Forskolin down-regulates type-1 plasminogen activator inhibitor and tissue-type plasminogen activator and their mRNAs in human fibrosarcoma cells

Forskolin down-regulates type-1 plasminogen activator inhibitor and tissue-type plasminogen activator and their mRNAs in human fibrosarcoma cells

103 Molecuiar and Cellular Endocrinology, 72 (1990) 103-l 10 Elsevier Scientific Publishers Ireland. Ltd. MOLCEL 02331 Forskolin down-regulates ty...

856KB Sizes 0 Downloads 29 Views

103

Molecuiar and Cellular Endocrinology, 72 (1990) 103-l 10 Elsevier Scientific Publishers Ireland. Ltd.

MOLCEL

02331

Forskolin down-regulates type-l plas~nogen activator inhibitor and tissue-type plasminogen activator and their mRNAs in human fibrosarcoma cells Birgitte Georg I, Andrea

Riccio * and Peter Andreasen

1,3

’ institute of Biochemists

C, Uniuersity of Copenhagen, DK-2200 Copenhagen N, Denmark, 2 Centro di EndocrinoIogia ed 0ncoiogia Sperimentale, Consiglio Nazionaie delle Richer&e, Naples, rtal,: and ’ Institute of Molecular Biology and Plant Physiology, University of A*rhus, DK-8000 krhus C, Denmark (Received

Key words:

Forskolin; Cyclic AMP; Type-l plasminogen activator

inhibitor

12 April 1990; accepted

of plasminogen

21 May 1990)

activators;

Human

fibrosarcoma

cell line HT-1080;

Tissue-type

Summary We have studied the effect of the adenylate cyclase-stimulating agent forskolin on expression of components of the plasminogen activation system in the human fibrosarcoma cell line HT-1080. By enzyme-linked immunosorbent assays, forskolin was found to cause a 2 to 4-fold decrease in intracellular and culture medium levels of type-l inhibitor of plasminogen activators (PAI-1) and tissue-type plasminogen activator (t-PA). This was true for cells not treated with other agents and for cells, in which the PAI- and t-PA levels had been increased 5 to lo-fold by treatment with dexamethasone. This down-regulation could be traced back to corresponding decreases in the cellular levels of PAI- and t-PA mRNAs. Of the two PAI- mRNAs, the 2.4 kb species was 5-fold decreased by forskolin in cells treated with dexamethasone, while the 3.4 kb transcript was unaffected; in cells not treated with dexamethasone, forskolin affected the two PAZ-1 transcripts in parallel. These studies show that in addition to the many inducers of PAI-1, PAI- gene expression is also subject to negative modulation by cyclic AMP. They also show that t-PA gene expression, in contrast to the induction by cyclic AMP observed in many other cell lines, may also be subject to negative regulation by cyclic AMP. Thus, hormonal agents acting with cyclic AMP as a second messenger may be involved in down-regulating PAI- and t-PA expression in vivo.

introduction Plasminogen activators are serine proteases, that are capable of catalyzing the conversion of the abundant extracellular proenzyme plasminogen to the active protease plasmin. Plasmin has a rela-

Address for correspondence: Peter A. Andreasen, Institute of Molecular Biology and Plant Physiology, 130 C.F. MBllers AIIC, University of Arhus C, DK-8000 Arhus C, Denmark.

0303-7207/90/$03.50

0 1990 Elsevier Scientific

Publishers

Ireland,

tively low substrate specificity, being able to degrade many extracellular proteins, including fibrin and extracellular matrix components like fibronectin, laminin and proteoglycans. Cellular release of the activators is therefore able to initiate localized extracellular proteolysis. Two types of mammalian plasminogen activators have been identified; namely urokinase-type plasminogen activator (uPA) and tissue-type plasminogen activator (t-PA). The plasminogen activation system also includes the specific and fast-acting plasmin inhibitor (YeLtd.

104

antiplasmin, two specific and fast-acting inhibitors of plasminogen activators, PAIand PAX-2, and a specific cell surface u-PA binding protein, the u-PA receptor or u-PAR (for reviews, see Reich, 1978; Collen, 1980; Dan0 et al., 1985; Blasi et al., 1987; Saksela and Rifkin, 1988; Andreasen et al., 1990). Plasminogen activation has been implicated in a variety of biological processes. u-PA has been associated with processes involving tissue remodelling, invasion and tissue destruction, while t-PA seems to be active primarily in thrombolysis, but also in prohormone processing and ovulation. Many of these processes are regulated by horcytokines and growth factors, and mones, hormonal regulation of many such processes occurs in parallel with hormonal regulation of occurrence of the activators (for reviews, see Dana et al., 1985; Saksela and Rifkin, 1988). The hormonal regulation of other components of the system in such processes is less clear. However, PAI- has, in cultured cells, been found to be induced by a variety of agents, including glucocorticoids, phorbol esters, thrombin, transforming growth factor-p, tumor necrosis factor-a, interleukin-1, epidermal growth factor and insulin (for reviews, see Sprengers and Kluft, 1987; Thorsen and Philips, 1987; Kruithof, 1988a, b; Andreasen et al., 1990). Some of the hormonally regulated in which plasminogen activation has processes, been implicated, involve hormones acting with cyclic AMP as a second messenger (see Dans et al., 1985; Saksela and Rifkin, 1988), but so far, little attention has been paid to cyclic AMP regulation of PAI-1. We report here that forskolin, acting on cells by increasing cellular cyclic AMP levels, causes a decrease in PAIprotein and mRNA in the human HT-1080 fibrosacroma cell line. In addition, we have found that also t-PA is down-regulated by forskolin in these cells, in contrast to what is the case in other cell lines. Materials

and methods

Materials Forskolin, 3-isobutyl-1-methyl-xanthine and 8Br-cyclic AMP were purchased from Sigma. A radioimmunoassay kit for cyclic AMP determina-

tion was from The Radiochemical Centre, Amersham, U.K. The probe used for t-PA mRNA comprises 2560 bp of human t-PA cDNA cloned into the PstI site of pBR322 (Waller, 1984). The glyceraldehyde phosphate dehydrogenase probe was the plasmid pGPDH5, containing a full-length cDNA for the rat enzyme inserted into the PstI site of pUC19 (Fort et al., 1985). All other materials were those previously described (Andreasen et al., 1986; Lund et al., 1987, 1988; Mayer et al., 1988; Georg et al., 1989). Cell culture The cell line HT-1080, derived from a human fibrosarcoma (ATCC CCL 121, American Type Culture Collection, Rockville, MD, U.S.A.), was cultured in 100 mm diameter Petri dishes as previously described (Andreasen et al., 1986). At confluency (lo-20 X lo6 cells per dish), the cells were washed 2 times with 0.01 M sodium phosphate, pH 7.4, 0.15 M NaCl, 1 mM CaCl, and 0.5 mM MgCl, (PBS), and then maintained under serumfree conditions for 48-72 h before experiments. After experiments, cells and conditioned medium were harvested, and cell extracts for enzyme-linked immunosorbent assay (ELISA) prepared as described previously (Lund et al., 1988). Cyclic AMP assay Cells were incubated with or without forskolin for various periods of time. The incubation was stopped by quickly removing the medium, placing the dish on ice, and adding 1 ml ice-cold 6% trichloroacetic acid in PBS. The cells were removed with a rubber policeman, sonicated, and particulate material was removed by centrifugation. The trichloroacetic acid-soluble fraction was extracted 3 times with ether, buffered further by the addition of one-tenth volume of lo-fold concentrated PBS, and analyzed for cyclic AMP by radioimmunoassay. Miscellaneous procedures ELISAs for PAI-1, t-PA and u-PA, RNA preparation and analysis by hybridization of Northern blots and slot blots were performed as previously described (Andreasen et al., 1987; Lund et al., 1987, 1988; Mayer et al., 1988; Georg et al., 1989).

105

Results Effect of forskolin on cellular concentrations of cyclic AMP Generally, forskolin increases intracellular levels of cyclic AMP (see review by Seamon and Daly, 1986). We measured the level of cyclic AMP in HT-1080 cells incubated with forskolin for various time periods. Forskolin was found to cause a rapid and sustained increase in the level of cyclic AMP in the cells. The level remained 4 to 5-fold elevated for at least 48 h (Fig. 1A). The cyclic AMP level was also measured in cells incubated for 48 h with different concentrations of forskolin. As shown in Fig. lB, forskolin caused an increase in the cyclic AMP level in concentrations between lo-’ and 10e5 M. Effect of forskolin on PAI-I and t-PA protein Table 1 shows a representative experiment on the effect of forskolin on HT-1080 cells. Forskolin was found to cause a 2 to 4-fold decrease in the amount of PAIaccumulated in serum-free conditioned medium over a 48 h period. Fig. 2 shows a time-course study, the PAIlevel being measured by ELISA after different times of incuba-

Fig. 1. Effect of forskolin on cyclic AMP level in HT-1080 cells. IN A, forskolin (final concentration 12.5 PM) was added to serum-free cultures of HT-1080 cells at the indicated time points before harvest. The insert shows the time course of the cyclic AMP level during the first minutes after forskolin addition, with an expanded time scale. In B, the indicated concentrations of forskolin were added to serum-free cultures of HT-1080 cells, which were then incubated for 48 h before harvest. Mean and range of the cyclic AMP level in two independent cultures are indicated.

TABLE

1

FORSKOLIN REGULATION OF PAI-1, u-PA AND t-PA PROTEIN LEVELS IN HT-1080 CELLS, AS MEASURED BY ELISA Confluent cell cultures were incubated under serum-free conditions for 48 h without additions, with 1 PM dexamethasone, with 24 PM forskolin, or with 1 PM dexamethasone plus 24 aM forskolin. After incubation, the conditioned media were harvested and analyzed for PAI-1, t-PA and u-PA by ELISA. Additions

PAI(pg per dish)

t-PA (pg per dish)

u-PA (ag per dish)

None Dexamethasone Forskolin Dexamethasone + forskolin

2.97+0.33 47.34& 6.10 1.02*0.16

0.22*0.01 0.65 +0.03 0.14+0.01

10.38+0.51 2.38+_0.13 9.00 + 0.43

24.82 * 1.97

0.27 + 0.02

1.04 f 0.06

tion of the cells with forskolin. Forskolin causes a steadily decreasing amount of PAIpresent in cell extracts and accumulated in the conditioned medium, with an approximately 12 h lag period. There is no further decrease after 72 h of treatment. Fig. 3 shows the concentration dependence of forskolin over a 96 h incubation period. The effect on the level of PAIprotein is seen to be half-maximal at a forskolin concentration of approximately 1 PM. This is similar to the concentration dependence for other effects of forskolin (see review by Seamon and Daly, 1986), and the PAIdecrease occurred over the same forskolin concentration range as the increase in the cellular cyclic AMP level (Fig. 1B). The amount of t-PA accumulated in the conditioned medium was also suppressed by forskolin (Table 1). A time-course study showed that after having reached a minimum level after incubation for 48 h, the t-PA level started to increase again, suggesting that prolonged incubations lead to attenuation of the forskolin effect on t-PA production (Fig. 4). An effect identical to that of forskolin could be achieved with 3 mM 8-Br-cyclic AMP plus 0.2 mM 3-isobutyl-1-methyl-xanthine (not shown). After treatment with forskolin, it was observed that some cells detached from the substratum, but the effect of forskolin was not just one of causing cell death and an ensuing decrease in PAI- production. This is verified by the following findings:

106

fu, B

I

6000 -

I

0

0

[email protected] 10-710-6 lo-510-' to-3 Forskah

Time

1hours)

Fig. 2. Time course of forskolin-induced decrease of PAIprotein in HT-1080 cells. Forskohn (final concentration 48 PM) was added to serum-free cultures of HT-1080 cells at the indicated time points before harvest. Cell extracts (A) and conditioned media (B) were analyzed for PAIby ELBA. The bars indicate standard deviations. The ELBA results are expressed as ng PAI- per dish. Note that the medium of the cultures was not changed during the 96 h incubation period. If instead the medium was changed after 48 h, and the cells then incubated for an additional 48 h, forskolin being absent or present for the whole 96 h incubation period, the ELISA values were: conditioned medium, control cells, 3.30+0.26 pg per dish; forskolin-treated cells, conditioned medium: 1.80 5 0.22 pg per dish; control cells, cell extract: 0.22*0.02 pg per dish; forskolin-treated cells, cell extract: 0.09 f 0.02 /.tg per dish.

there is no decrease in cellular production of u-PA (Table 1); the forskohn treatment does not lead to qualitative changes in the pattern of proteins appearing on Coomassie blue-stained sodium dodecyl sulfate (SDS)-polyacrylamide gels (data not shown); moreover, forskolin does not cause any significant changes in the number of cells per dish, as judged from the amounts of protein, DNA and RNA. Combined effect of forskolin and dexamethasone on PAI-I and t-PA protein Dexamethasone increases the production of PAI- and t-PA in HT-1080 cells (Andreasen et

10-810-7 10-b 10-~10~‘10-~

[MI

Fig. 3. Effect of different concentrations of forskolin on the amount of PAI- protein in HT-1080 cells. Confluent cultures of HT-1080 cells were incubated for 96 h with the indicated concentrations of forskolin. Cell extracts (A) and conditioned media (B) were analyzed for PAI- protein by ELISA. Other experimental conditions were as described in the legend to Fig. 2. Separate experiments (not shown) demonstrated that there was no further decrease in the PAI- concentration by increasing the forskolin concentration above 48 FM, which is the maximal one used in the experiment shown.

800

3 200

-0

50 Time

1’00

(hours1

Fig. 4. Time course of forskolin-induced decrease of t-PA protein in HT-1080 cells in the absence and presence of dexamethasone. Serum-free cultures of HT-1080 cells were incubated for 96 h in the absence (0) or the presence (0) of 1 pM dexamethasone. Forskolin (final concentration 12.5 PM) was added to the cultures at the indicated time points before harvest. Conditioned media were analyzed for t-PA by ELISA. Other experimental conditions were as described in the legend to Fig. 2.

107

al., 1986; Medcalf et al., 1988b). Forskolin was found to decrease the amount of PAI- and t-PA in cell extracts and conditioned medium also in the presence of dexamethasone, but the dexamethasone-induced increases of PAIand t-PA protein are not prevented by forskolin. The time course of the dexamethasone-induced increase of PAI- and t-PA protein in cell extracts and conditioned medium is indistinguishable in the presence and absence of forskolin (data not shown). Likewise, the forskolin-induced decrease in PAI- protein proceeded largely in parallel in the absence and presence of dexamethasone (data not shown). The time courses of the forskolin-induced decreases of t-PA protein in the absence and presence of dexamethasone were parallel during the first 48 h, but in the presence of dexamethasone, the t-PA level did not return to control levels after prolonged incubation times, as it did in the absence of dexamethasone (Fig. 4). In combination with dexamethasone, which strongly down-regulates u-PA in HT-1080 cells, forskolin causes a further decrease in the u-PA protein level (Table 1). This is reminiscent of the down-regulation of u-PA by cyclic AMP seen in other cell lines (see review by Dan0 et al., 1985) and we did therefore not study it in further detail.

Effect of forskolin

on PAI-I

and t-PA mRNAs

Fig. 5 shows Northern blots of total RNA from cells incubated for 48 h without additions with forskolin, with dexamethasone, and with the two in combination. Scanning of the Northern blots hybridized with PAI- cDNA showed a 2 to 5-fold decrease in the abundance of PAImRNA in forskolin-treated cells. As previously reported (Andreasen et al., 1987) dexamethasone causes a strong induction of PAImRNA. There are two PAImRNAs, arising by differential choice of polyadenylation sites (Loskutoff et al., 1987; Bosma et al., 1988; Riccio et al., 1988; Strandberg et al., 1988). Neither dexamethasone nor forskolin, when added separately, affect any of the two transcripts preferentially. However, the two in combination change the ratio between the two transcripts drastically; in the absence of the agents, the ratio is 2 in favor of the shorter transcript, but in their presence, it is 0.5. This corresponds to an approximately 4.5-fold reduction in the abundance of the short transcript in cells treated with dexamethasone plus forskolin as compared to cells treated with dexamethasone alone, while the long transcript is not affected. Forskolin causes a decrease in the abundance of the t-PA mRNA, to 0.6 times the control value.

3.4Kb2.3Kb-

a

b

c

d

a’

b’

c’

d’

a” b”

c” d”

Fig. 5. Northern blot analysis of PAI- and t-PA mRNA in HT-1080 cells incubated with forskolin and dexamethasone. Total RNA was isolated from cells incubated for 48 h without additions (a, a’, a”); with 48 pM forskolin (b, b’, b”); with 1 pM dexamethasone (c, c’, c”); and 48 pM forskolin plus 1 pM dexamethasone (d, d’, d”). 20 pg portions of RNA were electrophoresed in agarose gels under denaturing conditions and blotted onto nitrocellulose filters, which were hybridized to ‘*P-labelled PAI- cDNA (a-d), glyceraldehyde 3-phosphate dehydrogenase cDNA (a/-d’) or t-PA cDNA (a”-d”). The sizes of the PAL1 mRNAs, the glyceraldehyde 3-phosphate dehydrogenase mRNA and the t-PA mRNA are indicated on the left of the blots.

108

Fig. 6. Time course of forskolin-induced decrease in PAI- and t-PA mRNA. Total RNA was isolated from cells that had been incubated with forskolin for the indicated time periods The relative amounts of the two mRNAs at each time point were estimated by spectrophotometric scannings of autoradiograms of slot-blot filters hybridized with 32P-labelled PAIcDNA (A) and t-PA cDNA (B), after normalization against the corresponding relative amounts of glyceraldehyde phosphate dehydrogenase mRNA (see Fig. 5 and Materials and Methods for further details). The mRNA levels at time 0 have been set equal to 1.0. and the data at subsequent time points expressed as fold reduction.

Dexamethasone causes a 3.0-fold increase in t-PA mRNA in the absence of forskolin, and a 1.7-fold increase in the presence of forskolin (Fig. 5). In contrast, the level of glyceraldehyde 3-phosphate dehydrogenase mRNA is not affected by either dexamethasone or forskolin (Fig. 5). The time course of these changes in PAI- and t-PA mRNA was studied by slot-blot analysis of RNA isolated from cells that had been incubated for different time periods with forskolin (Fig. 6). Both mRNAs started to decrease after a lag period of around 2 h, and reached the new steady-state level after 12 h. As expected, the decreases in the mRNA levels occur faster than the corresponding decreases in the levels of the corresponding proteins. After prolonged incubation, the t-PA mRNA started to increase again, as also shown for t-PA protein in Fig. 4. Discussion In the present report, we describe studies of the effect of the cyclic AMP inducing agent forskolin on components of the plasminogen activation system in the human fibrosarcoma cell line HT-1080. We show, by the use of a specific and sensitive

ELISA, that forskolin decreases the amount of PAIprotein in conditioned medium and cell extracts of HT-1080 cells. The parallel changes of PAIin medium and cells indicate that it is not the release from the cells, but the biosynthesis of PAIthat is affected. The change in PAIprotein could be traced back to a decrease in cellular PAImRNA level. A variety of hormones, cytokines and growth factors have previously been reported to induce PAImRNA in cell lines (for a review, see Andreasen et al., 1990). The present report is the first one on down-regulation of PAImRNA, by an agent affecting the cellular cyclic AMP level. There have been previous indications that cyclic AMP may regulate PAIgene expression: follicle-stimulating hormone (FSH) and luteinizing hormone (LH), which are generally believed to have cyclic AMP as a second messenger, downregulate PAIactivity, as assayed by reverse zymography, in primary cultures of FSH-primed rat granulosa cells (Ny et al., 1985) and a 2-fold decrease in PAIantigen and abolishment of phorbol ester induction of PAIantigen in cultured human umbilical cord endothelial cells is also found after cyclic AMP treatment (Sante11 and Levin, 1988). Heaton et al. (1989) in a recent report, described an effect of cyclic AMP on the production of PAIby primary cultures of rat hepatocytes. Contrary to the findings with human fibrosarcoma cells in the present report, they found that cyclic AMP causes an increase in the level of PAIprotein and mRNA in the rat hepatocytes. This apparent discrepancy may be explained by species and/or cell specificity of the response of PAI- to cyclic AMP. Tumor necrosis factor-a (TNF-(Y) induces PAI1 in HT-1080 cells (Medcalf et al., 1988a; Georg et al., 1989) in the carcinoma cell line T-CAR1 (Georg et al., 1989) and in cultured human endothelial cells (van den Berg et al., 1988; van Hinsbergh et al., 1988). It has been reported that for some cellular actions of TNF-q cyclic AMP is an intracellular second messenger (Shirakawa et al., 1988; Zhang et al., 1988). However, forskolin decreases PAIproduction in HT-1080 cells, as reported here. This observation does not support the view that the TNF-cu effect on PAIoccurs

109

via cyclic AMP, and that this cytokine generally acts with cyclic AMP as a second messenger. We further demonstrate here a change in t-PA protein that can be traced back to a parallel change in t-PA mRNA. The observation of down-regulation of t-PA and its mRNA by forskolin is novel and remarkable; in other cells studied, t-PA has always been found to be induced by agents affecting cellular cyclic AMP levels (for reviews, see Dan0 et al., 1985; Saksela and Rifkin, 1988). Dexamethasone is a strong inducer of PAIprotein and mRNA in some cell lines, while in others, it is without effect (for a review, see Andreasen et al., 1990). We show here that forskolin decreases PAIprotein and mRNA in HT-1080 cells even in the presence of dexamethasone, but it does not prevent dexamethasone from causing a strong increase in PAI- production. As previously reported (Andreasen et al., 1986; Medcalf et al., 1988b), dexamethasone induces t-PA protein in HT-1080 cells, and we report here that forskolin decreases t-PA protein in the presence as well as in the absence of dexamethasone. The negatively regulatory events described here are likely to be at least partly due to effects on the transcription rate of the genes. Two c&acting elements, CRE and AP-2 element, have been implicated in cyclic AMP-dependent regulation of transcription of specific genes (for a review, see Roesler et al., 1988), and one may speculate that such elements are involved in the present effects of forskolin. It is also possible that a cascade of various gene activations takes place between the addition of forskolin and the effects on the PAIand t-PA genes. Alternatively, the changes in mRNA levels may be caused by changes in mRNA stability. Further studies are needed to elucidate fully the intracellular mechanisms of the described effects of forskolin. In the presence of dexamethasone, the forskolin-induced decrease in PAI- mRNA is mainly associated with the short PAItranscript. The 3’-end of the 3.4 kb PAI- transcript contains a 75 bp partly palindromic (AT)-rich sequence (Ginsburg et al., 1986; Ny et al., 1986; Pannekoek et al., 1986) similar to sequences present in a variety of other mRNAs (Treissman, 1985; Caput et al., 1986). Such a region has been reported to confer

cycloheximide-reversible instability to a heterologous mRNA (Shaw and Kamen, 1986). In many cell lines, cycloheximide induces an increase in PAI- mRNA, and this increase is mainly associated with the longer transcript (for a review, see Andreasen et al., 1989). The cycloheximide effect on PAImRNA may therefore have to do with an increased stability of the longer transcript. Likewise, the increase in the relative abundance of the long transcripts in the presence of forskolin and dexamethasone may in part be due to an increase in the relative stability of this transcript. Such an increase would counteract the total decrease in PAI- mRNA. At present, it is not clear which cell types and organs produce PAInor which hormones, cytokines and growth factors regulate PAIproduction in the intact organism (see Andreasen et al., 1990). The present findings indicate that the expression of the PAI- gene is susceptible to cyclic AMP, and that PAIexpression in vivo may be regulated by hormonal agents acting with cyclic AMP as a second messenger. In addition, the present results suggest that in vivo, hormones acting with cyclic AMP as a second messenger may be involved in down-regulation of t-PA expression. Acknowledgements Dr. S.J.F. Degen is thanked for providing the t-PA cDNA, and Dr. P. Fort for the glyceraldehyde phosphate dehydrogenase cDNA. The excellent technical assistance of Lisbeth 0sterskov Andersen, Hanne Baasch, Bente Johannessen and Klavs Tanning-Sorensen is gratefully acknowledged. This work was supported financially by the Danish Cancer Society and the Danish Medical Research Council. References Andreasen, P.A., Christensen, T.H., Huang, J.-Y., Nielsen, L.S., Wilson, E.L. and Dana, K. (1986) Mol. Cell. Endocrinol. 45, 137-147. Andreasen, P.A., Pyke, C., Riccio, A., Kristensen, P., Nielsen, LX, Lund, L.R., Blasi, F. and Dam,, K. (1987) Mol. Cell. Biol. 7, 3021-3025. Andreasen, P.A., Georg, B., Lund, L.R., Riccio, A. and Stacey, S.N. (1990) Mol. Cell. Endocrinol. 68, l-19.

110 Blasi, F., Vassalli, J.-D. and Dano, K. (1987) J. Cell Biol. 104, 801-804. Bosma, P.J., van den Berg, E.A. and Kooistra, T. (1988) J. Biol. Chem. 263, 9129-9141. Caput, D., Beutler, B., Hartog, K., Thayer, R., Browh-Shimer, S. and Cerami, A. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 1670-1674. Collen, D. (1980) Thromb. Haemostasis 43, 77-89. Dane, K., Andreasen, P.A., Grondahl-Hansen, J., Kristensen, P., Nielsen, L.S. and Skriver, L. (1985) Adv. Cancer Res. 44, 139-266. Georg, B., Helseth, E., Lund, L.R., Skandsen, T., Riccio, A., Drum, K., Unsgaard, G. and Andreasen, P.A. (1989) Mol. Cell Endocrinol. 61, 87-96. Ginsburg, D., Angela, R.Z., Yang, A.Y., Rafferty, U.M., Andreasen, P.A., Nielsen, L., Dana, K. and Gelehrter, T.D. (1986) J. Clin. Invest. 87, 1673-1680. Fort, P., Marty, L., Piechaczyk, M., El Sabrouty, S., Dani, C., Jeanteur, P. and Blanchard, J.M. (1985) Nucleic Acids Res. 13, 1431-1442. Heaton, J.H., Nebes, V.L., O’Dell, L.C., Morris, SM. and Gelehrter, T.D. (1989) Mol. Endocrinol. 3, 185-192. Kruithof, E.K.O. (1988a) in Tissue-Type Plasminogen Activator, Clinical and Physiological Aspects (Kluft, C., ed.), pp. 189-209, CRC Press, Boca Raton, FL. Kruithof, E.K.O. (1988b) Inhibitors of Plasminogen Activators (Dissertation), Faculty of Science, University of Lausanne, Lausanne. Loskutoff, D.J., Linders, M., Keijer, J., Veerman, H., van Heerikhuizen, H. and Pannekoek, H. (1987) Biochemistry 26, 3763-3768. Lund, L.R., Riccio, A., Andreasen, P.A., Nielsen, L.S., Kristensen, P., Laiho, M., Saksela, O., Blasi, F. and Dano, K. (1987) EMBO J. 6, 1281-1286. Lund, L.R., Georg, B., Nielsen, L.S., Mayer, M.. Dane, K. and Andreasen, P.A. (1988) Mol. Cell. Endocrinol. 60, 43-53. Mayer, M., Lund, L.R., Riccio, A., Skouv, J., Nielsen, L.S., Stacey, S.N., [email protected], K. and Andreasen, P.A. (1988) J. Biol. Chem. 263, 15688815693. Medcalf, R.L., Kruithof, E.K.O. and Schleuning, W.D. (1988a) J. Exp. Med. 168, 751-759. Medcalf, R.L., van den Berg, E. and Schleuning, W.-D. (1988b) J. Cell Biol. 106, 971-978. Ny, T., Bjersing, L., Hsueh, A.J.W. and Loskutoff, D.J. (1985) Endocrinology 116, 1666-1668.

Ny, T., Sawdey, M., Lawrence, D., Millan, J.L. and Loskutoff, D.J. (1986) Proc. Natl. Acad. Sci. U.S.A. 83, 6776-6780. Pannekoek, H., Veerman, H., Lambers, H., Diergaarde, P., Verweij, CL., van Zonneveld, A.-J. and van Mourik, J.A. (1986) EMBO J. 5, 2539-2544. Reich, E. (1978) in Molecular Basis of Biological Degradative Processes (Berlin, R.D., Hermann, H., Lepow, I.H. and Tanzer, J.M., eds.), pp. 155-169, Academic Press, New York. Riccio, A., Lund, L.R., Sartorio, R., Lania, A., Andreasen, P.A., Dane, K. and Blasi, F. (1988) Nucleic Acids Res. 16. 2805-2824. Roesler, W.J., van den Bark, G.R. and Hansson, R.W. (1988) J. Biol. Chem. 263, 9063-9066. Saksela, 0. and Rifkin, D.B. (1988) Annu. Rev. Cell Biol. 4, 93-126. Santell, L. and Levin, E.G. (1988) J. Biol. Chem. 263, 1680216808. Seamon, K.B. and Daly, J.W. (1986) Adv. Cyclic Nucleotide Protein Phosphorylation Res. 20, l-150. Shaw, G. and Kamen, R. (1986) Cell 46, 659-667. Shirakawa, F., Yamashita, U., Chedid. M. and Mizel, S.B. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 8201-8205. Sprengers, E.D. and Kluft, C. (1987) Blood 69. 381-387. Strandberg, L., Lawrence, D. and Ny, T. (1988) Eur. J. Biothem. 176, 609-616. Thorsen, S. and Philips, M. (1987) in Fundamental and Clinical Fibrinolysis (Castellino, F.J., Gaffney, P.J., Samama. M.M. and Takada, A., eds.), pp. 83-98, Elsevier, Amsterdam. Treissman, R. (1985) Cell 42, 889-902. van den Berg, E.A., Sprengers, E.D., Jaye, M., Burgess, W., Maciag, T. and van Hinsbergh, V.W.M. (1988) Thromb. Haemostasis 60, 63-67. van Hinsbergh, V.W.M., Kooistra, T., van den Berg, E.A., Princen, H.M.G., Fiers, W. and Emeis, J.J. (1988) Blood 72, 1467-1473. van Zonneveld, A.-J., Curriden, S.A. and Loskutoff, D.J. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 5525-5529. Waller, E.K. (1984) Ph.D. Thesis, Rockefeller University, New York. Zhang, Y., Lin, J.-X., Yip, Y.K. and Vilcek, J. (1988) Proc. Natl. Acad. Sci. U.S.A. 85, 6802-6805.