Abstract
Human organic anion-transporting polypeptide 1B1 (OATP1B1) is an important hepatic uptake transporter that can transport a wide variety of drugs. In the present study, we have generated and characterized a transgenic mouse model with specific and functional expression of human OATP1B1 (SLCO1B1) in the liver. Immunohistochemical staining revealed basolateral localization of transgenic OATP1B1 in the liver, whereas no expression of OATP1B1 was found in the kidney and small intestine. Using this transgenic model, the in vivo role of human OATP1B1 in the disposition of the anticancer drug methotrexate (MTX) was studied. In mice on a semisynthetic diet, the area under the plasma concentration-time curve for intravenous methotrexate in SLCO1B1 transgenic mice was 1.5-fold decreased compared with wild-type mice. Furthermore, the amount of MTX in the liver was markedly higher (∼2-fold) in the SLCO1B1 transgenic mice compared with wild-type mice, resulting in 2- to 4-fold higher liver-plasma ratios of MTX. Some murine liver Slco genes were markedly down-regulated on the semisynthetic diet compared with a standard diet, which probably reduced murine Oatp-mediated MTX uptake in the liver and therefore facilitated detection of the function of the transgenic OATP1B1. Taken together, these data demonstrate a marked and possibly rate-limiting role for human OATP1B1 in MTX elimination in vivo. Variation in OATP1B1 activity due to genetic polymorphisms, drug-drug interactions, and possibly dietary conditions may therefore play a role in the severity of MTX-related toxicity. SLCO1B1 transgenic mice could be a useful tool in studying the in vivo role of human OATP1B1 in drug pharmacokinetics.
Uptake transporters belonging to the superfamily of organic anion-transporting polypeptides (rodents, Oatp/Slco and human, OATP/SLCO) are nowadays recognized as important transmembrane proteins that can have a profound impact on the systemic pharmacokinetics, tissue distribution, and elimination of a wide range of drugs (König et al., 2006). Since the discovery of the first Oatp (rat Oatp1a1) in 1994 (Jacquemin et al., 1994), considerable effort has been put into the discovery of new substrates for Oatps/OATPs. However, most of these studies are performed in vitro, and models to investigate the role of Oatps/OATPs in vivo are limited. Therefore, the aim of this study was to obtain a useful model to study human OATP1B1 in vivo, by generating SLCO1B1 transgenic mice.
Human OATP1B1 (previously called OATP-C, LST-1, or OATP2; gene name, SLCO1B1) is highly expressed at the basolateral (sinusoidal) plasma membrane of hepatocytes and could play a key role in the uptake of compounds into the human liver (Abe et al., 1999; König et al., 2000; Tamai et al., 2000). In a recent study, however, SLCO1B1 mRNA was also detected in human enterocytes (Glaeser et al., 2007). OATP1B1 has a broad substrate specificity and seems to be involved in the transport of bile salts, bromosulphthalein, steroid conjugates, the thyroid hormones T3 and T4, and drugs like benzylpenicillin, rifampicin, pravastatin, pitavastatin, rosuvastatin, fexofenadine, and methotrexate (König et al., 2006; Matsushima et al., 2008). The importance of OATP1B1 in the therapeutic efficacy and toxic side effects of substrate drugs has been confirmed by several studies focusing on genetic polymorphisms in SLCO1B1. For example, a commonly occurring haplotype, SLCO1B1*15, containing single nucleotide polymorphisms A388G and T521C, has been associated with a strongly reduced transport functionality, markedly increased plasma levels (2-fold), and drastically reduced nonrenal clearance of the drugs pravastatin and pitavastatin in Japanese, Korean, and white populations (Nozawa et al., 2002; Nishizato et al., 2003; Niemi et al., 2004; Chung et al., 2005; Ho et al., 2007).
Methotrexate (MTX), a folate antimetabolite and a bicarboxylic organic anion, is widely used for the treatment of various types of cancer (i.e., breast cancer, head and neck cancer, lung cancer, and non-Hodgkin's lymphoma). It is also used to treat nonmalignant diseases, including psoriasis and rheumatoid arthritis (van Outryve et al., 2002; Wessels et al., 2008). Two independent studies show that OATP1B1 is able to transport MTX in vitro, suggesting the importance of a basolateral uptake transporter in MTX pharmacokinetics (Abe et al., 2001; Sasaki et al., 2004).
A, structure of the ApoE promoter-HCR1-driven expression cassette, containing human SLCO1B1 cDNA. Functional elements are represented approximately to scale. Translational start and stop codon and reading frame direction (bold arrow) for SLCO1B1 are indicated. Small arrows indicate the primers used for PCR detection. The probe used for Southern blot analysis (1.8 kb) is indicated below the cDNA. B, expression of human OATP1B1 in the liver, but not in kidney and small intestine, of SLCO1B1 transgenic versus wild-type mice fed with the standard diet (male) as detected by Western blotting (top panel). Two independently generated founder lines were analyzed (#1 and #2). HCLM, human crude liver membrane; Wt, wild-type; Tg, SLCO1B1 transgenic. Crude membrane protein (20 μg) was analyzed for all fractions. A molecular mass marker of 85 kD is indicated. Total protein staining (Ponceau S) confirmed equal loading across the lanes (bottom panel).
In this study, we have generated and characterized a transgenic mouse model that shows substantial and functional expression of human OATP1B1 specifically in the liver. Using this transgenic model, the role of human OATP1B1 in MTX disposition in vivo was studied. Our results indicate that, in vivo, OATP1B1 can be a rate-limiting factor in the clearance of MTX, illustrating the potential use of this model in assessing drug pharmacokinetics.
Materials and Methods
Animals. Mice were housed and handled according to institutional guidelines complying with Dutch legislation. The animals used in this study were male SLCO1B1 transgenic and wild-type mice of identical genetic background (FVB) between 9 and 14 weeks of age. Animals were kept in a temperature-controlled environment with a 12-h light/12-h dark cycle. Mice received a standard diet (AM-II; Hope Farms, Woerden, The Netherlands) and acidified water ad libitum. Three weeks before specified experiments, mice were fed with a semisynthetic diet (Reference diet 20% casein, 4068.02; Hope Farms).
Chemicals and Reagents. Methotrexate (100 mg/ml, Emthexate PF) was obtained from Pharmachemie (Haarlem, The Netherlands). Methoxyflurane (Metofane) was from Medical Developments Australia (Springvale, Victoria, Australia) and heparin (5000 IE/ml) was from Leo Pharma BV (Breda, The Netherlands). Bovine serum albumin, Fraction V, was obtained from Roche Diagnostics (Mannheim, Germany). Drug-free human plasma was obtained from healthy volunteers. The polyclonal antibody against human OATP1B1 (ESL clone) was a kind gift of Prof. Dr. D. Keppler (Deutsches Krebsforschungszentrum, Heidelberg, Germany) (König et al., 2000).
Transgene Construction. To achieve liver-specific expression of human OATP1B1, we generated a transgene construct as follows (Fig. 1A). pLIV-LE6 (kindly provided by Dr. J. Taylor, Gladstone Institute, University of California, San Francisco, CA) (Simonet et al., 1993) was completely digested with ClaI followed by a partial digestion with Asp718. The 7.6-kilobase (kb) fragment of the vector was isolated and used for ligation with human SLO1B1 cDNA later on. Polymerase chain reaction (PCR) was used to generate the SLCO1B1 construct with ClaI and Asp718 recognition sites at the 5′ and 3′ ends, respectively. The template, human SLCO1B1 cDNA in pBluescript SK-, was a kind gift from Dr. T. Abe (University Graduate School of Medical Science, Sendai, Japan). This cDNA represents the SLCO1B1*1a polymorphic variant, which is considered to be the wild-type variant (Iwai et al., 2004). The PCR product was digested with ClaI and Asp718 and ligated into the dephosphorylated 7.6-kb pLIV-LE6 fragment, yielding pLIV-LE6-SLCO1B1. Plasmid DNA was linearized, and irrelevant plasmid DNA was removed by restriction digestion with SalI and NgOM IV, followed by pronuclear injection into fertilized oocytes of FVB mice. Two-cell stage embryos were implanted into oviducts of pseudopregnant F1 fosters and carried to term.
PCR and Southern Blot Analysis. Transgenic founder lines were initially detected by PCR screen with forward 5′-GAAGGCTAACCTGGGGTGAG-3′ and reverse 5′-GGCAAAGCAGTCAAAACACC-3′ primers located within the apolipoprotein E (ApoE) intron 1 and SLCO1B1 cDNA, respectively, to yield a 450-base pair fragment (Fig. 1A). Southern blot analysis was used for the definitive identification of transgenic founders, as well as for the differentiation between heterozygous and homozygous individuals. DNA was extracted from the tail tips of mice (Laird et al., 1991), and a ∼1.8-kb EcoRV pLIV-LE6-SLCO1B1 fragment was used as a probe (Fig. 1A).
Western Blot Analysis. Crude membrane fractions from liver, kidney, and small intestine were prepared as described previously (Ogihara et al., 1996). The microsomal protein was quantified by the Bio-Rad protein assay based on the Bradford (1976) method (Bio-Rad, Veenendaal, The Netherlands), and 20 μg of the crude membrane protein was loaded on a 8% polyacrylamide Tris-HCL gel. After transfer, blots were probed with rabbit polyclonal antibody ESL (1:10,000) followed by horseradish peroxidase-labeled goat anti-rabbit secondary antibody (Dako Denmark A/S, Glostrup, Denmark). Equal protein loading across the lanes was confirmed by Ponceau S staining of the membranes after transfer.
Real-Time-PCR Analysis. RNA isolation from mouse livers, subsequent cDNA synthesis, and real-time (RT)-PCR using specific primers (QIAGEN, Hilden, Germany) for mouse Slco1a1, -1a4, -1b2, and -2b1 were performed as described previously (van Waterschoot et al., 2008b).
Clinical Chemical and Hematological Analysis. Standard clinical chemical analyses on EDTA plasma were performed on a Roche Hitachi 917 analyzer to determine levels of total and conjugated bilirubin, alkaline phosphatase, aspartate aminotransaminase, alanine aminotransaminase, γ-glutamyl transferase, lactate dehydrogenase, creatinine, ureum, Na+, K+, Ca2+, total protein, albumin, uric acid, cholesterol, and triglyceride. Hemoglobin, hematocrit, mean corpuscular volume, red and white blood cells, and platelets were analyzed in peripheral blood on a Beckman Coulter Ac·T Diff analyzer (Beckman Coulter, Fullerton, CA).
Immunohistochemical Analysis. Wild-type and SLCO1B1 transgenic livers were fixed in 4% phosphate-buffered formalin, embedded in paraffin, sectioned at 4 μm, and incubated with rabbit polyclonal ESL antibody (1:250 in phosphate-buffered saline) followed by horseradish peroxidase-labeled secondary antibody (Dako Denmark A/S). Nuclei were stained with hematoxylin and eosin according to standard procedures.
MTX Pharmacokinetics. MTX (100 mg/ml saline) was diluted to 2 mg/ml in saline and was injected as a single bolus into the tail vein of male mice (n = 3–4 for each group) at a dose of 10 mg/kg (5 μl/g b.wt.). Animals were killed at indicated time points by terminal bleeding through cardiac puncture under methoxyflurane anesthesia, and livers were isolated. Plasma fractions were isolated by centrifugation, and organs were homogenized in 4% bovine serum albumin using a Polytron homogenizer (Kinematica, Bohemia, NY). Levels of MTX and 7-hydroxy-methotrexate (7-OH-MTX) in plasma and homogenized organs were determined by high-performance liquid chromatography (HPLC) analysis as described previously (van Tellingen et al., 1989).
Pharmacokinetic Calculations and Statistics. The two-sided unpaired Student's t test was used to assess the statistical significance of differences between two sets of data. Results are presented as the means ± S.D. Differences were considered to be statistically significant when P < 0.05. Averaged concentrations for each time point were used to calculate the area under the plasma concentration versus time curve (AUC) from t = 0 to the last sampling time point by the linear trapezoidal rule. The S.E. of the AUC was calculated by the law of propagation of errors (Bardelmeijer et al., 2000). Results of the AUC measurements are presented as means ± S.E.
Immunolocalization of human OATP1B1 in the liver of SLCO1B1 transgenic mice. Paraffin-embedded liver of a SLCO1B1 transgenic mouse fed with the standard diet (male, founder line 1) was sectioned (4 μm) and stained with a rabbit polyclonal antibody against human OATP1B1 (brown). Nuclei were stained with hematoxylin/eosin (blue). The picture shows a basolateral staining pattern throughout the liver lobule of SLCO1B1 transgenic mice (A), which was strongest around the portal vein (B) and weaker (but positive) at the centrolobular region (C). Wild-type liver (male) did not show staining of OATP1B1 (D). Scale bars are indicated.
Results and Discussion
Transgenic Mice Show Liver-Specific and Stable Basolateral Expression of Human OATP1B1. Stable and specific expression of human OATP1B1 in the liver of transgenic mice was achieved by using an ApoE promoter-HCR1-driven expression cassette containing human SLCO1B1 cDNA (Fig. 1A). A similar expression cassette was used before to generate liver-specific CYP3A4 transgenic mice (van Herwaarden et al., 2005). Integration of the transgenic construct into the mouse genome was confirmed by PCR and Southern blot analysis (data not shown). Transgene transmission occurred at the expected Mendelian ratios, and two independent homozygous SLCO1B1 transgenic founder lines were generated. Homozygous SLCO1B1 transgenic mice were fertile and did not differ from wild-type mice in life span or body weights. Clinical chemical, hematological, and pathological analyses did not reveal any abnormalities. Crude membrane fractions of liver, small intestine, and kidney of SLCO1B1 transgenic mice and wild-type mice were analyzed for the expression of OATP1B1 by Western blotting (Fig. 1B). Both founder lines showed abundant expression of human OATP1B1 in their livers, which was roughly comparable with expression of OATP1B1 in a pooled human crude liver fraction. Detection of OATP1B1 in human crude liver fraction showed a main band approximately 80 kD, representing (N-glycosylated) OATP1B1 as was shown previously (König et al., 2000; Ho et al., 2006). OATP1B1 detection in the transgenic liver also revealed a main band at ∼80 kD (appearance of multiple bands in both human and transgenic samples probably reflects differences in N-glycosylation levels). We found no expression of the transgene in the small intestine and kidney of these mice (Fig. 1B). Immunohistochemical staining confirmed basolateral (sinusoidal) localization of human OATP1B1 throughout the liver lobules of transgenic mice (Fig. 2), as was shown for OATP1B1 expression in human liver (König et al., 2000; Ho et al., 2006). This supports the physiological relevance of this model. However, for SLCO1B1 transgenic livers, immunohistochemical staining was strongest around the portal vein (periportal; Fig. 2B), whereas weaker staining was found toward the central vein (centrolobular; Fig. 2C). Expression of transgenic OATP1B1 did not influence hepatic expression levels of murine Slco1a1, -1a4, -1b2, and -2b1 as measured by RT-PCR analysis (data not shown). Transgenic OATP1B1 expression was monitored over approximately five generations and was found to be stable (data not shown).
Expression Levels of EndogenousSlcoGenes in HumanSLCO1B1Transgenic Mice on Semisynthetic Diet. Pilot studies with high-dose MTX (50 mg/kg) in SLCO1B1 transgenic mice fed with a standard diet resulted in only minor differences between transgenic and wild-type mice [17% decrease in plasma AUC (6194 ± 220 versus 7506 ± 517 nmol · h l–1; P = 0.07) and maximally 2.0-fold increase in liver accumulation (P < 0.01) after intravenous administration; Supplemental Data 1]. These modest effects suggest that under standard conditions, the transgenic OATP1B1 activity does not go much beyond the endogenous murine Oatp activity. Therefore, we switched the mice from the standard diet to a semisynthetic diet, as we expected that this would result in down-regulation of some Oatps, because the semisynthetic diet contains less phytochemicals than the standard diet (composition of both diets are shown in Supplemental Data 2). Phytochemicals are well known inducers of detoxifying systems by activating pregnane X receptor, constitutive androstane receptor, and possibly other xenobiotic nuclear receptors (e.g., van Waterschoot et al., 2008a). RT-PCR analysis for a set of endogenous hepatic SLCO genes was performed to determine diet-dependent alterations in mRNA levels in the liver of wild-type and SLCO1B1 transgenic mice. We found that mouse Slco1a1, -1a4, -1b2, and -2b1 were indeed (markedly) down-regulated in livers of both wild-type and SLCO1B1 transgenic mice on the semisynthetic diet. These observed decreases were of the same order of magnitude in the two strains, with 1.7- and 2.6-fold (Slco1a1), 3.9- and 6.3-fold (Slco1a4), 2.5- and 1.7-fold (Slco1b2), and 1.5- and 1.5-fold (Slco2b1) decreases in wild-type and SLCO1B1 transgenic mice on the semisynthetic diet, respectively (Fig. 3). Because mouse Oatp1a4 is also a MTX transporter (Sasaki et al., 2004), the marked down-regulation of SLCO1a4 in mice fed with the semisynthetic diet might reduce background of murine Oatp-mediated MTX uptake in the liver. It is noteworthy that protein expression of the transgenic human OATP1B1 (which is controlled by the ApoE promoter) was not affected by the semisynthetic diet, as analyzed by Western blotting (data not shown).
Incidentally, our results show that expression of some Slco1 and -2 genes can be markedly affected by dietary conditions. Given the impact of OATP on drug disposition (see also below), it will be interesting to investigate whether this also applies in humans.
SLCO1B1Transgenic Animals Show Increased Hepatic Uptake of MTX and Lower Plasma Concentrations. To test the in vivo functionality of the transgene, we evaluated MTX disposition in SLCO1B1 transgenic versus wild-type mice fed with the semisynthetic diet. At various time points after intravenous administration of 10 mg/kg MTX, blood samples were taken and livers were isolated. The amounts of MTX, and its main metabolite 7-OH-MTX, were determined by HPLC analysis. Plasma AUC for MTX in SLCO1B1 transgenic mice was 1.5-fold decreased compared with wild-type mice (1261 ± 30.3 versus 1857 ± 112 nmol · hl–1; P < 0.05; Fig. 4A). The inset in Fig. 4 illustrates that the terminal elimination of plasma MTX is somewhat faster in SLCO1B1 transgenic mice compared with wild-type mice (semilog scale). This supports a role of transgenic OATP1B1 not just in short-term liver accumulation, but also in longer-term plasma clearance. Furthermore, the amount of MTX in the liver was markedly increased (∼2-fold) at all time points in the SLCO1B1 transgenic mice compared with wild-type mice (Fig. 4B). Liver to plasma ratios of MTX showed 2.2-, 2.6-, and 4.2-fold increases in mice expressing human OATP1B1 compared with wild-type mice at 15, 30, and 60 min after injection, respectively (P < 0.001; Fig. 4C). Plasma concentrations of 7-OH-MTX, which is primarily formed by aldehyde oxidase in the liver, were low and only significantly decreased in SLCO1B1 transgenic mice compared with wild-type mice 30 min after MTX administration (93.3 ± 24.3 versus 192.7 ± 67.8 nM; P < 0.05). 7-OH-MTX amount in the liver did not differ between transgenic and wild-type mice (data not shown).
Slco1a1, -1a4, -1b2, and -2b1 mRNA expression measured by RT-PCR. Results are expressed as the -fold change in expression of murine Slco genes in livers of wild-type and SLCO1B1 transgenic mice fed with the standard diet (male, n = 4) compared with mice fed with the semisynthetic diet (male, n = 4). Data were normalized against the expression of the endogenous control β-actin. Each sample was assayed in duplicate. **, P < 0.01; ***, P < 0.001.
Plasma and liver pharmacokinetics of MTX after intravenous injection (10 mg/kg) into wild-type and SLCO1B1 transgenic mice on the semisynthetic diet (male, n = 3–4). Plasma concentration versus time curves of methotrexate (A), methotrexate concentration in the liver (presented as percentage of dose) (B), and liver to plasma ratios of methotrexate (C) are shown. Each point/bar represents mean ± S.D. *, P < 0.05; **, P < 0.01; ***, P < 0.001.
MTX was earlier identified as a substrate for human OATP1B1 in vitro (Abe et al., 2001; Sasaki et al., 2004). To the best of our knowledge, this is the first study that shows that OATP1B1 is an important hepatic uptake transporter for MTX in vivo, with a rate-limiting role in MTX plasma elimination. Interindividual variation in MTX efficacy and toxicity correlating with plasma levels (Gorlick and Bertino, 1999) is a well recognized obstacle in the treatment of patients with cancer and rheumatoid arthritis. Because many genetic and functional variants in SLCO1B1 have been identified (Tirona et al., 2001; Nozawa et al., 2002; Nishizato et al., 2003; Niemi et al., 2004; Chung et al., 2005), the results of this study imply that variations in OATP1B1 activity, due to genetic polymorphism, dietary conditions, and perhaps drug-drug interactions, can have profound effects on plasma pharmacokinetics of MTX in patients and therefore partly explain interindividual variation. Thus, OATP1B1, in addition to other hepatic transporters like multidrug resistance protein 2 (Vlaming et al., 2006), might play an important role in MTX-related hepatic and/or plasma exposure-dependent toxicity. For example, MTX treatment is associated with acute and chronic liver damage (Hirvikoski et al., 1997; van Outryve et al., 2002). It would be interesting to see whether the observed toxicity could be correlated with genetic polymorphisms in SLCO1B1. At this moment, we can only speculate about this, and further research needs to investigate the clinical implications of MTX as an OATP1B1 substrate.
In conclusion, this study describes a novel liver-specific SLCO1B1 transgenic mouse model that provides an appropriate tool to study the role of human OATP1B1 in drug pharmacokinetics in vivo. In recent studies, Slco1b2 knockout mice have been described previously (Lu et al., 2008; Zaher et al., 2008). Because mouse Slco1b2 is orthologous to human SLCO1B1 and SLCO1B3 (Hagenbuch and Meier, 2004), Slco1b2 knockout mice might be a useful model to combine with our SLCO1B1 transgenic mice to generate a humanized model for analysis of human OATP1B1 function in vivo.
Acknowledgments
We thank Rahmen Bin Ali and Paul Krimpenfort for oocyte injection of the ApoE-SLCO1B1 cDNA construct; Rob Lodewijks and Enver Delic for analysis of blood and plasma samples; Martin van der Valk for histological and pathological examination of the mice; and Olaf van Tellingen for technical assistance with HPLC analyses.
Footnotes
-
This work was supported by GlaxoSmithKline [Grant S2918]; the Technical Sciences Foundation of the Netherlands Organization for Scientific Research (NOW/STW) [Grant BFA.6165]; and the Dutch Cancer Society [Grants 2000-2143, 2007-3764].
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.108.024315.
-
ABBREVIATIONS: OATP1B1, organic anion-transporting polypeptide 1B1; MTX, methotrexate; kb, kilobase; PCR, polymerase chain reaction; ApoE, apolipoprotein E; RT, real-time; 7-OH-MTX, 7-hydroxy-methotrexate; HPLC, high-performance liquid chromatography; AUC, area under plasma versus time curve; HCR1, hepatic control region-1.
-
↵
The online version of this article (available at http://dmd.aspetjournals.org) contains supplemental material. - Received September 1, 2008.
- Accepted November 13, 2008.
- The American Society for Pharmacology and Experimental Therapeutics
