Focal adhesion (FA) turnover has been demonstrated to play an important role in cell migration; however, the mechanism of FA turnover is complicated and requires further investigation. In this study, Rab11, which is involved in endosome recycling, was examined in terms of a direct regulatory function in FA formation during cell migration. Wild-type and dominant negative (DN) Rab11 or shRab11 were transfected into human HT1080 fibrosarcoma cells; the cell motility and migration abilities were determined, and localization of Rab11 and FA molecules was monitored by confocal microscopy. The results showed that Rab11 deficiency or the DN form inhibited sarcoma cell migration. Rab11 was also found to be co-localized with recycled β1 integrin and affected FA formation. We further employed immunofluorescence and immunoprecipitation to examine the physical interaction between Rab11 and focal adhesion kinase (FAK), and the results suggested that Rab11 affected cell migration by regulating FAK recycling to aid formation of an FA complex on the cell membrane.
Cell migration is involved in embryonic development, inflammation, and wound-healing and has recently been emphasized in cancer metastasis. It is also characterized by cell movement: the cell membrane protrusion extends to form a new focal adhesion (FA) in the cell leading edge, and FA retraction occurs at the rear of the cell, and the cell then moves forward.,
FA is an important structure for cell movement. A FA complex, which includes integrin, a transmembrane receptor that interacts with the extracellular matrix (ECM), delivers an extracellular signal to intracellular focal complex members such as focal adhesion kinase (FAK), Src, paxillin, and vinculin, and further regulates actin rearrangement.,, However, FA turnover is a dynamic process with frequent assembly and disassembly during cell migration. Disassembly of FA has been revealed to act through clathrin-dependent endocytosis and tubulin-mediated disassembly of FAK, and the process is also accompanied by internalization of integrin.,, The internalized integrin undergoes endocytosis and is then either transported to lysosomes for degradation or recycled back to the cell membrane., For vesicle recycling transport, Rab11, a subfamily of Rab GTPase, has been found to play a role in integrin recycling from perinuclear endosomes to the plasma membrane, and the recycling of integrin leads to the formation of new FA at the cell leading edge and promotes cell migration.,,,
FAK is a cytosolic kinase required for FA formation, and the importance of FAK activation has been demonstrated in cancer invasion and metastasis.,, Although the disassembly of FAK from a focal complex has been studied,,, the mechanism of recruitment to the membrane is still not clear. Rab11 expression has been demonstrated to promote cancer formation, such as breast cancer, colon cancer, lung cancer, and prostate cancer.,,, Rab11 is involved in many recycling transport functions, such as integrin, epidermal growth factor receptor (EGFR), and E-cadherin, and regulates signal pathways for cell migration and furthermore collective cell migration.,, The results of this study highlighted the novel role of Rab11 in terms of interaction with FAK and regulation of FAK transport to the FA for cell migration.
Materials and Methods
Cell culture and transfection
Human sarcoma HT1080 cells were cultured in Dulbecco's modified Eagle medium (DMEM) supplemented with 5% penicillin/streptomycin and 10% fetal bovine serum (FBS; GIBCO-BRL, Gaithersberg, MD, USA). The cells were incubated at 37°C with 5% CO2. Lipofectamine 2000 reagent (Invitrogen, CA, USA) was used for the transfection of green fluorescent protein (GFP)-tagged wild-type (WT) Rab11, dominant negative (DN, GDP locked S25N mutant) Rab11 (Addgene, MA, USA), and Rab11 shRNA (RNAi core, Academia Sinica, Taiwan) in HT1080 cells.
Western blotting and immunostaining
HT1080 cells (1.5 × 106) were plated on 10-cm dishes and transfected with GFP-Rab11 WT and GFP-Rab11 DN for 24 h, and Rab11 shRNA for 48 h. Cells were then lysed in radioimmunoprecipitation assay (RIPA) buffer with 25 mM Tris/150 mM NaCl (pH 7.4), including 1% NP-40, 5% glycerol, 0.5 M ethylenediaminetetraacetic acid (EDTA), 100 mM sodium orthovanadate 1:1000, Cocktail 1:50 and 100 mM phenylmethylsulfonyl fluoride (PMSF) 1:500. Lysates were centrifuged for 10 min at 12,300 × g, following which the supernatant was collected, and protein concentrations were measured using a BCA kit (Thermo Scientific, MA, USA). Proteins were separated by 12% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride (PVDF) membranes. The PVDF membranes were blotted with primary antibodies followed by appropriate secondary antibodies. The following primary antibodies were used: GFP (1:1000, Roche, NJ, USA), Rab11 (1:1000, Cell Signaling Technology, MA, USA), and phosphor-FAK (p-FAK) (1:1000, BD Biosciences, NJ, USA); cells were then incubated with an appropriate horseradish peroxidase-conjugated secondary antibody.
Integrin recycling assay
HT1080 cells (0.05 × 106) were plated on coverslips in a 24-well plate and transfected with GFP-Rab11 WT and GFP-Rab11 DN. Starved HT1080 cells were chilled on ice and incubated with active integrin antibody (12G10) (Abcam, MA, USA) for 30 min, then unbound antibodies were removed by washing. Cells were then incubated at 37°C for 30 min for integrin internalization, then incubated in medium-containing FBS, during which integrin underwent degradation or export. Integrin was revealed by immunofluorescence microscopy.
For the migration assay, transfected HT1080 cells (0.05 × 106) were transferred into the insert of a transwell chamber (BD Biosciences) without FBS supplementation, and the chambers were incubated in normal growth DMEM for 2 h. For the invasion assay, Matrigel (Corning, NY, USA) was coated on the transwell membrane and cells were incubated for 6 h. Cells on the top of the insert membrane were removed using a cotton swab, while migrated cells on the bottom of the insert membrane were stained with 4',6-diamidino-2-phenylindole (DAPI).
Live image analysis
For FA tracking, HT1080 cells seeded on coverslips were co-transfected with GFP-FAK plasmid and mCherry-Rab11 and incubated for 24 h. Real-time fluorescent signals were captured using a Zeiss Axio Observer microscope or a Zeiss LSM 700 confocal microscope with an incubation system. Images were scanned every 3 min.
Transfected HT1080 cells were plated on coverslips or on collagen (10 μg/mL) coated coverslips for 24 h. For nocodazole treatment, 10 μM nocodazole was applied for additional 1 h. Cells were fixed with 3.7% formaldehyde for 20 min, then washed three times with phosphate-buffered saline (PBS) and 0.1% Triton-X 100 in PBS for 1 min. Fixed HT1080 cells were incubated with primary antibodies Rab11 (1:100, Cell Signaling Technology), p-FAK (1:150, BD Biosciences), vinculin (1:100, Sigma, MO, USA), paxillin (1:100, Millipore, MA, USA) or Rac1 antibody (1:100, BD Biosciences) for 1 h at room temperature, and after washing three times with PBS, the HT1080 cells were incubated with an appropriate fluorescence-conjugated secondary antibody, Cy2-conjugated anti-rabbit antibody or Cy3-conjugated anti-mouse antibody, at room temperature for 1 h. The coverslips were then mounted on slides and observed using a Zeiss LSM 510 META confocal microscope (Zeiss, Germany). For vinculin and Rac1 expression level, the distribution area was the region of interest with ImageJ software (National Institutes of Health, Bethesda, MD, USA) and the expression intensities were counted and quantified.
HT1080 cells were transfected with GFP-tagged Rab11, and cell lysates were collected in RIPA buffer with 25 mM Tris/150 mM NaCl (pH 7.4), including 1% NP-40, 5% glycerol, 0.5 M EDTA, 100 mM sodium orthovanadate 1:1000, Cocktail 1:50 and 100 mM PMSF 1:500. Cell lysates were incubated with GFP antibody or nonspecific immunoglobulin G overnight at 4°C, and then with protein A/G (Thermo Scientific) for 4 h. The precipitated protein complex was subjected to SDS-gel, and Rab11 (1:1000, Cell Signaling Technology) and p-FAK (1:1000, BD Biosciences) antibodies were used to detect proteins by western blotting.
The results are presented as means ± standard error. In this study, the Student's t-test was used to perform statistical analysis. Differences at the *P < 0.05 and ***P < 0.001 levels were considered statistically significant.
Rab11 activity plays a role in chemo-attractive cell migration
HT1080 cells are the human sarcoma cells that is ideal for the study of cell migration and FA. To determine whether Rab11 affects cell migration in HT1080 cells, Rab11 was knocked-down with overexpression of Rab11 shRNA. Western blot showed that the Rab11 expression was decreased in Rab11 shRNA-transfected cells [Figure 1]a. Rab11 shRNA-overexpressed HT1080 cells were applied in a transwell assay to test the chemo-attractive cell migration ability, and the results showed that cell migration was suppressed in Rab11 knocked-down cells [Figure 1]b and [Figure 1]c. To further examine whether Rab11 also has a function in the cell invasion ability, Rab11 shRNA-transfected cells were subjected to a transwell assay with a Matrigel-coated inside of the chamber, and the results showed that Rab11 depletion also inhibited cell invasion [Figure 1]d.
Figure 1: Rab11 deficiency inhibited the migration and invasion abilities of HT1080 cells. (a) HT1080 cells transfected with Rab11 shRNA were subjected to Western blot, and the endogenous Rab11 expression was decreased in the shRNA-transfected cells (supplementary figure S1). (b) The cell migration ability was analyzed using a transwell assay, and the migrated cells were visualized using DAPI staining. Quantification of transwell cell migration (c) without Matrigel or (d) with Matrigel was performed to assess invasion. The results showed that reducing the Rab11 expression inhibited the migration and invasion abilities of HT1080 cells (means ± standard deviation were obtained from three independent experiments ***P < 0.001).
Rab11 is a GTPase protein, and protein activity is important for many cellular functions. GFP-tagged Rab11-WT or DN (Rab11-DN) plasmids were overexpressed in HT1080 cells, and the transfection efficiency was confirmed by Western blot and confocal microscopy [Figure 2]a and [Figure 2]b. The overexpression of Rab11-WT or Rab11-DN has no effect on cell viability [Figure 2]c. The transfected cells were transferred to a transwell chamber in order to conduct a chemo-attractive cell migration assay. The results showed that Rab11-WT overexpression promoted cell migration and invasion [Figure 2]d, [Figure 2]e, [Figure 2]f; however, the DN form reduced cell migration and invasion. These results suggested that Rab11 plays an important role in chemo-attractive cell migration and invasion.
Figure 2: Rab11 GTP activity is regulated in cell migration and invasion. Overexpression of Rab11 wild-type and dominant negative was induced in HT1080 cells. GFP-tagged Rab11 wild-type and dominant negative plasmids were transfected into HT1080 cells, and the GFP vector was used as the control. (a) The expression levels were detected by Western blot with an antibody against GFP (supplementary figure S2) and (b) fluorescence microscopy was used to identify the GFP signal. (c) Cell viability of Rab11 wild-type and dominant negative transfected cells. (d-f) HT1080 cells were transfected with wild-type Rab11 and dominant negative Rab11 in transwell migration chambers. (d) The migrated cells were fixed and stained with DAPI and visualized using a fluorescence microscope, and (e) the results of cell migration were quantified. (f) A transwell membrane was coated with Matrigel for the invasion assay, and the cell invasion ability was quantified (means ± standard deviation were obtained from three independent experiments *P < 0.05).
Integrin is important for cell migration, and the transport of integrin has been shown to be associated with Rab11; hence, we examined whether Rab11 GTP activity has a function in integrin recycling. An integrin antibody was applied to Rab11-WT- or DN-transfected cells to trace integrin localization. When the cells had been starved and chilled on ice to stop endocytosis, labeled integrin appeared on the cell membrane in the Rab11-WT- and DN-transfected cells [Figure 3]. When the cells were then incubated at 37°C, endocytosis of integrin was observed in both the Rab11-WT- and DN-overexpressed cells [Figure 3]. The cells with internalized integrin were then incubated with normal growth medium to stimulate integrin export, and the results showed that Rab11-DN inhibited integrin recycling to the cell membrane. Overall, the results demonstrated that the GTP activity of Rab11 is important for integrin recycling to the cell membrane.
Figure 3: Defective Rab11 activity inhibited integrin recycling. Starved HT1080 cells transfected with (a) GFP-Rab11 wild-type and (b) GFP-Rab11 dominant negative were chilled on ice and incubated with active integrin antibody (12G10) for 30 min, and the unbound antibodies were removed by washing. Cells were then incubated at 37°C for 30 min for integrin internalization, after which cells were incubated in normal growth medium for integrin export. Integrin was revealed by immunofluorescence microscopy. The results showed that integrin was bound to the cell surface when the cells were chilled on ice, insert box indicated the enlarged image. Integrin entered into either Rab11 wild-type- or dominant negative-overexpressed cells when the cells were incubated at 37°C. However, when the cells were treated with normal medium, integrin export was found to be co-localized with Rab11 on the cell surface (arrow heads), and accumulated in the cytosol in Rab11 dominant negative-transfected cells. Scale bar = 20 μm.
FA complexes are linked to the cytosolic domain of integrin, and in order to explore whether the influence of Rab11 on cell migration is related to FA turnover, Rab11-WT and Rab11-DN were overexpressed in HT1080 cells, and FA proteins were revealed by immunofluorescent staining. The results showed that Rab11-WT induced a higher level of FA complex protein vinculin in the cell leading edge, and the cells spread well; however, this was not the case in the cells that overexpressed Rab11-DN [Figure 4]a and [Figure 4]b.
Figure 4: Rab11 localized in cell focal adhesion areas. (a) GFP-Rab11wild-type- or dominant negative mutant-overexpressed HT1080 cells were spread on collagen-coated coverslips, then stained with vinculin (focal adhesion molecule). The results showed that Rab11 wildtype overexpression induced vinculin expression in focal adhesion areas in spreading cells; however, the Rab11 dominant negative mutant did not induce significant vinculin expression in the membrane. (b) The vinculin expression area was quantified (means ± standard deviation were obtained from three independent experiments). (c) HT1080 cells expressing Rab11-wild-type or Rab11-dominant negative were stained with Rac1 antibody. Confocal imaging showed that Rac1 translocated to the leading edge and localized with Rab11-wild-type, but not with Rab11-dominant negative. (d) The areas of Rac1 expression were quantified (means ± standard deviation were obtained from three independent experiments. ***P < 0.001). Scale bar = 20 μm.
Next, we localized the distribution of Rac1, which is an actin regulation protein important for lamellipodium formation in cell migration. The immunostaining images showed that in Rab11-WT-overexpressed cells, but not in Rab11-DN-overexpressed cells, Rac1 was recruited and accumulated at the cell leading edge, the difference between the cells being significant [Figure 4]c and [Figure 4]d. The results suggested that Rab11 activity plays a function in the recruitment of FA proteins.
Rab11 colocalized with phosphor-focal adhesion kinase
Although Rab11 activity was shown to be related to FA formation [Figure 4], vesicle transport and FA turnover are dynamic, and so in order to visualize the localization of Rab11 and FA proteins, endogenous Rab11 and p-FAK were revealed by immunofluorescence staining. The results showed that Rab11 in cytosolic compartments was co-localized with p-FAK [Figure 5].
Figure 5: Rab11 co-localized with phosphor-focal adhesion kinase. (a) Immunofluorescence microscopic imaging demonstrated that endogenous Rab11 and phosphor-focal adhesion kinase were co-localized in the cytosol or focal adhesion areas. (b) An enlarged image from the area within the square, arrow indicated the focal adhesion site. Scale bar = 20 μm.
Because FA turnover is a dynamic process, to stable the FA on the cell membrane and observe the function of Rab11, we treated cells with nocodazole, a microtubule inhibitor that can reduce microtubule-mediated FA disassembly. When HT1080 cells were transfected with Rab11-WT and Rab11-DN and treated with nocodazole, FA protein paxillin accumulated at FA sites, and GFP-tagged WT Rab11 was co-localized with paxillin at the FA, but DN Rab11 was not recruited to the FA [Figure 6]a. To further examine whether Rab11 interacts with FAK directly, an immunoprecipitation assay was employed, and the results showed that p-FAK can be immuno-precipitated with GFP-Rab11, which was recognized by an anti-GFP antibody [Figure 6]b.
Figure 6: Rab11 GTP was recruited to the focal adhesion site and interacted with focal adhesion kinase directly in HT1080 cells. GFP-Rab11 wild-type- or dominant negative-transfected HT1080 cells were treated with nocodazole to induce stable focal adhesion. (a) Cells were fixed and stained with an anti-paxillin antibody. Rab11 wild-type, but not the dominant negative, was localized in the paxillin focal adhesion. (b) An immunoprecipitation assay was used to demonstrate Rab11 and focal adhesion kinaseinteraction. HT1080 cells were transfected with GFP-Rab11 wild-type plasmid, and the cell lysate was incubated with GFP antibody fused with protein A/G beads overnight. The protein complex was separated by Western blot. Anti-Rab11 and anti-p-focal adhesion kinase antibodies were used to stain the polyvinylidene difluoride membrane. GFP-Rab11 wild-type interacted with p-focal adhesion kinase directly (supplementary figure S3). Scale bar = 20 μm.
We further used live imaging to visualize the dynamics of GFP-tagged Rab11 in HT1080 cells. The live images showed that Rab11 moved quickly from the cytosol to the cell membrane in the direction of migration [Figure 7]a and [Supplementary Movie S1]. To observe the interaction of Rab11 and FAK in FA formation, mCherry-Rab11 and GFP-tagged FAK were co-transfected in HT1080 cells, and the cells were treated with nocodazole to inhibit FA disassembly. The live images showed that with overexpression, FAK was revealed apparent at the FA sites, Rab11 moved to FA sites, accompanied by FAK formation [Figure 7]b and [Supplementary Movie S2]. These results demonstrated that Rab11 can regulate FA assembly and physically interacts with FAK.
Figure 7: Rab11 and focal adhesion kinase dynamics in HT1080 cells. (a) In HT1080 cells expressing GFP-Rab11-wild-type, the Rab11 signal was tracked using a live imaging system (Axio Observer microscope, Zeiss, Germany) with time-lapse recording every 3 min. The results showed that Rab11 translocated to the cell leading edge (white arrow). (b) Co-expression of mCherry-Rab11 and GFP-focal adhesion kinase was induced after nocodazole treatment in HT1080 cells. Cells were treated with 10 μM nocadazole for 1 h. Rab11 and focal adhesion kinase signals were tracked using a Zeiss LSM 700 confocal microscope. The time-lapse data showed that Rab11 (red color) interacted with focal adhesion kinase (green color) and translocated to the focal adhesion area (yellow arrow and white arrow showing two different sites as examples).
FA assembly and disassembly have been observed and shown to be important in cell migration. Cell migration requires the formation of a membrane protrusion at the leading edge in the direction of migration. With growth factor stimulation, the signal triggers the dynamics of integrin, which interacts with the ECM and is required for FA turnover; however, FA turnover is fast, the mechanism by which FA-associated molecules are recruited to FA sites is not clear. In this study, vesicle protein Rab11 was demonstrated to play an important role in integrin recycling to the cell membrane, and with the immunoprecipitation and the live image, it is for the first time the function of Rab11 in the transport of FAK to FA sites was confirmed.
Cell migration requires external signals such as growth factors or the ECM to stimulate cell migration behavior. Our results showed that Rab11 GTP activity is required for chemo-attractive cell migration in a transwell assay, and is more sensitive to matrix coating in cell invasion experiment. Overexpression of Rab11-WT also resulted in different level of morphology changes by membrane protrusions. These results indicated that the direction of cell migration requires a growth factor gradient and ECM through Rab11 GTP activity.
Rab11 has been shown to play a role in the transport of important membrane receptors, such as EGFR, integrin and E-cadherin, to the cell membrane, and also has a function in the homeostasis of the endosome/lysosome pathway. This study demonstrated for the first time the recycling role of Rab11 with FAK. Recent studies revealed that cross-talk of receptors has a convergence effect on cell signaling pathways.,, In cell migration, integrin and FAK signals are necessary components of the process, and the results of this study provided evidence to show that GTP activity of Rab11 may play a role in coordinating integrin and FAK translocation to maintain cell migration in the direction towards the chemo gradient.
Rab11 also plays a role in E-cadherin transport. E-cadherin is a junctional receptor that sticks epithelial cells together, and a decrease in or downregulation of E-cadherin has been shown to be associated with epithelial-mesenchymal transition, which is a key process in cell migration from a primary tumor. However, recent studies revealed that E-cadherin may also play an important role in collective cell migration, which requires E-cadherin to maintain cell-cell contact and downstream signals for cohort cell migration. Whether Rab11-mediated FA dynamics are also coordinated by E-cadherin in collective cell migration requires further investigation.
FA disassembly has been shown to be mediated by microtubules., The results of our study showed that when tubulin inhibitor nocodazole was applied to suppress FA disassembly, the function of Rab11 recycling was not affected; instead, Rab11-WT was co-localized with FA molecules and accumulated at FA sites. This result confirmed that Rab11 GTP activity plays a role in recycling in FA formation. However, although DN Rab11 cannot be recruited to FA sites after nocodazole treatment, FA formation was not affected, which indicated that Rab11 affects the majority of FA formation under normal culture conditions [Figure 4] and that FA complexes may also be regulated by other mechanisms.
This study provided evidence that Rab11 can recruit FAK to the membrane; therefore, Rab11 has critical functions in coordinating many important receptors that promote cell migration processes. Furthermore, the development of drug that targets recycling endosomes may be the potential strategy in the treatment of cancer metastasis.
Financial support and sponsorship
Ministry of Science and Technology (MOST 108-2320-B-029-002) to Wei-Ting Chao.
Rivero-Ríos P, Romo-Lozano M, Madero-Pérez J, Thomas AP, Biosa A, Greggio E, et al. The G2019S variant of leucine-rich repeat kinase 2 (LRRK2) alters endolysosomal trafficking by impairing the function of the GTPase RAB8A. J Biol Chem 2019;294:4738-58.