Molecular Biology of the Cell Vol. 3, 1057-1065, October 1992

Essay

The Extracellular Regulation of Growth Factor Action Robert Flaumenhaft and Daniel B. Rifkin Department of Cell Biology and Kaplan Cancer Center, New York University Medical Center, and the Raymond and Beverly Sackler Foundation Laboratory, New York, New York 10016

INTRODUCTION The concept that growth factors act as autocrine and paracrine regulators of cell growth and differentiation was introduced over a decade ago (Sporn and Todaro, 1980). Since then significant advances have been made in understanding the regulation of growth factor expression and interaction with high affinity plasma membrane receptors. More recently, structural analysis and in vitro experiments suggest that regulatory mechanisms other than growth factor expression and receptor binding dominate growth factor action. These data demonstrate that most growth factors do not simply diffuse passively from the site of their release to interact with the nearest cell surface receptor. Rather, many of these proteins contain structural features that foster their interaction with other molecules within the extracellular environment. These interactions confine the action of the growth factor to an appropriate place and time. Thus, the specificity of growth factor signaling is derived not only from its binding to cell surface receptors but also through critical interactions with other components of the extracellular environment. In this review we will describe recently appreciated mechanisms involved in the nonreceptor mediated extracellular regulation of growth factor activity. THE CENTRAL ISSUE: REGULATION OF ACTIVITY

Whereas substantial progress has been made in determining the tissue localization, biological effects, and structure of growth factors, the question of how the activity of these molecules is controlled once the proteins are released from the cell has just begun to be addressed. Many growth factors and their receptors have broad distributions and elicit their biological activities when present at pico- or nanomolar concentrations. The ubiquity and potency of growth factors suggest that forces other than simple diffusion may be required to regulate the stimulation of a discrete population of cells for an appropriate duration by a specific growth factor. In addition, some growth factors are present in plasma and tissues in latent forms at concentrations well above the effective dose of their active counterparts. Regulation of the release of such growth factors from the latent C 1992 by The American Society for Cell Biology

complexes must be critical to the organism to avoid errors in extracellular signaling that could contribute to inappropriate cell growth, differentiation, and function. REGULATION IN TIME AND SPACE To elicit an appropriate response, the availability of a growth factor for its cell surface receptor must be regulated in time and space. Two questions arise when considering the issue of regulation in time: 1) what mechanisms initiate growth factor action and 2) what factors control the duration of growth factor activity? The initiation of growth factor activity may be regulated by expression, secretion from the cell, and/or release from a latent complex. Expression and secretion involve intracellular processes and will not be addressed here. Release from an inactive complex may control the onset of activity of several factors. This mechanism for the initiation of a biological response has been observed in the regulation of other bioactive molecules such as clotting factors. In fact, some of the sequence motifs (e.g., kringle and epidermal growth factor-like repeats) present in surface-activatible clotting factors are also present in growth factors and their binding proteins (e.g., hepatocyte growth factor and the binding protein of latent transforming growth factor-f1) (Kanzaki et al., 1990; Tashiro et al., 1990). The availability of the reactants that catalyze the release of the active growth factor also effects the duration of growth factor activity because new factor will only be produced in the presence of the activating molecules. In addition, the duration of growth factor action can be controlled both by additional interactions that act directly on the growth factors (e.g., proteases, scavenger molecules, and dilution to substimulatory concentrations) and mechanisms that act on cells to alter their response to growth factor (e.g., molecules that down-modulate the response of a cell to a growth factor and/or cause a decrease in growth factor receptor number). The evaluation of the regulation of growth factor action in space must consider the question of how a protein is transported to and maintained at a particular location. Mechanisms exist both to maintain certain growth factors at the site of their release and to allow other species to act in locations some distance away from their release. Sequestration is accomplished through transmembrane 1057

R. Flaumenhaft and D.B. Rifkin

domains that anchor the growth factor in the plasma membrane (kit-ligand, transforming growth factor-a [TGF-a],7 epidermal growth factor [EGF], colony stimulating factor-1 [CSF-1], tumor necrosis factor-a [TNFa]) (Massague, 1990) or motifs that bind growth factors to extracellular matrix components (heparin-binding domains of basic fibroblast growth factor [bFGF] or heparin binding-EGF [HB-EGF], exon 6 of platelet-derived growth factor [PDGF]) (Baird et al., 1988; Higashiyama, 1991; Raines and Ross, 1992). Release from these sites often occurs in a regulated manner. For example, the release of bFGF or TGF-a from extracellular matrix and membrane stores can be mediated by specific extracellular proteases (Saksela and Rifkin, 1990; Pandiella and Massague, 1991). In addition, alternative splicing allows some growth factors to be synthesized in both membrane-anchored and soluble forms (Rathjen et al., 1990; Raines and Ross, 1992). Once a growth factor partitions into the soluble phase, it can diffuse to act on cells some distance away (Flaumenhaft et al., 1990). However, its destination is not only dictated by laws of simple diffusion, but also by the affinity of the factor for components of both the aqueous environment (e.g., scavenger molecules) and the matrix. In this manner, structural features of growth factors may target them to destinations other than their cell surface receptors. This regulated localization of growth factors appears to be an essential aspect of their activity. Analysis of growth factor structures demonstrates that many of these polypeptides occur either as high molecular weight complexes (consisting of the active growth factor and binding protein), as membrane-anchored and/or latent precursors, or in association with matrix proteins. Structural motifs within many of these growth factors interact with specific components of the extracellular environment. These interactions influence the initiation and duration of growth factor activity and direct the localization and targeting of the growth factors. Thus, comprehending the regulation of growth factor action requires knowledge of 1) the binding proteins with which growth factors associate, 2) the precursors from which they are derived, and 3) the matrix components to which they bind. These features of growth factor action are discussed below. BINDING PROTEINS Many growth factors have been isolated as complexes consisting of the factor and one or more binding pro'Abbreviations used: aFGF, acidic fibroblast growth factor; a2-M, alpha2-macroglobulin; bFGF, basic fibroblast growth factor; CSF-1, colony stimulating factor-1; EGF, epidermal growth factor; HB-EGF, heparin binding-EGF; HGF, hepatocyte growth factor; IGF, insulinlike growth factor; IGF-BP, insulin-like growth factor binding proteins; ILs, interleukins; INF--y, interferon-gamma; LAP, latency associated peptide; LTBP, latent TGF-,B binding protein; NGF, nerve growth factor; PDGF, platelet-derived growth factor; TGF-a, transforming growth factor a; TGF-#, transforming growth factor beta; TNF-a, tumor necrosis factor-a; VGF, vaccinia virus growth factor. 1058

teins. The binding protein may be a separate gene product, as in the case of insulin-like growth factor-binding proteins (IGF-BPs), or a part of the growth factor precursor, as in the case of the transforming growth factor beta (TGF-f) propeptide. In the first example, the synthesis of the binding protein and the growth factor are under separate control, whereas in the second case, the synthesis of both molecules is under the same control. The interaction of the binding proteins with purified growth factors may be specific (i.e., the binding protein will interact only with a single growth factor) or nonspecific (i.e., the binding protein will interact with a variety of growth factors). The potential roles for binding proteins in the extracellular regulation of growth factors are diverse (Hintz, 1990). They can 1) provide a relatively constant supply of growth factor, 2) alter the action of a factor by either potentiating or inhibiting its activity, 3) protect growth factors from degradation, and 4) potentially target growth factors to specific tissues. A particular binding protein may simultaneously serve several of these functions, and a particular growth factor may interact with different binding proteins. An example of a family of binding proteins that serve to provide cells with a constant supply of growth factor is the insulin-like growth factor binding proteins (IGFBPs). IGF-BPs were first identified as serum proteins that prolonged the half-life of the insulin-like growth factors (IGFs) (Cohen and Nissley, 1976; Kaufmann et al., 1977). The IGF-BPs extend the half-life of IGF in vitro and in vivo from a few minutes (for free IGF) to several hours (Cohen and Nissley, 1976; Zapf et al., 1986). This increase in half-life permits IGF to circulate at levels (- 1 ug/ml) two to three orders of magnitude higher than most peptide hormones (Hintz, 1990). In fact, the concentration of IGF in the bloodstream may depend primarily upon IGF-BP levels rather than the rate of IGF synthesis (Blum et al., 1989b). The carrier function of IGF-BPs not only affects the availability of IGFs, but may also alter the potency of IGF (Knauer and Smith, 1980; Mottala et al., 1986; Elgin et al., 1987; Bugsby et al., 1988). In addition, some IGF-BPs contain the amino acid sequence arg-gly-asp (RGD), a motif involved in cell surface protein-protein interactions (Hintz, 1990). This domain may direct the high molecular weight IGF complex to sites at the cell surface where contact with glycosaminoglycans (GAG) and proteases might mediate the liberation of IGF (Clemmons et al., 1983; Baxter, 1990). A second type of binding protein is represented by soluble receptors that bind their respective ligands with specificity and avidity similar to that of the corresponding cell-surface receptors. Soluble receptors have been demonstrated for EGF (Ullrich et al., 1984; Weber et al., 1984), nerve growth factor (NGF) (Zupan et al., 1989), tumor necrosis factor (TNF) (Gray et al., 1990; Schall et al., 1990), interleukins (ILs) 1 (Symons and Duff, 1990), 2 (Marcon et al., 1988), 4 (Mosley et al., 1989), 6 (Novick Molecular Biology of the Cell

The Extracellular Regulation of Growth Factor Action

et al., 1989), and 7 (Goodwin et al., 1990), interferongamma (INF--y) (Novick et al., 1989), IGF-II/mannose6-phosphate (Bobek et al., 1991), bFGF (ohnson et al., 1990), and colony-stimulating factors (CSFs) (Downing et al., 1989; Fukunaga et al., 1990). These soluble receptors are generated by either enzymatic or genetic mechanisms (Ullrich et al., 1984; Weber et al., 1984; Downing et al., 1989; Mosley et al., 1989; Fukunaga et al., 1990; Johnson et al., 1990; Symons and Duff, 1990). By competing with cell-associated receptors for ligand binding, soluble growth factor receptors can interfere with growth factor activity (Mosley et al., 1989; Gray et al., 1990; Schall et al., 1990). Alternatively, soluble receptors may prolong growth factor half-lives as has been demonstrated with other binding proteins. Thus, the modulation of soluble receptor levels may represent an additional mechanism for the regulation of growth factor activity. In addition to interacting with specific binding proteins, growth factors can complex with certain abundant plasma proteins. For example, alpha2-macroglobulin (a2-M) has been found to associate with nerve growth factor (NGF) (Ronne et al., 1979), EGF (Ito et al., 1984), PDGF (Huang et al., 1984), transforming growth factors,B1 and 32 (TGF-31 and ,B2) (O'Connor-McCourt and Wakefield, 1987; Danielpour and Sporn, 1990; LaMarre et al., 1991), IL-1,B (Borth and Luger, 1989), bFGF (Dennis et al., 1989), IL-6 (Matsuda et al., 1989), and TNF (Wollenberg et al., 1991). Proposed functions for a2-M in regulating growth factor activity include clearance of growth factor from the circulation (LaMarre et al., 1991), inhibition or stimulation of activity (Bonner et al., 1990; Shi et al., 1990), protection from proteolytic degradation (Ronne et al., 1979), and delivery of growth factor to distant sites (McCaffrey et al., 1989). Another potential role of binding proteins is to target the growth factor to specific cell or tissue targets. The latent TGF-f binding protein (LTBP) contains 16 EGFlike repeats, an RGD sequence, and a laminin B2 sequence (Kanzaki et al., 1990); motifs previously shown to be involved in protein-protein interactions (Ruoslahti and Pierschbacher, 1987; Sasaki and Yamada, 1987; Appella et al., 1988). LTBP has been shown to participate in the activation of latent TGF-,B1 (Flaumenhaft et al., unpublished data) and might serve to direct the latent high molecular weight TGF-,31 to a cell surface site where mature TGF-,B1 is released from the latent complex via proteolysis (Sato et al., 1990).

GROWTH FACTOR PRECURSORS Many growth factors are synthesized as either membrane-bound precursors inserted into the plasma membrane with their growth factor sequences exposed on the extracellular surface or as high molecular weight forms that must be altered to release the active species. An example of a growth factor synthesized as a soluble Vol. 3, October 1992

precursor that must be activated is TGF-i3. Growth factors synthesized as large precursors that remain associated with the plasma membrane include EGF (Gray et al., 1983; Mroczkowski et al., 1989), TGF-a (Derynck et al., 1984; Lee et al., 1985), CSF-1 (Stein et al., 1990), TNF-a (Perez et al., 1990), kit-ligand (Anderson et al., 1990), vaccinia virus growth factor (VGF) (Stroobant et al., 1985), and amphiregulin (Plowman et al., 1990). An advantage of this type of restriction on growth factor action is that during processes such as development, hematopoiesis, and cell killing, discrete signaling might occur that requires stimulation of one cell type exclusively by adjacent cells (Perez et al., 1990; Stein et al., 1990). A diffusible factor would not be appropriate in such cases, whereas the membrane-inserted forms of growth factors would have a very restricted action. Membrane-anchored growth factors have been shown to be active as transmembrane species and do not need to be released for full activity. For example, transmembrane TGF-a precursors are able to stimulate cell signaling and proliferation through binding to the EGF/TGF-a receptor (Brachman et al., 1989; Wong et al., 1989; Anklesaria et al., 1990). In addition to stimulating cell signaling, some membrane-anchored growth factors may promote cell-cell adhesion perhaps through EGF-like repeats (TGF-a, Anklesaria et al., 1990) or through other mechanism (kit-ligand, see below). Growth factors synthesized as precursors containing membrane-spanning domains can be solubilized by proteolysis as has been shown for TGF-a (Massague, 1990; Mueller et al., 1990; Pandiella and Massague, 1991). In addition, alternative splicing may generate both membrane-inserted and diffusible forms of growth factor (Brannan et al., 1991; Flanagan et al., 1991). Perhaps, the most striking evidence of a physiological role for membrane-bound growth factors is derived from experiments performed using mice with the steelDickie (Sld) mutation. The Sl gene product is the kitligand (Brannan et al., 1991), which exists in both soluble and membrane-anchored forms. Although the two forms differ in their cellular localization, they have similar signaling activities. However, the membrane-anchored form of the kit-ligand is able to mediate activities, such as cell-cell adhesion between COS-7 cells and mast cells, that the soluble form cannot (Flanagan et al., 1991). The supposition that membrane-bound kit-ligand is necessary for normal development is illustrated by mice that are homozygous for a mutant SId gene which encodes exclusively the kit-ligand form that lacks the transmembrane and cytoplasmic domains (Brannan et al., 1991; Flanagan et al., 1991). This soluble mutant form of kit-ligand is as active as wild type kit-ligand in biological assays and is expressed at similar levels. However, mice with this mutation are anemic, sterile, and have altered pigmentation. Other experiments performed in vitro demonstrate that membrane-anchored kit-ligand supports primordial germ cell growth, 1059

R. Flaumenhaft and D.B. Rifkin

whereas soluble kit-ligand does not (Dolci et al., 1991). Thus, the physiological consequences of the Sld mutation probably reflect the requirement of a membranebound form of kit-ligand in normal development. An example of growth factors secreted from cells in high molecular weight latent forms are the TGF-,Bs. The TGF-3s are synthesized as precursors that are processed intracellularly into mature TGF-# and a propeptide termed the latency associated peptide (LAP) (Derynck et al., 1985). Mature TGF-3 and LAP are secreted as an inactive complex termed latent TGF-# and remain associated in the extracellular environment through noncovalent interactions. TGF-# must be released from LAP to bind to its cell surface receptor. LAP regulates TGF,B activity in cell culture (Gentry and Nash, 1990) and in vivo (Wallick et al., 1990) and influences the distribution of TGF-,B in the organism, increasing the halflife of the growth factor 50-fold (Wakefield et al., 1990). Latent TGF-f circulates in plasma at concentrations well above those necessary for TGF-f to elicit biological activity (Danielpour et al., 1989). Thus, regulation of TGF,B activity is selectively mediated at the level of activation (i.e., release of mature TGF-f from the latent TGF-,3 complex). Because other members of the TGF-f superfamily are structurally similar to TGF-(3, the control of activity via noncovalent interaction with the propeptide may extend to these proteins as well. However, little data are as yet available on this aspect of control. If activation of the latent molecule controls the level of active TGF-f3, one would anticipate that this is a highly regulated event. In fact, most cells known to synthesize TGF-#1 do not activate the latent molecule when grown in culture. Many cells types, however, can be stimulated to activate latent TGF-fl when exposed to bioactive agents such as retinoic acid, antiestrogens, or growth factors (Knabbe et al., 1987; Glick et al., 1989; Oreffo et al., 1989; Colletta et al., 1990; Twardzik et al., 1990; Flaumenhaft et al., unpublished data). Activation of latent TGF-#1 also occurs in homotypic cultures of erythroleukemia cell lines (Piao et al., 1990) and in cocultures of endothelial and smooth muscle cells (or pericytes) (Antonelli-Orlidge et al., 1989; Sato and Rifkin, 1989). Several mechanisms for the activation of latent TGF-31 have been proposed. Because acid-treatment (-pH 3) of latent TGF-31 releases the active peptide (Lawrence et al., 1985), several groups have suggested that latent TGF-f31 might be activated in vivo in acidic compartments (Jullien et al., 1989; Pfeilschifter et al., 1990). However, it is difficult to conceive of how such a process would be controlled and how the mature growth factor would be released into the extracellular milieu. Activation of latent TGF-31 has also been accomplished by enzymatic alteration of carbohydrate side chains within LAP (Miyazono and Heldin, 1989) suggesting that these residues are important in maintaining latency. A mechanism of activation of latent TGF-#1 that may have broad physiological relevance is that 1060

mediated by proteases. Purified plasmin has been demonstrated to cleave latent TGF-,B1 within the LAP releasing active growth factor (Lyons et al., 1988, 1990). Furthermore, inhibitors of plasmin inhibit the activation of latent TGF-#1 observed upon coculturing endothelial and smooth muscle cells (Sato and Rifkin, 1989; Sato et al., 1990), treatment of endothelial cells with retinoids (Kojima and Rifkin, unpublished data), and treatment of endothelial cells with certain growth factors (Flaumenhaft et al., unpublished data). Binding of the latent TGF-# complex to mannose-6-phosphate/IGF II cation independent receptors and the latent TGF-,B binding protein also appear to be involved in the activation of latent TGF-,B in cocultures of endothelial and smooth muscle cells (Dennis and Rifkin, 1991; Flaumenhaft et al., unpublished data).

EXTRACELLULAR MATRIX A soluble growth factor may interact with a variety of molecules in the extracellular environment other than its cell-surface receptor. For most growth factors the most abundant extracellular binding molecules are components of the extracellular matrix (Ruoslahti and Yamaguchi, 1991). Some of the growth factors that bind to matrix components are fibroblast growth factors (FGFs), TGF-fl, HB-EGF, PDGF-AL, and hepatocyte growth factor (HGF). Growth factor matrix interactions can alter growth factor potency, retain either active or inactive growth factors in the vicinity of the cell, modify growth factor diffusion properties, increase growth factor stability, and protect growth factors from proteolytic degradation. An example of the effect of matrix on growth factor potency is observed with aFGF whose mitogenic activity is 20to 100-fold greater in the presence of heparin than in its absence (Thorton et al., 1983). In addition, growth factor interaction with matrix molecules may be required for the binding of the growth factor to its receptor as has been described for the interaction of bFGF and heparan sulfate with the FGF receptor (Rapraeger et al., 1991; Yayon et al., 1991). In contrast, TGF-,B1 activity is inhibited by interaction with the proteoglycans decorin or biglycan (Yamaguchi et al., 1990). This inhibition is of particular significance because TGF-,31 induces the expression of these small proteoglycans in certain cells (Okuda et al., 1990). Thus, TGF-31 activity may be controlled by a negative feedback system in which the induction of decorin and/or biglycan synthesis by TGF-l31 increases the expression of molecules that inhibit further TGF-,31 activity. The ability of the extracellular matrix to function as a slow release reservoir for growth factors, as has been shown for bFGF, may explain how a molecule released over a short period of time can stimulate processes, such as angiogenesis, that take days or weeks (Flaumenhaft et al., 1989). SimMolecular Biology of the Cell

The Extracellular Regulation of Growth Factor Action

ilarly, the binding of TGF-1l to the heparan and chondroitin sulfate proteoglycan, betaglycan, may concentrate TGF-f at the cell surface for presentation to high affinity receptors or aid in the clearance of TGF-3 (Andres et al., 1989). The potential importance of compartmentilization of growth factor activity via interaction with matrix has also been suggested for growth factors involved in bone maintenance (Hauschka et al., 1988), hematopoiesis (Gordon et al., 1988; Roberts et al., 1988), and lymphopoiesis (Kimura et al., 1991). The interaction of growth factors with matrix molecules can also affect the diffusion of the growth factor. This has been shown in vitro with bFGF where interaction with soluble glycosaminoglycans or proteoglycans may neutralize the charge of the basic growth factor. This permits the growth factor to partition into the mobile aqueous phase rather than the insoluble matrix. The complex of growth factor and GAG, even though it is larger than the free growth factor, can then diffuse further than the free growth factor (Flaumenhaft et al., 1990). The interaction of growth factors with matrix molecules can also stabilize the growth factor. For example, heparin protects aFGF and bFGF from inactivation by heat or acid (Gospodarowicz and Cheng, 1986) and proteases (Rosengart et al., 1988; Saksela et al., 1988; Sommer and Rifkin, 1989). The half-life of aFGF is increased 100-fold in the presence of heparin (Damon et al., 1989; Ortega et al., 1991). Growth factors that associate with matrix can often be isolated in both diffusible and matrix-bound forms. These different forms of growth factor can be generated by proteolytic processing and/or alternative splicing. PDGF-A occurs in two forms that arise from alternative splicing of exon 6 yielding PDGF-AL, which is primarily cell-associated, or PDGF-As, which is secreted (Betsholtz et al., 1986; Raines and Ross, 1992). The C-terminal extension encoded by exon 6 contains a stretch of basic amino acids that confers matrix-binding properties on PDGF-B and PDGF-AL (Robbins et al., 1985; LaRochelle et al., 1991) through interaction with heparan sulfate proteoglycans in the extracellular matrix (Raines and Ross, 1992). PDGF-AL found in conditioned medium probably derives from proteolysis of the retention sequence of the growth factor because inhibition of proteolytic processing prevents the release of PDGFAL into culture medium (LaRochelle et al., 1991). Alternative splicing also effects the cellular localization of differentiating inhibitory activity (Rathjen et al., 1990), vascular endothelial growth factor (via alternative splicing of a basic amino acid motif similar to exon 6 of PDGF) (Leung et al., 1989; Betsholtz et al., 1990), and int-1 (van Ooyen et al., 1984; Papkoff and Schryver, 1990). An additional mechanism by which both matrixbound and diffusible forms of growth factor can be produced is by solubilization of growth factor-binding matrix components. For example, bFGF binds avidly to Vol. 3, October 1992

heparan sulfate proteoglycans in the extracellular matrix. This retention of bFGF in the matrix permits longterm stimulation of target cells after brief exposure to the growth factor (Flaumenhaft et al., 1989). However, bFGF bound to heparan sulfates may remain associated with the cell surface only temporarily because heparan sulfates are released spontaneously from the matrix and cell surface. Furthermore, the release of bFGF-heparan sulfate complexes from the extracellular matrix can be mediated by agents (growth factors) that control pericellular protease (Saksela and Rifkin, 1990) or phospholipase activity (Brunner et al., 1991). The masking of heparin-binding sites of bFGF by soluble heparan sulfate prevents bFGF from rebinding to insoluble heparan sulfate proteoglycans (Flaumenhaft et al., 1990). Because the bFGF-heparan sulfate complex maintains the same biological potency as bFGF (Moscatelli, 1987; Saksela et al., 1988), this complex functions as a diffusible form of active bFGF. Thus, diffusible forms of growth factor can be generated by solubilization of the insoluble extracellular components to which they are bound.

CONCLUDING REMARKS Growth factors have been likened to symbols of an intercellular language (Sporn and Roberts, 1990). Investigators are beginning to appreciate, however, that these proteinaceous messages are not simply released into the extracellular environment to passively interact with cell surface receptors. We suggest that most growth factors are endowed with structures that direct the timing and localization of their activity in the extracellular environment. These structures can maintain growth factors in particular compartments (e.g., binding of bFGF to matrix through its heparin-binding domain or EGF to plasma membrane through its transmembrane domain), confer protease sensitivity on growth factors (e.g., the ala-ala-ala t val-val cleavage site of TGF-a) (Pandiella and Massague, 1991), render the factors unstable (e.g., free cysteines of aFGF) or latent (e.g. LAP of latent TGF1), and target growth factors to nonreceptor surface proteins (e.g. RGD of IGF-BP and EGF-like domains of LTBP). Thus, the extracellular activity of growth factors is not dictated simply by laws of passive diffusion; rather, their activities are regulated by motifs within their structures. Understanding the information within these motifs is becoming an important dimension of growth factor research. Although much work in understanding growth factor regulation has been accomplished in recent years with regard to the transcriptional control of growth factor expression, the studies described in this review and elsewhere (Nathan and Sporn, 1991) suggest that regulation of growth factor activity cannot be understood solely at the level of expression. For example, it is our opinion that an increase in the expression of a latent 1061

R. Flaumenhaft and D.B. Rifkin

growth factor, such as latent TGF-#, is not likely to impact directly on cell function because the latent factor is already present in plasma at a concentration above that necessary to elicit maximal biological activity. In contrast, an increase in the ability of the cell to release the active growth factor from its precursor is likely to have profound consequences on cellular physiology. Similarly, an active factor may be expressed but only perform its function if associated with a particular cellular compartment (e.g., membrane-bound macrophage growth factor [MGF]) or with a particular accessory molecule (e.g., bFGF associated with heparin). To understand how growth factor action is regulated in the extracellular environment, we must understand the structural code that conveys the regulatory information. Unraveling the structural code that determines the localization and duration of growth factor activity will involve determination of 1) the specificity of motifs within growth factors that direct factor to particular extracellular locations (which motifs interact with which extracellular components), 2) the structural features that affect the stability and protease sensitivity of growth factors, 3) the functional significance of structures within growth factors that interact with nonreceptor extracellular molecules (are they necessary for receptor recognition? are they required to increase local concentrations? do they mediate adhesion?), and 4) the potential regional variations in extracellular effectors of growth factor action. Such information will provide us with a clearer understanding of how growth factors act in complex environments and may enable us to manipulate effectively the activities of these potent regulators of cell function.

ACKNOWLEDGMENTS This research was supported by grants from the National Institutes of Health and the American Cancer Society.

REFERENCES Anderson, D.M., Lyman, S.D., Baird, A., Wignall, J.M., Eisenman, J., Rauch, C., March, C.J., Boswell, H.S., Gimpel, S.D., Cosman, D., and Williams, D.E. (1990). Molecular cloning of mast cell growth factor, a hematopoietin that is active in both membrane and soluble forms. Cell 63, 235-243. Andres, J.L., Stanley, K., Chiefetz, S., and Massague, J. (1989). Membrane-anchored and soluble forms of betaglycan, a polymorphic proteoglycan that binds transforming growth factor-#. J. Cell Biol. 109, 3137-3145.

Anklesaria, P., Teixido, J., Laiho, M., Pierce, J.H., Greenberger, J.S., and Massague, J. (1990). Cell-cell adhesion mediated by binding of membrane-anchored transforming growth factor a to epidermal growth factor receptors promotes cell proliferation. Proc. Natl. Acad. Sci. USA 87, 3289-3293. Antonelli-Orlidge, A.A., Saunders, K.B., Smith, S.R., and D'Amore, P. (1989). An activated form of transforming growth factor beta is produced by cocultures of endothelial cells and pericytes. Proc. Natl. Acad. Sci. USA 86, 4544-4548.

1062

Appella, E., Weber, I.T., and Blasi, F. (1988). Structure and function of epidermal growth factor-like regions in proteins. FEBS Lett. 231, 1-4. Baird, A., Schubert, D., Ling, N., and Guillemin, R. (1988). Receptor and heparin-binding domains of basic fibroblast growth factor. Proc. Natl. Acad. Sci. USA 85, 2324-2328. Baxter, R.C. (1990). Glycosaminoglycans inhibit formation of the 140 kDa insulin-like growth factor-binding protein complex. Biochem. J. 271, 773-777. Betsholtz, C., Johnsson, A., Heldin, C.-H., Westermark, B., Lind, P., Urdea, M.S., Eddy, R., Shows, T.B., Philpott, K., Mellor, A.K., Knott, T.J., and Scott, J. (1986). cDNA sequence and chromosomal localization of human platelet-derived growth factor A-chain and its expression in tumour cell lines. Nature 320, 695-699. Betsholtz, C., Rorsman, F., Westermark, B., Ostman, A., and Heldin, C.-H. (1990). Analogous alternative splicing. Nature 344, 299. Blum, W.F., Ranke, M.B., and Kietzman, K. (1989b). Excess of IGFbinding protein in chronic renal failure: evidence for relative resistance and inhibition of somatomedin activity. In: The Insulin-Like Growth Factor Binding Proteins, ed. S.L.S. Drop and R.L. Hintz, Amsterdam: Excerpta Medica, 93-102. Bobek, G., Scott, C.D., and Baxter, R.C. (1991). Secretion of soluble insulin-like growth factor-II/mannose 6-phosphate receptor by rat tissue in rat culture. Endocrinology 128, 2204-2206. Bonner, J.C., Badgett, A., Osornio-Vargas, A.R., Hoffman, M., and Brody, A.R. (1990). PDGF-stimulated fibroblast proliferation is enhanced synergistically by receptor-recognized alpha2-macroglobulin. J. Cell. Physiol. 145, 1-8. Borth, W., and Luger, T.A. (1989). Identification of alpha2-macroglobulin as a cytokine binding plasma protein. Binding of interleukin1 beta to 'F' alpha2-macroglobulin. J. Biol. Chem. 264, 5818-5825. Brachman, R., Linquist, P.B., Nagashima, M., Kohr, W., Lipari, T., Napier, M., and Derynck, R. (1989). Transmembrane TGF-a precursors activate EFG/TGF-a receptors. Cell 56, 691-700. Brannan, C.I., Lyman, S.D., Williams, D.E., Eisenman, J., Anderson, D.M., Cosman, D., Bedell, M.A., Jenkins, N.A., and Copeland, N.G. (1991). Steel-Dickie mutation encodes a c-kit ligand lacking transmembrane and cytoplasmic domains. Proc. Natl. Acad. Sci. USA 88, 4671-4674. Brunner, G., Gabrilove, J., Rifkin, D.B., and Wilson, E.L. (1991). Phospholipase C release of basic fibroblast growth factor from human bone marrow cultures as a biologically active complex with a phosphatidylinositol-anchored heparan sulfate proteoglycan. J. Cell Biol. 114, 1275-1283. Bugsby, W.H., Jr., Klepper, D.G., and Clemmons, D.R. (1988). Purification of a 31,000-dalton insulin-like growth factor binding protein from human amniotic fluid: isolation of two forms with different biological activities. J. Biol. Chem. 263, 14203-14270. Clemmons, D.R., Underwood, L.E., Chatelain, P.G., and Van Wyck, J.J. (1983). Liberation of immunoreactive somatomedin-C from its binding protein by proteolytic enzymes and heparin. J. Clin. Endocrinol. and Metab. 56, 384-389. Cohen, K.L., and Nissley, S.P. (1976). The serum half-life of somatomedin activity: evidence for growth factor dependence. Acta. Endocrinol. 83, 243-247. Colletta, A.A., Wakefield, L.M., Howell, F.V., Van Roozendaal, K.E.P., Danielpour, D., Ebbs, S.R., Sporn, M.B., and Baum, M. (1990). Antioestrogens induce the secretion of active TGF-,B from human fibroblasts. Br. J. Cancer 60, 405-409. Damon, D.H., Lobb, R.R., D'Amore, P.A., and Wagner, J.A. (1989). Heparin potentiates the action of acidic fibroblast growth factor by prolonging its biological half life. J. Cell. Physiol. 138, 221-226.

Molecular Biology of the Cell

The Extracellular Regulation of Growth Factor Action

Danielpour, D., Kim, K.-Y., Dart, L.L., Watanabe, S., Roberts, A.B., and Sporn, M.B. (1989). Sandwich enzyme-linked immunosorbent assays (SELISAs) quantitate and distinguish two forms of transforming growth factor-beta (TGF-1l and TGF-,B2) in complex biological fluids. Growth Factors 2, 61-71. Danielpour, D., and Sporn, M.B. (1990). Differential inhibition of transforming growth factor beta 1 and beta 2 activity by a2-macroglobulin. J. Biol. Chem. 265, 6973-6977. Dennis, P., and Rifkin, D.B. (1991). Cellular activation of latent transforming growth factor-# requires binding to the cation-independent mannose-6-phosphate/insulin-like growth factor type II receptor. Proc. Natl. Acad. Sci. USA 88, 580-584. Dennis, P.A., Saksela, O., Harpel, P., and Rifkin, D.B. (1989). Alpha2macroglobulin is a binding protein for basic fibroblast growth factor. J. Biol. Chem. 264, 7210-7216. Derynck, R., Jarrett, J.A., Chen, E.Y., Eaton, D.H., Bell, J.R., Assoian, R.K., Roberts, A.B., Sporn, M.B., and Goeddel, D.V. (1985). Human transforming growth factor-(# complementary DNA sequence and expression in normal and transformed cells. Nature 316, 701-705. Derynck, R., Roberts, A.B., Winkler, M.E., Chen, E.Y., and Goeddel, D.V. (1984). Human transforming growth factor-a: precursor structure and expression in E. coli. Cell 38, 287-297. Dolci, S., Williams, D.E., Ernst, M.K., Resnick, J.L., Brannan, C.I., Lock, L.F., Lyman, S.D., Boswell, H.S., and Donovan, P.J. (1991). Requirement for mast cell growth factor for primordial germ cell survival in culture. Nature 352, 809-811. Downing, J.R., Roussel, M.F., and Sherr, C.J. (1989). Ligand and protein kinase C downmodulate the colony-stimulating factor-i receptor by independent mechanisms. Mol. Cell. Biol. 9, 2890-2896. Elgin, R.G., Busby, W.H., and Clemmons, D.R. (1987). An insulinlike growth factor (IGF) binding protein enhances the biological response to IGF-1. Proc. Natl. Acad. Sci. USA 84, 3254-3258. Flanagan, J.G., Chan, D.C., and Leder, P. (1991). Transmembrane form of the kit ligand growth factor determined by alternative splicing and is missing in the 51 mutant. Cell 64, 1025-1035. Flaumenhaft, R., Moscatelli, D., and Rifkin, D.B. (1990). Heparin and heparan sulfate increase the radius of diffusion and action of basic fibroblast growth factor. J. Cell Biol. 111, 1651-1659. Flaumenhaft, R., Moscatelli, D., Saksela, O., and Rifkin, D.B. (1989). Role of extracellular matrix in the action of basic fibroblast growth factor: matrix as a source of growth factor for long-term stimulation of plasminogen activator and DNA synthesis. J. Cell. Physiol. 140, 75-81. Fukunaga, R., Sato, Y., Mizushimna, S., and Nagata, S. (1990). Three different mRNAs encoding human granulocyte colony-stimulating factor receptor. Proc. Natl. Acad. Sci. USA 87, 8702-8706. Gentry, L.E., and Nash, B.W. (1990). The pro domain of pre-protransforming growth factor beta 1 when independently expressed is a functional binding protein for the mature growth factor. Biochemistry 29, 6851-6857. Glick, A.B., Flanders, K.C., Danielpour, D., Yuspa, S.H., and Sporn, M.B. (1989). Retinoic acid induces transforming growth factor-,B2 in cultured keratinocytes and mouse epidermis. Cell Regul. 1, 87-97. Goodwin, R.G., Friend, D., Ziegler, S.F., Jerzy, R., Falk, B.A., Gimpel, S., Cosman, D., Dower, S.K., March, C.J., Namen, A.E., and Park, L.S. (1990). Cloning of the human and murine interleukin-7 receptors: demonstration of a soluble form and homology to a new receptor superfamily Cell 60, 941-951. Gordon, M.Y., Riley, G.P., Watt, S.M., and Greaves, M.F. (1988). Compartmentalization of a haematopoietic growth factor (GM-CSF) by glycosaminoglycans in the bone marrow microenvironment. Nature 326, 403-405. Vol. 3, October 1992

Gospodarowicz, D., and Cheng, J. (1986). Heparin protects basic and acidic FGF from inactivation. J. Cell. Physiol. 128, 475-484. Gray, A., Dull, T.J., and Ullrich, A. (1983). Nucleotide sequence of epidermal growth factor cDNA predicts a 128,000 molecular weight protein precursor. Nature 303, 722-725. Gray, P.W., Barrett, K., Chantry, D., Turner, M., and Feldman, M. (1990). Cloning of human tumor necrosis factor (TNF) receptor cDNA and expression of recombinant soluble TNF-binding protein. Proc. Natl. Acad. Sci. USA 87, 7380-7384. Hauschka, P.V., Chen, T.L., and Mavrakos, A.E. (1988). Polypeptide growth factors in bone matrix. In: Cell and Molecular Biology of Vertebrate Hard Tissue. Ciba Foundation Symposium 136. New York: John Wiley and Sons, 207-225. Higashiyama, S., Abraham, J.A., Miller, J., Fiddes, J.C., and Klagsbrun, M. (1991). A heparin-binding growth factor secreted by macrophagelike cells that is related to EGF. Science 251, 936-939. Hintz, R.L. (1990). Role of growth-hormone and insulin-like growth factor-binding proteins. Horm. Res. 33, 105-110. Huang, J.S., Huang, S.S., and Deuel, T. (1984). Specific covalent binding of platelet-derived growth factor to plasma alpha2-macroglobulin. Proc. Natl. Acad. Sci. USA 81, 342-346. Ito, F., Ito, S., and Shimizu, N. (1984). Transmembrane delivery of polypeptide growth factors bypassing the intrinsic cell surface receptors: synthesis and biological activity of a conjugate of epidermal growth factor with a2-macroglobulin. Cell Struct. Funct. 9, 105-115. Johnson, D.E., Lee, P.L., Lu, J., and Williams, L. (1990). Diverse forms of a receptor for acidic and basic fibroblast growth factor. Mol. Cell. Biol. 10, 4728-4736. Jullien, P., Berg, T.M., and Lawrence, D.A. (1989). Acidic cellular environments: activation of latent TGF-3 and sensitization of cellular responses to TGF-,B and EGF. Int. J. Cancer 43, 886-891. Kanzaki, T., Olofsson, A., Moren, A., Wernstedt, C., Hellman, U., Miyazono, K., Claesson-Welsh, L., and Heldin, C.H. (1990). TGF-,B1 binding protein: a component of the large latent complex of TGF-,B1 with multiple repeat sequences. Cell 61, 1051-1061. Kaufmann, U., Zapf, J., Torretti, B., and Froesch, E.R. (1977). Demonstration of a specific serum carrier protein of nonsuppressible insulinlike activity in vivo. J. Clin. Endocrinol. Metab. 44, 160-166. Kimura, K., Matsubara, H., Sogoh, S., Kita, Y., Sakata, T., Nishitani, Y., Watanabe, S., Hamaoka, T., and Fujiwara, H. (1991). Role of glycosaminoglycans in the regulation of T cell proliferation induced by thymic stroma-derived T cell growth factor. J. Immunol. 146, 26182624. Knabbe, C., Lippman, M.E., Wakefield, L.M., Flanders, K.C., Kasid, A., Derynck, R., and Dickson, R.B. (1987). Evidence that transforming growth factor-# is a hormonally regulated negative growth factor in human breast cancer cells. Cell 48, 417-428. Knauer, D.J., and Smith, G.L. (1980). Inhibition of biological activity of multiplication-stimulating activity by binding to its carrier protein. Proc. Natl. Acad. Sci. USA 77, 7252-7256. LaMarre, J., Hayes, M.A., Wollenberg, G.K., Hussaini, I., Hall, S.W., and Gonias, S.L. (1991). An alpha2-macroglobulin receptor-dependent mechanism for the plasma clearance of transforming growth factorbeta 1 in mice. J. Clin. Invest. 87, 39-44. LaRochelle, W.J., May-Siroff, M., Robbins, K.C., and Aaronson, S.A. (1991). A novel mechanism regulating growth factor association with the cell surface: Identification of a PDGF retention domain. Genes and Dev. 5, 1191-1199. Lawrence, D.A., Pircher, R., and Jullien, P. (1985). Conversion of a high molecular weight latent TGF-#3 from chicken embryo fibroblasts into a low molecular weight active TGF-#i under acidic conditions. Biochem. Biophys. Res. Commun. 133, 1026-1034.

1063

R. Flaumenhaft and D.B. Rifkin

Lee, D.C., Rose, T.M., Webb, N.R., and Todaro, G.J. (1985). Cloning and sequence analysis of a cDNA for rat transforming growth factora. Nature 313, 489-491. Leung, D.W., Cachianes, G., Kuang, W.-J., Goeddel, D.V., and Ferrera, N. (1989). Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 246, 1306-1312. Lyons, R.M., Gentry, L.E., Purchio, A.F., and Moses, H.L. (1990). Mechanism of activation of latent recombinant transforming growth factor 31 by plasmin. J. Cell Biol. 110, 1361-1367. Lyons, R.M., Keski-Oja, J., and Moses, H.L. (1988). Proteolytic activation latent transforming growth factor-,B from fibroblast conditioned medium. J. Cell Biol. 106, 1659-1665. Marcon, L., Fritz, M.E., Kurman, C., Jensen, J.C., and Nelson, D.L. (1988). Soluble Tac peptide is present in the urine of normal individuals and at elevated levels in patients with adult T cell leukemia. Clin. Exp. Immunol. 73, 29. Massague, J. (1990). Transforming growth factor-a. J. Biol. Chem. 265, 21393-21396. Matsuda, T., Hirano, T., Nagasawa, S., and Kishimoto, T. (1989). Identification of alpha2-macroglobin as a carrier protein for IL-6. J. Immunol. 142, 148-153. McCaffrey, T.A., Falcone, D.J., Brayton, C.F., Agarwal, L.A., Welt, F.G., and Weksler, B.B. (1989). Transforming growth factor-beta activity is potentiated by heparin via dissociation of the transforming growth factor-beta/alpha2-macroglobulin inactive complex. J. Cell Biol. 109, 441-448. Miyazono, K., and Heldin, C.H. (1989). Role for carbohydrate structure in TGF-#1 latency. Nature 338, 158-160. Moscatelli, D. (1987). High and low affinity binding sites for basic fibroblast growth factor on cultured cells: absence of a role for low affinity binding in the stimulation of plasminogen activator production by bovine capillary endothelial cells. J. Cell. Physiol. 131, 123-130. Mosley, B., Beckman, M.P., March, C.J., Idzerda, R.L., Gimpel, S.D., Vanden Bos, T., Friend, D., Alpert, A., Anderson, D., Jackson, J., Wignall, J.M., Smith, C., Gallis, B., Sims, J.E., Urdal, D., Widmer, M.B., Cosman, D., and Park, L.S. (1989). The murine interleukin-4 receptor: molecular cloning and characterization of secreted and membrane forms. Cell 59, 335-348. Mottala, C., MacDonald, R.G., Brackett, J.L., Mole, J.E., Anderson, J.K., and Czech, M.P. (1986). Purification and amino acid sequence of an insulin-like growth factor binding protein secreted by rat liver BRL-3A cells. J. Biol. Chem. 261, 11,180-11,188. Mroczkowski, B., Reich, M., Chen, K., Bell, G.I., and Cohen, S. (1989). Recombinant human epidermal growth factor precursor is a glycosylated membrane protein with biological activity. Mol. Cell. Biol. 9, 2771-2778. Mueller, S.G., Paterson, A.J., and Kudlow, J.E. (1990). Transforming growth factor a in arterioles: cell surface processing of its precursor by elastases. Mol. Cell. Biol. 10, 4596-4602. Nathan, C., and Sporn, M. (1991). Cytokines in context. J. Cell Biol. 113, 981-986. Novick, D., Engelmann, M., Wallach, D., and Rubinstein, M. (1989). Soluble cytokine receptors are present in normal human urine. J. Exp. Med. 170, 1409. O'Connor-McCourt, M.D., and Wakefield, L.M. (1987). Latent transforming growth factor-beta in serum. A specific complex with alpha2macroglobulin. J. Biol. Chem. 262, 14090-14099. Okuda, S., Languino, L.R., Ruoslahti, E., and Border, W. (1990). Elevated expression of transforming growth factor-, and proteoglycan production in experimental glomerulonephritis. Possible role in expression of the mesangial extracellular matrix. J. Clin. Invest. 86, 453-462.

1064

Oreffo, R.O.C., Mundy, G.R., Seyedin, S.M., and Bonewald, L.F. (1989). Activation of bone-derived latent TGF beta complex by isolated osteoclasts. Biochem. Biophys. Res. Commun. 158, 817-823. Ortega, S., Schaeffer, M.T., Soderman, D., DiSalvo, J., Linemeyer, D.L., Gimenez-Gallego, G., and Thomas, K.A. (1991). Conversion of cysteine to serine residues alters the activity, stability, and heparin dependence of acidic fibroblast growth factor. J. Biol. Chem. 266, 58425846. Pandiella, A., and Massague, J. (1991). Cleavage of the membrane precursor for transforming growth factor alpha is a regulated process. Proc. Natl. Acad. Sci. USA 88, 1726-1730. Papkoff, J., and Schryver, B. (1990). Secreted int-1 protein is associated with the cell surface. Mol. Cell. Biol. 10, 2723-2730. Perez, C., Albert, I., DeFay, Zachariades, N., Gooding, L., and Kriegler, M. (1990). A nonsecretable cell surface mutant of tumor necrosis factor (TNF) kills by cell-to-cell contact. Cell 63, 251-258. Pfeilschifter, J., Bonewald, L., and Mundy, G.R. (1990). Characterization of the latent transforming growth factor beta complex in bone. J. Bone Miner. Res. 5, 49-58. Piao, Y.F., Ichijo, H., Miyagawa, K., Ohashi, H., Takaku, F., and Miyazono, K. (1990). Latent form of transforming growth factor-31 acts as a potent growth inhibitor of a human erythroleukemia cell line. Biochem. Biophys. Res. Commun. 167, 27-32. Plowman, G.D., Green, J.M., McDonald, V.L., Neubauer, M.G., Disteche, C.M., and Todaro, G.J., and Shoyab, M. (1990). The amphiregulin gene encodes a novel epidermal growth factor-related protein with tumor-inhibitory activity. Mol. Cell. Biol. 10, 1969-1981. Raines, E.W., and Ross, R. (1992). Compartmentalization of PDGF on extracellular binding sites dependent on exon-6-encoded sequences. J. Cell Biol. 116, 533-543. Rapraeger, A.C., Krufka, A., and Olwin, B.B. (1991). Requirement of heparin sulfate for bFGF-mediated fibroblast growth and myoblast differentiation. Science 252, 1705-1708. Rathjen, P.D., Toth, S., Willis, A., Heath, J.K., and Smith, A.G. (1990). Differentiation inhibiting activity is produced in matrix-associated and diffusible forms that are generated by alternative promoter usage. Cell 62, 1105-1114. Robbins, K.C., Leaf, F., Pierce, J.A., and Aaronson, S.A. (1985). The v-sis/PDGF-2 transforming gene product localizes to cell membranes but is not a secretory protein. EMBO J. 4, 1783-1792. Roberts, R., Gallagher, J., Spooncer, E., Allen, T.D., Bloomfield, F., and Dexter, T.M. (1988). Heparan sulphate bound growth factors: a mechanism for stromal cell mediated haemopoiesis. Nature 332, 376378. Ronne, H., Anundi, H., Rask, L., and Peterson, P.A. (1979). Nerve growth factor binds to serum alpha2-macroglobulin. Biochem. Biophys. Res. Commun. 87, 330-336. Rosengart, T.K., Warren, J.V., Friesel, R., Clark, R., and Maciag, T. (1988). Heparin protects heparin-binding growth factor-1 from proteolytic inactivation in vitro. Biochem. Biophys. Res. Commun. 152, 432-440. Ruoslahti, E., and Pierschbacher, M.D. (1987). New perspectives in cell adhesion: RGD and integrins. Science 238, 491-497. Ruoslahti, E., and Yamaguchi, Y. (1991). Proteoglycans as modulators of growth factor activity. Cell 64, 867-869. Saksela, O., Moscatelli, D., Sommer, A., and Rifkin, D.B. (1988). Endothelial cell-derived heparan sulfate binds basic fibroblast growth factor and protects it from proteolytic degradation. J. Cell. Biol. 107, 743-751. Saksela, O., and Rifkin, D.B. (1990). Release of basic fibroblast growth factor-heparan sulfate complexes from endothelial cells by plasminogen activator-mediated proteolytic activity. J. Cell Biol. 110, 767-776. Molecular Biology of the Cell

The Extracellular Regulation of Growth Factor Action

Sasaki, M., and Yamada, Y. (1987). The laminin B2 chain has a multidomain structure homologous to the B1 chain. J. Biol. Chem. 262, 17111-17117. Sato, Y., and Rifkin, D.B. (1989). Inhibition of endothelial cell movement by pericytes and smooth muscle cells: activation of a latent transforming growth factor-fll-like molecule by plasmin during coculture. J. Cell Biol. 109, 309-315. Sato, Y., Tsuboi, R., Lyons, R.M., Moses, H.L., and Rifkin, D.B. (1990). Characterization of the activation of latent TGF-,B by co-cultures of endothelial cells and pericytes or smooth muscle cells: a self-regulating system. J. Cell Biol. 111, 757-763. Schall, T.J., Lewis, M., Koller, K.J., Lee, A., Rice, G.C., Wong, G.H., Gatanaga, T., Granger, G.A., Lentz, R., Raab, H., Kohr, W.J., and Goeddel, D.V. (1990). Molecular cloning and expression of a receptor for human tumor necrosis factor. Cell 61, 361-370. Shi, D.L., Savona, C., Gagnon, J., Cochet, C., Chambaz, E.M., and Feige, J.J. (1990). Transforming growth factor-beta stimulates the expression of alpha2-macroglobulin by cultured bovine adrenocortical cells. J. Biol. Chem. 265, 2881-2887. Sommer, A., and Rifkin, D.B. (1989). Interaction of heparin with human basic fibroblast growth factor: protection of the angiogenic protein from proteolytic degradation by a glycosaminoglycan. J. Cell. Physiol. 138, 215-220. Spom, M.B., and Roberts, A.B. (1990). The multifunctional nature of peptide growth factors. In: Peptide Growth Factors and Their Receptors, vol. 1, ed. M.B. Spom and A.B. Roberts, Berlin: Springer-Verlag 3-15.

Sporn, M.B., and Todaro, G.J. (1980). Autocrine secretion and malignant transformation of cells. New Engl. J. Med. 303, 878-880. Stein, J., Borzillo, G.V., and Rettenmier, C.W. (1990). Direct stimulation of cells expressin receptors for macrophage colony-stimulating factor (CSF-1) by a plasma membrane-bound precursor of human CSF-1. Blood 76, 1308-1314. Stroobant, P., Rice, A.P., Gullick, W.J., Cheng, D.J., Kerr, I.M., and Waterfield, M.D. (1985). Purification and characterization of vaccinia virus growth factor. Cell 42, 383-393. Symons, J.A., and Duff, G.W. (1990). A soluble receptor form of the interleukin-1 receptor produced by a human B cell line. FEBS Lett 272, 133-136. Tashiro, K., Hagiya, M., Nishizawa, T., Seki, T., Shimonishi, M., Shimizu, S., and Nakamuru, T. (1990). Deduced primary structure of

Vol. 3, October 1992

rat hepatocyte growth factor and expression of the mRNA in rat tissues. Proc. Natl. Acad. Sci. USA 87, 3200-3204.

Thorton, S.C., Mueller, S.N., and Levine, E.M. (1983). Human endothelial cells: use of purified heparin in cloning and long-term serial cultivation. Science 222, 623-625. Twardzik, D.R., Mikovits, J.A., Ranchalis, J.E., Purchio, A.F., Ellingsworth, L., and Ruscetti, F.W. (1990). Gamma-interferon-induced activation of latent transforming growth factor-$ by human monocytes. Annals N.Y. Acad. Sci. 593, 276. van Ooyen, A., and Nusse, R. (1984). Structure and nucleotide sequence of the putative mammary oncogene int-1; proviral insertions leave the protein-encoding domain intact. Cell 39, 233-240. Wakefield, L.M., Winokur, T.S., Hollands, R.S., Christopherson, K., Levinson, A.D., and Spom, M.B. (1990). Recombinant latent transforming growth factor fl1 has a longer plasma half-life in rats than active transforming growth factor f,l, and a different tissue distribution. J. Clin. Invest. 86, 1976-1984. Wallick, S.C., Figari, I.S., Morris, R.E., Levinson, A.D., and Palladino, M.A. (1990). Immunoregulatory role of transforming growth factor beta (TGF-fl) in development of killer cells: comparison of active and latent forms. J. Exp. Med. 172, 1777-1784. Weber, W., Gill, G.N., and Spiess, J. (1984). Production of an epidermal growth factor receptor-related protein. Science 224, 294-297. Wollenberg, G.K., LaMarre, J., Rosendale, S., Gonias, S.L., and Hayes, M.A. (1991). Binding of tumor necrosis factor to activiated forms of human plasma alpha2-macroglobulin. Am. J. Pathol. 138, 265-272. Wong, S.T., Winchell, L.F., McCune, B.K., Earp, H.S., Teixido, J., Massague, J., Herman, B., and Lee, D.C. (1989). The TGF-a precursor expressed on the cell surface binds to the EGF receptor on adjacent cells, leading to signal transduction. Cell 56, 495-506. Yamaguchi, Y., Mann, D., and Ruoslahti, E. (1990). Negative regulation of transforming growth factor-f by the proteoglycan decorin. Nature 346, 281-284. Yayon, A., Klagsbrun, M., Esko, J.D., Leder, P., and Omitz, D. (1991). Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high affinity receptor. Cell 64, 841-848. Zapf, J., Hauri, C., Waldvogel, M., and Froesch, E.R. (1986). Acute metabolic effect and half-lives of intravenously administered insulinlike growth factors I and II in normal and hypophysectomized rats. J. Clin. Invest. 77, 1768-1775. Zupan, A.A., Osbome, P.A., Smith, C.E., Siegel, N.R., Leimgruber, R.M., and Johnson, E.M., Jr. (1989). Identification, purification, and characterization of truncated forms of the human nerve growth factor. J. Biol. Chem. 264, 11714-11718.

1065