Homandberg G A, Kramer-Bjerke J, Grant D, Christianson G, Eisenstein R
Biochim Biophys Acta. 1986 Nov 7;874(1):61-71. doi: 10.1016/0167-4838(86)90102-0.
Heparin-binding fragments derived from the amino- and carboxyl-terminal regions of human plasma fibronectin appear to be at least relatively specific potent inhibitors of the growth of bovine aortic endothelial cells in culture by as yet unknown mechanisms. In order to understand better the sites which subserve this activity, we have compared the relative potency of other major fragments of fibronectin, most of which have dissimilar properties and do not bind heparin. We have also proteolytically digested and chemically modified the most potent of these fragments, the amino-terminal 29-kDa fragment, in order to test whether structural alterations that affect heparin-binding also affect the inhibitory property. Not all chemical modifications that abolished heparin-binding also abolished endothelial cell growth. Neither an amino-terminal 20-kDa nor a carboxyl-terminal 8-kDa subfragment of the 29-kDa fragment bound heparin; however, both were as inhibitory as native 29-kDa fragment. Reduction of the disulfides of the 20-kDa and 8-kDa fragments did not abolish inhibitory activity. We therefore conclude that the activity is not strictly conformation-dependent and that although the inhibitory activity is distributed throughout the 29-kDa segment, it can be expressed by an 8-kDa carboxyl-terminal segment containing residues of the last Type I loop structure.