The consequences of the interaction of angiogenic growth factors with heparin/HS in endothelium can result either in the inhibition or in the enhancement of their biological activity. Generally speaking, the binding of the growth factor to cell-associated HSPGs causes its storage in the ECM with consequent increase in local concentration, prolonged half-life, and decrease in the radius of diffusion. This will favor growth factor oligomerization, TK receptor interaction and signaling.

Conversely, the binding of the growth factor to soluble HSPGs may antagonize all or part of these effects, resulting in a potential angiostatic action. Exceptions can be represented by heparin/HS-derived oligosaccharides whose size and sequence allow a ternary interaction among soluble GAG, growth factor, and TK receptor. Thus, the biological activity of angiogenic growth factors on endothelial cells is controlled by a complex interplay among free and cell-associated heparin/HS. In this scenario, natural and synthetic heparin-related angiostatic compounds play their pharmacological action.

These considerations point to the importance of the accurate definition of the molecular bases of protein-heparin/HS interaction for the design of molecules endowed with angiogenic agonist or antagonist activity. As stated above, specific oligosaccharide sequences appear to be involved in the interaction with different growth factors and their receptors. In theory, this may allow the specific tailoring of molecules with selective action towards defined heparin-binding growth factors.

To this respect, the synthetic angiostatic compounds described above are quite "non-selective", being able to affect the activity of a variety of growth factors. The lack of selectivity may be responsible, at least in part, for the side effects of these compounds, including their anticoagulant activity. On the other hand, because of the apparent redundancy of angiogenic growth factors possibly involved in different pathological conditions, the multiple-target action of the heparin/HS-related angiostatic molecules characterized so far may represent a therapeutical advantage. This may not be the case for synthetic GAG analogs aimed to potentiate the activity of a defined growth factor for which the selectivity of action appears to be of paramount importance.

New methods are emerging for modeling carbohydrate interaction with protein combining site (164). These strategies will help to optimize the chemical structure of saccharide species and pharmacological analogues in order to wide their therapeutic window and to increase their specificity of action.
 

ACKNOWLEDGMENTS

The work from our laboratory described in this review was supported in part by M.U.R.S.T. (quota 60% to M.R and to M.P.) and by the Associazione Italiana per la Ricerca sul Cancro (Special Project Angiogenesis), Istituto Superiore di Sanità (AIDS Project), and European Communities (Human Capital Mobility Project "Mechanisms for the Regulation of Angiogenesis") to M.P.