1. Cancer Biology
  2. Cell Biology
Download icon

Ral GTPases promote breast cancer metastasis by controlling biogenesis and organ targeting of exosomes

  1. Shima Ghoroghi
  2. Benjamin Mary
  3. Annabel Larnicol
  4. Nandini Asokan
  5. Annick Klein
  6. Naël Osmani
  7. Ignacio Busnelli
  8. François Delalande
  9. Nicodème Paul
  10. Sébastien Halary
  11. Frédéric Gros
  12. Laetitia Fouillen
  13. Anne-Marie Haeberle
  14. Cathy Royer
  15. Coralie Spiegelhalter
  16. Gwennan André-Grégoire
  17. Vincent Mittelheisser
  18. Alexandre Detappe
  19. Kendelle Murphy
  20. Paul Timpson
  21. Raphaël Carapito
  22. Marcel Blot-Chabaud
  23. Julie Gavard
  24. Christine Carapito
  25. Nicolas Vitale
  26. Olivier Lefebvre
  27. Jacky G Goetz  Is a corresponding author
  28. Vincent Hyenne  Is a corresponding author
  1. INSERM UMR_S1109, Tumor Biomechanics, France
  2. Université de Strasbourg, France
  3. Fédération de Médecine Translationnelle de Strasbourg (FMTS), France
  4. Laboratoire de Spectrométrie de Masse BioOrganique (LSMBO), IPHC UMR 7178, CNRS, Université de Strasbourg, France
  5. INSERM UMR_S1109, Genomax, France
  6. CNRS, UMR 7245 MCAM, Muséum National d’Histoire Naturelle de Paris, France
  7. Université de Bordeaux, CNRS, Laboratoire de Biogenèse Membranaire, UMR 5200, France
  8. Centre National de la Recherche Scientifique, Université de Strasbourg, Institut des Neurosciences Cellulaires et Intégratives, France
  9. Plateforme Imagerie In Vitro, CNRS UPS 3156, France
  10. IGBMC Imaging Center CNRS (UMR7104)/ INSERM (U1258)/ Université de Strasbourg, France
  11. Team SOAP, CRCINA, INSERM, CNRS, Université de Nantes, Université d’Angers, France
  12. Integrated Center for Oncology, ICO, France
  13. Nanotranslational laboratory, Institut de Cancérologie Strasbourg Europe, France
  14. Équipe de synthèse pour l’analyse (SynPA), Institut Pluridisciplinaire Hubert Curien (IPHC), UMR7178, CNRS/Université de Strasbourg, France
  15. Vincent's Clinical School, Faculty of Medicine, University of New South Wales, Australia
  16. The Kinghorn Cancer Centre, Garvan Institute of Medical Research, Australia
  17. C2VN, INSERM 1263, Inrae 1260, Aix-Marseille Université, France
  18. CNRS SNC5055, France
Research Article
Cite this article as: eLife 2021;10:e61539 doi: 10.7554/eLife.61539
8 figures and 2 additional files

Figures

Figure 1 with 2 supplements
RalA and RalB control exosome secretion and multi-vesicular body (MVB) homeostasis.

(a–b) Nanoparticle tracking analysis of extracellular vesicles (EVs) isolated by ultracentrifugation (100,000 g pellet) from the supernatant of shCtl, shRalA, or shRalB 4T1 cells (a) or from various cell types treated with Ral inhibitors RBC8 (b, left) or BQU57 (b, right). 231: MDA-MB-231 cells. Each dot represents one experiment (a: 10 independent experiments; One-Way Anova followed by Bonferroni's Multiple Comparison Test; b: four to five independent experiments, Mann Whitney test). (c) Representative confocal images of 4T1 cells showing endogenous expression of RalA, RalB, and CD63 by immunofluorescence (left) and overexpression of GFP-RalA and GFP-RalB in cells incubated with Lysotracker (right). Scale bar: 10 μm; zoom: 2 μm. (d) Representative electron micrographs of 4T1 shCtl, shRalA and shRalB cells, with zoom on MVBs; Scale bar: 1 μm; zoom: 200 nm. Violin plots show quantification of the number of MVB per cytoplasm surface. Each dot represents one field of view; horizontal bars represent the average (76–88 fields of view; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test).

Figure 1—figure supplement 1
Ral knockdown efficiency and impact on EV secretion.

(a) Representative western blots showing tubulin, RalA (left) and RalB (right) expressions in 4T1 shControl, shRalA, and shRalB cells. (b) Graph showing the average diameter of the extracellular vesicles (EVs) isolated from 4T1 shControl, shRalA, and shRalB cells measured by nanoparticle-tracking analysis. Each dot represents one experiment (12 independent experiments; One-Way Anova followed by Bonferroni's Multiple Comparison Test). (c) Nanoparticle-tracking analysis of EVs isolated by ultracentrifugation (100,000 g pellet) from the supernatant of various breast cancer cell lines treated with the Ral inhibitor RBC8. Each dot represents one experiment (four independent experiments, Mann Whitney test, p value indicated on the graph).

Figure 1—figure supplement 2
Electron microscopy analysis of endosomes in the absence of RalA or RalB (a-b).

Graph showing the number of multi-vesicular body (MVB) (a) or endolysosomes (b) per cytoplasm surface in electron microscopy analysis performed by chemical fixation or high-pressure freezing on 4T1 shControl, shRalA, and shRalB cells. Each dot represents one field of view. Kruskal-Wallis test followed by Dunn's Multiple Comparison Test. (c) FACS analysis of 4T1 shControl, shRalA, and shRalB cells stained with Lysotracker. Each dot represents one experiment (five independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (d) Representative electron micrographs of MVBs in 4T1 shControl, shRalA, and shRalB cells after chemical fixation or high-pressure freezing. Scale bar: 200 nm. (e–f) Number of ILV per MVB surface (e) and MVB diameter (f) measured in electron microscopy analysis performed by chemical fixation or high-pressure freezing on 4T1 shControl, shRalA, and shRalB cells. Each dot represents one MVB; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test.

Figure 2 with 1 supplement
The RalA/B-PLD1-PA axis governs exosome secretion.

(a) Representative confocal images of 4T1 cells co-transfected with PLD1-GFP and tdTomato-RalA (upper panels) or tdTomato-RalB (lower panels) and incubated with Lysotracker. Scale bar: 10 μm; zoom: 2 μm. (b) Electron microscopy analysis of 4T1 cells treated with PLD1 or PLD2 inhibitor. Scale bar: 1 μm. Violin plots show quantification of the number of multi-vesicular body (MVB) per cytoplasmic surface. Each dot represents one field of view; horizontal bar represents the average (180–194 fields of view; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (c) Nanoparticle tracking analysis of extracellular vesicles (EVs) isolated by ultracentrifugation (100,000 g pellet) from the supernatant of 4T1 cells treated with PLD1 (CAY10593) or PLD2 (CAY10594) inhibitor. Each dot represents one experiment (three independent experiments; One-Way Anova permutation test followed by fdr multi-comparison permutation test). (d) Representative confocal images of shControl, shRalA and shRalB 4T1 cells transfected with PLD1-GFP. Scale bar: 10 μm; zoom: 2 μm. Graph shows the percentage of cells with high (>5) number of PLD1-GFP cytoplasmic puncta. (Each dot represents one experiment. Five independent experiments; Number of cells analyzed: shCtl (136), shRalA (170), shRalB (244); Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (e) Quantification of the Phosphatidic Acid (PA) / PhosphatidylCholine (PC) ratio in EVs isolated from shControl, shRalA, and shRalB cells (each dot represents one experiment; three independent experiments; One-Way Anova permutation test followed by fdr multi-comparison permutation test; fdr <0.1). (f) Model showing how RalA and RalB could control PLD1 localization on MVBs, thereby inducing the PA accumulation on MVBs, promoting MVB homeostasis and controlling exosome secretion.

Figure 2—figure supplement 1
PLD1 and PLD2 in 4T1 cells.

(a) PLD1 and PLD2 expression in 4T1 shControl cells quantified by qRT-PCR and normalized by gapdh expression. (b) Representative confocal images showing PLD1-GFP and PLD2-GFP subcellular localization in 4T1 cells stained with lysotracker. Scale bars: 10 μm (c) Representative confocal images of PLD2-GFP localization in shControl, shRalA, and shRalB cells. Scale bar: 10 μm (d) PA/PC ratio of species known to be targeted by PLD1 identified in extracellular vesicles (EVs) isolated from shControl, shRalA, and shRalB cells. Each dot represents one experiment; three independent experiments.

Figure 3 with 2 supplements
RalA and RalB promote lung metastasis in a non-cell autonomous fashion.

(a) Kaplan-Meier curve, obtained from TCGA 1097 cohort, showing the survival probability of patients with tumor breast invasive carcinoma having high or low RalA (pvalue: 5.15 e-03; pAdj: 1.35e-01) or RalB (pvalue: 1.77 e-05; pAdj: 5.99e-03) expression levels. (b) Representative images of immunohistochemistry against RalA or RalB performed on mammary primary tumors from patients with or without metastasis. Scale bar: 500 μm Graphs represent automated scoring of DAB staining. Each dot represents one patient; 10 patients per group; Student t-test. (c) Orthotopic injection of shControl, shRalA, and shRalB 4T1 cells in syngenic mice. Representative images of primary tumors at day 41. Scale bar: 1 cm. Graphs showing the primary tumor growth over time (Left) and the primary tumor weight at day 41. Each dot represents one mouse. (Two independent experiments; Left: Two-way Anova followed by Bonferonni post-test, Right: Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (d) Representative images of primary tumors stained with anti-Ki67 antibody. Scale bar: 50 μm. Graph indicates the % of Ki67-positive nuclei. Each dot represents one mouse. (six mice taken from two independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test. (e) Analysis of lung metastasis in mice from the orthotopic experiment presented in (c). Representative images of lung sections (Day 41) stained with hematoxilin eosin. Scale bar: 1 mm. Graphs show the number of metastatic foci per section (upper, One-Way Anova followed by Bonferroni's Multiple Comparison Test) and the metastatic surface per lung surface (lower; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). Each dot represents one section f) Pictures of wound healing closure at different time points. Scale bar: 150 μm. Graph represents the percentage of wound closure at 16 hr (three independent experiments; Kruskal-wallis test followed by Dunn’s multiple comparison test). (g) Pictures of 3D invasion assay after 15 days. Graph represents the invasive index. Scale bar: 100 μm.

Figure 3—figure supplement 1
Proliferation and apoptosis of 4T1 cells and tumors.

(a) Graph showing the percentage of cells expressing caspase 3 assessed by immunofluorescence on breast primary tumors. Each dot represents one mouse (eight mice taken from two independent experiments; One-Way Anova followed by Bonferroni's Multiple Comparison Test). (b) Graph showing the caspase3/7 expression levels on protein extracts from breast primary tumors. Each dot represents (12–19 mice from two independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test) (c) Graph showing the proliferation of 4T1 shControl, shRalA, and shRalB cells over time in an in vitro assay Two-way ANOVA followed by Bonferroni's Multiple Comparison Test (three independent experiments with n = 5). (d) Graph showing the cycle analysis of 4T1 shControl, shRalA, and shRalB cells. (e) Graph showing the effect of high and low doses of 4T1 extracellular vesicles (EVs) on permeabilization of a HUVEC monolayer in vitro. The graph represents the normalized amount of fluorescent dextran that crossed the endothelial barrier. Each dot represents one experiment (three independent experiments; One-Way Anova permutation test followed with fdr multicomparison test).

Figure 3—figure supplement 2
Soluble secretome of 4T1 shControl cells compared to 4T1 shRalA or 4T1 shRalB cells (three independent experiments; One-Way Anova permutation test followed with pairwise permutation test with fdr correction).
Figure 4 with 1 supplement
RalA and RalB control lung tropism of pro-metastatic tumor extracellular vesicles (EVs).

(a) Effect of a similar amount of EVs on HUVEC monolayer permeability in vitro. The graph represents the normalized amount of fluorescent dextran that crossed the endothelial barrier. Each dot represents one experiment (eight independent experiments; One-Way Anova followed by Bonferroni's Multiple Comparison Test). (b) Representative epifluorescence images of VE-cadherin (upper panels) and ZO1 (Lower panel) stainings on HUVECS cells treated with similar amounts of EVs. Scale bar: 20 μm; zoom: 2 μm. Graphs represent the disorganization of adherent (up) and tight (low) junctions (Three independent experiments; up; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (c) Metastasis priming experiment, Balb/c mice are first injected twice with tumor equal number of EVs (1.5 × 108 EVs), then intravenously with 4T1 luciferase cells and metastasis is then followed over time. Graph shows metastasis progression over time in mice pre-injected with PBS, or with equal number of EVs from shControl, shRalA or shRalB cells (7–10 mice per group; merge of two independent experiments; Two-way Anova followed by Bonferonni multiple comparison post test; stars indicate statistically significant differences at day 14). Right: In vivo and ex vivo representative images of mice and lungs at day 14. Scale bars: 1 cm. (d–e) Lung accumulation of equal number of fluorescent-labeled EVs (3 108 EVs), from shControl, shRalA or shRalB cells injected intravenously. (d) Representative ex vivo images and graph showing the total lung fluorescence 1 hr post-injection. Each dot represents one mouse. (Eight mice taken from two independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test.) (e) Representative confocal lung sections images and graph showing the percentage of EVs-positive fields. Each dot represents one section (three mice; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). Scale bar: 5 μm. (f–g) Arrest and internalization of equal number of EVs from shControl, shRalA, and shRalB cells on endothelial cells in vitro and in vivo. (f) Representative confocal Z-stacks of equal number of EVs after 1 hr or incubation with HUVEC monolayer. Scale bar: 25 μm. Each dot represents one field of view (each dot represents one field of view from three independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (g) Representative confocal Z-stacks the caudal plexus of Tg(Fli1:GFP) zebrafish embryos, where GFP is expressed in the endothelium, injected with similar number of EVs and imaged right after injection. Each dot represents one zebrafish (31–53 embryos from four independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). Scale bar: 20 μm.

Figure 4—figure supplement 1
4T1 extracellular vesicles (EVs) organotropism.

(a) Organs (left) and lungs sections (right) harvested from mice injected with PBS-MB or 4T1 MB-EVs 1 hr after intravenous injection. Scale bars: 10 μm. (b) Representative confocal images showing the identity of lung cells accumulating fluorescently labeled 4T1 EVs. Scale bars: 10 μm. Graphs show the percentage of endothelial cells, fibroblasts, macrophages and cells of unknown identity among EV-positive cells (108 positive cells from six mice taken from two independent experiments). (c) Graph showing liver accumulation of fluorescent EVs isolated from shControl, shRalA, and shRalB cells. Equal numbers of EVs were injected intravenously 1 hr before organ harvesting (eight mice taken from two independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test).

Figure 5 with 2 supplements
CD146/MCAM is under-expressed in RalA/B knockdown extracellular vesicles (EVs) and mediates their lung tropism.

(a) Venn diagram representing the RNA present in the EVs isolated from shControl, shRalA or shRalB cells (with a minimum of 10 reads per sample; RNA sequencing performed in triplicate). (b) Type of RNA associated identified in EVs isolated from shControl, shRalA, or shRalB cells. Left: RNA exclusively present in one type of EVs. Right: enriched RNAs (log2 fold change >2; p(adj.)<0,05). (c) GO terms of the proteins identified in EVs isolated from 4T1 cells by ultracentrifugation (100,000 g pellet) and illustration of some proteins known to be present in EVs. (d) Comparison of the protein content of EVs isolated from shControl, shRalA, and shRalB cells. The venn diagram represents proteins having different expression levels (Mass spectrometry performed in triplicate; FDR < 1%). (e) Analysis of the expression of CD146/MCAM, Clic4, and Glypican4 in EVs isolated from shControl, shRalA and shRalB cells by western blots. Each dot represents one experiment (four to six independent experiments; Kruskal-Wallis test followed by Dunn's Multiple Comparison Test). (f–g) Arrest, internalization, and organotropism of EVs treated with an anti-CD146 antibody and injected in the circulation of zebrafish embryos (f) or mouse (g). (f) Representative confocal Z-stacks the caudal plexus of Tg(Fli1:GFP) zebrafish embryos, where GFP is expressed in the endothelium, injected with equal number of EVs and imaged right after injection. Scale bar: 20 μm; Zoom scale bar: 5 μm. Each dot represents one zebrafish (46 embryos from four independent experiments; Mann Whitney test). (g) Representative confocal images of lung sections and graph showing the percentage of EVs-positive fields. Scale bar: 10 μm. Each dot represents one mouse (eight mice from two independent experiments; Mann Whitney test). (h) Metastasis priming experiment, Balb/c mice are injected twice with tumor equal number of EVs (1.5 × 108 EVs), pre-incubated with CD146 blocking antibody or isotype control, and then intravenously injected with 4T1 luciferase cells and metastasis is followed over time. Graph shows metastasis progression over time (14 mice per group; merge of two independent experiments; Two-Way Anova followed by Bonferonni multiple comparison post-test; stars indicate statistically significant differences at day 14). In vivo and ex vivo representative images of mice and lungs at day 14. Scale bars: 1 cm. (i) Kaplan-Meier curve, obtained from TCGA 1097 cohort, showing the survival probability of patients with tumor breast invasive carcinoma having high or low MCAM/CD146 expression levels (pvalue: 3.42 e-02; pAdj: 5.67e-01). (j) Model describing the role of RalA/B-dependent EVs in metastatic formation.

Figure 5—figure supplement 1
RNA content of extracellular vesicles (EVs) from shControl, shRalA, and shRalB cells.

(a) Volcano plots comparing the mRNA present in EVs isolated from shControl and shRalA cells. (b) Volcano plots comparing the mRNA present in EVs isolated from shControl and shRalB cells. (c) GO term analysis of the mRNA present in EVs from shControl, shRalA, and shRalB cells.

Figure 5—figure supplement 2
4T1 cells and extracellular vesicles (EVs) express CD146/MCAM long isoform.

(a) Western blots on 4T1 cell and EVs extracts using antibodies selectively recognizing CD146/MCAM short (left) and long (right) isoforms. (b) Anti-CD146 Elisa on EVs isolated from shControl, shRalA, and shRalB cells. Each dot represents one replicate. Horizontal line represents the median. (c) Representative confocal images of 4T1 cells showing endogenous expression of CD146 and CD63 by immunofluorescence. Scale bars: 10 μm; zoom: 2 μm.

Author response image 1
RAL expression.

RalA and RalB expression assessed by western blots (Left in multiple cell types and quantified (Right). Each dot represents one experiment.

Author response image 2
Effects of PLD doses on EV secretion.

Each experiment was repeated three times on 4T1 cells.

Author response image 3
Segmentation of TCGA data.

Additional files

Supplementary file 1

EVs RNA analysis.

Sheet a: RNAs overexpressed in EVs from 4T1 shCtl cells Vs EVs from shRalA cells Sheet b: RNAs overexpressed in EVs from 4T1 shRalA cells Vs EVs from shCtl cells Sheet c: RNAs overexpressed in EVs from 4T1 shCtl cells Vs EVs from shRalB cells Sheet d: RNAs overexpressed in EVs from 4T1 shRalB cells Vs EVs from shCtl cells

https://cdn.elifesciences.org/articles/61539/elife-61539-supp1-v2.xlsx
Supplementary file 2

EVs proteomic analysis.

Sheet a: Proteins identified in EVs from 4T1 shCtl cells Sheet b: Proteins overexpressed in EVs from 4T1 shCtl cells Vs EVs from shRalA cells Sheet c: Proteins overexpressed in EVs from 4T1 shRalA cells Vs EVs from shCtl cells Sheet d: Proteins overexpressed in EVs from 4T1 shCtl cells Vs EVs from shRalB cells Sheet e: Proteins overexpressed in EVs from 4T1 shRalB cells Vs EVs from shCtl cells Sheet f: Proteins overexpressed in EVs from 4T1 shCtl cells Vs EVs from shRalA cells and EVs from shRalB cells

https://cdn.elifesciences.org/articles/61539/elife-61539-supp2-v2.xlsx

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)