A Rhodopsin Transport Assay by High-Content Imaging Analysis

Summary

Here, we described a high-content imaging method to quantify the transport of rhodopsin mutants associated with retinitis pigmentosa. A multiple-wavelength scoring analysis was used to quantify rhodopsin protein on the cell surface or in the whole cell.

Abstract

Rhodopsin misfolding mutations lead to rod photoreceptor death that is manifested as autosomal dominant retinitis pigmentosa (RP), a progressive blinding disease that lacks effective treatment. We hypothesize that the cytotoxicity of the misfolded rhodopsin mutant can be alleviated by pharmacologically stabilizing the mutant rhodopsin protein. The P23H mutation, among the other Class II rhodopsin mutations, encodes a structurally unstable rhodopsin mutant protein that is accumulated in the endoplasmic reticulum (ER), whereas the wild type rhodopsin is transported to the plasma membrane in mammalian cells. We previously performed a luminescence-based high-throughput screen (HTS) and identified a group of pharmacological chaperones that rescued the transport of the P23H rhodopsin from ER to the plasma membrane. Here, using an immunostaining method followed by a high-content imaging analysis, we quantified the mutant rhodopsin protein amount in the whole cell and on the plasma membrane. This method is informative and effective to identify true hits from false positives following HTS. Additionally, the high-content image analysis enabled us to quantify multiple parameters from a single experiment to evaluate the pharmacological properties of each compound. Using this assay, we analyzed the effect of 11 different compounds towards six RP associated rhodopsin mutants, obtaining a 2-D pharmacological profile for a quantitative and qualitative understanding about the structural stability of these rhodopsin mutants and efficacy of different compounds towards these mutants.

Introduction

Protein misfolding is involved in muscular dystrophy, neural degenerations, as well as blinding diseases, including retinitis pigmentosa (RP)¹. RP is an inherited and progressive retinal degeneration associated with mutations in over 60 genes affecting the function and homeostasis of rod photoreceptors or the retinal pigmented epitheliums (RPEs)^2,3. No effective treatment is currently available for RP. Rhodopsin mutations account for about 25-30% of autosomal dominant (ad) RP cases. Among the more than 150 rhodopsin mutations⁴ (Human Gene Mutation Database, http:/www.hgmd.cf.ac.uk/), the Class II mutations cause the structural instability of the rhodopsin protein that contributes to the rod photoreceptor death and vision loss^5,6,7,8. The P23H is the most frequent rhodopsin mutation in North America, which is also a typical example of the Class II rhodopsin mutations^9,10. Due to its inherent structural instability, the misfolded rhodopsin is accumulated in the endoplasmic reticulum (ER) in mammalian cells, whereas the wild type rhodopsin is located on the plasma membrane⁵. The misfolded rhodopsin P23H mutant exhibits dominant negative cytotoxicity that is not due to haploinsufficiency, but is related to the activation of ER associated protein degradation pathway and the interrupted rod outer segment organization. To alleviate rod photoreceptor cell stress, one strategy is to stabilize the native folding of the mutant rhodopsin using a pharmacological chaperone.

To achieve this goal, we performed a cell-based high-throughput screen (HTSs)^11,12,13 using a β-galactosidase fragment complementation assay to quantify the P23H rhodopsin mutant transported on the plasma membrane. The robust and simple protocol of this HTS assay enabled us to explore the activities of about 79,000 small molecules for each screen. However, because this HTS assay reads luminescence signals, false positives including the β-gal inhibitors, colored or cytotoxic compounds are included in the hit list waiting to be identified by a secondary assay.

The traditional immunostaining and fluorescence imaging methods have been used for years to study the rhodopsin transport in mammalian cells^5,14,15,16. However, these conventional methods cannot be used to quantify pharmacological effects of more than 10 compounds towards rhodopsin transport because a reliable imaging analysis requires a large number of images taken under a highly consistent condition, which is not amendable by the conventional imaging methods. Here, we developed an immunostaining based high-content imaging protocol as a secondary assay to quantify the cell surface transport of misfolded rhodopsin mutants^11,13,17. To label rhodopsin on the plasma membrane, we skipped the step of cell membrane permeabilization and immunostained the rhodopsin mutants by a monoclonal (B6-30) anti-rhodopsin recognizing the N-terminal epitope of rhodopsin at the extracellular side of the cell membrane¹⁸. To visualize the mutant rhodopsin in the whole cell, we fused rhodopsin with the Venus fluorescence protein. By the quantification of the fluorescence intensities in different fluorescence channels, we are able to obtain multiple parameters from one single experiment including the total rhodopsin intensity in the whole cell, on the cell surface, and the ratio of rhodopsin fluorescence on the cell surface to that in the whole cell. Applying this method to stable cells expressing a total of six misfolded rhodopsin mutants, we can generate a pharmacological profile of multiple small molecule chaperones towards these mutants. In this protocol, all cells are immunostained in a 384-well plate and imaged using an automated imaging system under a highly consistent imaging condition. An image analysis is performed to each well, containing images of more than 600 cells to reduce variation due to the heterogeneity of the cells with varying cell shape and protein expression level. The workflow of this protocol is summarized in Figure 1. The advantage of this method is that we obtain high-resolution images as well as multi-parameter quantifications from the image-based analysis. In general, this protocol can be modified and applied to quantify the transport of any misfolded membrane protein of interest.

Protocol

NOTE: The rhodopsin transport assay.

1. Preparation and Culture of Cells

1. Revive cryo-preserved U2OS stable cells expressing the wild type (WT) or mutant mouse rhodopsin-Venus fusion proteins. Thaw the cells at 37 °C until only small ice crystals are left in the vial.
   NOTE: The U2OS cells are used in this protocol because there is no photoreceptor cell line available for in vitro studies and the pre-ciliary biosynthesis of rhodopsin is regulated by similar molecular mechanisms in mammalian cells. Additionally, the U2OS cells attach tightly to the bottom of the plate and have large cell bodies, so they are ideal for the rhodopsin transport assay. Seven U2OS stable cells expressing the WT, T4R, P23H, P53R, C110Y, D190N and P267L rhodopsin mutants are used here as an example to show the quality and reliability of the assay. Stable cells expressing other rhodopsin mutants or other membrane proteins can be used, depending on the goal of the assay. Stable cells expressing human rhodopsin can be used in this assay if available. The mouse rhodopsin was used here because mouse and human rhodopsin proteins share 95% homology, and the compounds identified by this assay will be tested in vivo using a mouse adRP model carrying the rhodopsin P23H mutation⁸. The stable cells were generated as described previously¹⁹. The WT and mutant rhodopsin transcripts were quantified by q-PCR, and the clones of stable cells expressing similar levels of WT and mutant rhodopsin transcripts were selected for this assay. The expression of rhodopsin proteins was confirmed by immunoblot and immunostaining. The passage of cells used in this assay should be lower than 20.
2. Transfer each line of cells into a 15 mL conical tube containing 10 mL of growth medium (high glucose Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS) and 5 µg/mL Plasmocin).
3. Centrifuge the tube at 200 x g for 5 min. Discard the supernatant and resuspend the cell pellet with 5 mL of cell growth medium for each cell line. Transfer each line of cell suspension into a 60 mm tissue culture dish and incubate at 37 °C with 5% CO₂.
4. Subculture the cells when the tissue culture dish reaches 90% confluence as described in reference¹².

2. Seeding Cells at 5,000 cells per Well

NOTE: Perform the following procedures in a tissue culture hood.

1. Treat the plate with poly-lysine.
   1. One day before seeding the cells, treat a black-wall and clear bottom 384-well plate with 20 µL/well of poly-lysine solution for 30 min.
   2. Aspirate the liquid and allow all wells to dry for 1 h. Put back the plate lid and store the plate at 4 °C overnight for cell seeding.
2. Prepare the cell suspension.
   1. Culture each cell line in growth medium in a 100 mm tissue culture dish until 90% confluence.
   2. Wash each dish with phosphate buffered saline (PBS) and detach the cells with 1 mL of 0.05% Trypsin at 37 °C.
   3. Resuspend each line of cells in 10 mL of assay medium (high-glucose DMEM with 10% FBS, 100 units/mL penicillin and 100 μg/mL streptomycin).
   4. Count the cells and dilute each line of cells to 1.25 x 10⁵ cells/mL with the assay medium.
3. Seed the cells.
   1. Use an electronic multichannel pipette or an automated reagent dispenser to dispense 40 µL/well of the cell suspension to three columns per cell line and fill columns 1-21 of the 384-well plate (Figure 2).
   2. Add 40 µL/well of U2OS (P23H-Rhodopsin-Venus) to columns 22-23 and 40 µL/well of U2OS (Rhodopsin-Venus) to column 24, as controls.
   3. Centrifuge the plate at 300 x g for 30 s to bring down all cells to the bottom of the 384-well plate.
      NOTE: Avoid physical shock to the plate after cell seeding. Otherwise, cells will be crushed to one corner of each well, and the plate will not be suitable for high-content imaging.
   4. Incubate the 384-well plate at 37 °C with 5% CO₂ for 3 h for the cells to attach to the bottom of the plate.

3. Treating the Cells with Compounds

1. Generate a plate map with treatment conditions as shown in Figure 2.
2. Prepare 300 µL of 5x working solutions of up to 15 compounds in a 96-well plate.
   1. Dilute cps 1 to 15 with the assay medium to five times of their final concentrations in wells A1 to G2 of the 96-well plate (Figure 2A). Add 300 µL of assay medium in well H2.
      NOTE: The final concentration of each compound is used at the most effective concentration for rescuing the transport of the P23H rhodopsin by the HTS^12,13.
   2. Add 100 µL per well of 2% dimethyl sulfoxide (DMSO) and 25 µM 9-cis-retinal that are diluted in the assay medium to columns 11 and 12, respectively. Add 9-cis-retinal in dim light.
3. Adding 10 µL per well of 5x working solutions to the 384-well plate cultured with the cells (Figure 2B).
   1. Use an electronic multi-channel pipette to add compounds 1, 3, 5, 7, 9, 11, 13, and 15 to rows A, C, E, G, I, K, M, and O from columns 1 to 21 (Figure 2). Add compounds 2, 4, 6, 8, 10, 12, 14 and M to rows B, D, F, H, J, L, N, and P from columns 1 to 21.
   2. Add 10 µL per well of 2% DMSO to columns 22 and 24. Add 10 µL per well of 25 µM 9-cis-retinal to column 23.
      NOTE: DMSO is used as a vehicle control because all the compounds are initially dissolved in DMSO as stocks. Column 22 containing DMSO treated cells expressing the P23H rhodopsin is used as the 0% control. 9-Cis-retinal treated cells expressing the P23H rhodopsin is used as the 100% control.
4. Cover the 384-well plate with aluminum foil and incubate the plate at 37 °C for 24 h.

4. Immunostaining without Membrane Permeabilization to Stain Rhodopsin Protein on the Cell Surface.

NOTE: Avoid any detergent in the entire immunostaining process to keep cell membrane intact.

1. Prepare 4% paraformaldehyde (PFA) by diluting the 16% PFA with PBS in a 1:4 volume ratio in a chemical fume hood. Transfer the 4% PFA in a reagent reservoir.
2. Take out the 384-well plate in a dark room with dim red light. Use an 8-channel aspirator connected to a vacuum collection bottle to gently aspirate the medium. Use an electronic multichannel pipette to add 20 µL per well of freshly prepared 4% PFA to the entire 384-well plate and incubate for 20 min at room temperature. To avoid cell detachment, always point the tips of the aspirator to the same side of each well during aspirations. Do not touch the middle bottom area of each well. When taking images, select fields to avoid the region touched by the aspirator. For incubations longer than 10 min, cover the 384-well plate with its lid to avoid evaporation.
   NOTE: Cells are fixed in a dark room to avoid photobleaching of the regenerated isorhodopsin that will affect the result of the 100% control. After fixation, the 384-well plate can be taken out under normal light.
3. Use an 8-channel aspirator to aspirate the PFA in each well and use an electronic multi-channel pipette to add 50 μL per well of PBS. Repeat two more times to perform three washes with PBS.
   CAUTION: The waste liquid containing PFA are collected in a capped bottle and is to be disposed of as a hazardous chemical waste after the experiment.
4. Block the cells by adding 20 µL per well of 5% goat serum to the entire 384-well plate and incubate at room temperature for 30 min.
5. Aspirate the 5% goat serum and add 15 µL per well of 20 µg mL^-1 B6-30 anti-rhodopsin antibody in 1% goat serum to rows A to O. Add 15 µL per well of 1% goat serum to row P for the secondary antibody-only control group.
6. Incubate the 384-well plate at room temperature for 90 min or at 4 °C overnight. Cover the 384-well plate with aluminum foil to avoid photobleaching of the fluorophores.
7. Wash the plate 3 times with 50 µL per well of PBS.
8. Aspirate PBS and add 15 µL per well of 5 µg mL^-1 Cy3-conjugated goat anti-mouse IgG antibody. Cover the 384-well plate with aluminum foil and incubate at room temperature for 1 h or at 4 °C overnight.
9. Wash the plate 3 times with 50 µL per well of PBS. Add 50 µL per well of PBS containing 1 µg mL^-1 Hoechst 33342 to stain nuclei at room temperature for 15 min.
10. Seal the 384-well plate with a transparent film before imaging. Cover the 384-well plate with aluminum foil and store at 4 °C for up to a week if images are taken immediately.

5. Imaging_.

NOTE: This high-content imaging procedure is adapted to the imaging system listed in the Table of Material. Procedures can be different if using other high-content imaging systems.

1. Remove the lid and put the 384-well plate into the high-content imager with A1 positioned on the top left corner of the plate.
2. Open the image acquisition software to set up parameters for image acquisition.
   1. Open the Plate Acquisition Setup window and create new setting or load an existing setup file.
   2. Select the 20x objective and set up pixel binning as 2 so the calibrated pixel size is 0.80 x 0.80 µm. Set the scan lines as 2,000 and the image size (W x H) as 1,000 x 1,000 pixels (800.00 x 800.00 µm) per site.
   3. Select the plate type to be imaged. Use the information provided by the manufacturer of the 384-well plate to fill in the plate dimensions.
   4. Select the wells to be imaged for the 384-well plate.
   5. Select four sites to be imaged per well. Avoid the side of the well touched by the aspiration tips.
   6. Select excitation lasers as 405, 488 and 561 nm. Select emission filters for DAPI, FITC and Texas Red channels. Optimize the laser power and gains of each channel to ensure the brightness of images taken from the positive control wells are less than the saturation threshold.
   7. Select well-to-well focus for autofocus. Select the first well acquired as the initial well for finding samples. Set site autofocus to all sites.
   8. Select four averages per line for each channel. Optimize the Z-offset value for each channel.
   9. Test 2-3 wells at the diagonal corners of the 384 well plate to make sure the images are on focus for all tested wells, images from all sites per well have cells with more than 40% confluence, and fluorescence intensities of all channels are about half saturated in the 100% control wells.
   10. Save the image acquisition method.
3. Run the entire plate. Watch the imager until it finishes capturing images from the first column of the plate to double-check the image quality before leaving the imager. Take the 384-well plate out and store at 4 °C for future use.
   NOTE: Depending on the number of sites selected per well and the channels taken, it takes 40 min to 3 h to finish imaging one 384-well plate.

6. Image Analysis.

1. Pull out the image data using a high-content image analysis software. Select one of the 100% control wells (column 23) to set up parameters.
2. Select the Multi-Wavelength Cell Scoring as the analysis method and start configuring settings.
   1. Define the nuclei using images from the DAPI channel. Preview to make sure the defined nuclei shapes in the selected well fits well with the nuclei images.
   2. Define the shape of cell in the FITC channel where rhodopsin-Venus is imaged.
   3. Define the rhodopsin cell surface stain areas in the Texas Red channel.
   4. Test the current algorithm in 5 wells to determine if the settings are optimized. Save the settings and close the Configuration window.
3. Run all the wells with the optimized analysis method.
4. Export Object Number as intact cell number. Export the Average Intensity of the FITC channel as Rhodopsin-Venus Intensity (Rhodopsin-Venus INT). Export the Average Intensity of the Texas Red channel as rhodopsin intensity on the cell surface (Rhodopsin INT on the cell surface).
5. Open the exported data in a spreadsheet software. Divide the rhodopsin intensity on the cell surface by the rhodopsin-Venus INT as MEM-to-total ratio. Calculate the Z’-factors for each parameter to evaluate the assay quality. Z′=1−3X(STD_{100% control+}STD_{0% control})/|Mean_{100% control}−Mean_{0% control}|.
   NOTE: A Z’-factor higher than 0 indicates a moderate assay sufficient for a high-content screen, and a Z’ factor between 0.5 and 1 suggests an outstanding assay required for a high-throughput screen^20,21.
6. Use the spreadsheet software to generate a two-color heat map for each parameter. Arrange the name of cell lines on the X-axis and the compounds on the Y-axis.

Results

We characterized the rhodopsin transport with three parameters: the rhodopsin-Venus intensity in the whole cell (Rhodopsin-Venus INT), the immunostaining intensity of rhodopsin on the plasma membrane (Rhodopsin INT on the cell surface), and the ratio of rhodopsin stain on the cell surface to rhodopsin-Venus intensity in the whole cell (MEM-Total Ratio). A representative result of the rhodopsin transport assay is shown in Figure 3 and Figu...

Discussion

Here, we showed a high-content imaging assay used for characterizing hits identified from a HTS. The only automation involved in these protocols is the high-content imager. The immunostaining and fluorescence imaging of rhodopsin have been used commonly to characterize the localization of rhodopsin^5,14,15,16. However, the quantification of images taken by the traditional imaging methods is limi...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
U2OS (rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
U2OS (T4R-rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
U2OS (P23H-rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
U2OS (P53R-rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
U2OS (C110Y-rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
U2OS (D190N-rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
U2OS (P267L-rhodopsin-Venus) cells	NA	NA	Stable cells generated from U2OS cells
DMEM high glucose	Genesee Scientific	25-500	With L-Glutamine, sodium pyruvate
Fetal bovine serum (FBS)	Gibco	16140071	Heat inactivated
Plasmocin	InvivoGen	ant-mpt	Mycoplasma elimination reagent
Penicillin-Streptomycin (100X)	Gibco	15140122	100X concentrated antibiotic solutions to prevent bacteria contamination of cell cultures
Trypsin-EDTA	Genesee Scientific	25-510	0.25%, 1mM EDTA in HBSS without calcium and magnesium
Poly-L-lysine solution	Sigma-Aldrich	P4707-50ML	Mol wt 70,000-150,000, 0.01%, sterile-filtered, BioReagent, suitable for cell culture
CellCarrier-384 Ultra Microplates	PerkinElmer	6057300	384-well tissue culutre-treated microplates with black well walls and an optically -clear cyclic olefin bottom for imaging cells in high content analysis
Sterile 96-well plate	Eppendorf	30730119	Tissue culture treated with lid flat bottom, sterile, free of detectable pyrogens, Rnase, DNase and DNA. Non-cytotoxic
Phosphate Buffered Sailine (PBS)	Invitrogen	AM9625	10 x PBS Buffer, pH 7.4
DMSO	Sigma-Aldrich	D4540	>99.5%, cell culture tested
9-cis-retinal	Sigma-Aldrich	R5754
Compounds tested	Selleckchem/Life Chemicals/Custom synthesized	NA	Compounds were purchased from different vendors or custom synthesized
B6-30 anti-rhodopsin antibody	Novus	NBP2-25160	Gift from Dr. Krzysztof Palczewski
Cy3-conjugated goat anti-mouse secondary antibody	Jackson ImmunoResearch Laboratories, Inc	115-165-146
16% paraformaldehyde	Thermo Fisher Scientific	28908	Methanol-free
10% Normal Goat Serum	Thermo Fisher Scientific	50062Z	Blocking buffer
Hoechst 33342, Trihydroch	Invitrogen	H3570	Nuclear staining solution
High-content imager	Molecular Devices	ImageXpress	ImageXpress® Micro Confocal High-Content Imaging System
MetaXpress high-content image acquisition and analysis software	Molecular Devices	MetaXpress	High-content image acquisition and analysis software
Multichannel pipette (0.5-10 µL)	Rainin	17013802	Manual 8-channel pipette, 0.5-10 µL
Multichannel pipette (0.5-10c	Rainin	17013805	Manual 8-channel pipette, 20-200 µL
Electronic multichannel pipette (10-200 μL)	Thermo Scientific	14-3879-56BT	Electronic multichanenel pipette for 96- and 384-well microplate pipetting tasks
50ml Reagent Reservoir	Genesee Scientific	28-125	Reagent reservior for multichannel pippte dispensing
8-Channel aspirator	ABC Scientific	EV503	8-Channel stainless steel adaptor for aspirating liquids from 96- or 384-well plates
Excel spreadsheet software	Microsoft	Excel2016	The spreadsheet software for data analysis and heatmap generation
Origin2018 scientific data analysis and graphing software	OriginLab	Origin2018	The data analysis software for generating the dose response curves