key: cord-0743870-0uf489iw authors: Cecon, Erika; Dam, Julie; Jockers, Ralf title: Detection of SARS-CoV-2 spike protein binding to ACE2 in living cells by TR-FRET date: 2021-11-22 journal: STAR Protoc DOI: 10.1016/j.xpro.2021.101024 sha: 35ec549cac89b3a931311b2d71fdc8e8bb45e4a2 doc_id: 743870 cord_uid: 0uf489iw The SARS-CoV-2 coronavirus infects human cells through the interaction of the viral envelope Spike protein (IPR044366) with the human Angiotensin-Converting Enzyme 2 (ACE2), expressed at the surface of target cells. Here, we describe a detailed protocol to measure the binding of the receptor binding domain (RBD) of Spike to ACE2 by time-resolved Fluorescence Resonance Energy Transfer (TR-FRET). The assay detects the Spike/ACE2 interaction in physiologically relevant cellular contexts and is suitable for high-throughput investigation of interfering small-molecule compounds and antibodies. Before you begin The protocol below describes the specific steps of the Spike/ACE2 binding assay in Human Embryonic Kidney 293 Cells (HEK293T) cells based on the TR-FRET technique to screen for inhibitory compounds. The assay is compatible with any cell type provided that the cells can be transfected to express the SNAP-tagged human ACE2 fusion protein. The TR-FRET assay is based on the energy transfer between an energy donor and energy acceptor that occurs if they are in close proximity to each other (less than 10 nm) (Mathis, 1995; Bazin, Trinquet and Mathis, 2002; Degorce et al., 2009) . Distinctively from classical FRET, the Time-Resolved (TR) aspect of the TR-FRET uses special donor fluorophores from the lanthanide family, displaying extended lifetime emission properties, which allows the introduction of a time delay between donor excitation and FRET signal measurement. This time-resolved feature results in measurement of the FRET signal at a time window of lowest background, following the decay of non-specific transient fluorescence signals, providing thus improved signal-to-noise ratio. This protocol uses the receptor binding domain (RBD) of the SARS-CoV-2 Spike protein labeled with the d2 fluorophore (energy acceptor, emission at 665 nm) and the human ACE2 labeled with the terbiumcryptate fluorophore (energy donor; excitation at 320 nm) to monitor the proximity between RBD and ACE2 through the resulting TR-FRET signal. The protocol describes a high-throughput(HT)-compatible assay format, but it can also be performed in a low-throughput format. Note: The advantage of performing the TR-FRET assay in the low-throughput / adherent cell format is that it allows to treat the cells for long incubation time with compounds (i.e. 16-24h incubation). A list of all reagents, buffer and equipment for this protocol are described in the Key Resources Table and Materials and Equipment sections. Make sure all materials required for the experiments are available and/or prepared in advance. Note: The experimental design should be established in advance in order to calculate the precise volume needed of each reagent. Please refer to Table 2 Timing: 30 min 3. In order to calculate the amounts of reagents required for the experiment, the experimental plate design needs to be set in advance. The data points should be performed in triplicates and it should always include the following groups: i. background ("BKG" -absence of d2labelled RBD); ii. total binding ("TB" -presence of d2-labelled RBD); iii. non-specific binding ("NSB" -presence of RBD-d2 AND excess of non-labelled RBD); iv. test (total binding of RBD-d2 in the presence of a compound to be tested for its effect on the RBD/ACE2 interaction). Figure 1 shows an example of plate template. Alternatives: We usually buy sterile cell culture medium bottles, but the medium can also be prepared from powder. In this case, after preparing the medium it should be filtered using a 0.22 μm filter prior to use. Final concentration Amount NaCl (5 M) 150 mM 1.5 mL ddH2O n/a 48.5 mL Total n/a 50 ML Note: Solution can be stored at stored at 4°C for up to 1 month. Reagent Final concentration Amount TagLite labelling medium 5x 1x 1 mL ddH2O n/a 4 mL Total n/a 5 mL Note: Prepare fresh TagLite buffer before each experiment. The final volume to be prepared will depend on the number of data points and should be calculated in advance. Final concentration Amount Lumi4-tb substrate stock (0.1 mM) 100 nM 5 µL TagLite labelling medium 1X n/a 5 mL Total n/a 5 mL Note: Prepare fresh SNAP substrate solution before each experiment. The final volume to be prepared will depend on the number of data points and should be calculated in advance. Final concentration Amount Spike(RBD) stock (37.6 µM) 3.5 µM 4.7 µL TagLite labelling medium 1X n/a 45.3 µL Total n/a 50 µL Note: Prepare fresh working solution before each experiment. The final volume to be prepared will depend on the number of data points and should be calculated in advance. Final concentration Amount Spike(RBD)-d2 stock (11 µM) 35 nM 3.2 µL TagLite labelling medium 1X n/a 996.8 µL Total n/a 1 mL J o u r n a l P r e -p r o o f Note: Prepare fresh working solution before each experiment. The final volume to be prepared will depend on the number of data points and should be calculated in advance. Step-by-step method details HEK293T cells seeding -Day 1 Timing: 30 min To obtain HEK293T cells expressing SNAP-ACE2, cells are transiently transfected using transfection reagents. Alternative: stable cell lines expressing SNAP-ACE2 can be used. 1. From the stock culture of HEK293T cells, remove the cell culture medium and add 5 mL of sterile PBS to wash the cells. 2. Remove the PBS and add 1 mL trypsin/EDTA solution. 3. Incubate at 37°C in the cell culture incubator for 2-5 min. 4. Add 9 mL of complete DMEM medium and resuspend the cells. Note: Pipette the medium with cells up-and-down using a 10mL pipette to better detach and dissociate the cells. 5. Collect the cell suspension in a 15 mL conical tube. 6. Take an aliquot of 10 µL of the cell suspension in a microcentrifuge tube to count the number of cells. 7. Spin down the 15 mL tube containing the cells at 180 × g for 5 min. 8. Meanwhile, mix the 10 μL cell suspension aliquot with 10 μL trypan blue solution and transfer the 20 µL into a cell counting slide. 9. Count live cells manually with a hemocytometer or with an automated cell counter to determine the cell density. a. Exclude from the counting trypan blue-stained dead cells. b. Calculate the required volume of cell suspension to obtain 2x10 6 cells / mL. 10. Following the centrifugation, aspirate the supernatant. 11. Resuspend the cell pellet in the proper volume of complete DMEM medium calculated to obtain 2x10 6 cells / mL. 12. Seed 1 mL of cell suspension solution in a new 10 cm cell culture dish containing at least 9 mL of complete DMEM medium. 13. Incubate cells for 18h-24h at 37°C in a CO2 (5%) incubator to allow cells to recover and attach to the plate. Note: Transfection can also be done in cells seeded in smaller dishes (i.e. 6-well plates) if the experimental design does not require a high number of cells (low number of data points). Cell transfection -Day 2 CRITICAL: Cells should not exceed 70% confluency before transfecting to assure an optimal transfection rate. We found that an incubation time of 48h (post-transfection) is optimal for the expression and plasma membrane translocation of membrane proteins. Note: If performing the TR-FRET assay in the low-throughput format or with adherent cells, the transfected cells can be plated 24h post-transfection (Day 3) into white 96-well plates at a density of 0.5x10 5 cells/well. Cells are then incubated for another 24h before starting the TR-FRET assay (Day 4). Fluorescent labelling of SNAP-ACE2 -Day 4 Timing: 1.5 hours This step consists in the covalent labelling of the N-terminal SNAP-tagged human ACE2 expressed at the cell surface of transfected cells with the TR-FRET-compatible terbium (Tb) fluorophore. The Lumi4-Tb SNAP substrate is a suicide substrate of the SNAP enzyme labelled with a terbium fluorophore (Keppler et al., 2003) . 28. Verify successful SNAP labeling a. Add 10 µL of the cell suspension per well in two wells (duplicate) of a with 384-well plate. b. Read the plate in a plate reader using the filter configuration compatible with the excitation and emission spectra of the Terbium fluorophore. Settings in the TECAN F500 plate reader to read the Tb donor signal are shown in Table 1 . Note: The Tb fluorophore emits at multiple wavelengths and it is, therefore, suitable as a donor fluorophore also for acceptors emitting at the green spectrum (see Cisbio websitehttps://fr.cisbio.eu/content/htrf-technology/). In both cases, the Tb (donor) signal is measured at 620 nm. CRITICAL: ACE2 labeling is essential for the TR-FRET assay. Only proceed to the next step if successful labeling of ACE2 is obtained. Use mock-transfected cells to access the background of Tb signal. From our experience, 5 to 10-fold signal-to-background ratio indicates successful SNAP labelling. An example of Tb donor signal in SNAP-ACE2-expressing cells and in mock-transfected cells is shown in Figure 2 . Troubleshooting 2 Timing: 2-3 hours 29. Distribute 10 µL per well of the Tb-labeled SNAP-ACE2 cells into a white 384-well plate. It is recommended to use an electronic pipette to increase speed and accuracy. Note: Labelled cells can be diluted by a factor of 3 to 5 in TagLite buffer if high Tb signal is obtained. Optimal cell density and/or Tb signal will depend on the cell surface expression of SNAP-ACE2 and can be tested in advance to define the best conditions to obtain a low background signal and high signalto-noise ratio. For example, in the experiment shown in Figure 2 , cells could be used at a dilution of 3.5 times, which would result in a Tb signal still 20-fold higher than background (i.e. mock-transfected cells). Optional: Remaining labeled cells can be frozen and stored (-20°C) to be used later, within 2 weeks. Cell suspension solution should be divided into aliquots to avoid repeated freeze-thaw cycles. 30. Add 2 µL of test compound at the desired concentration to the "test" wells. Note: Working solution should be 7x concentrated. In the "NSB" wells, add 2 µL of 3.5 µM non-labeled RBD (final concentration of 0.5 µM). 32. In the "BKG" and "TB" wells, add 2 µL of 1x TagLite buffer. a. Seal the plate with a plate sealing film and incubate at 20-22°C for 15 minutes. 33. Add 2 µL of Tag-lite buffer to "BKG" wells. b. Incubate the plate at 20-22°C for 2h. Note: An overview of the order of TR-FRET reagents to be added to the different wells is shown in Table 2 . 35. Read the plate in a TR-FRET-compatible plate reader. Settings in the TECAN F500 plate reader to read the TR-FRET signal are shown in Table 3 . CRITICAL: Perform at least triplicates for each data point (see Figure 1 for an example of plate template). Alternative: other TR-FRET-compatible plate readers, such as EnVision (PerkinElmer) can also be used. Optional: The TR-FRET signal being stable, several reads can be performed at different time points, once every hour, for example, until equilibrium is reached. In this case, it is important to fix the "Gain" setting of the donor channel (620 nm) to a fixed value in order to be able to compare the different readings. Alternative: Kinetics parameters can also be determined with slight changes in the protocol, as previously shown (Cecon et al., 2021) . To determine association kinetics parameters, read the TR-FRET signal continuously during 2h, starting immediately after adding RBD-d2. To determine dissociation kinetics parameters, allow binding of RBD-d2 to reach equilibrium (2h incubation) and only then add excess of non-labelled RBD to displace bound RBD-d2 and read the TR-FRET signal continuously during 2h, starting immediately after adding non-labelled RBD. Timing: 30 min 36. Once the reading is finished, export the results file as .csv file. 37. Calculate the TR-FRET signal, which is the ratio of the acceptor (RBD-d2, measured at 665 nm) over the donor (Tb-ACE2, measured at 620 nm) signals for each well. For convenience, the ratio is multiplied by 10000. An example of a TR-FRET data sheet result is shown in Table 4 . The specific signal for each well can be obtained by subtracting the NSB signal from the TR-FRET signal. TR-FRET signal = Signal 665 nm x 10 4 Signal 620 nm Specific Signal (for each well) = TR-FRET signal -NSB Note: By performing the ratio, the TR-FRET is normalized to the donor signal, counteracting thus any potential differences in the number of cells per well. Some plate readers allow to integrate the calculation equation, providing thus the final TR-FRET signal data. Note: the raw data value are arbitrary units and, therefore, it is not possible to perform direct comparison of raw data between different plate readers. If such comparison is envisioned, the foldchange of the specific signal should be used. 38. Use a spreadsheet or statistics software package to calculate the arithmetic mean of the triplicates. Note: Check that the Tb signal is homogeneous between wells to identify potential outliers. 39. Normalize the data to 0 -100 % scale by setting the mean of the "TB" group as 100% and the mean of the "NSP" group as 0 % using a statistics software package like GraphPad Prism. Note: As the raw values (fluorescence signal) are given in arbitrary units, the normalization to a percentage scale allows to pool together independent repetitions of the experiment. 40. Use a statistics software package like GraphPad Prism to calculate standard deviations and statistical significance and to plot the data in a graph. A successful TR-FRET experiment should show a clear difference between "TB" and "NSB" signals, with "NSB" signal being similar or very close to the "BKG" signal (Troubleshooting 4; Troubleshooting 5). The amplitude of the signal depends on the distance between the two fluorophores. In the TR-FRET mode for ACE2/RBD, a high "TB" TR-FRET signal with a low BKG is observed, as it can be easily competed with excess of non-labelled RBD, conferring a high amplitude of the specific signal ("TB"-"NSB"). The ratio between the TR-FRET signals for "TB" and "NSB" ("TB"/"NSB") can be expected to range between 5 (at Kd concentration of RBD-d2) to 30-fold (at saturating concentrations of RBD-d2). Of note, the "TB" TR-FRET signal is proportional to the concentration of RBD-d2, up to a saturation limit that we have detected to occur at approximately 100 nM of RBD-d2 (Cecon et al., 2021) . In the case that the objective of the assay is to screen the efficacy of candidate compounds to interfere with the RBD-ACE2 interaction, the concentration of RBD-d2 used should be around its affinity (Kd), i.e. low nanomolar range (for example, 5nM), as described in this protocol. Positive interfering compounds are the ones giving a specific TR-FRET signal significantly lower than the specific signal of RBD-d2 alone (i.e. without "test" compound). A representative bar graph of an expected result of an experiment including all the groups mentioned in this protocol ("BKG", "TB", "NSB" and "Test") is shown in Figure 3 . The protocol described here comprises an assay to detect the binding of RBD to ACE2. The choice of using the RBD domain of the spike protein was due to the reduced size of this peptide, compared to the full protein, which favors the proximity between the d2-fluorophore and the Tb-labelled ACE2. The assay has the advantage of providing a very high specific signal, being thus suitable to the identification of compounds interfering with this interaction. However, one cannot exclude that other regions of the Spike protein might be relevant for cell interaction and cell entry of SARS-CoV-2 virus. The use of d2-labeled full-length S1-S2 Spike protein for the TR-FRET assay is possible, as reported (Cecon et al., 2021) , but the amplitude of the signal is significantly lower than that observed for the RBD-d2, reaching 2-fold the "NSB" values. Such difference is probably due to the bigger size of the S1-S2 spike protein, which might increase the distance between the fluorophores. In addition, the use of recombinant proteins (RBD or S1-S2) may represent only a simplified model of SARS-CoV-2 interaction with cellular systems, as it has been described that the Spike protein is highly glycosylated and acts as a trimer (Xu et al., 2021) . Therefore, inhibitory compounds identified using the TR-FRET assay described here need to be validated using additional techniques such as cellular or in vivo infection models with SARS-CoV-2 virus. Finally, the TR-FRET signal is correlated to the distance between the acceptor and donor fluorophores. Therefore, a change in TR-FRET signal could reflect either a conformational change within the protein complex or an increase in association or dissociation of RBD and ACE2. Compounds of interest identified by this method can have an impact on those processes and would need to be characterized further for their exact properties. Problem 1: No detectable SNAP-ACE2 expression (Cell transfection -Day 2 step 14). Transfection efficiency can be indirectly accessed at step 28, when verifying the SNAP labelling. If low transfection rate is suspected to occur, transfection efficiency can be tested using different transfection reagents and verified by direct methods such as Western Blot (using antibodies against ACE2 or against the Flag-tag) or fluorescence imaging (using microscopy-compatible SNAP fluorophores for SNAP labeling). No detectable SNAP-ACE2 labeling with Tb fluorophore (Fluorescent labelling of SNAP-ACE2 -Day 4 step 28). Potential solution: Verify that the cells were efficiently transfected with SNAP-ACE2 construct by using techniques like western blot or immunofluorescence imaging. Make sure cells are not overconfluent at the moment of labelling and verify that they do not detach during the washing steps after labelling. Problem 3: Too much variation between the wells for the Tb fluorescent signal (Data Analysis step 38). Check the precision of the pipette used for the step of distribution of the cells in the 384-well plate. Make sure that the same pool of Lumi4-Tb-labelled cells is used for all wells in the experiment. Forty-eight hours post-transfection, cells were labelled with Lumi4-Tb substrate as described, and 10 µL of the cell suspension was used to measure the resulting Tb donor signal in the Tecan F500 plate reader. Gain was set to "optimal" and corresponded to a value of 92 in this experiment. Non-specific binding ("NSB") Test compounds ("Test") Cells 10 µL 10 µL 10 µL 10 µL TagLite buffer 1x 2 µL 2 µL --Test compounds (7x) ---2 µL Non-labeled RBD (7x) --2 µL -Seal the plate and incubate at 20-22°C for 15 minutes TagLite buffer 1x 2 µL ---RBD-d2 (7x) -2 µL 2 µL 2 µL Seal the plate and incubate at 20-22°C for 2 hours Read the plate in a TR-FRET-compatible plate reader Table 3 . TR-FRET reading settings using TECAN F500 plate reader. Reading settings in Tecan F500 plate reader Lag Time 150 µs * When reading TR-FRET multiple times (every hour or in a kinetics mode, for example) the Gain of the donor signal should be set to a fixed value corresponding to the value obtained in the first reading. Table 4 . Example of raw fluorescence signal data (in triplicates) and TR-FRET ratio calculation. J o u r n a l P r e -p r o o f Time resolved amplification of cryptate emission: a versatile technology to trace biomolecular interactions SARS-COV-2 spike binding to ACE2 in living cells monitored by TR-FRET HTRF: A technology tailored for drug discovery -a review of theoretical aspects and recent applications A general method for the covalent labeling of fusion proteins with small molecules in vivo Probing molecular interactions with homogeneous techniques based on rare earth cryptates and fluorescence energy transfer NIH Image to ImageJ: 25 years of image analysis Conformational dynamics of SARS-CoV-2 trimeric spike glycoprotein in complex with receptor ACE2 revealed by cryo-EM BKG BKG BKG Non-specific binding Non-specific binding Non-specific binding We thank all the members of the Jockers lab for the discussion at the initial phase of the project. This work was supported by Agence Nationale de la Recherche The authors declare no competing interests. No detectable differences between BKG and TB signals in the d2 fluorescent channel (Expected outcomes). Decrease the cell density used in the assay in order to decrease BKG signal. Make sure RBD-d2 is freshly prepared and added to the corresponding wells. Increase the RBD-d2 concentration or perform a saturation curve to verify that RBD-d2 is working as expected (see saturation curve in Cecon et al., 2021 ). No significant difference between TR-FRET signals for "TB" and "NSB" (Expected outcomes). Make sure that the concentration of non-labelled RBD is at least 100x higher than the concentration of RBD-d2. Check (by visual inspection) for the presence of aggregates in the RBD stock solution, as the real concentration of the protein might be reduced if insoluble aggregates are present. Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Ralf Jockers (ralf.jockers@inserm.fr). Plasmids generated in this study have been deposited to Addgene (SNAP-ACE2, id: 178592). .  This paper does not report original datasets.  This paper does not report original code.  Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.