Cell culture

HEK293 (human embryonic kidney cell line, ATCC Cat# CRL-3216, RRID:CVCL_0063) were obtained from ATCC and cultivated in DMEM/Ham’s F-12 medium (FG 4815, Biochrom, Berlin, Germany), supplemented with 10% fetal calf serum (FCS). Medium was changed to DMEM without FCS prior to addition of stimuli. Transfections were performed with Lipofectamine 2000 according to the manufacturer’s instructions (ThermoFisher Scientific, Dreieich, Germany).

Plasmid transfection for FRET and FLIM measurements

HEK293T were seeded in a 96-well glass bottom plate (Greiner) at a density of 10,000 cells per well and transfected with plasmids expressing AT1R-mTurquoise2, EGFR-mTurquoise2, MAS1-YPet, AT1R-YPet or interaction reducing mutants of AT1R-YPet. Transfection was performed using the TurboFect reagent (Thermo Fisher Scientific) according to the manufacturer`s recommendations. Imaging experiments were performed one day after transfection.

Förster resonance energy transfer (FRET)

Live-cell FRET imaging was performed at 37 °C using a Nikon A1R confocal microscope equipped with a 60 × oil immersion objective (plan apo lambda, Nikon, n.a. = 1.4), a PMT detector unit (Nikon, Minato, Japan) and a humidified O2/CO2 cage incubator (okolab, Ottaviano, Italy) as previously described [46]. Images were acquired and processed using the NIS-Elements FRET module (Nikon). Fluorescence of mTurquoise2-taged constructs (FRET donor) was exited using a 405 nm laser diode (Cube 405-100C, Coherent, Santa Clara, USA) and fluorescence emission was detected in the spectral range of the donor (465–500 nm, DD image) and the acceptor (525–555 nm, DA image), respectively. YPet-tagged constructs (FRET acceptor) were exited using the 514 nm laser line of an argon laser (Melles Griot, Bensheim, Germany) and detected in the spectral range of the acceptor (525–555 nm, AA image). Laser power and detector gain were set in a way to obtain best signal intensities while avoiding oversaturation within the region of interest (cell membranes). Calculation of FRET index was calibrated using donor and acceptor only samples to determine the correction factors for donor crosstalk (α) and the acceptor's direct excitation (β) in the DA image. Images displaying the color-coded FRET index were calculated as intensity of the corrected FRET image normalized by the intensity of the donor image according to the following formula (FRET index = 100% * (DA−αDD−βAA)/DD). FRET values were determined from membranous regions of the cells, only.

Fluorescence lifetime imaging microscopy (FLIM)

Fluorescence lifetime images were acquired using the FLIM upgrade kit (Picoquant, Berlin, Germany) for the Nikon 1AR confocal laser scanning microscope. Fluorescence of mTurquoise2 was excited using a pulsed laser source (PDL 828 Sepia II, Picoquant) at a wavelength of 444 nm and a repetition rate of 20 MHz. Single photons and their arrival times were detected (PicoHarp300, 483/35 filter, Picoquant) using time correlated single photon counting (TCSPC) method. To avoid pile-up effects, the excitation laser intensity was adjusted for each cell to keep maximum count rate below 2000 kcps. Photons were counted for up to 30 cycles at a capture rate of 1 frame per second. Decay profiles were analysed using SymPhoTime 64 software (Picoquant). Membranous regions of the cells were selected and fitted by employing either one- or two-exponential reconvolution fits using a measured instrument response function (IRF). IRF was measured from fluorescein quenched with saturating concentrations of potassium iodide. In case of FRET, two exponential fitting was the best fitting approach and therefore chosen for direct comparison. In diagrams, the amplitude weighted average lifetime is displayed. For the display of the average decay profiles, photon counts were normalized to the peak value and averaged for all cells measured.

Co-immunoprecipitation

HEK293T cells were grown for 24 h in 25 cm2 culture flasks with a densitiy of 1 × 106 cells in DMEM high-glucose supplemented with 10% FCS, 1% penicillin–streptomycin and 1% GlutaMAX (Thermo Fisher Scientific). pLVX-HA-AGTR1 was generated from pLVX-IRES-Neo (Takara catalogue # 632,181) and pcDNA3.1( +)-HA-hAGTR1 (Bloomsburg University cDNA Resource Center, USA, Catalog Number: #AGTR10TN01). PCR was used to introduce a SpeI restriction site at the N-Terminus of HA-hAGTR1 while keeping the NotI restriction site from pcDNA3.1( +)-HA-hAGTR1. The resulting PCR fragment was digested using SpeI and Not1 and cloned into pLVX-IRES-Neo. For expression of human influenza hemagglutinin (HA)-tagged AT1R as well as EGFP-tagged EGFR, transfection of the plasmids pLVX-HA-AGTR1 (Takara bio, San Jose, CA, USA) and pEGFP-EGFR (addgene #32751) alone or in combination was performed by mixing the respective plasmid DNA (2 µg) with 2 M CaCl2, water, and 2 × HBS at pH 7.07 and adding the mixture dropwise to the cell culture flasks. After 20 h, medium was changed once. Two days after transfection, proteins were harvested. For this, cells were lysed using a pre-made lysis buffer (Cell Signaling Technology, Danvers, MA, USA) containing 20 mmol/L Tris/HCl, pH 7.5, 1 mmol/L Na2EDTA, 1 mmol/L EGTA, 150 mmol/L NaCl, 1% Triton X-100, 2.5 mmol/L Na4P2O7, 1 mmol/L b-glycerophosphate, 1 mM Na3VO4, and 1 μg/ml leupeptin and a premade protease and phosphatase inhibitors single use cocktail (Thermo Fisher Scientific) as previously described [47]. The lysates were subsequently centrifuged for 5 min, 4 °C, at 1500×g to remove cell debris. Receptor-receptor interaction was subsequently determined by Western Blot analyses (description see below).

Co-immunoprecipitation of EGFR-EGFP or HA-AT1R fusion proteins was carried out utilizing the Immunoprecipitation Kit Dynabeads (Thermo Fisher Scientific, Waltham, MA, USA). First, the co-IP bead complex was prepared with 5 µg of the monoclonal GFP antibody JL-8 from Takara bio (order number 632381) or the human hemagglutinin (HA) antibody from Sigma-Aldrich (St. Louis, Missouri, USA; order number H3663). Each protein lysate (1 mg) was added to the HA antibody/Dynabead complex and incubated overnight. The bound proteins were eluted and prepared for Western blot analyses using 2 µl of the respective eluate.

Equal amounts of eluate (2 µl/lane) were separated by SDS-PAGE and transferred to a nitrocellulose membrane. Membranes were then incubated with primary antibody solution (anti-HA tag (clone C29F4), 1:1000, 5% BSA, Cell signaling technologies. Bound antibodies were detected by peroxidase-conjugated secondary antibodies and the ECL system (Amersham Bioscience, Amersham, UK).

In silico modeling

The inactive state of AT1R was derived from PDB ID: 4ZUD, crystallized in monomer form, while the active state was sourced from PDB ID: 6OS0. The protonation states were determined at pH 7.4 using the Protein Preparation Wizard in Schrodinger Biologics Suite 2023-3, employing the OPLS4 force field. AT1R homodimer modeling leveraged the structure of the CCR4 homodimer (PDB ID: 3OE0) as a foundation, facilitating the generation of molecular systems representing both inactive and active states of the TM4-TM5 interacting interface. For modeling the inactive state of the TM1-TM2 and TM8 interacting interface, the AT1R homodimer crystal structure (PDB ID: 6OS0), which is present in an active state, was employed. The structural models were generated using Protein Structure Alignment tool accessed via Maestro.

Due to the clashing observed in the inactive state of TM4-TM5, we proceeded to generate two mutation models, focusing on the active state. This was done using the Residue Scanning Tool in Maestro, following the alignment steps for both homodimers. Mutation model 1 (MUT1) features alterations including S189A, I193A, L197A, I201A, L202A, L205A, and F206A, while mutation model 2 (MUT2) involves mutations Y54A, F55A, F96A, Y99A, and L100A. The creation of these models leveraged homology modeling approaches and incorporated a relaxation step (with an RMSD of 0.3 Å), allowing for the concurrent mutation of the chosen residues.

Single cell reporter gene analysis by digital high content microscopy

We assessed activity of the transcription factor SRF by reporter gene assays. Changes in the expression of the reporter gene is a measure for transcription factor activation. Thus, we measured reporter gene activity under different conditions, calculated the changes versus control and denominated these values transcription factor activation (e.g. SRF activation). A detailed description of data acquisition and analysis with exemplary images is given in supplementary methods file. With this approach only signal from transfected cells are recorded, preventing confounding effects from non-transfected cells. Reporter for SRE (sequence GGATGTCCATATTAGGA) transcription factor was purchased from Qiagen, Hilden, Germany. We used the Cignal™ System (http://www.sabiosciences.com/reporterassays.php) with Monster-green fluorescent protein (MGFP) as reporter. The respective transfection control was red fluorescent protein (RFP) under the control of a constitutive CMV promoter. After transfection with Polyfect (Qiagen, Hilden, Germany) cells were incubated as described in the figure legends and reporter activity was determined as recommended by the manufacturer by digital fluorescence microscopy (Cytation 3, BioTek, Bad Friedrichshall, Germany or the PerkinElmer Operetta CLS™ high content screening system). To determine the cellular responses, first transfected cells were identified according to their red fluorescence (Ex 586/15 nm; Em 647/57 nm; DM 605 nm; LED 590 nm) and their number, mean fluorescence intensity, area, circularity and integral fluorescence intensity determined. Cell identification and determination of the parameters was performed with the Gen5 2.09 software (BioTek, Bad Friedrichshall, Germany). For this purpose, the object recognition parameters (background fluorescence, threshold fluorescence change, rolling ball diameter, object minimum and maximum size, light exposure time, light intensity, gain of image acquisition) were determined during three independent training experiments and subsequently applied to all experiments, making them comparable. Second, the mean green fluorescence intensity (Ex 469/35 nm; Em 525/39; DM 497 nm; LED 465 nm) of the red cells as well as the integral green fluorescence of red cells was determined, using the same routine as for red cells. Finally, red cells that were also green were identified and their number, mean fluorescence intensity, area, circularity and integral green fluorescence integral. The change in fraction of red cells (= transfected cells that could respond) that show a green signal (= active SRF) corresponds to the digital response (switching on of previously inactive cells). The change in green fluorescent intensity of green cells corresponds to the analogue response (enhancing the activity of already activated cells). The overall response to a stimulus is the change in green fluorescence (= SRF activity) of all red cells. This overall response results from the changes in digital and the analogue component (Δoverall = Δdigital x Δanalogue).

Morphological analysis

Cell circularity and cell area, as a surrogate for cell size, were determined from the RFP fluorescence images obtained by digital high content microscopy. Both parameters were calculated by the Gen5 3.11 software. First, cell area (A) and perimeter (P) were measured. Then, circularity (C) was calculated as C = 4 × π × A/P2. Theoretically, circularity can range from 0 to 1, representing perfectly linear to completely circular morphology, respectively. An increase in circularity results from cell dedifferentiation or contraction.

In-Cell-ELISA: single cell immunofluorescence imaging by digital microscopy

To investigate the expression of putative SRF-target genes or the phosphorylation of pERK1/2 in transfected cells only, cells were transfected with pEGFP in addition to AT1R, followed by stimulation as indicated. A detailed description of data acquisition and analysis with exemplary images is given in the supplementary methods file. With this approach only signals from transfected cells are recorded, preventing confounding effects from non-transfected cells, by contrast to immunoblotting aproaches. The approach was validated in our previous study [36] and combines the advantages of flexible transient transfection with the analysis of transfected cells only. After cell fixation with 4% formaldehyde for 24 h at 4 °C, cells were permeabilized (0.1% Triton X-100 in TBS; 37 mg/l Na-orthovanadate) and non-specific antibody binding was blocked using 5% donkey serum in permeabilization buffer. Primary antibodies from Cell Signaling Technologies, Frankfurt, Germany (phospho-ERK1/2 #9101, RRID:AB_331646, 1:1000; cFOS, # 2250, RRID:AB_2247211, GAPDH #2118, RRID:AB_561053, 1:1000) were diluted in 1% BSA in permeabilization buffer and incubated overnight at 4 °C. Donkey anti-rabbit AlexaFluor568 secondary antibody (#A10042, Invitrogen Life Technologies, Darmstadt, Germany) was then diluted 1:500 in 1% BSA in permeabilization buffer and incubated for 1 h in the dark at room temperature. Nuclei were stained by diluting DAPI in PBS at 1 µg/ml and applied for 10 min in the dark at room temperature. Digital microscopy was performed using a 20 × objective. Subsequently the images were analysed with the Gene 5 3.11 software (BioTek, Bad Friedrichshall, Germany) and in-build routines after adjusting the necessary parameters (background, threshold, object size, rolling ball size). The sequence of single cell analysis for was the following: 1. Identify transfected cells by green fluorescence (= EGFP fluorescence) red. 2. Determine cell number, mean cell area, mean fluorescence intensity 3. Determine mean red fluorescence (= protein of interest marked by AlexaFluor568-labelled antibody. 4. Identify transfected cells positive for the protein of interest. 5. Determine number of transfected cells positive for the protein of interest. 6. Determine the intensity of the protein of interest (red fluorescent level) of cells positive for the protein of interest. The sequence of single nucleus analysis for was the following: 1. Identify nuclei by DAPI fluorescence. 2. Identify the subpopulation of nuclei of transfected cells by green fluorescence (= EGFP fluorescence) red. 3. Determine nuclei number, mean nuclear area, mean fluorescence intensity 4. Determine mean red fluorescence (= protein of interest marked by AlexaFluor568-labelled antibody. 5. Identify nuclei of transfected cells positive for the protein of interest. 6. Determine number of nuclei of transfected cells positive for the protein of interest. 7. Determine the intensity of the protein of interest (red fluorescent level) of nuclei from transfected cells positive for the protein of interest.

Materials

The following plasmids were used: pDsRed2 (#632,406, Clontech, Mountain View, CA; 3.6 ng/cm2); pEGFR-C1 (#6084-1, Clontech Laboratories now Takara Bio USA, Göteborg, Sweden; 3.6 ng/cm2); pCMV6-XL4-AT1R (SC 108918, Origene, Rockville, MD; 36 ng/cm2); pCMV6-XL4-AT1R-MUT1 (obtained by site-directed mutagenesis from SC 108918, Origene, Rockville, MD; 36 ng/cm2) and pCMV6-XL4-AT1R-MUT2 (obtained by site-directed mutagenesis from SC 108918, Origene, Rockville, MD; 36 ng/cm2), pLVX-HA-AGTR1 (Takara bio, San Jose, CA, USA), pEGFP-EGFR (addgene #32751), pN1-AT1R-YPet (Takara bio, San Jose, CA, USA), pN1-AT1R-YPet-MUT1 (Takara bio, San Jose, CA, USA), pN1-AT1R-YPet-MUT2 (Takara bio, San Jose, CA, USA), pN1-EGFR-mTurquoise2 (Takara bio, San Jose, CA, USA), pN1-AT1R-YPet (Takara bio, San Jose, CA, USA). All cloning procedures regarding the generation of pN1-AT1R-YPet, pN1-EGFR-mTurquoise2 and pN1-MAS1-YPet as well as sequencing confirmation of these steps were carried out by Synbio Technologies (Monmouth Junction, NJ). Site-directed mutagenesis to generate two AT1R mutants AT1R-MUT1 (S189A, I193A, L197A, I201A, L202A, L205A, F206A; S026251-02-K319510), AT1R-MUT2 (Y54A, F55A, F96A, Y99A, L100A; S026251-01-K319394), AT1R-YPet-MUT1 (S189A, I193A, L197A, I201A, L202A, L205A, F206A; S020606-3) and AT1R-YPet-MUT2 (Y54A, F55A, F96A, Y99A, L100A; S020606-1) as well as sequencing confirmation of the mutations was performed by Synbio Technologies (Monmouth Junction, NJ). Unless stated otherwise all materials were purchased from Sigma, Munich, Germany.

Statistics

ANOVA or Kruskal–Wallis ANOVA on ranks followed by post hoc testing (e.g. Holm-Sidak or Dunn method), Student´s T-Test or Mann–Whitney rank sum test were used as applicable according to pre-test data analysis by SigmaPlot 12.5 (Systat Software, Inc., San Jose, CA) or STATplus (AnalystSoft Inc., Brandon, GB). A p-value < 0.05 was considered significant. Biometrical planning was performed with α = 0.05 and β = 0.8. Experiments on reporter gene expression, ERK1/2-phosphorylation, cFOS-expression and cell morphology were performed with five cells passages or more with 3 or more individually treated cell culture wells in each experiment. The numbers are given as N/n, where N represents the number of passages and n the number of individually treated cell culture wells. The number of passages was used for statistical testing. In each well the transfected cells (identified either by RFP or EGFP) were analysed on a single cell basis and the mean value of all cells analysed in one well used for further evaluation. Thus, the value of each individually treated cell culture well results from several transfected cells (in the range of 300–400 transfected cells per well). For more details see supplementary methods SM01–SM16. The box plots show the median, 10th, 25th, 75th, and 90th percentiles as vertical boxes with error bars. Line plots are presented as mean ± 95% confidence intervals. FRET and FLIM were determined in two independent experiments with 4 or more independent biological replicates in each experiment.