THROMBOSIS RESEARCH 67; 589-599,1992 0049-3848192 $5.00 + .OOPrinted in the USA. Copyright (c) 1992 Pergamon Press Ltd. All rights reserved.
BINDING OF RECOMBINANT VARIANTS OF HUNAN TISSUE-TYPE PLASMINOGEN ACTIVATOR (t-PA) TO HUNAN UMBILICAL VEIN ENDOTHELIAL CELLS
H.Stockinger', M. Kubbies, Boehringer Germany
Mannheim
R.Rudolph, A. Stern, U. Kohnert and S. Fischer GmbH, Biochemical Research Center, Penzberg,
(Received 27.2.1992; accepted in revised form 15.7.1992 by Editor H HCirmann)
ABSTRACT Endothelial cells synthesize and secrete hemostatic components like tissue-type plasminogen activator (t-PA) which is thought to be the major determinant of fibrinolytic activity in the blood. Most recently, a receptor on human protein for t-PA umbilical vein endothelial cells (HWEC) in culture has been described (1); there are, however, in addition low affinity binding sites for t-PA on HWEC. The sites of binding are of particular interest, because they are potential regulators of t-PA activity and clearance. We analysed the low affinity binding of recombinant t-PA (rt-PA) to normal diploid HWEC and to the permanent human cell lines Jurkat, Daudi, HL 60 and K562 by flow cytometry applying t-PA specific monoclonal antibodies. Using this test system binding of both recombinant glycosylated human t-PA produced in Chinese hamster ovary cells (CHO-t-PA) and of nonglycosylated t-PA, produced in E. coli (BM 06.021) was investigated. Analysis of the binding pattern to HWEC and other cell lines revealed that deglycosylation of full length rt-PA increases non-specific binding. Additionally, we investigated the binding properties of an unglycosylated t-PA deletion variant which comprises the kringle 2 and the (BM 06.022). Data obtained show that protease domains deletion of these domains most drastically reduces nonand other human cell lines. specific binding to HWEC
__ plasminogen activator tissue-type worcas: recombinant Key endothelial cells - human cell lines - flow cytometric analysis non-specific binding "corresponding Author 589
590
CELL-BINDING OF &PA VARIANTS
Vol. 67, No. 5
INTRODUCTION Thrombolytic therapy is an effective treatment for acute myocardial infarction (2). In this respect, tissue-type plasminogen activator (t-PA) has been shown to dissolve artery thrombi thereby improving myocardial function. Recently, t-PA produced by recombinant DNA technology (rt-PA) has become a widely used drug for the treatment of thromboembolic diseases (3-5). Human t-PA is a multidomain serine protease which exerts its dominant role during fibrinolysis by converting plasminogen into plasmin (6). The proteolytically active site of t-PA resides in the B-chain, whereas the A-chain consists of the finger domain, the EGF-domain and two kringle domains (7-9). The EGF/finger and kringle domains of the A-chain probably contain fibrin binding sites of t-PA. One of the major problems encountered upon therapeutic use of rt-PA is its rapid hepatic elimination from the plasma characterized by a half life of is probably about three minutes (10,ll). The rapid clearance mediated by a carbohydrate specific receptor on liver endothelial cells and by a sofar unknown receptor on liver parenchymal cells (12). Consequently, a nonglycosylated full length rt-PA from E. coli (BM 06.021) displays a longer half-life and a lower clearance in vivo than CHO-t-PA, the rt-PA from Chinese hamster ovary cells (13). In addition to binding of t-PA by liver parenchymal cells, it has been shown that human umbilical vein endothelial cells (HWEC) possess two types of t-PA acceptor sites with different affinity for t-PA (14). The sites of binding of t-PA to endothelial cells are of particular interest because they are potential regulators of t-PA activity and are probably involved in removing t-PA from the circulation (15). The aim of the present study was to investigate the binding of externally added full length glycosylated rt-PA produced from mammalian cell culture (CHO-t-PA; (16)) or of unglycosylated rt-PA produced in E. coli (BM 06.021; (17)) to HWEC and established human cell lines in vitro. The binding pattern of CHO-t-PA and BM 06.021 was compared with the binding of a recombinant nonglycosylated t-PA molecule produced in E. coli (BM 06.022) lacking the finger/EGF and kringle 1 domains (5). The results of this study should give us an insight which cell types might be involved in the clearance of rt-PA in vivo. MATERIALS
AND METHODS
Cell Culture. HWEC were isolated according to standard protocols (18) using 0,2 % collagenase Type II (Sigma). Cells were grown to confluence on gelatin-coated petri culture dishes and subsequently passaged. After each passage, aliquots of cells were frozen in Ml99 medium supplemented with 20 % fetal calf serum and 7 % dimethyl sulfoxide and stored in liquid nitrogen. After thawing, cells were grown to confluence in a mixture of RPM1 1640/M199 medium (50/50) supplemented with 20% fetal calf serum and heparin (10 mg/ml) and endothelial cell growth factor (20 pg/ml). Only cells from the 8th to 10th passage were used for binding studies. Endothelial cells were characterized by the presence of von Willebrand Factor (vWF)(19) and angiotensin converting enzyme (ACE)(ZO) in the culture supernatant, using an ACE test kit and a vWF ELISA test kit
Vol. 67, No. 5
CELL-BINDING OF rt-PA VARIANTS
591
(Boehringer Mannheim). The percentage of HWEC in the cell culture determined by using a factor VIII-specific antibody (Pan Systems, Germany) was in the range of 50 - 70%. Human cell lines. The T-cell line Jurkat, the B-cell line Daudi, the promyelocytic cell line HL 60 and the erythroleukemic cell line K562 were cultivated in RPM1 1640 containing 10% fetal calf serum. Flow cvtometric analvsis. HWEC were grown to confluence. Afterwards, medium was removed and the cell layer was washed three times in phosphate buffered saline (PBS) to remove fetal calf serum. The rt-PA molecules were added to the cell culture at various concentrations and incubated at 37°C for different times. Subsequently, cells were washed again in cold PBS and incubated on ice with the respective t-PA specific monoclonal antibodies (10 /.&g/ml). Thereafter, cells were washed again and a fluorescein isothiocyanate (FITC) labeled anti mouse Ig antibody was added. After incubating the samples for 30 minutes on ice, cells were washed again in cold PBS, detached from culture plates with 0.2% ethylenediaminetetraacetate (EDTA, pH 8). Bound t-PA was determined by flow cytometry. Flow cytometric analysis was performed using an Ortho Cytofluorograph 50H (Ortho Instruments, MA) connected to a 2151 computer for data aquisition and storage. The cells were excited the with 488 nm line forward/side and scatter (50 mw) characteristic was used to gate on viable cells (21). FITC fluorescence was selected using a 510130 nm bandpass. The histograms obtained were transferred to a PC/AT computer and a subtraction analysis of the controls and positive cells was performed using the MPLUS software (Phoenix Flow Systems, Sorrento Valley, CA). Monoclonal antibodies. The binding of rt-PA's to HWEC or other human cell lines was determined with the following epitope-specific antibodies: B-chain specific: EPA-BC monoclonal (Boehringer Mannheim), MA2-G6 (22), TCl VPA and TC2 VPA (all from Technoclone, Vienna, Austria). A-chain specific: MAl-C8 prevents binding of t-PA to clots generated in normal human plasma (22), TC4 VPA and TC7 VPA are kringle 2 specific and TC3 VPA and TC6 VPA are finger and EGF like domain specific (all from Technoclone). TCl VPA and TC4 VPA are non-competitive inhibitors of plasminogen activation in the presence of CNBr fragments of fibrinogen while TC3 VPA and TC7 VPA are competitive inhibitors. t-PA molecules used for bindins studies. CHO-t-PA was isolated and purified from the supernatant of transfected Chinese Hamster Ovary BM cells (16). Full length rt-PA BM 06.021 and the deletion Variant 06.022 were produced in E. coli (5,13). During the production process initially insoluble aggregates are solubilized, refolded in vitro and purified as described (5). The specific activity of CHOt-PA was 977 units per mg and ml, of BM 06.021 was 800 units pe!rmg and ml and of BM 06.022 was 542 units per mg and ml. RESULTS Bindinc analvsis bv flow cytometry of BM 06.021 to HWEC In Fig. 1 the flow cytometric histograms of the anti-t-PA/anti
CELL-BINDING OF rt-PA VARIANTS
592
Vol. 67, No. 5
mouse FITC fluorescence is shown on a linear scale after gating on intact cells according to their scatter characteristics (excluding unspecific binding of monoclonal antibodies to dead cells;)(21). In panel A and B the broken lines correspond to the controls with anti t-PA/anti mouse FITC incubation without rt-PA preincubation of labelling shows HUVEC. Addition of rt-PA prior to antibody (solid increasing FITC fluorescence intensities of the cells lines), significantly overlapping with the control region. A subtraction technique of the two corresponding histograms was used to quantitate the number of positive cells (shadowed areas). In principle, the two different monoclonal antibodies for the B-chain (EPA-BC, panel A) and for the A-chain (MAl-C8, panel B) give similar results of about 60% positively labeled cells. Using a dual laser technique and Hoechst 33342 viable labelling of cellular DNA (23) we found no cell-cycle specific expression of t-PA binding sites to HUVEC (data not shown).
rl
140+
i
i
FLUORESCENCE
INTENSITY
FIG. 1 Flow cytometric analysis of the t-PA binding to HUVEC using monoclonal antibodies to t-PA followed by FITC conjugated anti mouse Ig, respectively. The control cells were not preincubated with rt-PA (broken lines). The FITC-fluorescence histograms were recorded on linear scale (x-axis: channel number; y-axis: cell number. Panel A: EPA-BC (Bchain specific); panel B: MAl-C8 (A-chain specific). Flow cvtometric analysis of the bindinu kinetics of CHO-t-PA. BM 06.021 and BM 06.022 to HUVEC HUVEC (106/ml) were incubated with BM 06.021, BM 06.022 or CHO-t-PA (440 nM) at 37'C as described in Materials and Methods, and binding was followed up over 2 hours using flow cytometric analysis. Data in Fig. 2 show that there is a time-dependent increase of the BM 06.021 binding to HUVEC within the first hour. After 90 minutes a plateau is reached with a maximum binding of 60 % positive cells. This can be detected with antibody MAl-C8 (A-chain specific) or antibody EPA-BC (B-chain specific). A further increase in the
Vol. 67, No. 5
593
CELL-BINDING OF t-t-PA VARIANTS
incubation time did not increase the number of positive cells. The analysis of binding of CHO-t-PA (440 nM) to HWEC by flow cytometry can only be demonstrated with the antibody EPA-BC, which stains about 15% of HWEC. After incubation with BM 06.022 less than 10% of HWEC are positively stained with the antibodies MAl-C8 or EPABC. Non-specific binding of the rt-PA's to HWEC was further characterized by incubating the cells with increalsing concentrations (55-3600 nM) of BM 06.021, BM 06.022 or CHO-t-PA for 1 hour. Subsequently, binding of the rt-PA's was determined by flow cytometric analysis with the antibody EPA-BC. Data presented in Fig. 3 show that a saturation in the relative binding of BM 06.021 to 55% of HWEC is reached at about 2 PM. However, even the highest concentration of CHO-t-PA or BM 06.022 tested revealed only marginal binding of these rt-PA's to less than 20% of HWEC. The percentage of positively stained cells after incubation with CHO-tPA increases from about 10 % obtained with 55 nM to about 20 % positive cells obtained at a concentration of 3.5 PM. The binding of BM 06.022 to HWEC is below 10% of positive cells for all concentrations tested.
0
30
60 time
90
120
150
(min)
FIG. 2 Analysis of the binding kinetics of CHO-t-PA, BM 06.021 and BM 06.022 to HWEC. lo6 HWEC per ml were incubated for 30, 60, 90 and 120 minutes with 440 nM BM 06.021 (o-o, o-o), BM 06.022 (5-O) or CHO-t-PA (Ls-A ) at 37"C, harvested, washed and then incubated for 30 min on ice with MAl-C8 (closed circles), or with EPA-BC (open circles, open triangles, open squares), cells were washed and FITC respectively. Subsequently, labeled anti mouse Ig was added for further 30 min on ice. The percentage of positively stained cells versus time of incubation is shown. Bindins of CHO-t-PA, BM 06.021 and BM 06.022 to human cell lines. Next, we tested whether recombinant t-PA variants bind to human cell lines of various origin. The T-cell line Jurkat, the B-cell line HL 60 and the cell Daudi, promyelocytic line the erythroleukemic cell line K562 were incubated with rt-PA's (440 nM). Bound molecules were determined with antibody EPA-BC by flow cytometric analysis. As one can see from the data in Tab. 1 about
594
CELL-BINDING OF rt-PA VARIANTS
Vol. 67, No. 5
50% of Daudi, Jurkat and K562 cells exhibit a positive binding after incubation with BM 06.021. The lowest number of positive cells (29.5%) is obtained for the HL 60 cell line. The binding pattern of CHO-t-PA and BM 06.022 to human cell lines is almost identical. The highest binding can be demonstrated for the K562 and Jurkat cell line, with 21.9 and 15.6 percent of positive cells, respectively. For HL 60 and Daudi cells, less than 10 percent of cells are positive for binding of CHO-t-PA or BM 06.022.
100
1000
rt-tPA
(nmol/l)
FIG. 3 Concentration dependent binding of CHO-t-PA, BM 06.021 and BM 06.022 to HUVEC as revealed by flow cytometric analysis. lo6 HUVEC per ml were incubated for 2 hours with increasing concentrations of BM 06.021 (o-o), CHO-t-PA (o-o), or BM 06.022 (A-A) at 37“C. Subsequently, cells were washed and incubated with EPA-BC. Preparation of samples for flow cytometric analysis was done as described in Materials and Methods. The percentage of positive cells versus concentration of rt-PA is shown. TABLE 1 Binding of rt-PA to HUVEC and established analyzed by flow cytometry. (% positive
human cell lines
cells)
lo6 cells per ml were incubated with 440 nM rt-PA for 1 hour at 37OC, subsequently washed and incubated with the B-chain
Vol. 67, No. 5
CELL-BINDING OF rt-PA VARIANTS
595
specific monoclonal antibody EPA-BC. Specific binding recorded by flow cytometric analysis as described Materials and Methods.
was in
Analvsis of bindina of BM 06.021 to HWEC with domain snecific monoclonal antibodies. We next wanted to know whether binding of BM 06.021 to YWEC influences the recognition of bound rt-PA by epitope-specific antibodies. Binding was measured by flow cytometric analysis and the results of two independent experiments are shown in Figure 4. In general, the percentage of positively stained cells in the second experiment is higher than in the first experimental setup. About 22 to 38% positively stained HWEC can be obtained with the B-chain specific antibodies EPA-BC and TC2 VPA. Using another Bchain specific antibody (TCl VPA) which stains only about 5% to 10% of HWEC after incubation with BM 06.021, results in a reduction of more than 70%. A similar phenomenon can be obtained with the antibody TC3 VPA which is directed against the EGF/finger domain of t-PA. The binding of this antibody to HWEC-bound BM 06.021 is more than 70% reduced compared to results obtained with a second EGF/finger domain specific antibody TC6 VPA. The binding of two independent kringle 2 specific antibodies, TC4 VPA and TC7 VPA is almost identical with about 25% positively stained cells.
1
FIG.4
: 3 4 5 6 7 EPITOPE SPECIFIC ANTIBODIES
Epitope analysis of binding of BM 06.021 to HWEC by monoper ml were incubated for 2 clonal antibodies. 106HWEC hours at 37“C with 440 nM of BM 06.021. Afterwards, cells were washed in ice-cold PBS and incubated for 30 min on ice antibodies. specific monoclonal epitope with t-PA Preparation of samples for flow cytometric analysis was done as described in Materials and Methods. The results of two independent experiments are shown. (1. experiment: open The columns). crosshatched experiment: columns, 2. following antibodies have been used: Lane 1: EPA-BC, Lane 2: TC2 VPA, Lane 3: TCl VPA, all B-chain specific. Lane 4: TC4 VPA, Lane 5: TC7 VPA, all kringle 2 specific. Lane 6: TC3 VPA, Lane 7: TC6 VPA, all EGF/finger domain Specific.
596
CELL-BINDING OF &PA VARIANTS
Vol. 67. No. 5
DISCUSSION
In the present study, we analysed the binding of the rt-PA's CHOt-PA, BM 06.021 and BM 06.022 to HWEC and other human Cell lines by flow cytometry. CHO-t-PA is a glycosylated human rt-PA obtained after in vitro cultivation of transfected Chinese hamster ovary cells, whereas BM 06.021 is a non-glycosylated &-PA produced in E. coli. BM 06.022 is a t-PA deletion variant comprising the kringle 2 and protease domain, and produced in E. coli, too. Binding studies were performed by flow cytometry using t-PA specific monoclonal antibodies. Data obtained show that a plateau of binding, with more than 60% can be obtained after 90 min of positively stained HWEC, incubation with 440 nM BM 06.021. This could be demonstrated with A-chain or B-chain specific antibodies. For BM 06.022, the nonglycosylated t-PA deletion variant, membrane bound t-PA can be detected on less than 5% of HWEC, indicating that binding of nonglycosylated t-PA's occurs most likey via the EGF/finger and kringle 1 domain. This assumption is also supported by reports from by Beebe et a1.(24). They blocked binding of t-PA to HWEC preincubation of the enzyme with either linear peptides derived from the EGF/finger domain or with monoclonal antibodies against one of these domains. The glycosylated CHO-t-PA binds marginaly to HWEC, only. In contrast to the analysis performed with BM 06.021, binding of CHOt-PA to HWEC can only be revealed with the B-chain specific antibody EPA-BC. The negative result of the other B-chain specific antibodies could be explained by steric hindrance. This assumption is supported by the fact that even with an almost tenfold increase of CHO-t-PA no further increase in the number of positively stained HWEC could be obtained. Endocytosis of CHO-t-PA by HWEC can be excluded, because after rt-PA incubation and cell fixation in paraformaldehyde a similar low number of positive cells was obtained (data not shown). Finally, it may also be possible that the low number of positive cells is due to a reduced affinity of CHO-t-PA than of BM 06.021 to HWEC. The results obtained reveal that the unglycosylated full length rtPA BM 06.021 has a higher tendency to bind non-specifically to HWEC than the glycosylated variant CHO-t-PA. The deletion variant BM 06.022 which lacks carbohydrate side chains, shows the lowest tendency to bind to HWEC. The results indicate that glycosylation of the A-chain reduces non-specific association of rt-PA's to HWEC. The most drastic reduction in non-specific binding of rt-PA to HWEC, however, was obtained by deletion of the EGF/finger and kringle 1 domain of the A-chain as in BM 06.022, suggesting that epitopes within the A-chain are mostly involved in non-specific binding. More recently, two binding sites of t-PA to HWEC have been described. A high affinity binding site of t-PA to HWEC can be attributed to the B-chain of t-PA at concentrations below 10 nM tPA (25), suggesting that in our study we analyzed low affinity binding of rt-PA's to HWEC. In addition, we investigated the binding pattern of CHO-t-PA, BM 06.021 and BM 06.022 to human cell lines of different origin. The T-cell line Jurkat and the B-cell line Daudi exhibit the highest number of cells binding CHO-t-PA or BM 06.021. An intermediate number of pOSitiVe CellS was detected on K562 cells and HWEC. The
Vol. 67, No. 5
CELL-BINDING
OF rt-PA VARIANTS
597
lowest number of positive cells was determined for HL 60 cells. The data indicate that deglycosylation of full length r-t-PA increases the tendency of non-specific binding not only to HWEC but also to other human cell lines. BM 06.022, however, does not bind to HWEC, Jurkat and HL 60 cells, and binds only marginaly to Daudi and K562. The binding of BM 06.021 to HWEC was characterized in more detail by epitope specific monoclonal antibodies. Data obtained show that the B-chain specific antibody TCl VPA stains only one third of HWEC in comparison to a second B-chain specific antibody TC2 VPA, although the affinity of TCl VPA to t-PA is higher than of TC2 VPA. These data suggest that the epitope of BM 06.021 recognized by TCl VPA is close to the binding site of BM 06.021 to HWEC. In conclusion we demonstrate, that the t-PA variant BM 06.021 binds to more human cells including HWEC than CHO-t-PA and BM 06.022. Glycosylation of full length rt-PA results in reduced numbers of positive cells. The best result in preventing non-specific binding of rt-PA can be obtained, however, by deletion of the EGF/finger and kringle 1 domains, as shown for BM 06.022. REFERENCES 1.
HAJJAR, K. A. The endothelial cell tissue plasminogen receptor. J. Biol. Chem. 32, 21962-21970, 1991.
2.
VERSTRAETE, M.,BERNARD, R., BORY, M., BROWER, R.W., COLLEN, D., DE BONO, D.P., ERBEL, R., HUHMANN, W., LENNANE, R.J., LUBSEN, J ., MATHEY, D., MEYER, J., MICHELS, H.R., RUTSCH, W.! SCHARTL, M ., SCHMIDT, W., UEBIS, R. and VON ESSEN, R. Randomised trial of intravenous recombinant tissue-type plasminogen activator streptokinase in acute myocardial versus intravenous infarction. Lancet 1, 842-847, 1985.
3.
PENNICA, D., HOLMES, W.E., KOHR, W.J., HARKINS, R.N., VEHAR, G.A., WARD, C.A., BENNETT, W.F., YELVERTON, E., SEEBURG, P.H., HEYNECKER, H.L., GOEDDEL, D.V. and COLLEN, D. Cloning and expression of human tissue type plasminogen activator cDNA in E. coli. Nature 301, 214-221, 1983.
4.
COLLEN, D., STASSEN, J.M., MARAFINO, B.J., JR., BUILDER, S., DECOCK, F., OGEZ, J., TAJIRI, D., PENNICA, D., BENNETT, W.F., of human SALWA, J. and HOYNG, C.F. Biological properties tissue-type plasminogen activator obtained by expression of recombinant DNA in mammalian cells. J. Pharm. Exntl. Therap. 231, 146-152, 1984.
5.
KOHNERT, U., RUDOLPH, R., VERHEIJEN, J.H., WEENING-VERHOEFF, E.J.D., STERN, A., OPITZ, U., MARTIN, U., LILL, H., PRINZ, H., LECHNER, M., KRESSE, G-B., BUCKEL, P. and FISCHER, S. Deletion of the 3 N-terminal domains still maintains the biochemical properties of the protease and the kringle 2 in the refolded tPA deletion variant BM 06.022. Prot. Ensineerins 5, 93-100, 1992.
6.
HESSEL,
L.W. and KLUFT, C. Advances
in clinical
activator
fibrinolysis.
598
CELL-BINDING OF &PA VARIANTS
Thromb. Haemostas.
Vol. 67, No. 5
43, 77-89, 1986.
7.
VON HEYNE, G. Patterns of amino Eur. J. Biochem. cleavage sites.
acids near signal-sequence 133, 17-21, 1983.
8.
NY, T., ELGH, F. and LUND, B. The structure of the human tissue type plasminogen activator gene: correlation of intron and exon structures to functional and structural domains. Proc. Natl. Acad. Sci. USA. 81, 5335-5339, 1984.
9.
BANYAI, L., VARADI, A. and PATTHY, L. Common evolutionary of the fibrin-binding structures of fibronectin and tissue-type plasminogen activator. FEBS Lett. 163, 37-41, 1983.
10. COLLEN, D., STASSEN, J.M. and VERSTRAETE, M. Thrombolysis with human extrinsic (tissue-type) plasminogen activator in rabbits with experimental jugular vein thrombosis. J. Clin. Invest. 71, 368-376, 1983. 11. MARTIN, U., FISCHER, S., KOHNERT, U., RUDOLPH, R., SPONER, G., properties of an STERN, A. and STREIN, K. Pharmacokinetic Escherichia-coli-producedrecombinantplasminogenactivator (BM 06.022) in rabbits. Thrombosis Research 62, 137-146, 1991. D.C., OTTER, M., KUIPER, J. and BERKEL, TH.J.C. 12. RIJKEN, Receptor-mediated endocytosis of tissue-type plasminogen activator (t-PA) by live cells. Thromb.Res. Sunnl.X, 63-71, 1990. 13. MARTIN, U., FISCHER, S., KOHNERT, U., LILL, H., RUDOLPH, R., SPONER, G. and STREIN, K. Properties of a novel plasminogen activator (BM 06.022) produced in Escherichia coli. 2. Kardiol. 79: Sunnl. 3, 167-170, 1990. 14. ESTREICHER, A., WOHLWEND, A., BELIN, D., SCHLEUNING, W.D. and VASSALLI, J.D. Characterisation of the cellular binding site for the urokinase-type plasminogen activator. J. Biol. Chem. 264, 1180-1189, 1989. 15. RUSSELL, M.E., QUERTERMOUSE, T., DECLERCK, P.J., COLLEN, D., HABER, E. and HOMCY, C.J. Binding of tissue-type plasminogen activator with human endothelial cell monolayers. J.Biol. Chem. 265, 2569-2575, 1990. 16. MACARTNEY, H.W. Verfahren zur Abtrennung PA. EP 0 302 503 AZ, 1988.
und Reinigung
von t-
17. MARTIN, U., FISCHER, S., KOHNERT, U., RUDOLPH, R., SPONER, G., STERN, A. and STREIN, K. Coronary thrombolytic properties of a novel recombinantplasminogen activator (BM 06.022) in a canine model. J. Card. Pharm. 18, 111-119, 1991. 18. JAFFE, E.A., NACHLMAN, R.L., BECKER, C.G. and MINICK, C.R. Culture of human endothelial cells derived from umbilical veins. Identification by morphologiacl and immunological criteria. J. Clin. Invest. 52, 2745-2756, 1973.
Vol. 67, No. 5
CELL-BINDING OF &PA VARIANTS
599
19. JAFFE, E.A. Endothelial cells and the biology of factor VIII. N. Enul. J.Med. 296, 377-383, 1977. 20. NEELS, H.M., VAN SANDE, M.E. and SCHARPE, S.L. Sensitive calorimetric assay for angiotensin converting enzyme in serum. Clin.Chem. 29, 1399-1403, 1983. 21.
COMBRIER, E., METEZEAU, P., RONOT, X., GACHELIN, H. and ADOLPHE, M. Flow cytometric assessment of cell viability: a multifaceted analysis. Cvtotechnolouv 2, 27-37, 1989.
22. HOLVOET, P., LIJNEN, H.R. and COLLEN, D. Characterisation of functional domains in human tissue-type plasminogen activator with the use of monoclonal antibodies. Eur. J. Biochem. 158, 173-177, 1986. 23. KUBBIES, M. and STOCKINGER, H. Cell-cycle dependent DHFR and tPA production in cotransfected, MTX amplified CHO cells revealed by dual-laser flow cytometry. Exe. Cell Res. 188, 267271, 1990. 24. BEEBE, D.P. Binding of tissue plasminogen activator to human umbilical vein endothelial cells. Thromb. Res. 46, 241-254, 1987. 25. BARNATHAN, E.S., KUO, A., VAN DER KEYL, H., MCCRAE, K.R., KARSEN, G.R. and CINES, D.B. Tissue-type plasminogen activator binding to human endothelial cells. J. Biol. Chem. 263, 77927799, 1988.
BINDING OF RECOMBINANT VARIANTS OF HUNAN TISSUE-TYPE PLASMINOGEN ACTIVATOR (t-PA) TO HUNAN UMBILICAL VEIN ENDOTHELIAL CELLS
H.Stockinger', M. Kubbies, Boehringer Germany
Mannheim
R.Rudolph, A. Stern, U. Kohnert and S. Fischer GmbH, Biochemical Research Center, Penzberg,
(Received 27.2.1992; accepted in revised form 15.7.1992 by Editor H HCirmann)
ABSTRACT Endothelial cells synthesize and secrete hemostatic components like tissue-type plasminogen activator (t-PA) which is thought to be the major determinant of fibrinolytic activity in the blood. Most recently, a receptor on human protein for t-PA umbilical vein endothelial cells (HWEC) in culture has been described (1); there are, however, in addition low affinity binding sites for t-PA on HWEC. The sites of binding are of particular interest, because they are potential regulators of t-PA activity and clearance. We analysed the low affinity binding of recombinant t-PA (rt-PA) to normal diploid HWEC and to the permanent human cell lines Jurkat, Daudi, HL 60 and K562 by flow cytometry applying t-PA specific monoclonal antibodies. Using this test system binding of both recombinant glycosylated human t-PA produced in Chinese hamster ovary cells (CHO-t-PA) and of nonglycosylated t-PA, produced in E. coli (BM 06.021) was investigated. Analysis of the binding pattern to HWEC and other cell lines revealed that deglycosylation of full length rt-PA increases non-specific binding. Additionally, we investigated the binding properties of an unglycosylated t-PA deletion variant which comprises the kringle 2 and the (BM 06.022). Data obtained show that protease domains deletion of these domains most drastically reduces nonand other human cell lines. specific binding to HWEC
__ plasminogen activator tissue-type worcas: recombinant Key endothelial cells - human cell lines - flow cytometric analysis non-specific binding "corresponding Author 589
590
CELL-BINDING OF &PA VARIANTS
Vol. 67, No. 5
INTRODUCTION Thrombolytic therapy is an effective treatment for acute myocardial infarction (2). In this respect, tissue-type plasminogen activator (t-PA) has been shown to dissolve artery thrombi thereby improving myocardial function. Recently, t-PA produced by recombinant DNA technology (rt-PA) has become a widely used drug for the treatment of thromboembolic diseases (3-5). Human t-PA is a multidomain serine protease which exerts its dominant role during fibrinolysis by converting plasminogen into plasmin (6). The proteolytically active site of t-PA resides in the B-chain, whereas the A-chain consists of the finger domain, the EGF-domain and two kringle domains (7-9). The EGF/finger and kringle domains of the A-chain probably contain fibrin binding sites of t-PA. One of the major problems encountered upon therapeutic use of rt-PA is its rapid hepatic elimination from the plasma characterized by a half life of is probably about three minutes (10,ll). The rapid clearance mediated by a carbohydrate specific receptor on liver endothelial cells and by a sofar unknown receptor on liver parenchymal cells (12). Consequently, a nonglycosylated full length rt-PA from E. coli (BM 06.021) displays a longer half-life and a lower clearance in vivo than CHO-t-PA, the rt-PA from Chinese hamster ovary cells (13). In addition to binding of t-PA by liver parenchymal cells, it has been shown that human umbilical vein endothelial cells (HWEC) possess two types of t-PA acceptor sites with different affinity for t-PA (14). The sites of binding of t-PA to endothelial cells are of particular interest because they are potential regulators of t-PA activity and are probably involved in removing t-PA from the circulation (15). The aim of the present study was to investigate the binding of externally added full length glycosylated rt-PA produced from mammalian cell culture (CHO-t-PA; (16)) or of unglycosylated rt-PA produced in E. coli (BM 06.021; (17)) to HWEC and established human cell lines in vitro. The binding pattern of CHO-t-PA and BM 06.021 was compared with the binding of a recombinant nonglycosylated t-PA molecule produced in E. coli (BM 06.022) lacking the finger/EGF and kringle 1 domains (5). The results of this study should give us an insight which cell types might be involved in the clearance of rt-PA in vivo. MATERIALS
AND METHODS
Cell Culture. HWEC were isolated according to standard protocols (18) using 0,2 % collagenase Type II (Sigma). Cells were grown to confluence on gelatin-coated petri culture dishes and subsequently passaged. After each passage, aliquots of cells were frozen in Ml99 medium supplemented with 20 % fetal calf serum and 7 % dimethyl sulfoxide and stored in liquid nitrogen. After thawing, cells were grown to confluence in a mixture of RPM1 1640/M199 medium (50/50) supplemented with 20% fetal calf serum and heparin (10 mg/ml) and endothelial cell growth factor (20 pg/ml). Only cells from the 8th to 10th passage were used for binding studies. Endothelial cells were characterized by the presence of von Willebrand Factor (vWF)(19) and angiotensin converting enzyme (ACE)(ZO) in the culture supernatant, using an ACE test kit and a vWF ELISA test kit
Vol. 67, No. 5
CELL-BINDING OF rt-PA VARIANTS
591
(Boehringer Mannheim). The percentage of HWEC in the cell culture determined by using a factor VIII-specific antibody (Pan Systems, Germany) was in the range of 50 - 70%. Human cell lines. The T-cell line Jurkat, the B-cell line Daudi, the promyelocytic cell line HL 60 and the erythroleukemic cell line K562 were cultivated in RPM1 1640 containing 10% fetal calf serum. Flow cvtometric analvsis. HWEC were grown to confluence. Afterwards, medium was removed and the cell layer was washed three times in phosphate buffered saline (PBS) to remove fetal calf serum. The rt-PA molecules were added to the cell culture at various concentrations and incubated at 37°C for different times. Subsequently, cells were washed again in cold PBS and incubated on ice with the respective t-PA specific monoclonal antibodies (10 /.&g/ml). Thereafter, cells were washed again and a fluorescein isothiocyanate (FITC) labeled anti mouse Ig antibody was added. After incubating the samples for 30 minutes on ice, cells were washed again in cold PBS, detached from culture plates with 0.2% ethylenediaminetetraacetate (EDTA, pH 8). Bound t-PA was determined by flow cytometry. Flow cytometric analysis was performed using an Ortho Cytofluorograph 50H (Ortho Instruments, MA) connected to a 2151 computer for data aquisition and storage. The cells were excited the with 488 nm line forward/side and scatter (50 mw) characteristic was used to gate on viable cells (21). FITC fluorescence was selected using a 510130 nm bandpass. The histograms obtained were transferred to a PC/AT computer and a subtraction analysis of the controls and positive cells was performed using the MPLUS software (Phoenix Flow Systems, Sorrento Valley, CA). Monoclonal antibodies. The binding of rt-PA's to HWEC or other human cell lines was determined with the following epitope-specific antibodies: B-chain specific: EPA-BC monoclonal (Boehringer Mannheim), MA2-G6 (22), TCl VPA and TC2 VPA (all from Technoclone, Vienna, Austria). A-chain specific: MAl-C8 prevents binding of t-PA to clots generated in normal human plasma (22), TC4 VPA and TC7 VPA are kringle 2 specific and TC3 VPA and TC6 VPA are finger and EGF like domain specific (all from Technoclone). TCl VPA and TC4 VPA are non-competitive inhibitors of plasminogen activation in the presence of CNBr fragments of fibrinogen while TC3 VPA and TC7 VPA are competitive inhibitors. t-PA molecules used for bindins studies. CHO-t-PA was isolated and purified from the supernatant of transfected Chinese Hamster Ovary BM cells (16). Full length rt-PA BM 06.021 and the deletion Variant 06.022 were produced in E. coli (5,13). During the production process initially insoluble aggregates are solubilized, refolded in vitro and purified as described (5). The specific activity of CHOt-PA was 977 units per mg and ml, of BM 06.021 was 800 units pe!rmg and ml and of BM 06.022 was 542 units per mg and ml. RESULTS Bindinc analvsis bv flow cytometry of BM 06.021 to HWEC In Fig. 1 the flow cytometric histograms of the anti-t-PA/anti
CELL-BINDING OF rt-PA VARIANTS
592
Vol. 67, No. 5
mouse FITC fluorescence is shown on a linear scale after gating on intact cells according to their scatter characteristics (excluding unspecific binding of monoclonal antibodies to dead cells;)(21). In panel A and B the broken lines correspond to the controls with anti t-PA/anti mouse FITC incubation without rt-PA preincubation of labelling shows HUVEC. Addition of rt-PA prior to antibody (solid increasing FITC fluorescence intensities of the cells lines), significantly overlapping with the control region. A subtraction technique of the two corresponding histograms was used to quantitate the number of positive cells (shadowed areas). In principle, the two different monoclonal antibodies for the B-chain (EPA-BC, panel A) and for the A-chain (MAl-C8, panel B) give similar results of about 60% positively labeled cells. Using a dual laser technique and Hoechst 33342 viable labelling of cellular DNA (23) we found no cell-cycle specific expression of t-PA binding sites to HUVEC (data not shown).
rl
140+
i
i
FLUORESCENCE
INTENSITY
FIG. 1 Flow cytometric analysis of the t-PA binding to HUVEC using monoclonal antibodies to t-PA followed by FITC conjugated anti mouse Ig, respectively. The control cells were not preincubated with rt-PA (broken lines). The FITC-fluorescence histograms were recorded on linear scale (x-axis: channel number; y-axis: cell number. Panel A: EPA-BC (Bchain specific); panel B: MAl-C8 (A-chain specific). Flow cvtometric analysis of the bindinu kinetics of CHO-t-PA. BM 06.021 and BM 06.022 to HUVEC HUVEC (106/ml) were incubated with BM 06.021, BM 06.022 or CHO-t-PA (440 nM) at 37'C as described in Materials and Methods, and binding was followed up over 2 hours using flow cytometric analysis. Data in Fig. 2 show that there is a time-dependent increase of the BM 06.021 binding to HUVEC within the first hour. After 90 minutes a plateau is reached with a maximum binding of 60 % positive cells. This can be detected with antibody MAl-C8 (A-chain specific) or antibody EPA-BC (B-chain specific). A further increase in the
Vol. 67, No. 5
593
CELL-BINDING OF t-t-PA VARIANTS
incubation time did not increase the number of positive cells. The analysis of binding of CHO-t-PA (440 nM) to HWEC by flow cytometry can only be demonstrated with the antibody EPA-BC, which stains about 15% of HWEC. After incubation with BM 06.022 less than 10% of HWEC are positively stained with the antibodies MAl-C8 or EPABC. Non-specific binding of the rt-PA's to HWEC was further characterized by incubating the cells with increalsing concentrations (55-3600 nM) of BM 06.021, BM 06.022 or CHO-t-PA for 1 hour. Subsequently, binding of the rt-PA's was determined by flow cytometric analysis with the antibody EPA-BC. Data presented in Fig. 3 show that a saturation in the relative binding of BM 06.021 to 55% of HWEC is reached at about 2 PM. However, even the highest concentration of CHO-t-PA or BM 06.022 tested revealed only marginal binding of these rt-PA's to less than 20% of HWEC. The percentage of positively stained cells after incubation with CHO-tPA increases from about 10 % obtained with 55 nM to about 20 % positive cells obtained at a concentration of 3.5 PM. The binding of BM 06.022 to HWEC is below 10% of positive cells for all concentrations tested.
0
30
60 time
90
120
150
(min)
FIG. 2 Analysis of the binding kinetics of CHO-t-PA, BM 06.021 and BM 06.022 to HWEC. lo6 HWEC per ml were incubated for 30, 60, 90 and 120 minutes with 440 nM BM 06.021 (o-o, o-o), BM 06.022 (5-O) or CHO-t-PA (Ls-A ) at 37"C, harvested, washed and then incubated for 30 min on ice with MAl-C8 (closed circles), or with EPA-BC (open circles, open triangles, open squares), cells were washed and FITC respectively. Subsequently, labeled anti mouse Ig was added for further 30 min on ice. The percentage of positively stained cells versus time of incubation is shown. Bindins of CHO-t-PA, BM 06.021 and BM 06.022 to human cell lines. Next, we tested whether recombinant t-PA variants bind to human cell lines of various origin. The T-cell line Jurkat, the B-cell line HL 60 and the cell Daudi, promyelocytic line the erythroleukemic cell line K562 were incubated with rt-PA's (440 nM). Bound molecules were determined with antibody EPA-BC by flow cytometric analysis. As one can see from the data in Tab. 1 about
594
CELL-BINDING OF rt-PA VARIANTS
Vol. 67, No. 5
50% of Daudi, Jurkat and K562 cells exhibit a positive binding after incubation with BM 06.021. The lowest number of positive cells (29.5%) is obtained for the HL 60 cell line. The binding pattern of CHO-t-PA and BM 06.022 to human cell lines is almost identical. The highest binding can be demonstrated for the K562 and Jurkat cell line, with 21.9 and 15.6 percent of positive cells, respectively. For HL 60 and Daudi cells, less than 10 percent of cells are positive for binding of CHO-t-PA or BM 06.022.
100
1000
rt-tPA
(nmol/l)
FIG. 3 Concentration dependent binding of CHO-t-PA, BM 06.021 and BM 06.022 to HUVEC as revealed by flow cytometric analysis. lo6 HUVEC per ml were incubated for 2 hours with increasing concentrations of BM 06.021 (o-o), CHO-t-PA (o-o), or BM 06.022 (A-A) at 37“C. Subsequently, cells were washed and incubated with EPA-BC. Preparation of samples for flow cytometric analysis was done as described in Materials and Methods. The percentage of positive cells versus concentration of rt-PA is shown. TABLE 1 Binding of rt-PA to HUVEC and established analyzed by flow cytometry. (% positive
human cell lines
cells)
lo6 cells per ml were incubated with 440 nM rt-PA for 1 hour at 37OC, subsequently washed and incubated with the B-chain
Vol. 67, No. 5
CELL-BINDING OF rt-PA VARIANTS
595
specific monoclonal antibody EPA-BC. Specific binding recorded by flow cytometric analysis as described Materials and Methods.
was in
Analvsis of bindina of BM 06.021 to HWEC with domain snecific monoclonal antibodies. We next wanted to know whether binding of BM 06.021 to YWEC influences the recognition of bound rt-PA by epitope-specific antibodies. Binding was measured by flow cytometric analysis and the results of two independent experiments are shown in Figure 4. In general, the percentage of positively stained cells in the second experiment is higher than in the first experimental setup. About 22 to 38% positively stained HWEC can be obtained with the B-chain specific antibodies EPA-BC and TC2 VPA. Using another Bchain specific antibody (TCl VPA) which stains only about 5% to 10% of HWEC after incubation with BM 06.021, results in a reduction of more than 70%. A similar phenomenon can be obtained with the antibody TC3 VPA which is directed against the EGF/finger domain of t-PA. The binding of this antibody to HWEC-bound BM 06.021 is more than 70% reduced compared to results obtained with a second EGF/finger domain specific antibody TC6 VPA. The binding of two independent kringle 2 specific antibodies, TC4 VPA and TC7 VPA is almost identical with about 25% positively stained cells.
1
FIG.4
: 3 4 5 6 7 EPITOPE SPECIFIC ANTIBODIES
Epitope analysis of binding of BM 06.021 to HWEC by monoper ml were incubated for 2 clonal antibodies. 106HWEC hours at 37“C with 440 nM of BM 06.021. Afterwards, cells were washed in ice-cold PBS and incubated for 30 min on ice antibodies. specific monoclonal epitope with t-PA Preparation of samples for flow cytometric analysis was done as described in Materials and Methods. The results of two independent experiments are shown. (1. experiment: open The columns). crosshatched experiment: columns, 2. following antibodies have been used: Lane 1: EPA-BC, Lane 2: TC2 VPA, Lane 3: TCl VPA, all B-chain specific. Lane 4: TC4 VPA, Lane 5: TC7 VPA, all kringle 2 specific. Lane 6: TC3 VPA, Lane 7: TC6 VPA, all EGF/finger domain Specific.
596
CELL-BINDING OF &PA VARIANTS
Vol. 67. No. 5
DISCUSSION
In the present study, we analysed the binding of the rt-PA's CHOt-PA, BM 06.021 and BM 06.022 to HWEC and other human Cell lines by flow cytometry. CHO-t-PA is a glycosylated human rt-PA obtained after in vitro cultivation of transfected Chinese hamster ovary cells, whereas BM 06.021 is a non-glycosylated &-PA produced in E. coli. BM 06.022 is a t-PA deletion variant comprising the kringle 2 and protease domain, and produced in E. coli, too. Binding studies were performed by flow cytometry using t-PA specific monoclonal antibodies. Data obtained show that a plateau of binding, with more than 60% can be obtained after 90 min of positively stained HWEC, incubation with 440 nM BM 06.021. This could be demonstrated with A-chain or B-chain specific antibodies. For BM 06.022, the nonglycosylated t-PA deletion variant, membrane bound t-PA can be detected on less than 5% of HWEC, indicating that binding of nonglycosylated t-PA's occurs most likey via the EGF/finger and kringle 1 domain. This assumption is also supported by reports from by Beebe et a1.(24). They blocked binding of t-PA to HWEC preincubation of the enzyme with either linear peptides derived from the EGF/finger domain or with monoclonal antibodies against one of these domains. The glycosylated CHO-t-PA binds marginaly to HWEC, only. In contrast to the analysis performed with BM 06.021, binding of CHOt-PA to HWEC can only be revealed with the B-chain specific antibody EPA-BC. The negative result of the other B-chain specific antibodies could be explained by steric hindrance. This assumption is supported by the fact that even with an almost tenfold increase of CHO-t-PA no further increase in the number of positively stained HWEC could be obtained. Endocytosis of CHO-t-PA by HWEC can be excluded, because after rt-PA incubation and cell fixation in paraformaldehyde a similar low number of positive cells was obtained (data not shown). Finally, it may also be possible that the low number of positive cells is due to a reduced affinity of CHO-t-PA than of BM 06.021 to HWEC. The results obtained reveal that the unglycosylated full length rtPA BM 06.021 has a higher tendency to bind non-specifically to HWEC than the glycosylated variant CHO-t-PA. The deletion variant BM 06.022 which lacks carbohydrate side chains, shows the lowest tendency to bind to HWEC. The results indicate that glycosylation of the A-chain reduces non-specific association of rt-PA's to HWEC. The most drastic reduction in non-specific binding of rt-PA to HWEC, however, was obtained by deletion of the EGF/finger and kringle 1 domain of the A-chain as in BM 06.022, suggesting that epitopes within the A-chain are mostly involved in non-specific binding. More recently, two binding sites of t-PA to HWEC have been described. A high affinity binding site of t-PA to HWEC can be attributed to the B-chain of t-PA at concentrations below 10 nM tPA (25), suggesting that in our study we analyzed low affinity binding of rt-PA's to HWEC. In addition, we investigated the binding pattern of CHO-t-PA, BM 06.021 and BM 06.022 to human cell lines of different origin. The T-cell line Jurkat and the B-cell line Daudi exhibit the highest number of cells binding CHO-t-PA or BM 06.021. An intermediate number of pOSitiVe CellS was detected on K562 cells and HWEC. The
Vol. 67, No. 5
CELL-BINDING
OF rt-PA VARIANTS
597
lowest number of positive cells was determined for HL 60 cells. The data indicate that deglycosylation of full length r-t-PA increases the tendency of non-specific binding not only to HWEC but also to other human cell lines. BM 06.022, however, does not bind to HWEC, Jurkat and HL 60 cells, and binds only marginaly to Daudi and K562. The binding of BM 06.021 to HWEC was characterized in more detail by epitope specific monoclonal antibodies. Data obtained show that the B-chain specific antibody TCl VPA stains only one third of HWEC in comparison to a second B-chain specific antibody TC2 VPA, although the affinity of TCl VPA to t-PA is higher than of TC2 VPA. These data suggest that the epitope of BM 06.021 recognized by TCl VPA is close to the binding site of BM 06.021 to HWEC. In conclusion we demonstrate, that the t-PA variant BM 06.021 binds to more human cells including HWEC than CHO-t-PA and BM 06.022. Glycosylation of full length rt-PA results in reduced numbers of positive cells. The best result in preventing non-specific binding of rt-PA can be obtained, however, by deletion of the EGF/finger and kringle 1 domains, as shown for BM 06.022. REFERENCES 1.
HAJJAR, K. A. The endothelial cell tissue plasminogen receptor. J. Biol. Chem. 32, 21962-21970, 1991.
2.
VERSTRAETE, M.,BERNARD, R., BORY, M., BROWER, R.W., COLLEN, D., DE BONO, D.P., ERBEL, R., HUHMANN, W., LENNANE, R.J., LUBSEN, J ., MATHEY, D., MEYER, J., MICHELS, H.R., RUTSCH, W.! SCHARTL, M ., SCHMIDT, W., UEBIS, R. and VON ESSEN, R. Randomised trial of intravenous recombinant tissue-type plasminogen activator streptokinase in acute myocardial versus intravenous infarction. Lancet 1, 842-847, 1985.
3.
PENNICA, D., HOLMES, W.E., KOHR, W.J., HARKINS, R.N., VEHAR, G.A., WARD, C.A., BENNETT, W.F., YELVERTON, E., SEEBURG, P.H., HEYNECKER, H.L., GOEDDEL, D.V. and COLLEN, D. Cloning and expression of human tissue type plasminogen activator cDNA in E. coli. Nature 301, 214-221, 1983.
4.
COLLEN, D., STASSEN, J.M., MARAFINO, B.J., JR., BUILDER, S., DECOCK, F., OGEZ, J., TAJIRI, D., PENNICA, D., BENNETT, W.F., of human SALWA, J. and HOYNG, C.F. Biological properties tissue-type plasminogen activator obtained by expression of recombinant DNA in mammalian cells. J. Pharm. Exntl. Therap. 231, 146-152, 1984.
5.
KOHNERT, U., RUDOLPH, R., VERHEIJEN, J.H., WEENING-VERHOEFF, E.J.D., STERN, A., OPITZ, U., MARTIN, U., LILL, H., PRINZ, H., LECHNER, M., KRESSE, G-B., BUCKEL, P. and FISCHER, S. Deletion of the 3 N-terminal domains still maintains the biochemical properties of the protease and the kringle 2 in the refolded tPA deletion variant BM 06.022. Prot. Ensineerins 5, 93-100, 1992.
6.
HESSEL,
L.W. and KLUFT, C. Advances
in clinical
activator
fibrinolysis.
598
CELL-BINDING OF &PA VARIANTS
Thromb. Haemostas.
Vol. 67, No. 5
43, 77-89, 1986.
7.
VON HEYNE, G. Patterns of amino Eur. J. Biochem. cleavage sites.
acids near signal-sequence 133, 17-21, 1983.
8.
NY, T., ELGH, F. and LUND, B. The structure of the human tissue type plasminogen activator gene: correlation of intron and exon structures to functional and structural domains. Proc. Natl. Acad. Sci. USA. 81, 5335-5339, 1984.
9.
BANYAI, L., VARADI, A. and PATTHY, L. Common evolutionary of the fibrin-binding structures of fibronectin and tissue-type plasminogen activator. FEBS Lett. 163, 37-41, 1983.
10. COLLEN, D., STASSEN, J.M. and VERSTRAETE, M. Thrombolysis with human extrinsic (tissue-type) plasminogen activator in rabbits with experimental jugular vein thrombosis. J. Clin. Invest. 71, 368-376, 1983. 11. MARTIN, U., FISCHER, S., KOHNERT, U., RUDOLPH, R., SPONER, G., properties of an STERN, A. and STREIN, K. Pharmacokinetic Escherichia-coli-producedrecombinantplasminogenactivator (BM 06.022) in rabbits. Thrombosis Research 62, 137-146, 1991. D.C., OTTER, M., KUIPER, J. and BERKEL, TH.J.C. 12. RIJKEN, Receptor-mediated endocytosis of tissue-type plasminogen activator (t-PA) by live cells. Thromb.Res. Sunnl.X, 63-71, 1990. 13. MARTIN, U., FISCHER, S., KOHNERT, U., LILL, H., RUDOLPH, R., SPONER, G. and STREIN, K. Properties of a novel plasminogen activator (BM 06.022) produced in Escherichia coli. 2. Kardiol. 79: Sunnl. 3, 167-170, 1990. 14. ESTREICHER, A., WOHLWEND, A., BELIN, D., SCHLEUNING, W.D. and VASSALLI, J.D. Characterisation of the cellular binding site for the urokinase-type plasminogen activator. J. Biol. Chem. 264, 1180-1189, 1989. 15. RUSSELL, M.E., QUERTERMOUSE, T., DECLERCK, P.J., COLLEN, D., HABER, E. and HOMCY, C.J. Binding of tissue-type plasminogen activator with human endothelial cell monolayers. J.Biol. Chem. 265, 2569-2575, 1990. 16. MACARTNEY, H.W. Verfahren zur Abtrennung PA. EP 0 302 503 AZ, 1988.
und Reinigung
von t-
17. MARTIN, U., FISCHER, S., KOHNERT, U., RUDOLPH, R., SPONER, G., STERN, A. and STREIN, K. Coronary thrombolytic properties of a novel recombinantplasminogen activator (BM 06.022) in a canine model. J. Card. Pharm. 18, 111-119, 1991. 18. JAFFE, E.A., NACHLMAN, R.L., BECKER, C.G. and MINICK, C.R. Culture of human endothelial cells derived from umbilical veins. Identification by morphologiacl and immunological criteria. J. Clin. Invest. 52, 2745-2756, 1973.
Vol. 67, No. 5
CELL-BINDING OF &PA VARIANTS
599
19. JAFFE, E.A. Endothelial cells and the biology of factor VIII. N. Enul. J.Med. 296, 377-383, 1977. 20. NEELS, H.M., VAN SANDE, M.E. and SCHARPE, S.L. Sensitive calorimetric assay for angiotensin converting enzyme in serum. Clin.Chem. 29, 1399-1403, 1983. 21.
COMBRIER, E., METEZEAU, P., RONOT, X., GACHELIN, H. and ADOLPHE, M. Flow cytometric assessment of cell viability: a multifaceted analysis. Cvtotechnolouv 2, 27-37, 1989.
22. HOLVOET, P., LIJNEN, H.R. and COLLEN, D. Characterisation of functional domains in human tissue-type plasminogen activator with the use of monoclonal antibodies. Eur. J. Biochem. 158, 173-177, 1986. 23. KUBBIES, M. and STOCKINGER, H. Cell-cycle dependent DHFR and tPA production in cotransfected, MTX amplified CHO cells revealed by dual-laser flow cytometry. Exe. Cell Res. 188, 267271, 1990. 24. BEEBE, D.P. Binding of tissue plasminogen activator to human umbilical vein endothelial cells. Thromb. Res. 46, 241-254, 1987. 25. BARNATHAN, E.S., KUO, A., VAN DER KEYL, H., MCCRAE, K.R., KARSEN, G.R. and CINES, D.B. Tissue-type plasminogen activator binding to human endothelial cells. J. Biol. Chem. 263, 77927799, 1988.
