Tumour cell blebbing and extracellular vesicle shedding: key role of matrikines and ribosomal protein SA

Abstract

Background: Carcinogenesis occurs in elastin-rich tissues and leads to local inflammation and elastolytic proteinase release. This contributes to bioactive matrix fragment (Matrikine) accumulation like elastin degradation products (EDP) stimulating tumour cell invasive and metastatic properties. We previously demonstrate that EDPs exert protumoural activities through Hsp90 secretion to stabilised extracellular proteinases.

Methods: EDP influence on cancer cell blebbing and extracellular vesicle shedding were examined with a videomicroscope coupled with confocal Yokogawa spinning disk, by transmission electron microscopy, scanning electron microscopy and confocal microscopy. The ribosomal protein SA (RPSA) elastin receptor was identified after affinity chromatography by western blotting and cell immunolocalisation. mRNA expression was studied using real-time PCR. SiRNA were used to confirm the essential role of RPSA.

Results: We demonstrate that extracellular matrix degradation products like EDPs induce tumour amoeboid phenotype with cell membrane blebbing and shedding of extracellular vesicle containing Hsp90 and proteinases in the extracellular space. EDPs influence intracellular calcium influx and cytoskeleton reorganisation. Among matrikines, VGVAPG and AGVPGLGVG peptides reproduced EDP effects through RPSA binding.

Conclusions: Our data suggests that matrikines induce cancer cell blebbing and extracellular vesicle release through RPSA binding, favouring dissemination, cell-to-cell communication and growth of cancer cells in metastatic sites.

Fig. 1

EDPs cause fast membrane bleb formation in HT-1080 cells. a Scanning electron microscopy of control (Ctl) and EDP-treated HT-1080 cells after 40 min of incubation. Scale bar: 10 µm. b EDP dose-dependent induction of blebbing was evaluated by counting 10 fields per well under a phase contrast optical microscope after a 40 min incubation. Results were expressed as mean ± S.E.M. Data from one experiment, representative of three independent experiments, are shown. c HT-1080 cells were treated with 50 µg/mL EDPs from 0 to 120 min. The quantification of the blebbing cell number was evaluated by counting 10 fields per well under a phase contrast optical microscope. Results were expressed as mean ± S.E.M. Data from one experiment, representative of three independent experiments, are shown. d EDPs stimulate cell membrane blebbing in 3D collagen matrix. GFP-Hsp90 transfected HT-1080 cells were seeded in a 3D collagen matrix in the presence of EDPs. 3D confocal imaging was realised after 24 h of incubation to study the HT-1080 cell morphology. In the presence of EDPs, HT-1080 cells adopted an amoeboid morphology with many membrane blebs and shed extracellular vesicles (white arrows). Scale bar: 10 µm. e Real-time imaging and analysis of bleb formation in GFP-Hsp90-transfected cell. Live-cell imaging by microscope coupled with confocal Yokogawa spinning disk of control and EDP-treated HT-1080 cells. f Analysis of bleb expansion and retraction on cell sections (red line Y-axis) on the 3D real-time sequences. Snapshots of the time series at 12 s intervals from which the analysis were generated (Z-axis/Y-axis). Data from one experiment, representative of 10 independent experiments are shown. g Graph of a bleb extension and retraction following the cell section analysis. Results are means of a representative experiment (n = 10) repeated on five independent experiments. h Snapshots of time-lapse phase-contrast microscopy movies of HT-1080 blebbing cell in presence of EDPs. Scale bar: 10 µm. *p < 0.05, **p < 0.01, ***p < 0.001, significantly different from control

Fig. 2

EDP-induced membrane blebbing leads to tumour extracellular vesicle shedding. a 3D Time-lapse snapshots of cell-to-cell communication between two blebbing cells in the presence of EDPs. Snapshot of 3D data was done using Imaris. White arrows indicate tumour shed extracellular vesicle. Scale bar: 10 µm. b Tumour extracellular vesicle binding and uptake by target cells. DiOC18(3) (green) stained EDP-dependent tumour shed extracellular vesicles were incubated for 6 h with modified calcein red-orange (red) staining tumour HT-1080 cells, HUVEC or human skin fibroblasts. Confocal imaging was realised to study the extracellular vesicle localisations. Scale bar: 10 µm c Extracellular vesicles were prepared from cell-conditioned medium by centrifugation and ultracentrifugation after 24 h of incubation. The extracellular vesicle (EV) protein content per mL of HT-1080 cell culture media in presence or absence of EDPs was measured by the Bradford micro-assay. Data are expressed as mean ± S.E.M. Values from five independent experiments, each performed in triplicate. *p < 0.05, significantly different from control. d Diameters of the isolated extracellular vesicles were measured by optical microscopy and expressed as a percentage of extracellular vesicles. e Structure of isolated extracellular vesicles (black arrows) visualised by transmission electron microscopy. The diameter of observed vesicles ranged from 300 to 700 nm. Scale bar: 0.5 µm. f Confocal microscopy analysis of MMP-2, MMP-14, P-ERM, Hsp90, αvβ3 integrin and actin in EDP-induced membrane blebs. White arrows indicate membrane blebs. Scale bar: 5 µm. g One microgram of extracellular vesicle (EV) proteins and 100 µg of cell extract proteins were analysed by western blot. h Quantification of Matrigel invasion by HT-1080 cells in presence of EDPs or extracellular vesicles (EVs) (amount of EVs corresponding to 1 µg of extracellular vesicle protein) after 6 h of incubation. Data are expressed as mean ± S.E.M. Values from three independent experiments, each performed in triplicate. *p < 0.05, significantly different from control

Fig. 3

EDP-induced blebbing depends on calcium influx and on RhoA/ROCK/MLC signalling pathway. a Measure of cytosolic calcium concentration using FURA-2 loaded HT-1080 cells in response to EDP stimulations. Calcium flux is reported as ratio 350 nm/380 nm fluorescence of FURA-2 in twenty HT-1080 cells. Data from one experiment, representative of five experiments are shown. Calcium-free media (0 Ca) were used to confirm the key role of the extracellular calcium. b Measure of cytosolic calcium concentration using FURA-2 loaded HT-1080 cells in presence of nifedipine after stimulation with EDPs. Calcium flux is reported as ratio 340 nm/380 nm fluorescence of FURA-2 in HT-1080 cells. The results shown are representative experiments (n = 3). Nifedipine was used at 10 µM. c Blebbing inhibition in HT-1080 cells in the presence of EDTA, evaluated by counting 10 fields per well under a phase contrast optical microscope after EDP stimulation for 40 min. Data from one experiment, representative of three independent experiments, are shown. d Immunolocalisation by confocal microscopy of RhoA, ROCK2, ROCK1, P-LIMK1/2, cofilin, P-ERM, calpain1, actin and snapshot of mCherry-MLC transfected cells after treatment with EDPs versus untreated control HT-1080 cells (Ctl). Snapshot of 3D data was done using Imaris. Scale bar: 10 µm. e Schematic representation of the signalling pathways leading to EDP-induced blebbing and extracellular vesicle shedding. f Images from confocal microscopy analysis of cleaved caspase-3 and annexin-5 distributions after treatment with doxorubicin and EDPs. Cells were cultured on glass slides, fixed with paraformaldehyde and labelled with an anti-cleaved caspase-3 antibody (green) and annexin5-AF568 (red). Scale bar: 10 µm

Fig. 4

Correlation between EDP-stimulated Hsp90 secretion and EDP-induced membrane blebbing. a Different cell types were treated with EDPs and extracellular Hsp90 was analysed by western blot. Actin level attested the equal amount of protein loaded in the control and EDP. Data from one experiment, representative of three independent experiments, are shown. b Quantitative evaluation of Hsp90 secretion by different cell types. c Quantification of the blebbing cell number was evaluated by counting 10 fields per well under a phase contrast optical microscope after EDP stimulation for 40 min. Data from one experiment, representative of three independent experiments, are shown. d Correlation between Δblebbing and ΔHsp90 secretion in different cell types. Δblebbing = EDP-dependent blebbing – unstimulated blebbing. ΔHsp90 = EDP-dependent Hsp90 secretion – unstimulated Hsp90 secretion. Linear regression test was performed for each analysis and the R and p values are indicated below the graph. e Blebbing inhibition in HT-1080 cells in the presence of blebbistatin (50 µM), evaluated by counting 10 fields per well under a phase contrast optical microscope after EDP stimulation for 40 min. Data from one experiment, representative of three independent experiments, are shown. f HT-1080 cells were pretreated with blebbistatin (50 µM) for 1 h then stimulated by EDPs. Cell culture media were analysed for Hsp90, MMP-2, MMP-9 and uPA secretions (24 h of treatment) by western bot, gelatin zymography and gelatin-plasminogen zymography. g Blebbing inhibition in HT-1080 cells in the presence of Y27632 (25 µM), evaluated by counting 10 fields per well under a phase contrast optical microscope after EDP stimulation for 40 min. Data from one experiment, representative of three independent experiments, are shown. h HT-1080 cells were pretreated with Y27632 (25 µM) for 1 h then stimulated by EDPs. Cell culture media were analysed for Hsp90, MMP-2, MMP-9 and uPA secretions (24 h of treatment) by western bot, gelatin zymography and gelatin-plasminogen zymography

Fig. 5

VGVAPG and AGVPGLGVG synthetic elastin peptides induced membrane bleb formation. HT-1080 cells were treated with three different synthetic peptides at concentrations ranging from 10⁻¹¹ to 10⁻³ M for 40 min. The quantification of the blebbing cell number was evaluated by counting 10 fields per well under a phase contrast optical microscope. Results were expressed as mean ± S.E.M. Data from one experiment, representative of three independent experiments, are shown

Fig. 6

EDP-induced blebbing involves the RPSA elastin-laminin receptor. Identification of the RPSA protein as the EDP receptor by affinity chromatography. a Eluted samples were submitted to silver staining and b western blot using RPSA antibodies. c Optical section realised by confocal microscopy of HT-1080 cells incubated with Tamara-AG9 (AG9) at 5 × 10⁻⁵ M for 1 h at +4 °C and after immunolocalisation of RPSA protein. Colocalisations were studied with Colocalisation plugin of ImageJ. Scale bar: 20 µm. d Blebbing quantification in HT-1080 cells in the presence or not of RPSA-blocking monoclonal antibody (10 µg/mL) evaluated by counting 10 fields per well under a phase contrast optical microscope after EDPs and AGVPGLGVG stimulations for 40 min. Data from one experiment, representative of three independent experiments, are shown. Results (mean ± S.E.M) were expressed as percentage of control (EDPs untreated cells). **p < 0.01, ***p < 0.001. e Real-time PCR analysis of RPSA 48 h after treatment with RPSA siRNA vs negative control siRNA-treated cells (Ctl siRNA). Results (mean ± S.E.M; n = 3) were expressed as the ratio of RPSA mRNA to EEF1a1 mRNA. f RPSA siRNA transfected HT-1080 cells incubated with TAMRA-AGVPGLGVG (AG9) at 5 × 10⁻⁵ M for 1 h at +4 °C. Confocal imaging was realised to study the TAMRA-AGVPGLVGV and RPSA localisations on HT-1080 cells. Scale bar: 10 µm. g RSPA expression was analysed in different cell types by western blot. Data from one experiment, representative of three independent experiments, are shown. h Quantitative evaluation of RPSA expression by different cell types. i Correlation between Δblebbing and ratio RPSA/actin in different cell types. Δblebbing = EDP-dependent blebbing – unstimulated blebbing. Linear regression test was performed for each analysis and the R and p values are indicated in the graph. j Optical section obtained by confocal microscopy of control HT-1080 cell and EDP-induced blebbing HT-1080 cell to analysis RPSA localisation; western blot for RPSA realised with 10 µg of HT-1080 cell extract and extracellular vesicle proteins. Scale bar: 10 µm