High-content screening of GPCR activation with MetaXpress,™ AcuityXpress™ and the Transfluor® assay system
METAXPRESS APPLICATION NOTE
By Susan Catalano, Ph.D., Acumen Pharmaceuticals, Sylvia de Bruin, Joe Bosworth, Ph.D.,
Neal Gliksman, Ph.D., Kimberly Best, Ph.D. and Paula Rickert, Ph.D., Molecular Devices Corporation,
1311 Orleans Drive, Sunnyvale, CA 94089.
INTRODUCTION
G-protein coupled receptors (GPCRs are the largest class of pharmaceutical
targets and, as a result, assays for detecting GPCR receptor agonists
and antagonists play a major role in screening operations. Recently Xsira
Pharmaceuticals introduced the Transfluor assay, a novel high-content
assay that monitors GPCR activation by measuring GPCR desensitization
and recycling. The Transfluor assay is now exclusively available from
Molecular Devices.
The Transfluor Assay, validated on over 90 different GPCRs, uses a common
pathway independent of downstream G-protein signaling which is shared
by virtually all GPCRs, representing all classes of GPCRs and G proteins.
The assay uses image analysis to quantitate internalization of GFP-tagged
beta-arrestin that associates with the receptor of interest following
activation.1 (See Figure 1.)
HIGH-CONTENT SCREENING OF GPCR ACTIVATION WITH METAXPRESS,
ACUITYXPRESS AND THE TRANSFLUOR ASSAY SYSTEM
Unstimulated cells display diffuse cytoplasmic beta-arrestin-GFP fluorescence.
When an over-expressed GPCR is activated, beta-arrestin targets the receptor
for internalization, resulting in the appearance of small fluorescent
clathrin-coated pits. If the receptor has a high affinity for beta-arrestin,
the cells subsequently form fluorescent endocytic vesicles, which are
larger and brighter than pits. (See Figure 2.) The Transfluor Assay has
been widely adopted as both a primary and secondary screen in many pharmaceutical
drug discovery operations as a functional measurement of physiological
events downstream of GPCR activation.
MetaXpress imaging software fully automates image
acquisition and analysis of the Transfluor assay with all of Molecular
Devices’ high-content screening systems, including the ImageXpress®
5000A and Discovery-1™ instruments. Optional application modules
provide assay-specific turnkey analysis of common high-content assays.
The Transfluor module, described in this application note, was specifically
designed and optimized for analyzing images from Transfluor assays.
MetaXpress is integrated with the enterprise-level MDCStore™ database for managing
images, cell-by-cell segmentation results and annotations. The AcuityXpress
cellular informatics platform is also seamlessly integrated with the MDCStore
database, providing powerful data visualization and statistical tools
for curve fitting, evaluating assay quality, and identifying hits, false
positives, outliers and trends. Used together, MetaXpress, AcuityXpress
and the Transfluor assay allow you to enhance your screening potential
and rapidly discover GPCR modulators.
Practice of Transfluor assays requires a license available only through
Molecular Devices. Contact your local sales representative for details
on licensing and available Transfluor assay reagents.
CUSTOM CELL-BY-CELL ANALYSIS
MetaXpress image acquisition and analysis software optionally includes
assay-specific application modules, designed for easy, interactive optimization
of segmentation parameters. The simplified interface allows simple configuration
and testing of analysis settings for your particular assay.(See Figure
3, Panel A.)
The MetaXpress Transfluor Module uses the proprietary
Adaptive Background Correction™ system to distinguish pits, vesicles
and nuclei from their local surroundings so they can be selectively measured
and counted, even in the presence of uneven staining or protein expression
levels. This can be difficult to do using ordinary thresholding methods
because the intensity of the pits can be marginally above background,
and the GFP-tagged beta-arrestin expression levels can vary from cell
to cell.
With the Transfluor Module, designed for cell-by-cell analysis
of Transfluor data, pits and vesicles can be detected and distinguished
simultaneously. The optional nuclear detection provides accurate counting
of cells for data normalization. Three user-adjustable parameters are
utilized for the image analysis. The “Approximate min width”
and “Approximate max width” parameters are modified as needed
for the expected size of the pits, vesicles or nuclei. The “Intensity
above local background” parameter is adjusted for optimal discrimination
of each pit or vesicle from the local background. This significantly improves
the ability to identify and accurately quantitate actual pit and vesicle
formation even if the intensity varies between cells in the image.
Multiple measurements can be logged to the database, including the number
of pits or vesicles, total area covered by pits or vesicles, average and
integrated intensity of the granules, and nuclear area and intensity.
Results are provided on a cell-by-cell basis as well as summarized by
image and site. Highlighting a row of the cellular results table immediately
highlights the corresponding cell and overlay in the image. (See Figure
3, Panels B and C.)
HIGH-CONTENT SCREENING OF GPCR ACTIVATION WITH METAXPRESS, ACUITYXPRESS
AND THE TRANSFLUOR ASSAY SYSTEM
MATERIALS
Practice of Transfluor Assays requires a license from Molecular
Devices. Transfluor reagents are available to Transfluor licensees from
Molecular Devices. Contact Molecular Devices for details.
Note: All solutions should be freshly prepared immediately before each experiment.
Beta-arrestin-GFP
and Beta-2-adrenergic receptor (ß2AR)-expressing U2OS cells (Molecular
Devices)
Corning Costar 384-well microplates (Cat. #3712)
U2OS
media: MEM (Gibco Cat. #11095) with 10% heat-inactivated FCS (Gibco Cat.
#10082), 10 µg/mL gentamicin (Gibco Cat. #15710), 10 mM HEPES (Gibco
Cat. #11344), 0.4 mg/mL Zeocin (Gibco Cat. #R250) and 0.4 mg/mL G418 (Gibco
Cat. #10131)
Ascorbic acid media: 5.7 mM solution of ascorbic acid
(sodium L-ascorbate, SIGMA Cat. #A7631) in serum- and antibiotic-free
media (MEM with 10 mM HEPES)
Isoproterenol stock solution: 175 µM
stock solution of (-)-isoproterenol (+)-bitartrate salt (SIGMA Cat. #I2760)
in ascorbic acid media
Propranolol stock solution: 125 µM solution
of (s)-(-)-propranolol hydrochloride (SIGMA Cat. #P8688) in serum- and
antibiotic-free media (MEM with 10 mM HEPES)
DPBS (SIGMA Cat. #D8537)
4% methanol-free formaldehyde: 16% methanol-free formaldehyde (Polysciences
Cat. #18814) diluted 1:4 in DPBS
DAPI, dilactate (Molecular Probes D-3571)
METHODS
Agonist dose-response assay
Step1. Plate cells at 4,500 cells/well in 25 µL of U2OS media.
Incubate overnight at 37°C, 5% CO2.
Step 2. Perform 10 five-fold serial dilutions of the isoproterenol
stock solution in the ascorbic acid media to form an 11-point series of
3.5X isoproterenol dose-response solutions. Use ascorbic acid media alone
as a negative control. (See Table 1.)
| Table 1. Concentrations of the Isoproterenol
3.5X Solutions Required for the Agonist Dose-Response Assay |
| 3.5X Concentration |
Final Concentration |
| 175 µM |
50 µM |
| 35 µM |
10 µM |
| 7 µM |
2 µM |
| 1.4 µM |
400 nM |
| 280 nM |
80 nM |
| 56 nM |
16 nM |
| 11.2 nM |
3.2 nM |
| 2.2 nM |
640 pM |
| 448 pM |
128 pM |
| 89.6 pM |
25.6 pM |
| 17.9 pM |
5.1 pM |
| 0 µM |
0 µM |
Step 3. Add 10 µL of the appropriate 3.5X isoproterenol dose-response
solution or negative control to each well.
Step 4. Incubate at 37°C, 5% CO2 for 40–45
minutes (vesicle formers) or 30 minutes (pit formers).
Antagonist dose-inhibition assay
Step 1. Plate cells at 4,500 cells/well in 23 µL of U2OS
media. Incubate overnight at 37°C, 5% CO2.
Step 2. Prepare a 175 nM 3.5X isoproterenol stimulation solution
by diluting the 175 µM isoproterenol stock 1:1000 in the ascorbic
acid media.
Step 3. Perform ten five-fold serial dilutions of the propranolol
stock solution in MEM with Hepes to form a series of 12.5X propranolol
dose-inhibition solutions. Use MEM/Hepes alone as a negative control.
(See Table 2.)
| Table 2. Concentrations of the Propanolol
12.5X Solutions Required for the Antagonist Dose-Inhibition Assay |
| 12.5X Concentration |
Final Concentration |
| 125 µM |
10 µM |
| 25 µM |
2µM |
| 5µM |
400µM |
| 1µM |
80 nM |
| 200 nM |
16 nM |
| 40 nM |
3.2 nM |
| 8 nM |
640 pM |
| 1.6 nM |
128 pM |
| 320 pM |
25.6 pM |
| 64 pM |
5.1 pM |
| 12.8 pM |
1.0 pM |
| 0 µM |
0 µM |
Step 4. Add 2 µL of the appropriate 12.5X propranolol dose-inhibition
solution or negative control to each well.
Step 5. Incubate at 37°C, 5% CO2 for 30 minutes.
Step 6. Add 10 µL of the 175 nM isoproterenol stimulation
solution or ascorbic acid media (unstimulated control) to each well.
Step 7. Incubate at 37°C, 5% CO2 for 40–45
minutes (vesicle formers) or 30 minutes (pit formers).
Sample fixing and staining
Step 1. Stop assay by adding 35 µL 4% methanol-free formaldehyde
to each well. Cover and incubate for 45 minutes at room temperature.
Step 2. Wash each well with 50 µL DPBS.
Step 3. Add 8 µL of 0.05 mg/mL DAPI to each well. Incubate
at room temperature for 15 minutes.
Step 4. Rinse wells twice with 75 µL DPBS, leaving in 75
µL DPBS.
Step 5. Cover plate with a plate seal and store at 4°C, protected
from light, until image acquisition.
Image acquisition
Step 1. Using MetaXpress, acquire unbinned images on an ImageXpress
5000A or Discovery-1 imaging system using DAPI and FITC filter sets
and a 20x Plan Fluor or Plan Fluor ELWD objective. The results described
here were obtained from imaging two sites per well.
Image analysis
Step 1. In MetaXpress, retrieve your images of interest from
the database and analyze them with the Transfluor assay module. Select
the appropriate source images, minimum and maximum widths, and intensity
above local background for both the pit/vesicle staining and the nuclear
staining.
Step 2. In AcuityXpress, annotate your plates with the appropriate
compounds and concentrations.
Step 3. Create a dataset containing your plates of interest and
perform curve fitting to obtain EC50 and IC50
values for each compound.
ASSAY RESULT
SU2OS cells expressing beta-arrestin-GFP and either wild-type ß2AR
(pit formers) or enhanced ß2AR (vesicle formers) were treated
with either isoproterenol (agonist) or propranolol (antagonist). The
enhanced ß2AR has a modified C-terminal tail to increase its affinity
for beta-arrestin, resulting in fluorescent vesicle formation.
Images were acquired on an ImageXpress 5000A system as described above,
and analyzed with the Transfluor module. Similar results were obtained
on a Discovery-1 system (data not shown). The accuracy of the module
is illustrated in Figure 4 (pit phenotype) and Figure 5 (vesicle phenotype).
The images show that the algorithm properly identifies pits and vesicles
in either phenotype for cells stimulated with varying concentrations
of the agonist isoproterenol. Unstimulated cells display minimal pits
or vesicles.
HIGH-CONTENT SCREENING OF GPCR ACTIVATION WITH METAXPRESS, ACUITYXPRESS
AND THE TRANSFLUOR ASSAY SYSTEM
DOSE-RESPONSE AND DOSE-INHIBITION ANALYSIS
In both pit- and vesicle-forming cell lines, the dose-response of the
agonist isoproterenol and the dose-inhibition of the antagonist propranolol
were measured with the MetaXpress Transfluor module. Each dose was run
across six replicate wells. Figure 6 shows a plate heat map generated
by AcuityXpress, demonstrating the dose-dependent response of the assay.
For EC50 and IC50 calculations, Hill curve fitting was performed on
pit or vesicle area per cell, averaged across all replicate wells on
the plate, plotted against compound concentration. AcuityXpress automatically
generates curve fits for your entire dataset, encompassing multiple
compounds and multiple plates, if appropriate. (See Figure 7.) Table
3 provides a summary of the agonist EC50 and antagonist IC50 for both
cell lines from two independent experiments. These values correspond
well to reports in the literature2 of isoproterenol stimulation of the
beta-2-adrenergic receptor with an EC50 of 1–10 nM and propranolol
antagonism with an IC50 = 0.5–5 nM.
Table 3. Measured EC50 and IC50 Values
for Beta-2-Adrenergic
Receptor Stimulation and Inhibition |
| |
Isoproterenol EC50 (µM) |
Propranolol IC50 (µM) |
| Pit Phenotype |
| Experiment 1 |
4.7 |
4.3 |
| Experiment 2 |
8.1 |
4.3 |
| Vesicle Phenotype |
| Experiment 1 |
3.4 |
3.7 |
| Experiment 2 |
6.3 |
3.7 |
ASSAY STATISTICAL SUMMARY
The Z' factors (see Table 4) were calculated for both experiments.3
The Z' factor, a measure of assay quality, reflects both the data variation
and the signal dynamic range. The Z' factors measured for this experiment
are > 0.6, easily within the desirable range for a cell-based imaging
assay.
Table 4. Z' Statistical Summary for
Agonist Stimulation
and Antagonist Inhibition |
| |
Agonist Dose-Response Z' |
Agonist Dose-Inhibition Z' |
| Pit Phenotype |
| Experiment 1 |
0.8 |
0.9 |
| Experiment 2 |
0.9 |
0.8 |
| Vesicle Phenotype |
| Experiment 1 |
0.8 |
0.8 |
| Experiment 2 |
0.7 |
0.9 |
DISCUSSION
The MetaXpress Transfluor module successfully identifies cells and measures
pits and vesicles formed by beta-arrestin-GFP translocation to activated
GPCRs in the Transfluor assay. Independent experiments with the beta-2-adrenergic
receptor, both wild-type and enhanced, produce consistent EC50 and IC50
values with high Z' factors. These results demonstrate that the Transfluor
assay as analyzed by MetaXpress is a highly robust and reproducible
assay for GPCR activation.
The rapid acquisition, analysis and data visualization capabilities
of MetaXpress and AcuityXpress combined with the universality of the
Transfluor Assay offer a uniquely powerful approach for screening GPCR
modulators.
REFERENCES
1. Oakley, R.H., Hudson, C.C., Cruickshank, R.D., Meyers, D.M., Payne,
R.E. Jr., Rhem, S.M., Loomis, C.R. The cellular distribution of fluorescently
labeled arrestins provides a robust, sensitive, and universal assay
for screening G-protein-coupled receptors.
Assay Drug Devel Tech 2003;1:21–30.
2. Oakley, R.H., Laporte, S.A., Holt, J.A., Barak, L.S., Caron, M.G.
Association of beta-arrestin with G-protein-coupled receptors during
clathrin-mediated endocytosis dictates the profile of receptor resensitization.
J Biol Chem 1999;274:32248–32257.
3. 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;4:67–73.
|