vitro angiogenesis model system in which mi- crovascularfragments and myofibroblasts (Mfs) isolatedfrom rat epididymal lipid tissues were grown in co-culture.
American Journal of Pathology,
Vol. 142, No. 4, April 1993 Copyright X) American Society for Investigative Pathology
Platelet-Derived Growth Factor Indirectly Stimulates Angiogenesis in Vitro
Noboru Sato,*t Julie G. Beitz,* Junji Kato,* Mikio Yamamoto,t Jeffrey W. Clark,* Paul Calabresi,* and A. Raymond Frackelton, Jr.* From the Department of Medicine, * Roger Williams Cancer Center and Brown University, Providence, Rhode Island;
and the Department of Biochemistry,t National Defense Medical College, Tokorozawa, Saitama, Japan
We evaluated the effect ofplatelet-derived growth factor (PDGF) on capiUaryformation using an in vitro angiogenesis model system in which microvascularfragments and myofibroblasts (Mfs) isolated from rat epididymal lipid tissues were grown in co-culture. In this system Mfs induce capiUary formation by producing an endothelial ceUl growth factor and by secreting extracelular matrix components that cause endothelial ceUls to form cordlike structures. Addition of PDGF enhances in vitro capiUary growth. Although some recently described microvascular endothelial ceUs display PDGF receptors and respond to PDGF, we found no evidence for direct PDGF action on the rat epididymal microvascular endothelial ceUls. Rather, we found that PDGF increased the proliferation of Mfs, as well as the production of Mf-derived endothelial ceUgrowth factor and matrix coUagen type L Our results suggest that even in cases where the microvasculature lacks PDGF receptors, PDGF may accelerate capiUary formation by activating connective tissue cells in the vicinity of endothelial ceUls. (Am J Pathol 1993, 142:1119-1130)
Platelet-derived growth factor (PDGF), the major growth factor released from stimulated platelets at the site of vascular insults, is mitogenic for most mesenchymally derived cells, including fibroblasts, smooth muscle cells, and osteoblasts, and for some neoplastic cells.1-3 Not surprisingly, PDGF has been implicated in several physiological and pathological processes, including wound repair, embryogenesis, atherosclerosis, and tumor growth. 1-3 Whereas active
formation of new capillary vessels is frequently observed during these events (for review, see ref. 4), the role of PDGF in neovascularization has not been well characterized. Although previous studies have reported that endothelial cells lack PDGF responsiveness and PDGF receptors, those studies dealt primarily with macrovascular endothelial cells.5- 11 Recently, however, functional PDGF receptors have been demonstrated on microvascular endothelial cells derived from human adipose tissues, human carcinoid tumors, rat brain, and rat liver.12-17 Interestingly, coexpression of PDGF-B chain and PDGFreceptor messenger (m)RNA has been demonstrated in hyperproliferating microvascular endothelial cells growing near malignant glioma cells secreting PDGF.17 Thus, there is growing evidence suggesting roles for PDGF as an autocrine or paracrine modulator of microvascular cells involved in neovascularization. Observations of neovascular events in vitro and in vivo demonstrate that the microvascular endothelium participates in a cascade of events beginning with the degradation of endothelial cell basement membranes, followed by the migration of endothelial cells toward an angiogenic stimulus, cell proliferation, sprout development, and the eventual formation of tubular structures.18 The entire process is believed to be mediated by a host of soluble endothelial mitogens on the one hand and insoluble extracellular matrix components on the other, all of which direct the formation and organization of new capillary networks. 4,19,20 Moreover, the biological activities of certain heparin-binding endothelial mitogens, such as basic fibroblast growth factor (bFGF) may be modulated by their ability to bind to low-affinity heparin sulfate proteoglycans present in the extracellular matrix of target cells. Thus, bFGF may remain stored within the matrix in a stable form, until its release is signaled by such events as basement membrane injury (in wounding) or
enzymatic degradation (by tumor proteases).2122
Supported in part by a grant from Farmitalia Carlo Erba. Accepted for publication September 22, 1992. Address reprint requests to Dr. Noboru Sato, Tsukuba Research Laboratories, Nippon Glaxo LTD, 43 Wadai, Tsukuba-City 300-42, Japan.
1119
1120
Sato et al
AJP April 1993, Vol. 142, No. 4
We have been evaluating an in vitro model of capillary network formation in which microvessel fragments from rat epididymal adipose tissue are cultured with rat epididymal myofibroblasts (Mfs). Under the influence of an endothelial cell growth factor and extracellular matrix components secreted by Mfs, endothelial cells proliferate, elongate, and form tubelike structures that subsequently join into capillary networks, the morphology of which closely resembles that of capillary blood vessels in vivo.23 Here we report that in this co-culture system, PDGF-BB enhances capillary growth. Further, we examine mechanisms by which PDGF may be acting in this system, exploring potential sequelae of PDGF action on both the Mf and endothelial cell components of this in vitro model.
Materials and Methods Materials Recombinant human platelet-derived growth factor (PDGF-BB) was obtained from Amgen Biologicals (Thousand Oaks, CA), and [1251]PDGF-BB was purchased from Collaborative Research, Inc. (Bedford, MA). Neutralizing antibody to bFGF was purchased from R&D Systems, Inc. (Minneapolis, MN).
Cell Cultures Bovine adrenal capillary endothelial cells (BCECs) were isolated in our laboratory as described previously24 or were the gift of Dr. J. Folkman (The Children's Hospital, Boston, MA). Bovine aortic endothelial cells (BAECs) were prepared as described.24 These cells were maintained in complete Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 100 pg/ml of endothelial mitogen purchased from Biomedical Technologies, Inc. (Stoughton, MA) and were used between passages 5 to 10. Mfs and microvascular fragments were prepared from rat epididymal fat pads by collagenase digestion and subsequent sequential fractionations using filter meshes and Percoll as described.23
Cell Proliferation Assay The effect of PDGF on the growth of Mfs or BCECs was examined as follows. Using 24-well tissue culture plates, rat Mfs or BCECs were plated in triplicate at densities of 5 x 103 or 104 cells per well, with 0.5 ml of medium 199 or DMEM supplemented
with 2.5% FBS, respectively. After 3 days, the medium was replaced with fresh medium containing 2.5% FBS and 0 to 10 ng/ml of recombinant PDGFBB. The medium and growth factor were replenished 3 days later. Cells were harvested by trypsinization at 0, 3, and 7 days after PDGF addition, and enumerated using a Coulter counter (Coulter Electronics, Inc., Hialeah, FL). Data were analyzed by ANOVA using the Statview statistical package (Abacus Concepts, Inc., Berkeley, CA).
In Vitro Capillary Growth Assay An in vitro capillary growth promotion assay was performed as previously described.25 In brief, microvascular fragments and Mfs isolated from rat epididymal fat pads were seeded together at densities of 7 x 103/well and 5 x 104/well in 12-well tissue culture plates in medium 199 supplemented with 10% FBS. Mfs grew actively and formed a monolayer, after which endothelial cells began to sprout from the microvessel fragments and link together to form long, anastomosing cords extending over the Mf monolayer. These cords could be stained with anti-factor VIII antibody, indicating that they consisted of endothelial cells.23 Electron microscopy of these cords revealed tubular structures containing lumina that were composed of three to five endothelial cells joined by intercellular junctions.23 For these studies, after Mfs were nearly confluent, the medium was replaced with 1.5 ml of fresh medium 199 supplemented with 2.5% FBS and containing test materials. The sprouting of endothelial cell cords from the microvessel fragments was photographically recorded (magnification x40), and the density of cords per defined area (1.7 x 1.1 mm) was determined as described.25 Five defined areas in every duplicate culture were photographed at random (n = 10). The cellular cord densities were determined by superimposing a transparent rectangular grid (6.9-mm spacing) on the light micrographs (final magnification x90), and then counting the total number of intersections between cellular cords and grids within the defined area. To determine if PDGF directly affected rat microvascular endothelial cells, we modified the in vitro capillary growth assay as follows: confluent cultures of Mfs were air-dried for 1 hour at room temperature, thereby killing Mfs, leaving Mf-derived extracellular matrix and fixed cellular material behind. Rat microvascular fragments were plated on the Mf-derived matrix, and capillary network formation was observed under several different culture
Angiogenic Stimulation by PDGF 1121 AJP Aprl 1993, Vol. 142, No. 4
conditions, including: 1) the usual culture medium supplemented with 2.5% FBS; 2) the usual medium plus medium conditioned by Mfs; 3) the usual medium plus medium conditioned by Mfs grown in the presence of 10 ng/ml PDGF; 4) the usual medium supplemented with 10 or 30 ng/ml PDGF; or 5) the usual medium supplemented with 10 ng/ml PDGF plus medium conditioned by Mfs. Data were analyzed by ANOVA using the Statview statistical package.
PDGF Binding Assay Mfs were seeded into 24-well tissue culture plates (2 x 104/well) and grown until confluent in medium 199 containing 2.5% FBS. [1251]PDGF binding was carried out using a radioreceptor assay as described.812 Briefly, cell monolayers were washed once with 1 ml of ice-cold phosphate-buffered saline, and then incubated with 250 pi of medium 199 containing 0.1% bovine serum albumin and varying concentrations of [1251]PDGF-BB (0 to 10 ng/ml; 0 to 480,000 cpm) for 4 hours at 4 C. Cell monolayers were then washed three times with ice-cold phosphate-buffered saline to remove unbound [1 251]PDGF, and remaining cell-associated radioactivity was extracted with 1% Triton X-100 and quantitated by y counting. Nonspecific binding was determined by inhibiting specific binding using a 100-fold excess of nonradiolabeled recombinant PDGF-BB. Cell numbers were determined after trypsinization of parallel cultures.
Immunoprecipitation of PDGF-lnduced Tyrosine-Phosphorylated Proteins Confluent monolayers of rat Mfs or BCECs seeded in 6-well tissue culture plates were rinsed twice with 1 ml of phosphate-free medium, incubated with 1 ml of phosphate-free DMEM containing [32P]orthophosphate (1 mCi/well) for 3 hours at 37 C and then exposed to recombinant PDGF-BB (10 ng/well) for 10 minutes. The medium was removed, and the cells were extracted at 0 C in 0.5 ml of 1% Triton X-100 buffer containing phosphatase, proteinase, and kinase inhibitors.26 27 Insoluble debris was removed by centrifugation (8000g for 15 minutes at 4 C) and phosphotyrosine-containing cellular proteins were isolated from the supernatant extract by microbatch immunoaffinity chromatography using the 1 G2-antiphosphotyrosine monoclonal antibody and specifically eluting with the hapten, phenyl phosphate.26 27 In some instances, glycosylated
PDGF receptors were further purified by affinity chromatography using the lectin, wheat germ agglutinin, eluting with 0.3 mol/L N-acetylglucosamine as described.28-29 The purified proteins were further resolved by sodium dodecyl sulfate (SDS)polyacrylamide gel (7.5%) electrophoresis under reducing conditions30 and visualized by autoradiography using preflashed XAR-5 film (Kodak) and a Lightning Plus (Dupont) intensifying screen, exposing at -70 C for 20 hours.
Endothelial Cell Growth Factor Production by Mfs The effect of PDGF on the production of Mf-derived endothelial cell growth factor was examined as follows: Mfs freshly prepared from rat epididymal tissue were cultured in 6-well tissue culture plates in 3 ml of medium 199 supplemented with 10% FBS. After the Mfs had reached confluency, the cells were cultured in 3 ml of medium 199 supplemented with 2.5% FBS in the presence or absence of 30 ng/ml of recombinant PDGF-BB for 5 days, after which the conditioned medium of the cultures was collected. The activity of endothelial cell growth factor in the Mf-conditioned medium was examined using BAEC proliferation as an endpoint. BAECs at passage 10 were seeded in 24-well tissue culture plates at a density of 104 cells per well and grown in DMEM supplemented with 10% FBS. The medium was replaced the following day with 0.5 ml of the usual medium now containing Mf-derived conditioned medium (25%). After 5 days, the cells were harvested by trypsinization and enumerated using a Coulter counter. In some instances, BAEC proliferation was examined in the presence of Mfconditioned medium containing 30 pg/ml of antibFGF neutralizing antibody.
Collagen Synthesis by Mfs Collagen synthesis was measured using the method of Tsuruoka et al.31 Mfs were grown in 6-well tissue culture plates in medium 199 containing 10% FBS. After Mfs had reached confluency, they were washed with phosphate-buffered saline, then cultured for 3 days in 2 ml of DMEM supplemented with 2.5% FBS in the presence or absence of recombinant PDGF-BB (0, 1, or 10 ng/ml). After the medium was aspirated, cell monolayers were washed three times with serum-free DMEM and incubated for 24 hours in 1 ml of serum-free DMEM containing 50 pg/mI ascorbic acid (a co-factor for
1122
Sato et al
AJP April 1993, Vol. 142, No. 4
the hydroxylation of proline residues in collagen),32 50 pg/ml ,B-aminopropionitrile (an inhibitor of collagen cross-linking)33 and 5 pCi L[2,3-3H]proline (ICN Biologicals, Costa Mesa, CA). Next, 0.1 ml of 2 N NaOH was added to each well and proteins in the medium, extracellular matrix, and cells were collected by scraping with a rubber policeman and by pipetting. The proteins were precipitated by adding 0.2 ml of 50% trichloroacetic acid (w/v)/5% tannic acid (w/v) per well on ice. Precipitable proteins were washed three times with 10% trichloroacetic acid (w/v)/1% tannic acid (w/v), once with cold acetone, and then dissolved in 1 ml of 0.05 N NaOH. The solution was divided into two equal portions and added to an equal volume of 0.2 mol/L Tris-HCI buffer, pH 7.5, containing 100 mmol/L CaCI2 and 2 mmol/L N-ethylmaleimide, and the mixture was incubated for 3 hours at 37 C with or without 11 units of purified bacterial collagenase type Ill (Advance Biofactures Co., Lynbrook, NY). After incubation, trichloroacetic acid was added to the mixture to a final concentration of 10% (w/v) to precipitate unhydrolyzed proteins. The radioactivity of each sample was quantitated by y counting. Mf-induced collagen synthesis was expressed as the radioactivity present in collagenase-untreated proteins minus the radioactivity present in collagenase-treated proteins. To characterize the type of collagen produced by PDGF-stimulated Mfs, 3H-labeled TCAprecipitable proteins isolated as described above were resolved by SDS-polyacrylamide (7.5%) gel electrophoresis under reducing conditions.30
probe (1 x 106 cpm/ml) had been added. After hybridization, the filter was washed twice for 15 minutes each in 2x SSC and 0.1% SDS at room temperature, twice for 15 minutes each in 2x SSC, and 0.1% SDS at 42 C, and once for 30 minutes in 0.2x SSC and 0.1% SDS at 65 C. The filter was autoradiographed using Kodak XAR-5 film (Eastman Kodak, Rochester, NY) at -70 C with intensifying screens for 12 hours. For sequential hybridization of the filter, the prior probe was removed by treating the membrane in 50% formamide and 6x SSC at 65 C for 30 minutes. A rat f3-actin oligonucleotide probe (Oncogene Science, Uniondale, NY) was used as a loading control. The oligomer was 5' end labeled with [y-32P]ATP (3,000 Ci/mmol, Amersham) with T4 polynucleotide kinase and purified on a NACS prepac column (Bethesda Research Laboratories Inc., Gaithersburg, MD). The filter was prehybridized for 2 hours at 42 C in 30% formamide, 6x SSC, 1Ox Denhardt's, 0.5% SDS, and 100 pg/ml of denatured, sonicated salmon testis DNA. Hybridization was performed for 16 hours at 42 C in 30% formamide, 6x SSC, 0.5% SDS, and 32P-labeled cDNA probe (1 x 106 cpm/ml). After hybridization, the filter was washed twice for 15 minutes each in 6x SSC and 0.1% SDS at room temperature, twice for 15 minutes each in 2x SSC and 0.1% SDS at 42 C, and once for 30 minutes in 2x SSC and 0.1% SDS at 50 C. The filter was autoradiographed as
before.
Results Mf-Derived Collagen mRNA Isolation and Northern Hybridization
Stimulation of In Vitro Capillary Growth by PDGF
Total cellular RNA was isolated from cultured cells using the guanidinium thiocyanate/cesium chloride method. Ten pg of RNA was electrophoresed on a 1.1% formaldehyde denaturing agarose gel and transferred onto a Hybond-N-nylon membrane (Amersham). An 800-kb EcoRI restriction fragment from a rat type 1 collagen complementary (c)DNA clone provided by Y. Yamada was labeled with [a_32p]dCTP (3,000 Ci/mmol, Amersham) by random hexamer priming. The filter was prehybridized for 2 hours at 42 C in 50% formamide, 5x standard saline citrate (SSC), 5x Denhardt's (1 x being 0.02% Ficoll, polyvinylpyrrolidone, and bovine serum albumin fraction V), 0.5% SDS, and 100 pg/mI of denatured, sonicated salmon testis DNA. Hybridization was performed for 16 hours at 42 C in a fresh mixture of the same solution to which 32P-labeled cDNA
The development of new capillary networks from the microvessel fragments in our co-culture system is dependent upon the presence of an intact Mf monolayer. We have reported that Mfs secrete an endothelial cell growth factor into the medium that is capable of supporting the proliferation of endothelial cells from a variety of sources, including human veins, arteries, and bovine capillaries, and induces the migration of bovine capillary endothelial cells in vitro.21 This factor is a 40-kd protein that does not bind to heparin, has a pl of 6.5 to 7.0, aggregates in the presence of Mg'2, and is sensitive to reducing agents.34 To test if PDGF affected capillary network formation in our model system, 3-day-old cocultures of Mfs and microvascular fragments were treated with PDGF-BB for 7 days. Exogenous PDGF-BB (5 or 10 ng/ml) stimulated capillary tubule
Angiogenic Stimulation by PDGF
1123
AJP Apil 1993, Vol. 142, No. 4
formation by nearly 4-fold (P < 0.001 at 5 ng/ml) as compared to control cultures that did not receive PDGF-BB (Figures 1 and 2). Capillary formation was increased using as little as 1 ng/ml (P < 0.001) of PDGF-BB and was nearly maximal at 5 ng/ml. PDGF might be functioning by stimulating Mfs, endothelial cells, pericytes, or any combination of these cells. To examine these possible mechanisms, we first took advantage of the fact that fixed Mfs alone can support capillary development.23 Rat microvessel fragments were plated on an Mfderived extracellular matrix (after Mfs had been killed by air-drying) and cultivated under several different culture conditions (see Methods). After 6 days, capillary network formation was quantitated. As shown in Figure 3, when microvascular fragments were cultured in the usual culture medium alone and in the absence of a viable Mf cell monolayer, no capillary growth was observed by day 6. The addition of 10 or 30 ng/ml of recombinant PDGF-BB did not stimulate new capillary formation under these conditions, indicating that PDGF-BB by itself was not capable of supporting the proliferation of rat microvascular endothelial cells. On the other hand, the addition of Mf-conditioned medium resulted in marked capillary network formation (P < 0.01), suggesting that Mfs were capable of secreting an endothelial cell growth factor into the medium which, in turn, stimulated capillary tubules to sprout from the microvessel fragments. Addition of PDGF to this conditioned media caused no increment in network formation, whereas conditioned media from Mfs grown in the presence of exogenous PDGF supported a 1.9-fold increase (P < 0.01) in capillary growth. We conclude, then, that
120 (0
0
ca
co
801. I.
E
401.
T hL1 MMMMMIMM
MMIMMEMZ WOMMMZZZ ISZZZ M
PDGF enhances capillary formation, at least in part, by increasing the amount of a soluble, Mf-derived factor that seems to act directly on the endothelial cells.
Mfs Possess Functional PDGF Receptors Next, we sought further evidence for the direct action of PDGF-BB on Mfs. We asked first whether Mfs express high-affinity cell-surface receptors for PDGF. Scatchard analysis of 1251-labeled PDGF-BB binding showed Mfs to display a single class of PDGF binding with a 26 pmol/L kd and 6 x 104 sites/cell (Figure 4). An important consequence of PDGF binding to its receptor is the activation of the PDGF-receptor-associated protein-tyrosine kinase. 28,35 The receptor must then phosphorylate itself to be able to interact with and tyrosine-phosphorylate critical cellular proteins believed to be second messengers in the signal transduction pathway.36-38 We asked, then, whether PDGF affected any of these critical tyrosine phosphorylations in Mfs. To address this question, Mfs were labeled with 32p,, exposed to recombinant PDGF-BB for 10 minutes, and then extracted with 1% Triton X-100 in a buffer containing kinase, phosphatase, and proteinase inhibitors. Phosphotyrosine-containing glycoproteins were then purified from this extract by sequential antiphosphotyrosine and wheat-germ lectin affinity chromatography (see Methods). In Mfs, PDGF-BB stimulated the phosphorylation of several proteins that could be isolated by anti-phosphotyrosine antibody, including a prominent 180-kd protein which, based on its molecular weight and ability to bind to the lectin, wheat germ agglutinin, is likely to be the PDGF receptor (Figure 5). Recent studies with anti-PDGF-receptor immunoblotting have confirmed this identity (L. Lu and A.R. Frackelton, unpublished results). Inasmuch as Mfs possessed PDGF receptors that seem to properly transduce the PDGF signal, we asked whether PDGF was mitogenic for Mfs. The addition of as little as 1 ng/ml PDGF-BB caused a marked increase in Mf proliferation (P < 0.01), whereas 10 ng/ml PDGF-BB increased Mf proliferation several-fold (P < 0.001, Figure 6). PDGF at 30 ng/ml was no more potent than PDGF at 10 ng/ml (data not shown).
0
1 0 5 10 Concentration of PDGF (ng/ml)
Figure 1. Stimulation of in vitro capillary development by recombinant PDGF-BB. Rat Mfs and microvascular fragments were cocultured in the presence or absence of PDGF-BB (O to 10 ng/ml). Capillary network formation was quantitated 5 days after PDGF-BB addition as described in Methods. Values are mean + SD (n 10). =
PDGF Effects on Mf Production of an Endothelial Cell Growth Factor We have demonstrated that an Mf-derived endothelial cell growth factor secreted into the culture me-
1124
Sato et al
AJP April 1993, Vol. 142, No. 4
Figure 2. Effect of recombinatnt PDGF-BB on in vitro capillary formation. Light micrographs show the new capillary cords that have sprouted from the microvesselfragments after 5 days of culture in the absence (A) or presence (B) of PDGF-BB (10 ng/ml). Magnification x100.
dium plays a critical role in the development of new capillary networks in the co-culture system.23 We asked, then, if PDGF affected Mf secretion of an endothelial growth factor. Conditioned medium was collected from Mf cell monolayers cultured in the presence or absence of 30 ng/ml of recombinant PDGF-BB for 5 days and added to cultures of BAECs at a final concentration of 25%. The proliferation of BAECs exposed for 5 days to Mfconditioned media was increased 1.4-fold over basal levels (P < 0.05), but 1.9-fold (P < 0.01) following exposure to conditioned medium from PDGF-treated Mfs (Figure 7). Addition of 30 pg/ml of anti-bFGF neutralizing antibody to the conditioned medium from PDGF-treated Mfs did not affect its endothelial cell growth promoting activity, although this antibody did block bFGF-stimulated proliferation of endothelial cells (Figure 7 and data not shown). Thus, the factor secreted from Mfs was not bFGF. It is also unlikely that the Mf-derived factor was PDGF: BCECs, which are stimulated to proliferate by Mf-conditioned medium,23 were not stimu-
lated to proliferate by 1 or 10 ng/ml PDGF-BB (data not shown) and failed to demonstrate any tyrosine phosphorylation in response to PDGF-BB (Figure 5). Similarly, BAEC proliferation was not stimulated by exogenous PDGF-BB (data not shown).
PDGF Effects on Collagen Synthesis by Mfs We have shown that Mfs produce extracellular matrix component(s) that contribute to the formation of tubular structures by proliferating endothelial cells in the co-culture system.23'39 This finding corroborates reports by other investigators that collagen(s) mediate certain changes in endothelial cell shape that eventually result in the formation of tubelike structures.4041 Therefore, we postulated that PDGF, in addition to increasing production of endothelial cell growth factors by Mfs, may affect collagen synthesis by Mfs, thereby facilitating capillary network formation in vitro. To determine the collagen types
Angiogenic Stimulation by PDGF 1125
AJP Apil 1993, Vol. 142, No. 4
0
c
=
E
?-V
(-)
Mf-CM
MIUPDGW
CM
PrGF
PXF
Mt-Um
(longlm) (3ongAnI +PDGF 0 nglml
Figure 3. Effect of recombinant PDGF-BB or Mf-conditionied medium on capillary netuork formation arising from rat microvessel fragments cultuired on Mf-derived extracellular matrix. Microvessel fragments were seeded on air-dried (killed) Mfs and cultuired itn the uisual culture medium (-), in the usual medium plus an equal volume of Mf-conzditioned medium (Mf-CM), in the usual mediuim pluis an equal volume of medium conditioned by Mfs grown in the presence of 10 ng/ml PDGF (Mf]PDGFI CM), in the uisual mediuim supplemented uith 10 or 30 ng/ml PDGF (PDGF 10 ng/ml, PDGF 30 ng/ ml), or in the usual medium supplemented with 10 ng/ml PDGF pluis an equal volume ofMf-conditioned mediuim (Mf-CM + PDGF 10 tgl ml). Capillary netu'ork formation u'as quantitated 5 days later as in Figure 1. Values are mean + SD (n = 10).
produced in Mf cultures, [3H]proline-labeled, acidprecipitable proteins isolated as described in Methods were analyzed by SDS-polyacrylamide (7.5%) gel electrophoresis under reducing conditions. As
0.20 C,
uz 0
c
0.10 -0 U-
CD
0
0.00
2 PDGF total (ng/ml) Figure 4. Scatchard analysis of J25I-labeled PDGF-BB binding to Mfs. Mfs were seeded into 24-well tissue culture plates and grown until confluent in medium 199 containing 2.50% FBS. [125IIPDGF bindinig u'as performed using a radioreceptor assay as described in Methods. Briefly, cell monolayers were uashed with ice-cold medium, incubated u'ith various concentrationts of /'25IIPDGFfor 4 hours at 4 C, and washed with ice-cold PBS to remove unbound f12'5]PDGF. Remaining cell-associated radioactivity uas extracted and quantitated by coulnting. Nonspecific binding, determined in the presence of a 100-fold excess of nonradiolabeled recombinant PDGF-BB, was
