This study was funded by the Swiss National Science Foundation SNF 146249. The authors have no other affiliations or financial involvement with any organization or entity with a financial interest. MA is employed by Mymetics SA, which provided support in the form of salary to MA and specific research materials for the preparation of nanoparticles, but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of MA are articulated in the "author contributions" section.
‡ CvG and FB are joint senior authors on this work.
The respiratory tract with its ease of access, vast surface area and dense network of antigen-presenting cells (APCs) represents an ideal target for immune-modulation. Bio-mimetic nanocarriers such as virosomes may provide immunomodulatory properties to treat diseases such as allergic asthma. In our study we employed a triple co-culture model of epithelial cells, macrophages and dendritic cells to simulate the human airway barrier. The epithelial cell line 16HBE was grown on inserts and supplemented with human blood monocyte-derived macrophages (MDMs) and dendritic cells (MDDCs) for exposure to influenza virosomes and liposomes. Additionally, primary human nasal epithelial cells (PHNEC) and EpCAM+ epithelial progenitor cell mono-cultures were utilized to simulate epithelium from large and smaller airways, respectively. To assess particle uptake and phenotype change, cell cultures were analyzed by flow cytometry and pro-inflammatory cytokine concentrations were measured by ELISA. All cell types internalized virosomes more efficiently than liposomes in both mono- and co-cultures. APCs like MDMs and MDDCs showed the highest uptake capacity. Virosome and liposome treatment caused a moderate degree of activation in MDDCs from mono-cultures and induced an increased cytokine production in co-cultures. In epithelial cells, virosome uptake was increased compared to liposomes in both mono- and co-cultures with EpCAM+ epithelial progenitor cells showing highest uptake capacity. In conclusion, all cell types successfully internalized both nanocarriers with virosomes being taken up by a higher proportion of cells and at a higher rate inducing limited activation of MDDCs. Thus virosomes may represent ideal carrier antigen systems to modulate mucosal immune responses in the respiratory tract without causing excessive inflammatory changes.
With its ease of access, the vast surface area and extended network of dendritic cells (DCs), the respiratory tract represents a promising target for inhaled immune-modulatory approaches by bio-mimetic nanocarriers such as virosomes and liposomes [
DCs are the most effective antigen-presenting cells (APCs) found in the respiratory tract that capture, process and present antigens [
Besides DCs, other APCs are present in the respiratory tract including B cells and macrophages [
Airway epithelial cells (AECs) provide not only barrier function but also play an important role in maintenance of pulmonary homeostasis [
Bio-mimetic antigen nanocarriers such as virosomes or liposomes are promising compounds for inhaled, non-invasive, and specifically tailored immune-modulation. Influenza virosomes and liposomes are spherical vesicles, comprised of lipids and close to 100 nm in diameter. Not only do virosomes provide carrier function by incorporating or encapsulating antigens, but they additionally are potent immune stimulators due to influenza virus envelope proteins integrated in the membrane [
The aim of this study was to establish a human model of the respiratory tract to study the interplay between epithelial cells, macrophages, and DCs, as well as interactions with bio-mimetic nanocarriers such as liposomes and virosomes to closely represent the human situation
The fluorchrome Atto647N NHS ester (1.7 μmol in DMSO; Sigma) was conjugated to 8.5 μmol OPPE (1-Oleoyl-3-palmitoyl-rac-glycero-2-phosphoethanolamine; Bachem, Bubendorf, Switzerland) in 90 mM dodecyl octaethylene glycol ether (OEG; Sigma)-PBS pH 7.4.
Influenza virosomes were prepared as follows. In short: per ml of final formulation, 8 mg of DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine; Merck, Darmstadt, Germany) and 1 mg of OPPE were dissolved in 100 mM OEG in PBS pH 7.4 (52.7 mM phosphate, 82 mM NaCl). Inactivated influenza A/Brisbane/59/2007 H1N1 virus was mixed with PBS and centrifuged at 100.000xg for 1h at 18°C. The pellet of inactivated influenza virus was resuspended with 100 mM OEG-PBS pH 7.4 for 10 min followed by sonication for 1 min at 30°C. This mixture was centrifuged at 100.000xg for 1h at 18°C to pellet down the nucleocapsid complex. The supernatant containing the solubilized influenza membrane proteins and lipids was used for virosome formulation and mixed with phospholipids at a concentration of 0.2 mg/ml hemagglutinin (HA). Virosome formation took place after removal of OEG detergent using 0.375 g per ml of formulation of SM2 Bio-Beads (BioRad) twice for 1h and once for 30 min at room temperature whilst mixing at 100 rpm. Fluorescent virosomes were obtained by adding Atto647-PE as indicated prior to detergent removal to enable peptide incorporation. Liposomes were prepared similarly by leaving out the influenza virus component but following the same procedure. At the end of the process virosomes and liposomes were sterile filtered on 0.22 μm units (Millex-GP; Merck Millipore).
Nanocarriers were thoroughly characterized prior to use. Particle size, homogeneity, as well as the amount of HA were analyzed. Size determination was performed in PBS pH 7.4 by dynamic light scattering (DLS) using a Zetasizer Nano S instrument and by nanoparticle tracking analysis (NTA) on a NanoSight NS300 instrument (both from Malvern Instruments, Malvern, UK). Samples were routinely measured for endotoxin by performing limulus amebocyte lysate (LAL) test.
Influenza HA concentration of virosomes was determined by SDS-PAGE using a 4–20% precast mini-protean TGX gel (BioRad) and Coomassie Brilliant Blue R-250 (BioRad) staining with a One-Color Protein Molecular Weight Marker (Odyssey, LiCor). HA concentration was confirmed by Spotblots and Western Blots using nitrocellulose (0.2 μm pore size; Life Technologies) and rabbit anti-HA and rabbit anti-OVA serum followed by secondary goat-anti-rabbit IRDye 800CW (LiCor Biosciences), visualized and quantified using the LiCor Odyssey imaging system (Lincoln, Nebraska, USA). Selected samples were also quantified in a SRID (single radial immunodiffusion) assay to confirm the HA concentration.
Cells were obtained from buffy coats of healthy individuals provided by blood donors from the Regional Red Cross Blood Donation Centre (Bern, Switzerland). Peripheral blood mononuclear cells (PBMCs) were obtained by Pancoll density centrifugation and monocytes were isolated by CD14 positive selection using MACS microbeads (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). MDDCs were generated by culturing in RPMI 1640 medium (Invitrogen) supplemented with 10% fetal calf serum (FCS), 1% L-glutamine (2 mM, Invitrogen), 1% penicillin/streptomycin (100 U/ml, Invitrogen), 10 ng/ml GM-CSF (R&D Systems) and 10 ng/ml IL-4 (R&D Systems) for 7 days at 37°C/5% CO2.
Cells were obtained from buffy coats of healthy individuals provided by blood donors from the Regional Red Cross Blood Donation Centre (Bern, Switzerland). PBMCs were obtained by Pancoll density centrifugation and monocytes were isolated by Monocyte Isolation Kit II (Miltenyi Biotex GmbH). MDMs were generated by culturing in RPMI 1640 medium (Invitrogen) supplemented with 10% FCS, 1% L-glutamine (2 mM, Invitrogen), 1% penicillin/streptomycin (100 U/ml, Invitrogen), 10 ng/ml M-CSF (R&D Systems) for 8 days at 37°C/5% CO2. Where indicated, macrophages were incubated with LPS (100 ng/ml) and IFN-γ (20 ng/ml) for 24h to induce M1 or incubated with IL-4 (20 ng/ml) for 24h to induce differentiation into M2 [
Differentiated SV-40 transformed human bronchial epithelial cell line (16HBE14o- cells) was maintained in MEM 1X, with Earle’s Salts, without L-Glutamine (Gibco BRL Life Technologies Invitrogen AG, Basel, Switzerland) supplemented with 1% glutamine/penicillin/streptomycin (Gibco BRL) and 10% FCS (Gibco BRL). About 5x106 cells were seeded in collagen coated (Pure Col, Purified Bovine Collagen Solution, Advanced Biomatrix, 1:50) T75 flasks (Falcon, US) for further cell culture at 37°C/5% CO2.
Ten healthy adult volunteers were recruited for a brushing of the inferior surface of the middle turbinate of both nostrils in order to obtain airway epithelial cells. This was performed by using a cytological brush (Dent-O-Care, No 720, London, UK) [
Primary human nasal epithelial cells grown on inserts (as described above) were differentiated in Maintenance medium (PneumaCult-ALI supplemented with 6 μl Maintenance, 3 μl Hydrocortison, 1.2 μl Heparin (Stemcell), 0.9 μl Primorcin (InvivoGen, US)) given to the lower chamber, while the apical side of the epithelial cells remained facing air. Cells were cultured for 8 weeks for full differentiation.
To prospectively isolate lung pericytes and epithelial cells, we used lung specimens obtained from patients following surgical resection for early stage lung cancer. All patients gave informed written consent for usage of surgical material for research purposes, which was approved by Ethics Commission of the Canton of Bern, CH (KEK-BE:042/2015). Preparation of lung tissue and cell sorting was performed as previously described with modifications [
The triple co-cultures were prepared as previously described [
To study uptake of virosomes and liposomes, cells of interest were cultured for 18h at 37°C in presence of virosomes, liposomes or controls. One ml of suspended cells was incubated for 18h with virosomes (5 μg HA), liposomes, or control (PBS) in the same dilution as virosomes. Uptake was determined by measuring Atto647 signal by flow cytometry (LSRII, BD Biosciences, Franklin Lakes, NJ, USA). Data was analyzed using FlowJoX (TreeStar, Ashland, OR, USA) software.
MDDC and MDM phenotype was determined by flow cytometry after treatment with virosomes, liposomes or appropriate controls for 18h. One ml of cell suspension was incubated with virosomes (5 μg HA), liposomes in the same dilution, or the following controls: medium only, PBS in the same dilution as virosomes, LPS (100 ng; Sigma) or inactivated influenza virus A/Brisbane (5 μg HA). 16HBE cells were collected by trypsinisation (Gibco BRL), MDMs and triple co-cultures were resuspended by Accutase digestion (Gibco BRL) and by cell scraping. Cells were treated with FcR block on ice followed by viability staining with Fixable Viability Dye eFluor506 (eBioscience, Vienna, Austria) for 30 min on ice. As positive controls, cells heated at 65°C for 15 min or frozen at -80°C for 30 min were used. Unless indicated otherwise, antibodies were purchased from eBioscience and utilised with appropriate isotype controls. For MDMs: CD14-Alexa Fluor 700, CD68-PE-eFluor610, CD163-PerCPeFluor710, CD40-APC-Cy7 (Biolegend, London, UK), HLA-DR-Brilliant Violet 785 (BioLegend), CD86-Brilliant Violet 605 (BioLegend), CD80-Brilliant Violet 421 (BioLegend, CD206-PE-Cy7, CD36-APC-Cy7 (BioLegend). For MDDCs: CD11c-Pe-Cy7, CD1c-Alexa Fluor 700 (BioLegend), CD83-PE-Cy7 (BioLegend), PD-L1-eFluor450, PD-L2-PerCP-eFluor710 and CCR7-APC-eFluor780. For epithelial cells and triple co-cultures additionally: CD209-PE-Cy7 (DC-Sign), CD86-PE and CD326-Brilliant Violet 650 (EpCAM; BioLegend). Flow cytometry was performed by using flow cytometry SORP LSR II (BD Biosciences).
Cells were treated with 20 μg/ml Brefeldin A (eBioscience) to stop protein transport. Subsequently, cells were stained for surface markers as mentioned above. Cells were fixed in 1% formalin solution followed by intracellular staining with the following anti-cytokine antibodies with appropriate isotype controls diluted in permeabilization buffer (PBS + 0.1% saponin + 10% FCS): IL-10-Alexa Fluor 488, IL-12-PE (all eBioscience).
Medium of the upper chamber of the triple co-cultures and supernatant of 16HBE mono-cultures were collected and production of IL-8, IL-1β and TNF-α was analysed by employing ELISA Kits from R&D Systems (Minneapolis, US) according to the manufacturer’s specifications. Optical density was measured with a microplate reader (Tecan reader) at a wavelength of 450nm.
Particle uptake was confirmed using laser scanning microscopy (LSM) to detect Atto647 emission in cells. PHNECs were fixed in 70% ethanol and permeabilized with 0.2% Triton X-100 (Sigma-Aldrich). Cells were stained with primary antibodies mouse anti-β-tubulin (1:100; Life Technologies), rabbit anti-occludin (1:50; Molecular Probes) or rabbit anti-mucin (1:50; Santa Cruz Biotechnology) overnight at 4°C. The following secondary antibodies were used for 1h at room temperature: goat anti-mouse Alexa488 (1:200; Molecular Probes), goat anti-rabbit Alexa546 (1:200; Life Technologies) and phalloidin Atto390 (1:25; ATTO-TEC). Cells were washed and embedded in Aquatex mounting medium (Merck KGaA, Darmstadt, Germany). Optical sections were taken with a Zeiss LSM 710 (Carl Zeiss AG, Feldbach, Switzerland) with a 40x oil objective (Plan-Apochromat 40x/1.40 Oil) and a digital zoom of 1.4x. Image processing was performed using Imaris (Bitplane AG, Zurich, Switzerland) software.
Statistical analyses were conducted using R version 3.2.1 [
Influenza virosomes and liposomes were extensively characterized for particle size, homogeneity, as well as HA content and endotoxin contamination. Particle size was measured by DLS and NTA that routinely provided a hydrodynamic diameter (particle diameter plus water shell) of 90–96 nm, and particle sizes of 84–86 nm for all nanocarrier formulations respectively (
Samples | DLS |
NTA |
|
---|---|---|---|
Size ± SD (nm) |
PDI |
Size ± SD (nm) |
|
Virosome | 95.0 ± 0.6 | 0.03 | 85.3 ± 0.8 |
Virosome-Atto647 | 95.6 ± 0.7 | 0.04 | 84.9 ± 0.5 |
Liposome | 90.0 ± 1.7 | 0.02 | 85.5 ± 0.8 |
Liposome-Atto647 | 90.8 ± 2.5 | 0.03 | 84.0 ± 0.4 |
Influenza virosomes and liposomes with or without conjugated fluorochrome Atto647 were analyzed for particle size by DLS and NTA. One representative measurement from two independent formulations of virosomes and liposomes is shown.
1 Dynamic light scattering
2 Nanoparticle tracking analysis
3 Hydrodynamic diameter
4 Poly Dispersity Index
5 Modal particle size.
Potential cytotoxic effects were determined prior to analysis of particle uptake and phenotype change. Cells were incubated with virosomes, liposomes or appropriate controls. Cells kept for 15 min at 65°C or stored at -80°C for 30 min served as positive controls and showed a high cell death rate (data not shown). Less than 10% of dead cells were detected in all cell types after exposure to particles compared to PBS treated cells in culture (
In order to investigate whether virosomes and liposomes are taken up by different cells of the respiratory tract and to compare their dynamics in mono- and triple co-culture, cells were incubated with virosomes, liposomes or a PBS control. For triple cell co-cultures gating was performed for size, single cells, live cells and split into EpCAM+ cells for 16HBE and EpCAM- cells. EpCAM- cells were further divided into DC-Sign+ (for MDDCs) and DC-Sign- (for MDMs;
In MDDCs uptake of both nanocarrier types was observed with significantly more cells capturing virosomes than liposomes in both mono- and co-culture (p<0.001;
MDDCs (DCs), MDMs (macrophages), 16HBE, PHNEC and human epithelial progenitor cells (EpCAM+) were incubated with either virosomes (VIRO), liposomes (LIPO) or control (PBS) for 18h at 37°C. Uptake of virosomes and liposomes was determined by measuring Atto647 signal by flow cytometry. Frequency
In co-culture, MDMs showed higher uptake frequency for virosomes than liposomes (p<0.001;
Significantly more 16HBE cells in both mono- and co-culture captured virosomes than liposomes (p<0.001;
Human epithelial progenitor cells (EpCAM+) internalized more virosomes than liposomes according to MFI (
Particle uptake in PHNEC was confirmed by LSM. Cells were stained for β-Tubulin (cilia), phalloidin (actin cytoskeleton) and either occludin (tight junctions;
Uptake of liposomes
The effect of liposomes (A) or virosomes (B) exposure was analyzed by LSM. Micrographs were obtained from three-dimensional stacks of consecutive optical sections and analyzed with Imaris software. Green: cilia (β-Tubulin), red: liposomes/virosomes, blue: actin cytoskeleton, yellow: mucin. xy-projections (top panel) and xz-projections (lower panel) (1) merged image (2) cilia and liposomes/virosomes (3) mucin and liposomes/virosomes. One representative experiment from three independent experiments is shown.
We next investigated whether treatment virosomes or liposomes induce MDMs, MDDCs and 16HBE phenotype changes or activation by expression of co-stimulatory surface molecules in both mono- and co-cultures.
16HBE cells were analysed by flow cytometry for surface expression of HLA-DR, CD40, CD80 and CD86, but no significant changes were detected (
For MDDCs we analyzed the phenotypic and co-stimulatory markers HLA-DR, CD40, CD80, CD86, CD83, PD-L1, PD-L2, CCR7 and intracellular cytokines IL-10 and IL-12 with respect to their relevant isotype controls. The marker CD86 showed significant upregulation after treatment with virosomes in mono-culture compared to PBS (p<0.01) and in co-culture after PBS (p<0.001) or liposome (p = 0.05) treatment compared to mono-culture (
Cells in mono- (MO) or co-culture (CO) were incubated for 18h with either virosomes (VIRO), liposomes (LIPO) or controls (PBS, as shown). Expression of surface molecule markers HLA-DR, CD40, CD80, and CD86 were measured by flow cytometry. Figures show the receptor expression in frequency
MDMs were analysed for HLA-DR, CD40, CD80, CD86, CD163, IL-10 and IL-12. The main effect observed was higher CD86 expression per cell in co-culture compared to mono-culture (p<0.01;
Cells in mono- (MO) or co-culture (CO) were incubated for 18h with either virosomes (VIRO), liposomes (LIPO) or controls (PBS, as shown). Expression of surface molecule markers HLA-DR, CD40, CD80, CD86 and CD163 were measured by flow cytometry. Figures show the receptor expression in frequency
Minimal changes in the cytokine profile were observed in 16HBE mono- and co-cultures after treatment with virosomes or liposomes. Although an increase in IL-8 secretion was observed in co-cultures after treatment with virosomes (p = 0.034), the effect was also observed with PBS (p = 0.011) and therefore might not be related to nanocarriers. No significant changes were measured for IL-1β (
Cells in mono- (MO) or co-culture (CO) were incubated for 18h with either liposomes (LIPO), virosomes (VIRO) or controls (PBS, as shown). Supernatants were collected to perform IL-1β and IL-8 ELISA. Data represents six independent experiments. Statistical significance was determined by ANOVA followed by Tukey’s HSD post hoc test to investigate individual paired comparisons. *p<0.05.
Taken together, we observed minimal differences between virosome- and liposome-induced phenotypic and co-stimulatory marker expression in different cell types, either in mono- or co-culture. MDDCs overall underwent moderate activation compared to MDMs and 16HBE cells.
Following polarization into M1 and M2 macrophages we analyzed uptake in these cells for both liposomes and virosomes. There was no significant difference in frequency of virosome or liposome uptake (
Cells in mono-cultures were differentiated for 24h into M1 and M2 type macrophages before being incubated for 18h with either virosomes (VIRO), liposomes (LIPO) or controls (PBS). Uptake of virosomes and liposomes was determined by measuring Atto647 signal by flow cytometry. Frequency
MDMs differentiated into M1 and M2 were further analysed for phenotype changes after addition of virosomes or liposomes to detect whether exposure to these nanocarriers is affecting phenotype and function of macrophages that have undergone polarization.
Cells in mono-culture were differentiated for 24h into M1 or M2 macrophages before being incubated for 18h with either virosomes (VIRO), liposomes (LIPO) or controls (PBS). Expression of surface molecule markers HLA-DR, CD80, CD86, CD36, CD163 and CD206 were measured by flow cytometry. Figures show the receptor expression in frequency
In recent years, influenza virosomes have successfully been developed for influenza and hepatitis A vaccines and as carrier systems for heterologous antigens for malaria, HIV or other pathogens [
Our
Though MDDCs in our
The uptake of virosomes by MDDCs is an essential result provided herein, since DCs have the ability to prime naïve T-lymphocytes [
Overall, there was more uptake of virosomes and liposomes in APCs compared to epithelial cells, consistent with other studies employing a triple co-culture model exposed to other particles [
The virosomal bilayer contains viral envelope proteins such as hemagglutinin and neuraminidase that both bind to sialic acid residues and trigger highly efficient receptor-mediated uptake. Liposomes, on the contrary, lack such viral proteins and thus are taken up mostly by macropinocytosis or endocytosis [
In our co-culture experiments a moderate increase of HLA-DR and CD86 expression occurred after treatment with both virosomes and liposomes, as well as with PBS, when compared to mono-cultures. A possible explanation may be the interplay of the different cell types via cell-cell contact or the release of cytokines facilitating activation in the co-culture model compared to mono-cultures. As an example, secretion of cytokines like thymic stromal lymphopoietin (TSLP) by epithelial cells may play an important role [
In conclusion, our data underlines that the triple co-culture model provides an appropriate experimental system that realistically simulates the complexity of the airway barrier, enabling to investigate how particle-cell interactions and the interplay between different cell types affects responses to novel treatments developed for the pulmonary administration. To improve the current model, particles could be applied by means of a microsprayer or air-liquid interface cell exposure system [
After incubating cells with influenza virosomes (VIRO, with (△) and without (○) Atto647), liposomes (LIPO, with (△) and without (○) Atto647) or control (PBS) for 18h, viability was tested employing a fluorescence viability dye and measuring its signal with flow cytometry. Data represents at least six independent experiments relative to PBS. Statistical significance was determined by ANOVA followed by Tukey’s HSD post hoc test to investigate individual paired comparisons.
(EPS)
Cells of interest were gated according to forward and sideward scatter (FSC/SSC). Doublets were excluded by gating for single cells (FCS-W/FCS-H). Viability was determined by using two positive controls (65°C for 15 min and -80°C for 30 min).
(EPS)
Cells were incubated with either virosomes (VIRO), liposomes (LIPO) or control (PBS) for 18h at 37°C. Uptake of virosomes and liposomes was determined by measuring Atto647 signal by flow cytometry. Frequency and MFI are shown relative to PBS. Data represents four independent experiments. Statistical significance was determined by ANOVA followed by Tukey’s HSD post hoc test to investigate individual paired comparisons.
(EPS)
Cells in mono- (MO) or co-culture (CO) were incubated for 18h with either virosomes (VIRO), liposomes (LIPO) or controls (PBS, as shown). Expression of surface molecule markers HLA-DR, CD40, CD80, CD86 was measured by flow cytometry. Figures show the receptor expression in frequency of at least six independent experiments. Statistical significance was determined by ANOVA followed by Tukey’s HSD post hoc test to investigate individual paired comparisons.
(EPS)
Cells were incubated for 18h with either virosomes (VIRO, with (△) and without (○) Atto647), liposomes (LIPO, with (△) and without (○) Atto647) or controls (PBS, as shown). Expression of surface molecule markers CD83, PD-L1, PD-L2, CCR7 and intracellular cytokines IL-10 and IL-12 was measured by flow cytometry. Figures show the receptor expression in frequency
(EPS)
Cells were incubated for 18h with either virosomes (VIRO, with (△) and without (○) Atto647), liposomes (LIPO, with (△) and without (○) Atto647) or controls (PBS, as shown). Expression of intracellular cytokines IL-10 and IL-12 was measured by flow cytometry. Figures show expression in frequency
(EPS)
Cells were incubated for 18h with medium (DCs only), or positive controls LPS and inactivated virus A/Brisbane/59/2007 H1N1 (A/B). Expression of surface molecule markers HLA-DR, CD40, CD80, CD86, CD83, PD-L1, PD-L2, CCR7 and intracellular cytokines IL-10 and IL-12 was measured by flow cytometry in MDDCs. Figures show expression in frequency
(EPS)
Cells were incubated for 18h with medium (DCs only), or positive controls LPS and inactivated influenza virus A/Brisbane/59/2007 H1N1 (A/B). Expression of surface molecule markers HLA-DR, CD40, CD80, CD86, CD163 and intracellular cytokines IL-10 and IL-12 was measured by flow cytometry. Figures show the receptor expression in frequency
(EPS)
We gratefully acknowledge the expert technical assistance provided by Sandra Barnowski and Seraina Beyeler. Flow Cytometry experiments were performed with the support of the FACS Lab at the University of Bern, Switzerland. Microscopy acquisition and analysis were performed with the support of the Live Cell Imaging Core Facility of the Department of Clinical Research coordinated by the Microscopy Imaging Center at the University of Bern, Switzerland.