| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cancer Biology Research Group,
* Department of Medical Science,
¶ Department of Biochemistry and Molecular Biology and Oncology, University of Calgary Calgary, Alberta, Canada T2N4N1
1Correspondence: Departments of Biochemistry and Molecular Biology and Oncology, Faculty of Medicine, University of Calgary, 3330 Hospital Dr., N.W., Calgary, Alberta T2N 4N1, Canada. E-mail: waisman{at}ucalgary.ca
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: annexin • S100A10 • p11 • plasminogen • plasmin • urokinase • cancer
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. In addition, plasmin activates several matrix-degrading metalloproteinases (MMPs) (3
, 4)
. Therefore, the generation of plasmin at the cell surface is a key event in the destruction of basement membrane and ECM that is necessary for the invasion of tumor cells through physiological barriers (5
, 6)
.
The presence of specific receptors for uPA and plasminogen at the cell surface is responsible for the spatial and temporal regulation of the conversion of plasminogen to plasmin (7
, 8)
. The cell surface receptor for uPA, uPAR, acts as a scaffold for the conversion of the zymogen pro-uPA to a catalytically active form, uPA. Subsequently, the bound uPA converts receptor-bound plasminogen to plasmin. Binding of plasminogen to its cell surface receptors is thought to be rate-limiting for efficient activation of plasminogen by uPA (9
, 8)
.
The identity of the cellular receptor(s) for plasminogen that participates in uPA-dependent plasminogen activation has been unclear. Typically, plasminogen binds to cells with low affinity (Kd=0.3–2 µM) and high capacity (104–107 binding sites per cell). The plasminogen binding sites on cells are heterogeneous in nature, and proteins and nonproteins such as glycosaminoglycans and gangliosides participate in plasminogen binding. However, a series of studies established the paradigm that only a small subset of cellular plasminogen receptors, those that possess a carboxy-terminal lysine residue, participate in cell surface plasminogen activation (reviewed in ref 10
). The identity of the plasminogen receptor(s) that participate in plasminogen activation has not been established. Candidate plasminogen receptors possessing carboxy-terminal lysines include p11 (11
12
13)
, cytokeratin-8 (14
15
16)
, TIP49a (17)
, and 
-enolase (18
19
20)
.
Mainly characterized as an intracellular protein, p11 (S100A10) is continuously expressed on the surface of different types of cells along with its binding partner, annexin II (12
, 21
, 22)
. Despite an abundance of in vitro kinetic data (reviewed in refs 23
, 24
), the issue of whether or not p11 plays an important role in the regulation of cellular plasmin generation or activity has never been addressed. To do so, we transfected human HT1080 fibrosarcoma cells with antisense or sense p11 cDNA and examined the effects of changes in the expression of the p11 on plasminogen activation. Our data establish that p11 accounts for 
80–90% of the plasmin generating capacity of the HT1080 cells. More important, these changes in the p11-regulated plasmin generation correlated with changes in cellular invasiveness and metastatic potential.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Plasmid transfection and retrovirus infection
The pLin-sp11 and pLin-ap11were transfected to a PA317 packaging cell line (ATCC, Manassas, VA, USA) using the calcium precipitation method. The packaging cells were enriched and selected by growth in the presence of 0.5 mg/mL G418 for 2 wk. At confluence, conditioned media were collected and used for transduction of HT1080 fibrosarcoma cells (ATCC, Manassas, VA, USA). The stably transduced HT1080 cells were selected with 300 µg/mL G418 for 2 months and clonal cell lines selected by p11 protein expression levels as assessed by Western analysis. The stable cell lines transduced with sense p11, antisense p11, or empty vector were termed S7, AS5, and Vec-1, respectively. HT1080 fibrosarcoma cells were transfected with pLin, pLin-sp11, or pLin-ap11 by using LIPOFECTAMINE 2000 Reagent (Gibco BRL, Rockville, MD, USA), selected with 300 µg/mL G418 (Bioshop Canada, Burlington, ON); clonal cell lines were selected by p11 protein expression.
Northern blot analysis
Total RNA was isolated from S7, AS5, and Vec cells using the RNeasy Kit (manufacturer’s instruction, QIAGEN, Valencia, CA, USA). Expression of p11 mRNA in these cells was examined according to the NorthernMax-Gly procedure (Ambion, Austin, TX, USA). Total RNA (20 µg) was electrophoretically separated by glyoxal agarose gel electrophoresis and transferred to a nylon membrane. Hybridization was performed with a radioactive cDNA probe for p11 (106 cpm/mL) labeled with 32P--dCTP using Ready-To-Go DNA Labeling Beads (Amersham Pharmacia Biotech, Piscataway, NJ, USA).
Immunoblot analysis
Cells were lysed with RIPA buffer containing 1 mM PMSF, 5 µg/mL leupeptin, and 5 µg/mL aprotinin. The cell lysates (25 µg), reconstituted in reducing or nonreducing Laemmli buffer, were subjected to SDS-PAGE and transferred to a nitrocellulose membrane. To detect p11, the membrane was fixed in 4% paraformaldehyde at room temperature for 10 min. After blocking with a 5% skim milk solution, the membrane was incubated with 0.25 ng/mL of an anti-human p11 mAb (BD Transduction Laboratories, San Jose, CA, USA), 0.5 ng/mL of an anti-human uPA pAb (American Diagnostica, Greenwich, CT, USA), 0.5 ng/mL of an anti-human tPA mAb (American Diagnostica), 0.5 ng/ mL of an anti-human PAI-1 mAb (American Diagnostica), 0.5 ng/ mL of an anti-human PAI-2 mAb (American Diagnostica), 0.5 ng/ mL of an anti-human uPAR pAb (Santa Cruz Biotechnology, Santa Cruz, CA, USA), or 1 ng/mL of an anti-human -tubulin mAb (Oncogene Science, Mineola, NY, USA). These mAbs were detected with 0.2 ng/mL horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG and a Super signal detection kit (Pierce, Rockford, IL, USA).
Detection of cell surface annexin II and p11
Subconfluent cells were biotinylated by incubation with DPBS (Gibco BRL, Rockville, MD, USA) containing 0.5 mg/mL Sulfo-NHS-Biotin (Pierce) for 30 min at room temperature. After washing with DPBS, cells were lysed with RIPA buffer containing 1 mM phenylmethylsulfonylfluoride (PMSF), 1 µg/mL leupeptin, and 1 µg /mL aprotinin. The lysate was incubated with 20 µL of streptavidin-agarose beads (Sigma) or control agarose beads at 4°C overnight. After washing with ice-cold TBS containing 0.5 mM CaCl2, Tween-20, and 1 mM PMSF, bound proteins were eluted with SDS-sample buffer and subjected to immunoblot analysis using an anti-annexin II mAb (Transduction Laboratories, San Jose, CA, USA).
For p11 detection, cells were surface-biotinylated and the total cell extract was precleared with 20 µL of protein A/G Plus-agarose beads (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The precleared sample was incubated with 4 µg of anti-p11 mAb (Transduction laboratory) on ice overnight. The mixture was subsequently incubated at 4°C with 50 µL of protein A/G Plus-agarose beads at 4°C for 4 h on a platform rotator, centrifuged, and the pellet was washed five times with TBS containing 0.5 mM CaCl2, 0.05% Tween-20, and 1 mM PMSF. The pellet was then subjected to SDS-PAGE and probed with streptavidin HRP.
Immunofluorescence staining
HT1080 cells were cultured on fibronectin-coated glass coverslips. To stain for surface antigens, cells were washed twice with DPBS and fixed with 4% paraformaldehyde in PBS for 10 min at 4°C. To detect intracellular antigens, some cells were permeabilized with cold 100% methanol for 10 min at 4°C. After washing with DPBS, cells were blocked with 3% bovine serum albumin in PBS for 1 h at 4°C, incubated with primary antibody for 1 h, washed four times with DPBS, and incubated with cy3-conjugated second antibody. Cells were washed and mounted with Prolong Antifade (Molecular Probes, Eugene, OR, USA) reagent, observed on a Zeiss Axioskop microscope (Oberkochen, Germany), and visualized using a Zeiss Axioskop microscope (Oberkochen, Germany) and a digital camera (RS Photometrics, Tucson, AZ, USA). Confocal microscopy was performed on a Lica DM RXA2 microscope (Princeton Instruments Scientific), CCD camera (ST-138 controller KAF1600 chip) with detector cooled to -40°C. Analysis was performed with a computer-controlled camera and microscope with in-house software, C++ Snapin for Winview (Princeton Instruments). Typically, 0.3 µm optical slices were examined.
Cell surface plasminogen binding assay
Recombinant human [Glu]plasminogen (American Diagnostica) was radioiodinated as described previously (25)
. Plasminogen (10 µM) was incubated for 3 min at room temperature with 37 MBq (=1 mCi) of Na125I and three Iodo-Beads (Pierce) in PBS. Free Na125I and protein were separated using a PD-10 column (Sephadex G-25, Amersham Pharmacia Biotech, Uppsala, Sweden) equilibrated, and eluted with PBS. The specific activity of the protein preparations ranged from 1000 to 2000 cpm/pmol of protein. Confluent cells in 24-well plates were rinsed with ice-cold DPBS and incubated with 10 nM radioiodinated plasminogen and 0.49 µM cold plasminogen in the presence or absence of 10 mM 
-ACA at 4°C for 1 h. Radioactivity of cells was counted after washing the cells with ice-cold DPBS three times.
Plasminogen activation assay on cell surface
The kinetics of cell-mediated plasminogen activation was determined by measuring amidolytic activity of the plasmin generated from plasminogen. Cells were seeded on 24-well culture plates. The reaction was conducted with the substrate H-D-norleucyl-hexahydrotyrosyl-lysine-p-nitroanilide (Spectrozyme #251, American Diagnostica) at a final concentration of 100 µM in phenol red-free DMEM. The reaction was initiated by the addition of 0.5 µM [Glu]-plasminogen to 80% confluent cells and was monitored at 405 nm in a Perkin Elmer HTS 7000 Bioassay reader (Shelton, CT, USA).
Zymography
Confluent cells (1x105/well) were cultured in serum-free DMEM in the presence or absence of 0.5 µM human [Glu]plasminogen. Cell debris was removed from the collected supernatant by centrifugation (2000 g, 4°C, 10 min). Cell-bound PA was prepared by incubation of the cells with 200 µL of 50 mM glycine-HCl in 0.1 M NaCl (pH 3.0) for 3 min at room temperature and consequent neutralization with 50 µL of 0.5 M Tris-HCl buffer (pH 7.8). The eluate was collected and cleared by centrifugation. The conditioned medium was subjected to nonreduced SDS-PAGE containing 1 mg/mL of gelatin in the absence or presence of 13 µg/mL plasminogen. After electrophoresis, the gels were washed twice with 200 mL of 50 mM Tris-HCl, 150 mM NaCl, 2.5% (v/v) Triton X-100, pH 7.4 for 2 h at room temperature and three times with water for 5 min. Gelatin gels were incubated at 37°C overnight in 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 2.5% (v/v) Triton X-100. Gels containing plasminogen were incubated in 100 mM glycine-NaOH, pH 8.3, for 3 h at 37°C. Gels were stained with Coomassie Brilliant Blue R-250 0.5% (w/v) in 45% (v/v) methanol, 10% (v/v) acetic acid and destained in the same solution without dye. Protease activity was visualized as a clear zone against a blue background.
Invasion assay
Cell invasiveness was assayed using a modified Transwell system with Matrigel-coated polycarbonate filters (8 µm pore size, Costar, Cambridge, MA, USA). After rehydration of filters, cells (2x105) were seeded in the upper chamber in DMEM containing 10% (v/v) FBS. DMEM supplemented with 10% (v/v) FBS was placed in the lower chamber. After 5 h of incubation at 37°C in 5% CO2/95% air (v/v), the media in the upper chamber was changed with serum-free DMEM containing 0.25 µM [Glu]plasminogen. After 18 h of incubation, cells in the lower chamber were stained with Diff-Quick Stain Set (DADE International, Miami, FL, USA) and counted under microscopic fields.
Proteolysis of smooth muscle cell-derived ECM
Human aortic smooth muscle cells (AoMC, Clonetics, San Diego, CA, USA) were grown in AoMC growth media provided by the manufacturer. AoMC were seeded in 24-well tissue culture plates (1x105), after which the media was supplemented with 1 µCi/mL [3H]-glucosamine HCl (Perkin Elmer Life Sciences, Boston, MA), USA. Five days later, cells were removed by addition of 0.25 mM NH4OH, 0.5% Triton X-100 for 5 min at room temperature. The labeled ECM was washed with DPBS three times and stored at 4°C. Cells (1x105/well) were plated on the rehydrated matrix in DMEM containing 10% FBS and incubated for 5 h. After being washed with serum-free DMEM, the cells were incubated with serum-free DMEM with or without 0.5 µM plasminogen at 37°C in 5% CO2/95% air (v/v) for various times. The media was then collected and subjected to liquid scintillation counting. To determine the value for 100% degradation, the matrix was subjected to hydrolysis with 10 µM trypsin.
Cell migration
Cell migration was conducted with the QCM Cell Migration Assay Kit (Chemicon International, Tumecula, CA, USA) and performed according to manufacturer’s instruction. Cells (5x104 cells/well) were seeded in DMEM containing 10% FBS and incubated for 5 h. Media in the upper chamber was changed with serum-free media with or without 0.2 µM [Glu]plasminogen and incubated for 4 or 15 h at 37°C in 5% CO2/95% air (v/v). The cells that had migrated to the lower chamber were detached, lysed, and detected by CyQuant GR dye.
Mouse lung metastasis
S7, AS5, and Vec-1 cells (1x106 cells/0.1 mL) were injected intravenously into the tail vein of SCID mice (n=6) (Jackson Laboratories, Bar Harbor, ME, USA). Mice were killed 5 wk after inoculation and extent of lung metastasis was determined as number of foci and lung mass.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Stable integration of the G418-p11 transgene in the cellular genome resulted in expression of the transgene mRNA by AS5 and S7 cell lines (Fig. 1A
). This resulted in an increase in total cellular levels of p11 in the S7 cells and a decrease in total cellular levels of p11 in the AS5 cell line (Fig. 1B
). The p11 protein level of the S7 clone was only slightly higher than the Vec-1 clone. As a control, the protein level of tubulin was similar in the three cell lines (Fig. 1B
). The protein level of annexin II (p36), the primary binding partner of p11, was similar.
|
To confirm that the cell lines had altered extracellular levels of p11, we labeled cell surface proteins with a biotin-conjugated, protein-labeling reagent, Sulfo-NHS-Biotin, collected the biotin-labeled proteins by avidin affinity chromatography, and immunoprecipitated p11 from the crude extract of biotinylated surface proteins. We found that the S7 clone contained higher extracellular levels of p11 than the Vec-1 cells, whereas p11 was virtually undetectable on the surface of the AS5 cell line (Fig. 1C
). In contrast, extracellular levels of annexin II remained unchanged in the cell lines.
Viral transduction of cells has been reported to inappropriately activate the transcription of cellular genes unrelated to the transgene (26)
. Therefore, we used Lipofectamine reagent to transfect the HT1080 cells with pLin-ap11, pLin-sp11, and the empty pLin vector. The G 418-resistant cells were cloned and stable cell lines were established. We established two p11 antisense clonal cell lines (AS6, AS7), three p11-sense clonal cell lines (S8, S9, S10), and vector control cells (Vec-2) by this procedure. Western blot analysis demonstrated that compared with the Vec-2 cells or untransfected cells, the pLin-sp11 cell lines had increased levels of p11 (Fig. 1D
). We were unable to detect p11 levels in the p11 antisense cell lines. We found that the total cellular levels of annexin II remained essentially constant in all cell lines.
Clonal selection of cancer cells such as HT1080 can result in the isolation of different clones with considerable variation of expression of proteins involved in the plasminogen activation cascade (27)
. Although our selection of the cell lines was based on their p11 levels, we could not rule out the possibility we had inadvertently selected cells with altered levels of plasminogen activators or plasminogen activator inhibitors. However, as shown in Fig. 1E
, Western blot analysis of the cell lines established that these cell lines had comparable protein levels of PAI-1, PAI-2, tPA, uPA, and uPAR.
We analyzed the extracellular levels and distribution of p11 by immunofluorescence microscopy of the HT1080 cell lines. As exemplified by the Vec-1 cells, extracellular p11 was observed in discrete patch-like structures on the cell surface whereas a typical cytoplasmic staining pattern was observed for intracellular p11. A similar staining pattern has been observed for the cell surface p11 of breast carcinoma, glioma, and HUVEC cells (23
, 28)
. We observed that p11 antisense cell lines exhibited a loss of p11 from the cell surface. For example, p11 was barely detectable on the extracellular surface of the AS5 cell line (Fig. 2
). In contrast, the p11-sense cell lines showed significant increase in extracellular p11; for example, the S7 clone showed more intense immunoreactivity than the Vec-1 cells. Since only the permeabilized cells stained for the Golgi protein golgin-97, the extracellular localization of p11 could not be attributed to the inadvertent permeabilization of the cells during the staining process. We stained the HT1080 cells for uPAR. As shown in Fig. 2
, we noted that p11 and uPAR colocalized on the surface of these cells. Collectively, these data confirmed that extracellular protein levels of p11 were lowered in the three p11 antisense cells lines and increased in the four p11-sense cell lines. Development of the stable HT1080 cell lines with altered cell surface p11 afforded the opportunity to examine the role of p11 in plasminogen binding and activation.
|
p11 modulates cellular plasmin generation
The binding of plasminogen to cells involves interaction of the lysine binding kringle domains of plasminogen with the carboxy-terminal lysines of plasminogen receptors. This binding interaction can be inhibited by lysine analogs such as -aminocaproic acid (-ACA) or by CpB treatment, which removes the carboxy-terminal lysines of plasminogen receptors. To examine whether changes in the extracellular levels of p11 corresponded to changes in the plasminogen binding capacity of the HT1080 cells, we incubated representative cell lines with plasminogen and after washing the cells with buffer, determined the levels of bound plasminogen. As shown in Fig. 3A
, the S7 clone showed a 39% enhancement of plasminogen binding compared with the Vec-1 cells. Plasminogen binding to these cell lines was reduced by either -ACA treatment. Conversely, binding of plasminogen to the AS5 cells was reduced by 28% compared with the Vec-1 cells.
|
We then examined the ability of the cell lines to convert plasminogen to plasmin. Like many cancer cells, HT1080 cells constitutively secrete plasminogen activators. The cellular plasmin formation was monitored by washing the cells with serum-free media and the initial rates of plasmin formation were determined after addition of plasminogen and a colorimetric plasmin substrate. We found that plasmin production of the S7 and AS5 cell lines corresponded to their extracellular p11 levels over a wide range of plasminogen concentrations (Fig. 3B
). Next, we performed a comprehensive analysis of all our cell lines. Cell lines were incubated with a physiological plasminogen concentration of 2 µM and their plasmin generating activity was compared. We found that cellular plasmin formation by the S7 cells was threefold higher than the Vec-1 cells whereas plasmin production by the AS5 cells was inhibited by almost 80% compared with Vec-1 cells. Most striking, we found that plasminogen activation by AS6 and AS7 cell lines was barely detectable, showing an 95% reduction compared with the Vec-2 cells (Fig. 3C
). In contrast, the S8, S9, and S10 cell lines showed a 2.5-fold increase in plasmin generation compared with Vec-2 cells.
We measured proteolytic activity released into media by representative cell lines. Cells were incubated in the presence or absence of 0.5 µM plasminogen for 5 h and the conditioned media was subjected to gelatin zymography. In the absence of plasminogen, very little proteolytic activity was observed (Fig. 3D
). However, when the cell lines were incubated with plasminogen, the resultant increase in gelatinolytic activity was highest for the S7 clone and lowest for the AS5 clone. The molecular mass of the major band of proteolytic activity was 80 kDa, consistent with the release of predominantly plasmin into the media by the cells.
Regulation of plasmin activity by p11
In addition to cleaving several ECM proteins, plasmin is capable of autoproteolysis. This self-destruct mechanism is thought to be important in order to prevent collateral tissue damage by accumulation of plasmin in the tissues. Our in vitro studies suggested that besides stimulating plasmin production, p11 could stimulate plasmin autoproteolysis (13)
. When we incubated the cell lines with plasminogen and monitored levels of plasmin(ogen) protein by Western blot analysis (Fig. 4A
) and plasmin activity by gelatin zymography (Fig. 4B
), we observed time-dependent autoproteolysis of plasmin. Three hours after plasminogen addition, we found the S7 cell line had converted most of the exogenous plasminogen to plasmin (Fig. 4A
) and was expressing a high level of plasmin activity (Fig. 4B
). In contrast, the AS5 clone had not produced measurable levels of plasmin protein and its plasmin activity was not detectable. The control Vec-1 cells had produced a small amount of plasmin protein and consequently showed a low level of plasmin activity during this period. Twelve hours later, the plasmin protein produced by the S7 clone had been autoproteolyzed and its gelatinolytic activity had decreased. The AS5 clone had produced only a small amount of plasmin protein and displayed low levels of plasmin activity compared with the Vec-1 cells. These results establish that p11 stimulates plasmin formation and autoproteolysis on the cell surface.
|
Plasminogen activator activity of cell lines
HT1080 cells constitutively secrete tissue plasminogen activator (tPA) and the urokinase-type plasminogen activator (uPA). The uPA is considered the predominant plasminogen activator secreted by cancer cells. The uPA is secreted by cancer cells is in its single-chain proenzyme form, pro-uPA. The pro-uPA is rapidly converted to its active two-chain form by cell-bound plasmin (8)
. uPA produced by the action of plasmin on pro-uPA then converts plasminogen to plasmin, participating in an exponential double-reciprocal zymogen activation cycle.
We used zymography to compare the levels of plasminogen activator activity of our cell lines. These assays were performed with acid-washed cell extracts to detect cell-bound uPA or conditioned media to detect uPA released into the media. As reported by others, we detected uPA activity only by reverse zymography. We found that compared with untransfected HT1080 cells or the vector control cells, p11 antisense cell lines had lower levels of uPA activity whereas the sense p11 cell lines had elevated uPA activity (Fig. 5A, B
). Since the cell lines secrete similar levels of tPA and uPA protein (Fig. 1E
), this indicates that more pro-uPA was converted to uPA by the p11-sense cell lines whereas much less pro-uPA was converted to uPA by the p11 antisense cell lines. Thus, activation of pro-uPA by the cell lines is dependent on their plasmin levels. Therefore, p11 indirectly regulates the plasminogen activator activity of the cell lines by regulating plasmin levels of the cells and hence the plasmin-dependent conversion of pro-uPA to uPA.
|
p11 Regulates ECM hydrolysis by HT1080 cells
Smooth muscle cells elaborate an ECM (SM-ECM) rich in glycoproteins such as fibronectin and other structural proteins, such as collagen and elastin. The SM-ECM is a valuable model system for studying the kinetics and mechanism of destruction of the ECM by tumor cells (29)
. Many tumor cells (including HT1080 cells but not normal fibroblasts) degrade the SM-ECM. Labeled SM-ECM was produced by culturing rat smooth muscle cells with [3H]-glucosamine; after removal of the smooth muscle cells, the HT1080 cell lines were seeded on the matrix. Plasminogen was added and the radioactivity released from SM-ECM was monitored over time. As expected, the wild-type cells and vector control cells rapidly hydrolyzed the SM-ECM (Fig. 6
). Consistent with their decreased plasminogen activation capability, the amount of SM-ECM hydrolyzed by the p11 antisense cell lines was 30% that of the control cells. The p11-sense cell lines showed an threefold increase in SM-ECM hydrolysis vs. the control cells.
|
p11 Modulates cancer cell invasiveness
Invasion of cancer cells through the ECM involves activation of cell surface proteases as well as changes in cell motility. Since our data established that the ability of the HT1080 cells to convert plasminogen to plasmin was dependent on the extracellular levels of p11, it was important to examine the possibility that p11 could contribute to the invasiveness of the HT1080 cells. We used our HT1080 cell lines and a Matrigel assay system to study how p11 contributed to the invasiveness of the HT1080 cells. This assay system consists of an upper and lower chamber separated by an 8 µm pore polycarbonate filter coated with Matrigel, which acts as a basement membrane. The cell lines were seeded on the Matrigel in the upper chamber and the number of cells that transversed the Matrigel and appeared on the underside of the Matrigel was determined. When the experiment was performed in the absence of plasminogen, the invasiveness of the three representative cell lines was similar (Fig. 7A
). In contrast, we found that in the presence of plasminogen, the invasiveness of the AS5 cell line was decreased by 60% compared with the vector control cells whereas the invasiveness of the S7 cell line was increased by 2.5-fold. When the plasminogen-dependent rates of invasiveness were determined by subtracting the rates determined in the presence and absence of plasminogen, the plasminogen-dependent invasiveness of the AS5 cell line was decreased 92% whereas invasiveness of the S7 cell line was increased by 3.5-fold.
|
Changes in the invasiveness of the HT1080 cells could be due to changes in the cell surface proteolytic activity or motility of the cells, or both. The motility of the cell lines was examined by determining their migration through the polycarbonate filter in the absence of Matrigel. The cells were placed on the upper side of the 8-µm-pore sized filter and the number of viable cells present on the under of the filter was determined. As shown in Fig. 7B
, AS5, S7, and Vec-1 cells had similar migration rates, indicating that changes observed in cellular invasiveness were not due to differences in cellular motility.
The metastatic potential of cancer cells is often related to their proteolytic and invasive capacity. To determine whether the difference in protease-dependent invasiveness of the cell lines translated to differences in metastatic potential, we investigated the ability of the clonal cell lines to extravasate and form solid tumors in SCID mice. AS5, S7, or Vec-1 cells were injected into the tail vein of SCID mice and the formation of lung metastatic foci was monitored. We found that the number of metastatic foci in the lungs decreased by 3-fold for the AS5 cell line and by 16-fold for the S7 cell line compared with Vec-1 cells (Fig. 7C
). These results indicate that the ability of the tumor cells to intravasate and form tumors is directly related to p11 expression.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 5
, 30
, 31
)).
uPA is the primary plasminogen activator secreted by human HT1080 fibrosarcoma cells and is almost entirely responsible for the cell surface activation of plasminogen (8)
. The uPA is localized to the cell surface by virtue of its tight binding to its receptor, uPAR. Similarly, plasminogen binds to a heterogeneous group of protein and nonprotein receptors at the cell surface. Although interaction between the cell surface receptors for uPA and plasminogen plays a key role in the spatial and temporal regulation of plasmin generation, it is the plasminogen receptors that are rate-limiting for this reaction (9)
. Despite this critical role plasminogen receptors play in the regulation of plasmin production, little progress has been made in identifying the plasminogen receptor(s) directly involved in plasmin production.
The primary difficulty in studying the role of plasminogen receptors in plasmin production has been the clonal variation of expression of the genes for the plasminogen activators and plasminogen activator inhibitors within a population of cultured cancer cells. For example, cultured HT1080 cells have been shown to consist of distinct populations of clones secreting as much as a 10-fold difference in the levels of plasminogen activator and 15–fold differences in plasminogen activator inhibitors (27)
. Thus, transient transfection or inhibitory antibody strategies aimed at studying the regulation of cellular plasmin generation have been difficult to interpret because of this cellular heterogeneity. To circumvent this problem, we used clonal cell lines to study plasmin regulation in HT1080 cells. Each clonal cell line was extensively characterized and only cell lines with comparable protein levels of plasminogen activators—tPA and uPA, uPAR, and plasminogen activator inhibitors PAI-1 and PAI-2—were further characterized. Therefore, any changes in plasmin generation we observed could be correlated with changes in the levels of p11 in the clonal cell lines.
Regulation of cell surface plasmin generation by p11
In earlier studies we identified p11 as a plasminogen binding protein that stimulated the tPA- and uPA-catalyzed conversion of plasminogen to plasmin in vitro. However, it was unclear whether p11 played a role in plasminogen regulation in vivo. In the current report, we have shown that loss of p11 from the cell surface of HT1080 cells results in a loss of plasminogen binding (Fig. 3A)
and a dramatic loss of plasmin production by these cells (Fig. 3B, C)
. Depending on the method of delivery of the pLin-p11 antisense vector to the cells, the loss in plasmin activity ranged from 76% to 95%. On the other hand, we have found that increased extracellular p11 levels correspond to increased extracellular plasminogen binding as well as enhanced plasminogen activation (Fig. 3A, B, C)
.
Our observation that p11 could account for as much as 95% of the plasmin generated by HT1080 cells was unexpected. Whereas plasminogen may bind to all cell surface proteins with carboxyl-terminal lysine residues, only the binding of plasminogen to proteins with carboxyl-terminal residues that are colocalized to uPAR are thought to be important for plasminogen activation. Although the number of plasminogen receptors is 4 to 5 orders of magnitude higher than that of uPAR, bound plasminogen is rate-limiting for cell surface-mediated activation of plasminogen by urokinase (8
, 9)
. Our observation that loss in p11 from the surface of HT1080 cells results in an 28% loss in plasminogen binding but a 76% loss in plasmin production (Fig. 3A, C)
suggests that p11 may be one of the few plasminogen receptors that is accessible to uPAR. Considering the small increase in extracellular p11 observed between the S7 and Vec-1 cells, it was interesting that the rate of plasminogen conversion more than doubled. This result is compatible with the suggestion that the p11-plasminogen complex is rate-limiting in terms of uPA-dependent plasmin production in these cells.
Therefore, changes in the extracellular levels of p11 have such dramatic effects on plasmin production because of three key properties of p11. First, p11 possesses the requisite carboxy-terminal lysine residues that serve as the hallmark of a plasminogen receptor, and loss of these residues blocks the activity of the protein in vitro (32)
. Second, the interaction of p11 with plasminogen results in the induction of an activator-susceptible conformational change in plasminogen (12)
. Third, we have observed that p11 colocalizes with uPAR on the surface of HT1080 cells (Fig. 2)
. Therefore, p11 may be a potent regulator of plasminogen activation because it binds plasminogen via its carboxyl-terminal lysine residues, induces a conformational change in plasminogen, and colocalizes with uPAR.
In addition to the stimulation of plasmin production by p11, we have shown that p11 stimulates plasmin autoproteolysis in vitro, which presumably prevents collateral tissue damage that might occur due to action of constitutively activated cell-bound plasmin. We have now confirmed this observation in situ. As shown in Fig. 4A and B
, enhanced p11 levels on the surface of the HT1080 cells results in a transient pulse of plasmin activity due to the stimulation of plasmin production and autoproteolysis by p11. However, since the rates of plasmin autoproteolysis are determined in the absence of plasmin inhibitors that would be present in vivo, it is difficult to evaluate the physiological significance p11-dependent regulation of plasmin autoproteolysis.
Regulation of plasminogen activator activity by p11
Human tumor cells are commonly found to secrete the single-chain zymogen pro-uPA (33)
. Pro-uPA, which is rapidly converted to its active two-chain form by cell-bound plasmin, converts plasminogen to plasmin thus participating in an exponential activation cycle. It has been established that HT1080 cells require the presence of plasminogen to facilitate the conversion of cell surface pro-uPA to uPA (8)
. We observed that although the total cellular levels of uPA (pro-uPA and uPA) were similar among our cell lines, the antisense p11 cell lines showed much less uPA activity (Fig. 5)
. Therefore, this loss in the ability of the antisense p11 cell lines to convert pro-uPA to uPA was most likely secondary to the loss in the plasmin generating activity in these cell lines.
Role of annexin II in uPA-dependent plasminogen activation
The results of the current study show that plasminogen activation by HT1080 cells is regulated by extracellular p11. We observed in our antisense p11 cell lines that the cell surface annexin II remained unchanged (Fig. 1B)
despite the loss of as much as 90% of cellular plasminogen activation (Fig. 3B,C)
. This was surprising because it was originally reported that annexin II was an important regulator of plasminogen activation (34
, 35)
. These authors concluded that p11 was not present on the endothelial cell surface and that the cleavage of annexin II by an unidentified protease resulted in the generation of a new carboxy-terminal lysine residue on the protein. Subsequent reports have shown, however, that plasminogen binding to cells does not correlate with annexin II (36)
and that extracellular annexin II does not express carboxy-terminal lysine residues (17)
. An AIIt-cathepsin B complex has been immunoprecipitated from the surface of human breast carcinoma and glioma cells, and immunofluorescence staining of these cells has confirmed that p11 is present as AIIt-cathepsin B complex (22
, 28)
. Other studies have shown that the annexin II-p11 complex can be immunoprecipitated from the surface of endothelial cells (12)
. Therefore, based on the available evidence, it is unlikely that the annexin II plays a role in the uPA-dependent plasminogen activation. It is unclear, however, whether isolated p11 or p11 complexed with annexin II on the cell surface (AIIt) is responsible for the plasminogen regulation.
p11 Regulates cellular invasiveness
Metastasis requires dissolution of the ECM and basement membranes; this remodeling event is due to the activity of proteolytic enzymes. Among the proteases implicated in tumor cell dissemination is serine proteinase plasmin. By binding plasminogen to specific protein receptors on its surface and converting it to plasmin, the cell can harness an enzyme with broad substrate recognition to perform local proteolytic events necessary for tumor growth, invasion, and metastasis. As shown in Fig. 6
, 7A
, the loss in p11 from the cell surface dramatically decreases the matrix proteolysis and invasiveness of the cells. Most interesting was the observation that the loss in p11 from the extracellular surface corresponded to a loss in metastatic potential, i.e., the ability of the HT1080 cells to extravasate and form tumors in the mouse lung. Future studies will be necessary to determine whether this loss in metastatic potential is due solely to the decreased p11-dependent plasmin production by the cells. Finally, we demonstrated that increases in cell surface p11 increased the invasiveness and metastatic potential of the cell lines. Our observation that changes in extracellular p11 levels result in changes in plasmin generation is therefore consistent with the proposed relationship between plasmin activity and cancer cell invasiveness and metastasis potential. Although our data are consistent with a role for p11-mediated plasmin production in metastasis, we cannot rule out the possibility that p11 could affect metastasis by other mechanisms. For example, p11 might affect the production of other cellular proteases such as the matrix metalloproteinases or promote the increased survivability of the HT1080 cells in the circulation or in the tissue. Future studies are required to resolve this issue.
It was interesting that although the loss of p11 from the cell surface affected cell invasion, it failed to affect plasminogen-stimulated cell migration. It is likely that other plasminogen receptors may regulate plasminogen-stimulated cell migration.
A model for the role of p11 in uPA-dependent plasminogen activation
Although our studies establish that p11 is the major plasminogen receptor on HT1080 cells, further studies on other cell types will be required before these results can be considered of widespread physiological significance. Based on our current data, we propose a revision to the current model for the generation of plasmin at the cell surface of the HT1080 cell (37
38
39)
. First, plasminogen binds to its cell surface receptor, p11. This binding involves the interaction of the carboxy-terminal lysines of p11 with the lysine binding kringle domains of plasminogen. Second, the formation of cell-bound plasmin from plasminogen is initiated by the action of trace amounts of active uPA, which is bound to its receptor, uPAR. Third, the p11-bound plasmin catalyzes the activation of uPAR-bound pro-uPA. Fourth, the uPAR-bound uPA activates p11-bound plasminogen, resulting in the production of p11-bound plasmin, which activates additional uPAR-bound pro-uPA, thus initiating a feed-forward activation cascade.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 22, 2002. Accepted for publication October 17, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Laumonnier, T. Syrovets, L. Burysek, and T. Simmet Identification of the annexin A2 heterotetramer as a receptor for the plasmin-induced signaling in human peripheral monocytes Blood, April 15, 2006; 107(8): 3342 - 3349. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kwon, C.-S. Yoon, W. Jeong, S. G. Rhee, and D. M. Waisman Annexin A2-S100A10 Heterotetramer, a Novel Substrate of Thioredoxin J. Biol. Chem., June 24, 2005; 280(25): 23584 - 23592. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhang, D. K. Fogg, and D. M. Waisman RNA Interference-mediated Silencing of the S100A10 Gene Attenuates Plasmin Generation and Invasiveness of Colo 222 Colorectal Cancer Cells J. Biol. Chem., January 16, 2004; 279(3): 2053 - 2062. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. MacLeod, M. Kwon, N. R. Filipenko, and D. M. Waisman Phospholipid-associated Annexin A2-S100A10 Heterotetramer and Its Subunits: CHARACTERIZATION OF THE INTERACTION WITH TISSUE PLASMINOGEN ACTIVATOR, PLASMINOGEN, AND PLASMIN J. Biol. Chem., July 3, 2003; 278(28): 25577 - 25584. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||
