Protrudin-mediated ER-endosome contact sites promote MT1-MMP exocytosis and cell invasion

Abstract

Cancer cells break tissue barriers by use of small actin-rich membrane protrusions called invadopodia. Complete invadopodia maturation depends on protrusion outgrowth and the targeted delivery of the matrix metalloproteinase MT1-MMP via endosomal transport by mechanisms that are not known. Here, we show that the ER protein Protrudin orchestrates invadopodia maturation and function. Protrudin formed contact sites with MT1-MMP-positive endosomes that contained the RAB7-binding Kinesin-1 adaptor FYCO1, and depletion of RAB7, FYCO1, or Protrudin inhibited MT1-MMP-dependent extracellular matrix degradation and cancer cell invasion by preventing anterograde translocation and exocytosis of MT1-MMP. Moreover, when endosome translocation or exocytosis was inhibited by depletion of Protrudin or Synaptotagmin VII, respectively, invadopodia were unable to expand and elongate. Conversely, when Protrudin was overexpressed, noncancerous cells developed prominent invadopodia-like protrusions and showed increased matrix degradation and invasion. Thus, Protrudin-mediated ER-endosome contact sites promote cell invasion by facilitating translocation of MT1-MMP-laden endosomes to the plasma membrane, enabling both invadopodia outgrowth and MT1-MMP exocytosis.

Figure 1.

FYCO1-positive LE/Lys localize to invadopodia. (A) MDA-MB-231 cells were grown on coverslips, stained with antibodies against TKS5 and cortactin, and analyzed by confocal microscopy. Phalloidin/Alexa Fluor 647 was used to detect F-actin. A section from a confocal z-stack shows invadopodia. Orthographic sections show invadopodia on the ventral side of the cell. (B) MDA-MB-231 cells were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with anti-TKS5 and phalloidin/Alexa Fluor 647 (actin) to visualize invadopodia, and analyzed by confocal microscopy. A section from a confocal z-stack shows invadopodia correlating with degraded gelatin (black areas). 3D view and orthographic sections of the same cell are shown. (C) Colocalization of FYCO1 with LE/Lys markers in MDA-MB-231 cells. Cells were grown on coverslips; stained with antibodies against FYCO1, RAB7, or LAMP1; and analyzed by confocal microscopy. (D) MDA-MB-231 cells were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with antibodies against TKS5 and FYCO1, and analyzed by superresolution microscopy (Airyscan). Z-stacks and Imaris surface 3D renderings from two independent cells show FYCO1-positive LE/Lys in close apposition to TKS5-positive invadopodia correlating with degraded gelatin. Data are representative of at least 16 captures.

Figure S1.

Characterization of MDA-MB-231 cells stably expressing GFP-Protrudin. (A) MDA-MB-231 cells stably expressing GFP-Protrudin siRNA resistant to oligo #1 (sires#1) was generated by lentiviral transduction. MDA-MB-231 and MDA-MB-231-GFP-Protrudin sires#1 cells were transfected with two different siRNAs targeting Protrudin (#1 and #2) or control siRNA for 4 d and subjected for Western blotting to detect protein levels of Protrudin, GFP-Protrudin, and actin (loading control) using anti-Protrudin and anti-actin antibodies, respectively. (B) Cells transfected as in A were seeded on coverslips, immunostained for LAMP1, and analyzed by confocal microscopy. Note that LAMP1-positive LE/Lys cluster perinuclearly in Protrudin-knockdown cells as expected, but not in Protrudin/GFP-Protrudin–positive cells, verifying the functionality of GFP-Protrudin.

Figure S2.

Invadopodia reformation is impaired in Protrudin-depleted cells. (A) MDA-MB-231 and MDA-MB-231-GFP-Protrudin-sires#1 cells were transfected with siRNA targeting Protrudin (oligo #1) or control siRNA. 4 d after transfection, cells were serum starved for 4 h and treated with Src inhibitor (10 µM PP2) for the last 30 min to remove invadopodia. Cells were stimulated for 1 h with serum containing medium supplemented with HGF (50 ng/ml) to allow reformation of invadopodia or serum free medium (SFM) as a negative control. Cells were stained with antibodies against TKS5 and analyzed by high-content microscopy. Note that TKS5-positive invadopodia only reform in the serum and HGF-treated cells and that cells expressing siRNA-resistant GFP-Protrudin reform invadopodia comparable to control cells. Graphs represent quantifications of different features of invadopodia reformation. Each plotted point represents the mean value of one experiment. Values represent mean ± SD. *, P < 0.05, one-way ANOVA, Tukey’s post hoc test. n = 3 independent experiments. In total, >2,800 cells were analyzed per condition. (B) Graphs represent quantifications of different features of invadopodia reformation from Fig. 2 B. Each plotted point symbolizes one image representing the average value of typically 10–15 cells. Values represent mean ± SD. *, P < 0.05; ***, P < 0.001, one-way ANOVA, Tukey’s post hoc test. n = 25 images per condition from three independent experiments. (C) Real-time PCR for verification of SYT7 knockdown from three independent experiments in MDA-MB-231 cells. **, P < 0.01, one-sample t test.

Figure 2.

Protrudin-depleted cells fail to regrow invadopodia. (A) MDA-MB-231 and MDA-MB-231-GFP-Protrudin-sires#1 cells grown on coverslips were transfected with siRNA targeting Protrudin (oligo #1) or control siRNA. 4 d after transfection, cells were serum starved for 4 h and treated with Src inhibitor (10 µM PP2) for the last 30 min to remove invadopodia. Cells were stimulated for 1 h with serum containing medium supplemented with HGF (50 ng/ml) to allow reformation of invadopodia or SFM as a negative control. Cells were stained with antibodies against TKS5 and cortactin and analyzed by confocal microscopy. Micrographs show reformation of TKS5-positive invadopodia in serum- and HGF-treated cells. Note that cells expressing siRNA-resistant GFP-Protrudin reform invadopodia as in control cells. Graphs represent quantifications of different features of invadopodia reformation. Each plotted point symbolizes one image representing the average value of typically 15–20 cells. Values represent mean ± SD. *, P < 0.05; ***, P < 0.001, one-way ANOVA, Tukey’s post hoc test. n = 9 (serum starvation) or n = 15 images per condition from three independent experiments. (B) MDA-MB-231 and MDA-MB-231-GFP-Protrudin-sires#1 cells were siRNA transfected with siRNA targeting Protrudin (oligo #1) or control siRNA. 4 d after transfection, the cells were seeded on unconjugated-gelatin coated coverslips in SFM for 2 h to allow attachment before treating with Src inhibitor (10 µM PP2) for 30 min. Serum containing medium with HGF (50 ng/ml) was added and cells were allowed to reform invadopodia for 4 h, immunostaining and confocal imaging as in A. Note that cells expressing siRNA-resistant GFP-Protrudin reform invadopodia as in control cells (GFP not shown). For quantifications of micrographs, see Fig. S2 B. n.s, not statistically significant.

Figure 3.

Protrudin-KO cells display small and short invadopodia. (A) Cell lines as indicated in the figure were grown on coverslips; stained with anti-TKS5, anti-GFP, and phalloidin/Alexa Fluor 647 (actin); and analyzed by confocal microscopy. Note the undersized invadopodia in cells lacking Protrudin. (B) Cell lines as indicated in the figure were grown on coverslips, stained with anti-TKS5, anti-GFP and phalloidin/Alexa Fluor 647 (actin) and analyzed by confocal microscopy. Micrographs show orthographic and parallel maximum projections of z-stacks. In the parallel projections, the fluorescent intensity of TKS5 and actin were measured along a dotted line overlaying two invadopodia in each cell line, showing a reduced intensity of bothTKS5 and actin in cells lacking Protrudin as indicated in the intensity plots. (C) MDA-MB-231 and MDA-MB-231-Protrudin KO#2 cells were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with anti-TKS5 and phalloidin/Alexa Fluor 647 (actin), and examined by superresolution microscopy (Airyscan). Z-stacks of individual cells were captured. Micrographs show section #9 from the parallel plane, and orthographic sections of the areas indicated by the dotted line. The graph shows the average length of TKS5-positive invadopodia per cell, quantified from Airyscan confocal z-stacks as described in Materials and methods. Each plotted point represents one cell, mean ± SD; parental, n = 19 cells; Protrudin KO, n = 15 cells from three independent experiments. ***, P < 0.0001, unpaired two-sided t test.

Figure S3.

Characterization of Protrudin-KO cell lines with or without stable expression of GFP-Protrudin, antibody validation, and verification of siRNA-mediated protein depletion. (A) KO of Protrudin in MDA-MB-231 cells was performed by CRISPR/Cas9-mediated genome editing. Stable expression of GFP-Protrudin in two KO clones was generated by lentiviral transduction. The indicated cell lines were either seeded on coverslips for confocal microscopy or lysed and subjected for Western blotting to detect protein expression levels of Protrudin. Cells grown on coverslips were immunostained for LAMP1. For Western blotting, anti-Protrudin and anti-actin (loading control) were used. Note that LAMP1-positive LE/Lys cluster perinuclearly in Protrudin-KO cells as expected, but not in Protrudin/GFP-Protrudin positive cells, verifying the functionality of GFP-Protrudin. (B) MDA-MB-231 and MDA-MB-231-Protrudin-KO#2 cells were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with anti-TKS5 and phalloidin/Alexa Fluor 647 (actin), and examined by confocal microscopy (Airyscan; same dataset as in Fig. 3 C). The number of TKS5-positive invadopodia per cell was quantified. Each plotted point represents one cell, and values represent mean ± SD. Parental, n = 19 cells; Protrudin KO#2, n = 15 cells from three independent experiments, unpaired two-sided t test. (C) Verification of MT1-MMP antibody specificity by confocal microscopy and Western blotting. MDA-MB-231 cells were transfected with siRNA targeting MT1-MMP or control siRNA for 3 d. Cells were either lysed and subjected for Western blotting or seeded on coverslips and stained with anti-MT1-MMP and Alexa Fluor 488/Phalloidin (actin). Note that the anti-MT1-MMP signal is virtually gone in knockdown cells. (D) Western blots verifying knockdown of Protrudin, TKS5, RAB7, or FYCO1 in MDA-MB-231 cells using lysates with the indicated siRNA-transfected cells and the respective siRNA control treatments. Actin or vinculin was used as a loading control.

Figure 4.

Invadopodia formation is stimulated by Protrudin overexpression and requires SYT7-dependent membrane fusion. (A) RPE-1 cells with or without stable overexpression (OE) of Protrudin were grown on coverslips coated with Oregon Green gelatin for 4 h in serum containing medium and stained with anti-TKS5 and phalloidin/Alexa Fluor 647 (actin). Confocal micrographs show TKS5- and actin-positive invadopodia correlating with degraded gelatin. The graphs show the amount of cells with invadopodia and the size of invadopodia. Each plotted point represents the average of one independent experiment. Shown is mean ± SEM, n = 3. Total number of cells analyzed per condition: percentage of cells with invadopodia, >200 cells; invadopodia size, >45 cells. *, P < 0.05; **, P < 0.01, unpaired two sided t test. The Western blot shows the expression level of Protrudin in the two cell lines. (B) MDA-MB-231 cells grown on coverslips were transfected with siRNA targeting SYT7 or control siRNA. 1 d after transfection, cells were serum starved for 4 h and treated with Src inhibitor (10 µM PP2) for the last 30 min to remove invadopodia. Cells were stimulated for 1 h with serum containing medium supplemented with HGF (50 ng/ml) to allow reformation of invadopodia. Cells were stained with antibodies against TKS5 and cortactin and analyzed by confocal microscopy. Micrographs show reformation of TKS5-positive invadopodia. Graphs represent quantifications of different features of invadopodia reformation. Each plotted point symbolizes one image representing the average value of typically 15 cells. Values represent mean ± SD. ***, P < 0.001, unpaired two-sided t test. n = 15 images per condition from three independent experiments.

Figure 5.

Protrudin forms contact sites with MT1-MMP–positive LE/Lys and regulates their intracellular localization. (A) Confocal micrographs of MDA-MB-231 and RPE-1 cells showing that endogenous MT1-MMP localizes to RAB7-positive endosomes. Arrows show colocalization of MT1-MMP and RAB7 in RPE-1 cells. Manders overlap coefficient shows that 56% (±12 SD) of MT1-MMP positive pixels overlap with Rab7-positive pixels in MDA-MB-321 cells (n = 15 cells), and 33% (±8 SD) in RPE-1 cells (n = 13 cells). (B) Confocal micrographs of MDA-MB-231-GFP-Protrudin cells showing contact sites between GFP-Protrudin and endogenous MT1-MMP. Graphs show three examples of contact sites between GFP-Protrudin in the ER and MT1-MMP–positive endosomes visualized by fluorescence intensity profiles. (C) Still image from Video 1 of RPE-1-GFP-Protrudin-MT1-MMP-pHuji cells showing that GFP-Protrudin forms contact sites with MT1-MMP-pHuji–positive vesicles. (D) MDA-MB-231 and MDA-MB-231-Protrudin KO#2 cells were grown on coverslips, immunostained for MT1-MMP and TKS5, and analyzed by confocal microscopy. Z-stacks and Imaris surface 3D renderings show the subcellular localization of MT1-MMP in relation to TKS5-positive invadopodia. Note the perinuclear clustering of MT1-MMP and the undersized TKS5-positive invadopodia in Protrudin-KO#2 cells. (E) MDA-MB-231 and MDA-MB-231-Protrudin-KO cells were grown on coverslips, immunostained for MT1-MMP, and analyzed by confocal microscopy. Images and graph show the perinuclear clustering of MT1-MMP–positive endosomes in Protrudin-KO cells. Each plotted point represents one cell. Values represent mean ± SD. Number of cells analyzed: parental, n = 30; KO#1, n = 20; KO#2, n = 28 from three independent experiments. ***, P < 0.001, Kruskal-Wallis, Dunn’s post hoc test. Note that the big variation in MT1-MMP localization seen between parental cells is diminished in Protrudin-KO cells where MT1-MMP vesicles mainly cluster perinuclearly.

Figure 6.

MT1-MMP exocytosis depends on Protrudin. (A) RPE-1-GFP-Protrudin sires#1-MT1-MMP-pHuji cells were seeded in MatTek dishes and transfected with siRNA against Protrudin (oligo #1 or #2) or control siRNA. 2 d after transfection, cells were imaged twice per second for 2 min by TIRF microscopy. Shown are representative TIRF micrographs where individual exocytic events (summed over a 2-min interval) are indicated by a white dot. Widefield micrographs and Western blot analysis show the expression level of GFP-Protrudin and knockdown efficiencies. The graph shows the number of MT1-MMP exocytic events per cell area per minute. For each condition, n = 18 movies were analyzed from three independent experiments. Boxplot whiskers show minimum to maximum. *, P < 0.05; ***, P < 0.001, one-way ANOVA, Tukey’s post hoc test. Note that siProtrudin#2 depletes both endogenous and exogenous Protrudin and reduces the amount of MT1-MMP exocytosis, whereas siProtrudin#1 only depletes the endogenous pool of Protrudin, leaving GFP-Protrudin (sires#1) to maintain MT1-MMP exocytosis. (B) MDA-MB-231-MT1-MMP-pHuji-TKS5-GFP cells were seeded in MatTek dishes and transfected with siRNA against Protrudin (oligo #1) or control siRNA. Western blots show protein expression levels and verify the Protrudin knockdown in the stable cell line. 4 d after transfection, cells were imaged by TIRF microscopy. Shown are representative TIRF micrographs of MT1-MMP-pHuji surface accumulation. Widefield micrographs show TKS5-GFP–positive invadopodia. Plasma membrane exposure of MT1-MMP-pHuji was quantified by segmenting bright pHuji spots from the TIRF images (mask). Graphs show whiskers (minimum to maximum) from n = 30 (siProtrudin) and n = 33 (siControl) TIRF images. ***, P < 0.001, Mann–Whitney test.

Figure 7.

Protrudin promotes degradation of extracellular gelatin in a dose-dependent fashion. (A) MDA-MB-231 cells were transfected with siRNA targeting Protrudin, FYCO1, RAB7, TKS5, or control siRNA. The cells were grown on coverslips coated with Oregon Green gelatin for 4 h and analyzed by confocal microscopy. Representative micrographs show degradation of fluorescent gelatin indicated by black areas. (B) Quantifications of images in A. The graph shows the relative area of gelatin degradation in the different knockdown conditions compared with siControl. Each individual data point represents the average of one independent experiment. Shown is mean ± SEM, n = 3 experiments (Protrudin, FYCO1#1, RAB7, and TKS5), n = 4 experiments FYCO1#2. **, P < 0.01; ***, P < 0.001, one-sample t test. Number of cells in total: siControl(Protrudin/TKS5), 1,145; siProtrudin#1, 800; siTKS5, 633; siControl(FYCO1#1), 1,064; siFYCO1#1, 975; siControl(FYCO1#2), 893; siFYCO1#2, 1,007; siControl(RAB7), 577; siRAB7, 512. (C) MDA-MB-231 and MDA-MB-231-GFP-Protrudin-sires#1 cells were transfected with siRNA targeting Protrudin (oligo #1 or #2) or control siRNA. 4 d after transfection, cells were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with Rhodamine-Phalloidin (actin), and analyzed by confocal microscopy (knockdown verification in Fig. S1 A). Representative micrographs show degradation of fluorescent gelatin indicated by black areas. (D) Quantifications of images in C. The graph shows area of gelatin degradation in the different knockdown conditions compared with siControl. Note that the reduced gelatin degradation in Protrudin-depleted cells is rescued in the presence of GFP-Protrudin. Each individual data point represents the average of one independent experiment. Values represent mean ±SEM, n = 4. ***, P < 0.001, one-way ANOVA, Tukey’s post hoc test. In total, >700 cells were analyzed per condition. (E) RPE-1 cells or RPE-1-Protrudin-overexpressing cells were grown on coverslips coated with Oregon Green gelatin for 6 h with addition of HGF (50 µm) for the last 2 h, stained with Rhodamine-Phalloidin (actin), and analyzed by confocal microscopy. Representative micrographs show degradation of fluorescent gelatin indicated by black areas. (F) Quantification of images in E. The graph shows relative area of gelatin degradation. Each individual data point represents the average of one independent experiment. Values represent mean ± SEM, n = 5. **, P < 0.01, one sample t test. In total, >1,600 cells were analyzed per condition.

Figure S4.

GFP-Protrudin rescues the loss of gelatin degradation in Protrudin-KO cells. (A) MDA-MB-231 parental or MDA-MB-231-Protrudin-KO (KO#1 and KO#2) cells were grown on coverslips coated with Oregon Green gelatin for 4 h in the absence or presence of the MMP inhibitor GM6001 (20 µM; characterization of cell lines in Fig. S3 A). Confocal micrographs show degradation of the fluorescent gelatin indicated by black areas. (B) Quantifications of experiments displayed in A. The graph shows the relative area of gelatin degradation. Each individual data point represents the average of one independent experiment. Values represent mean ± SD, n = 4 experiments. **, P < 0.01; ***, P < 0.001, one-sample t test. Number of cells in total: parental, 1,156; Protrudin KO#1, 1,146; Protrudin KO#2, 1,174. With MMP inhibitor: n = 1; number of cells: parental, 175; Protrudin KO#1, 163; Protrudin KO#2, 206. (C) MDA-MB-231-Protrudin-KO#1 and MDA-MB-231-Protrudin-KO#1-GFP-Protrudin cells were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with Rhodamine-Phalloidin (actin), and analyzed by confocal microscopy (characterization of cell lines in Fig. S3 A). Representative micrographs show degradation of fluorescent gelatin indicated by black areas. (D) Quantification of images in C. The graph shows relative area of gelatin degradation. Each individual data point represents the average of one independent experiment. Values represent mean ± SD, n = 4. ***, P < 0.001, one-sample t test. In total, >900 cells were analyzed per condition. (E) The indicated cell lines were grown on coverslips coated with Oregon Green gelatin for 4 h, stained with Rhodamine-Phalloidin (actin), and analyzed by confocal microscopy (characterization of cell lines in Fig. S3 A). Representative micrographs show degradation of fluorescent gelatin indicated by black areas. (F) Quantification of images in E. The graph shows relative area of gelatin degradation. Each individual data point represents the average of one independent experiment. Values represent mean ± SD, n = 4. **, P < 0.01, one-sample t test. In total, >750 cells were analyzed per condition.

Figure 8.

Protrudin mediates cleavage of collagen-I. (A) MDA-MB-231 cells were embedded in fluorescent type I collagen (cyan, 2.0 mg/ml) for 24 h and stained for the cleaved collagen neo epitope using anti-Col-3/4C. The confocal micrograph shows the presence of collagen cleavage sites lining the collagen-I fibrils indicating the specificity of the antibody. (B) Protrudin KO decreases pericellular collagenolysis in MDA-MB-231 cells. MDA-MB-231, MDA-MB-231-Protrudin KO#1, and MDA-MB-231-Protrudin KO#1-GFP-Protrudin cells were embedded in fluorescent type I collagen (cyan, 2.0 mg/ml) for 24 h and stained with anti-Col-3/4C. Representative micrographs show the sum projections of confocal z-stacks from each condition. The graph shows quantification of the area of pericellular collagenolysis (µm²) from sum projections of wide field images. Each plotted point represent the average of one collagen droplet. n = 7 droplets (10 z-stacks per droplet) were analyzed from five different experiments. Values represent mean ± SEM. *, P < 0.05; ***, P < 0.001, one-way ANOVA, Tukey’s post hoc test. In total, >700 cells were analyzed per condition. (C) RPE-1 cells or RPE-1-Protrudin-overexpressing cells were embedded in fluorescent type I collagen (cyan, 2.0 mg/ml) for 24 h and stained with anti-Col-3/4C and Alexa488-Phalloidin (actin). Representative micrographs show the sum projections of confocal z-stacks from each condition. The graph shows quantification of the relative area of pericellular collagenolysis from sum projections of wide-field images. Each plotted point represents the average of one collagen droplet. n = 7 droplets (10 z-stacks per droplet) were analyzed from three different experiments. Values represent mean ± SEM. **, P < 0.01, one-sample t test. In total, >290 cells were analyzed per condition.

Figure 9.

Protrudin promotes cell invasion. The cell lines indicated were allowed to invade into plugs of fibronectin-supplemented Matrigel or collagen I for 4 d, stained with Calcein, and imaged by confocal microscopy. Optical sections (Δz = 10 µm) are shown for each condition. Graphs show the percentage of cell invasion over 40 µm. Each plotted point represents the average invasion of one plug. (A) In total n = 9 plugs (five z-stacks per plug) were analyzed from three independent experiments. Values represent mean ± SD. ***, P < 0.001, one-way ANOVA, Dunnett’s post hoc test. (B) In total, n = 18 plugs (five z-stacks per plug) were analyzed from six independent experiments. Values represent mean ± SD. ***, P < 0.001, one-sample t test. (C) In total, n = 12 plugs (five z-stacks per plug) were analyzed from four independent experiments. Values represent mean ± SD. *, P < 0.05, one-sample t test. (D) In total n = 9 plugs (5 z-stacks per plug) were analyzed from 3 independent experiments. Values represent mean ± SD. *, P < 0.05, one-sample t test. (E) Single tumor spheroids from RPE-1 and RPE-1-Protrudin–overexpressing cells were embedded in Matrigel and allowed to invade. Phase contrast micrographs show spheroids from days 0 and 2. Graphs represent percent increase of spheroid area and distance of the longest protrusion from the outer rim of the spheroid dense core. Values represent mean ± SD. Each plotted point represents one spheroid. In total, n = 34 spheroids from six different experiments. ***, P < 0.001, Mann–Whitney. OE, overexpression.

Figure S5.

GFP-Protrudin expression rescues cell invasion of Protrudin-KO cells, and the prognostic value of Protrudin in different cancer types. (A) The indicated cell lines were allowed to invade into plugs of fibronectin-supplemented Matrigel for 4 d, stained with Calcein, and imaged by confocal microscopy. Optical sections (Δz = 10 µm) are shown for each condition. Graphs show the relative cell invasion over 20 µm. Each plotted point represents the average invasion of one independent experiment, n = 4. In total, 12 plugs (five z-stacks per plug) were analyzed. Values represent mean ± SD. **, P < 0.01, one-sample t test. (B–D) Survival curves are plotted for breast, gastric, and ovarian cancer using the publicly available database Kaplan-Meier Plotter for high and low expression of Protrudin (ZFYVE27 Affymetrix ID 225281_at). (B) Analysis of lymph node–positive, estrogen receptor–negative, progesterone receptor–negative breast cancer patient cohorts (hazard ratio [HR], 1.88; 95% confidence interval [CI], 1.09–3.24. (C) Analysis of metastasis-free gastric cancer patient cohorts (HR, 1.62; 95% CI, 1.22–2.14). (D) Analysis of ovarian cancer patient cohorts (HR, 1.44; 95% CI, 1.19–1.76).

Figure 10.

Model for the function of the Protrudin pathway in invadopodia formation and exocytosis of MT1-MMP. When the Protrudin pathway is active (right), the ER protein Protrudin makes contact sites with RAB7 and phosphatidylinositol 3-phosphate(PI3P)–positive LE/Lys, which contain MT1-MMP. In the ER–LE/Lys contact sites, the microtubule motor kinesin-1 is handed over from Protrudin to the RAB7-binding kinesin-1 adaptor protein FYCO1. This enables the translocation of the LE/Lys along microtubules toward immature invadopodia at the plasma membrane. SYT7-dependent fusion of LE/Lys with the invadopodial plasma membrane provides membrane for the maturing invadopodia and ensures that MT1-MMP is exposed at the cell surface. This enables invadopodia growth and degradation of the ECM and facilitates cell invasion. Upon depletion of any of the components of the Protrudin pathway (left), MT1-MMP–containing LE/Lys cluster perinuclearly. This prevents MT1-MMP exocytosis, invadopodia maturation, and cell invasion. MTOC, microtubule organizing center.