Automated Triplexed Hepatocyte-Based Viability and CYP1A and -3A Induction Assays

Abstract

Cytochrome P450 (CYP) enzymes are key players in drug metabolism. Therefore, it is essential to understand how these enzymes can be affected by xenobiotics with regards to induction and toxicity to avoid potential drug–drug interactions. Typically, information has been gathered by combining data from multiple experiments, which is time-consuming and labor intensive, and interassay variability may lead to misinterpretation. Monitoring CYP induction and cytotoxicity by xenobiotics using an automated, multiplexed format can decrease workload and increase data confidence. Here the authors demonstrate the ability to monitor CYP1A and CYP3A4 induction, combined with a cytotoxicity measurement, from a single microplate well using cryopreserved human hepatocytes. The assay procedure was automated in a 384-well format, including cell manipulations, compound titration and transfer, and reagent dispensing, using simple robotic instrumentation. EC₅₀ and E_max values were derived for multiple known CYP1A and -3A4 inducers. Induction and toxicological responses in the triplex system were validated based on literature values from conventional single-parameter assays. Validation and pharmacology data confirm that multiplexed cell-based CYP assays can simplify workload, save time and effort, and generate biologically relevant data.

Keywords

cytochrome P450 hepatocytes induction viability multiplexing automation luminescence liquid handling detection

Introduction

Adverse drug–drug interactions (DDIs) are of serious concern to the pharmaceutical industry and associated regulatory agencies. One DDI parameter is induction, the upregulation of enzyme levels that increase metabolism, reducing a concomitant drug’s therapeutic efficacy. Many CYP enzymes have proven to be inducible, including CYP1A1, -1A2, -2B6, -2C9, -2C19, and -3A4.¹ This is accomplished primarily through ligand binding and subsequent activation of various receptors. Aryl hydrocarbon receptor (AHR),² constitutively active receptor (CAR), and pregnane X receptor (PXR)³ are responsible for transcription of many enzymes and transporters associated with drug metabolism and excretion. However, all CYP enzymes and transporters do not need to be measured to determine a drug’s induction potential. To quantify this risk, the Food and Drug Administration (FDA) draft guidance states that CYP3A appears to be sensitive to all known coinducers of CYP2C and CYP2B. Therefore, to evaluate whether an investigational drug induces the major drug-metabolizing enzymes CYP1A, -2B6, -2C8, -2C9, -2C19, or -3A, the initial in vitro induction evaluation may include only CYP1A2 and CYP3A.¹ Although there are cell lines and reporter gene assays that can be used, human hepatocytes are the gold standard to screen in vitro for CYP induction.⁴ Cryopreserved human hepatocytes, in particular, are a reliable model system to evaluate the potential of a new chemical entity (NCE) to induce these enzymes.⁴ This can be accomplished by measuring enzymatic activity with a specific substrate and comparing the induced activity to the basal activity of the enzyme. Typically, this information has been gathered from separate wells of cultured hepatocytes and by combining the data from multiple experiments. This process can take several days or weeks to perform, which increases labor time and requires significant reagent expense given the high cost of primary hepatocytes. It may also lead to erroneous conclusions when combining the data due to the variability within each experiment. A multiplexed format provides a method to obtain several end points from a single microplate well, which may attenuate these concerns. However, multiple manual manipulations in a single assay may prove to be a demanding process and may still lead to errors and inaccurate data. Therefore, by automating the assay using appropriate liquid handling, the multiplexed assay performance becomes an efficient process, which can deliver high reproducibility among replicates and increased confidence in the generated data.

Here we demonstrate the ability to monitor CYP1A and CYP3A4 induction in human cryoplateable hepatocytes, combined with cytotoxicity measurements, from a single microplate well triplex format. The assay procedure was automated in a 384-well format, including cell manipulations, compound titration and transfer, and reagent dispensing, using robotic instrumentation. Four known CYP1A inducers and four known CYP3A4 inducers were used as control compounds to validate the triplex assay. Results show how measuring three critical responses from a single sample streamlines workflow, derives maximum value from human hepatocytes, and provides a means for reaching more biologically relevant interpretations.

Materials and Methods

Materials

P450-Glo CYP3A4 Assay with Luciferin-IPA and CellTiter-Glo Luminescent Cell Viability Assay were obtained from Promega Corporation (Madison, WI). Each P450-Glo CYP3A4 assay kit contains Luciferin-IPA substrate (3 mM), lyophilized Luciferin Detection Reagent, and Reconstitution Buffer with Esterase. Each CellTiter-Glo assay kit contains CellTiter-Glo buffer and lyophilized CellTiter-Glo substrate.

Female human cryoplateable hepatocytes (donors 1 and 2), male human cryoplateable hepatocytes (donor 3), InVitro GRO CP medium, InVitro GRO HI medium, InVitro GRO KHB, and Torpedo Antibiotic Mix were obtained from Celsis In Vitro Technologies (Baltimore, MD).

Resorufin ethyl ether, rifampicin, 3-methylcholanthrene, β-napthoflavone, phenylbutazone, salicylamide, and DMSO were purchased from Sigma-Aldrich (St. Louis, MO). Omeprazole, lansoprazole, SR12813, and dexamethasone were purchased from Tocris Bioscience (Ellisville, MO).

Liquid handling

An EL406 Combination Washer Dispenser from BioTek Instruments (Winooski, VT) was used to dispense cells and reagents, as well as perform media exchanges and media removal in the cell plates. The instrument contains a 96-tube wash manifold, a peristaltic pump, and two syringe pumps for dispensing. The peristaltic pump was used to dispense cells, media, and diluted Luciferin-IPA substrate to the 384-well cell plates; the wash manifold was used to aspirate media from the cell plates; and the two syringe pumps were used to dispense assay reagents to the same plates.

A Precision Microplate Sample Processor from BioTek Instruments was used to perform compound titrations and dispense compounds to the cell plate wells. The instrument was also used to transfer media containing Luciferin-IPA substrate from cell plates to white luminescent plates and dispense reconstituted Luciferin Detection Reagent to the plates. The instrument contains a single-channel pipetting head, an eight-channel bulk reagent dispenser, and an eight-channel pipetting head.

Luminescent and fluorescent signal detection

All assay plates were read using a Synergy H4 Hybrid Multi-Mode Microplate Reader from BioTek Instruments. The luminescence mode was used to measure the P450-Glo and CellTiter-Glo assay signals, using a 1.0-s integration time. Automatic sensitivity adjustment was used to detect the plate well containing the highest luminescent signal. The gain setting is automatically adjusted to the well’s signal. Fluorescent mode was used to measure the resorufin signal from the ethoxyresorufin-O-deethylase (EROD) assay, using a 530/20-nm excitation filter, a 590/35-nm emission filter, and a 550-nm cutoff dichroic mirror.

Methods

Cell preparation

Cryopreserved hepatocytes were thawed and resuspended in InVitro Gro CP medium as described in the Celsis IVT instructions. The cells were counted using Trypan blue exclusion and diluted to 4 × 10⁵ viable cells/mL for plating. A volume of 25 µL/well was dispensed into the 384-well collagen I cell culture microplates for a seeding density of approximately 180 000 viable cells per cm². All cell plates were then incubated in a 37 °C/5% CO₂ tissue culture incubator for 48 h before compound addition. After 24 h, medium was removed from the cell plates, and an equal volume of fresh InVitro Gro CP medium was added to each plate.

CYP assays and substrate preparation

The CYP3A4 assay uses the luminogenic CYP3A4 substrate Luciferin-IPA, which is very selective for CYP3A4 over other CYP enzymes.^5,6 This pro-luciferin is converted by CYP3A4 to an active form of luciferin that is detected with a luciferin detection reagent that contains a luciferase enzyme that generates a signal proportional to the amount of CYP reaction product formed. The CYP1A assay uses the resorufin ethyl ether pro-fluorescent CYP1A substrate. This substrate is converted by CYP1A1 and CYP1A2 to a fluorescent resorufin product that produces a signal proportional to CYP1A activity. The assays are referred to here as CYP3A4 and CYP1A in deference to the substrate selectivities. The Luciferin-IPA and resorufin ethyl ether were incubated with the cells for 60 and 30 min, respectively ( Table 1 ). These incubation times have previously been shown to be within the linear range for the CYP3A4 and CYP1A reactions (data not shown). The 3-mM Luciferin-IPA stock solution was diluted 1:1000 in InVitro Gro HI medium to create a final working concentration of 3 µM. A 100% DMSO stock of salicylamide (an inhibitor of secondary resorufin metabolism) and resorufin ethyl ether was each created at 3 M and 1 mM, respectively. The salicylamide stock solution was then diluted in InVitro Gro KHB buffer to create a 3-mM working concentration. The resorufin ethyl ether stock solution was diluted in InVitro Gro KHB buffer containing 3 mM salicylamide to create a 2-µM working concentration.

Table 1.

Automated Triplexed 384-Well CYP Induction Protocol

Step	Parameter	Value	Description
1	Cells	25 µL	Cell dispensing
2	Media	25 µL	Media transfer
3	Compounds	25 µL	CYP induction
4	Luciferin-IPA	50 µL	Luciferin-IPA substrate addition
5	Incubation	60 min	37 °C/5% CO₂
6	Media transfer	30 µL	CYP3A4 reaction preparation
7a^a	Luciferin detection reagent	30 µL	1:1 volume ratio of luciferin detection reagent and Luciferin-IPA in media
8a^a	Incubation	20 min	Room temperature
9a^a	Detection	Luminescent mode	Multimode reader, 1.0-s integration time, auto sensitivity
7b^b	Media replacement	25 µL	Media with Luciferin-IPA substrate replaced with 3 mM salicylamide in buffer
8b^b	Incubation	10 min	37 °C/5% CO₂
9b^b	Media replacement	25 µL	3 mM salicylamide in buffer replaced with 2 µM 7-ER in buffer with 3 mM salicylamide
10b^b	Incubation	30 min	37 °C/5% CO₂
11b^b	Detection	Fluorescent	Multimode reader, 530/20 excitation filter; 590/35 emission filter; 550 cutoff mirror
12b^b	Luminescent viability reagent	25 µL	CellTiter-Glo addition
13b^b	Incubation	10 min	Room temperature
14b^b	Detection	Luminescent mode	Multimode reader, 1.0-s integration time, auto sensitivity
Step	Notes
1	Hepatocytes dispensed to cell plates (10K cells/25 µL). Plates then placed in 37 °C/5% CO₂ incubator for 24 h.
2	Spent media removed and replaced by fresh media. Plates then placed in 37 °C/5% CO₂ incubator for an additional 24 h.
3	Compounds titrated and transferred into cell plates, following media removal. Plates then placed in 37 °C/5% CO₂ incubator for 24 h. Process repeated to ensure fresh compounds were incubated with the cells for a total of 48 h.
4	Cell-permeant Luciferin-IPA substrate added to the cells following media removal.
5	Plates incubated in 37 °C/5% CO₂ incubator for 60 min.
6	Portion of media containing oxidized and nonoxidized Luciferin-IPA substrate transferred to white luminescent plates to perform P450-Glo assay.
7a	Luciferase reaction initiated.
8a	Plates shaken at 1000 RPM; 30 s, covered and incubated at room temperature for 20 min.
9a	Luminescent signal quantified using luminescent mode of multimode reader. Signal produced proportional to CYP3A4 activity in each well.
7b	Salicylamide added to inhibit phase II conjugating enzymes and block metabolism of resorufin.
8b	Plates placed in 37 °C/5% CO₂ incubator for 10 min.
9b	Cell-permeant resorufin ethyl ether substrate added to the cells following media removal to initiate CYP1A2 reaction.
10b	Plates placed in 37 °C/5% CO₂ incubator for 30 min.
11b	Fluorescent signal quantified using filter-based detection system of reader. Signal produced proportional to CYP1A2 activity in each well.
12b	CellTiter-Glo reagent added to initiate cell viability assay.
13b	Plates shaken at 1000 RPM; 30 s, covered and incubated at room temperature for 10 min.
14b	Luminescent signal quantified using luminescent mode of multimode reader. Signal produced proportional to number of viable cells within each well.

Steps 7 to 9a performed with white luminescent plates.

Steps 7 to 14b performed with black cell plates.

Luminescent detection reagent preparation

Reconstituted Luciferin Detection Reagent (LDR) was prepared by adding Reconstitution Buffer with Esterase to the lyophilized LDR. Reconstituted CellTiter-Glo reagent (CTG) was prepared by adding CellTiter-Glo buffer to the lyophilized CellTiter-Glo substrate.

Compounds

All compounds were dissolved in 100% DMSO. Omeprazole (Omep), β-napthoflavone (β-NF), phenylbutazone (Pb), and dexamethasone (Dex) were prepared at 100 mM; lansoprazole (Lpz) was prepared at 60 mM; 3-methylcholanthrene (3-MC) and rifampicin (Rif) were prepared at 10 mM; and SR12813 (SR) was prepared at 1 mM. Omep and Rif were diluted to 10 µM in 0.5% DMSO for the Z′ factor validation. For the EC₅₀ determination, all test compounds were serially diluted 1:2 in 100% DMSO, starting at the concentration prepared. Each titration series was then diluted 1:200 in media to create the final concentrations used to test for CYP induction. Final DMSO concentrations were 0.5%.

Data analysis

P450-Glo CYP3A4 and resorufin CYP1A signals were measured in relative luminescent units (RLU) and relative fluorescent units (RFU), respectively. Signals from vehicle-alone treated cells are referred to as basal. The average of the no–cell fluorescent or luminescent background was subtracted from the individual signals to give net values. Net values were then divided by the CTG viability signal, as measured by RLU, from the same well to provide values normalized to cell number. Fold induction (treated wells/basal wells) was calculated from net, normalized values. The CTG basal values were set as 100% cell viability, and treated cell viability was calculated relative to this value.

The coefficient of variation (CV) for the CTG signal in the hepatocyte dispensing validation was computed by dividing the standard deviation (σ) among the replicates by the mean (µ) of the same replicate group. This fraction is expressed as a percentage to create the following formula: (σ_(Replicates)/µ_(Replicates)) × 100.

Z′ factors (Z′) were determined from 48 replicate wells containing 10 µM inducer (positive control) and 48 replicate wells containing 0 µM inducer (negative control).⁷ EC₅₀s were computed using a one-site hyperbola equation: Y = Y_max × X/ (EC₅₀ + X) + C. The zero compound value was set as a constant (C) equal to 1, Y = fold induction, X = [inducer], and EC₅₀ = half maximal induction, with Prism Software, version 5.03, from GraphPad Software (La Jolla, CA).

Automated workflow

Assay setup

Table 1 describes the general automated multiplexed assay protocol. Labcon (Petaluma, CA) 200-µL robotic tips: sterile, low retention were used for diluent addition and serial dilutions and to transfer the dilution series to 384-well cell plates. Dilutions were carried out in 96-well format using Corning (Corning, NY) 96-well clear, round-bottom, polypropylene plates. The hepatocytes were plated in collagen I 384-well black, clear-bottom microplates. Media from incubations containing Luciferin-IPA were transferred to and combined with LDR in Corning 384-well white, flat-bottom, polystyrene, nontreated plates.

Cells, media, compounds, salicylamide, resorufin ethyl ether, and CTG were dispensed at 25-µL volumes, and Luciferin-IPA in medium was dispensed at a 50-µL volume. The medium transfer following the substrate incubation and LDR addition were completed using a 30-µL volume. All assay plates were mixed on an orbital shaker for 30 s at approximately 1000 RPM following the LDR and CTG additions to increase reaction consistency among replicates. According to the manufacturer’s protocols, LDR is stable at room temperature for 24 h, and CTG is stable at room temperature for up to 8 h.

Hepatocyte dispensing validation

The EL406’s ability to properly dispense hepatocytes in a 384-well format was validated by analyzing the CV of the luminescent cell viability signal from replicate wells. Hepatocytes were dispensed at 25 µL, into columns 1 to 22 of the 384-well collagen I–coated black, clear-bottom microplates described previously. A total of 10 000 cells were dispensed into each well of the 384-well plates. The cell plates were then placed into a 37 °C/5% CO₂ cell incubator for 24 h. After incubation, 25 µL of reconstituted CTG was dispensed to each test plate well. The plates were covered, shaken at approximately 1000 RPM for 30 s, and incubated for 10 min at room temperature. The luminescent signal from each well was then quantified using the Synergy H4 microplate reader. The CV was then determined.

Z′ factor validation

The quality of the automated assay was assessed before proceeding with the compound profiling. Hepatocytes from donor 1 were dispensed into the first eight rows of a collagen I 384-well cell plate per Table 1 . The cells were incubated for 48 h, with fresh media applied to the cells after 24 h. Forty-eight replicates of 10 µM or 0 µM inducer, diluted in InVitro Gro HI media, were then added to the 384-well cell plate following media removal. The final DMSO concentration was 0.5%. Omep was the CYP1A assay inducer, and Rif was the CYP3A4 assay inducer. Wells containing 10 µM inducer represent the positive control, whereas the wells containing 0 µM inducer represent the negative control. The process was repeated after 24 h, and the remainder of the assay was performed per Table 1 . Fold induction by the 10-µM inducer and the Z′ factor for each automated assay were then computed.

EC₅₀ and E_max validation

The automated triplex assay was run as described in Table 1 using known inducers of CYP1A (Omep, β-NF, 3-MC, Lpz) and CYP3A4 (Pb, Rif, SR, Dex). Compound serial 1:2 dilutions were performed using the Precision for half maximal (EC₅₀) and maximum fold induction (E_max) determinations. Each concentration–response curve included a vehicle control point. The assay was run in a 384-well format.

CYP induction profiling

To test the CYP profiling triplex validity and reproducibility, assays were run using three hepatocyte donors. Three separate runs were performed on separate days with hepatocytes from donor 1, whereas one run was performed with hepatocytes from donors 2 and 3. Hepatocytes were dispensed in a 384-well format with the EL406 and incubated for 48 h, including medium exchange. Following the incubation, serial 1:2 titrations were performed with eight separate compounds to create the titration curves. Following media removal, the titrated compounds were added to the wells. Final starting concentrations were 500 µM for Omep, β-NF, Pb, and Dex; 300 µM for Lpz; 50 µM for 3-MC and Rif; and 5 µM for SR. The remainder of the automated triplex assay was performed as described above.

Results and Discussion

Hepatocyte dispensing validation

A target cell density of 180 000 hepatocytes per cm² was used to achieve a confluence above 90%, which is concordant with literature values.⁸ To assess automated hepatocyte dispensing precision to obtain equivalent cell density between wells, a CV was calculated from the luminescent cell viability signals detected among the 352 replicate wells tested in the 384-well format. The CV value measured was 4.34%, which compares favorably with previously reported values for automated reagent dispensing of 4% to 10%.⁹

Z′ factor validation

The 384-well CYP1A and CYP3A4 induction assay quality was assessed by performing Z′ factor determinations. Z′ factor, as a statistical parameter, is commonly used for assay quality assessment in drug discovery applications. Therefore, it is widely adopted as a measurement of automated assay quality where a value ≥0.5 is considered an excellent assay, providing a basis for accurate pharmacological measurements. The Z′ values were 0.73 and 0.77 for CYP1A induction by Omep and CYP3A4 induction by Rif, respectively, and these values indicate high-quality assays.⁷

Automated triplex assay performance

To evaluate the automated triplex assay performance, dose–response treatments with known inducers of CYP1A and CYP3A4 enzymes were carried out along with a cell viability measurement ( Fig. 1 ). Relative viable cell signals measured with this assay, which correlates adenosine triphosphate (ATP) content with viable cell number, are not influenced by the CYP substrates applied at the concentrations used in the triplex format (data not shown). Three independent assay tests using donor 1 hepatocytes and single tests with donor 2 and 3 hepatocytes were performed. Fold stimulation was determined at each concentration across a dose range of the two compounds. The calculation of fold induction (induced signal/basal signal) requires assay chemistries that are sensitive enough to significantly detect basal activity signals over assay background. The CYP assays gave robust basal measurements that varied among the donors with an average signal/noise for CYP3A4 of 49 for the three runs with donor 1, as well as 32 and 14 for donors 2 and 3 (S/N = signal mean/standard deviation of the background). For the CYP1A assays, the basal S/N values for the three runs were 25 for donor 1, as well as 14 and 11 for donors 2 and 3. In each case, basal signals were significantly higher than background as evidenced by unpaired t test p values <0.0001.

Fig. 1.

Fold induction/cell viability. A dilution series was created for each compound, using a 1:2 dilution scheme. (A) Omeprazole: 500–1 µM. (B) β-Napthoflavone: 500–1 µM. (C) 3-Methylcholanthrene: 50–0.1 µM. (D) Lansoprazole: 300–0.6 µM. (E) Phenylbutazone: 500–1 µM. (F) Rifampicin: 50–0.1 µM. (G) SR12813: 5–0.01 µM. (H) Dexamethasone: 500–1 µM. CYP1A and -3A4 fold induction (left y-axis) and % cell viability (right y-axis) plotted for each compound concentration (x-axis). Basal levels (0 µM compound) are shown as left-most point on the x-axis. Error bars represent standard deviation of n = 4 values for run 3 with donor 1 hepatocytes.

We tested the triplex system using eight compounds that are well documented as CYP1A2 and -3A4 inducers via associated AHR and PXR nuclear receptors. Omep, β-NF, and 3-MC are recognized as preferred CYP1A2 inducers, as noted in the FDA Guidance for Industry,¹ due to their extensive references,¹⁰ and from regulatory submissions. In addition, Lpz has been cited as a potent CYP1A2 inducer.¹¹ For PXR activation and CYP3A4 induction, Rif is cited as a preferred inducer, whereas Dex is considered an acceptable inducer according to the FDA Guidance for Industry.¹ Rif is well documented as a positive control for PXR activation and subsequent CYP3A4 induction.¹⁰ Dex,^10,12 SR,¹³ and Pb¹⁴ are also potent CYP3A4 inducers.

Figure 1 represents the data generated with the eight inducers for a single run with a single hepatocyte donor. Maximal increases between 11- and 48-fold were seen in the CYP1A assay from the known AHR receptor activators across the concentration–response curve, with no discernable activation seen with the CYP3A4 assay. The opposite is true with the known PXR activators, where a maximal 5- to 30-fold increase was seen in the CYP3A4 assay and minor activation seen with the CYP1A assay at the highest concentrations. Notably, Pb had a fold induction value for CYP1A of 5.9 at 500 µM. Substantial negative cell viability effects were observed at high concentrations with β-NF and Lpz. This loss in cell viability at approximately 100-µM compound concentrations and higher corresponds to marked changes in apparent fold induction trends for each compound.

The test compound induction potency can be expressed as an EC₅₀ that represents the concentration where half maximal induction is observed. These values were calculated from curve fits of the ascending portion of the induction profiles ( Fig. 2 ). The three donors varied in basal, E_max, and EC₅₀, as is commonly observed when different donors are compared ( Table 2 ).¹⁴ However, in considering Rif and Omep respectively as prototypical CYP1A and CYP3A4 inducers, variability for induction from run to run and donor to donor was minimal at less than twofold. Specifically, for the three runs with donor 1, CVs for EC₅₀ and E_max were 63% and 24% for CYP1A Omep induction and 24% and 32% for CYP3A4 Rif induction. Across the three donors, CVs for EC₅₀ and E_max were 44% and 51% for CYP1A Omep induction and 62% and 13% for CYP3A4 Rif induction. The EC₅₀ values measured here are consistent with previously reported values for AHR activation by Omep of 5.9 µM,¹⁵ as well as for PXR activation by Rif of 0.8 µM.¹⁶ The EC₅₀ values can also be arranged to create rank orders of potency for CYP1A: 3-MC < β-NF ≈ Omep < Lpz, or CYP3A4: SR < Rif < Dex < Pb.

Table 2.

Compound EC₅₀/E_max Values

	EC₅₀ (µM)				E_max (Fold Induction)
CYP1A induction	Omep	β-NF	3-MC	Lpz	Omep	β-NF	3-MC	Lpz
Donor 1, runs 1–3	7.6 ± 4.8	5.9 ± 3.5	0.5 ± 0.1	13.2 ± 8.7	32.3 ± 7.9	25.9 ± 4.2	44.1 ± 7.1	13.7 ± 2.4
Donors 1–3	5.2 ± 2.3	4.7 ± 2.5	0.3 ± 0.2	5.8 ± 6.5	20.3 ± 10.4	20.1 ± 5.1	26.3 ± 15.5	9.6 ± 3.6
CYP3A4 Induction	Pb	Rif	SR	Dex	Pb	Rif	SR	Dex
Donor 1, runs 1–3	14.1 ± 1.6	2.1 ± 0.5	0.3 ± 0.1	9.0 ± 1.9	19.3 ± 5.6	23.4 ± 7.4	18.4 ± 5.5	8.0 ± 2.9
Donors 1–3	21.2 ± 6.5	1.3 ± 0.8	0.15 ± 0.1	6.6 ± 2.7	16.0 ± 2.9	20.4 ± 2.6	14.6 ± 3.5	6.1 ± 1.8

Values are mean ± standard deviation. EC₅₀ values were derived from curve fits as shown in Figure 2 ; E_max are concentrations where highest fold inductions were observed. Omep, omeprazole; β-NF, β-napthoflavone; Pb, phenylbutazone; Dex, dexamethasone; Lpz, lansoprazole (Lpz); 3-MC, 3-methylcholanthrene; Rif, rifampicin; SR, SR12813.

Fig. 2.

Dose-dependent CYP induction. The ascending phase of dose-dependent CYP1A and CYP3A4 induction responses from run 3 with donor 1 are shown. Fold induction values were calculated for each of four replicate samples at each drug concentration; the mean ± SD is shown. Fold induction is the fluorescent (CYP1A/EROD) or luminescent (CYP3A4/Luciferin-IPA) value divided by the average of the vehicle-treated controls (n = 4). The calculated EC₅₀ values are presented in Table 2 .

The data may be used in more complex calculations from other in vitro induction models such as PXR transactivation, where effect may be calculated from E_min, E_max, EC₅₀, and Hill coefficient values.¹⁷ As well, this system may be redesigned to increase the compounds tested by limiting to one concentration for rank ordering of compounds or calculating induction potential as a positive control percentage. These benefits are compounded by the fact that both CYP1A and CYP3A4 data were generated from the same well, providing a direct comparison without potential confounding variability from separate well conditions, compound dosing, and signal generation.

Data quality was enhanced in the triplex assay by providing a viability component and generating induction data from the same well. In this way, data were normalized to account for any hepatocyte dispensing and culturing variations within the same experiment or from experiments carried out on different days. Besides an experimental variation control, viability can be used to determine compound toxicity within the concentration– response curve to aid in induction data interpretation. For instance, Lpz produced an E_max for CYP1A induction at 37.5 µM concentration and showed a substantial decline above that concentration ( Fig. 1 ). Viability also declined between 75 and 300 µM, indicating that decreased fold-induction in this range was influenced by Lpz toxicity. A similar trend was observed for β-NF. Conversely, the other compounds caused little or no loss in viability across the concentration–response curve. Nevertheless, at high concentrations, they did cause slight (Rif, SR, Dex) or substantial (Omep, 3-MC, Pb) declines in fold induction compared to E_max. This may be interpreted as an induction mechanism suppression or CYP inhibition.¹⁸ It is also noteworthy that modest basal CYP3A4 inhibition was observed at high Omep and Lpz concentrations. Omep is a known CYP3A4 substrate that apparently acts here as a competitive CYP3A4/Luciferin-IPA reaction inhibitor.¹⁹ The apparent inhibition of CYP3A4 by Lpz must be understood in light of its toxicity in the inhibitory concentration range.²⁰ Conversely, CYP1A induction was observed at the high end of the CYP3A inducer concentration ranges (Pb, SR, Dex). Pb presented a more complex action, where E_max for CYP3A4 was observed at 125 µM with lower CYP3A4 induction at 250 and 500 µM, whereas CYP1A induction was observed at 250- and 500-µM concentrations. Nevertheless, effects observed at high concentrations that are not explained in terms of cytotoxicity should be interpreted with caution due to the potential for off-target effects or toxicities that fall short of causing ATP depletion.

The triplex assay described here offers a straightforward and robust way to assess test compound effects on CYP1A and CYP3A4 enzymes, as well as any potential toxic effects that these compounds might have. The induction data generated from known inducers provided an assay quality and utility assessment for human cryopreserved hepatocytes in a 384-well format. The miniaturized format allowed for quantitative E_max and EC₅₀ measurements from fewer hepatocytes than used in a traditional larger well plate format. The procedure was easily automated using instrumentation that ensures sterile manipulation of cells, media, and assay components. Hepatocyte dispensing, Z′ factor, and EC₅₀ validation data illustrate the automated assay’s capability to deliver accurate and reproducible data. The mixing of viability, CYP1A, and CYP3A4 assays in an automated and miniaturized format provides complex data sets in a single pass. This approach optimizes data extraction from hepatocytes, a costly reagent; minimizes resource time; and provides a basis for more biologically relevant data interpretations than can be achieved with a single-parameter assay. The combination of assays and instrumentation provides a practical solution for automated CYP induction profiling.

References

U.S. Food and Drug Administration. Guidance for Industry: Drug Interaction Studies—Study Design, Data Analysis, and Implications for Dosing and Labeling; U.S. Department of Health and Human Services: Rockville, MD, 2006.

Willson

T. M.

Kliewer

S. A.

PXR and CAR Drug Metabolism. Nat. Rev. Drug Discov. 2002, 1(4), 259–266.

Nebert

D. W.

Dalton

T. P.

Okey

A. B.

Gonzalez

F. J.

Role of Aryl Hydrocarbon Receptor-Mediated Induction of the CYP1 Enzymes in Environmental Toxicity and Cancer. J. Biol. Chem. 2004, 279(23), 23847–23850.

Roymans

Annaert

Van Houdt

Weygers

Noukens

Sensenhauser

Silva

Van Looveren

Hendrickx

Mannens

. Expression and Induction Potential of Cytochromes P450 in Human Cryopreserved Hepatocytes. Drug Metab. Dispos. 2005, 33(7), 1004–1016.

Cali

J. J.

Sobol

Uyeda

H. T.

Meisenheimer

CYP3A4 P450-Glo® Assays with Luciferin-IPA: The Most Sensitive and Selective Bioluminescent CYP3A4 Assay. Promega Corporation Web site. http://www.promega.com/resources/articles/pubhub/cellnotes/cyp3a4-p450-glo-assays-with-luciferin-ipa-the-most-sensitive-and-selective-bioluminescent-cyp3a4/ (accessed Apr 15, 2011).

A. P.

Evaluation of Luciferin-Isopropyl Acetal as a CYP3A4 Substrate for Human Hepatocytes: Effects of Organic Solvents, Cytochrome P450 (P450) Inhibitors, and P450 Inducers. Drug Metab. Dispos. 2009, 37(8), 1598–1603.

Zhang

J. H.

Chung

T. D.

Oldenburg

K. R.

A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays. J. Biomol. Screen. 1999, 4(2), 67–73.

A. P.

Brent

J. A.

Pham

Fackett

Ruegg

C. E.

Silber

P. M.

Cryopreserved Human Hepatocytes: Characterization of Drug-Metabolizing Enzyme Activities and Applications in Higher Throughput Screening Assays for Hepatotoxicity, Metabolic Stability, and Drug-Drug Interaction Potential. Chem. Biol. Interact. 1999, 121(1), 17–35.

Liu

Southall

Titus

S. A.

Inglese

Eskay

R. L.

Shinn

Austin

C. P.

Heilig

M. A.

Zheng

A Multiplex Calcium Assay for Identification of GPCR Agonists and Antagonists. Assay Drug Dev. Technol. 2010, 8(3), 367–379.

10.

Bjornssen

T. D.

Callaghan

J. T.

Einolf

H. J.

Fischer

Gan

Grimm

Kao

King

S. P.

Miwa

. The Conduct of In Vitro and In Vivo Drug-Drug Interaction Studies: A Pharmaceutical Research and Manufacturers of America (PhRMA) Perspective. Drug Metab. Dispos. 2003, 31(7), 815–832.

11.

Bowen

W. P.

Carey

J. E.

Miah

McMurray

H. F.

Munday

P. W.

James

R. S.

Coleman

R. A.

Brown

A. M.

Measurement of Cytochrome P450 Gene Induction in Human Hepatocytes Using Quantitative Real-Time Reverse Transcriptase–Polymerase Chain Reaction. Drug Metab. Dispos. 2000, 28(7), 781–788.

12.

Ogg

M. S.

Williams

J. M.

Tarbit

Goldfarb

P. S.

Gray

T. J.

Gibson

G. G.

A Reporter Gene Assay to Assess the Molecular Mechanisms of Xenobiotic-Dependent Induction of the Human CYP3A4 Gene In Vitro. Xenobiotica 1999, 29(3), 269–279.

13.

Ekins

Erickson

J. A.

A Pharmacophore for Human Pregnane X Receptor Ligands. Drug Metab. Dispos. 2002, 30(1), 96–99.

14.

Lin

J. H.

A. Y.

Interindividual Variability in Inhibition and Induction of Cytochrome P450 Enzymes. Ann. Rev. Pharmacol. 2001, 41, 535–567.

15.

Kanebratt

K. P.

Andersson

T. B.

HepaRG Cells as an In Vitro Model for Evaluation of Cytochrome P450 Induction in Humans. Drug Metab. Dispos. 2008, 36(1), 137–145.

16.

Sahi

Hamilton

Sinz

Barros

Huang

S. M.

Lesko

L. J.

LeCluyse

E. L.

Effect of Troglitazone on Cytochrome P450 Enzymes in Primary Cultures of Human and Rat Hepatocytes. Xenobiotica 2000, 30(3), 273–284.

17.

Ripp

S. L.

Mills

J. B.

Fahmi

O. A.

Trevena

K. A.

Liras

J. L.

Maurer

T. S.

de Morais

S. M.

Use of Immortalized Human Hepatocytes to Predict the Magnitude of Clinical Drug-Drug Interactions Caused by CYP3A4 Induction. Drug Metab. Dispos. 2006, 34(10), 1742–1748.

18.

Kajosaari

L. I.

Laitila

Neuvonen

P. J.

Backman

J. T.

Metabolism of Repaglinide by CYP2C8 and CYP3A4 In Vitro: Effect of Fibrates and Rifampicin. Basic Clin. Pharmacol. Toxicol. 2005, 97(4), 249–256.

19.

Abelö

Andersson

T. B.

Antonsson

Naudot

A. K.

Skânberg

Weidolf

Stereoselective Metabolism of Omeprazole by Human Cytochrome P450 Enzymes. Drug Metab. Dispos. 2000, 28(8), 966–972.

20.

Pearce

R. E.

Rodrigues

A. D.

Goldstein

J. A.

Parkinson

Identification of the Human P450 Enzymes Involved in Lansoprazole Metabolism. J. Pharmacol. Exp. Ther. 1996, 277(2), 805–816.

Automated Triplexed Hepatocyte-Based Viability and CYP1A and -3A Induction Assays

Abstract

Keywords

Introduction

Materials and Methods

Materials

Liquid handling

Luminescent and fluorescent signal detection

Methods

Cell preparation

CYP assays and substrate preparation

Luminescent detection reagent preparation

Compounds

Data analysis

Automated workflow

Assay setup

Hepatocyte dispensing validation

Z′ factor validation

EC50 and Emax validation

CYP induction profiling

Results and Discussion

Hepatocyte dispensing validation

Z′ factor validation

Automated triplex assay performance

References

EC₅₀ and E_max validation