s
Microparticles: mediators of cellular and environmental homeostasis
Böing, A.N.
Publication date
2011
Link to publication
Citation for published version (APA):
Böing, A. N. (2011). Microparticles: mediators of cellular and environmental homeostasis.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.
Chapter 8
Coagulant tissue factor is not raft associated
Anita N. Böing, Chi M. Hau, Jan van Marle, Auguste Sturk and Rienk Nieuwland
Abstract
Introduction. The subcellular distribution of tissue factor (TF) is still debated. Therefore,
we studied the localization of TF in cell lysates, purified plasma membranes and microparticles.
Methods. Human TF was cloned and expressed by human MIA PaCa-2 cells. Cells and
microparticles were lysed in Triton X-100-containing buffer. Plasma membranes were purified by Percoll-gradient ultracentrifugation after cell disruption. Rafts were isolated by OptiPrep gradient ultracentrifugation. Fractions were analyzed for TF antigen, TF coagulant activity, flotillin (raft marker), caveolin (caveolar raft marker), protein disulfide isomerase (PDI), and coatomer protein complex subunit-β (β-COP; Golgi membrane marker). Co-localization was studied by confocal microscopy.
Results. Of the total cellular amount of TF, less than 10% was associated with plasma
membranes, of which less than 10% triggered thrombin generation and fibrin formation. Plasma membrane fractions 2-6, which contained the coagulant form of TF, lacked flotillin, caveolin and PDI. In contrast, the non-coagulant form of TF was associated with flotillin, caveolin and PDI. In microparticles, the coagulant form of TF was also not associated with flotillin, caveolin or PDI.
Conclusions. Cell lysates, plasma membranes and microparticles contain coagulant and
non-coagulant forms of TF. The coagulant form of TF is not associated with rafts in plasma membranes and microparticles.
8
Introduction
Tissue factor (TF) is a 45 kDa trans-membrane protein that initiates coagulation after binding coagulation factor VII(a). With regard to the cellular distribution of TF, Mandal and coworkers showed that approximately 25% of the total cellular TF is exposed at the cell surface1. Thus, only a small proportion of the total cellular TF is accessible to factor VII(a).
To which extent this surface-exposed TF is associated with rafts in plasma membranes, however, has not been studied.
Rafts are defined in the literature as Triton X-100 resistant, cholesterol-enriched membrane microdomains, which compartmentalize cellular functions, including vesicular trafficking and signal transduction2. Two common types of rafts are present in cells,
non-caveolar rafts (also known as planar rafts), and non-caveolar rafts. Caveolar rafts contain caveolea, which are small, flask-shaped invaginations of the membrane that are formed when caveolin-1 becomes integrated into the microenvironment of the raft3;4. Another
raft-associated protein is flotillin, which is present in both caveolar and non-caveolar rafts5;6. To
which extent (coagulant) TF is raft associated, is still unclear. TF was enriched in raft fractions of a total cell lysate of a monocyte cell-line7. Furthermore, TF and caveolin
co-localized in fibroblasts, suggesting that TF localizes in caveolar rafts1. In total cell lysates
of human embryonic kidney cells, however, TF was not associated with rafts8.
TF can be present in a coagulant and in a non-coagulant form. The non-coagulant TF can be converted into coagulant TF, and the actual activity of coagulant TF can be regulated. Various mechanisms are described to influence both processes, such as dimerization of the TF antigen, the presence of anionic phospholipids, or the association with rafts. The latter is controversial, since disrupture of rafts has been reported to decrease as well as to increase the coagulant activity of TF8;9. In addition, the coagulant activity of
TF was shown to be downregulated after translocation of TF in complex with factors VIIa and Xa into caveolar rafts on the surface of endothelial cells10. Thus, to which extent the
coagulant form of TF is associated with (caveolar) rafts and whether the coagulant activity of TF is regulated by association with (caveolar) rafts, remains unclear.
In 2006, protein disulfide isomerase (PDI) was shown to oxidize and reduce the disulfide bond between Cys186 and Cys209 in the extracellular domain of TF, thereby
Furthermore, antibodies against PDI inhibited fibrin formation in a mouse vessel wall injury model, suggesting that PDI may regulate the coagulant activity of TF in vivo12;13.
The precise role of PDI in the conversion of non-coagulant TF into coagulant TF, is still under debate. There is no direct evidence that free thiols within TF are available and accessible for the proposed redox reactions14, and the concurrent presence of PDI and TF
has been questioned since cell membranes of epithelial cancer cells containing the coagulant form of TF lacked PDI15, and PDI was not secreted by activated platelets or
endothelial cells16.
Thus, at present there is no consensus on the cellular distribution of (coagulant) TF and PDI. Therefore, in the present study, we investigated the subcellular localization of the TF antigen, TF coagulant activity and PDI in purified plasma membranes and microparticles.
Materials and Methods
Cloning of human TF from HUVEC
HUVEC (Human Umbilical Vein Endothelial Cells) were isolated and cultured as described previously, but human serum (10%) was replaced by Fetal Calf Serum (FCS; 20%)17.
HUVEC were activated with interleukin-1α (IL-1α; 5 ng/mL; Sigma, St. Louis, MO) for one hour. Subsequently, cells were harvested and RNA was isolated (RNeasy Mini Kit; Qiagen, Hilden, Germany) and used for cDNA preparation (First Strand cDNA synthesis kit; Roche, Basel, Switserland). The TF open reading frame was amplified using a 5’primer: GGGGACAAGTTTGTACAAAAAAGCAGGCTCCACCATGGAGACCCC TGCCTGG and a 3’primer: GGGGACCACTTTGTACAAGAAAGCTGGGTTTTATGA AACATTCAGTGGGGAGT. Using Gateway technology (Invitrogen; Carlsbad, CA), the resulting PCR fragment was shuttled via pDON/Zeo (Invitrogen), into pEGFP-N3 (Clontech, Mountain View, CA). The resulting vector expresses the TF mRNA/protein under control of the CMV promoter.
Transfection of MIA PaCa-2 cells with TF
MIA PaCa-2 cells were a gift from the Department of Neurogenetics (Academic Medical Center; Amsterdam, The Netherlands). Cells were cultured in DMEM (Invitrogen)
8
supplemented with 10% FCS (PAA, Pasching, Austria), penicillin and streptomycin (10 units/mL and 10 µg/mL, respectively; Invitrogen). Before transfection, cells were cultured in 75 cm2 culture flask for three days. Thereafter, seven µg DNA and 70 µL Lipofectin
(Invitrogen) were added to the cells and after 24 hours the culture supernatant was removed and replaced by fresh culture medium. After 48 hours, cells were harvested.
Raft isolation from cell lysates
After transfection (48 hours), MIA PaCa-2 cells were washed twice with ice cold PBS. Then 1.0 mL lysis buffer, containing 10 µL protease inhibitors and 1% (v/v) Triton X-100 (Caveolae/rafts isolation kit, Sigma-Aldrich, Zwijndrecht, The Netherlands) was added and cells were harvested. Cells were incubated on ice for one hour followed by centrifugation (5 minutes, 2000g, 4 °C). The cell lysate (supernatant) was loaded on an OptiPrep gradient (5-35%; Caveolae/rafts isolation kit), followed by ultracentrifugation for 16 hours at 137,000g (polyallomer tube, Beckman; SW41Ti; Beckman XL90 ultracentrifuge). After centrifugation, nine fractions of 1 mL were collected.
Isolation of rafts from purified plasma membranes
TF-transfected MIA PaCa-2 cells (48 hours) were washed twice with ice cold PBS and harvested by scraping with PBS containing protease inhibitors (5 mL). Thereafter, the cells were disrupted in a nitrogen bomb (70 atm, 20 minutes, on ice). After disruption, the total cell lysate was centrifuged to remove cellular debris (15 minutes at 1500g at 4 °C). The supernatant was collected (5 mL) and mixed with ice cold PBS (10 mL), Percoll (8.6 mL; GE Healthcare, Uppsala, Sweden) and aquadest (1.4 mL) at pH 7.4, and centrifuged for 23 minutes at 79,000g at 4 °C (polycarbonate tubes, Beckman; 70Ti rotor; Beckman XL90 ultracentrifuge). After ultracentrifugation, the upper part of the fluid (11 mL), containing both plasma membranes and intracellular membranes, was collected and mixed with ice cold PBS (22 mL) and Percoll (18.9 mL), and the pH was carefully adjusted to 9.6. Subsequently, the mixture was again centrifuged for 23 minutes at 79,000g at 4 °C. After ultracentrifugation, the upper part (11 mL), containing the purified plasma membranes18,
minutes at 200,000g at 4 °C. After isolation of the plasma membranes, rafts were isolated as described for cell lysates.
Isolation of rafts from microparticles
Supernatant of transfected MIA PaCa-2 cells was collected and centrifuged (10 minutes at 180g) to remove detached cells. Cell-free supernatants (aliquots of 1 mL) were frozen in liquid nitrogen and stored at -80 °C until use. Supernatant (13 x 1 mL) was thawed on melting ice, and microparticles were isolated by centrifugation for 60 minutes at 18,890g. The microparticle-free supernatant was removed (975 µL) and microparticle pellets (25 µL each) were pooled. PBS was added to 1 mL and microparticles were pelleted by centrifugation for 60 minutes at 18,890g. Thereafter, microparticle-free supernatant was removed (975 µL), microparticles were resuspended in PBS (225 µL), and again centrifuged to pellet the microparticles. Then, 225 µL microparticle-free supernatant was removed and 1 mL lysis buffer containing protease inhibitors and Triton X-100 was added. Thereafter, the same procedure was followed for raft isolation as described for whole cell lysates.
TF ELISA
After ultracentrifugation, 100 µL of each fraction was frozen to determine the TF antigen by TF ELISA (American Diagnostica, Stamford, CT). The ELISA was performed as described by the manufacturer.
Western Blot
Proteins of each fraction were precipitated with trichloroacetic acid (20% final concentration; Sigma-Aldrich) and the protein concentration was determined by a Coommassie blue protein detection assay (Thermo Scientific, Rockford, IL). From each fraction, equal amounts of protein or equal amounts of volume were dissolved in reduced sample buffer, loaded on 8-16% gradient PAGE gels (Biorad, Hercules, CA) and transferred to PVDF membrane (Millipore, Billerica, MA). Blots were incubated with anti-TF (4503; American Diagnostica), anti-caveolin-1 (Caveolae/rafts isolation kit, Sigma-Aldrich), anti-flotillin (BD transduction laboratories, San Jose, CA), anti-PDI (Thermo
8
Scientific) and anti-β-COP (Abcam, Cambridge, UK). As secondary antibodies, mouse (GAM)-horseradish peroxidase (HRP) (Dako, Glostrup, Denmark) or goat-anti-rabbit (GAR)-HRP (Dako, only used for anti-caveolin) were used. To visualize the bands, membranes were incubated with a 5-fold diluted peroxidase substrate (LumiLight, Roche Diagnostics, Almere, The Netherlands) for 5 minutes, followed by analysis of luminescence using a LAS3000 luminescent image analyzer (Fuji, Valhalla, NY).
Thrombin generation assay
Fractions (20 µL) were preincubated with and without antibodies directed against factor (F) VII(a) (5 µL; Sanquin, Amsterdam, The Netherlands). After 30 minutes, defibrinated and microparticle-free normal pool plasma (120 µL), also preincubated with or without anti-FVII (20 µL), was added to the preincubated fractions. Thrombin generation was initiated by addition of calciumchloride (0.1 mol/L), and every minute a subsample (3 µL) was collected and added to pefachrome (Pentapharm, Basel, Switserland). After six minutes, thrombin generation was stopped by addition of citric acid (1 mol/L) and thrombin formation was monitored by measuring ρ-nitroaniline at λ = 405 nm. The area under curve (AUC) was quantified using GraphPad Prism for Windows, release 5 (Prism, San Diego, CA). From each experiment (n=3), inhibition by anti-FVII was calculated as follows: (AUC of thrombin generation induced by fraction X in the absence of anti-FVII minus AUC of thrombin generation induced by fraction X in the presence of anti-FVII) / AUC of thrombin generation induced by fraction X in the absence of anti-FVII.
Fibrin generation assay
Fractions (20 µL) were added to microparticle-free normal pool plasma (70 µL) and the mixture was preincubated with and without antibodies against FVII(a) (3 µL) for 5 minutes at 37 °C. Then, the reaction was started with calciumchloride (0.1 mol/L) and fibrin clot formation was monitored by measuring the optical density of the plasma at λ=405 nm in a microtiterplate reader at 37 °C (Molecular Devices Corp, Sunnyvale, CA). The absorbance measurements were recorded every minute during one hour. From each experiment (n=2), the factor of inhibition of the 50% Vmax by anti-FVII was calculated as follows: (50%
Vmax of fraction X in the presence of anti-FVII minus 50% Vmax of fraction X in the absence of anti-FVII) / 50% Vmax of fraction X in the absence of anti-FVII.
In control experiments, we compared the activity and specificity of anti-FVII(a) and anti-TF in microparticle-free normal plasma to inhibit thromborel- and kaolin-induced thrombin generation and fibrin formation. Both anti-FVII(a) and anti-TF completely inhibited thromborel induced thrombin generation and fibrin formation, but did not inhibited kaolin-induced thrombin generation or fibrin formation. Since the antibody against FVII can be used at a lower concentration (1.0 µg/mL) than the antibody against TF (7.8 µg/mL) to completely inhibit TF-initiated coagulation, we used anti-FVII(a) in all experiments.
TF localization by confocal fluorescence microscopy
To investigate the co-localization of TF and caveolin, and co-localization of TF and PDI, we performed confocal fluorescence microscopy. MIA PaCa-2 cells (2.5 x 104) were
cultured on gelatine (0.75%)-coated cover slips for two days before transfection. Thereafter, one µg of DNA encoding TF and 10 µL Lipofectin (Invitrogen) were added to the cells and after 24 hours the culture supernatant was removed and replaced by fresh culture medium. After 48 hours, cells were washed once with PBS (Invitrogen). Then, cells were fixed with 2% paraformaldehyde (Electron Microscopy Science; Hatfield, PA) for 10 minutes, followed by washing with PBS for three times. Thereafter, cells were incubated for 30 minutes with PBS with 1% v/v FCS (PAA) and 0.1% v/v Triton X-100 (Sigma-Aldrich). Cells were labelled with various combinations of antibodies: (1) anti-TF (mouse monoclonal, 4503; American Diagnostica) and anti-caveolin (rabbit polyclonal, Sigma-Aldrich), (2) TF (mouse monoclonal clone HTF1; Becton Dickinson (BD)) and anti-caveolin (rabbit polyclonal, Sigma-Aldrich) or (3) anti-TF (rabbit polyclonal, product number 4502; American Diagnostica) and anti-PDI (mouse monoclonal, Thermo Scientific), for 1.5 hour at room temperature. Cells were washed thrice with PBS/FCS/Triton X-100 for five minutes. Then, cells labelled with TF 4503 and anti-caveolin were incubated with GAM-FITC (Dako) and GAR-Cy3 (Jackson ImmunoResearch, West Grove, PA) for 30 minutes at room temperature. Cells labelled with anti-TF BD and anti-caveolin were incubated with GAM-FITC (Dako) and GAR-Cy3
8
(Jackson ImmunoResearch) for 30 minutes at room temperature. Cells labelled with anti-TF 4502 and anti-PDI were incubated with GAR-FITC (Jackson ImmunoResearch) and GAM-Cy3 (Jackson ImmunoResearch) for 30 minutes at room temperature. After incubation, cover slips were washed three times with PBS/FCS/Triton. Then, cover slips were embedded in Vectashield with DAPI (Vector Laboratories, Burlingame, CA).
Cells were imaged with a SP2-AOBS confocal system (Leica Microsystems, Wetzlar, Germany). For all imaging a 40* NA1.25 objective was used. The pinhole setting was kept at 1 Airy, resulting in a XY resolution and a Z resolution of approximately 170 nm and 500 nm, respectively. Excitation and detection of DAPI, FITC and Cy3 were performed as follows: 405 nm/ 420-470 nm, 488 nm/500-570 nm and 561 nm/575-670 nm. To avoid cross talk, imaging was done sequentially. All images were adapted to the full dynamic range of the system (8 bit). Co-localization was imaged using Leica Multicolour and Dye Finder modules (Leica).
Results
TF localization in cell lysates normalized by protein concentration
To investigate whether TF is enriched in rafts from total cell lysates, we loaded similar amounts of protein from all gradient fractions on gel, which was used for Western blotting of TF, flotillin and caveolin (Figure 1A). TF, flotillin and caveolin were present in fractions 2-4, indicating that TF, flotillin and caveolin are enriched in these fractions and suggesting that the TF antigen is entirely associated with caveolar rafts. These data confirm the results reported by del Conde et al7. Subsequently, we also determined the absolute quantity of TF
that is present in each gradient fraction using a TF ELISA. Surprisingly, less than 10% of total TF was present in fractions 2-4, whereas the bulk of TF localized in fractions 7-9 (Figure 1B). The absence of the TF antigen in fractions 7-9 on Western blot (Figure 1B) is explained by the high protein concentrations in these fractions, shown in the insert in Figure 1B. Since loading of an equal amount of protein of each fraction on gel does not reflect the real (absolute) distribution of TF, in all following experiments the fractions were normalized per volume, unless indicated otherwise.
Figure 1. Distribution of TF in cell lysates.
A. Western blots incubated with anti-TF, anti-flotillin (general marker of rafts) and anti-caveolin (marker of caveolar rafts) from a representative experiment, showing their relative distribution in Triton X-100 OptiPrep gradient fractions of cell lysates (fractions normalized per µg protein). B. The absolute quantities of TF in Triton X-100 OptiPrep gradient fractions of cell lysates as determined by ELISA (fractions normalized per volume). Insert B: The protein concentration of each gradient fraction.
Absolute localization of TF antigen and activity in cell lysates
When similar volumes of each fraction were studied by Western blot, the bulk of caveolin was present in fractions 2-4, but also fractions 8-9 contained small bands of caveolin, suggesting that cell lysates may actually contain two different types of caveolar rafts; type I (fractions 2-4) and type II (fractions 8-9) (Figure 2A). The TF antigen was present in fractions 3-9, but the bulk of TF antigen (>90%) was present in fractions 7-9 (Figure 2A and B), fractions which also contained type II caveolar rafts, flotillin, PDI and coatomer protein complex subunit beta (β-COP; a marker of Golgi membranes), suggesting that most of the TF in these fractions may be of intracellular membrane origin.
In addition, we investigated the localization of the coagulant form of TF by thrombin generation and fibrin formation. In all experiments (thrombin generation n=3, fibrin formation n=2), fractions 5-6 contained TF/factor VII-dependent coagulant activity (Figures 2C and D). Fraction Cell lysate (µg) TF Caveolin Flotillin 1 2 3 4 5 6 7 8 9 Fraction A B 1 2 3 4 5 6 7 8 9 0 5 000 10000 15000 20000 Fract ion p rot e in ( µ g /m L ) 1 2 3 4 5 6 7 8 9 0 10000 20000 30000 TF a n ti ge n ( pg) Fraction Cell lysate (µg) TF Caveolin Flotillin 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Fraction A B 1 2 3 4 5 6 7 8 9 0 5 000 10000 15000 20000 Fract ion p rot e in ( µ g /m L ) 1 2 3 4 5 6 7 8 9 0 10000 20000 30000 TF a n ti ge n ( pg)
8
Figure 2. Distribution of TF antigen and activity, flotillin, caveolin and PDI in cell lysates.
A. Western blot incubated with anti-TF, anti-flotillin, anti-caveolin, anti-PDI and anti-β-COP, showing their absolute distribution in Triton X-100 OptiPrep gradient fractions of cell lysates (fractions normalized per volume). B. TF present in each Triton X-100 OptiPrep gradient fraction expressed as percentage of the total amount of TF in cell lysates, as determined by ELISA (fractions normalized per volume; n=3). C-D. The coagulant activity of TF from the Triton X-100 OptiPrep gradient fractions determined by thrombin generation (C; n=3) and fibrin formation (D; n=2). Of each fraction, thrombin generation and fibrin formation are expressed as the factor of inhibition by anti-FVII of the area under curve (mean ± SD) or 50% Vmax (mean), respectively.
A C TF Caveolin Flotillin PD I 1 2 3 4 5 6 7 8 9 Fraction ß-C OP 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 Fraction TF a n ti ge n (% ) 1 2 3 4 5 6 7 8 9 0.0 0.5 1.0 1.5 th ro m b in ge n e ra ti on (i n h ib it ion by a n ti F V II ) Fraction Fraction 1 2 3 4 5 6 7 8 9 0 2 4 6 8 10 fi br in g e ne ra ti on (i nhi bi ti o n b y a n ti F V II ) B D A C TF Caveolin Flotillin PD I 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Fraction ß-C OP 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 Fraction TF a n ti ge n (% ) 1 2 3 4 5 6 7 8 9 0.0 0.5 1.0 1.5 th ro m b in ge n e ra ti on (i n h ib it ion by a n ti F V II ) Fraction Fraction 1 2 3 4 5 6 7 8 9 0 2 4 6 8 10 fi br in g e ne ra ti on (i nhi bi ti o n b y a n ti F V II ) B D
Evidently, the bulk of the TF antigen and the coagulant activity of TF are present in different fractions. Furthermore, the coagulant form of TF co-localizes hardly with rafts or PDI, whereas the non-coagulant form of TF co-localizes with (type II) caveolar rafts and PDI.
Co-localization of TF, caveolin and PDI
Since lysates of total cells contain all cellular compounds, the real localization of surface exposed TF is difficult to extrapolate from such studies. Therefore, we investigated the co-localization of TF with caveolin (Figures 3A and 3B) and PDI (Figure 3C) by confocal microscopy. Figure 3A shows two representative cells, demonstrating that TF (green) is localized both at the cell membrane and within the cells. Caveolin (red) is localized mainly in the nucleus, and more weakly in the cytoplasm and at the cell membrane. Co-localization of TF and caveolin (yellow) was absent or at best very faint. When we used a different antibody to detect TF (Figure 3B), again no co-localization of TF and caveolin was visible. PDI (Figure 3C, red) was detectable only in the cytoplasm. Thus, also no co-localization of TF and PDI was detectable. However, since only a minor fraction of TF may be present in the plasma membrane, the concentration of TF in these membranes may be too low to detect co-localization with either caveolin or PDI. To overcome this problem, we investigated the distribution of TF antigen, TF activity and PDI in purified plasma membranes.
8
Figure 3. Co-localization of TF, caveolin and PDI.
The co-localization of TF with caveolin (n=10, A and B) and the co-localization of TF with PDI (n=5, C) were investigated by confocal microscopy. Cells were incubated with anti-TF monoclonal antibody (green label) from Becton Dickinson (A) or American Diagnostica (B) plus polyclonal anti-caveolin (red label). C. Cells incubated with polyclonal anti-TF (green label, American Diagnostica) and monoclonal anti-PDI (red label). localization would show up as a yellow label. Co-localization is not substantially present, and the yellow color due to the overlap of the two cells demonstrates the validity of the approach.
A B C A B C
Localization of TF in plasma membranes
First, we studied whether TF is enriched in caveolar rafts in purified plasma membranes by loading equal amounts of protein on gel and blotted for TF, caveolin and flotillin. TF, caveolin and flotillin, however, were all below detection limits (data not shown).
To investigate the absolute localization of plasma membrane-associated TF, we loaded similar volumes of each fraction on gel and blotted for TF, flotillin, caveolin, PDI and β-COP. In purified plasma membranes, caveolin was most prominent present in fractions 7-8, indicating that especially the type II caveolar raft is present in plasma membranes (Figure 4A). Most of the TF antigen (>90%; Figures 4A and B) was present in fractions 7-8, which also contained flotillin, caveolin and PDI. This implicates that the bulk of TF antigen and PDI co-localize with caveolar rafts (type II) in plasma membranes. β-COP was below the detection limit, confirming the successful purification of the plasma membranes by removal of intracellular membranes. In all experiments (n=3), the TF antigen was below the detection limit of the Western blot in fraction 9, but was nevertheless clearly detectable with the TF ELISA. At present, we have no explanation for this discrepancy.
The coagulant activity of TF was determined by thrombin generation (Figure 4C) and fibrin formation (Figure 4D), and was present in fractions 2-6 but not in fractions 7-9. In contrast, flotillin, caveolin and PDI were absent or hardly detectable in fractions 2-6 (Figure 4A), indicating that the coagulant form of TF is not caveolar raft-associated and does not co-localize with PDI in purified plasma membranes.
Thus, plasma membranes concurrently contain the coagulant form of TF and the non-coagulant form of TF. The coagulant form of TF does not co-localize with caveolar rafts or PDI, whereas the non-coagulant form of TF strongly co-localizes with caveolar rafts and PDI.
8
Figure 4. Distribution of TF antigen and activity, flotillin, caveolin and PDI in purified plasma membranes.
A. Western blot incubated with anti-TF, anti-flotillin, anti-caveolin, anti-PDI and anti-β-COP, showing their absolute distribution in Triton X-100 OptiPrep gradient fractions of Percoll-gradient purified plasma membranes (fractions normalized per volume). B. TF present in each Triton X-100 OptiPrep gradient fraction, expressed as percentage of the total amount of TF in plasma membranes as determined by ELISA (fractions normalized per volume; n=3). C-D. The coagulant activity of TF from the Triton X-100 OptiPrep gradient fractions determined by thrombin generation (C; n=3) and fibrin formation (D; n=2). Of each fraction, thrombin generation and fibrin formation are expressed as the factor of inhibition by anti-FVII of the area under curve (mean ± SD) or 50% Vmax (mean), respectively. TF Caveolin Flotillin PDI 1 2 3 4 5 6 7 8 9 Fraction A ß-COP C D B 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 Fracti on TF an ti g e n (% ) 1 2 3 4 5 6 7 8 9 0.0 0.5 1.0 1.5 th ro m b in ge n e ra ti o n (i n h ib it io n b y a n ti F V II ) Fracti on 1 2 3 4 5 6 7 8 9 0 2 4 6 8 10 fi br in g e n e ra ti o n (i nh ib it io n b y a nti FV II ) Fraction D TF Caveolin Flotillin PDI 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Fraction A ß-COP C D B 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 Fracti on TF an ti g e n (% ) 1 2 3 4 5 6 7 8 9 0.0 0.5 1.0 1.5 th ro m b in ge n e ra ti o n (i n h ib it io n b y a n ti F V II ) th ro m b in ge n e ra ti o n (i n h ib it io n b y a n ti F V II ) Fracti on 1 2 3 4 5 6 7 8 9 0 2 4 6 8 10 fi br in g e n e ra ti o n (i nh ib it io n b y a nti FV II ) fi br in g e n e ra ti o n (i nh ib it io n b y a nti FV II ) Fraction D
Localization of TF in microparticles
Finally, we studied the distribution of TF within isolated microparticles. The TF antigen was present in fractions 7-8 on Western blot (Figure 5A) and in fractions 6-9 by ELISA (Figure 5B). These fractions also contained flotillin (fractions 6-9) and PDI (fractions 7-9), whereas caveolin and β-COP were not detectable (Figure 5A). Since caveolin was not detectable in microparticle fractions, it seems unlikely that microparticles contain substantial amounts of caveolar rafts. As β-COP was below the detection limit, we assume that flotillin, which can be present in rafts of intracellular as well as plasma membranes19,
originates mainly from plasma membrane-derived non-caveolar rafts. Thus, TF-containing microparticles contain mainly non-caveolar rafts.
Before OptiPrep fractionation, the microparticles strongly initiated FVII-dependent thrombin generation and fibrin formation (data not shown). After fractionation, the coagulant activity of TF was hardly detectable by thrombin generation (Figure 5C). In contrast, fraction 4 clearly triggered FVII-dependent fibrin formation (Figure 5D), indicating that the coagulant form of TF is present in a non-raft fraction. So, also microparticles contain the coagulant form of TF and the non-coagulant form of TF. The coagulant form of TF seems not to be associated with rafts or PDI, and the non-coagulant form of TF co-localizes with non-caveolar rafts and PDI.
8
Figure 5. Distribution of TF antigen and activity, flotillin, caveolin and PDI in microparticles.
A. Western blot incubated with anti-TF, anti-flotillin, anti-caveolin, anti-PDI and anti-β-COP, showing their absolute distribution in Triton X-100 OptiPrep gradient fractions of isolated microparticles (fractions normalized per volume). B. TF present in each Triton X-100 OptiPrep gradient fraction, expressed as percentage of the total amount of TF in microparticles as determined by ELISA (fractions normalized per volume; n=3). C-D. The coagulant activity of TF from the Triton X-100 OptiPrep gradient fractions determined by thrombin generation (C; n=3) and fibrin formation (D; n=2). Of each fraction, thrombin generation and fibrin formation are expressed as the factor of inhibition by anti-FVII of the area under curve (mean ± SD) or 50% Vmax (mean), respectively.
TF Caveolin PDI Flotillin 1 2 3 4 5 6 7 8 9 Fraction A ß-COP C D B 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 Fraction TF a n ti g e n (% ) Fraction 1 2 3 4 5 6 7 8 9 0.0 0.5 1.0 1.5 th ro m b in g e n e ra ti on (i n h ib it io n b y a n ti F V II ) 1 2 3 4 5 6 7 8 9 0 5 10 15 20 25 fi b ri n ge ne ra ti o n ( in hi bi ti on b y a n ti FV II ) Fraction TF Caveolin PDI Flotillin 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Fraction A ß-COP C D B 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 Fraction TF a n ti g e n (% ) Fraction 1 2 3 4 5 6 7 8 9 0.0 0.5 1.0 1.5 th ro m b in g e n e ra ti on (i n h ib it io n b y a n ti F V II ) 1 2 3 4 5 6 7 8 9 0 5 10 15 20 25 fi b ri n ge ne ra ti o n ( in hi bi ti on b y a n ti FV II ) Fraction
Discussion
Our results show that the representation of the raft results is evidently of paramount importance. In line with previous studies, we observed enrichment of TF in rafts when similar amounts of protein are loaded on gel. Presentation of data in this manner, however, does not reflect the true localization of TF nor its coagulant activity. Therefore, in the presented studies we used similar volumes of each fraction to determine the localization of TF.
We found that less than 10% of the total cellular TF is present in plasma membranes (Figure 6A). Furthermore, less than 10% of this plasma membrane-associated TF is capable of triggering extrinsic thrombin generation and fibrin formation, and this coagulant form of TF is not associated with rafts (Figure 6B). More than 90% of the plasma membrane-associated TF is present in a non-coagulant form, which localizes with caveolar rafts (Figure 6B), confirming earlier studies that caveolar rafts may be involved in maintaining TF in a non-coagulant state8;10;20.
Figure 6. Schematic representation of the distribution of TF antigen and activity in cells and rafts within the plasma membrane.
A. Distribution of TF present within cells and associated with the plasma membrane (PM). B. Distribution of the non-coagulant form of TF (TFnc) and the coagulant form of TF (TFc) in rafts within the plasma membrane.
A PM T F c T F nc rafts non-rafts B A PM T F c T F nc rafts non-rafts B
8
We found that the non-coagulant form of TF localizes in the same gradient fractions as PDI. At present, the role of PDI in the regulation of the coagulant activity of TF is still debated14-16. When we treated fractions containing the non-coagulant form of TF with
HgCl2 to mimic the oxidizing activity of PDI21, still no coagulant activity was detectable
(data not shown), indicating that a conformational change of TF induced by PDI is insufficient to convert non-coagulant TF into coagulant TF. We speculate that the conversion of the non-coagulant form of TF into the coagulant form of TF may involve two steps. First, the non-coagulant form of TF interacts with PDI within the caveolar rafts, resulting in a redox reaction-induced conformational change of TF. In the second step, this TF translocates from rafts to the non-raft membrane fraction, which is enriched in phosphatidylserine compared to rafts. In the presence of phosphatidylserine or other anionic phospholipids, TF thus becomes coagulant active.
We also investigated the distribution of TF in microparticles. Previously, del Conde et al. described that, compared to the cells, the monocyte-derived microparticles are enriched in TF and virtually depleted of the membrane antigen CD45. Since they found that TF and flotillin were present in the same raft fractions of monocyte lysates, they concluded that monocyte-derived microparticles are enriched in raft proteins. Furthermore, when they incubated monocytes with methyl-β-cyclodextrin to disrupt the rafts, they observed a reduced release of microparticles induced by calcium ionophore, from which they concluded that “monocyte-derived TF-bearing microvesicles arise from lipid rafts or from regions of high raft content”7. In that study, however, the presence of rafts in microparticles
was not studied directly. Similar to the findings of Versteeg et al.22, we also found that
microparticles contain detectable amounts of flotillin. In addition, we found no detectable amounts of caveolin, indicating that, if microparticels arise from raft areas in the plasma membrane, microparticles are likely to originate from non-caveolar rafts.
Our present findings show that microparticles contain coagulant as well as non-coagulant forms of TF. To which extent both forms of TF concurrently occur at the same microparticle, however, cannot be concluded from our present results. Given the fact that the non-coagulant form of TF also localizes in the same fractions as flotillin in microparticles, we suggest that rafts may not only suppress the coagulant activity of TF in
The coagulant form of TF seems not to be associated with rafts or PDI, but additional studies will be essential to elucidate the precise role of rafts and PDI on the regulation of the coagulant function of TF.
Acknowledgements
The authors want to thank prof. dr. J.W.N. Akkerman (Department of Clinical Chemistry and Haematology; University Medical Center Utrecht, Utrecht, The Netherlands) for providing a nitrogen bomb.
References
1. Mandal SK, Pendurthi UR, Rao LV. Cellular localization and trafficking of tissue factor. Blood 2006;107:4746-4753.
2. Simons K, Ikonen E. Functional rafts in cell membranes. Nature 1997;387:569-572. 3. Anderson RG. The caveolae membrane system. Annu.Rev.Biochem. 1998;67:199-225.
4. Smart EJ, Graf GA, McNiven MA et al. Caveolins, liquid-ordered domains, and signal transduction. Mol.Cell Biol. 1999;19:7289-7304.
5. Bickel PE, Scherer PE, Schnitzer JE et al. Flotillin and epidermal surface antigen define a new family of caveolae-associated integral membrane proteins. J.Biol.Chem. 1997;272:13793-13802.
6. Salzer U, Prohaska R. Stomatin, flotillin-1, and flotillin-2 are major integral proteins of erythrocyte lipid rafts. Blood 2001;97:1141-1143.
7. Del Conde I, Shrimpton CN, Thiagarajan P, Lopez JA. Tissue-factor-bearing microvesicles arise from lipid rafts and fuse with activated platelets to initiate coagulation. Blood 2005;106:1604-1611.
8. Dietzen DJ, Page KL, Tetzloff TA. Lipid rafts are necessary for tonic inhibition of cellular tissue factor procoagulant activity. Blood 2004;103:3038-3044.
9. Mandal SK, Iakhiaev A, Pendurthi UR, Rao LV. Acute cholesterol depletion impairs functional expression of tissue factor in fibroblasts: modulation of tissue factor activity by membrane cholesterol. Blood 2005;105:153-160.
10. Sevinsky JR, Rao LV, Ruf W. Ligand-induced protease receptor translocation into caveolae: a mechanism for regulating cell surface proteolysis of the tissue factor-dependent coagulation pathway. J.Cell Biol. 1996;133:293-304.
11. Ahamed J, Versteeg HH, Kerver M et al. Disulfide isomerization switches tissue factor from coagulation to cell signaling. Proc.Natl.Acad.Sci.U.S.A 2006;103:13932-13937.
12. Cho J, Furie BC, Coughlin SR, Furie B. A critical role for extracellular protein disulfide isomerase during thrombus formation in mice. J.Clin.Invest 2008;118:1123-1131.
13. Reinhardt C, von Bruhl ML, Manukyan D et al. Protein disulfide isomerase acts as an injury response signal that enhances fibrin generation via tissue factor activation. J.Clin.Invest 2008;118:1110-1122.
14. Bach RR, Monroe D. What is wrong with the allosteric disulfide bond hypothesis? Arterioscler.Thromb.Vasc.Biol. 2009;29:1997-1998.
15. Pendurthi UR, Ghosh S, Mandal SK, Rao LV. Tissue factor activation: is disulfide bond switching a regulatory mechanism? Blood 2007;110:3900-3908.
8
16. van Nispen Tot Pannerden HE, van Dijk SM, Du V, Heijnen HF. Platelet protein disulfide isomerase is localized in the dense tubular system and does not become surface expressed after activation. Blood 2009;114:4738-4740.
17. Abid Hussein MN, Meesters EW, Osmanovic N et al. Antigenic characterization of endothelial cell-derived microparticles and their detection ex vivo. J.Thromb.Haemost. 2003;1:2434-2443. 18. Fauvel J, Chap H, Roques V, Levy-Toledano S, Douste-Blazy L. Biochemical characterization
of plasma membranes and intracellular membranes isolated from human platelets using Percoll gradients. Biochim.Biophys.Acta 1986;856:155-164.
19. Gkantiragas I, Brugger B, Stuven E et al. Sphingomyelin-enriched microdomains at the Golgi complex. Mol.Biol.Cell 2001;12:1819-1833.
20. Mulder AB, Smit JW, Bom VJ et al. Association of smooth muscle cell tissue factor with caveolae. Blood 1996;88:1306-1313.
21. Chen VM, Ahamed J, Versteeg HH et al. Evidence for activation of tissue factor by an allosteric disulfide bond. Biochemistry 2006;45:12020-12028.
22. Versteeg HH, Ruf W. Tissue factor coagulant function is enhanced by protein-disulfide isomerase independent of oxidoreductase activity. J.Biol.Chem. 2007;282:25416-25424.
