Journal of Biomolecular Screening
The online version of this article can be found at:
2007 12: 1029 originally published online 7 November 2007 J Biomol Screen
Somerville, Steve Nadler and Taosheng Chen
Michele Agler, Margaret Prack, Yingjie Zhu, Janet Kolb, Kimberly Nowak, Rolf Ryseck, Ding Shen, Mary Ellen Cvijic, John
A High-Content Glucocorticoid Receptor Translocation Assay for Compound Mechanism-of-Action
On behalf of:
Journal of Biomolecular Screening
can be found at:
Journal of Biomolecular Screening
Additional services and information for
What is This?
- Nov 7, 2007 OnlineFirst Version of Record
- Dec 17, 2007 Version of Record >>
by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 by guest on October 31, 2013 jbx.sagepub.com jbx.sagepub.comjbx.sagepub.com jbx.sagepub.comjbx.sagepub.com jbx.sagepub.comjbx.sagepub.comjbx.sagepub.com jbx.sagepub.com jbx.sagepub.com jbx.sagepub.comjbx.sagepub.com jbx.sagepub.comjbx.sagepub.com Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from Downloaded from
A High-Content Glucocorticoid Receptor Translocation
Assay for Compound Mechanism-of-Action Evaluation
MICHELE AGLER,1MARGARET PRACK,1YINGJIE ZHU,1JANET KOLB,1
KIMBERLY NOWAK,1ROLF RYSECK,2DING SHEN,3MARY ELLEN CVIJIC,3
JOHN SOMERVILLE,2STEVE NADLER,2and TAOSHENG CHEN4
Ligand-induced cytoplasm to nucleus translocation is a critical event in the nuclear receptor (NR) signal transduction cas-
cade. The development of green fluorescent proteins and their color variants fused with NRs, along with the recent develop-
ments in automated cellular imaging technologies, has provided unique tools to monitor and quantify the NR translocation
events. These technology developments have important implications in the mechanistic evaluation of NR signaling and pro-
vide a powerful tool for drug discovery. The unique challenges for developing a robust NR translocation assay include cyto-
toxicity accompanied with chronic overexpression of NRs, basal translocation induced by serum present in culture medium,
and interference from endogenous NRs, as well as subcellular dynamics. The authors have developed a robust assay system
for the glucocorticoid receptor (GR) that was applied to a panel of nuclear receptor ligands. Using a high-content imaging
system, ligand-induced, dose-dependent GR nuclear translocation was quantified and a correlation with other conventional
assays established. (Journal of Biomolecular Screening 2007:1029-1041)
glucocorticoid receptor, nuclear translocation, high content screening, green fluorescent proteins, androgen receptor
© 2007 Society for Biomolecular Scienceswww.sbsonline.org 1029
involved in development and homeostasis. Advances in the
knowledge of regulation of this target class have come from stud-
ies of the function and structure of the receptors, their tissue
expression patterns, and effects on important regulatory path-
ways, as well as changes in expression and regulation that under-
lie their role in diseased physiology.1,2The actions of these
receptors are determined by a diverse and multistep set of factors.
Nuclear receptors (NRs) have been classified broadly by their lig-
ands, the availability of ligand and coregulators (coactivators and
corepressors), and the underlying protein-protein interactions and
HE NUCLEAR HORMONE RECEPTOR GENE FAMILY encodes
related proteins that regulate transcription of target genes
signaling events that control transcription.3,4Identification and
characterization of therapeutic modulators have been the focus of
NR-based drug discovery efforts. Various assay methodologies
have been used to identify and study NR modulators either with
the full-length receptor or the ligand binding domain (LBD) of the
receptor, in a cell-free or cell-based format.5-9Over the past sev-
eral years, nonsteroidal receptor modulators with tissue selectiv-
ity have been identified for many NRs, which offer the potential
for improved pharmacological profiles over synthetic steroids as
Upon ligand binding, some members of the steroid hormone
receptor family, such as the glucocorticoid receptor (GR), the
androgen receptor (AR), and the mineralocorticoid receptor
(MR), translocate from the cytoplasm to the nucleus, where
they bind to their specific hormone response elements in the
promoter of their target genes and activate transcription. Other
NRs, such as the estrogen receptor (ER), the progesterone
receptor (PR), the heterodimeric receptors such as the vitamin
D receptor (VDR) and the retinoid X receptor (RXR), and some
orphan nuclear receptors (oNHRs), are predominantly located
in the nucleus in the absence of ligand.
The use of the green fluorescent protein (GFP)12,13to construct
fusion proteins allows real-time analysis of intact cells and extends
the potential assay formats for steroid hormone receptors to
include nuclear translocation for the identification of agonists and
antagonists.14The unliganded GFP-GR fusion is completely local-
ized in the cytoplasm and rapidly translocates to the nucleus in a
Research and Development, Bristol-Myers Squibb:1Lead Discovery & Profiling,
Wallingford, Connecticut; 2Immunology & Oncology Drug Discovery, Princeton,
New Jersey; and 3Lead Evaluation, Princeton, New Jersey.
4Department of Chemical Biology & Therapeutics, St. Jude Children’s Research
Hospital, Memphis, Tennessee.
This article was presented as an oral presentation at the Society for Biomolecular
Screening 11th Annual Conference, September 2005, Geneva, Switzerland.
Received Jun 29, 2007, and in revised form Aug 2, 2007. Accepted for publi-
cation Sep 7, 2007.
Journal of Biomolecular Screening 12(8); 2007
ligand-dependent manner with ligand specificity similar to that of
the native GR.15Several investigators have shown subcellular traf-
ficking of NRs,including AR and MR in various cell lines,16,17and
work by Nishi et al.18using fluorescence resonance energy trans-
fer (FRET) microscopy has shown that, upon translocation to the
nucleus,GR and MR can form heterodimers. Other receptors such
as ER, which are predominately localized in the nucleus in the
unliganded state, undergo an alteration in receptor distribution
within the nucleus from a uniform GFP-ER pattern to a punctuate
pattern upon ligand addition.19The ligand-induced intranuclear
reorganization of ER, although not as dramatic as the cytoplasm-
to-nuclear translocation of AR and GR in response to their corre-
sponding ligands, is a measurable signaling event with important
physiological consequences. Subcellular trafficking of regulatory
components can be imaged and quantified using high-content
screening (HCS) technology. This technology consists of sophisti-
cated image analysis algorithms that can convert images obtained
by automated fluorescence microscopy into quantitative numerical
data. This enables the quantification of translocation events, sub-
cellular distribution of proteins, object intensity, and shape
changes on a cell-by-cell basis. HCS technology has been used for
phenotypic screening and compound mechanism-of-action stud-
ies, and it has opened up a new biology-driven approach to drug
We describe here the development and validation of a GR
nuclear translocation assay for ligand mechanism-of-action
studies. This assay has also been extended to include AR translo-
cation. Ligand-induced cytoplasm-to-nucleus translocation is an
early step in the transcription process and can be used to identify
agonists, antagonists, and other modulators of the receptor, and it
can be evaluated in a variety of cellular backgrounds. Because
chronic overexpression of NRs could potentially be unhealthy to
the host cells, a protocol using yellow fluorescent protein–GR
(YFP-GR) transiently expressed in COS-7 cells was developed
Using the Cellomics ArrayScan™ HCS system (www.cel-
lomics.com), ligand-induced GR nuclear translocation events
were imaged and quantified. Dexamethasone (Dex) induced a
rapid, robust, and dose-dependent cytoplasm-to-nucleus GR
translocation in COS-7 cells transfected with the YFP-GR plas-
mid. The EC50 for Dex in the translocation assay was 2 to 3 nM,
which is in close agreement to the results reported from a whole-
cell Dex binding assay.22These results supported the use of a
GR high-content nuclear translocation assay for compound
mechanism-of-action studies. We have used this assay to evaluate
a library of known steroids in a high-throughput screening (HTS)
mode. Our data indicate that this is a robust assay, miniaturized to
a 384-well format, that is amenable for screening larger collec-
tions of compounds when integrated on an automated screening
system.23We also compare and discuss assay correlation
between translocation, ligand binding, and cell-based transacti-
vation assays. Finally, the assay demonstrates the utility of GFP
fusions and the use of transient transfection assays for image-
based HCS for drug discovery.
MATERIALS AND METHODS
Chemicals and reagents
Known steroids (prednisolone, prednisolone-21-acetate, hydro-
cortisone, triamcinolone acetonide, flunisolide, budesonide, proges-
terone, RU-486 [mifepristone], spironolactone, corticosterone,
dexamethasone, β-estradiol) were obtained from Sigma-Aldrich
(St. Louis, MO) and geldanamycin from EMD (San Diego, CA).
All cell culture media and reagents were obtained from Invitrogen
(Rockville, MD). Charcoal/dextran-treated fetal bovine serum
was obtained from Hyclone (Logan, UT), fetal bovine serum
from Invitrogen, and FuGENE 6 from Roche Applied Sciences
(Indianapolis, IN). Molecular biology reagents were from
Invitrogen Life Technologies (Rockville, MD), BD Biosciences
Clontech (Mountain View, CA), and Qiagen (Valencia, CA).
Black, clear-bottom, tissue culture–treated 384-well assay
plates (Falcon #3962) for the translocation assay were obtained
from BD Biosciences (Franklin Lakes, NJ), and 96-well, white
clear-bottom plates for the luciferase assays were obtained from
Corning (#3903) (Corning,NY). The 1536-well,black solid plates
(PerkinElmer #60052358) for the fluorescence polarization (FP)
assay were obtained from Greiner Bio-One (Longwood, FL).
Formaldehyde (37%, v/v) was purchased from J. T. Baker
(Phillpsburg, NJ), and Hoescht 33342 was purchased from
Molecular Probes (Eugene, OR). The luciferase Steady-Glo
reagents were from Promega (Madison, WI), and PMA was from
EMD (San Diego, CA).
Preparation of recombinant fusion protein constructs
The coding regions of the human GR (NM_000176) and the
human AR (NM_000044) were cloned using reverse transcrip-
tase PCR (RT-PCR) as follows.
A set of human RNA was commercially obtained from BD
Biosciences Clontech (bone marrow, brain, colon, heart, kidney,
liver, lung, mammary gland, ovary, pancreas, placenta, prostate,
skeletal muscle, small intestine, spleen, stomach, testis, thymus,
thyroid, uterus), and cDNA was prepared according to the manu-
facturer (SuperScript First-Strand Synthesis System for RT-PCR,
Invitrogen Life Technologies). For the PCR reactions, the individ-
ual cDNA preparations were pooled, and an aliquot of the mixture
was used (approximately 50-ng input RNA/100-µL reaction).
For PCR amplification, a premix volume of 600 to 800 µL was
set up and distributed into 12 wells for a gradient thermocycler
reaction (Eppendorf, Mastercycle Gradient) with between 50 and
70 µL per well. For the reaction, Pfu polymerase and buffer con-
ditions were used according to Stratagene’s directions except that
DMSO to a final concentration of 10% was added. PCR conditions
Agler et al.
1030 www.sbsonline.orgJournal of Biomolecular Screening 12(8); 2007
were as follows: initial 2 min at 95 °C, 40 cycles at 95 °C, 30 s;
gradient midpoint at 60 °C with ±2 °C difference per well (span-
ning approximately between 50 °C and 70 °C) as annealing tem-
perature for 1.5 min, and 2.5-min elongation at 72 °C. This was
followed by a final 10 min at a 72 °C incubation step.
PCR products were gel isolated using the QIAquick Gel
Extraction Kit from Qiagen and cloned into the pENTR/D-TOPO
vector according to Invitrogen Life Technologies. After sequence
verification, the clones were combined with the desired expres-
sion vectors using the Gateway LR Clonase enzyme mix accord-
ing to the manufacturer’s directions.
Vectors used for expression are derivatives of pcDNA3.1
modified to use the gateway technology. Specifically, the N-
termini have either eGFP or YFP (BD Biosciences Clontech)
followed by the gateway conversion cassette to allow in-frame
Primers were designed according to the pENTR Directional
TOPO Cloning KIT. The primers used are as follows:
Primers for GR: “forward primer” CACCTATGGATCC-
and “reverse primer” ATTAAGGCAGTCACTTTTGAT-
Primers for AR: “forward primer” CACCGGATCCAGCTCAAG-
GATGGAAGTGCAG and “reverse primer” AAGCTTTCA-
Translocation assay. For all the translocation experiments,
COS-7 cells were maintained in RPMI 1640 supplemented
with 5% fetal bovine serum in a 5% CO2incubator at 37 °C.
Prior to transfection, the cells were seeded in assay medium,
which consisted of phenol red–free RPMI 1640 media (supple-
mented with 5% charcoal/dextran-treated fetal bovine serum).
Reporter assay. Hela cells (human cervical adenocarcinoma)
were stably transfected with a chimera receptor joining the Gal 4
DNA binding domain and human GR ligand binding domain, as
well as a plasmid that contained a Gal 4- luciferase reporter gene.
The resulting stable cell line,NP-1,was maintained in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 10%
charcoal-treated fetal bovine, 1% penicillin/streptomycin, 1×
MEM nonessential amino acids, 1 mg/mL G418, and 0.3 µg/mL
purimycin. For the assay, cells were seeded into assay plates in
assay media that were the same as cell culture medium without
G418 and purimycin and containing DMEM high glucose with-
out phenol red.
Optimization of a transient transfection protocol
A transient transfection protocol was developed and optimized
for the GR translocation assay in COS-7 cells and further
expanded to include AR translocation. To shorten the time between
the introduction of the DNA construct into the cells and compound
stimulation, as well as to lower the background (complete serum
can induce GR translocation to the nucleus), COS-7 cells were
seeded directly into P75 flasks (2,000,000 cells/P75 flask) in assay
medium and incubated at 37 °C with 5% CO2for 16 h before trans-
fection. The media were then replaced with fresh assay media prior
to transfection and the cells transfected with a mixture of diluted
FuGENE 6 and pcDNA3-YFP-GR. For each flask,the transfection
mixture was prepared by mixing 30 µL of FuGENE 6 into 500 µL
of phenol red–free, serum-free RPMI medium followed by the
addition of 10 µg of pcDNA3-YFP-GR. The mixture was then
incubated at room temperature for 15 min before the addition of
500 µL of phenol red–free,serum-free RPMI medium and the sub-
sequent overlaying of the transfection mixture onto the cells. The
transfected cells were incubated at 37 °C for 6 h, trypsinized, and
seeded into 96-well plates (5000 cells per well in 90 µL of assay
media) or into 384-well plates (2000 cells per well in 25 µL of
assay media). Cells were incubated for an additional 16 h at 37 °C,
5% CO2before stimulation with compounds.
Image acquisition and analysis
Transfected COS-7 cells plated in 384-well plates were treated
with compounds diluted in water/DMSO to give a final assay con-
centration of 1% (v/v) DMSO. Then, 5 µL of compound was
transferred to the cell plates using a 384-well CyBi™ (CyBi Inc.,
Woburn, MA) into columns 1 to 20 with the appropriate concen-
tration of DMSO in columns 21 to 24. Dex (100 nM) was
included on each assay plate in column 23 as an internal control.
Initially, different time points poststimulation (5-150 min) were
evaluated,but 1 h was chosen as the optimal incubation time point
for ease of plate processing. The cells were fixed by adding 25 µL
of fixation buffer (3.7% v/v formaldehyde containing 10 µg/mL
Hoechst 33342 in Dulbecco’s phosphate-buffered saline [DPBS])
directly onto the assay medium and incubated for 15 min at room
temperature followed by 2 washes with DPBS.
This was followed by the addition of 50 µL DPBS to the cell
layer, and the plates were sealed with a transparent plate seal
(PerkinElmer #6005185). Imaging was carried out using the
Cellomics ArrayScan IV™ High Content System (Pittsburg, PA)
equipped with a Xenon-arc lamp. YFP fluorescence images were
acquired with a 10× Plan Fluor 0.5 numerical aperture objective
using Omega dichroic Emitter Pair XF104 filters with an excita-
tion λ = 500 nm (25-nm bandpass) and an emission λ = 545 nm
(35-nm bandpass). The nuclear translocation BioApplication (ver-
sion 2) was used for image analysis. The difference in fluores-
cence intensity between the cytoplasm and the nucleus was used
to quantify the nuclear translocation event. Data were analyzed
using a customized HTS data analysis software package. IC50 val-
ues were determined using curve-fitting software XL-Fit (IDBS
Inc., Emeryville, CA).
High-Content Glucocorticoid Receptor Translocation Assay
Journal of Biomolecular Screening 12(8); 2007www.sbsonline.org 1031
Luciferase assays were performed according to the manufac-
turer’s instructions (Promega) with the slight modification of
reducing the volume of substrate added to the cell plate. The NP-
1 assay was used to identify agonists or antagonists of the human
GR ligand binding domain. Activation of the receptor chimera
transactivates the Ga14 luciferase reporter in a stable Hela cell
line. Cells were seeded at 10,000 cells per well into 96-well,white
clear-bottom plates and incubated overnight in assay media. For
agonist assays, compounds were added and plates incubated in a
cell culture incubator for 18 h at 37 °C, 5% CO2. For antagonist
assays, after compound addition, 10 nM final concentration of
Dex was added before incubation. Luciferase activity was mea-
sured after a 20-min incubation with the addition of Steady-Glo
reagents and read on the Envision™ plate reader (PerkinElmer,
Boston, MA). The agonist effect of compounds was normalized
and expressed as 100 × (Sample – Blank)/(Total – Blank); the
antagonist effect of the compounds was normalized and expressed
as 100 × [1 – (Sample – Blank)/(Total – Blank)]. The Sample is
defined as the luciferase activity in the presence of a test com-
pound; Blank is referred to as the luciferase activity in the presence
of the DMSO control, and Total represents the signal induced by
100 nM Dex.
Fluorescence polarization assay
The GR competition assay was purchased from the Invitrogen
Corporation (Madison, WI) as a complete kit (#2893) that
included buffer, ligand, and protein. This assay uses a TAMARA-
labeled tracer, Fluoromone™ GS Red, which binds to the human
recombinant GR with a Kdof 0.3 nM. The assay was carried out
in the presence of a GR-stabilizing peptide as specified in the
protocol supplied by the vendor. The fluorescence polarization
measurement was taken using the ViewLux™ Imaging System
(PerkinElmer) at an excitation λ = 525 nm (20-nm bandpass) and
an emission λ = 598 nm (25-nm bandpass).
To prevent compound carryover due to the potency of the
known steroid compounds, an acoustic dispensing system from
Labcyte Inc. (Sunnyvale, CA) was used to transfer the com-
pounds in 100% (v/v) DMSO into the 1536-well assay plates as
40-nL aliquots in quadruplicate on the same plate. This system
uses acoustic wave energy applied to the bottom of a multiwell
plate to transfer droplets of fluid from a source plate to the
Compounds were titrated to calculate concentration response
curves (CRCs) in 10 half-log serial dilutions starting at a final con-
centration of 1 µM. All other reagents (buffer, 4 nM GR, 1 nM
Fluoromone™ GS Red) were delivered using the Flexdrop liquid
dispenser (PerkinElmer) in a total volume of 8 µL. The plates were
incubated at room temperature for 2 h prior to imaging on a
Viewlux™. Polarization was calculated as millipolarization units
(mP), and the data were analyzed using a customized HTS data
analysis software package. Inhibition was expressed as 100 × [1 –
(Sample – Blank)/(Total – Blank)],where Sample is the activity of
the test compounds, Blank is the activity of the DMSO control,
and Total is the activity in the presence of 1 µM Dex.
RESULTS AND DISCUSSION
Live-cell imaging of the YFP-GR nuclear
YFP-GR was transiently transfected into COS-7 cells, and the
GR agonist Dex was used to perform proof-of-concept assays to
validate the nuclear translocation method. COS-7 cells were cho-
sen for these experiments because they express low levels of
endogenous corticosteroid receptors, which could contribute to
increased background and slower nuclear translocation rates.16In
addition, 5% charcoal/dextran-treated fetal bovine serum (FBS)
was used to eliminate any endogenous corticosteroid ligand that
may be found in FBS. The COS-7/YFP-GR cells exhibited
strong fluorescence in the cytoplasm that was independent of lig-
and. Real-time GR cytoplasm-to-nucleus translocation following
Dex stimulation in COS-7 cells expressing YFP-GR was first
observed using a Nikon Eclipse TE300 microscope with a cam-
era by recording images at 30-, 60-, 240-, and 480-s intervals
(Fig. 1). Nuclear translocation of GR was rapid, and the kinetics
of GR translocation was similar to that initially reported by Htun
et al.15in murine adenocarcinoma cells (1471.1) and more
recently by Funato et al.25in SY5Y cells. Furthermore, Nishi
et al.18demonstrated that transiently transfecting GFP fusion pro-
teins (GR and MR) into COS-1 cells maintained their functional
abilities as well as expression patterns and molecular masses.
Although there are commercially available cell lines stably
expressing GFP-GR, we have not evaluated the utilization of
these stable cell lines. In general, we believe that stable cell lines
have advantages over transient transfection for HTS and HCS
when constitutive overexpression of target proteins does not
result in cell growth inhibition. However, when overexpression
of target proteins causes inhibition of cellular growth, transient
transfection is a reasonable approach. In the case of NRs, when
multiple NRs are used to evaluate ligand specificity, using the
same parental cell line transiently transfected with the NR of
interest not only is convenient but also provides the identical cel-
lular context for comparison of NR signaling.
Image analysis and quantification of the
Dex-induced YFP-GR translocation
Compound-induced translocation for the GR-YFP fusion
protein was quantified from triplicate wells after image acqui-
sition at ×10 magnification using the Cellomics ArrayScan™.
The nuclear translocation BioApplication was used for analysis
of acquired images and calculation of the nuclear-cytoplasmic
differences in YFP fluorescence.26This system was used to
Agler et al.
1032 www.sbsonline.org Journal of Biomolecular Screening 12(8); 2007
scan multiple fields in each well to obtain images from a mini-
mum of 500 cells per well. Hoechst staining was used to indi-
cate the position of the nucleus. Quantification of the amount
of YFP staining found in the nucleus was determined from the
intensity of yellow fluorescence averaged over the region iden-
tified by the Hoechst stain. To delineate a representative region
of cytoplasm, an annular ring 4 pixels wide was drawn around
the nuclear boundary where GR-YFP is primarily in the cyto-
plasm of unstimulated cells. The difference in fluorescence
intensity between the cytoplasm and the nuclear region was cal-
culated for each cell and then averaged across all cells imaged
per well. EC50 values were calculated from the well-averaged
data under each assay condition.
Figure 2A shows representative images from unstimulated
and 100-nM Dex-stimulated cells. In unstimulated cells where
there is a low level of GR activation, the GR-YFP is primarily
in the cytoplasm. This is hallmarked by the presence of cells
with nuclear regions that show low YFP fluorescence. In Dex-
stimulated cells,YFP-GR is present primarily in the region cor-
responding to the nucleus (Fig. 2A). This agrees well with the
previously reported observations.27
To determine the time course of YFP-GR translocation for
Dex, a potent GR ligand, the EC50was determined after 0.25 to 4
h of treatment (Fig. 2B). Potent effects of Dex were evident by the
0.25-h measurement. These data are in agreement with previous
studies in GR-GFP transiently transfected COS cells where Dex-
induced translocation was evident as early as 5 min after ligand
addition.28The Y maximum for each time point, which converges
and plateaus by 1 h of incubation, would suggest that transloca-
tion was complete, but apparent potency of Dex increases with
increased Dex treatment time. The log (EC50) determined for Dex
was linear after 1 h of Dex incubation (Fig. 2C). Compound incu-
bation was therefore standardized to 1 h for future experiments.
The time course of ligand stimulation that we observed has been
confirmed by extensive studies reported on GR,16,18,27MR,16,18,29
Geldanamycin effects on nuclear translocation
To further validate the nuclear translocation assay, we decided
to compare agonist-induced and antagonist-induced GR translo-
cation, as well as investigate the effect of chaperone proteins
such as heat shock protein on GR translocation. As shown in
Figures 1 and 2 with the transient expression of YFP-GR in
COS-7 cells,GR resides in the cytoplasm in the unliganded state.
There it associates with multiprotein chaperone complexes that
contain heat shock proteins (Hsp90/Hsp70), as well as other pro-
teins such as immunophilins and FK-binding proteins.31,32The
addition of the agonist Dex results in a rapid translocation of
YFP-GR from the cytoplasm to the nucleus in which a hyper-
speckled pattern is initially seen at 0.5 h (Fig. 3A). The observa-
tion of this hyper-speckled pattern is compound concentration
and time dependent, and it is not consistently observed through-
out the population of transfected cells. This could be due to the
varying levels of expression of the YFP-GR in the transiently
transfected COS-7 cell population. In fact, the effect of protein
expression level on NR translocation has been previously
reported. Using fluorescence recovery after photobleaching
(FRAP) experiments, Marcelli et al.17observed a decrease in
mobility with high expression levels of GFP-AR as compared to
High-Content Glucocorticoid Receptor Translocation Assay
Journal of Biomolecular Screening 12(8); 2007www.sbsonline.org1033
protein–GR (YFP-GR). 10–5 M Dex in 1% DMSO (v/v) was added to the cells and images taken at indicated time points poststimulation. The
assay was carried out at room temperature, in real time, on live cells.
Dex induces a rapid glucocorticoid receptor (GR) nuclear translocation in COS-7 cells transiently transfected with yellow fluorescent
low expression levels in response to agonist treatment. No dif-
ference in mobility was observed with antagonist treatment.
Because of the inconsistent observation of this hyper-speckled
pattern in our transiently transfected cell population, we could
not use this parameter to quantify intranuclear conformational
changes. As shown in Figure 3A, the effect on GR translocation
by the antagonist, RU486, was at a slower rate than that for
the agonist Dex, and a diffused YFP pattern was observed.
Agler et al.
1034 www.sbsonline.org Journal of Biomolecular Screening 12(8); 2007
Intensity Difference (YFP)
Hoechst. The composite image represents the fusion of blue (nucleus) and green (GR) fluorescence. (B) Concentration response curves at dif-
ferent time points post-Dex stimulation in transiently transfected COS-7/yellow fluorescent protein–glucocorticoid receptor (YFP-GR). Log of
Dex concentration (nM) is plotted against the difference in nuclear-cytoplasmic intensity of YFP-GR. (C) Time course of Dex potency determi-
nation. The EC50for Dex was determined and plotted at indicated times after compound addition.
(A) Representative images of unstimulated cells (DMSO) and 100-nM Dex-stimulated cells. The nucleus was counterstained with
Translocation was complete around 2 h in the presence of RU486
as compared with Dex, which was complete at approximately
1 h, as determined from CRCs over time (Fig. 2B). Similar to
the response to Dex, the rate of GR translocation in response to
RU486 was also concentration dependent; at higher concentra-
tions, translocation was rapid but slower at lower concentrations.
Because of the variation of protein expression levels and the
dependency on compound concentration, especially with antago-
nists, the selection of a 1-h compound treatment time described
above for compound screening is appropriate.
Using the YFP-GR transient transfection translocation assay,
we investigated whether Hsp90 inhibitors such as geldanamycin
(GA) could distinguish agonist from antagonist activity. To ensure
complete translocation of GR into the nucleus, a 1-h compound
incubation of Dex or RU486 at 60 nM was chosen. GA (1 µg/mL)
was added either 1 h after ligand treatment or 15 min prior to lig-
and treatment. As shown in Figure 3B, GA addition after ligand
incubation reduced the GR translocation by 12% for Dex but
caused a 46% reduction in the presence of RU486,confirming the
slower rate of antagonist-mediated translocation. Pretreatment
with GA for 15 min prior to ligand addition resulted in a 54.5%
and a 91.9% reduction in GR translocation for Dex and RU486,
respectively. Therefore, GA significantly blocked the antagonist-
induced translocation of GR. In agreement with this observation,
previous reported work on the effect of GA on AR translocation
suggests that antagonists induce a different AR conformation and
that antagonist-bound AR remains associated with Hsp90 even
upon translocation.17Heat shock proteins form a heterocomplex
with GR in the cytoplasm and are translocated into the nucleus
upon ligand induction. GA acts by inhibiting the adenosine
triphosphate (ATP)–dependent conformational change of Hsp90
in the cells, which is required for nuclear receptor binding.33
Using immunofluorescence techniques, Czar et al.34have shown
that GA inhibits the GR trafficking to the nucleus by destabilizing
receptor-bound Hsp90 complexes, whereas Tago et al.35reported
that Hsp90 also regulates nuclear retention of GR. Recently
Kakar et al.36identified an export motif in the LBD of GR that
binds heat shock proteins in the cytoplasm. A truncated GR with-
out the export motif showed mostly nuclear localization, suggest-
ing a role for Hsp90 in subcellular localization of the receptor.
However, the key proteins involved in the chaperone complex of
shuttling nuclear receptors across the nuclear membrane still need
to be clearly defined.
Based on the initial results during the development of the
assay, we decided to carry out a small validation screening run
containing nine 384-well compound plates to determine assay
robustness and reproducibility across multiple plates. COS-7
cells were transiently transfected with the YFP-GR plasmid as
described in Materials and Methods. For the validation run,
transfected cells were dispensed into the assay plates using a
Multidrop. Test compounds or DMSO controls (1.0% v/v) were
added and incubated for 1 h at 37 °C, after which the plates
were fixed and washed with a plate washer. Nuclear transloca-
tion was imaged, and the nuclear-cytoplasmic intensity differ-
ence was calculated. As shown in Figure 4, the average signal
(induced by 100 nM Dex) to background (DMSO) ratio was
>7, with an average Z′ > 0.6. Both the signal-to-background
(S/B) ratio and the Z′ were consistent across the plates within
the run and were maintained in experiment runs after valida-
tion. The use of the plate washer after fixation did not affect the
performance of the assay across the multiplate validation run,
as measured by the Z′. The Z′ is a statistical measurement of
the signal window as a fraction of the distance between the
High-Content Glucocorticoid Receptor Translocation Assay
Journal of Biomolecular Screening 12(8); 2007www.sbsonline.org1035
cells transiently transfected with the yellow fluorescent protein–
glucocorticoid receptor (YFP-GR) and stimulated with Dex (60 nM) or
the antagonist RU486 (60 nM) and measured at 0.5 to 4 h poststimula-
tion. (B) Evaluation of the effect of geldanamycin (GA; 1 µg/mL) on
nuclear translocation of GR either in the presence of Dex or RU486 (60
nM). Cells were either incubated first with Dex or RU486 for 1 h fol-
lowed by an additional 1-h incubation with GA (Dex + Geld; RU486 +
Geld), GA for 15 min followed by an additional 1-h incubation with lig-
and (Geld + Dex; Geld + RU486), or ligand alone for 1 h (Dex; RU486).
Results are presented as the mean of control (100 nM Dex) ± SD nor-
malized to DMSO from 4 independent samples.
(A) Higher resolution images (×20) of the nuclei of COS-7
means of the distribution. A perfect Z′ is 1, and an acceptable
Z′ should be greater than 0.5.37 The HCS GR translocation
assay is economical to run, with an estimated cost per well in a
384-well format of $0.04. This includes the cost of transfection
reagents, cell culture, assay plates, and buffers. Throughput
was approximately 10,000 data points per instrument per day.
The rate-limiting step in running an HCS screen is the scanning
speed of the imager. Newer imagers that can scan and acquire
images simultaneously such as the Evotec Opera (www.evotec-
technologies.com; PerkinElmer, Waltham, MA) will extend
HCS assays into HTS capacity.
Correlation between GR binding, translocation, and
Activation of the GR receptor is a multistep process. Prior
to hormone binding, GR resides in the cytoplasm hetero-
complexed with heat shock proteins. Upon ligand binding, GR
translocates to the nucleus. In the nucleus, one of the functions
of GR is to bind to DNA promotor elements, which in turn acti-
vates or represses transcription. Ligand-activated DNA binding
results in transactivation of the promoters containing GR-
responsive elements (GRE), which contributes toward the glu-
cocorticoid side effects associated with steroid therapy.
Development of a safer GR modulator would result in little or
no GR-mediated transactivation of GR-dependent genes, thus
significantly improving the therapeutic ratio compared to cur-
rently used steroids. Discovery of new GR modulators requires
assessment of the entire activation and translocation process. A
panel of known GR ligands was assessed using a FP competition
assay and a cellular transcriptional transactivation assay. The
results obtained were correlated to the GR translocation assay.
The FP competition assay measures binding to a purified
recombinant GR receptor via displacement of a red-shifted labeled
glucocorticoid ligand. The red-shifted ligand was used to reduce
compound interference due to intrinsic compound fluorescence.
The FP assay was run on a fully automated system in a miniatur-
ized 1536-well format to reduce reagent consumption and
decrease costs. The assay was validated by determining 1) stabil-
ity of the reagents on the robotic system, 2) the consistency of the
assay window over multiple runs, and 3) the Z′ factor. The GR FP
assay was robust with a coefficient of variance for the totals
(Dex) of 3.9% and the blank (DMSO) of 10.5% and an average
Z′ of 0.68.
The GR reporter cell line H-Gal hGR NP1-1 was obtained
from H. Gronemeyer (IGBMC, Strasbourg, France). The recep-
tor is a chimera between the GAL4 DNA binding domain and
the human GR ligand binding domain. Upon binding ligand,
the chimera activates the cognate (17-mer) 5-β-globin promoter
in front of luciferase. The assay identifies compounds that pro-
mote binding of the chimera to the GAL4 DNA binding site in
the β-globin promoter. Transcriptional activity with 10 nM Dex
in a final DMSO concentration of 0.1% for 20 h at 37 °C in 5%
CO2resulted in a 9.5-fold S/B ratio in the agonist assay and a
8-fold S/B ratio in the antagonist assay. This assay was robust
with an average Z′ > 0.8 in both assay formats.
Table 1 compares the results of the GR multistep process of
binding, translocation, and transactivation for 10 known
steroids. The FP competition assay uses purified human recom-
binant glucocorticoid receptor, and results for selected ligands
such as estradiol and corticosterone are more potent than pre-
viously described using crude extracts.7 Results from the bind-
ing and transactivation assays support the data obtained from
the YFP-GR translocation assay in COS-7 cells in that the same
rank order of potency was observed for the 10 known steroid
agonists tested. The potency measurements obtained for all
molecules tested showed significant (p < 0.05) correlation
between GR translocation and Dex displacement in the binding
assay. The Spearman rank correlation coefficient between these
2 variables was 0.7. The transactivation assay permits discrim-
ination between GR agonists and antagonists. Potency determi-
nations from the translocation assay were highly correlated
(p < 0.01) with potency measurements from the transactivation
assay in the agonist mode. An EC50for GR translocation was
measured for 2 (progesterone and RU486) of the 3 antagonists
tested. At the 1-h compound incubation time point and concen-
tration range tested, β-estradiol did not give an EC50value.
Rank order of the antagonists was apparent between the 3 assay
formats. However, there was a greater difference in antagonist
potency measured between the 3 assays than for the agonists.
Agler et al.
1036www.sbsonline.orgJournal of Biomolecular Screening 12(8); 2007
described in Materials and Methods, a validation run containing 9 plates
was carried out to determine reproducibility across plates. Data are
graphed as the average signal (S) (from 100 nM dexamethasone, dia-
monds) and blanks (B) from 1% DMSO (v/v) (squares). Averaged Z′
factor (triangles) was >0.6, which is defined as [1 – (3SS + 3SB)/(S – B)],
where S and B are the mean values of signal and background, and SSand
SBare the standard deviations of S and B, respectively. YFP, yellow flu-
Using the glucocorticoid receptor (GR) translocation protocol
EC50values obtained for several agonists in the HCS GR
translocation assay under screening conditions were similar to the
EC50values reported for GR translocation using an enzyme frag-
ment complementation assay.38Data presented here agree with
Pariante et al.39that RU486 and Dex both show similar potent
binding to GR and the ability of RU486 to induce GR transloca-
tion. However, RU486 is less effective at promoting nuclear
translocation, and the potent antagonist effect seen in the transac-
tivation assay is the result of inhibition of nuclear GR effects.
Cross-talk and ligand specificity of NRs
To determine ligand specificity and better understand cross-
talk between nuclear receptors using our assay format, nuclear
translocation assays for AR, ERα, and ERβ were performed
using their known steroid ligands. After transfection in COS-7
cells, the YFP was predominately observed to be cytoplasmic
for GR and AR prior to ligand addition, and upon stimulation,
translocation to the nucleus was observed. For both ER iso-
forms, the YFP was observed in the nucleus upon transfection
regardless of ligand stimulation (data not shown). Previous
studies by Tanaka et al.28confirmed that overexpression of
YFP-MR did not contribute to the nuclear expression of the
protein. Our transient assay system is in agreement with these
findings that without ligand treatment for YFP-GR and YFP-
AR, no nuclear translocation was observed.
Glucocorticoid and mineralocorticoid are 2 classes of the
adrenal corticosteroid hormone that exert different physiologi-
cal responses but act through similar receptors and DNA
response elements. GR has the closest peptide homology to
MR than to PR or AR in all species.40This similarity of recep-
tors is most apparent by looking at common ligands. To deter-
mine the effect of ligand treatment and cross-talk among
steroid hormone receptors, YFP-GR and YFP-AR were tran-
siently transfected into COS-7 cells as discussed in Materials
and Methods. Exposure of the cells for 1 h at 100 nM Dex,
R1881 (the synthetic androgen), and aldosterone (the MR lig-
and) resulted in the nuclear translocation of the YFP fusion pro-
teins that was concentration dependent, as seen in Figure 5. For
AR, the EC50 for R1881 was 0.28 nM, but activity was observed
for GR (>300 nM). The curve for R1881 on the nuclear translo-
cation of GR was not classic and showed no plateau at the high-
est concentration tested of 0.33 µM. Recently, R1881 was
observed to have potent antagonist activity for MR (0.75 nM)
by antagonizing the aldosterone-induced MR transactivation
activity.41Aldosterone has high affinity for GR translocation
(36.3 nM) and no activity on AR translocation.42Nuclear
translocation in response to Dex for GR was consistent with
previous results, but Dex was ineffective at promoting AR
We then used the transient assay format in COS-7 cells to
assess the specificity of a panel of known steroid ligands, both
agonists and antagonists, for GR and AR receptors, as well as
MR agonists and antagonists. As can be seen in Figure 6,
known steroid ligands were incubated for 1 h posttransfection
at a 100-nM final concentration in 1% DMSO (v/v), scanned on
the Cellomics ArrayScan™, and calculated as percentage con-
trol referenced to DMSO treatment. The rank order of ligands
High-Content Glucocorticoid Receptor Translocation Assay
Journal of Biomolecular Screening 12(8); 2007www.sbsonline.org1037
Correlation between GR Binding, Translocation, and Transactivation
12.4 ± 1.1
13.3 ± 1.1
9.4 ± 1.4
69.3 ± 10.5
5.4 ± 0.7
7.3 ± 0.7
3.1 ± 0.5
90.1 ± 0.8
56.9 ± 19.4
34 ± 5
56 ± 9
17 ± 2
169 ± 33
18 ± 2
20 ± 3
15 ± 3
22 ± 4
617 ± 207
17 ± 4
43.71 ± 0.96
118.10 ± 4.41
2.71 ± 0.18
41.13 ± –2.73
1.65 ± 0.03
2.12 ± 0.08
0.50 ± 0.02
666.3 ± 19.5
1651.0 ± 26.8
16.45 ± 4.35
Protocols for each assay are described in Materials and Methods. GR, glucocorticoid receptor.
a. Mean ± SD of 3 experiments. Data were calculated as percentage control normalized to 100 nM Dex. The GR translocation assay was run at 1-h compound incubation posttrans-
fection, fixed, and imaged on the ArrayScan™.
b. Mean ± SD of 20 experiments. Data were calculated as percentage inhibition relative to 1 µM Dex.
c. Kiwas calculated from the average IC50values from the GR fluorescence polarization binding assay.
d. Mean ± SD of 3 experiments except for Dex and RU486, where n = 6. Data were calculated for the agonist assay as percentage control, relative to 100 nM Dex. For the antago-
nist assay, percentage inhibition was calculated relative to 10 nM Dex. The > symbol represents the highest concentration the compound was tested in the assay.
with respect to potency on AR translocation was R1881 >
RU486 > β-estradiol > spironolactone > progesterone > budes-
onide, followed by the antagonists casodex and hydroflutamide
(Fig. 6A). One possibility for the low antagonist activity seen
here and reported by others is that translocation to the nucleus
for antagonists is at a slower rate and was incomplete at the 60-
min compound incubation time used in these experiments and
at the concentration tested.17This is similar to our previous
results looking at the antagonist activity of RU486 on GR
translocation (Fig. 3). In our hands, the EC50for RU486 on AR
translocation, which has been found to have AR agonist activ-
ity, was 6.23 nM as compared to 0.19 nM for R1881 in these
experiments. The relative affinities of RU486 and R1881 in this
assay differ by a factor of 32, which is in agreement with the
transactivation results reported by Song et al.43In addition,
besides RU486, only progesterone showed any significant cross-
reactivity with both AR and GR. Progesterone, along with β-
estradiol, has been reported to be transactivation competent in
an AR-dependent promoter assay.29
Next we monitored GR selectivity on nuclear translocation for
the panel of known steroids at 100 nM. As shown in Figure 6B,
GR translocation under the same conditions showed equal affin-
ity for Dex, prednisolone, corticosterone, budesonide, and RU486
and lower affinity for aldosterone, R1881, progesterone, spirono-
lactone, and β-estradiol. The higher affinity ligands showed sim-
ilar efficacy at 100 nM and translocated >90% of the GR receptor.
The biological response of a ligand, especially that of antagonists
and their effect on translocation, may be due to different regula-
tory factors. These include immunophilins and immunophilin-
like phosphate PP5, as well as chaperone proteins such as Hsp70,
Hsp90, Hsp40, p23, and NLS.31,44Thus, the continuous shuttling
of proteins across the nuclear membrane is a complicated process.
Our assay system detected ligands of both high and low affinity,
which demonstrates that this assay could be used in a screening
format for the identification of novel nuclear receptor modulators.
Engineered chimeras for constitutive nuclear receptors such
as the THR,VDR, ER, and orphan receptors can be constructed
using GR, which is predominantly cytoplasmic prior to ligand
stimulation. A protocol for designing these chimeras has been
Agler et al.
1038 www.sbsonline.org Journal of Biomolecular Screening 12(8); 2007
(YFP-AR) or yellow fluorescent protein–glucocorticoid receptor (YFP-GR) construct and treated with various ligands (GR ligand, Dex; AR lig-
and, R1881; mineralocorticoid receptor [MR] ligand, aldosterone) from 0.33 µM down in a 10-point titration in triplicate for 1 h at 37 °C, fixed
and imaged. Data were calculated according to the Materials and Methods section. The Z′ for AR and MR were 0.72 and 0.64, respectively.
Cross-talk among nuclear receptors. COS-7 cells were transiently transfected with either yellow fluorescent protein–androgen receptor
developed by Martinez and Hager.45Using the methods we
have developed and optimized for transient expression of GR
and applied to AR in COS-7 cells, the possibility exists of iden-
tifying modulators, both agonists and antagonists of orphan
nuclear receptors. The complexity of steroid receptor function
and protein interactions, as well as the tissue-specific recruit-
ment of coactivators and corepressors, poses a challenge to
drug discovery in identifying nonsteroidal receptor modulators.
Successful identification of leads for drug discovery requires a
panel of assays that trace the mechanism of action of a target path-
way.46,47We conclude here that, for a small set of known ligands,
ligand affinity as indicated by the FP assay correlates well with
potency in the HCS translocation assay. Previous HTS experience
with biochemical screens suggests that identification of binders
does not always predict cellular activity, as an isolated protein
screen will identify moieties that are not selective to the target or
not cell permeable.47A screening effort using an HCS transloca-
tion assay would be expected to identify compounds that not only
bind to GR but also promote translocation. However, leads iden-
tified in this assay would contain a mixture of agonists and antag-
onists, which could be triaged with the reporter transactivation
assay run in both agonist- and antagonist-seeking mode. However,
reporter assays may not be receptor dependent (receptor-
independent activation), and cytotoxic compounds can show up
as false positives in antagonist assays or as false negatives in ago-
nist assays. HCS assays provide other information unavailable
from FP and reporter assays, such as subcellular localization
where protein-protein interactions occur. Within this assay triage
strategy, the HCS translocation assay provides an inexpen-
sive assay format compared to purchasing commercially avail-
able reagents for FP and reporter assays. Ideally,depending on the
goal of a screening campaign, multiple assay formats should be
used during the hit identification and evaluation process.
Although stable cell lines are an advantage for screening,
many times the overexpressed receptor can be toxic to the cell.
We have demonstrated that the transient approach is both repro-
ducible and robust, and it can be run in a 384-well plate format.
Scaling up transient transfections for HTS,although complex,has
been performed for reporter assays.48,49Previous studies using
GFP-GR have shown that when GFP-GR is expressed in living
cells, it is competent for normal transactivation of GR-responsive
promoter activity.15,16,50The unliganded GFP-GR localizes in the
cytoplasm and translocates to the nucleus in a hormone-dependent
manner similar to that of the native GR receptor, which is in
agreement with our observation presented in this article. All these
support the biological relevance of the overexpressed system most
researchers use in the field.50,51Our assay system has the potential
to be modified using viral-based methods for more efficient
expression in target-specific cell lines.52It may also be possible to
multiplex different color variants of fluorescent proteins tagged to
different NRs to investigate heterodimer formation and the
dynamics of nuclear-cytoplasmic shuttling using our system. The
limitation of running an HCS translocation assay in an HTS mode
is not the cell production or the data analysis but the HCS imager
Using high-content imaging technologies, the nongenomic
function of the GR receptor—such as effects on G-protein-coupled
receptors, mitogen-activated protein kinases, Wnt signaling, and
transforming growth factor β family, to name a few—can be mul-
tiplexed with GR translocation to begin to understand the signal-
ing pathways that are induced by ligand activation.53-56Newer
high-resolution, confocal, high-throughput imaging platforms
High-Content Glucocorticoid Receptor Translocation Assay
Journal of Biomolecular Screening 12(8); 2007www.sbsonline.org1039
R1881 (200 nM)
% Control (DMSO)
R1881 (200 nM)
% Control (DMSO)
fluorescent protein–androgen receptor (YFP-AR) or (B) yellow fluorescent protein–glucocorticoid receptor (YFP-GR) constructs and treated with
a final concentration of 100 nM of a known steroid in 1% DMSO (v/v), except for R-1881, which was at 200 nM for 1 h at 37 °C. Results are
presented as the mean of the cytoplasm to nuclear difference ± SD normalized to DMSO from 3 independent samples.
Nuclear translocation of known steroid agonists and antagonists. COS-7 cells were transiently transfected with either the (A) yellow
with advanced image analysis algorithms will also enable the
development of nuclear receptor signaling panels, selectivity
assays, membrane signaling events, and HTS assays for drug dis-
covery in an automated high-throughput mode.57
We gratefully acknowledge contributions from David Connors
and Suki Jayachandra for cell culture support, Tim Spicer for
automation support, David Stock for discussions on the statistical
analysis of data, and the compound management group at BMS
for providing compound samples and plates. We thank Martyn
Banks, director of Lead Discovery, Profiling and Compound
Management, for continuing support for high-content screening
and reviewing this article.
1. Mangelsdorf DJ, Thummel C, Beato M, Herlich P, Schutz G, Umesonn K,
et al: The nuclear receptor superfamily: the second decade. Cell 1995;
Gronemeyer H, Gustafsson J, Laudet V: Principles for modulation of the
nuclear receptor superfamily. Nat Rev 2004;3:950-964.
Perissi V, Rosenfeld M: Controlling nuclear receptors: the circular logic
of cofactor cycles. Nat Rev 2005;6:542-554.
Chen T, Xie W, Agler M, Banks M: Coactivators in assay design for
nuclear hormone receptor drug discovery. Assay Drug Dev Tech 2003;1:
Lin S, Bock CL, Gardner DB, Webster JC, Favata MF, Trzasko JM, et al:
A high-throughput fluorescent polarization assay for nuclear receptor
binding utilizing crude receptor extract. Anal Biochem 2002;300:15-21.
Parker GJ, Law TL, Lenoch FJ, Bolger RE: Development of high through-
put screening assays using fluorescence polarization: nuclear receptor-
ligand-binding and kinase/phophatase assays. J Biomol Screen2000;5:77-88.
Glickman JF, Wu X, Mercuri R, Illy C, Bowen B, He Y, et al: A compar-
ison of ALPHAScreen, TR-FRET and TRF as assay methods for FXR
Nuclear Receptors. J Biomol Screen 2002;7:3-10.
Zhou, G, Cummings R, Hermes J, Moller D: Use of homogeneous time-
resolved fluorescence energy transfer in the measurement of nuclear recep-
tor activation. Methods 2001;25:54-61.
Grover GS, Turner BA, Parker CN, Meier J, Lala DS, Lee PH:
Multiplexing nuclear receptors for agonist identification in a cell-based
reporter gene high-throughput screen. J Biomol Screen 2003;8:239-246.
Hermkens PHH, Kamp S, Lusher S,Veeneman GH: Non-steroidal steroid
receptor modulators. IDrugs 2006;9:488-494.
Mohler ML, He Y, Wu A, Hong S, Miller DD: Dissociated non-steroidal
glucocorticoids: tuning out untowards effects. Exp Opin Ther Patents
Kain SR: Green fluorescent protein (GFP): applications in cell-based
assays for drug discovery. DDT 1999;7:304-312.
McNally JG, Muller WG, Walker D, Wolford R, Hager GL: The gluocor-
ticoid receptor: rapid exchange with regularity sites in living cells.
Walker D, Htun H, Hager GL: Using inducible vectors to study intracel-
lular trafficking of GFP-tagged steroid/nuclear receptors in living cells.
15. Htun H, Barsony J, Renyi I, Gould DL, Hager CL: Visualization of glu-
cocorticoid receptor translocation and intracellular organization in living
cells with a green fluorescent protein chimera. Proc Natl Acad Sci USA
Nishi M, Ogawa H, Ito T, Matsuda K, Kawata M: Dynamic changes in
subcellular localization of mineralocorticoid receptors in living cells: in
comparison with glucocorticoid receptor using dual-color labeling with
green fluorescent spectral variants. Mol Endocrinol 2001;15:1077-1092.
Marcelli M, Stenoien DL, Szafran AT, Simeoni S, Agoulnik IU, Weigel
NL, et al: Quantifying effects of ligands on androgen receptor nuclear
translocation, intranuclear dynamics and solubility. J Cell Biochem
Nishi M, Tanaka M, Matsuda K, Sunaguchi M, Kawata M: Visualization
of glucocorticoid receptor and mineralocorticoid receptor interactions in
living cells with GFP-based fluorescence resonance energy transfer.
J Neurosci 2004;24:4918-4927.
Htun H, Holth LT,Walker D, Davie JR, Hager GL: Direct visualization of
the human estrogen receptor alpha reveals a role for ligand in the nuclear
distribution of the receptor. Mol Biol Cell 1999;10:471-486.
Taylor DL, Giuliano K: Multiplexed high content screening assays create
a systems cell biology approach to drug discovery. Drug Discov Today
Lang P, Yeow K, Nichols A, Scheer A: Cellular imaging in drug discov-
ery. Nat Rev/Drug Discov 2006;5:343-356.
Huizenga NATM, De Herder WW, Koper JW, de Lange P, v Lely DAJ,
Brinkmann AO, et al: Decreased ligand affinity rather than glucocorticoid
receptor down-regulation in patients with endogenous Cushing’s syn-
drome. J Endocrinol 2000;142:472-476.
Heydorn A, Lundholt BK, Praestegaard M, Pagliaro L: Protein transloca-
tion assays: key tools for accessing new biological information with high-
throughput microscopy. Methods Enzymol 2006;414:513-530.
Olechno J, Ellson R, Browning B, Stearns R, Mutz M, Travis M, et al:
Acoustic auditing as a real-time, non-invasive quality control process for
both source and assay plates. Assay Drug Develop Tech 2005;3:425-437.
Funato H, Kobayashi A, Watanabe Y: Differential effects of antidepres-
sants on dexamethasone-induced nuclear translocation and expression of
glucocorticoid receptor. Brain Res 2006;1117:125-134.
Ding GJF, Fischer P, Boltz RC, Schmidt JA, Colaianne JJ, Gough A, et al:
Characterization and quantitation of NF-κB nuclear translocation induced by
interleukin-1 and tumor necrosis factor-α. J Biol Chem 1998;273:28897-
Nishi M,Takenaka N,Morita N,Ito T,Ozawa H,Kawata M:Real-time imag-
ing of glucocorticoid receptor dynamics in living neurons and glial cells in
comparison with non-neural cells. Eur J Neurosci 1999;11:1927-1936.
Tanaka M, Nishi M, Morimoto M, Sugimoto T, Kawata M: Imaging
analysis of mineralocorticoid receptor and importins in single living cells
by using GFP color variants. Cell Tissue Res 2005;320:447-453.
Fejes-Toth G, Pearce D, Naray-Fejes-Toth A: Subcellular localization of
mineralocorticoid receptors in living cells: effects of receptor agonists
and antagonists. Proc Natl Acad Sci USA 1998;95:2973-2978.
Tyagi RK, Lavrovsky Y, Ahn SC, Song CS, Chatterjee B, Roy AK:
Dynamics of intracellular movement and nucleocytoplasmic recycling of
the ligand-activated androgen receptor in living cells. Mol Endocrinol
Pratt WB, Morishima Y, Murphy M, Harrell M: Chaperoning of gluco-
corticoid receptors. HEP 2006;172:111-138.
Cheung J, Smith DF: Molecular chaperone interactions with steroid
receptors: an update. Mol Endocrinol 2000;14:939-949.
Agler et al.
1040www.sbsonline.orgJournal of Biomolecular Screening 12(8); 2007
33.Roe SM, Prodromou C, O’Brien R, Ladbury JE, Piper PW, Pearl LH:
Structural basis for inhibition of the Hsp90 molecular chaperone by the
antitumor antibiotics radcicol and geldanamycin. J Med Chem 1999;
Czar MJ, Galigniana MD, Silverstein AM, Pratt WB: Geldanamycin, a
heat shock protein 90-binding benzoquinone ansamycin, inhibits steroid-
dependent translocation of the glucocorticoid receptor from the cyto-
plasm to the nucleus. Biochem 1997;36:7776-7785.
Tago K, Tsukahara F, Naruse M, Yoshioka Y, Takano K: Regulation of
nuclear retention of glucocorticoid receptor by nuclear HSP90. Mol Cell
Kakar M, Kanwal C, Davis JR, Li H, Lim CS: Geldanamycin, an
inhibitor of Hsp90, blocks cytoplasmic retention of progesterone recep-
tors and glucocorticoid receptors via their respective ligand binding
domains. AAPS J 2006;8:E718-E728.
Zhang JH, Chung TD, Oldenburg KR: A simple statistical parameter for
use in evaluation and validation of high throughput screening assays.
J Biomol Screen 1999;4:67-73.
Fung P, Peng K, Kobel P, Dotimas H, Kauffman L, Olson K, et al: A
homogeneous cell-based assay to measure nuclear translocation using β-
galactosidase enzyme fragment complementation. Assay Drug Dev Tech
Pariante CM, Pearce BD, Pisell TL, Miller AH: The steroid receptor
antagonist RU40555 and RU486 activate glucocorticoid receptor translo-
cation and are not excreted by the steroid hormones transporter in L929
cells. J Endocrinol 2001;169:309-320.
Hu X, Funder JW: The evolution of mineralocorticoid receptors. Mol
Takeda AN, Pinon GM, Bens M, Fagart J, Rafestin-Oblin ME,
Vandewalle A: The synthetic androgen methyltrienolone (R1881) acts as
a potent antagonist of the mineralocorticoid receptor. Mol Pharm 2007;
Buckingham J: Glucocorticoids: exemplars of multi-tasking. Br J Pharm
Song LN, Coghlan M, Gelmann EP:Antiandrogen effects of mifepristone
on coactivatior and corepressor interactions with the androgen receptor.
Mol Endocrinol 2004;18:70-85.
Pilipuk GP,Vinson GP, Sanchez CG, Galigniana MD: Evidence for NL1-
independent nuclear translocation of the mineralocorticoid receptor.
Martinez ED, Hager GL: Development of assays for nuclear receptor
modulators using fluorescently tagged proteins. Methods Enzymol 2006;
46.Garrett MD, Walton MI, McDonald E, Judson I, Workman P: The con-
temporary drug development process: advances and challenges in pre-
clinical and clinical development. Prog Cell Cycle Res 2003;5:145-158.
Moore K, Rees S: Cell-based versus isolated target screening: how lucky
do you feel? J Biomol Screen 2001;6:69-74.
Chen J, Lake MR, Sabet RS, Niforatos W, Pratt SD, Cassar SC, et al: Utility
of large-scale transiently transfected cells for cell-based high throughput
screens to identify transient receptor potential channel A1 antagonists.
J Biomol Screen 2007;12:61-69.
Zhu Z, Puglish J, Connors D, Stewart J, Herbst J, Marino A, et al: Use of
cryopreserved transiently transfected cells in high-throughput pregnane X
receptor transactivation assay. J Biomol Screen 2007;12:248-254.
Kawata M: Subcellular steroid/nuclear receptor dynamics. Arch Histol
Hager GL, Lim CS, Elbi C, Baumann CT: Trafficking of nuclear recep-
tors in living cells. J Steroid Biochem Mol Biol 2000;74:249-254.
Berno V, Hinojos CA,Amazit L, Szafran AT, Mancini MA: High-resolution,
high-throughput microscopy analyses of nuclear receptor coregulator func-
tion. Methods 2006;414:188-210.
Kino T, Tiulpakov A, Ichijo T, Chheng L, Kozasa T, Chrousos G: G pro-
tein B interacts with the glucocorticoid receptor and suppresses it tran-
scriptional activity in the nucleus. J Cell Biol 2005;166:885-896.
Ichijo T, Voutetakis A, Cotrim AP, Bhattachryya N, Fujii M, Chrousos GP,
et al: The smad6-histone deacetylase 3 complex silences the transcriptional
activity of the glucocorticoid receptor. J Biol Chem 2005;280:42067-42077.
Takayama S, Rogatsky I, Schwarcz LE, Darimont BD: The glucocorti-
coid receptor represses cyclin D1 by targeting the TCF-B-catenin com-
plex. J Biol Chem 2006;281:17856-17863.
Yemelyanov A, Czwornog J, Chebotaev D, Karseladze A, Kulevitch E,
Yang X: Tumor suppressor activity of glucocorticoid receptor in the
prostate. Oncogene 2007;26:1885-1896.
Haney SA, LaPan P, Pan J, Zhang J: High-content screening moves to the
front of the line. DDT 2006;11:889-894.
Address correspondence to:
Michele Agler, Ph.D.
Bristol Myers Squibb
5 Research Parkway
Wallingford, CT 06492
High-Content Glucocorticoid Receptor Translocation Assay
Journal of Biomolecular Screening 12(8); 2007 www.sbsonline.org 1041