visualized on a 2% agarose gel containing ethidium bromide, and the intensity of each band was determined by densitometric analysis using Kodak ID image analysis software (Eastman Kodak Co.)
Liver Function Analysis. Serum aspartate aminotransferase (AST) levels and alanine aminotransferase (ALT) levels were determined using VITROS ALT and AST slides on a VITROS DT60 AutoAnalyzer (Ortho-Clinical Diagnostics, Raritan, NJ).
X-Gal Histochemistry. Frozen liver sections (6 μm) were fixed in 0.5% glutaraldehyde and stained for β-galactosidase activity as previously described (Croyle et al., 2002).
Data Analysis. One-way analysis of variance with a Bonferroni/Dunn post hoc analysis was used to determine statistical differences between individual groups (SuperANOVA; Abacus Concepts, Berkley, CA). Differences were determined to be significant when the probability of chance explaining the results was reduced to less than 5% (P < 0.05).
Effect of Administration of High Doses of AdlacZ on CYP3A Protein Expression, Catalytic Activity, and mRNA Levels. A replication defective recombinant adenoviral vector expressing the E. coli β-galactosidase transgene, the most commonly used in situ reporter in preclinical development of gene transfer protocols (Cui et al., 1994) and routinely used in our laboratory, was administered in these studies. One day after administration of either 5.7 × 1011 or 5.7 × 1012 vp/kg of this vector, CYP3A1/2 protein expression was significantly suppressed by 59 and 60%, respectively, compared with the vehicle treatment group (P ≤ 0.01, Fig. 1A). After 4 days, CYP3A1/2 protein expression was suppressed by 35, 40, and 42% following administration of 5.7 × 1010, 5.7 × 1011, and 5.7 × 1012 vp/kg, respectively (P ≤ 0.05, Fig. 1A). Protein levels displayed signs of recovery at 14 days. In vitro analysis of hepatic CYP3A2 activity was similar to immunoblot data. Production of the primary metabolite of CYP3A2 testosterone hydroxylation, 6β-hydroxytestosterone (Waxman et al., 1983), was significantly suppressed by 45% as early as 6 h after administration of 5.7 × 1012 vp/kg (P ≤ 0.01, Fig. 1B). One day after treatment, catalytic activity of CYP3A2 was suppressed by 30, 39, and 38% for each of the three doses, respectively (P ≤ 0.05, Fig. 1B). This suppression continued through 4 days with levels of 6β-hydroxytestosterone being reduced by 48, 36, and 41% compared with those animals in the vehicle treatment group (P ≤ 0.01, Fig. 1B). Following administration of 5.7 × 1012 vp/kg, CYP3A2 mRNA levels were significantly suppressed as early as 6 h (P ≤ 0.01, Fig. 1C). At 4 days, animals treated with 5.7 × 1011 and 5.7 × 1012 vp/kg had mRNA levels significantly lower than those of control animals (30 and 61%, respectively, P ≤ 0.05, Fig. 1C). The pattern of suppression for CYP3A2 mRNA production was similar to protein expression and activity at 14 days for all treatment groups.
Effect of Administration of High Doses of AdlacZ on CYP2C11 Protein Expression, Catalytic Activity, and mRNA Levels. Western blot analysis revealed that 1 day after administration of 5.7 × 1010, 5.7 × 1011, and 5.7 × 1012 vp/kg CYP2C11 protein expression was significantly suppressed by 41, 52, and 75%, respectively, compared with vehicle-treated animals (P ≤ 0.05, Fig. 2A). Four days after administration, levels remained suppressed by 48, 56, and 67%, in order of increasing dose (P ≤ 0.01, Fig. 2A). At 14 days, protein levels showed signs of recovery with no significant suppression noted in any of the three treatment groups. In vitro analysis of CYP2C11 catalytic activity was examined by measuring the formation of the testosterone metabolite 2α-hydroxytestosterone (Cheng and Schenkman, 1983). Six hours after administration of the highest dose, CYP2C11 activity levels were suppressed by 56% compared with vehicle-treated animals (P ≤ 0.01, Fig. 2B). Activity levels remained suppressed for this treatment group through 14 days with the greatest suppression, 78%, seen at 4 days (P ≤ 0.01, Fig. 2B). Following administration of 5.7 × 1010 and 5.7 × 1011 vp/kg, 2α-hydroxytestosterone production was suppressed by 22 and 19%, respectively, 1 day after administration (P ≤ 0.05, Fig. 2B). Suppression continued through 4 days, 33 and 35%, respectively, with levels returning to baseline by day 14 (P ≤ 0.05, Fig. 2B). Treatment with each of the three higher doses of AdlacZ resulted in suppressed levels of CYP2C11 mRNA starting at 6 h and continuing through 14 days (Fig. 2C). At 6 h, levels were decreased by 44, 59, and 62% in order of increasing dose compared with the vehicle treatment group (P ≤ 0.05, Fig. 2C). Levels remained consistently suppressed throughout the duration of the study. At the 1- and 4-day time points, mRNA levels were 54, 54, and 71% suppressed and 49, 45, and 54% suppressed compared with vehicle-treated animals (P ≤ 0.01, Fig. 2C). Fourteen days after administration, mRNA levels remained 39, 41, and 44% suppressed (P ≤ 0.01, Fig. 2C). Overall, administration of 5.7 × 1012 vp/kg resulted in the largest suppression of the three treatment groups.
Effect of Administration of Low Doses of AdlacZ on CYP3A Protein Expression, Catalytic Activity, and mRNA Levels. The results described above clearly demonstrated that 1 day after administration the lowest of the three vector doses (5.7 × 1010 vp/kg) significantly suppressed CYP3A2 catalytic activity and CYP2C11 protein expression, catalytic activity, and mRNA. As a result of these findings, a study was initiated to investigate the effect of three additional doses (5.7 × 106, 5.7 × 108, and 5.7 × 109 vp/kg) on the expression and function of these P450 isoforms at a single time point, 1 day after administration. None of these doses of AdlacZ altered CYP3A1/2 protein expression at the selected 1-day time point (Fig. 3A). Catalytic activity and mRNA production was also not significantly altered following administration of the three lower doses (P ≤ 0.05, Fig. 3, B and C).
Effect of Administration of Low Doses of AdlacZ on CYP2C11 Protein Expression, Catalytic Activity, and mRNA Levels. Administration of 5.7 × 106, 5.7 × 108, and 5.7 × 109 vp/kg significantly increased levels of CYP2C11 protein expression 86, 71, and 107%, respectively, compared with the vehicle treatment group (P ≤ 0.01, Fig. 4A). With respect to in vitro catalytic activity, 2α-hydroxytestosterone levels were 110, 118, and 53% above those of vehicle-treated animals (P ≤ 0.01, Fig. 4B). Levels of CYP2C11 mRNA remained unchanged 1 day after administration of 5.7 × 106 vp/kg. Treatment with 5.7 × 108 and 5.7 × 109 vp/kg effectively suppressed mRNA levels by 30 and 35% of vehicle-treated controls (P ≤ 0.01, Fig. 4C).
Effect of Administration of AdlacZ on Serum Transaminase Levels. Serum ALT and AST levels were measured to assess liver toxicity. Elevations of these enzymes occur with hepatic inflammation, injury, or disease (Henderson and Moss, 2001). Administration of the highest dose of AdlacZ (5.7 × 1012 vp/kg) significantly elevated ALT levels at all time points (Fig. 5A). Six hours after administration, levels were elevated 3-fold above baseline and 4-fold above baseline at 1 day (P ≤ 0.01, Fig. 5A). The greatest elevation, a 13.5-fold increase, was seen 4 days after administration. Levels remained elevated at 14 days, 4.5-fold above baseline, but tended toward recovery (P ≤ 0.01). AST levels were significantly elevated in all treatment groups (Fig. 5B). Six hours after administration of the higher doses (5.7 × 1010-5.7 × 1012 vp/kg) AST levels were 7-, 11-, and 4-fold above those of animals treated with vehicle, respectively (P ≤ 0.01, Fig. 5B). Levels remained significantly elevated at 1 day, 4-, 4-, and 3.5-fold and at 4 days, 4-, 4.5-, and 6-fold above baseline, respectively (statistical significance noted in the 5.7 × 1012 vp/kg group only). At 14 days, the two highest treatments continued to have elevated AST levels, both 3-fold greater than baseline (P ≤ 0.01). Following administration of the three lower doses (5.7 × 106, 5.7 × 108, and 5.7 × 109 vp/kg) AST levels were elevated 3-, 3-, and 4-fold, respectively, compared with vehicle-treated animals (P ≤ 0.01, Fig. 5D).
Histological Evaluation of Transgene Expression. Hepatic tissue sections were stained with X-gal to evaluate the degree of transgene expression after a single dose of AdlacZ (Fig. 6). Approximately 10% of the hepatocytes of an animal treated with 5.7 × 1010 vp/kg were positive for β-galatosidase (Fig. 6B). A dose of 5.7 × 1011 vp/kg successfully transduced 40% of hepatocytes (Fig. 6C), whereas 95% hepatocytes expressed the β-galactosidase transgene after a dose of 5.7 × 1012 vp/kg (Fig. 6D). Staining of tissues from animals treated with a saline bolus demonstrated that endogenous expression of β-galactosidase was not present in hepatic tissue (Fig. 6A). Administration of the doses ranging from 5.7 × 106-5.7 × 109 vp/kg resulted in undetectable levels of transgene (data not shown).
The data presented here show that a single intravenous dose of 5.7 × 1012 vp/kg AdlacZ significantly reduced levels of CYP3A2 protein expression over 4 days and catalytic activity and mRNA levels over the course of 14 days (P ≤ 0.05, Fig. 1, A-C). Administration of 5.7 × 1010 and 5.7 × 1011 vp/kg had comparable effects, although both the onset and depth of suppression varied. Studies investigating the effect of infection and inflammation on drug metabolism have primarily assessed P450 expression and activity for an acute time period of 48 to 72 h after exposure (Renton, 2000; Morgan, 2001). In the current study, we not only extended the time period to assess P450 expression, but found that levels were altered for 14 days after a single dose of virus. Clinically, this degree of suppression could have profound effects if there was a similar effect on the human equivalent CYP3A isoform, CYP3A4. In recent years, adenoviral vectors have become one of the most promising vectors in cancer therapy trials where a combination of virotherapy and chemotherapy are often applied (Kanerva and Hemminki, 2004). A variety of chemotherapeutic and antinausea medications used for cancer treatment, including vinorelbine, docetaxel, and granisetron, are metabolized by CYP3A4 (Rendic, 2002); thus, a suppression of this enzyme following vector administration could lead to altered pharmacokinetics and adverse drug reactions for an extended period of time in patients already enduring poor health.
Doses in the range of 5.7 × 1010 to 5.7 × 1012 vp/kg of AdlacZ suppressed CYP2C11 protein expression, catalytic activity, and mRNA levels as early as 6 h after administration (P ≤ 0.05, Fig. 3, A-C). This continued throughout the 14-day study period. Doses of AdlacZ in the range of 5.7 × 106 to 5.7 × 109 vp/kg significantly increased both CYP2C11 protein expression and in vitro catalytic activity (P ≤ 0.01, Fig. 4, A and B). Although this was somewhat unexpected, a similar biphasic pattern has been described in the context of glucocorticoid administration and CYP2C11. Low concentrations of glucocorticoid have been shown to significantly increase levels of 2C11 mRNA and protein expression, whereas higher concentrations suppress CYP2C11 (Iber et al., 1997). Bacterial and viral infections activate the hypothalamic-pituitary-adrenal axis (HPAA), which responds by releasing endogenous glucocorticoids (Turnbull et al., 2003). The manner by which the HPAA is activated impacts the amount of glucocorticoids released. Preclinical studies in several animal models have shown that adenoviral doses below 1010 viral particles are sequestered by Kupffer cells, whereas doses above this threshold saturate the Kupffer cell barrier and allow the remaining viral particles to transduce hepatocytes (Tao et al., 2000). Thus, we envision a model where doses in the range of 5.7 × 106 to 5.7 × 109 vp/kg, sequestered by Kupffer cells, activated the central limb of the HPAA and induced it to produce lower levels of glucocorticoids than when activated by higher doses of AdlacZ, resulting in an increase in CYP2C11. A similar model describing serum glucocorticoid levels in rats treated with bacterial lipopolysaccharide, which encompass this transition between 2C11 induction and repression, has been described (Iber et al., 1997).
Although we observed a biphasic regulation of CYP2C11 protein expression and activity at the 1-day time point, mRNA levels did not reflect this phenomenon (P ≤ 0.01, Fig. 4C). In light of the fact that recombinant adenoviruses induce many different cell-signaling pathways, we believe that our data may result from induced CYP2C11 protein synthesis by serum glucocorticoid levels coupled with the activation of cell-signaling pathways such as the mitogen-activated protein kinase pathway, also sensitive to glucocorticoids (Lasa et al., 2002), which confer altered patterns of translation and protein degradation later in the infection process. The kinetics of this sequence of events and dynamics between this pathway and CYP2C11 are currently under investigation in our laboratory.
Although the results presented here demonstrate that a single dose of adenovirus significantly alters the expression and function of enzymes responsible for the metabolism and clearance of xenobiotics, the exact mechanism responsible for this effect is unclear. In vitro and in vivo mechanistic studies suggest cytokines such as interleukin-1β, interleukin-6, tumor necrosis factor-α, and interferon-γ associated with the acute phase of the immune response are responsible for aberrations in drug metabolism after exposure to infectious agents (Morgan, 2001). Based on the established time course for cytokine production after treatment with recombinant adenovirus in a murine model (Tarpnell and Shanley, 2002), we believe that changes in CYP3A2 and CYP2C11 6 and 24 h after administration of high doses of adenovirus in the present study is related to the release of cytokines and chemokines during the acute phase of the immune response. Faced with a limited number of samples, we were unable to characterize secretion patterns of animals used in this study. Additional studies to address the role of cytokine production in our system are warranted.
The effect of adenoviral administration observed at later time points could be related in part to the second phase of the immune response. During this phase, cytotoxic T cells attack cells expressing newly synthesized viral proteins and transgene products (O'Riordan, 2002). Given this information, one might expect that the long-term suppression of CYP3A2 and CYP2C11 is the result of hepatocellular apoptosis. Physical signs of this phenomenon, however, were only detected in tissue sections from animals treated with the highest dose of recombinant virus (data not shown). Serum transaminases, which serve as an indirect measure of tissue damage, also do not support this hypothesis. ALT, largely concentrated in the liver, serves as a good measure for hepatic tissue damage (Henderson and Moss, 2001). Elevated ALT levels were only detected in animals treated with 5.7 × 1012 vp/kg, although all six dosing groups significantly altered P450 levels (P ≤ 0.01, Fig. 5, A and C). This is not surprising due to the fact that >90% of the hepatocytes were saturated with the β-galactosidase transgene in these animals (Fig. 6D). In contrast, AST found in a variety of tissues including the liver was consistently elevated in all groups at the 1-day time point regardless of the effect of AdlacZ on P450 expression and function (P ≤ 0.01, Fig. 5, B and D). The differences in elevations of AST after treatment with adenovirus warrants further investigation of P450 alterations in nonhepatic tissues.
The possibility that the transgene product is responsible for the observed alterations in P450 cannot be overlooked. Shifts in the activity of cellular machinery from maintenance of physiological homeostasis to production of transgene product and the toxicity of elevated levels of transgene present in hepatocytes may also contribute to the alterations in P450 that we observed, especially at the later time points as the β-galactosidase could still be detected in tissue sections of animals treated with the highest three doses of virus (data not shown). Additional studies defining the relationship between multiple cellular signaling pathways known to influence P450 transcription, such as those mediated by nuclear factor kappa B, Janus kinase signal transducer and activator transcription factor, and mitogen-activated protein kinase (Tian et al., 1999; Badger et al., 2003; Yasunami et al., 2004), which are also altered during adenovirus infection (Nemerow, 2002), may provide further insight into the long-term effect of a single dose of virus on P450 expression and function.
It is also important to note that the transgene used in the present study, β-galactosidase, elicits an immune response in the murine model (Everett et al., 2003). Although it would never be used in the clinic, this transgene is routinely used in preclinical studies due to the ease with which vector efficacy can be assessed (Cui et al., 1994). A potential risk of treating genetic diseases by gene replacement is a host immune response to the transgene regardless of the vector used for delivery (Zhou et al., 2004). The immune response against viral genes, capsid proteins, and the transgene remains to be a significant hurdle in the treatment of genetic disease and presents a rather complicated system in which to assess the mechanism by which recombinant adenoviruses alter hepatic P450. Additional in vivo studies including vectors containing less immunogenic “self” transgenes, such as murine erythropoietin or insulin, and those devoid of all viral coding sequences will allow us to determine whether the transcription of transgenes and viral proteins are key contributing factors in the down-regulation of P450.
In this report, we have shown for the first time the effects of a model gene therapy vector on the hepatic P450 enzyme system. To date, the area of vector delivery and its effects on drug metabolism has not been investigated. The data presented here should be considered in the development of safe and effective therapeutic regimens for patients participating in gene therapy or vaccine trials and in the assessment of potential problems that may arise with concomitant use of xenobiotics and gene therapy vectors. In addition, these findings may offer some explanation for some idiosyncratic adverse events previously described in gene therapy trials. However, additional studies to elucidate specific mechanisms by which viral infection and the immune response modulates the expression and function of several key enzymes responsible for drug metabolism are warranted.
- Received August 2, 2004.
- Accepted October 19, 2004.
This work was supported in part by a special grant from The University of Texas at Austin, Office of the Vice President for Research (M.A.C.) Grants GM69870 (M.A.C.) and GM60910 (L.J.B.) from the National Institutes of Health. S.M.C. was the recipient of a University of Texas at Austin Continuing Fellowship.
Data from this manuscript has been presented at the 2003 American Association of Pharmaceutical Scientists meeting in Salt Lake City, Utah.
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
ABBREVIATIONS: Ad, adenovirus; P450, cytochrome P450; vp, virus particles; DEPC, diethylpyrocarbonate; X-gal, 5-bromo-4-chloro-3-indolyl-β-d-galactoside; AdlacZ, first-generation recombinant adenovirus expressing E. coli β-galactosidase under the control of a cytomegalovirus promoter; pfu, plaque-forming units; NFDM, nonfat dry milk; RT-PCR, reverse transcription-polymerase chain reaction; AST, aspartate aminotransferase; ALT, alanine aminotransferase; HPAA, hypothalamic-pituitary-adrenal axis.
- The American Society for Pharmacology and Experimental Therapeutics