Wounding or proteolysis may be envisioned to perforate the basement membrane, a relatively thin scaffold approximately several hundred nanometers thick. Subsequently, EC would have to breach the basement membrane and migrate only small distances or, depending on the circumstances, even not at all, to encounter fibrillar collagens or fibrin (Figure 2, lower left and right panels). Thus, in many tissues the EC layer of the microvasculature or small vessels exist within several mm of light to sometimes heavy arrays of extracellular fibrillar collagens (e.g., Figure 1). During the wounding process, or in the vicinity of some tumors, fibrin may be heavily deposited throughout the tissue, and there may be cases where fibrin pools literally form adjacent to the resident ECs. Fibrillar collagens and fibrin each exert powerful proangiogenic activities on EC and are thus considered to comprise the second phase of angiogenesis regulation, and function to catalyze capillary tube construction (Figure 2, lower right panel). Because angiogenesis can occur during development in tissues lacking fibrillar collagen or fibrin (e.g., in the brain), we speculate that angiogen-esis is not absolutely dependent on the presence of such polymers. Instead, we propose that fibrillar collagens or fibrin may greatly accelerate or catalyze capillary tube formation during instances when rapid revascularization is required to support tissue survival and regeneration, as in wound healing.
Type I collagen is the most abundant protein in the human body and in other vertebrates. It is synthesized by vascular cells and most other cell types and is a common and, sometimes, predominant extracellular component of many interstitial tissues. Type I collagen is synthesized as a soluble triple-helical procollagen precursor of @ 300 Kd, is proteolytically processed upon secretion, and assembles in a staggered fashion into the type I collagen fibril, the form most commonly found in tissues. A role for type I collagen in angiogenesis has been known for many years; initial observations showed that inhibition of collagen metabolism and cross-linking disrupted angiogenesis in vivo, and subsequently, type I collagen was shown to be an ideal angiogenic scaffold in vitro. Many investigators have developed in vitro collagen-induced angiogenesis assays. The most common type involves culturing EC within a collagen gel sandwich into which cells migrate, proliferate, undergo cell-to-cell interactions, and then form capillary-like tubes, usually within two to three days. A modification of this system was devised by our group, where an apical collagen gel is placed on a confluent EC monolayer in the presence of a serum-free medium, and tubes form as rapidly as 12 hours. Because morphogenesis occurs in the absence of significant cell migration and proliferation, we consider this to be an endpoint or tube formation assay. Using this system, we have shown that angiogenesis in response to a type I collagen gel is surprisingly specific and relies largely on interactions between the EC a2pi integrin receptor and only one of the three potential glycine-phenylalanine-hydroxyproline-glycine-glutamic acid-arginine (GFPGER) integrin-binding sites on the collagen fibril. This stands although the collagen fibril contains numerous sites for potential interactions with cell surfaces and their associated molecules, including those for the binding of aipi or a2pi integrins, discoidin domain receptors, various proteoglycans, and fibronectin. Further evidence from our work suggests that a2pi-collagen liga-tion correlates with activation of the intracellular signaling molecule p38MAPK and downregulation of FAK, associated with sites of focal adhesion-matrix interactions. Ongoing work attempts to determine if these signaling pathways may transduce the signal from the level of cell-collagen interactions to the intracellular signaling events that induce tube formation.
Fibrin is an extracellular polymer generated during thrombogenesis from the precursor fibrinogen . Fibrinogen has an M of @ 340 Kd and comprises two identical sub-units, each of which consists of three distinct polypeptides called the Aa, Bp, and g chains; the two halves of the molecule are associated by several disulfide bonds at the N-terminal regions. During clot formation, thrombin cleaves the A and B fibrinopeptides, and the resultant product of @ 45 nM in length assembles in a half-staggered fashion into fibrin polymers or fibrils. As with type I collagen, fibrin has been shown by Jose Martinez and coworkers to be a potent inducer of angiogenesis in vitro. EC monolayers exposed apically to fibrin gels form capillary-like tubes within 24 hours. It was found that an interaction between the EC VE-cadherin and the beta 15-42 sequence of fibrin was necessary for fibrin-induced tube morphogenesis.
Common Features of Type I Collagen and Fibrin Polymers
When considering whether type I collagen and fibrin act in common ways to promote angiogenesis, it may be useful to compare the molecular and supramolecular features of these polymers. On a gross level, both polymers form classic gels, which when placed apically on EC monolayers induce capillary morphogenesis. At the level of the primary protein sequence, the two molecules share no significant homology; however, both are high-molecular-weight proteins that assemble into cable-like polymers exhibiting regular periodicities. The average type I collagen fibril diameter is 82.5 nm, the lateral repeat distance is 67 nm, and the spacing between monomers is approximately 1.0 to 1.5 nm. For fibrin, the average fibril diameter is 85 nm, the lateral repeat distance is 22.5 nm, and spacing between monomers is about 5 nm. It may be significant that the average fibril diameters are quite similar and that the other dimensions are well within an order of magnitude of each other.
We envision that these molecules may play at least two roles in promoting angiogenesis. First, both polymers present a rigid, linear substrate for cell interactions that may catalyze the side-by-side positioning of EC into tube-like arrays, a function reminiscent of that suggested by the matri-cal track theory of ECM-induced angiogenesis, put forth by Vernon and colleagues . This theory proposes that EC may exert tractional forces on a planar, relatively homogeneous fibrillar matrix scaffold, which results in a gathering of alignment of fibrillar matrix cables between cells. These cables in turn function as matrical tracks, which guide ECs to extend processes and join together to ultimately form a polygonal cellular or capillary network. In the case of the induction of angiogenesis by a collagen gel, it is possible that the EC must first act on the three-dimensional random array of collagen fibrils within the gel to gather or align multiple fibrils together, which in turn may promote linear cell-to-cell associations. In addition, we propose that a second distinct role these polymers may play is to promote the proper spatial interactions between specific EC surface receptors and regularly repeating ligand-binding sites on the polymers, resulting in cell surface receptor clustering and the ensuing activation of signaling pathways necessary for capillary morphogenesis. It can be speculated that this event may trigger EC responses such as capillary lumen formation and the expression of genes necessary to initiate and maintain capillary differentiation. In the case of type I collagen, the a2pi integrin is likely involved, and for fibrin, VE-cadherin may be the relevant receptor. Such receptor clustering may be dictated by as-yet-undetermined molecular repeat distances, which may be either common to these two polymers or unique to each type of receptor-ligand interaction. In this regard, we will briefly consider only the potential mechanism of a2pi integrin clustering by the type I collagen fibril.
Assuming anywhere from one to three of the proposed sites for a2pi-binding are available on the fibril surface, it may be calculated to contain from approximately 10 to 40 a2pi integrin receptor-binding sites across its width, and along its length, to contain such sites at 67-nm intervals. Because the region of the integrin receptor that engages type I collagen is @ 7.0 nm in diameter, across the fibril of @ 80 nm in width there is only enough space for a maximum of 10 receptors to occupy their ligand-binding sites; in this scenario the receptors clearly have the potential to be tightly arranged or clustered at the same cell surface location. On the other hand, receptors bound to sites available every 67 nm along the length of the fibril may be too sparsely distributed to promote their clustering and activation.
Another level of complexity that must be investigated is the matter of whether EC must interact with multiple collagen fibrils to initiate angiogenesis, and if so, how fibril arrangement may influence their EC interactions and capillary morphogenesis. Such information will contribute to the rational design of angiogenic polymers for various applications in tissue engineering, as discussed following. Last, it is worth addressing the fact that some other molecules, such as laminin and type IV collagen mixtures, have been reported to support tube formation in vitro, which at first glance seems inconsistent with the model described previously. However, in vivo, EC may never be exposed to high concentrations of only one or two basement membrane components, and moreover, in the intact basement membrane, many of the biologically active domains of the matrix scaffold may be inaccessible to EC. Finally, some ECM molecules, although incapable of supporting angiogenesis when present as components of the native macromolecular assembly, may be capable of engaging the appropriate cell surface receptors and thus promoting angiogenesis, when isolated and presented to EC at high concentrations.
Was this article helpful?
This ebook provides an introductory explanation of the workings of the human body, with an effort to draw connections between the body systems and explain their interdependencies. A framework for the book is homeostasis and how the body maintains balance within each system. This is intended as a first introduction to physiology for a college-level course.