December 1, 2012 (Vol. 32, No. 21)
Kathleen E. Sampson
Michael D. Mitchell
Novel Caco-2 Knockout Cell Lines Accurately Identify Efflux Transporter Interactions
Membrane drug transporters have been identified as a determinant of drug disposition in the body, potentially affecting absorption, pharmacokinetics, drug-drug interactions, and safety profiles. The major efflux transporters of the ABC family, including P-glycoprotein (P-gp, ABCB1, MDR1 gene), BCRP (ABCG2), and MRP2 (ABCC2), are localized to barrier tissues of the body such as intestine, liver, kidney, blood-brain barrier, and placenta, where they efflux a wide range of xenobiotics and chemicals such as statins, macrolide antibiotics, angiotensin blockers, and chemotherapeutic agents, affecting exposure and clearance in vivo.
Recently, the FDA and EMA have issued recommendations and guidelines for drug transporter studies in the pharmaceutical and biotech industries. New chemical entities are now routinely screened for potential transporter recognition and impact; interaction at clinical concentrations may indicate additional clinical trials during drug development.
Current cell-based transporter studies may involve cell lines such as Caco-2 intestinal enterocytes that express multiple transporters, requiring the use of transporter-specific substrates or inhibitors. However, substrates are often recognized by multiple transporters at different affinities, and the specificity of inhibitors is often uncertain or unknown, leading to mistaken interpretations of transporter interaction.
Utilizing CompoZr® zinc finger nucleases (ZFNs) from Sigma® Life Science, MDR1, BCRP, and MRP2 efflux transporter genes were targeted for ZFN-mediated knockout in a Caco-2 cell line. The resultant panel of single and double knockout cells show disruption of gene sequence as well as complete loss of transporter function in bidirectional transport assays to at least 40 passages post ZFN genomic-modification.
In this article, we describe the functional characterization of these cells using a set of industry-standard compounds targeting single transporters, as well as compounds displaying overlapping transporter specificity. These novel cells are powerful tools to elucidate transporter interactions without dependence on chemical inhibitors and clarify the potential impact of specific efflux transporters in drug disposition.
Cell Culture: Caco-2 subclone C2BBe1cells were obtained from ATCC and cultured in DMEM High Glucose with 10% Fetal Bovine Serum, sodium pyruvate, non-essential amino acids, L-Glutamine, and Pen/Strep. Medium was changed every 2−3 days. Cells were passaged at least once/week.
Cell Line Modification: The CompoZr protocol for cell-line modification was used to deliver ZFN pairs into the C2BBe1 cells by nucleofection. Viable cells were single -cell sorted by flow cytometry and resultant colonies were tested for mutations. Genomic DNA was amplified using ZFN Cel 1 primers followed by PCR on target regions using nested primers. DNA sequence of the target regions was analyzed to confirm gene disruption through deletion or insertion.
Bidirectional Transport Assay: C2BBe1 cells (wildtype (WT), or knockout (KO)) were cultured on 24-well Costar Transwell® plates for 20−22 days. Apical and basolateral chambers were rinsed twice and then pre-incubated at 37ºC for 30−60 min with Transport Buffer (HBSS, 25 mM glucose, 10 mM HEPES, pH 7.4). Test articles at 5 μM in Transport Buffer were added to donor chambers with fresh transport buffer added to opposite chambers. Plates were incubated at 37ºC for 2 h. Samples from both donor and receiver chambers were quantified for test articles by LC-MS/MS or fluorescence. Lucifer yellow (LY) permeability was examined post-assay to confirm monolayer integrity.
Analysis by LC-MS/MS: Concentration of test articles in donor and receiver chamber samples were analyzed by LC-MS/MS using an API-4000 Q Trap mass spectrometer with a Turbo V atmospheric pressure electrospray ionization source (AB SCIEX). Samples (40 μL) were injected onto a Fortis C8 column (2.1 × 50 mm, 5 μm) and eluted by a mobile-phase gradient specific for each test article (mobile phase A: 4 mM ammonium formate; mobile phase B: 4 mM ammonium formate in 90% (v/v) acetonitrile). Flow rate was 0.5 mL/min. MS conditions: positive or negative ionization mode (4.5 kV spray voltage); source temperature of 450ºC with multiple reaction monitoring specific for each analyte and internal standard (tolbutamide) parent-product ion pairs. Peak areas of analyte, and internal standard and resulting ratios were quantified using Analyst 1.5 (AB SCIEX).
Analysis of CDCF by Fluorescence: The nonfluorescent 5(6)-carboxy-2’,7’-dichlorofluorescein-diacetate (CDCFDA) is taken up by passive diffusion into Caco-2 cells, where it undergoes hydrolysis to fluorescent CDCF. The fluorescent compound CDCF is then effluxed by MRP2. The receiver samples or fully hydrolyzed donor samples were quantified by fluorescence at 485 nm emission, 538 nm excitation, and compared to linear standard curves of CDCF to determine concentration.
Efflux Ratio (ER) Calculations: The apparent permeability (Papp, cm/sec) of test articles were determined for both Apical → Basolateral (A-B) and Basolateral → Apical (B-A) directions in the bidirectional transport assays. The efflux ratio (ER) was determined from:
ER = Papp B-A/Papp A-B
whereby an ER greater than 2 indicates an active transport process and a reduced ER to <2 in KO cell lines identifies test article as substrate to lost transporter.
Polarized transport in C2BBe1 (WT) and MDR1, BCRP, or MRP2 single and double knockout (KO) cell lines was measured in Transwell® bidirectional transport assays using probe substrates at 5 μM for 2 hr at 37 ºC. Contents of both chambers were quantitated using LC-MS/MS, and permeability and efflux ratios (ER) were calculated.
The ERs for MDR1 substrates, digoxin and erythromycin, were reduced to 1.4 and 1.04 in the MDR1 KO cells from 17.7 and 16.8 in the WT cells, respectively (Figure 1). The ER for the MRP2 substrate, 5(6)-carboxy-2’,7’-dichlorofluorescein was reduced to 1.92 in the MRP2 KO from 30.0 in the WT cells (Figure 2).
ERs for BCRP substrates, estrone 3-sulfate and nitrofurantoin, were reduced to 1.76 and 1.68 in the BCRP KO cells from 22.7 and 13.2 in the WT, respectively (Figure 3). Double knockout cell lines (MDR1/BCRP KO, MDR1/MRP2 KO, and MRP2/BCRP KO) were used to identify substrates of multiple transporters. Cimetidine was confirmed as a substrate of both MDR1 and BCRP demonstrating an ER of 0.97 using the MDR1/BCRP KO cell line and ERs of >2 for all other cell lines (data not shown).
We have generated stable MDR1, BCRP, and MRP2 single and double knockout Caco-2 cell lines using CompoZr ZFN technology. The transporter KO cell lines show disrupted DNA sequence due to ZFN activity and loss of transporter function using specific substrates in the bidirectional transport assay format.
These novel knockout cell lines can identify specific transporters by comparison of transport between the WT and the KO cell lines. Digoxin and erythromycin, estrone 3-sulfate and nitrofurantoin, and CDCF were confirmed as substrates of MDR1, BCRP, and MRP2, respectively, using the single and double KO cell lines. Cimetidine was identified as a substrate of both MDR1 and BCRP in the MDR1/BCRP KO cell line.
With regulatory agencies placing greater scrutiny on identifying drug-transporter interactions, these innovative, predictive transporter KO cell lines can clarify complex drug-transporter interactions, revealing previously unidentified transporter involvement. Sigma’s transporter KO cell lines are becoming a valuable tool for application in drug discovery transporter interaction assessment without dependence on the uncertainties of chemical inhibition specificities.