The authors have declared that no competing interests exist.
Read the manuscript and provided the suggestions: RCN SK CTE URP. Conceived and designed the experiments: RCN URP LVMR. Performed the experiments: RCN SK. Analyzed the data: RCN SK LVMR. Contributed reagents/materials/analysis tools: CTE URP. Wrote the paper: RCN LVMR.
Recent studies have established that factor VIIa (FVIIa) binds to the endothelial cell protein C receptor (EPCR). FVIIa binding to EPCR may promote the endocytosis of this receptor/ligand complex. Rab GTPases are known to play a crucial role in the endocytic and exocytic pathways of receptors or receptor/ligand complexes. The present study was undertaken to investigate the role of Rab GTPases in the intracellular trafficking of EPCR and FVIIa. CHO-EPCR cells and human umbilical vein endothelial cells (HUVEC) were transduced with recombinant adenoviral vectors to express wild-type, constitutively active, or dominant negative mutant of various Rab GTPases. Cells were exposed to FVIIa conjugated with AF488 fluorescent probe (AF488-FVIIa), and intracellular trafficking of FVIIa, EPCR, and Rab proteins was evaluated by immunofluorescence confocal microscopy. In cells expressing wild-type or constitutively active Rab4A, internalized AF488-FVIIa accumulated in early/sorting endosomes and its entry into the recycling endosomal compartment (REC) was inhibited. Expression of constitutively active Rab5A induced large endosomal structures beneath the plasma membrane where EPCR and FVIIa accumulated. Dominant negative Rab5A inhibited the endocytosis of EPCR-FVIIa. Expression of constitutively active Rab11 resulted in retention of accumulated AF488-FVIIa in the REC, whereas expression of a dominant negative form of Rab11 led to accumulation of internalized FVIIa in the cytoplasm and prevented entry of internalized FVIIa into the REC. Expression of dominant negative Rab11 also inhibited the transport of FVIIa across the endothelium. Overall our data show that Rab GTPases regulate the internalization and intracellular trafficking of EPCR-FVIIa.
The endothelial cell protein C receptor (EPCR) is the cellular receptor for protein C (PC) and activated protein C (APC), and is mainly present on the endothelial cell lining of larger blood vessels
A subfamily of Ras-like small GTPases, termed as Rab GTPases, have been shown to play a critical regulatory role in both endocytic and exocytic pathways of protein trafficking by regulating vesicular membrane transport and membrane fusion events
In the present study, we investigated whether Rab GTPases regulate the internalization, intracellular trafficking, and recycling of EPCR and EPCR bound ligand. We show that Rab 4, Rab 5, and Rab 11 control the intracellular trafficking of EPCR and FVIIa at different stages. Overall, our data suggest that Rab GTPases play important roles in the endosomal sorting/recycling of EPCR and provide information on a potential mechanism for regulation of EPCR levels on the cell surface and EPCR-dependent transcytosis.
Rabbit polyclonal antibodies against Rab5, Rab4, Rab7 and Rab11 were purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Secondary antibodies conjugated with Oregon Green or Rhodamine Red, and Alexa Fluor 488 (AF488) labeling kit were obtained from Invitrogen Corp. (Carlsbad, CA). Mouse monoclonal antibodies against human EPCR (JRK-1494/blocking mAb and JRK-1500/non-blocking mAb) were prepared as described earlier
Primary human umbilical vein endothelial cells (HUVEC), EBM-2 basal medium, and growth supplements were purchased from Lonza (Walkersville, MD). Endothelial cells were cultured in EBM-2 basal medium supplemented with growth supplements, 1% penicillin/streptomycin, and 5% fetal bovine serum. Generation of CHO cells stably expressing EPCR (CHO-EPCR) was described previously
Wild type (WT) Rab5A, its constitutively active (CA) form (Rab5AQ67L), and dominant negative (DN) variant (Rab5AS34N) were kindly provided by Brian Knoll (University of Houston, Houston, TX). Rab11 and its variants, Rab11Q70L and Rab11S25N, were provided by David Sabatini (New York University School of Medicine, New York, NY). Rab4A cDNA was provided by Marino Zerial (Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany). Rab7 and its variants, Rab7Q67L and Rab7T22N, were provided by Juan Bonifacino (National Institute of Child Health & Human Development, Bethesda, MA). All of the above plasmid inserts were transferred into adenoviral shuttle vector pacAD5CMV K-N pA using standard cloning techniques. Rab4AQ67L and Rab4AS34N variants were generated by site-directed mutagenesis method using “Quick change II XL” site directed mutagenesis kit (Stratagene, La Jolla, CA) and pacAD5CMV K-N- pA Rab4A as the template.
HEK 293 cells seeded in 60 mm dishes (80% confluent) were cotransfected with 1 µg of adenoviral backbone DNA and 5 µg of pac1-digested linearized pacAD5CMV K-N pA containing Rab DNA using Fugene HD transfection reagent according to the manufacturer’s protocol (Roche Diagnostics Corp. Indianapolis, IN). After 7–8 days of transfection, HEK 293 cells showing cytopathic effects were lysed by repeated freeze/thaw cycles and centrifuged at 3,000×g to collect the supernatant containing primary adenoviral stock, and this primary adenoviral stock was used to infect HEK 293 cells in 6–8 T-75 flasks to generate high titer viruses. Viral titers were determined according to the manufacturer’s protocol using “Quick Titer Adenovirus Titer Immunoassay” kit (Cell Biolabs, Inc. San Diego, CA). In general, HUVEC were transduced with 20 MOI/cell whereas 50 MOI/cell were used for transduction in CHO-EPCR cells.
FVIIa was labeled with AF488 fluorescent probe using micro scale protein labeling kit (Invitrogen Corporation, Carlsbad, CA) as described recently
CHO-EPCR cells cultured on fibronectin-coated glass cover slips were infected for 48 h with adenoviruses encoding one of the Rab variants or control adenovirus. After 48 h, the cells were washed with buffer B (10 mM HEPES, 0.15 M NaCl, 4 mM KCl, 11 mM glucose, pH 7.5 buffer containing 5 mM CaCl2, 1.0 mM MgCl2, and 1 mg/ml BSA), and incubated with AF488-FVIIa (50 nM) in buffer B at 4°C (on an ice-bath in a cold room) for 1 h to allow binding of the ligand to EPCR with no or minimal internalization of the bound ligand. At the end of the 1 h incubation, the unbound ligand was removed; cells were washed twice with cold buffer B, and then were transferred to 37°C to induce internalization. At varying time intervals the cells were fixed, permeabilized, and processed for immunofluorescence confocal microscopy. For steady-state internalization studies, HUVEC were incubated with AF488-FVIIa (50 nM) at 37°C for a fixed time.
Fixed and permeabilized cells were subjected to immunostaining for EPCR and various Rabs as described recently by us
Internalization, recycling, and degradation of 125I-labeled FVIIa were determined as described recently
A transwell permeable system (3-µm pore size, polyester membrane, 12-mm diameter; Corning, NY) was used to evaluate the transport of FVIIa from apical to basal surface. Briefly, upper chamber inserts were coated with 0.05% fibronectin (Sigma, St Louis, MO USA) for 30 min, washed once with serum-free medium, and air-dried. HUVEC were seeded in the upper chamber (50,000 cells/well) and allowed to grow for 48 h in EBM-2 growth medium. After 48 h, HUVEC were infected with either control adenovirus or adenovirus encoding wild-type, constitutively active or dominant negative Rab11 (20 moi/cell). After culturing cells further for 72 h, the cells were washed twice with serum-free medium, and serum-free EBM-2 medium supplemented with 2% BSA was added to both upper and bottom chambers. FVIIa (10 nM) was added to the upper chamber. The cells were allowed to incubate for 2 h at 37°C and 5% CO2. At the end of 2 h, the medium from the bottom chamber was removed and the FVIIa that transcytosed into the bottom chamber was determined in FXa generation assay using saturating concentrations of relipidated TF.
The images were processed using LSM Zen 2009 (Zeiss) software and imported to Adobe Photoshop for compilation of figures. When mean fluorescence was determined, typically fluorescence values of 10 to 20 ROI were used for determining FVIIa accumulation at the REC, and 30 to 50 ROI for determining FVIIa levels at the cell surface (for recycling to the plasma membrane). Unpaired t-test was used to calculate whether an experimental value significantly differs from the control value.
Our recent studies
CHO-EPCR cells were exposed to AF488-FVIIa (50 nM) for 1 h at 4°C. At the end of 1 h, the supernatant was removed, cells were washed quickly with Ca2+/Mg2+-containing buffer to remove the unbound ligand and then transferred to 37°C to induce internalization of the surface bound ligand. At the end of 0, 5, and 15 min, cells were fixed, permeabilized, and immunostained with mouse monoclonal or rabbit polyclonal antibodies against EPCR, Rab5A, Rab4A, or Rab11 followed by Rhodamine Red-labeled anti-mouse/anti-rabbit IgG as a secondary reporter antibody. The cells were imaged as described in methods. Left panel, immunostaining of EPCR, Rab5A, Rab4A or Rab11; middle panel, fluorescence of AF488-FVIIa. The right panel depicts the merged images of left and middle panels. The insets show the magnified view of the boxed regions.
To investigate the role of various Rab GTPases in EPCR-mediated endocytosis and intracellular trafficking of FVIIa, CHO-EPCR cells were transduced with either control adenoviruses or adenoviruses encoding WT, CA, or DN variants of Rab4A, Rab5A, Rab7, or Rab11. The transduced cells were exposed to AF488-FVIIa, and the endocytosis and intracellular trafficking of EPCR and FVIIa were analyzed. In cells transfected with control adenovirus, at 4°C, AF488-FVIIa bound to the cell surface and colocalized exclusively with EPCR at the cell surface (
CHO-EPCR cells transduced with control adenoviruses were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C. After removing the unbound ligand, cells were transferred to 37°C to induce internalization of the surface bound ligand. The cells were immunostained for EPCR and analyzed for immunofluorescence of EPCR and fluorescence of AF488-FVIIa. The two right panels are digitally enlarged images of the inset, and the arrow indicates the accumulation of AF488-FVIIa and EPCR in the REC. Please note that in this and other figures involving CHO-EPCR cells, the images in top three panels were a single chosen section from z-stack and the images in the bottom three panels were reconstructed composite of all z-stacks. We chose this presentation to show FVIIa trafficking from the surface to REC via endosomes more illustratively, as different compartments reside in different planes. Bar scale shown here and in other figures for CHO-EPCR cells represent 10 µm.
FVIIa internalization and trafficking in CHO-EPCR cells expressing WT Rab5A was very similar to that was observed with control CHO-EPCR cells, with the exception that endosomal structures were slightly larger in cells transduced with WT Rab5A (data not shown). It has been shown that overexpression of Rab5 WT or Rab5 CA mutants led to the formation of enlarged early endosomal structures due to the enhanced endosome-endosome fusion mediated by GTP bound Rab5
CHO-EPCR cells transduced with recombinant adenoviruses to express constitutively active Rab5A were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C. After removing the unbound ligand, cells were transferred to 37°C to induce internalization of the surface bound ligand. The cells were immunostained for EPCR and Rab5, and analyzed for immunofluorescence of EPCR and Rab. The cells were also analyzed for fluorescence of AF488-FVIIa. The two right panels are the digitally enlarged images of a small portion of the merged image of AF488-FVIIa with Rab5 or EPCR staining, respectively. The arrow indicates the accumulation of AF488-FVIIa in the REC and the arrow head shows the trapping of AF488-FVIIa in the enlarged endosomal structures.
CHO-EPCR cells transduced to express dominant negative Rab5A mutant were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C, and internalization was induced at 37°C as described in Fig. 3. The cells were immunostained for EPCR and Rab5. They were then analyzed for immunofluorescence of EPCR and Rab, and fluorescence of AF488-FVIIa. The 5th and 6th panels are the digitally enlarged images of the insets of the 2nd and 4th panels, respectively. The extreme right panel was the same panel that was shown in
CHO-EPCR cells transduced with control adenoviruses (brown) or recombinant adenoviruses encoding wild-type (blue), constitutively active (green), or dominant negative (red) Rab5A were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C, and allowed to internalize for varying time periods at 37°C. FVIIa that accumulated in the REC was quantified by measuring the pixel density of the fluorescence of AF488-FVIIa in this compartment. * denotes that the value significantly differs from the values obtained in cells expressing endogenous Rab5A or wild-type Rab5A (
HUVEC transduced with control adenovirus or recombinant adenoviruses to express constitutively active or dominant negative Rab5A variants were incubated with AF488-FVIIa (50 nM) for 30 min at 37°C. Permeabilized cells were immunostained for EPCR and Rab5A, and imaged to localize EPCR, Rab5A, and AF488-FVIIa. The last two panels are the digitally enlarged images of the insets in 2nd and 3rd panels. Arrows indicate the accumulation of AF488-FVIIa in the REC. Therrowhead indicates entrapment of the ligand in abnormally enlarged early endosomal structures. The bar shown on images of HUVEC in this and other figures represents 20 µm length.
Rab4A is known to play a critical role in the recycling of receptor/ligand complexes from early/sorting endosomes back to the cell surface. Overexpression of WT Rab4A in CHO-EPCR cells impaired trafficking of AF488-FVIIa from early endosomes to the REC as significantly less ligand accumulation was found at the REC in these cells even after 15 to 30 min following the onset of internalization (
CHO-EPCR cells were transduced with adenoviruses encoding wild-type Rab4A. The transduced cells were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C, and then transferred to 37°C to induce internalization of the surface bound ligand. The permeabilized cells were immunostained for EPCR and Rab4A, and immunofluorescence and fluorescence of AF488-FVIIa were analyzed by confocal microscopy. The two right two panels are the digitally enlarged images of the insets of 2nd and 3rd panels to clearly illustrate differences in the intracellular localization of FVIIa at different time intervals. The arrow indicates the accumulation of AF488-FVIIa in the REC.
The experimental procedure and the image acquisition were essentially the same as described in Fig. 7 except that CHO-EPCR cells were transduced with adenoviruses encoding constitutively active Rab4A, instead of wild-type Rab4A.
The experimental procedure is essentially the same as described in Fig. 7 except that CHO-EPCR cells were transduced with adenoviruses encoding dominant negative mutant of Rab4A, instead of wild-type Rab4A.
CHO-EPCR cells transduced with control adenoviruses (brown) or recombinant adenoviruses encoding wild-type (blue), constitutively active (green), or dominant negative (red) Rab4A were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C, and allowed to internalize for varying time periods at 37°C. FVIIa accumulated in the REC was quantified by measuring the pixel density of the fluorescence of AF488-FVIIa in this compartment. * denotes that the value significantly differs from the values obtained in cells expressing endogenous Rab4A (
CHO-EPCR cells transduced with control adenoviruses (brown) or recombinant adenoviruses encoding wild-type (blue), constitutively active (green), or dominant negative (red) Rab4A were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C, and allowed to internalize for varying time periods at 37°C. The recycling of the ligand to the cell surface was quantified by measuring the pixel intensity of AF488-FVIIa on the plasma membrane at different time intervals.
We also analyzed the steady state internalization of AF488-FVIIa in HUVEC transduced to overexpress Rab4A WT, CA, or DN variants. Similar to that observed in CHO-EPCR cells, the trafficking of AF488-FVIIa from early endosome to the REC was impaired in HUVEC expressing WT or CA mutant, and therefore the amount of the ligand accumulated at the REC was lower in these cells compared to control cells (
HUVEC transduced to express wild-type, constitutively active, or dominant negative Rab4A variants were incubated with AF488-FVIIa (50 nM) for 30 min at 37°C. Then, the cells were fixed, permeabilized, immunostained for EPCR and Rab4, and imaged to localize EPCR, Rab4, and AF488-FVIIa. Arrows mark the presence of AF488-FVIIa and EPCR in REC at the juxtanuclear position.
Rab11 has been shown to associate with the pericentriolar REC and play a critical role in the recycling of receptor/ligand complexes from this compartment back to the cell surface. Overexpression of WT Rab11 in CHO-EPCR cells showed no significant effect in EPCR-FVIIa trafficking (
CHO-EPCR cells transduced with adenovirus encoding wild-type Rab11 were exposed to AF488-FVIIa (50 nM) for 1 h at 4°C. Then, the unbound ligand was removed and the cells were transferred to 37°C to induce internalization of the surface bound ligand. After varying times at 37°C, the cells were fixed, permeabilized and immunostained for EPCR and Rab11. The cells were imaged for immunofluorescence of EPCR and Rab11, and fluorescence of AF488-FVIIa. The two right panels are digitally enlarged images of insets of 2nd and 3rd panels, respectively, to provide a better illustration of differences in the intracellular localization of FVIIa at varying time intervals. Arrows indicate the accumulation of AF488-FVIIa and EPCR in the REC at the juxtanuclear region.
The experimental procedure and the image acquisition were essentially the same as described in
The experimental procedure and the image acquisition were essentially the same as described in
CHO-EPCR expressing endogenous Rab11 (brown) or CHO-EPCR cells transduced to overexpress wild-type Rab11 (blue) or constitutively active Rab11 (green) were incubated with AF488-FVIIa (50 nM) for 1 h at 4°C, and allowed to internalize for varying time periods at 37°C. FVIIa accumulated at varying times in the REC was quantified by measuring the pixel density of the fluorescence of AF488-FVIIa in this compartment. * denotes that the value significantly differs from the values obtained in cells expressing endogenous Rab11 or wild-type Rab11 (
HUVEC transduced with control or recombinant adenoviruses to express wild-type, constitutively active or dominant negative Rab11 were incubated with AF488-FVIIa (50 nM) for 30 min at 37°C. Then, the cells were fixed, permeabilized, and immuonstained for EPCR and Rab11, and imaged to observe localization of EPCR, Rab11 and AF488-FVIIa. The two right panels are digitally enlarged images of the insets of 2nd and 3rd panels, respectively. Arrows indicate the accumulation of AF488-FVIIa and EPCR in REC at the juxtanuclear region.
In additional studies, we investigated the effect of overexpression of WT, CA, and DN Rab7 variants on EPCR-FVIIa endocytosis and trafficking. No noticeable differences were observed in EPCR-FVIIa endocytosis or intracellular trafficking of internalized EPCR and FVIIa in HUVEC overexpressing Rab7 WT, CA or DN variants (
HUVEC transduced with control or recombinant adenoviruses to express wild-type, constitutively active or dominant negative Rab7 were incubated with AF488-FVIIa (50 nM) for 30 min at 37°C. Then, the cells were fixed, permeabilized, and immuonstained for EPCR and Rab7, and imaged to observe localization of EPCR, Rab7, and AF488-FVIIa. Right two panels are digitally enlarged images of the insets of 2nd and 3rd panels. Arrow indicates the accumulation of AF488-FVIIa and EPCR at REC at the juxtanuclear region.
Since Rab11 was shown to regulate transcytotic migration of internalized ligands from apical to basal surfaces in polarized epithelial cells
HUVEC cultured in transwells were infected with control adenovirus or adenoviruses encoding wild-type Rab11 or Rab11 variants (20 moi/cell), and grown to reach full confluency. Then, FVIIa (10 nM) was added to the upper chamber. Two hours after adding FVIIa, the medium from the bottom chamber was removed and measured for FVIIa activity levels in factor X activation assay. The data shown in the figure represent mean ± SEM from 3 to 6 independent experiments. *indicates a statistical significant difference from the value obtained with HUVEC infected with control virus (
To further strengthen the above data obtained from microscopic studies, we attempted to quantify, more objectively, differences in the endocytosis and recycling of FVIIa and EPCR following expression of various Rab GTPases and their variants by monitoring FVIIa and EPCR using125I-labeled FVIIa and EPCR mAb. However, these studies failed to yield robust and conclusive data. A probable explanation for this is that small amounts of FVIIa and EPCR endocytosed and trafficked would be much easier to detect by confocal microscopy as they will have targeted organization into distinctive membrane compartments. Furthermore, basal non-specific/EPCR-independent binding and internalization does not interfere in image analysis as they do not result in targeted organization and dense accumulations. Internalization and recycling assays using radiolabeled ligands measure global differences, and thus it may be difficult to capture distinctive differences in intracellular trafficking mediated by Rab GTPase variants using these methods.
Our recent studies showed that FVIIa binding to EPCR promoted the endocytosis of EPCR via dynamin and caveolar-dependent pathways and the endocytosed receptor-ligand complexes accumulated in the REC before being targeted back to the cell surface
Colocalization analyses of the internalized FVIIa with various Rab proteins (Rab5, Rab4, Rab11 and Rab7) at different time intervals following FVIIa internalization showed that immediately following internalization (at 5 min), FVIIa colocalizes extensively with Rab5 positive endosomes, suggesting entry of internalized FVIIa into these early endosomal compartments. At this early time, internalized FVIIa also colocalizes with Rab4 positive endosomal structures, indicating targeted trafficking of the internalized ligand into the sorting endosomes as well. However, the colocalization efficiency between Rab4 and FVIIa is lower than that observed between Rab5 and FVIIa. These data suggest that some but not all of the internalized FVIIa is sorted to Rab4 positive sorting endosomes. As expected from our earlier study
Rab5 localizes to the plasma membrane and early endosomal structures, and plays a critical role in the endocytosis of receptor and receptor/ligand complexes from the plasma membrane to the early endosomes
Rab4 has been shown to play a vital role in the recycling of receptor or ligand from sorting endosomes back to the cell surface
When we examined the role of Rab11 in EPCR-FVIIa trafficking, we found that the expression of the dominant negative mutant Rab11 resulted in accumulation of FVIIa throughout the cytoplasm and very little in the REC. This suggests that Rab11 regulates the trafficking of FVIIa and EPCR from the early endosome to the REC. Interestingly, the expression of constitutively active Rab11 mutant not only led to accumulation of FVIIa and EPCR in the REC, but also resulted in retention of FVIIa and EPCR for a longer period of time within this compartment. It has been shown that hydrolysis of GTP bound to Rab11 GTPase is essential for the recycling of transferrin receptor from the REC back to the cell surface
In contrast to Rab5, Rab4, and Rab11, expression of Rab7, either the constitutively active or dominant negative variant, did not alter the kinetics of EPCR-FVIIa endocytosis or its trafficking. Rab7 acts downstream of Rab5 in regulating the membrane transport from early to late endosomes
Although in the present study, we have limited our investigation to EPCR-mediated FVIIa trafficking, it is likely that Rab GTPases regulate other ligands of EPCR, i.e., FVII, protein C and APC, in a similar fashion. It may be pertinent to note here that our earlier studies showed a similar pattern of internalization and cellular localization of FVII, FVIIa, protein C, and APC
Overall, our data presented herein indicate that Rab GTPase activity plays a role in regulating EPCR and FVIIa levels at the cell surface by controlling the rate at which the receptor and receptor-ligand complex are processed through the endosomal compartments. The ability of Rab GTPases to regulate EPCR trafficking suggests that mutations leading to altered Rab GTPase activity and/or differences in Rab GTPase protein levels may affect EPCR function by altering the dynamics of its endocytosis, intracellular trafficking, and plasma membrane recycling. A number of studies have associated various human diseases with the expression of mutant Rab GTPases
The authors thank Brian Knoll, University of Houston, Houston, TX for providing plasmid constructs of Rab 5A and its variants; David Sabatini, New York University School of Medicine, New York, NY for providing plasmid constructs of Rab11 and its variants; Marino Zerial, Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany for providing Rab4A cDNA; and Juan Bonifacino, National Institute of Child Health & Human Development, Bethesda, MA for providing plasmid constructs of Rab7 and its variants. We are thankful for Janet Arras for proof-reading the manuscript.