The document describes experiments characterizing a HepaRG cell line with MRP2 (an efflux transporter) knocked out using zinc finger nuclease technology. Western blot and immunostaining showed loss of MRP2 protein expression in knockout cells. Assays found control cells accumulated the fluorescent MRP2 substrate CDCF in bile canaliculi, while knockout cells did not, demonstrating functional loss of MRP2. The MRP2 knockout cell line allows investigation of MRP2-associated drug toxicity without its protective efflux.

1. Figure 6. Differentiated 5F control and MRP2 KO HepaRG cells were treated in triplicate with 3uM CDCFDA in standard
Ca2+ HBSS buffer with or without 50uM MK571 inhibitor. The plate was imaged with a GE IN Cell Analyzer 2200 high
content imaging system. The cell images were analyzed for mean fluorescent intensity with GE IN Cell Analyzer 2200
Workstation software. Error bars represent standard deviation of mean fluorescent intensity among 48 fields of view taken
from triplicate wells of each test condition. The bottom of the figure shows a representative image of CFDA fluorescence
accumulation for each test condition. The images demonstrate a significant accumulation of CDCF fluorescence in bile
canaliculi of 5F control cells, while bile canaliculi of MRP2 KO cells remain dark and lack accumulation. When both cell
types are treated with the MK571 inhibitor compound, there is a dramatic increase of CDCF fluorescence inside the cells.
This suggests that CDCF is not solely an MRP2 substrate in differentiated HepaRG liver cells. The images also demonstrate
that the average mean fluorescent intensity of CDCF in the 5F control and the MRP2 KO cells are very similar in value due
to CDCF accumulation in the cytoplasm of the MRP2 KO cells.
Figure 7. Zoomed-In images of HepaRG 5F
control and MRP2KO cells from Figure 6,
demonstrates a significant accumulation of
CDCF fluorescence (GREEN) in bile canaliculi
(arrows) of 5F control cells, while bile canaliculi
(arrows) of MRP2 KO cells remain dark and lack
accumulation.
Figure 2. Immunochemical
staining to confirm loss of MRP2
expression from bile canaliculi in
MRP2 KO versus control cells. No
MRP2 expression was observed in
MRP2 KO cells. Bile canaliculi
(arrows) are clearly visible in
both MRP2 KO and control cells.
Images acquired using GE IN Cell
Analyzer 2200 high content
imaging system.
Figure 1. Western Blot Analysis of
differentiated HepaRG 5F control, MDR1KO,
and MRP2 KO cell lines. Demonstrating the
loss of MRP2 protein in the MRP2 KO HepaRG
cells.
Protein Expression
Western Blot Analysis
0
2000
4000
6000
8000
10000
12000
5F Control MRP2 KO 5F Control
MK571
MRP2 KO
MK571
Generation of MRP2 Efflux Transporter Knock-Out in HepaRG Cell Line
Jennifer Pratt, Maureen Bourner, Toni Steiner, Yongling Xiao, and David Thompson
Life Science and Technology Center, MilliporeSigma, St. Louis, MO 63103
A HepaRG cell line with a total disruption of the ABCC2 gene encoding multidrug resistance protein 2 (MRP2) was generated using Sigma’s Zinc
Finger Nuclease (ZFN) technology. Both gene and protein expression for the MRP2 efflux transporter were altered as shown by DNA sequence and
Western analysis when compared to HepaRG control cell line. The purpose of this study was to prove that the HepaRG MRP2 knock-out (KO) cells
lack function for this efflux transporter.
Staining of differentiated control and HepaRG MRP2 KO cells suggests robust bile canaliculi formation in both control and knock-out cells, and loss of MRP2 protein within bile canaliculi of knock-out cells. The non-fluorescent reagent 5(6)-carboxy-2’,7’-dichlorofluorescein diacetate (CDCFDA) is known to be passively
taken up by cells and hydrolyzed by intracellular esterases to the fluorescent molecule CDCF which can then be exported into the bile canaliculi by MRP2. The HepaRG MRP2 KO cell line has demonstrated a loss of accumulation of CDCF in bile canaliculi when compared to control cell line.
Characterization of the HepaRG MRP2 KO cell line was performed predominately by fluorescent imaging of MRP2, phalloidin (F-actin) counterstain,
and CDCF in bile canaliculi of differentiated cells.
Materials
HepaRG cell lines (5F control, MRP2KO, and MDR1 KO), growth and differentiation supplements were from Sigma Aldrich. Trans-It mRNA
transfection reagent was purchased from Mirus Bio (Madison, WI). Glutamax was purchased from Life Technologies (Carlsbad, CA), Williams E,
Pen/Strep, DMSO, Accumax, substrates, and inhibitors were from Sigma Aldrich (St. Louis, MO). Western Blot reagents and antibodies. Phalloidin
Alexa594, ibiTreat 24-well black-walled imaging plates from ibidi (Madison, WI), BD Cytofix/Cytoperm™ Kit from BD Biosciences (San Jose, CA),
phalloidin-Alexa594 from Molecular Probes (Eugene, OR)
Differentiation of HepaRG cells
A sandwich culture model of differentiated HepaRG cells was used for all experiments. The cells were seeded on 24-well collagen-coated plates
(0.4X10E6 cells/well) and maintained in growth media for 5 days. Differentiated cells analyzed for transport of CDCF and immunochemical staining
by high-content image analysis were plated at the same density and with same media conditions, but in an ibidi 24-well specialized imaging plate.
Cells were then maintained in recovery media for 2 days, followed by differentiation for 12 days in differentiation media. After 12 days, a Matrigel
matrix(0.25mg/ml) was added onto the cells for 3days. The assay was conducted on day 21.
Western blot analysis
Western Blot Analysis of 5F control, MDR1KO, and MRP2 KO differentiated HepaRG cell lines was used to measure protein expression.
Electrophoresis was performed with approximately 10ug of protein per well from cell lysates on a NuPage Novex 4-12% Bis Tris gel from Thermo
Fisher Scientific (Waltham,MA) in MOPS buffer according to manufacturer’s instructions. Performed protein transfer with BioRad TransBlot Turbo
Transfer System using PVDF membrane and mixed MW protocol by BioRad (Hercules, CA). Western Blot detection was performed with an overnight
incubation at 40C with 250X dilution of primary antibodies - rabbit monoclonal anti-MRP2(12559, Cell Signaling Technology Danvers,MA), rabbit
monoclonal anti-MDR1(ab170904, Abcam Cambridge, UK), and polyclonal rabbit anti-GAPDH (G9545, Sigma Aldrich, MO). Secondary antibody
detection was performed for 1 hour at room temperature using Goat anti-rabbit peroxidase (A0545, Sigma Aldrich, St. Louis, MO) at a 10,000X
dilution. Super Signal West Dura detection reagent was used for detection (Thermo Fisher Scientific,Waltham,MA) (Figure 1)
Transporter assays
For transport of digoxin, differentiated HepaRG cells in sandwich culture were pre-incubated in standard Ca2+ or Ca2+-free HBSS for 30 minutes (all
Ca2+-free HBSS buffer mentioned also contained 0.1mM EGTA from Sigma Aldrich) , this was then followed by a 30 minute incubation of 10uM
digoxin dissolved in standard Ca2+ HBSS. The reaction was stopped by adding ice-cold HBSS and washed 3 times with HBSS. The intracellular
accumulation of compound was determined by LC/MS following extraction with methanol. Protein concentrations were determined using BCA
protein assay after RIPA buffer treatment of the monolayer. Transporter activity was expressed as pmol per min per mg protein (pmol/min/mg
protein) before calculating for the biliary excretion index (Figure 5).
For transport of CDCF, differentiated HepaRG cells in sandwich culture were pre-incubated for 30minutes with standard Ca2+ HBSS buffer or Ca2+-
free HBSS with or without 50uM MK571 inhibitor. The cells were then treated in triplicate with 3uM CDCFDA in standard Ca2+ HBSS buffer with or
without the same inhibitor. The reaction was stopped by adding ice-cold HBSS and washed 3 times with HBSS. The intracellular accumulation of
compound was imaged with the GE IN Cell Analyzer 2200 high content imaging system. The cell images were analyzed for mean fluorescent
intensity with GE IN Cell Analyzer 2200 Workstation software. The imager collected 16 fields of view per well using a 20X objective and wavelength
excitation/emission optimal for CDCF. The reported well-by-well summary was used for subsequent analysis. (Figure 4, Figure 6, and Figure 7).
Biliary excretion index (BEI) was calculated by using the following formula BEI = [(Uptake(+Ca) – Uptake(-Ca))/Uptake(+Ca)] X 100.
Immunochemical and phalloidin staining
Differentiated HepaRG cells in sandwich culture were fixed and permeabilized using protocol and reagents from BD Cytofix/Cytoperm™ Kit.
Immunofluorescence was done with a primary antibody - a mouse monoclonal anti-MRP2(ab3373, Abcam Cambridge, UK ), 1:100 dilution two
hours; and a secondary antibody - goat F(ab’)2 anti-mouse IgG (Alexa488) from Life Technologies (Carlsbad, CA), 1:2000 dilution for 1 hour. F-
actin was revealed by 0.2nM phalloidin-Alexa594 (Figure 2).
For experiment with CDCF transport followed by f-actin detection with phalloidin counterstain, cells were differentiated in 24-well collagen-coated
plates treated for CDCF transport activity as described above, then cells were fixed, permeabilized and stained with phalloidin-alexa594. For this
data, images were collected using a Nikon Eclipse TS100 inverted microscope, MetaVue software and a 20X objective by bright field and
fluorescence. Filter sets used were optimal for CDCF (GREEN,ex465-495/em515-555) and phalloidin-Alexa594 (RED,ex540-580/em600-660)
(Figure 3).
The expression and function of both sinusoidal and canalicular drug transporters in differentiated HepaRG cells have
been shown to be comparable to primary human hepatocytes. The loss of MRP2 protein expression in the HepaRG MRP2
KO cells has been proven. Generation of this cell line allows for detailed investigations of MRP2-associated drug-induced
liver injury (DILI) including; drug transport and accumulation, drug-drug interactions, MRP2 substrate assessment, and
metabolism-mediated DILI.
PURPOSE
METHODS
RESULTS
CONCLUSIONS
M1022
MilliporeSigma and the M logo are trademarks of Merck KGaA, Darmstadt, Germany. Durapore, Millipore Express and Viresolve are registered trademarks of Merck KGaA, Darmstadt, Germany.
© 2016 EMD Millipore Corporation, Billerica, MA USA. All rights reserved. Lit. No. PBXXXXXXXX XXX
The life science business of Merck KGaA, Darmstadt, Germany operates as MilliporeSigma in the U.S. and Canada
www.sigmaaldrich.com
Biliary Excretion Index (BEI) was performed under a use license for B-CLEAR® technology from Qualyst Transporter Solutions (QTS).
CFDA Accumulation in Bile Canaliculi & Cytoplasm
Mean Fluorescent Intensity
Average of 48 Images
FluorescentUnits
0
5
10
15
20
25
30
35
40
45
50
5F Control MRP2 KO
MDR1 Activity - Digoxin
MRP2
MDR1
GAPDH
HepaRG Cell Line
Protein
Specific
Antibody
MRP2 Merged
5F Control
MRP2 KO
F-Actin
Bile Canaliculi MRP2 Merged
CDCF
CDCF
5F Control
MRP2 KO
Loss of MRP2 Protein Expression Disposition of CDCF in Presence of the MRP Inhibitor, MK571
Efflux Transporter Activity: 5F vs MRP2 KO
Lack of Accumulation of CDCF in
Bile Canaliculi from MRP2 KO Cells
Figure 3. Differentiated 5F control and MRP2 KO HepaRG cells were treated for 30
minutes with 3uM CDCFDA dissolved in +Ca HBSS buffer. Images were collected using
a Nikon Eclipse TS100 microscope and a 20X objective by bright field and fluorescence,
using a filter optimal for CDCF detection(GREEN). The same CDCF treated cells were
then fixed, permeabilized, and stained with Phalloidin-Alexa594 for F-Actin visualization.
Image was taken with filter optimal for Alexa594 to visualize the punctate F-actin stain
which indicates bile canaliculi structures (RED). This demonstrates that although MRP2
KO cells form bile canaliculi, they lack accumulation of CDCF in these structures,
indicating a hepatocyte culture that lacks MRP2 function.
Figure 4 . After a 30 minute pre-incubation in
standard Ca2+ or Ca2+-free HBSS, cumulative
uptake of 3uM CDCFDA dissolved in standard
Ca2+ HBSS for 30 minutes was determined,
Error Bars represent the STDEV for BEI of
CDCF for n = 3.
F-Actin
Bile Canaliculi
F-Actin
Bile Canaliculi
F-Actin
Bile Canaliculi
0
10
20
30
40
50
60
70
80
5F Control MRP2 KO
MRP2 Activity - CDCF
BEI
BEI
65%
36%
31%
35%
Figure 5 . After a 30 minute pre-incubation in
standard Ca2+ or Ca2+-free HBSS, cumulative
uptake of 10uM digoxin dissolved in standard
Ca2+ HBSS for 30 minutes was determined,
Error Bars represent the STDEV for BEI of
digoxin for n = 3.
SPECIAL THANKS TO
Kelly Keys and Lillian Vickery (Flow cytometry sorting and analysis), James Blasberg (LC/MS),
Tim Brayman (cell line metabolism validation), and Thomas Juehne (High content imaging)
CDCF
CDCF
5F Control
MRP2 KO
Common Human Liver Transporters