Heparin amplifies platelet-derived growth factor (PDGF)- BB-induced PDGF alpha -receptor but not PDGF beta -receptor tyrosine phosphorylation in heparan sulfate-deficient cells. Effects on signal transduction and biological responses.

Platelet-derived growth factor (PDGF) induces mitogenic and migratory responses in a wide variety of cells, by activating specific receptor tyrosine kinases denoted the PDGF alpha- and beta-receptors. Different PDGF isoforms bind in a distinct manner to glycosaminoglycans, particularly heparan sulfate. In the present study, we show potentiation by exogenous heparin of PDGF-BB-induced PDGF alpha-receptor tyrosine phosphorylation in heparan sulfate-deficient Chinese hamster ovary (CHO) 677 cells. This effect was not seen for PDGF-AA treatment, and heparin lacked a potentiating effect on PDGF-BB stimulation of the PDGF beta-receptor. Heparin did not affect the affinity of PDGF-BB binding for the PDGF receptors on CHO 677 cells. The PDGF-BB-stimulated PDGF alpha-receptor phosphorylation was enhanced in a dose-dependent fashion by heparin at low concentration. The effect was modulated by 2-O- and 6-O-desulfation of the polysaccharide. Maximal induction of PDGF alpha-receptor tyrosine phosphorylation (6-fold) in CHO 677 cells was achieved by treatment with a heparin decasaccharide, but shorter oligosaccharides consisting of four or more monosaccharide units were also able to augment PDGF alpha-receptor phosphorylation, albeit at higher concentrations. Heparin potentiated PDGF-BB-induced activation of mitogen-activated protein kinase and protein kinase B (Akt) and allowed increased chemotaxis of the CHO 677 cells toward PDGF-BB. In conclusion, heparin modulates PDGF-BB-induced PDGF alpha-receptor phosphorylation and downstream signaling, with consequences for cellular responsiveness to the growth factor.

capillary endothelial cells and neurons (reviewed in Refs. 1 and 2). Classically, PDGF is a dimeric molecule consisting of disulfide-bonded A and B polypeptides that assemble into homo-and heterodimers; i.e. PDGF-AA, PDGF-BB, and PDGF-AB (3). PDGF transduces cellular responses by binding to two related protein tyrosine kinase receptors, the PDGF ␣and ␤-receptors. PDGF-AB and PDGF-BB bind to both PDGF ␣and ␤-receptors with similar affinity (4), in contrast to PDGF-AA, which binds only to the PDGF ␣-receptor (5)(6)(7). Recently, additional PDGF-related polypeptides were identified and denoted PDGF-C and -D. These novel isoforms do not appear to form heterodimers but exist as PDGF-CC and -DD (8 -10), which bind to PDGF ␣and ␤-receptors, respectively.
Upon binding of PDGF, the receptors dimerize, leading to autophosphorylation of tyrosine residues in trans between two receptor molecules in the dimer. The phosphorylated tyrosine residues, in combination with surrounding amino acid residues, form binding sites for signaling proteins equipped with Src homology (SH) 2 domains. The PDGF receptors are known to associate with members of the Src family of cytoplasmic tyrosine kinases, phospholipase C-␥, the regulatory p85 subunit of phosphoinositide 3-kinase, the adaptor Grb2, and the Src homology-containing phosphatase 2 (Shp-2) (Ref. 11 and reviewed in Refs. 12 and 13). Binding of SH2 domain proteins in turn leads to initiation of signaling cascades involving mitogen-activated protein kinase (MAPK) and protein kinase B (PKB/Akt) that provide survival signals for the cell.
The PDGF-A chain appears as two variants, a longer form (PDGF-A L ) and a shorter form (PDGF-A S ), that are generated through alternative splicing of exons 6 and 7 of the PDGF-A gene. The PDGF-B chain, on the other hand, is proteolytically processed into a shorter form and a longer form. The PDGF-A S and the short form of PDGF-B are effectively secreted into the medium, whereas PDGF-A L and the long form of PDGF-B are retained at the cell surface (14,15). The retention of the long PDGF polypeptides is due at least in part to binding to glycosaminoglycans, particularly those of heparan sulfate proteoglycans (16,17). Heparan sulfate proteoglycans are expressed on most cell types but are also secreted and deposited in the extracellular matrix. The binding of proteins to HS and other glycosaminoglycans is largely electrostatic in nature and involves the negatively charged carboxyl and sulfate groups in the HS chains and basic amino acid residues in the protein.  (18). Notably, the sulfation patterns of HS are tissue-specific, developmentally regulated, and apparently designed to accommodate selective interactions with a spectrum of proteins (19). In the present study, we have investigated effects of heparin on PDGF-BB-stimulated PDGF receptor activation and determined the influence of oligosaccharide chain length and O-sulfation on receptor activation and downstream signaling events.
Scatchard Analysis-Confluent CHO 677 cells were washed with PBS-B/BSA (PBS plus 0.077 mM CaCl 2 and 0.083 mM MgSO 4 supplemented with 1% BSA) and incubated for 1 h on ice with increasing amounts of unlabeled PDGF-BB (24,300 Da; Peprotech) in the presence or absence of 100 ng/ml heparin. The cells were then incubated for 1 h with 1 ng/ml 125 I-PDGF-BB (20,000 cpm/ng; Amersham Biosciences). The cells were washed three times with PBS-B/BSA and then lysed for 15 min on ice in 20 mM Tris-HCl, pH 7.5, 1% Triton X-100, and 10% glycerol. Cell-associated 125 I was estimated using a gamma counter.
Flow Cytometry-For detection of cell surface HS and chondroitin sulfate (CS), CHO KI and CHO 677 cells were suspended at a concentration of 1 ϫ 10 6 cells/ml in RPMI 1640 medium and 10% fetal calf serum containing 10 g/ml anti-HS antibody 10E4 (21) or an anti-CS antibody (catalogue number C8035; Sigma) for 1 h on ice. Cells were washed with RPMI 1640/10% fetal calf serum, incubated with fluorescein isothiocyanate-labeled goat anti-mouse IgG (Dako) for 30 min on ice, and then washed with PBS/2% BSA. The cells were analyzed by fluorescence-activated cell sorting.
Glycosaminoglycan (GAG) Preparations-Purification of heparin from pig intestinal mucosa (22) and selective chemical O-desulfation followed by re-N-sulfation of bovine lung heparin (23) were performed as described. In the 2-O-desulfated heparin, 1% of the iduronic acid residues were 2-O-sulfated, whereas Ͼ80% of the glucosamine residues were 6-O-sulfated. In the 6-O-desulfated preparations, the degree of glucosamine 6-O sulfation was Ͻ10%, but the treatment also resulted in the removal of ϳ30% of the 2-O sulfate groups. The preparations were subjected to high-resolution gel filtration and sterile-filtered to remove possible contamination. There was no sign of toxicity of these preparations in the tissue culture. Chemical depolymerization of bovine lung heparin was performed by limited deamination with nitrous acid at pH 1.5 as described. Heparin fragments were radiolabeled by reduction with NaB 3 H 4 (24) and then separated with regard to size by gel chromatography on a column (1 ϫ 146 cm) of BioGel P-10 in 0.5 M NH 4 HCO 3 . All nonlabeled GAGs were quantified by colorimetric determination of hexuronic acid using the meta-hydroxydiphenyl method with glucoronic acid as a standard (25). A factor of 3 was arbitrarily employed to convert values to saccharide mass.
Filter Binding Assay-Radiolabeled heparin, 2-O-desulfated heparin, and preferentially 6-O-desulfated heparin (0.75 M) of Ն20-mer size were incubated at room temperature for 60 min with PDGF (0 -10 M) in a final volume of 40 l of Tris-HCl, pH 7.4, 150 mM NaCl, and 0.1% BSA. Protein, along with protein-bound oligosaccharides, was trapped on nitrocellulose filters (2.5 cm, diameter) (Schleicher & Schuell) through vacuum suction, whereas nonbound oligosaccharides were washed off with phosphate-buffered saline (26). The protein-bound oligosaccharides were dissociated from the filter in 2 ml of 2 M NaCl and quantitated by scintillation counting.
PDGF-BB Treatment and Immunoprecipitation-Cell cultures at 70% confluence in 10-cm dishes were serum-starved for 16 h in Ham's F-12 medium. Cells were treated for 1 h on ice and for 10 min at 37°C with PDGF-BB at the indicated concentrations in the absence or presence of various polysaccharides and then rinsed with ice-cold PBS. Cells were lysed in Nonidet P-40 lysis buffer (1% Nonidet P-40, 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 10% glycerol, 1 mM Na 3 VO 4 , 1% Trasylol (Bayer), and 1 mM phenylmethylsulfonyl fluoride). Clarified supernatants were incubated with PDGF ␣or ␤-receptor antibodies (Santa Cruz Biotechnology) for 2 h at 4°C and then incubated for 45 min with 40 l of immobilized protein A (EC Diagnostics, Uppsala, Sweden). Beads were washed twice with lysis buffer and once with water. Samples were subjected to SDS-PAGE under reducing conditions, followed by electroblotting to Hybond-C extra membranes (Amersham Biosciences).
Immunoblot Analysis-The membranes were blocked for 1 h at room temperature in TBS-T containing 5% BSA or 5% nonfat milk. Antiphosphotyrosine antibody PY99 or anti-PDGF receptor antibodies (Santa Cruz Biotechnology) were diluted in 0.2% Tween 20/0.1% BSA or 0.2% Tween 20/5% milk and incubated overnight on the membranes, which were then washed in TBS-T. Membranes were incubated for 1 h with the appropriate secondary antibody diluted in 0.2% Tween 20/0.1% BSA or 0.2% Tween 20/5% milk. After several washes in TBS-T, immune reactivity was visualized by an enhanced chemiluminescence detection system (Amersham Biosciences). Before reprobing, filters were stripped in 62.5 mM Tris-HCl, pH 6.7, 2% SDS, and 100 mM ␤-mercaptoethanol at 50°C for 30 min.
Co-precipitation-Heparin (5 g/ml) was incubated with 50 ng/ml PDGF-BB in 0.5 or 0.15 M NaCl in Nonidet P-40 lysis buffer, end over end for 1 h at 4°C, followed by a 2-h incubation of a mixture of two HS proteoglycan monoclonal antibodies, 10E4 (21) and HepSS-1 (Seikagaku), or anti-PDGF-BB anti-serum, a kind gift from Dr. Carl-Henrik Heldin (Ludwig Institute, Uppsala, Sweden). Rabbit anti-mouse immunoglobulins (Dako) and mouse serum (a kind gift from Mikael Karlsson, Department of Genetics and Pathology, Rudbeck Laboratory) were used as controls. Immunoprecipitation and immunoblotting were performed as described above.
Cell Migration Assay-The migration capacity of CHO KI and CHO 677 cells was investigated using a modified Boyden chamber with a micropore nitrocellulose filter (8-m thick, 8-m pore size) precoated with 50 g/ml type I collagen solution overnight. Subconfluent cells were starved in Ham's F-12 medium containing 0.25% BSA (starvation medium) for 16 h. Cells were detached using a nonenzymatic cell dissociation solution (Sigma) for 20 min, washed, and resuspended in starvation medium. Cells were loaded into the upper Boyden chamber wells (25 ϫ 10 3 cells/well) with or without 1 g/ml heparin. Ham's F-12 medium containing 10 ng/ml PDGF-BB was used as a chemoattractant in the lower wells. The migration assay was run for 4 h at 37°C, and then the membrane was fixed in ice-cold methanol and stained with Giemsa solution, and cells on the upper side were removed mechanically. Cells on the lower side were counted in a microscope (ϫ20) in three separate fields. All samples were analyzed in triplicate on four separate occasions.

Tyrosine Phosphorylation of PDGF ␣and ␤-Receptors in PDGF-BB-treated HS-deficient CHO Cells-
We analyzed HSdeficient CHO 677 cells for their ability to respond to PDGF-BB stimulation with increased tyrosine phosphorylation of PDGF ␣and ␤-receptors. Both receptor types were expressed on the cells, and ligand stimulation led to increased tyrosine phosphorylation of the receptors, indicative of activation of their intrinsic tyrosine kinase activities (Fig. 1A). To verify the phenotype of the cells with regard to cell surface proteoglycans, fluorescence-activated cell-sorting analysis was performed after incubation of the cells with antibodies against HS or, as a control, against CS. As seen in Fig. 1B, the CHO 677 cells lacked expression of HS but showed CS expression comparable to that of the wild-type CHO KI cells. Based on indications in the literature that HS-related polysaccharides may affect PDGF function, we decided to examine in more detail the potential effects of heparin on PDGF receptor activation. PDGF-BB was chosen as the ligand because it binds to both PDGF receptors.
Effect of Heparin on PDGF-BB-induced PDGF ␣and ␤-Receptor Tyrosine Phosphorylation in HS-deficient CHO Cells-We first characterized the heparin binding ability of the commercial 24,300-Da PDGF-BB in nitrocellulose filter trapping assays. PDGF-BB clearly bound to 3 H-labeled heparin oligosaccharides in a dose-dependent fashion. Thus, incubation of 0.75 M heparin (Ն20-mer fragments) with PDGF-BB at physiological ionic strength resulted in saturation of the saccharide at a ϳ5-fold molar excess of the protein and an estimated dissociation constant in the micromolar range. Similar results were obtained with 2-O-or 6-O-desulfated heparin oligomers, although these experiments could not be pursued to define the relative affinities of the various saccharides for the growth factor. Nevertheless, the results obtained suggest that significant proportions of the PDGF-BB added to cells in subsequent experiments were complexed to heparin or to its partially O-desulfated derivatives. Furthermore, stable complex formation between PDGF-BB and heparin was obtained in co-immunoprecipitation experiments. Immunoprecipitation was performed using antibodies against PDGF-BB or a mixture of antibodies (10E4 and HepSS-1) against HS proteoglycans, followed by immunoblotting for PDGF-BB. The 10E4 antibody probably recognizes an L-iduronic acid epitope. The HepSS-1 antibody epitope is likely an N-O-sulfated glucoronic acid-rich sequence that recognizes N-sulfates, but not free amino groups, L-iduronic acid, N-acetyl groups, or O-sulfates. There was appreciable co-precipitation of PDGF-BB with heparin, using the anti-HS monoclonal antibodies, whereas the isotype-matched control mouse serum essentially failed to precipitate PDGF-BB (Fig. 2). Washing the immobilized precipitate with 0.5 M NaCl eliminated co-precipitation with the HS antibodies.
Next, HS-deficient CHO 677 cells were treated with PDGF-BB and heparin at increasing concentrations to analyze the effects on PDGF ␣and ␤-receptor tyrosine phosphorylation. Cells were lysed, divided equally, and immunoprecipitated with antibodies specific for PDGF ␤- (Fig. 3A) or ␣-receptors (Fig. 3B), and samples were subjected to immunoblotting with anti-phosphotyrosine antibody. The results showed that PDGF ␣-receptor tyrosine phosphorylation was amplified by heparin in a dose-dependent manner, with a maximal 4-fold effect at 100 ng/ml heparin. At higher concentrations of heparin, the level of PDGF ␣-receptor tyrosine phosphorylation returned to the basal level. In contrast, PDGF-BB stimulated PDGF ␤-receptor tyrosine phosphorylation efficiently in the absence of heparin, and the addition of the polysaccharide did not augment the reaction, set in relation to the loading control. Similar results were obtained in at least three repeated experiments; the augmenting effect of heparin on PDGF ␤-receptor activation was small or nonexistent, whereas the effect on the PDGF ␣-receptor was stable and significant. Furthermore, there was a small effect or no effect of heparin on PDGF-AAstimulated PDGF ␣-receptor tyrosine phosphorylation (Fig.  3C). PDGF-AA does not bind appreciably to the PDGF ␤-receptor, and this combination was therefore not tested. Heparin alone, in the absence of PDGF, did not induce phosphorylation of either PDGF receptor (Fig. 3C).
For comparison, wild-type CHO KI cells were treated similarly with heparin in the presence or absence of PDGF-BB, followed by immunoprecipitation of PDGF receptors and immunoblotting (Fig. 4). In these HS-expressing cells, heparin had no effect on PDGF-BB-induced PDGF ␣or ␤-receptor tyrosine phosphorylation.
The possibility that heparin may increase the affinity of PDGF-BB binding for the PDGF ␣-receptor was tested in a Scatchard analysis. CHO 677 cells were incubated in the presence of 1 ng/ml 125 I-PDGF-BB and increasing concentrations of unlabeled ligand. As shown in Fig. 5, the affinity of PDGF-BB binding to PDGF receptors expressed on CHO 677 cells was similar in the presence and absence of heparin.

Effect of Heparin Desulfation on PDGF-BB-induced Receptor
Tyrosine Phosphorylation-We further tested the effects of selectively desulfated heparin preparations on PDGF-BB-induced PDGF ␣and ␤-receptor tyrosine phosphorylation. CHO 677 cells were treated with PDGF-BB in the absence or presence of 2-O-desulfated heparin (Fig. 6, A and B) or 6-O-desulfated heparin (Fig. 6, C and D) at different concentrations. The cells were lysed, and lysates were immunoprecipitated with PDGF ␤- (Fig. 6, A and C) or ␣-receptor (Fig. 6, B and D) antibodies. There was no appreciable effect of 2-O-or 6-Odesulfated heparin on the PDGF ␤-receptor, in agreement with the lack of effect of native heparin on this receptor (cf. Fig. 3A). On the other hand, 2-O-as well as 6-O-desulfated heparin amplified PDGF-BB-induced PDGF ␣-receptor phosphorylation, albeit to a lower extent than native heparin. However, contrary to the pattern seen with native heparin, there was no dose-dependent decrease in PDGF ␣-receptor activation at higher concentrations of the desulfated heparin preparations.

Heparin Fragments of Four Monosaccharide Units Amplify PDGF-BB-induced PDGF ␣-Receptor
Phosphorylation-We examined the effect of heparin oligosaccharide size on PDGF-BBinduced PDGF receptor tyrosine phosphorylation. As expected, heparin fragments from four monosaccharide units to fulllength heparin showed little or no effect on PDGF ␤-receptor tyrosine phosphorylation in the CHO 677 cells (Fig. 7A). In contrast, PDGF ␣-receptor tyrosine phosphorylation in the same cells was induced 4-fold by PDGF-BB in the presence of the 4-mer at 100 ng/ml (Fig. 7B). Treatment with the decasac-charide fragment led to 6-fold amplification of PDGF ␣-receptor tyrosine phosphorylation by PDGF-BB, whereas longer saccharide fragments were slightly less efficient.
Heparin  and PKB/Akt phosphorylation in a dose-dependent manner (Fig. 8), indicative of HS-modulated activation of these signaling components in the intact, PDGF-BB-stimulated cell.
To determine whether co-treatment with heparin would increase cellular responsiveness to PDGF-BB, we analyzed directed migration of CHO 677 cells, compared with CHO KI cells, under different conditions as shown in Fig. 9. The HSexpressing CHO KI cells displayed a relatively high basal migration and only a slight increase in migration toward PDGF-BB; there was no appreciable additional stimulation when cells also received heparin. The CHO 677 cells also migrated poorly toward PDGF-BB, but the addition of heparin significantly augmented the chemotactic migration of the cells.

DISCUSSION
In this study, we show that the addition of heparin augmented PDGF-BB-induced activation of the PDGF ␣-receptor, but not PDGF ␤-receptor, in HS-deficient CHO 677 cells expressing endogenous PDGF receptors. Moreover, heparin promoted PDGF-BB-induced intracellular signaling and increased chemotaxis of such cells. Heparin alone had no effect on PDGF receptor tyrosine phosphorylation. The lack of effect of heparin on PDGF ␤-receptor tyrosine phosphorylation is in agreement with previous data (27). Moreover, heparin had no effect on PDGF-AA-stimulated PDGF ␣-receptor tyrosine phosphorylation.
The effect of heparin on the PDGF-BB-stimulated PDGF ␣-receptor did not appear to be due to increased binding affinity. However, the CHO 677 cells express both PDGF ␣and ␤receptors, and it is possible that PDGF-BB binding to the PDGF ␤-receptor obscured any heparin-related changes in affinity for the PDGF ␣-receptor. The dose-dependent effect of heparin was maximal at around 100 ng/ml heparin, whereas higher concentrations of 1-5 g/ml lacked effect. Short oligosaccharide fragments (4-mers and longer) were active in this model. Heparin lacking either the 2-O-sulfate groups on iduronic acid units or the 6-O-sulfate groups on glucosamine units retained the ability to augment PDGF-BB-induced PDGF ␣-receptor activation, although the effect was reduced, and the dose-response pattern was changed. Notably, there was no absolute requirement for heparin in PDGF ␣-receptor activation by PDGF-BB. This is in agreement with previous data on chlorate-treated fibroblasts deficient in cell surface HS, which still respond to PDGF-BB with increased mitogenic activity (28).
Previous reports in the literature indicate that PDGF-BB does indeed bind heparin and that biological responses to PDGF may depend on interaction with HS. Thus, heparinbinding fragments from fibronectin (29) or apolipoprotein E (30) negatively modulate proliferative responses to PDGF-BB. Furthermore, PDGF-BB may be deposited in the matrix through binding to heparan sulfate proteoglycans because treatment with heparatinase I allows release of biologically active growth factor (16). Binding of heparin to the long PDGF-AA isoform is dependent on N-, 2-O-, and 6-O-sulfation (31); the minimal size binding to PDGF-AA L is an octasaccharide. The long PDGF-A isoform contains an 18-amino acid residue polybasic stretch encoded by the alternatively spliced exon 6 in the PDGF-A chain gene. The PDGF-B isoform contains a similar but not identical polybasic stretch encoded by exon 6 in the PDGF-B chain gene. This stretch is proteolytically removed to generate the mature processed PDGF-BB. The removal of the polybasic stretch does not preclude heparin binding because the short PDGF-AA isoform, which lacks this stretch, still binds heparin, although with reduced affinity (32). Furthermore, three basic residues in the loop III receptorbinding domain present in the short and long form of PDGF-BB have been identified as important for heparin binding (33). The commercially available short form of PDGF-BB (M r 24,300) used in this study presumably lacked the exon 6-encoded polybasic heparin-binding sequence but nevertheless retained binding capacity for heparin and modified heparin fragments as shown in a filter binding assay.
The mode of action of heparin/HS in relation to PDGF-BB and its PDGF ␣-receptor remains unclear. The effect appears highly specific for this combination of ligand and receptor be- cause heparin did not modulate PDGF-AA-stimulated PDGF ␣-receptor tyrosine phosphorylation. This is compatible with the observation that PDGF-AA and -BB bind with different affinities and induce different conformational changes in the PDGF ␣-receptor extracellular domain (34). The possibility that heparin/HS may physically interact not only with growth factors but also with their receptors has been argued for fibroblast growth factors (FGFs) and the corresponding receptors (FGFRs) (35)(36)(37)(38). Indeed, x-ray crystallography studies of ternary complexes show heparin oligosaccharides in contact with both FGF and FGFR proteins (35,39). We do not know whether PDGF receptor ectodomains bind heparin/HS. The interaction between heparin and FGFR-1 appears to involve a basic stretch, denoted K18K, in the FGFR-1 extracellular domain (residues Lys 160 to Lys 177 ) (40). Although there is no obvious polybasic region in the PDGF ␣-receptor extracellular domain, there is some sequence similarity between the FGFR-1 K18K sequence and regions in the PDGF ␣-receptor (data not shown); such a similarity is not recorded for the PDGF ␤-receptor in homology searches. Additional studies are needed to show whether the PDGF ␣-receptor binds heparin with any measurable affinity.
Our observation that PDGF ␣-receptor activation by PDGF-BB is augmented not only by full-sized heparin but also by relatively short oligosaccharides would seem to argue against a bridging function for the saccharide in a ternary complex with growth factor and receptor. However, we note that similar (and unexplained) effects of small saccharides have been observed also in connection with FGF action (41,42). Several additional possibilities may be considered. Saccharide binding may change the conformation of PDGF-BB in such a way that its interaction with the PDGF ␣-receptor, but not with the PDGF ␤-receptor, is promoted. Alternatively, receptor binding of the saccharide may selectively make the PDGF ␣-receptor more receptive to the growth factor. On the other hand, the involvement of a more extended "bridging" domain could explain why the receptor-stimulatory effect decreases at higher heparin concentrations; under these conditions, the probability of binding PDGF-BB and its receptor to the same polysaccharide chain will decrease. Conversely, the partially O-desulfated heparin derivatives may present fewer binding sites along the chain, thus explaining the persistent stimulation at higher saccharide concentration. Finally, we cannot exclude the possibility that the selective effect of heparin on PDGF ␣-receptor phosphorylation is mediated by an as yet unidentified protein ligand(s).
Notably, FGF-2 stimulation of FGF receptor-1 in the absence of heparin/HS elicits FGF receptor activation and signal transduction, but the spectrum of autophosphorylation sites employed and the range of signal transduction pathways that become initiated are limited compared with stimulation in the presence of heparin. 2 These data indicate that signal transduction by receptor tyrosine kinases can be directed by heparinmediated changes in receptor conformation or by effects on other properties of the kinase. Our data indicate that signal transduction and cellular responses to PDGF-BB are augmented by heparin. Whether this effect is purely quantitative or also qualitative is an interesting issue for future studies.