EXPERIMENTAL

CELL

RESEARCH

P9@127-132

(1990)

Regulatory Activities of Endothelial Extracell Mediation by Transforming Growth IF LE~LEY K. NEWTON, W. K. ALFRED Department

YUNG,

L. CREE

PETTIGREW,

AND

PETER

of Neuro-oncology, The University of Texas M. D. Anderson Cancer Center and Department The University of Texas Health Science Center at Houston, Houston, Texas 77030

A.

STT’ECK?

of Neurology,

sponses, including regulation of ~el~~~a~ proliferation, modulation of extracellular matrix ( ) accumulation, immunoregulation, and di entiatisu [I-3]. Several distinct gene products of TGF-P family have been identified; each initiates e similar responses, but these gene products also display independent activities, as illustrated by the immu~osu~~r~ssive activities observed only with TGF-/I2 [4, 5]. Additionally, the multifunctional cellular responses to TGF-P appear to be dependent on that factor’s specific interaction with various cell types, such as mesenchymal and epithelial cells exhibiting either growth-stimulator-y or -inhibitory effects, respectively, in response to mechanism(s) of action by which growth control and other biologic &ions is unknown TGF-P has been shown to interact with specific high affinity cell surface receptors quent pathway(s) has not been elu general, TGF-fl has been shown to rice the accumulation of ECM by stimulating the synthesis of specific matrix eomponents and also by inhibiting matrix degradation [l, 2, 10, PI]. The ECM provides structural support a.nd directs a number of biological functions similar to those mediated by TGF-/3 [12]. Cellular prolifer and differentiation have been shown to [E&15]. The matrix components present in the also provides the framework for cell o~i~nt,ation, adhesion, and morphology, and plays irn~ort~~t roles in the selective filtration of solutes. The major components of ECM consist of collagens, elastins, hyaluronic acid, glycoproteins, and proteoglycans, of which the latter two classes of constituents have been irn~~~at~?d in reflecting or regulating cellular proliferations F cific substructures of heparan sulfate ( shown to possess growth ~nb~~~t~~~ various glycoproteins and gly~Q~o~j~~~~tes exhibit either growth-inhibitory or -stimula&ory properties [16-191. status and because Since TGF-fi modulates EC ECM exhibits some properties similar to tjhose elicited by TGF-/3, we initiated an invest’ to examine may be mewhether the cellular responses to

The potential of transforming growth factor-p (TGF@) to modulate the growth of endothelial cells via alterations in the cell’s extracellular matrix (ECM) was examined. Rat brain endothelial cells were cultured in the presence or absence of TGF-b, and subsequently ECM was prepared from the cell cultures by hypotonic lysis of the cells. Untreated endothelial cells were then cultured on the various matrices. Cells grown on TGF+?treated ECM showed a significant decrease in cell numher (41 i 6% mean growth inhibition at 6 days, P < 0.005 by paired T-test) compared with cells grown on untreated ECM. The growth inhibitory activity of the ECM was depleted by 9 days of culture, and resumption of exponential cell growth was observed. A similar phenomenon was observed if anti-TGF-j3 neutralizing antibodies were incubated with the ECM. When the TGF-/3treated matrix was exposed to a brief dithiothreitol treatment in order to inactivate residual TGF-fi, an approximately equal degree of growth inhibition was observed initially, but the reversal of inhibition occurred at an earlier time point than that with unreduced TGFB-treated matrix. Analyses of the composition of matrices synthesized in the presence or absence of TGF-/3 re,vealed about a twofold increase in the accumulation of various radioactive metabolic precursors in the TGF-@treated matrices. However, no qualitative alterations in the matrix or cellular-associated proteins or glycoproteins were observed, as analyzed by polyacrylamide gel eleetrophoresis in sodium dodecyl sulfate. An increase in cell-associated heparan sulfate, however, was observed in TGF-P-treated cells. The results suggest that certain growth regulatory effects of TGF-8 may be mediated, at least in part, by alterations in the ECM. o 1990 Academic press, kc.

INTRODUCTION Transforming growth factor-/3 (TGF-P) has been shown to elicit a wide variety of different biological re1 To whom correspondence and reprint requests should be addressed at The University of Texas, M. D. Anderson Cancer Center, Department of Neuro-oncology, 118,1515 Holcombe Blvd., Houston, TX 17030. 127

All

CopyCgRt, 0 1990 rights of reproduction

QQ14-4827/90 $3.00 by Academic Press, Inc. in any form reserved.

NEWTON

128

diated, at least in part, by alterations in the ECM. To examine this hypothesis, we used endothelial cells, which have been previously shown to synthesize significant quantities of ECM and also to be responsive to the effects of TGF-P [21-241. Our results suggest that induced alterations in ECM have a role in mediation of growth regulation by TGF-P. MATERIALS

AND

METHODS

Cell culture. Rat brain endothelial (RBE) cells were isolated using techniques previously described [25]. Briefly, new-born rat brains were obtained, washed, minced, and then dissociated using a loosefitting Dounce homogenizer. The cell extract was collected by centrifugation (500g for 10 min), resuspended in 2 vol of cold PBS containing 20% dextran (A& - 84,000; Sigma, St. Louis, MO), and repelleted at 1700g for 15 min. The supernatant was discarded and the pellet was washed again with a cold 15% dextran solution, followed by resuspension in 10 ml of Dulbecco’s modified minimal essential medium (DMEM) containing 25 mM Hepes, pH 7.2. The microvessels were passed through a lOO+m nylon mesh screen and retained on a 50-pm screen. The purity of the preparation was examined by phase-contrast microscopy. The microvessels were then dissociated by collagenase treatment (0.5%) in DMEM for 30 min at 37°C. The cells were then plated on gelatin-coated dishes in culture medium of Dulbecco’s minimal essential medium/Hams F12 medium (DME/F12; l/l; v/v) containing 2% human platelet-poor plasma (PPP), 2% fetal bovine serum (FBS, Hyclone, Ogden UT), and 100 wg/ml endothelial cell growth factor (ECGF, Collaborative Research, Boston, MA) and supplemented with 10 U/ml heparin (growth medium). Colonies that exhibited endothelial morphology were subcultured and subsequently cloned, and frozen cell stocks were maintained. Cells were routinely cultured in growth medium without heparin on gelatin-coated plates. The cells were characterized for endothelial properties including morphology, nonthrombogenic cell surface [25], and expression of factor VIII antigen [26]. The cells were routinely examined and found to be free of mycoplasma and virus contamination. Cells were used between in vitro passages 3-12. For TGF-P treatment, cells in logarithmic growth were treated with or without TGF-P (R&D Systems, Minneapolis, MN; TGF-/31 from porcine platelets) at various concentrations (0.01 to 5 rig/ml). AntiTGFP neutralizing antibody (R&D Systems, AB-lo-NA) was preincubated with TGF$ or ECM for 1 h at room temperature at concentrations ranging from 10 to 100 pg/ml. Cell number was determined after various periods of time in culture by harvesting the cells in 0.125% trypsin and 2 mA4 EDTA and counting them in a hemocytometer. Preparation and characterization of endothelial ECM. RBE cells were grown on gelatin-coated 60-mm tissue culture dishes or 12-mm glass coverslips in the presence or absence of TGF-P (2 rig/ml). The TGF-P was added to the cultures when the cells were 70-80% confluent, which allowed for an additional cell division and a confluent cell monolayer to be obtained. The monolayers were maintained lo14 days postconfluence to allow for matrix accumulation. The cultures were washed twice with Ca’+- and Mg2+-free PBS, followed by approximately five to seven washes with 5 m&f Hepes, pH 7.4. The plates were monitored for cell lysis by inverted phase microscopy. The plates were then washed an additional three times with PBS. For cellular proliferation studies, untreated RBE cells were plated (2.5 X lo4 cells/dish) onto the various matrices in growth medium. Cell number determinations were performed after appropriate time intervals. The ECM was also subjected to various treatments to partially characterize its composition and biological activities. Certain ECMcontaining plates were treated with 2 mM dithiothreitol for 10 min at room temperature. Alternatively, some of the ECM plates were incu-

ET AL. bated with chondroitinase ABC (0.1 U/ml), Streptomyces hyaluronidase (5 TRU/ml), heparitinase (5 mu/ml; ICN, Lisle, IL), or collagenase (10 wg/ml; Sigma), for 30 min at 37°C. After the various treatments, the plates were washed an additional three times with PBS prior to addition of RBE cells. For metabolic radioactive labeling of ECM and ECM-producing cells, the cells were cultured the last 48 h prior to lysis with either [3H]glucosamine, [aH]galactose, [aH]fucose, [3H]leucine, [‘Hlproline (10 &i/ml; ICN, Irvine, CA), [?!?]methionine (10 &i/ml), or [9]sulfate (50 &i/ml; DuPont, Boston, MA). The cells and ECM were harvested as described above, except that the Hepes solution confluoride (PMSF), 2 mM EDTA, tained 1 mM phenylmethylsulfonyl and 0.1% aprotinin. The ECM was obtained by scraping and solubilization in sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) sample buffer (2% SDS, 10% glycerol, 0.125 M Tris PO,, [pH 6.71, 2 mM EDTA, and 1% 2-mercaptoethanol.) The extracts were then subjected to SDS-PAGE as previously described [27] based on equal quantities of macramolecular radioactivity. Characterization of glycosaminoglycans. The glycosaminoglycans (GAGS) were isolated and characterized as previously described [27]. The GAGS were identified and quantitated by ion-exchange chromatography or cellulose acetate electrophoresis and by sensitivity to chondroitinase ABC, Streptomyces hyaluronidase, or nitrous acid as previously described [29, 301. The degradation fragments from the various treatments were analyzed by previously described procedures [29-311.

RESULTS

AND

DISCUSSION

Endothelial cells were isolated, cloned, and characterized from rat brain. The RBE cells exhibited a cobblestone morphology (Fig. 1) and initially expressed factor VIII antigen, although its presence was decreased upon prolonged in vitro passage. The cells consistently displayed a nonthrombogenic cell surface. The effects of TGF-P on the proliferation of RBE cells were initially examined in monolayer culture. Growth inhibition was initially observed at a TGF-fl concentration of >0.025 rig/ml with a medium effective dose (ID& of -0.75 rig/ml, and a maximal inhibition (>90%) was seen at doses > 1 rig/ml in the continued presence of endothelial growth factor and optimal culture conditions. Growth inhibition was consistently obtained at a dose of 2 rig/ml and this concentration of TGF-P was used in the majority of studies. The preincubation of TGF-P with neutralizing antibodies or DTT (2 mM) was shown to inhibit the growth modulation activities of TGF-P (Table 1). The morphology of RBE cells was dramatically altered in response to TGF-& with the cells increasing four- to sixfold in size (Fig. 1). Scanning election microscopy showed similar results. The morphological change also appeared to be associated with induction of cell senescence as previously reported [23], because removal of the enlarged cells by trypsinization and replating onto new culture dishes did not result in their resumption of cellular growth. The growth inhibitory effects of TGF-P were mimicked in part by the ECM synthesized by RBE cells treated with TGF-P. Confluent RBE cells were cultured lo-14 days in the presence or absence of TGF-fi, and the ECM was obtained by hypotonic lysis of the cells in the

TGF-P-MEDIATED

GROWTH

REGULATION

FIG. 1. The morphologies of RBE cells grown on gelatin-coated plates (A and B) and on ECM synthesized by RBE cells (6, D, and E). Cells were untreated (A) or treated (B) with TGF-P (2 pg/ml) for 6 days. The matrices were prepared from cells grown in the absence (C) or presence (D, E) of TGF-P, and the untreated RBE cells were cultured on the various matrices. The ECM in D was exposed to a brief treatment with DTT before culturing of the cells. Magnification X70.

presence or absence of protease inhibitors (Table 1). Attempts to use EDTA or EGTA to remove the cells from the ECM were ineffective, resulting in 10 to 20% of the cells remaining attached to the matrix. Untreated RBE cells were then added to the various matrix-coated plates and cell number was determined after predetermined time intervals. Cells grown on untreated matrix showed a logarithmic increase in cell number over 14 days in culture (Fig. 2). Pn contrast, RBE cells plated onto ECM from TGF-P-treated cells showed an initial growth inhibition at 8 days or -85% growth inhibition compared witb that in control cells based on cell numbers. A partial reversal of the growth inhibition occurred after longer periods in culture, with the cells exhibiting an exponential growth rate nearly similar to that initially observed with the control cells (Fig. 2).

This observation suggests that the e cts of the TGF-/3ine the possibilsynthesized ECM are not toxic. To e ity that residual TGF-/3 in the matrix was responsible for the growth inhibition, the matrices were ,treated with dithiothreitol (DTT) prior to inoc~~a~i~~ of the cells. TT-treated TGF-/? The growth of RBE cells on the matrix was similar to that on the no trix except that the reversal of curred at a time period approxim tween times for control and TGF-P matrices (Fig. 2). To further examine the possible effect of residual TGF-P in the matrix, the ECM was preinc d with anti-TGF-8 neutralizing antibodies and the ells were continuously cultured in the presence of the antibody on the matrix (Table 1). Only a partial reversal o bition was observed in RBE cells culture

130

NEWTON

ET AL.

TABLE Growth

Response

of RBE

Cells

Cultured

on ECM

TGF-/3 b matrix

Treatment’

+ + + + + -

None DTT Ab (10 ad Ab (100 /x4 DTT + Ab (100 pg) TGF-fl (2 ng) TGF-@ + Ab (2 ng + 100 pg) TGF-/3 + DTT (2 ng + 2 mM) Chondroitinase ABC (0.1 U) St. Hyaluronidase Heparitinase Collagenase (100 rg) Trypsin (10 rg)

1

after

Treatment

Cell no.’ (X104) 64+ 33k 67k 37f 3Ok 37+ 40t 26+ 622 66-t 38+43 -t 52+ 41+ 54+ 57k

+ + + + + -

3 6 5 2 9 5 2 7 5 8 5 10 9 2 3 8

of the Matrix

with

% of control 100 51 100 57 48 59 62 40 96 100 59 67 81 64 84 89

Various

Agents % of matrix remainingd (aH/?S) 100 100 99199 98/99 nd nd nd nd nd nd 94186 97191 30122 91/95 10115 12/16

a The ECM was prepared and treated as described under Materials and Methods. DTT refers to a lo-min treatment with dithiotheritol and Ab is an abbreviation for anti-TGF-P neutralizing antibody, which was preincubated with the ECM for 1 h and was continuously present in the culture. All quantities of reagents are based on the amount per 1 ml. b The RBE matrix was prepared in the presence or absence of TGF-@ (2 rig/ml) for 14 days. All treatments were performed on matrices synthesized both with and without TGF-0, but the - matrix that was similar (within 10%) to control (no treatment) was not reported. ’ The cell number was determined after 6 days of exposure of the RBE cells to the various matrices. d The percentage of matrix remaining was determined by prelabeling certain of the matrix-synthesizing cells with [3H]leucine and [3”S]SO; (10 $X/ml each) for 3 days prior to harvesting the matrix. The various treatments were performed and then the amount of radioactive material still associated with the culture plate was determined by solubilization in SDS-sample buffer. The ratio refers to the amount (%) of [3H]leucine/[35S]SO; macromolecular material associated with the matrix.

ence of the antibody. Furthermore, the combined treatments of DTT and anti-TGF-/3 antibody of the TGF-j% synthesized matrix failed to restore growth of the RBE cells, indicating that the TGF-/3 matrix contained a nonreducible growth inhibitory activity. The earlier reversal of the growth inhibition after DTT or antibody

0

5

Time

10

15

(days)

FIG. 2. The kinetics of increase in cell number of RBE cells cultured on ECM obtained from RBE cells grown in the absence (0) or presence (0) of TGF-P. Some of the TGF-P-treated matrix was also briefly treated with DTT before addition of new RBE cells (m). At various times, parallel cultures were trypsinized and the cells were collected for cell number determination. Data points are the mean of analyses on triplicate cultures.

treatment suggests that another reduction-insensitive component of the ECM may be in part responsible for the growth regulation. The proliferative rates of the RBE cells on the untreated matrix and during reversal phases on either the TGF-P or DTT-treated TGF-6 matrices are similar and logarithmic. The decrease or plateauing of the cell number observed at late time periods with the TGF-fi matrices may be due in part to peeling of the matrix and subsequent cell loss upon prolonged incubation. Additionally, there was no difference in initial plating efficiencies of the RBE cells on the various matrices. Furthermore, several (five) independent repetitions of these studies all yielded similar results, although time prior to growth reversal was somewhat variable (6 days * 1.5 days for DTT-treated TGF-P matrix, and 8 days -+ 2 days for TGF-/3 matrix), although growth on the DTT-treated matrix always reversed more quickly than on treated ECM. A morphological difference was observed between RBE cells grown on control ECM (Fig. 1C) and RBE cells cultured on TGF-P matrices with or without DTT treatment (Figs. 1D and lC, respectively). The cell shape changes were different from those observed with TGF-P treatment, with the TGF-P matrix cells exhibiting a stellate form rather than the “pancake” morphology.

TGF-@-MEDIATED

GROWTH

TABLE

REGULATION

131

2

Analysis of 35S-Labeled Glycosaminoglycans in RBE Cell Compartments after Treatment with or without TGF-P

Cell compartments” Medium Medium Matrix Matrix Cell Cell

Exposure TGF-0 + + +

to

Heparan

sulfateb (So) 12 9 76 66 31 44

Chondroitin sulfate and dermatan sulfate (%I 88 91 24 34 68 56

Distribution among cellular compartments” (%j 78 73 19 23 3 4

a The RBE cells were labeled with [35§]sulfate for 48 h in the presence or absence of TGF-6, and the GAGS were isolated from the different cellular locations as described in the text. * Percentage represents the amount of total radioactivity associated with the separated GAGS by ion exchange cbromatograpby and analysis by treatment with various degradative enzymes. The radioactivity is based on cpmlyg of protein determined from the ceil fraction. The standard error is less than 10% of the reported value. ’ The percentage of total radioactivity incorporated into macromolecular material found associated with the various cellular fractions. The distribution percentage was the average of two independent studies and the standard error was less than 10%.

Several studies were initiated to examine the potential mechanism of TGF-/3 modulation of ECM-mediated cellular proliferation. The incorporation of various radioactive metabolic precursors into RBE cells and matrix with or without TGF-/3 treatment was quantitated. A two- to threefold increase in the accumulation of all of the precursors was observed in the matrix of TGF-Ptreated RBE cells compared to that of untreated cells. The various cellular radioactive extracts were then analyzed by SDS-PAGE followed by fluorography. No quantitative or qualitative differences in radioactively labeled cellular components were found between TGF&treated or untreated matrix or cellular fractions using the different metabolic labels based on equal amounts of macromolecular radioactivity loaded per lane (data not shown). Additionally, TGF-P-treated matrices were treated with various degradative enzymes. Exposure of the matrices to either Streptomyces hyaluronidase or chondroitinase ABC has little effect on the inhibitory activity of the matrix, whereas trypsin or heparitinase treatment appeared to cause a partial reversal of the inhibition. However, whether the effects of the enzymes were acting directly or indirectly on degradation of the potential inhibitor(s) or matrix was not determined. The composition of [35S]sulfate-labeled GAGS from the various cellular fractions of RBE cells treated or untreated with TGF-/3 was analyzed by ion exchange chromatography. Treatment of the RBE cells with TGF-fi did not affect the overall distribution of the GAGS among the various cell compartments (medium, matrix, and cell-associated, Table 2). The composition of GAGS released into the medium was not significantly affected by TGF-0 exposure to the cells. However, TGF-p treatment resulted in a decrease in the amount of heparan sulfate (HS) found in the matrix fraction, whereas an increase in the quantity of HS was noted in the cell-associated fraction. The amount of chondroitin

sulfatet’dermatan sulfate (CSIDS) was inversely proportional to the quantity of HS. Furthermore, TGF-P treatment was shown not to affec ize of the GAGS synthesized by the RBE cells (HS, - 33,000 I 6,000 kDa; CS/DS, M, - 40,000 + 8,000 - 18’,000 + 3,000 kDa). The ratio of N- to Q- sulfat sidws of 35S-labeled HS was estimated to be 41 + for N-sulfation and 59 +- 4% for O-sulfation regar of TGF-6 treatment. Additionally, no major qua ative alterations were observed in the elution profiles of the diamination fragments of the various Bio Gel PlO columns Glycosaminoglycans, in general, a have been shown to be altered by o to modulate growth regulation. HS has been o to inhibit cellular proliferation of a number o although it has been reported transformed fibroblasts [32-35] sis of the different structural for dependent on the growth status o but not proliferating smooth muscle cehs express a growth inhibitory HS [36]. Recently, ml&ear levels and possibly structures of HS have been reported to be altered by changes in growth conditions of rat bepatocytes [lS, 37j. Alternatively, Yamaguchi and cently shown that overexpressio tic CS/DS containing proteoglycan, in transfected fibroblasts results in decreased cellular pro~~fera!tion, as evidenced by a lower saturation density. Tbe expression of decorin was shown to be increased by the treatment of the cells with TGF-0 [39]. Ad~t~~~all~~ the cells were observed to increase their surface area (size) in direct relation to the amounts of decorin expressed. However, the increased expression of decorin appeared to mimic he R contact inhibition of cell growth. In csnt was growth inhibition on the TGF-P-treated

NEWTON

132

served in the absence of cell contact, suggesting that independent mechanisms may be utilized. Recently, TGF-P was suggested to induce collagen deposition which was directly implicated to play a role in growth regulation of NRK cells [40]. Regardless of the precise biochemical nature of the nonreducible growth inhibitor(s) these results suggest that alterations in ECM components may function as growth regulatory agents in addition to their other biological properties. The studies were supported by grants from the John S. Dunn Foundation, The Preuss Foundation, The University Cancer Foundation, and by NIH Grant ROl CA42729. The scanning electron microscopy was generously performed by Dr. G. Nicolson and supported by NC1 Core Grant P30-CA16672. The authors also thank Mary Sirrieh for her assistance in the preparation of this manuscript and Dr. Paula Belloni for her help in isolation and characterization of the endothelial cells.

ET AL. 15. 16. 17.

Ishihara, 467-476.

19. 20.

Wieser, R. J., and Oesch, F. (1986) J. Cell Biol. 103, 361-367. Matuoka, K., Namba, M., and Mitsui, Y. (1987) J. Cell Biol. 104, 1105. Gospodarowicz, D., Delgado, D., and Vlodovsky, I. (1980) Proc. Natl. Acad. Sci. USA 77,4094-4098. Kinsella, M. G., and Wight, T. N. (1986) J. Cell Biol. 102,679687.

21. 22. 23.

1. 2. 3. 4.

5.

6. 7.

8. 9. 10. 11.

12. 13. 14.

Sporn, M. B., Roberts, A. B., Wakefield, L. M., and de Crombugghe, B. (1987) J. Cell Biol. 105,1039-1045. Sporn, M. B., Roberts, A. B., Wakefield, L. M., and Assorian, R. K. (1986) Science 233,532-534. Roberts, A. B., Thompson, N. L., Heine, U., Flanders, C., and Sporn, M. B. (1988) &it. J. Cancer 57,594-600. Wrann, M., Bodmer, S., de Martin, R., Siepl, C., Hofer-Warbinek, R., Frei, K., Hofer, E., and Fontana, A. (1987) EMBO J. 6,1633-1636. Kehrl, J. H., Wakefield, L. M., Roberts, A. B., Jakowlew, S., Alvarez-Mon, M., Rerynch, R., Sporn, M. B., and Fauci, A. S. (1986) J. Exp. Med. 163, 1037-1050. Roberts, A. B., Anzano, M. A., Lamb, L. C., Smith, J. M., and Sporn, M. B. (1981) Proc. N&Z. Acad. Sci. USA 78,5339-5343. Holley, R. W., Bohlen, P., Fava, R., Boldwin, J. H., Kleeman, G., and Armour, R. (1980) Proc. Natl. Acad. Sci. USA 77, 59895992. Cheifetz, S., Weatherbee, J. A., Tsang, M. L. S., Anderson, J. K., Mole, J. E., Lucas, R., and Massague, J. (1987) Cell 48,409-415. Wakefield, L. M., Smith, D. M., Masui, T., Harris, C. C., and Sporn, M. B. (1987) J. Cell Biol. 105,965-975. Ignotz, R., and Massague, J. (1986) J. Biol. Chem. 261, 64436446. Keski-Oja, J., Raghow, R., Sawdey, M., Laskutoff, D. J., Pastlethwaite, A. E., Kang, A. H., and Moses, H. L. (1988) J. Biol. Chen. 263,3111-3115. Hay, E. D. (1981) J. Cell Biol. 91, 205-223. Ekblom, P., Vestweber, D., and Kemler, R. (1986) Annu. Rev. Cell Biol. 2, 27-47. Bernfield,

Received

March

M., and Banerjee, 5, 1990

S. D. (1982) Dev. Biol. 90,291-305.

25. 26. 27. 28.

M., and Conrad,

H. E. (1989) J. Cell. Physiol.

138,

18.

24.

REFERENCES

Gospodarowicz, D., Feiroro, N., Schweiguer, L., and Neufeld, G. (1987) Endocrol. Rev. 8,95-114. Castellot, J. J., Choay, J., Lormeau, J. C., Petilou, M., Sache, E., and Karnovsky, M. J. (1986) J. Cell Biol. 102,1979-1984. Wright, T. C., Castellot, J. J., Petilou, M., Lormeau, J. C., Choay, J., and Karnovsky, M. J. (1989) J. Biol. Chem. 264, 1534-1542.

Takehara, K., Le Roy, E. C., and Grotendorst, G. R. (1987) Cell 49,415-422. Baird, A., and Durkin, T. (1986) Biochem. Biophys. Res. Commun. 138,476-482. Belloni, P. N., Carney, D. H., and Nicolson, G. L. (1990) Microvascular Res., in press. Jaffe, E. A., Hoger, L. W., and Nachman, R. L. (1973) J. Clin. Invest. 52,2757-2764. Laemmli, U. K. (1970) Nature (London) 227, 680-683. Steck, P. A., Cheong, P. H., Nakajima, M., Yung, W. K. A., Moser, R. P., and Nicolson, G. L. (1987) Biochemistry 26,10201027.

29. Saito, H., Yamagata,

T., and Suzuki,

S. (1968) J. Biol. Chem.

243,1536-1542. 30.

Shively,

J. E., and Conrad, H. E. (1976) Biochemistry

15,3932-

3942. 31. 32. 33. 34. 35. 36. 37.

Winterbounre, D. J., andMora, P. T. (1981) J. Biol. Chem. 256, 4310-4320. Rutka, J. T., Giblin, J. R., Apodoca, G., DeArmond, S. J., Stein, R., and Rosenblum, M. L. (1987) Cancer Res. 47, 3515-3522. Reilly, C. F., Fritze, L. S., and Rosenberg, R. D. (1986) J. Cell Physiol. 129, 11-19. Matuoka, K., Mitsui, Y., Murota, S., and Namba, M. (1985) Cell Biol. Znt. Rep. 9,815-824. Ratner, N., Bunge, R. P., and Glaser, L. (1985) J. Cell Biol. 101, 744-754. Fritze, L. M. S., Reilly, C. F., and Rosenburg, R. D. (1988) J. Cell Biol. 100,1041-1049. Fedarko,

N. S., and Conrad, H. E. (1986) J. Cell Biol. 102,587-

599. 38. 39. 40.

Yamaguchi, Y., and Ruoslahti, E. (1988) Nature (London) 336, 244-246. Basols, A., and Massague, J. (1988) J. Biol. Chem. 263, 30393045. Nugent, M. A., and Newman, l&060-18,067.

M. J. (1989) J. Biol. Chem. 264,

Growth regulatory activities of endothelial extracellular matrix: mediation by transforming growth factor-beta.

The potential of transforming growth factor-beta (TGF-beta) to modulate the growth of endothelial cells via alterations in the cell's extracellular ma...
3MB Sizes 0 Downloads 0 Views