Extracellular Sequestration and Release of FGF2

Another major issue associated with the release pathway of FGF-1 and FGF-2 is the nature of their sequestration by HSPG in the extracellular matrix and basement membrane of endothelial cells from different sources (79,134-136). The model in which FGFs might be stored in a latent form that can be locally activated by heparin or heparanase, or by exposure to their high-affinity receptors, offers an attractive possibility for the regulation of FGF activities in normal and pathological situations.

The authors' studies on the control of cell proliferation by its local environment focused on the interaction of cells with the ECM produced by cultured corneal and vascular endothelial cells (53,79). This ECM closely resembles the subendothelium in vivo in its morphology and molecular composition. Vascular endothelial cells, plated in contact with the subendothelial ECM, no longer require the addition of soluble FGF in order to proliferate (79). This observation, together with the presence of HS as a major glycosami-noglycan (GAG) in the subendothelial ECM, raised the possibility that ECM contains heparin-binding growth factors that are tightly bound and stabilized by the ECM-HS. Indeed, FGF-2 was extracted from the subendothelial ECM produced in vitro (79) and in vivo (136), suggesting that ECM may serve as a reservoir for FGF-2. Immunoreactive FGF-2 was identified in BM of the cornea (136) and in BM underlying endothelial (137) and epithelial (138) cells. It was suggested that intracellular FGF may be released into the ECM in response to mild cell damage and certain stress conditions associated with tissue injury, irradiation, inflammation, shear force, heat shock, and tumor necrosis. High amounts of FGF-2 are contained in platelets and macrophages. Since these cells release their entire cell constituents upon activation, it is conceivable that FGFs are released by these cells during inflammation, hypoxia, or ischemia. The released factor may then be sequestered from its site of action by means of binding to HS (139), and possibly to FGF-2 receptor proteins in the ECM (47), and saved for emergencies, such as wound repair and neovascularization (53,140). It appears that FGF-2 binds primarily to HS in ECM and basement membrane, since the majority of the bound growth factor was displaced by heparin, HS, or HS-degrading enzymes (i.e., heparanase), but not by unrelated GAGs or GAG-degrading enzymes (139-141).

The involvement of sulfate groups in FGF-2 sequestration by the subendothelial ECM was studied by growing the ECM-producing cells in the presence of chlorate, a potent inhibitor of sulfation (61). Both the FGF-2 content and growth-promoting activity of sulfate depleted ECM were less than 10% of native ECM, indicating that sulfate moieties of HS are involved in FGF-2 sequestration and growth promoting activity of the ECM. FGF-2 is also sequestered by HS on cell surfaces, as revealed by immunohistochemistry (137), release by glycosyl phosphatidylinositol specific phospholipase C (PI-PLC) (142,143), and displacement by heparin from the luminal surface of blood vessels (144). Heparanase, an endoglycosidase that specifically degrades HS, was found to be a most efficient specific releaser of active FGF-2 from ECM (145). The authors' studies suggest that heparanase activity expressed by metastatic tumor cells and activated cells of the immune system may not only function in cell migration and invasion, but at the same time may elicit an indirect neovascular response by means of releasing the ECM-resident FGF (53,140). Apart from HS-degrading enzymes, active FGF-2 is released from ECM by thrombin (146) and by plasmin (147), as a noncovalent complex with HSPG.

Despite the ubiquitous presence of FGF-2 in the ECM and basement membranes of tissues, EC proliferation in these tissues is usually very low, with turnover time measured in years. This raised the question of how FGF-2, and possibly other growth factors, are prevented from acting on the vascular endothelium continuously, and in response to what signals do they become available for stimulation of capillary EC proliferation? One possibility is that FGF-2 may be stored in a latent inactive form or bound to truncated high-affinity FGF receptors, identified in the BM of retinal vascular endothelial cells

(47). Restriction of FGFs in ECM and BM prevents their systemic action on the vascular endothelium, thus maintaining a very low rate of EC turnover and vessel growth. On the other hand, release of FGF-2 from storage in ECM may elicit a localized EC proliferation and neovascularization in processes such as wound healing, inflammation, and tumor development (53,140,148).

Based on a recent study on the involvement of FGF-BP in tumor growth and angio-genesis (133), it is conceivable that, during tumor progression, FGF-BP expression is upregulated, and the protein is secreted into the microenvironment. There it can bind FGF-2 molecules that are inactive because of their immobilization within the ECM. The displaced (soluble), and now biologically active FGF-2 molecules are free to mediate various functions, such as angiogenesis (149). In other words, secretion of FGF-BP by tumors may flip the angiogenic switch during tumor progression. It should also be noted that, under the normal in vivo situation, the lack of EC proliferation in response to ECM-resident FGF-2 may simply be caused by the closely apposed and contact inhibited configuration of the cells. Once the cells are released from contact inhibition (i.e., in response to stress conditions and tissue injury), cells that remain bound to the ECM, but are no longer growth arrested, become susceptible to stimulation by the ECM-bound FGF-2, until they regain their characteristic contact inhibited cobblestone morphology.

0 0

Post a comment