LongTerm Stability of Primary Rat Hepatocytes in ... - Semantic Scholar
J BIOCHEM MOLECULAR TOXICOLOGY Volume 27, Number 3, 2013
Long-Term Stability of Primary Rat Hepatocytes in Micropatterned Cocultures Okechukwu Ukairo,1 Chitra Kanchagar,1 Amanda Moore,1 Julianne Shi,1 Jeannemarie Gaffney,1 Simon Aoyama,1 Kelly Rose,2 Stacy Krzyzewski,1 Jack McGeehan,1 Melvin E. Andersen,2 Salman R. Khetani,3 and Edward L. LeCluyse2 1 Hepregen 2 The
Corporation, Suite 1500, Medford MA 02155, USA
Hamner Institutes for Health Sciences, Research Triangle Park NC 27709, USA
3 Colorado
State University, Mechanical and Biomedical Engineering, Fort Collins CO 80523-1374, USA; E-mail: [email protected]
Received 8 October 2012; revised 17 November 2012; accepted 19 November 2012
ABSTRACT: Primary hepatocytes display functional and structural instability in standard monoculture systems. We have previously developed a model in which primary hepatocytes are organized in domains of empirically optimized dimensions and surrounded by murine embryonic fibroblasts (HepatoPacTM ). Here, we assess the long-term phenotype of freshly isolated and cryopreserved rat hepatocytes in a 96-well HepatoPac format. The viability, cell polarity (actin microfilaments, bile canaliculi), and functions (albumin, urea, Phase I/II enzymes, transporters) of fresh and cryopreserved rat hepatocytes were retained in HepatoPac at similar levels for at least 4 weeks as opposed to rapidly declining over 5 days in collagen/MatrigelTM sandwich cultures. Pulse or continuous exposure of rat HepatoPac to GW-7647, a selective agonist of PPARα, caused reproducible induction of CYP4A1 and 3hydroxy-3-methylglutaryl-CoA synthase over 4 weeks. In conclusion, rat HepatoPac in a 96-well format can be used for chronic dosing of highly functional hepatocytes and assessment of perturbed hepatocellular C 2013 Wiley Periodicals, Inc. J Biochem Mol pathways. Toxicol 27:204–212, 2013; View this article online at wileyonlinelibrary.com. DOI 10.1002/jbt.21469
KEYWORDS: Cocultures; Micropatterning; Rat Hepatocytes; Sandwich Cultures; Stromal Cells,
INTRODUCTION Rats are used in assessment of long-term effects of chemicals and drugs [1]. However, in vivo studies in rats with a large number of compounds can be C
Correspondence to: Salman R. Khetani. 2013 Wiley Periodicals, Inc.
expensive, require significant compound scaleup for dosing, are too slow for real-time feedback in a screening campaign, and at times are ill-suited for elucidating detailed molecular mechanisms. In the case of the liver, several in vitro models such as liver slices, microsomes, cell lines, and isolated primary hepatocytes have been developed to complement in vivo studies to address the aforementioned issues [2–5]. Liver slices have limited viability and are not amenable to high-throughput screening; cell-free microsomes lack the dynamic gene expression and intact cellular machinery required for dynamic cellular responses; and carcinoma-derived cell lines and immortalized hepatocytes display an abnormal repertoire and levels of liver-specific functions [6]. Therefore, isolated primary hepatocytes are widely considered to be the best choice for in vitro drug/chemical screening applications because they are simple to use and their cytoarchitecture remains intact [4,7]. However, primary hepatocytes are difficult to maintain in culture for more than a few days under conventional culture conditions that rely only on extracellular matrix (ECM) manipulations such as ECM-sandwich models utilizing collagen and/or MatrigelTM [8]. Thus, such conventional cultures are inadequate for evaluating the chronic effects of xenobiotics at physiologically relevant (low) dose levels and there is an urgent need for more predictive liver models. Khetani and Bhatia have previously developed a liver model with precise microscale architecture and optimal stromal interactions (micropatterned cocultures or HepatoPacTM ) that displays phenotypic stability of primary human hepatocytes for several weeks in vitro as assessed by major categories of liver-specific functions and gene expression [9]. In HepatoPac, primary hepatocytes are patterned onto ECM domains 204
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
of empirically optimized microscale dimensions and then surrounded by 3T3-J2 murine embryonic fibroblasts in a 24-well format. Under these culture conditions, phase I and II enzymes as well as transporters in primary human hepatocytes are expressed at high levels for several weeks, which has proven useful for improved detection of drug metabolites over suspension cultures and enzyme fractions [10]. Human HepatoPac outperforms conventional culture models (monolayer on rigid collagen, collagen gel sandwich, Matrigel overlay, Matrigel substratum) with respect to the magnitude and longevity of liver-specific functions. However, the ability of the HepatoPac culture model to stabilize the long-term functionality of primary rat hepatocytes from both fresh and cryopreserved sources in a higher throughput 96-well format remains undetermined. Here, we hypothesized that (a) the HepatoPac culture technique could stabilize the morphology and phenotypic functions of both fresh and cryopreserved rat hepatocytes (Sprague-Dawley) in vitro as opposed to a precipitous decline in sandwich cultures; (b) such stability could be obtained in a 96-well high-throughput format; and (c) stable rat HepatoPac cultures could be subsequently used to assess perturbation of hepatocellular pathways upon chronic administration with xenobiotics. We selected widely used and commercially available Sprague-Dawley fresh and cryopreserved rat hepatocytes to test our hypotheses. Finally, the effect of GW-7647, a potent and selective agonist of human and rodent PPARα [11] (promotes tumor progression), on short- and long-term induction of CYP4A1 [12] and 3-hydroxy-3-methylglutaryl-CoA synthase (HMGCS2) [13] gene expression over a 4-week period was examined to determine the long-term stability and robustness of the nuclear receptor response pathways.
MATERIALS AND METHODS Culture Preparation Freshly isolated and cryopreserved primary rat (Sprague-Dawley) hepatocytes were purchased from Life Technologies (Carlsbad, CA). Cryopreserved hepatocyte vials were thawed at 37◦ C for 90–120 seconds followed by dilution with 50 mL of prewarmed hepatocyte culture medium (HCM) (Hepregen Corporation, Medford, MA). The cell suspension was spun at 50×g for 5 min. The supernatant was discarded, cells were resuspended in HCM, and viability was assessed using trypan blue exclusion (typically 80–95%). Liverderived nonparenchymal cells, as judged by their size (∼10 μm in diameter) and morphology (nonpolygoJ Biochem Molecular Toxicology
DOI 10.1002/jbt
205
nal), were consistently found to be less than 1% in these preparations. To create micropatterned cocultures (HepatoPacTM , Hepregen Corporation) in 96-well plates, a hepatocyte pattern was first produced by seeding hepatocytes on rat-tail collagen (BD Biosciences, Franklin Lakes, NJ) type I-patterned substrates that mediate selective cell adhesion [9]. The cells were washed with medium 4–6 h later to remove unattached cells (leaving ∼5000 attached hepatocytes on 13 collagencoated islands within each well of a 96-well plate) and incubated in HCM. 3T3-J2 murine embryonic fibroblasts [14] were seeded 12–18 h later to create cocultures. Culture medium was replaced every 2 days (∼65 μL per well of a 96-well plate.). For sandwich cultures, approximately 56,000 hepatocytes were diluted in a previously reported serumsupplemented hepatocyte medium [7] and seeded per well in a 96-well format, where each well was coated with rat-tail collagen type I for 2 h and washed twice with 1× phosphate buffered saline (PBS) prior to seeding. Four hours after seeding, cells were washed twice with culture medium and then 0.25 mg/mL Matrigel (BD Biosciences) was diluted in a previously reported serum-free hepatocyte medium [7] and applied to the cultures overnight. Medium was changed daily with serum-free medium for the duration of the study.
Immunofluorescent Staining Cultures were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA) for 15 min at room temperature (RT), washed three times with 1× PBS, and stored at 4◦ C for up to 2 weeks prior to further processing. Fixed cells were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO) for 10 min on a rocker at RT, washed three times with 1× PBS, and then incubated in blocking buffer containing 1% BSA (bovine serum albumin; Sigma-Aldrich) and 5% goat serum (Life Technologies, Carlsbad, CA) in 1× PBS for 30 min at RT. Next, cells were incubated at RT with primary antibody for MRP-2 (Alexis Biochemicals, San Diego, CA) diluted to 1:200 in 1× PBS with 0.2% BSA. After washing 5 times with 1× PBS containing 0.1% Tween-20 (wash buffer), cells were incubated with fluorescein-labeled secondary antibody (Life Technology) diluted at 1:500 in 1× PBS with 0.2% BSA for 1 at RT, and then washed 5 times with wash buffer. For staining actin, rhodamine-labeled phalloidin (Life Technologies) in 1× PBS supplemented with 1% BSA and 5% goat serum was added to cells for 20 min at RT. Cultures were counterstained with 0.5 μg/mL Hoechst (Molecular Probes, Eugene, OR) dye for 5 min at RT, followed by three washes in 1× PBS. Specimens were observed and recorded using a Nikon Eclipse TE200
206
UKAIRO ET AL.
microscope equipped with a SPOT digital camera (SPOT Diagnostic Equipment, Sterling Heights, MI), and PixeLINK uScope Software was used for digital image acquisition.
Hepatocyte Functionality Assays Aliquots of culture medium were collected from culture wells at different time points and stored at –20◦ C. The urea concentration was assayed in culture supernatant with a colorimetric endpoint assay utilizing diacetyl monoxime with acid and heat (Stanbio Labs, Boerne, TX). The albumin content in spent medium was measured using an albumin-specific enzyme-linked immunosorbent assay (MP Biomedicals, Irvine, CA) with horseradish peroxidase detection and 3,3’,5,5’-tetramethylbenzidine (TMB, Fitzgerald Industries, Concord, MA) as the substrate [9]. Overall cytochrome P450 activity was measured by direct incubation of cells for 1 h with 40 μM of 7-benzyloxy-4trifluoromethylcoumarin (BFC; BD Biosciences), which is converted to 7-hydroxytrifluoromethylcoumarin (7-HFC) by CYP450s. Reactions were stopped by collecting the incubation medium from plates, and medium samples were then incubated with βglucuronidase/arylsulfatase (Roche Applied Science, Indianapolis, IN) for 4 h at 37◦ C. Fluorimetric quantification of 7-HFC formed during incubations was performed as previously described [15]. Total cell content was estimated via phase contrast micrographs by manually counting the number of hepatocytes in three representative islands from each condition 1 day after seeding, and then multiplying average cell count per island by total number of islands designed for each well using microfabrication technologies (i.e., an average of 13 collagen-coated islands within each well of a 96well plate). For sandwich cultures, the number of cells seeded per well was used for normalization, which is an overestimate of the actual number of hepatocytes that survive after a few days of culture. Results were expressed as picomoles of metabolite formed per hour per million cells. In addition, the metabolism of model CYP substrates (Sigma-Aldrich) in rat HepatoPac was examined over several weeks. To evaluate CYP3A1/3A2, CYP2D2, CYP2C6, CYP2B1, and CYP4A activities over time, cells were exposed to 25 μM midazolam or 3 μM CYP3A-Glo by Promega (Madison, WI), 10 μM dextromethorphan, 25 μM diclofenac, 500 μM bupropion, and 100 μM lauric acid, respectively, for 30 min at 37◦ C [16–19]. Supernatants were collected, and the concentration of the appropriate metabolite was determined via liquid chromatography/mass spectrometry (Integrated Analytical Services, San Francisco, CA). The 4-hydroxylation of mi-
Volume 27, Number 3, 2013
dazolam, 4-demethylation of dextromethorphan, 4hydroxylation of diclofenac, bupropion hydroxylation, and 12-hydroxylation of lauric acid was assessed via LC-MS/MS (Integrated Analytical Solutions, Berkeley, CA). Phase II enzyme metabolism in rat HepatoPac was assessed by glucuronidation and sulfation of 50 μM 7-hydroxycoumarin (7-HC) to 7-HC-glucuronide and 7-HC-sulfate, respectively. Conversion of CYP3AGlo into luminescent metabolite was quantified as per manufacturer’s (Promega) instructions using a Centro XS luminometer (Berthold Technologies, Germany).
Transporter Assays Hepatic transport studies were conducted using two different substrates: an anionic compound, estradiol 17 β-D-glucuronide (E2 17ßG) and bile acid taurocholate, which are actively transported by sinusoidal proteins OATP (organic anion transport protein) and NTCP (sodium taurocholate transport protein), respectively [20]. Cell cultures were preincubated with calcium-supplemented buffer (Life Technologies) for 10 min at 37◦ C before adding 1 μM cold substrate plus 0.04 μCi/well tritium-labeled substrates (Perkin-Elmer, Boston, MA) for 10 min. Uptake reactions were stopped by placing plates on ice and subsequent washing of wells 3 times with ice-cold calcium supplemented buffer. Plates were sealed and stored at –20◦ C until analyzed by liquid scintigraphy. Cells were treated with 1× PBS with 1% Triton X-100, shaken for 20 min, harvested and transferred to a scintillation vial containing liquid scintillation fluid. Radioactivity in each sample was counted using an LS6500 multipurpose scintillation counter (Beckman Coulter, Indianapolis, IN). The uptake rate of individual substrates is reported as disintegrations per minute (DPM) per million cells.
Gene Expression Profiling Hepatocyte gene expression was analyzed using R -based quantitative real-time PCR (qRT-PCR, TaqMan Life Technologies). Cell lysis and RNA purification R mini kit (Qiagen, were performed using the RNeasy Valencia, CA). Transcription to cDNA was accomplished using the high capacity cDNA reverse transcription kit according to manufacturer’s instructions (Life Technologies), and 2 μg was used for each qPCR reaction. FAM-labeled primer/probe sets for genes of interest and Path-IDTM qPCR Master Mix were obtained from Life Technologies. RT-PCR was perR realplex instrument (Epformed on a Mastercycler pendorf, Hamburg, Germany), and data were analyzed using the realplex software. Gene expression was normalized to the housekeeping genes α-actin and β-tubulin. J Biochem Molecular Toxicology
DOI 10.1002/jbt
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
207
FIGURE 1. Morphology and polarity of primary rat hepatocytes in HepatoPac over the course of several weeks. Hepatocytes maintained prototypical polygonal shape, distinct nuclei/nucleoli, and bile canaliculi (arrows) for at least 4 weeks (days 5, 15, and 27 shown) in HepatoPac as assessed via phase contrast microscopy (A). Fluorescently labeled actin microfilaments in hepatocytes were primarily associated with the peripheral margins of the hepatocytes, including the edges of bile canaliculi (B). Fluorescently labeled multidrug resistance protein 2 (MRP2), an efflux transporter, was localized on the bile canaliculi between hepatocytes (C). Representative images shown are for cryopreserved hepatocytes, but similar trends were seen with freshly isolated hepatocytes. Scale bar is 100 μm. Inset shows a cluster of a few hepatocytes in each panel.
Dosing with GW-7647 Rat HepatoPac cultures were dosed with the PPAR agonist GW-7647 (Sigma-Aldrich) at 0.01, 0.1, and 1 μM (DMSO concentration at 0.1% vol/vol in all conditions), and its effect on CYP4A1 and 3-hydroxy3-methylglutaryl-CoA synthase (HMGCS2) transcript expression in rat HepatoPac as a function of dose and time of exposure was evaluated. Cell lysates were collected for gene expression analyses following 6 and 72 h incubation of cultures with compound. Compound incubations were carried out once per week for 4 weeks in separate rat HepatoPac cultures (96-well format). In addition, rat HepatoPac cultures were incubated with 1 μM GW-7647 continuously for 4 weeks and redosed every 2 days with each medium change. Cell lysates were collected at 1, 2, 3, and 4 weeks for CYP4A1 and HMGCS2 expression analyses as described above.
RESULTS We verified previous findings in the literature that primary rat hepatocytes showed a gradual decline in prototypic morphology in sandwich culJ Biochem Molecular Toxicology
DOI 10.1002/jbt
tures (Supplemental Figure 1 in the Supporting Information) over the course of 10 days [8]. In contrast, fresh and cryopreserved hepatocytes from the same donors as that used for sandwich cultures, maintained in vivo-like polygonal shape, distinct nuclei/nucleoli and visible bile canaliculi in HepatoPac for at least 4 weeks in vitro (Figure 1A). Actin microfilaments in hepatocytes in HepatoPac were primarily associated with the peripheral margins of the hepatocytes, including the edges of bile canaliculi (Figure 1B), as reported previously in hepatocytes in vivo [21]. In addition, multidrug resistance protein 2 (MRP2) protein, a transporter [22], was localized on the bile canaliculi between hepatocytes (Figure 1C), which is indicative of polarized hepatocytes. Next, major liver-specific functions were assessed in the two culture models, specifically albumin secretion (a marker of liver protein synthesis), urea secretion (a marker of nitrogen metabolism), and overall CYP450 activity (nonspecific) using a fluorometric substrate. HepatoPac cultures took ∼5 days to stabilize functionally and retained high levels of functions for at least 4 weeks in vitro (Figures 2A–2C). On the other hand, sandwich cultures showed a gradual decline in CYP450 activity and after 5 days of culture, activity
208
UKAIRO ET AL.
Volume 27, Number 3, 2013
FIGURE 2. Major liver-specific functions in HepatoPac and sandwich cultures created using freshly isolated and cryopreserved primary rat hepatocytes. Albumin secretion (A), urea synthesis (B), and overall CYP450 activity (C) in HepatoPac is shown over the course of 4 weeks. A decline in overall CYP450 activity is shown for sandwich cultures (rigid collagen with Matrigel overlay) over the course of 9 days (D); however, similar trends were observed with urea secretion. CYP450 activity was assessed via the hydroxylation of 7-benzyloxy-4-trifluoromethylcoumarin into 7- hydroxy-4-trifluoromethylcoumarin (7-HFC).
was ∼5% of day 1 sandwich and overall HepatoPac levels (Figure 2D). Urea secretion in sandwich cultures showed near identical kinetics of decline as CYP450 activity; however, albumin secretion remained relatively stable in sandwich cultures, but 8–10 fold lower than HepatoPac when normalized to cell numbers (data not shown). Furthermore, fresh and cryopreserved hepatocytes showed relatively similar levels and kinetics of functions in both culture models. Because cryopreserved hepatocytes are a more convenient and on-demand source for creation of cultures for in vitro investigations [23], they were further characterized in the HepatoPac culture model for activities of multiple CYP450 isoforms as well as Phase II-mediated conjugation reactions (Figure 3). CYP2B, 3A, and 4A isoforms took 8–10 days to reach maximal levels and remained relatively stable for ∼4 weeks. CYP2C and 2D, on the other hand, reached maximal levels in ∼6 days and showed a 50% decline in activities between 18 and 26 days of culture. CYP3AGlo (luminescent-based assay from Promega, Madison, WI) showed similar kinetics as formation of 4-hydroxy midazolam, suggesting that the rapid luminescent assay is suitable for probing CYP3A activity. Phase II glucuronidation and sulfation of 7-hydroxycourmarin took ∼6 days to reach maximal levels and remained relatively stable for ∼4 weeks in the HepatoPac system.
Next, the time-dependent accumulation of radioactive estradiol 17 β-D-glucuronide (E2 17ßG, a substrate for organic anion transport protein or OATP) and taurocholate (a substrate for sodium taurocholate transport protein or NTCP) was assessed in the culture models (Figures 4A–4B). Fresh and cryopreserved hepatocytes in HepatoPac showed remarkably similar uptake activities for ∼4 weeks. The data indicate that maximal levels of uptake activities of these substrates was achieved by day 15-21 of HepatoPac culture and remained stable until at least day 27; however, by day 9 of culture, ∼70% to 75% of the maximal uptake activities were observed. Sandwich cultures, on the other hand, showed differences in uptake activities across fresh and cryopreserved hepatocytes after 2 days of culture (activities in cryopreserved hepatocytes were ∼40% of activities in fresh ones). Furthermore, the uptake activities of the fresh hepatocytes in sandwich cultures after 2 days of culture were 70% and 115% of HepatoPac levels for E2 17ßG and taurocholate, respectively; however, by the 5th day of culture, uptake activities in sandwich cultures had declined to 18% and 8% of HepatoPac levels for E2 17ßG and taurocholate, respectively (Figure 4C). We next evaluated the functionality of the bile canaliculi network in HepatoPac via transporter-mediated secretion of the MRP2 (ABCC2 or ATP-binding cassette, sub-family C (CFTR/MRP), member 2) substrate, 5-(and-6)- carboxy-2’,7’-dichlorofluorescein diacetate J Biochem Molecular Toxicology
DOI 10.1002/jbt
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
209
FIGURE 3. Activities of major CYP450 and phase II conjugation enzymes in HepatoPac created using cryopreserved rat hepatocytes. Enzyme activities over the course of 4 weeks were assessed via luciferin release from Promega’s CYP3A-Glo assay (A), midazolam hydroxylation by CYP3A (A), diclofenac hydroxylation by CYP2C (B), lauric acid hydroxylation by CYP4A (B), dextromethorphan demethylation into dextrorphan by CYP2D (C), bupropion hydroxylation by CYP2B (C), and phase II-mediated glucuronidation and sulfation of 7-hydroxycoumarin or 7-HC (D).
FIGURE 4. Transporter activities in HepatoPac and sandwich cultures created using freshly isolated and cryopreserved rat hepatocytes. Hepatocytes in HepatoPac take up prototypical transporters substrates, estradiol 17 β-D-glucuronide (E2 17ßG) (A) and bile acid taurocholate (B) over the course of 4 weeks. A decline in uptake of taurocholate is shown for sandwich cultures (rigid collagen with Matrigel overlay) over the course of 9 days (C); however, near identical trends were observed with E2 17ßG. Functional bile canaliculi were present between hepatocytes in HepatoPac (D) as assessed via transporter-mediated secretion of the fluorescent dye, 5-(and-6)- carboxy-2’,7’-dichlorofluorescein diacetate succinimidyl ester, into bile canaliculi. Image from cultures 2 weeks old is shown; however, similar trends were seen for at least 4 weeks.
succinimidyl ester, into bile canaliculi between hepatocytes. After about 20 min of pretreatment of cells with the dye, uniform, intense fluorescence was observed in nearly all the canalicular networks throughout the hepatocyte islands. No fluorescent staining was observed between hepatocytes and the fibroblasts (Figure 4D). J Biochem Molecular Toxicology
DOI 10.1002/jbt
Finally, we evaluated the utility of the HepatoPac model using cryopreserved rat hepatocytes for chronic/repeat dosing with enzyme inducers and assessment of known upregulated pathways. Incubation of rat HepatoPac for 6 or 72 h with varying concentrations of GW-7647, a potent PPARα agonist, caused a
210
UKAIRO ET AL.
Volume 27, Number 3, 2013
FIGURE 5. GW-7647 dose- and time-dependent induction of CYP4A1 and HMGCS2 transcripts in HepatoPac created using cryopreserved rat hepatocytes. Dose- and time-dependent upregulation of CYP4A1 (A) and HMGCS2 (B) mRNA was seen after exposure of HepatoPac to GW-7647. The effect of culture age (weeks 1–4) on GW-7647-mediated upregulation of CYP4A1 and HMGCS2 transcripts is shown after exposure of HepatoPac cultures for 6 h at a GW-7647 dose of 0.1 μM (C). Continuous exposure (fresh compound every 2 days) with GW-7647 at 1 μM caused upregulation of CYP4A1 and HMGCS2 transcripts over the course of 4 weeks (D). VC is vehicle controls.
dose- and time-dependent induction of PPARα target genes CYP4A1 and HMGCS2 (Figures 5A and 5B). After 6 h of exposure, both CYP4A1 and HMGCS2 showed ∼9 fold induction over vehicle controls at 1 μM of GW7647, while after 72 h of incubation the maximal fold induction was 71 and 29 for CYP4A1 and HMGCS2, respectively. Induction of these genes after 6 h exposure to 1 μM GW-7647 was seen at different culture ages; however, 4 week old cultures showed fold induction levels that were ∼50% of levels seen in the first 3 weeks of culturing (Figure 5C). Furthermore, continuous exposure of HepatoPac to 1 μM GW-7647 (with medium change every 2 days) enabled assessment of induction kinetics of both transcripts over the course of 4 weeks (Figure 5D). CYP4A1 remained induced 27 fold after 1 and 2 weeks of continuous exposure and that dropped by ∼50% to ∼12.5 fold and 14 fold after 3 and 4 weeks of exposure, respectively. HMGCS2, on the other hand, showed a 107 fold induction after 1 week of exposure, 29 fold after 2 weeks, 15 fold after 3 weeks, and an increase to 42 fold after 4 weeks.
DISCUSSION Predicting the effects of compounds in humans from animal or in vitro data continues to be a crit-
ical challenge facing the chemical and pharmaceutical industries [24, 25]. Progress in implementing costeffective screening strategies for an ever-increasing array of chemicals depends on physiologically relevant in vitro models of animal and human livers [26, 27]. However, current model systems such as primary hepatocyte monocultures and sandwich cultures are limited in their ability to maintain hepatocyte functionality over an extended period of time [2, 7]. Improved, longlasting hepatocyte cultures could revolutionize drug screening. In undertaking this study, we looked to demonstrate the potential of the HepatoPac micropatterned coculture system to mimic the native liver architecture and retain liver-specific functions over an extended period of time. We report here that rat hepatocytes cultured in HepatoPac demonstrate functional behavior at higher levels and for significantly longer periods of time than those maintained under the conventional two-dimensional (2D) static sandwich culture conditions. Specifically, in both primary and cryopreserved rat hepatocytes cultured in HepatoPac, cell viability, histotypic cell architecture, phase 1 and 2 metabolic activity, gene expression, and albumin/urea production were maintained up to 4 weeks and found to be superior to standard 2D sandwich culture.
J Biochem Molecular Toxicology
DOI 10.1002/jbt
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
The long-term stability of the classical peroxisomeproliferator activated receptor pathway in rat HepatoPac was determined by evaluating mRNA levels for known PPARα target genes, CYP4A1 [12] and HMGCS2 [13], after treatment with the selective agonist, GW-7647. The PPARα pathway remained responsive to agonist in HepatoPac for at least 4 weeks and exhibited dose- and time-dependent activation. We did, however, observe that the magnitude of target gene expression dropped by ∼50% when 4 week old cultures were subjected to short-term (hours) exposure to GW7646, suggesting potential downregulation of PPARα expression and/or its cofactors. HepatoPac’s design is based on a standard (static fluid) 96-well plate format, and the optimized micropatterned cell configuration and cell ratios support most of the major functions of primary hepatocytes, including key phase 1 and 2 biotransformation pathways. In the HepatoPac system, hepatocytes are thought to be responding to enhancements in cell–cell interactions created by the paracrine and heterotrophic effects of the hepatocyte-stromal coculture environment [28, 29]. We and others have shown that 3T3 fibroblasts and other nonliver derived stroma can induce functions in both animal and human hepatocytes, suggesting possible conservation of mechanisms [9, 29, 30]. SpragueDawley rat hepatocytes were chosen here given commercial availability of fresh and cryopreserved lots and their historical use as a toxicology species; however, other studies with Lewis [31] and Wistar Han rats (unpublished) suggest compatibility of HepatoPac with multiple strains. The results from these studies indicate that HepatoPac represents a robust model for the examination of prolonged effects of xenobiotics on the perturbation of hepatic pathways. The cumulative effects of parent compound and major metabolites formed during the course of continuous incubation in the HepatoPac system could be more easily determined, providing more relevant information about the pharmaco and toxicokinetic properties of compounds. The magnitude and dynamics of perturbations on nuclear receptor pathways, such as PPARα, upon chronic exposure to compounds, could be better compared to in vivo chronic dosing studies and utilized for more accurate in vitro to in vivo correlations. Finally, responsiveness of rat HepatoPac to stimulation with other nuclear receptor agonists as well as bioactivated compounds such as acetaminophen [32] is under investigation.
SUPPORTING INFORMATION Supplemental Figure 1 is available from the corresponding author on request. This figure shows the J Biochem Molecular Toxicology
DOI 10.1002/jbt
211
morphology of collagen/Matrigel sandwich cultures (days 2, 5, and 10 postseeding) created using freshly isolated Sprague-Dawley primary rat hepatocytes in a 96-well format. See the Methods section for details on the preparation of sandwich cultures.
ACKNOWLEDGMENTS Funding for this project was provided by the LongRange Research Initiative of the American Chemistry Council (ACC).
REFERENCES 1. Klaassen CD, Casarett LJ, Doull J. Casarett and Doull’s toxicology: The basic science of poisons. New York: McGraw-Hill; 2013. 2. Gebhardt R, Hengstler JG, Muller D, Glockner R, Buenning P, Laube B, Schmelzer E, Ullrich M, Utesch D, Hewitt N, Ringel M, Hilz BR, Bader A, Langsch A, Koose T, Burger HJ, Maas J, Oesch F. New hepatocyte in vitro systems for drug metabolism: metabolic capacity and recommendations for application in basic research and drug development, standard operation procedures. Drug Metab Rev 2003;35(2–3):145–213. 3. Sivaraman A, Leach JK, Townsend S, Iida T, Hogan BJ, Stolz DB, Fry R, Samson LD, Tannenbaum SR, Griffith LG. A microscale in vitro physiological model of the liver: predictive screens for drug metabolism and enzyme induction. Curr Drug Metabol 2005;6(6):569–591. 4. Guillouzo A. Liver cell models in vitro toxicology. Environ Health Pers 1998;106(Suppl 2):511–532. ´ MJG, Houston JB, Hallifax D, Brown 5. Hewitt NJ, Lechon HS, Maurel P, Kenna JG, Gustavsson L, Lohmann C, Skonberg C and others. Primary hepatocytes: current understanding of the regulation of metabolic enzymes and transporter proteins, and pharmaceutical practice for the use of hepatocytes in metabolism, enzyme induction, transporter, clearance, and hepatotoxicity studies. Drug Metab Rev 2007;39(1):159–234. 6. Wilkening S, Stahl F, Bader A. Comparison of primary human hepatocytes and hepatoma cell line Hepg2 with regard to their biotransformation properties. Drug Metab Disposition: Biol Fate Chem 2003;31(8):1035–1042. 7. Lecluyse EL. Human hepatocyte culture systems for the in vitro evaluation of cytochrome P450 expression and regulation. Eur J Pharm Sci 2001;13(4):343–368. 8. LeCluyse E, Bullock P, Madan A, Carroll K, Parkinson A. Influence of extracellular matrix overlay and medium formulation on the induction of cytochrome P-450 2B enzymes in primary cultures of rat hepatocytes. Drug Metab Dispos 1999;27(8):909–915. 9. Khetani SR, Bhatia SN. Microscale culture of human liver cells for drug development. Nat Biotechnol 2008;26(1):120–126. 10. Wang WW, Khetani SR, Krzyzewski S, Duignan DB, Obach RS. Assessment of a Micropatterned Hepatocyte Coculture System to Generate Major Human Excretory and Circulating Drug Metabolites. Drug Metab Dispos: Biol Fate Chem 2010;38(10):1900–1905.
212
UKAIRO ET AL.
11. Lee JY, Hashizaki H, Goto T, Sakamoto T, Takahashi N, Kawada T. Activation of peroxisome proliferatoractivated receptor-alpha enhances fatty acid oxidation in human adipocytes. Biochem Biophys Res Commun 2011;407(4):818–822. 12. Elcombe CR, Elcombe BM, Foster JR, Chang SC, Ehresman DJ, Butenhoff JL. Hepatocellular hypertrophy and cell proliferation in Sprague-Dawley rats from dietary exposure to potassium perfluorooctanesulfonate results from increased expression of xenosensor nuclear receptors PPARalpha and CAR/PXR. Toxicology 2012;293(1– 3):16–29. 13. Rakhshandehroo M, Hooiveld G, Muller M, Kersten S. Comparative analysis of gene regulation by the transcription factor PPARalpha between mouse and human. PLoS One 2009;4(8):e6796. 14. Rheinwald JG, Green H. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 1975;6(3):331– 343. ´ ´ MJ. 15. Donato MT, Jim´enez N, Castell JV, Gomez-Lech on Fluorescence-based assays for screening nine cytochrome P450 (P450) activities in intact cells expressing individual human P450 enzymes. Drug Metabol Dispos: Biol Fate Chem 2004;32(7):699–706. 16. Kobayashi K, Urashima K, Shimada N, Chiba K. Substrate specificity for rat cytochrome P450 (CYP) isoforms: screening with cDNA-expressed systems of the rat. Biochem Pharmacol 2002;63(5):889–896. 17. Madan A, DeHaan R, Mudra D, Carroll K, LeCluyse E, Parkinson A. Effect of cryopreservation on cytochrome P450 enzyme induction in cultured rat hepatocytes. Drug Metab Dispos 1999;27(3):327–335. 18. Burke MD, Thompson S, Weaver RJ, Wolf CR, Mayer RT. Cytochrome P450 specificities of alkoxyresorufin Odealkylation in human and rat liver. Biochem Pharmacol 1994;48(5):923–936. 19. Pekthong D, Desbans C, Martin H, Richert L. Bupropion hydroxylation as a selective marker of rat CYP2B1 catalytic activity. Drug Metab Dispos 2012;40(1):32–38. 20. Consortium IT, Giacomini KM, Huang S-M, Tweedie DJ, Benet LZ, Brouwer KLR, Chu X, Dahlin A, Evers R, Fischer V and others. Membrane transporters in drug development. Nat Rev Drug Disc 2010;9(3):215– 236.
Volume 27, Number 3, 2013
21. Hooser SB, Beasley VR, Waite LL, Kuhlenschmidt MS, Carmichael WW, Haschek WM. Actin filament alterations in rat hepatocytes induced in vivo and in vitro by microcystin-LR, a hepatotoxin from the blue-green alga, Microcystis aeruginosa. Vet Pathol 1991;28(4):259–266. 22. Li N, Bi Y-A, Duignan DB, Lai Y. Quantitative expression profile of hepatobiliary transporters in sandwich cultured rat and human hepatocytes. Molec Pharm 2009;6(4):1180– 1189. 23. Bi Y-A, Kazolias D, Duignan DB. Use of cryopreserved human hepatocytes in sandwich culture to measure hepatobiliary transport. Drug Metabol Dispos: Biol Fate Chem 2006;34(9):1658–1665. 24. Chen M, Vijay V, Shi Q, Liu Z, Fang H, Tong W. FDAapproved drug labeling for the study of drug-induced liver injury. Drug Discov Today 2011;16(15–16):697–703. 25. Kaplowitz N. Idiosyncratic drug hepatotoxicity. Nature Rev Drug Discov 2005;4(6):489–499. ´ ´ MJ, Lahoz A, Gombau L, Castell JV, Do26. Gomez-Lech on nato MT. In vitro evaluation of potential hepatotoxicity induced by drugs. Curr Pharma Des 2010;16(17):1963– 1977. 27. Rawlins MD. Cutting the cost of drug development? Nature Rev Drug Discov 2004;3(4):360–364. 28. Khetani SR, Chen AA, Ranscht B, Bhatia SN. Tcadherin modulates hepatocyte functions in vitro. FASEB J 2008;22(11):3768–3775. 29. Khetani SR, Szulgit G, Del Rio JA, Barlow C, Bhatia SN. Exploring interactions between rat hepatocytes and nonparenchymal cells using gene expression profiling. Hepatology (Baltimore, Md) 2004;40(3):545–554. 30. Bhatia SN, Balis UJ, Yarmush ML, Toner M. Effect of cellcell interactions in preservation of cellular phenotype: cocultivation of hepatocytes and nonparenchymal cells. FASEB J 1999;13(14):1883–1900. 31. Bhatia SN, Balis UJ, Yarmush ML, Toner M. Probing heterotypic cell interactions: hepatocyte function in microfabricated co-cultures. Journal of biomaterials science. Polym Ed 1998;9(11):1137–1160. 32. Masubuchi N, Makino C, Murayama N. Prediction of in vivo potential for metabolic activation of drugs into chemically reactive intermediate: correlation of in vitro and in vivo generation of reactive intermediates and in vitro glutathione conjugate formation in rats and humans. Chem Res Toxicol 2007;20(3):455–464.
J Biochem Molecular Toxicology
DOI 10.1002/jbt
Long-Term Stability of Primary Rat Hepatocytes in Micropatterned Cocultures Okechukwu Ukairo,1 Chitra Kanchagar,1 Amanda Moore,1 Julianne Shi,1 Jeannemarie Gaffney,1 Simon Aoyama,1 Kelly Rose,2 Stacy Krzyzewski,1 Jack McGeehan,1 Melvin E. Andersen,2 Salman R. Khetani,3 and Edward L. LeCluyse2 1 Hepregen 2 The
Corporation, Suite 1500, Medford MA 02155, USA
Hamner Institutes for Health Sciences, Research Triangle Park NC 27709, USA
3 Colorado
State University, Mechanical and Biomedical Engineering, Fort Collins CO 80523-1374, USA; E-mail: [email protected]
Received 8 October 2012; revised 17 November 2012; accepted 19 November 2012
ABSTRACT: Primary hepatocytes display functional and structural instability in standard monoculture systems. We have previously developed a model in which primary hepatocytes are organized in domains of empirically optimized dimensions and surrounded by murine embryonic fibroblasts (HepatoPacTM ). Here, we assess the long-term phenotype of freshly isolated and cryopreserved rat hepatocytes in a 96-well HepatoPac format. The viability, cell polarity (actin microfilaments, bile canaliculi), and functions (albumin, urea, Phase I/II enzymes, transporters) of fresh and cryopreserved rat hepatocytes were retained in HepatoPac at similar levels for at least 4 weeks as opposed to rapidly declining over 5 days in collagen/MatrigelTM sandwich cultures. Pulse or continuous exposure of rat HepatoPac to GW-7647, a selective agonist of PPARα, caused reproducible induction of CYP4A1 and 3hydroxy-3-methylglutaryl-CoA synthase over 4 weeks. In conclusion, rat HepatoPac in a 96-well format can be used for chronic dosing of highly functional hepatocytes and assessment of perturbed hepatocellular C 2013 Wiley Periodicals, Inc. J Biochem Mol pathways. Toxicol 27:204–212, 2013; View this article online at wileyonlinelibrary.com. DOI 10.1002/jbt.21469
KEYWORDS: Cocultures; Micropatterning; Rat Hepatocytes; Sandwich Cultures; Stromal Cells,
INTRODUCTION Rats are used in assessment of long-term effects of chemicals and drugs [1]. However, in vivo studies in rats with a large number of compounds can be C
Correspondence to: Salman R. Khetani. 2013 Wiley Periodicals, Inc.
expensive, require significant compound scaleup for dosing, are too slow for real-time feedback in a screening campaign, and at times are ill-suited for elucidating detailed molecular mechanisms. In the case of the liver, several in vitro models such as liver slices, microsomes, cell lines, and isolated primary hepatocytes have been developed to complement in vivo studies to address the aforementioned issues [2–5]. Liver slices have limited viability and are not amenable to high-throughput screening; cell-free microsomes lack the dynamic gene expression and intact cellular machinery required for dynamic cellular responses; and carcinoma-derived cell lines and immortalized hepatocytes display an abnormal repertoire and levels of liver-specific functions [6]. Therefore, isolated primary hepatocytes are widely considered to be the best choice for in vitro drug/chemical screening applications because they are simple to use and their cytoarchitecture remains intact [4,7]. However, primary hepatocytes are difficult to maintain in culture for more than a few days under conventional culture conditions that rely only on extracellular matrix (ECM) manipulations such as ECM-sandwich models utilizing collagen and/or MatrigelTM [8]. Thus, such conventional cultures are inadequate for evaluating the chronic effects of xenobiotics at physiologically relevant (low) dose levels and there is an urgent need for more predictive liver models. Khetani and Bhatia have previously developed a liver model with precise microscale architecture and optimal stromal interactions (micropatterned cocultures or HepatoPacTM ) that displays phenotypic stability of primary human hepatocytes for several weeks in vitro as assessed by major categories of liver-specific functions and gene expression [9]. In HepatoPac, primary hepatocytes are patterned onto ECM domains 204
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
of empirically optimized microscale dimensions and then surrounded by 3T3-J2 murine embryonic fibroblasts in a 24-well format. Under these culture conditions, phase I and II enzymes as well as transporters in primary human hepatocytes are expressed at high levels for several weeks, which has proven useful for improved detection of drug metabolites over suspension cultures and enzyme fractions [10]. Human HepatoPac outperforms conventional culture models (monolayer on rigid collagen, collagen gel sandwich, Matrigel overlay, Matrigel substratum) with respect to the magnitude and longevity of liver-specific functions. However, the ability of the HepatoPac culture model to stabilize the long-term functionality of primary rat hepatocytes from both fresh and cryopreserved sources in a higher throughput 96-well format remains undetermined. Here, we hypothesized that (a) the HepatoPac culture technique could stabilize the morphology and phenotypic functions of both fresh and cryopreserved rat hepatocytes (Sprague-Dawley) in vitro as opposed to a precipitous decline in sandwich cultures; (b) such stability could be obtained in a 96-well high-throughput format; and (c) stable rat HepatoPac cultures could be subsequently used to assess perturbation of hepatocellular pathways upon chronic administration with xenobiotics. We selected widely used and commercially available Sprague-Dawley fresh and cryopreserved rat hepatocytes to test our hypotheses. Finally, the effect of GW-7647, a potent and selective agonist of human and rodent PPARα [11] (promotes tumor progression), on short- and long-term induction of CYP4A1 [12] and 3-hydroxy-3-methylglutaryl-CoA synthase (HMGCS2) [13] gene expression over a 4-week period was examined to determine the long-term stability and robustness of the nuclear receptor response pathways.
MATERIALS AND METHODS Culture Preparation Freshly isolated and cryopreserved primary rat (Sprague-Dawley) hepatocytes were purchased from Life Technologies (Carlsbad, CA). Cryopreserved hepatocyte vials were thawed at 37◦ C for 90–120 seconds followed by dilution with 50 mL of prewarmed hepatocyte culture medium (HCM) (Hepregen Corporation, Medford, MA). The cell suspension was spun at 50×g for 5 min. The supernatant was discarded, cells were resuspended in HCM, and viability was assessed using trypan blue exclusion (typically 80–95%). Liverderived nonparenchymal cells, as judged by their size (∼10 μm in diameter) and morphology (nonpolygoJ Biochem Molecular Toxicology
DOI 10.1002/jbt
205
nal), were consistently found to be less than 1% in these preparations. To create micropatterned cocultures (HepatoPacTM , Hepregen Corporation) in 96-well plates, a hepatocyte pattern was first produced by seeding hepatocytes on rat-tail collagen (BD Biosciences, Franklin Lakes, NJ) type I-patterned substrates that mediate selective cell adhesion [9]. The cells were washed with medium 4–6 h later to remove unattached cells (leaving ∼5000 attached hepatocytes on 13 collagencoated islands within each well of a 96-well plate) and incubated in HCM. 3T3-J2 murine embryonic fibroblasts [14] were seeded 12–18 h later to create cocultures. Culture medium was replaced every 2 days (∼65 μL per well of a 96-well plate.). For sandwich cultures, approximately 56,000 hepatocytes were diluted in a previously reported serumsupplemented hepatocyte medium [7] and seeded per well in a 96-well format, where each well was coated with rat-tail collagen type I for 2 h and washed twice with 1× phosphate buffered saline (PBS) prior to seeding. Four hours after seeding, cells were washed twice with culture medium and then 0.25 mg/mL Matrigel (BD Biosciences) was diluted in a previously reported serum-free hepatocyte medium [7] and applied to the cultures overnight. Medium was changed daily with serum-free medium for the duration of the study.
Immunofluorescent Staining Cultures were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA) for 15 min at room temperature (RT), washed three times with 1× PBS, and stored at 4◦ C for up to 2 weeks prior to further processing. Fixed cells were permeabilized with 0.1% Triton X-100 (Sigma-Aldrich, St. Louis, MO) for 10 min on a rocker at RT, washed three times with 1× PBS, and then incubated in blocking buffer containing 1% BSA (bovine serum albumin; Sigma-Aldrich) and 5% goat serum (Life Technologies, Carlsbad, CA) in 1× PBS for 30 min at RT. Next, cells were incubated at RT with primary antibody for MRP-2 (Alexis Biochemicals, San Diego, CA) diluted to 1:200 in 1× PBS with 0.2% BSA. After washing 5 times with 1× PBS containing 0.1% Tween-20 (wash buffer), cells were incubated with fluorescein-labeled secondary antibody (Life Technology) diluted at 1:500 in 1× PBS with 0.2% BSA for 1 at RT, and then washed 5 times with wash buffer. For staining actin, rhodamine-labeled phalloidin (Life Technologies) in 1× PBS supplemented with 1% BSA and 5% goat serum was added to cells for 20 min at RT. Cultures were counterstained with 0.5 μg/mL Hoechst (Molecular Probes, Eugene, OR) dye for 5 min at RT, followed by three washes in 1× PBS. Specimens were observed and recorded using a Nikon Eclipse TE200
206
UKAIRO ET AL.
microscope equipped with a SPOT digital camera (SPOT Diagnostic Equipment, Sterling Heights, MI), and PixeLINK uScope Software was used for digital image acquisition.
Hepatocyte Functionality Assays Aliquots of culture medium were collected from culture wells at different time points and stored at –20◦ C. The urea concentration was assayed in culture supernatant with a colorimetric endpoint assay utilizing diacetyl monoxime with acid and heat (Stanbio Labs, Boerne, TX). The albumin content in spent medium was measured using an albumin-specific enzyme-linked immunosorbent assay (MP Biomedicals, Irvine, CA) with horseradish peroxidase detection and 3,3’,5,5’-tetramethylbenzidine (TMB, Fitzgerald Industries, Concord, MA) as the substrate [9]. Overall cytochrome P450 activity was measured by direct incubation of cells for 1 h with 40 μM of 7-benzyloxy-4trifluoromethylcoumarin (BFC; BD Biosciences), which is converted to 7-hydroxytrifluoromethylcoumarin (7-HFC) by CYP450s. Reactions were stopped by collecting the incubation medium from plates, and medium samples were then incubated with βglucuronidase/arylsulfatase (Roche Applied Science, Indianapolis, IN) for 4 h at 37◦ C. Fluorimetric quantification of 7-HFC formed during incubations was performed as previously described [15]. Total cell content was estimated via phase contrast micrographs by manually counting the number of hepatocytes in three representative islands from each condition 1 day after seeding, and then multiplying average cell count per island by total number of islands designed for each well using microfabrication technologies (i.e., an average of 13 collagen-coated islands within each well of a 96well plate). For sandwich cultures, the number of cells seeded per well was used for normalization, which is an overestimate of the actual number of hepatocytes that survive after a few days of culture. Results were expressed as picomoles of metabolite formed per hour per million cells. In addition, the metabolism of model CYP substrates (Sigma-Aldrich) in rat HepatoPac was examined over several weeks. To evaluate CYP3A1/3A2, CYP2D2, CYP2C6, CYP2B1, and CYP4A activities over time, cells were exposed to 25 μM midazolam or 3 μM CYP3A-Glo by Promega (Madison, WI), 10 μM dextromethorphan, 25 μM diclofenac, 500 μM bupropion, and 100 μM lauric acid, respectively, for 30 min at 37◦ C [16–19]. Supernatants were collected, and the concentration of the appropriate metabolite was determined via liquid chromatography/mass spectrometry (Integrated Analytical Services, San Francisco, CA). The 4-hydroxylation of mi-
Volume 27, Number 3, 2013
dazolam, 4-demethylation of dextromethorphan, 4hydroxylation of diclofenac, bupropion hydroxylation, and 12-hydroxylation of lauric acid was assessed via LC-MS/MS (Integrated Analytical Solutions, Berkeley, CA). Phase II enzyme metabolism in rat HepatoPac was assessed by glucuronidation and sulfation of 50 μM 7-hydroxycoumarin (7-HC) to 7-HC-glucuronide and 7-HC-sulfate, respectively. Conversion of CYP3AGlo into luminescent metabolite was quantified as per manufacturer’s (Promega) instructions using a Centro XS luminometer (Berthold Technologies, Germany).
Transporter Assays Hepatic transport studies were conducted using two different substrates: an anionic compound, estradiol 17 β-D-glucuronide (E2 17ßG) and bile acid taurocholate, which are actively transported by sinusoidal proteins OATP (organic anion transport protein) and NTCP (sodium taurocholate transport protein), respectively [20]. Cell cultures were preincubated with calcium-supplemented buffer (Life Technologies) for 10 min at 37◦ C before adding 1 μM cold substrate plus 0.04 μCi/well tritium-labeled substrates (Perkin-Elmer, Boston, MA) for 10 min. Uptake reactions were stopped by placing plates on ice and subsequent washing of wells 3 times with ice-cold calcium supplemented buffer. Plates were sealed and stored at –20◦ C until analyzed by liquid scintigraphy. Cells were treated with 1× PBS with 1% Triton X-100, shaken for 20 min, harvested and transferred to a scintillation vial containing liquid scintillation fluid. Radioactivity in each sample was counted using an LS6500 multipurpose scintillation counter (Beckman Coulter, Indianapolis, IN). The uptake rate of individual substrates is reported as disintegrations per minute (DPM) per million cells.
Gene Expression Profiling Hepatocyte gene expression was analyzed using R -based quantitative real-time PCR (qRT-PCR, TaqMan Life Technologies). Cell lysis and RNA purification R mini kit (Qiagen, were performed using the RNeasy Valencia, CA). Transcription to cDNA was accomplished using the high capacity cDNA reverse transcription kit according to manufacturer’s instructions (Life Technologies), and 2 μg was used for each qPCR reaction. FAM-labeled primer/probe sets for genes of interest and Path-IDTM qPCR Master Mix were obtained from Life Technologies. RT-PCR was perR realplex instrument (Epformed on a Mastercycler pendorf, Hamburg, Germany), and data were analyzed using the realplex software. Gene expression was normalized to the housekeeping genes α-actin and β-tubulin. J Biochem Molecular Toxicology
DOI 10.1002/jbt
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
207
FIGURE 1. Morphology and polarity of primary rat hepatocytes in HepatoPac over the course of several weeks. Hepatocytes maintained prototypical polygonal shape, distinct nuclei/nucleoli, and bile canaliculi (arrows) for at least 4 weeks (days 5, 15, and 27 shown) in HepatoPac as assessed via phase contrast microscopy (A). Fluorescently labeled actin microfilaments in hepatocytes were primarily associated with the peripheral margins of the hepatocytes, including the edges of bile canaliculi (B). Fluorescently labeled multidrug resistance protein 2 (MRP2), an efflux transporter, was localized on the bile canaliculi between hepatocytes (C). Representative images shown are for cryopreserved hepatocytes, but similar trends were seen with freshly isolated hepatocytes. Scale bar is 100 μm. Inset shows a cluster of a few hepatocytes in each panel.
Dosing with GW-7647 Rat HepatoPac cultures were dosed with the PPAR agonist GW-7647 (Sigma-Aldrich) at 0.01, 0.1, and 1 μM (DMSO concentration at 0.1% vol/vol in all conditions), and its effect on CYP4A1 and 3-hydroxy3-methylglutaryl-CoA synthase (HMGCS2) transcript expression in rat HepatoPac as a function of dose and time of exposure was evaluated. Cell lysates were collected for gene expression analyses following 6 and 72 h incubation of cultures with compound. Compound incubations were carried out once per week for 4 weeks in separate rat HepatoPac cultures (96-well format). In addition, rat HepatoPac cultures were incubated with 1 μM GW-7647 continuously for 4 weeks and redosed every 2 days with each medium change. Cell lysates were collected at 1, 2, 3, and 4 weeks for CYP4A1 and HMGCS2 expression analyses as described above.
RESULTS We verified previous findings in the literature that primary rat hepatocytes showed a gradual decline in prototypic morphology in sandwich culJ Biochem Molecular Toxicology
DOI 10.1002/jbt
tures (Supplemental Figure 1 in the Supporting Information) over the course of 10 days [8]. In contrast, fresh and cryopreserved hepatocytes from the same donors as that used for sandwich cultures, maintained in vivo-like polygonal shape, distinct nuclei/nucleoli and visible bile canaliculi in HepatoPac for at least 4 weeks in vitro (Figure 1A). Actin microfilaments in hepatocytes in HepatoPac were primarily associated with the peripheral margins of the hepatocytes, including the edges of bile canaliculi (Figure 1B), as reported previously in hepatocytes in vivo [21]. In addition, multidrug resistance protein 2 (MRP2) protein, a transporter [22], was localized on the bile canaliculi between hepatocytes (Figure 1C), which is indicative of polarized hepatocytes. Next, major liver-specific functions were assessed in the two culture models, specifically albumin secretion (a marker of liver protein synthesis), urea secretion (a marker of nitrogen metabolism), and overall CYP450 activity (nonspecific) using a fluorometric substrate. HepatoPac cultures took ∼5 days to stabilize functionally and retained high levels of functions for at least 4 weeks in vitro (Figures 2A–2C). On the other hand, sandwich cultures showed a gradual decline in CYP450 activity and after 5 days of culture, activity
208
UKAIRO ET AL.
Volume 27, Number 3, 2013
FIGURE 2. Major liver-specific functions in HepatoPac and sandwich cultures created using freshly isolated and cryopreserved primary rat hepatocytes. Albumin secretion (A), urea synthesis (B), and overall CYP450 activity (C) in HepatoPac is shown over the course of 4 weeks. A decline in overall CYP450 activity is shown for sandwich cultures (rigid collagen with Matrigel overlay) over the course of 9 days (D); however, similar trends were observed with urea secretion. CYP450 activity was assessed via the hydroxylation of 7-benzyloxy-4-trifluoromethylcoumarin into 7- hydroxy-4-trifluoromethylcoumarin (7-HFC).
was ∼5% of day 1 sandwich and overall HepatoPac levels (Figure 2D). Urea secretion in sandwich cultures showed near identical kinetics of decline as CYP450 activity; however, albumin secretion remained relatively stable in sandwich cultures, but 8–10 fold lower than HepatoPac when normalized to cell numbers (data not shown). Furthermore, fresh and cryopreserved hepatocytes showed relatively similar levels and kinetics of functions in both culture models. Because cryopreserved hepatocytes are a more convenient and on-demand source for creation of cultures for in vitro investigations [23], they were further characterized in the HepatoPac culture model for activities of multiple CYP450 isoforms as well as Phase II-mediated conjugation reactions (Figure 3). CYP2B, 3A, and 4A isoforms took 8–10 days to reach maximal levels and remained relatively stable for ∼4 weeks. CYP2C and 2D, on the other hand, reached maximal levels in ∼6 days and showed a 50% decline in activities between 18 and 26 days of culture. CYP3AGlo (luminescent-based assay from Promega, Madison, WI) showed similar kinetics as formation of 4-hydroxy midazolam, suggesting that the rapid luminescent assay is suitable for probing CYP3A activity. Phase II glucuronidation and sulfation of 7-hydroxycourmarin took ∼6 days to reach maximal levels and remained relatively stable for ∼4 weeks in the HepatoPac system.
Next, the time-dependent accumulation of radioactive estradiol 17 β-D-glucuronide (E2 17ßG, a substrate for organic anion transport protein or OATP) and taurocholate (a substrate for sodium taurocholate transport protein or NTCP) was assessed in the culture models (Figures 4A–4B). Fresh and cryopreserved hepatocytes in HepatoPac showed remarkably similar uptake activities for ∼4 weeks. The data indicate that maximal levels of uptake activities of these substrates was achieved by day 15-21 of HepatoPac culture and remained stable until at least day 27; however, by day 9 of culture, ∼70% to 75% of the maximal uptake activities were observed. Sandwich cultures, on the other hand, showed differences in uptake activities across fresh and cryopreserved hepatocytes after 2 days of culture (activities in cryopreserved hepatocytes were ∼40% of activities in fresh ones). Furthermore, the uptake activities of the fresh hepatocytes in sandwich cultures after 2 days of culture were 70% and 115% of HepatoPac levels for E2 17ßG and taurocholate, respectively; however, by the 5th day of culture, uptake activities in sandwich cultures had declined to 18% and 8% of HepatoPac levels for E2 17ßG and taurocholate, respectively (Figure 4C). We next evaluated the functionality of the bile canaliculi network in HepatoPac via transporter-mediated secretion of the MRP2 (ABCC2 or ATP-binding cassette, sub-family C (CFTR/MRP), member 2) substrate, 5-(and-6)- carboxy-2’,7’-dichlorofluorescein diacetate J Biochem Molecular Toxicology
DOI 10.1002/jbt
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
209
FIGURE 3. Activities of major CYP450 and phase II conjugation enzymes in HepatoPac created using cryopreserved rat hepatocytes. Enzyme activities over the course of 4 weeks were assessed via luciferin release from Promega’s CYP3A-Glo assay (A), midazolam hydroxylation by CYP3A (A), diclofenac hydroxylation by CYP2C (B), lauric acid hydroxylation by CYP4A (B), dextromethorphan demethylation into dextrorphan by CYP2D (C), bupropion hydroxylation by CYP2B (C), and phase II-mediated glucuronidation and sulfation of 7-hydroxycoumarin or 7-HC (D).
FIGURE 4. Transporter activities in HepatoPac and sandwich cultures created using freshly isolated and cryopreserved rat hepatocytes. Hepatocytes in HepatoPac take up prototypical transporters substrates, estradiol 17 β-D-glucuronide (E2 17ßG) (A) and bile acid taurocholate (B) over the course of 4 weeks. A decline in uptake of taurocholate is shown for sandwich cultures (rigid collagen with Matrigel overlay) over the course of 9 days (C); however, near identical trends were observed with E2 17ßG. Functional bile canaliculi were present between hepatocytes in HepatoPac (D) as assessed via transporter-mediated secretion of the fluorescent dye, 5-(and-6)- carboxy-2’,7’-dichlorofluorescein diacetate succinimidyl ester, into bile canaliculi. Image from cultures 2 weeks old is shown; however, similar trends were seen for at least 4 weeks.
succinimidyl ester, into bile canaliculi between hepatocytes. After about 20 min of pretreatment of cells with the dye, uniform, intense fluorescence was observed in nearly all the canalicular networks throughout the hepatocyte islands. No fluorescent staining was observed between hepatocytes and the fibroblasts (Figure 4D). J Biochem Molecular Toxicology
DOI 10.1002/jbt
Finally, we evaluated the utility of the HepatoPac model using cryopreserved rat hepatocytes for chronic/repeat dosing with enzyme inducers and assessment of known upregulated pathways. Incubation of rat HepatoPac for 6 or 72 h with varying concentrations of GW-7647, a potent PPARα agonist, caused a
210
UKAIRO ET AL.
Volume 27, Number 3, 2013
FIGURE 5. GW-7647 dose- and time-dependent induction of CYP4A1 and HMGCS2 transcripts in HepatoPac created using cryopreserved rat hepatocytes. Dose- and time-dependent upregulation of CYP4A1 (A) and HMGCS2 (B) mRNA was seen after exposure of HepatoPac to GW-7647. The effect of culture age (weeks 1–4) on GW-7647-mediated upregulation of CYP4A1 and HMGCS2 transcripts is shown after exposure of HepatoPac cultures for 6 h at a GW-7647 dose of 0.1 μM (C). Continuous exposure (fresh compound every 2 days) with GW-7647 at 1 μM caused upregulation of CYP4A1 and HMGCS2 transcripts over the course of 4 weeks (D). VC is vehicle controls.
dose- and time-dependent induction of PPARα target genes CYP4A1 and HMGCS2 (Figures 5A and 5B). After 6 h of exposure, both CYP4A1 and HMGCS2 showed ∼9 fold induction over vehicle controls at 1 μM of GW7647, while after 72 h of incubation the maximal fold induction was 71 and 29 for CYP4A1 and HMGCS2, respectively. Induction of these genes after 6 h exposure to 1 μM GW-7647 was seen at different culture ages; however, 4 week old cultures showed fold induction levels that were ∼50% of levels seen in the first 3 weeks of culturing (Figure 5C). Furthermore, continuous exposure of HepatoPac to 1 μM GW-7647 (with medium change every 2 days) enabled assessment of induction kinetics of both transcripts over the course of 4 weeks (Figure 5D). CYP4A1 remained induced 27 fold after 1 and 2 weeks of continuous exposure and that dropped by ∼50% to ∼12.5 fold and 14 fold after 3 and 4 weeks of exposure, respectively. HMGCS2, on the other hand, showed a 107 fold induction after 1 week of exposure, 29 fold after 2 weeks, 15 fold after 3 weeks, and an increase to 42 fold after 4 weeks.
DISCUSSION Predicting the effects of compounds in humans from animal or in vitro data continues to be a crit-
ical challenge facing the chemical and pharmaceutical industries [24, 25]. Progress in implementing costeffective screening strategies for an ever-increasing array of chemicals depends on physiologically relevant in vitro models of animal and human livers [26, 27]. However, current model systems such as primary hepatocyte monocultures and sandwich cultures are limited in their ability to maintain hepatocyte functionality over an extended period of time [2, 7]. Improved, longlasting hepatocyte cultures could revolutionize drug screening. In undertaking this study, we looked to demonstrate the potential of the HepatoPac micropatterned coculture system to mimic the native liver architecture and retain liver-specific functions over an extended period of time. We report here that rat hepatocytes cultured in HepatoPac demonstrate functional behavior at higher levels and for significantly longer periods of time than those maintained under the conventional two-dimensional (2D) static sandwich culture conditions. Specifically, in both primary and cryopreserved rat hepatocytes cultured in HepatoPac, cell viability, histotypic cell architecture, phase 1 and 2 metabolic activity, gene expression, and albumin/urea production were maintained up to 4 weeks and found to be superior to standard 2D sandwich culture.
J Biochem Molecular Toxicology
DOI 10.1002/jbt
Volume 27, Number 3, 2013
RAT HEPATIC MICROPATTERNED COCULTURES
The long-term stability of the classical peroxisomeproliferator activated receptor pathway in rat HepatoPac was determined by evaluating mRNA levels for known PPARα target genes, CYP4A1 [12] and HMGCS2 [13], after treatment with the selective agonist, GW-7647. The PPARα pathway remained responsive to agonist in HepatoPac for at least 4 weeks and exhibited dose- and time-dependent activation. We did, however, observe that the magnitude of target gene expression dropped by ∼50% when 4 week old cultures were subjected to short-term (hours) exposure to GW7646, suggesting potential downregulation of PPARα expression and/or its cofactors. HepatoPac’s design is based on a standard (static fluid) 96-well plate format, and the optimized micropatterned cell configuration and cell ratios support most of the major functions of primary hepatocytes, including key phase 1 and 2 biotransformation pathways. In the HepatoPac system, hepatocytes are thought to be responding to enhancements in cell–cell interactions created by the paracrine and heterotrophic effects of the hepatocyte-stromal coculture environment [28, 29]. We and others have shown that 3T3 fibroblasts and other nonliver derived stroma can induce functions in both animal and human hepatocytes, suggesting possible conservation of mechanisms [9, 29, 30]. SpragueDawley rat hepatocytes were chosen here given commercial availability of fresh and cryopreserved lots and their historical use as a toxicology species; however, other studies with Lewis [31] and Wistar Han rats (unpublished) suggest compatibility of HepatoPac with multiple strains. The results from these studies indicate that HepatoPac represents a robust model for the examination of prolonged effects of xenobiotics on the perturbation of hepatic pathways. The cumulative effects of parent compound and major metabolites formed during the course of continuous incubation in the HepatoPac system could be more easily determined, providing more relevant information about the pharmaco and toxicokinetic properties of compounds. The magnitude and dynamics of perturbations on nuclear receptor pathways, such as PPARα, upon chronic exposure to compounds, could be better compared to in vivo chronic dosing studies and utilized for more accurate in vitro to in vivo correlations. Finally, responsiveness of rat HepatoPac to stimulation with other nuclear receptor agonists as well as bioactivated compounds such as acetaminophen [32] is under investigation.
SUPPORTING INFORMATION Supplemental Figure 1 is available from the corresponding author on request. This figure shows the J Biochem Molecular Toxicology
DOI 10.1002/jbt
211
morphology of collagen/Matrigel sandwich cultures (days 2, 5, and 10 postseeding) created using freshly isolated Sprague-Dawley primary rat hepatocytes in a 96-well format. See the Methods section for details on the preparation of sandwich cultures.
ACKNOWLEDGMENTS Funding for this project was provided by the LongRange Research Initiative of the American Chemistry Council (ACC).
REFERENCES 1. Klaassen CD, Casarett LJ, Doull J. Casarett and Doull’s toxicology: The basic science of poisons. New York: McGraw-Hill; 2013. 2. Gebhardt R, Hengstler JG, Muller D, Glockner R, Buenning P, Laube B, Schmelzer E, Ullrich M, Utesch D, Hewitt N, Ringel M, Hilz BR, Bader A, Langsch A, Koose T, Burger HJ, Maas J, Oesch F. New hepatocyte in vitro systems for drug metabolism: metabolic capacity and recommendations for application in basic research and drug development, standard operation procedures. Drug Metab Rev 2003;35(2–3):145–213. 3. Sivaraman A, Leach JK, Townsend S, Iida T, Hogan BJ, Stolz DB, Fry R, Samson LD, Tannenbaum SR, Griffith LG. A microscale in vitro physiological model of the liver: predictive screens for drug metabolism and enzyme induction. Curr Drug Metabol 2005;6(6):569–591. 4. Guillouzo A. Liver cell models in vitro toxicology. Environ Health Pers 1998;106(Suppl 2):511–532. ´ MJG, Houston JB, Hallifax D, Brown 5. Hewitt NJ, Lechon HS, Maurel P, Kenna JG, Gustavsson L, Lohmann C, Skonberg C and others. Primary hepatocytes: current understanding of the regulation of metabolic enzymes and transporter proteins, and pharmaceutical practice for the use of hepatocytes in metabolism, enzyme induction, transporter, clearance, and hepatotoxicity studies. Drug Metab Rev 2007;39(1):159–234. 6. Wilkening S, Stahl F, Bader A. Comparison of primary human hepatocytes and hepatoma cell line Hepg2 with regard to their biotransformation properties. Drug Metab Disposition: Biol Fate Chem 2003;31(8):1035–1042. 7. Lecluyse EL. Human hepatocyte culture systems for the in vitro evaluation of cytochrome P450 expression and regulation. Eur J Pharm Sci 2001;13(4):343–368. 8. LeCluyse E, Bullock P, Madan A, Carroll K, Parkinson A. Influence of extracellular matrix overlay and medium formulation on the induction of cytochrome P-450 2B enzymes in primary cultures of rat hepatocytes. Drug Metab Dispos 1999;27(8):909–915. 9. Khetani SR, Bhatia SN. Microscale culture of human liver cells for drug development. Nat Biotechnol 2008;26(1):120–126. 10. Wang WW, Khetani SR, Krzyzewski S, Duignan DB, Obach RS. Assessment of a Micropatterned Hepatocyte Coculture System to Generate Major Human Excretory and Circulating Drug Metabolites. Drug Metab Dispos: Biol Fate Chem 2010;38(10):1900–1905.
212
UKAIRO ET AL.
11. Lee JY, Hashizaki H, Goto T, Sakamoto T, Takahashi N, Kawada T. Activation of peroxisome proliferatoractivated receptor-alpha enhances fatty acid oxidation in human adipocytes. Biochem Biophys Res Commun 2011;407(4):818–822. 12. Elcombe CR, Elcombe BM, Foster JR, Chang SC, Ehresman DJ, Butenhoff JL. Hepatocellular hypertrophy and cell proliferation in Sprague-Dawley rats from dietary exposure to potassium perfluorooctanesulfonate results from increased expression of xenosensor nuclear receptors PPARalpha and CAR/PXR. Toxicology 2012;293(1– 3):16–29. 13. Rakhshandehroo M, Hooiveld G, Muller M, Kersten S. Comparative analysis of gene regulation by the transcription factor PPARalpha between mouse and human. PLoS One 2009;4(8):e6796. 14. Rheinwald JG, Green H. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing colonies from single cells. Cell 1975;6(3):331– 343. ´ ´ MJ. 15. Donato MT, Jim´enez N, Castell JV, Gomez-Lech on Fluorescence-based assays for screening nine cytochrome P450 (P450) activities in intact cells expressing individual human P450 enzymes. Drug Metabol Dispos: Biol Fate Chem 2004;32(7):699–706. 16. Kobayashi K, Urashima K, Shimada N, Chiba K. Substrate specificity for rat cytochrome P450 (CYP) isoforms: screening with cDNA-expressed systems of the rat. Biochem Pharmacol 2002;63(5):889–896. 17. Madan A, DeHaan R, Mudra D, Carroll K, LeCluyse E, Parkinson A. Effect of cryopreservation on cytochrome P450 enzyme induction in cultured rat hepatocytes. Drug Metab Dispos 1999;27(3):327–335. 18. Burke MD, Thompson S, Weaver RJ, Wolf CR, Mayer RT. Cytochrome P450 specificities of alkoxyresorufin Odealkylation in human and rat liver. Biochem Pharmacol 1994;48(5):923–936. 19. Pekthong D, Desbans C, Martin H, Richert L. Bupropion hydroxylation as a selective marker of rat CYP2B1 catalytic activity. Drug Metab Dispos 2012;40(1):32–38. 20. Consortium IT, Giacomini KM, Huang S-M, Tweedie DJ, Benet LZ, Brouwer KLR, Chu X, Dahlin A, Evers R, Fischer V and others. Membrane transporters in drug development. Nat Rev Drug Disc 2010;9(3):215– 236.
Volume 27, Number 3, 2013
21. Hooser SB, Beasley VR, Waite LL, Kuhlenschmidt MS, Carmichael WW, Haschek WM. Actin filament alterations in rat hepatocytes induced in vivo and in vitro by microcystin-LR, a hepatotoxin from the blue-green alga, Microcystis aeruginosa. Vet Pathol 1991;28(4):259–266. 22. Li N, Bi Y-A, Duignan DB, Lai Y. Quantitative expression profile of hepatobiliary transporters in sandwich cultured rat and human hepatocytes. Molec Pharm 2009;6(4):1180– 1189. 23. Bi Y-A, Kazolias D, Duignan DB. Use of cryopreserved human hepatocytes in sandwich culture to measure hepatobiliary transport. Drug Metabol Dispos: Biol Fate Chem 2006;34(9):1658–1665. 24. Chen M, Vijay V, Shi Q, Liu Z, Fang H, Tong W. FDAapproved drug labeling for the study of drug-induced liver injury. Drug Discov Today 2011;16(15–16):697–703. 25. Kaplowitz N. Idiosyncratic drug hepatotoxicity. Nature Rev Drug Discov 2005;4(6):489–499. ´ ´ MJ, Lahoz A, Gombau L, Castell JV, Do26. Gomez-Lech on nato MT. In vitro evaluation of potential hepatotoxicity induced by drugs. Curr Pharma Des 2010;16(17):1963– 1977. 27. Rawlins MD. Cutting the cost of drug development? Nature Rev Drug Discov 2004;3(4):360–364. 28. Khetani SR, Chen AA, Ranscht B, Bhatia SN. Tcadherin modulates hepatocyte functions in vitro. FASEB J 2008;22(11):3768–3775. 29. Khetani SR, Szulgit G, Del Rio JA, Barlow C, Bhatia SN. Exploring interactions between rat hepatocytes and nonparenchymal cells using gene expression profiling. Hepatology (Baltimore, Md) 2004;40(3):545–554. 30. Bhatia SN, Balis UJ, Yarmush ML, Toner M. Effect of cellcell interactions in preservation of cellular phenotype: cocultivation of hepatocytes and nonparenchymal cells. FASEB J 1999;13(14):1883–1900. 31. Bhatia SN, Balis UJ, Yarmush ML, Toner M. Probing heterotypic cell interactions: hepatocyte function in microfabricated co-cultures. Journal of biomaterials science. Polym Ed 1998;9(11):1137–1160. 32. Masubuchi N, Makino C, Murayama N. Prediction of in vivo potential for metabolic activation of drugs into chemically reactive intermediate: correlation of in vitro and in vivo generation of reactive intermediates and in vitro glutathione conjugate formation in rats and humans. Chem Res Toxicol 2007;20(3):455–464.
J Biochem Molecular Toxicology
DOI 10.1002/jbt
