Jararhagin disruption of endothelial cell anchorage is enhanced in collagen enriched matrices

Hemorrhage is one of the most striking effects of bites by viper snakes resulting in fast bleeding and ischemia in affected tissues. Snake venom metalloproteinases (SVMPs) are responsible for hemorrhagic activity, but the mechanisms involved in SVMP-induced hemorrhage are not entirely understood and the study of such mechanisms greatly depends on in vivo experiments. In vivo, hemorrhagic SVMPs accumulate on basement membrane (BM) of venules and capillary vessels allowing the hydrolysis of collagen IV with consequent weakness and rupture of capillary walls. These effects are not reproducible in vitro with conventional endothelial cell cultures. In this study we used two-dimension (2D) or three-dimension (3D) cultures of HUVECs on matrigel and observed the same characteristics as in ex vivo experiments: only the hemorrhagic toxin was able to localize on surfaces or internalize endothelial cells in 2D cultures or in the surface of tubules formed on 3D cultures. The contribution of matrigel, fibronectin and collagen matrices in jararhagin-induced endothelial cell damage was then analyzed. Collagen and matrigel substrates enhanced the endothelial cell damage induced by jararhagin allowing toxin binding to focal adhesions, disruption of stress fibers, detachment and apoptosis. The higher affinity of jararhagin to collagen than to fibronectin explains the localization of the toxin within BM. Moreover, once located in BM, interactions of jararhagin with α2β1 integrin would favor its localization on focal adhesions, as observed in our study. The accumulation of toxin in focal adhesions, observed only in cells grown in collagen matrices, would explain the enhancement of cell damage in these matrices and reflects the actual interaction among toxin, endothelial cells and BM components that occurs in vivo and results in the hemorrhagic lesions induced by viper venoms.

Collagen binding is a key factor for the hemorrhagic activity of snake venom metalloproteinases

Snake venom metalloproteinases (SVMPs) are multifunctional enzymes involved in several symptoms following snakebite, such as severe local hemorrhage. Multidomain P-III SVMPs are strongly hemorrhagic, whereas single domain P-I SVMPs are not. This indicates that disintegrin-like and cysteine-rich domains allocate motifs that enable catalytic degradation of ECM components leading to disruption of capillary vessels. Interestingly, some P-III SVMPs are completely devoid of hemorrhagic activity despite their highly conserved disintegrin-like and cysteine-rich domains. This observation was approached in the present study by comparing the effects of jararhagin, a hemorrhagic P-III SVMP, and berythractivase, a pro-coagulant and non-hemorrhagic P-III SVMP. Both toxins inhibited collagen-induced platelet aggregation, but only jararhagin was able to bind to collagen I with high affinity. The monoclonal antibody MAJar 3, that neutralizes the hemorrhagic effect of Bothrops venoms and jararhagin binding to collagen, did not react with berythractivase. The three-dimensional structures of jararhagin and berythractivase were compared to explain the differential binding to collagen and MAJar 3. Thereby, we pinpointed a motif within the Da disintegrin subdomain located opposite to the catalytic domain. Jararhagin binds to both collagen I and IV in a triple helix-dependent manner and inhibited in vitro fibrillogenesis. The jararhagin-collagen complex retained the catalytic activity of the toxin as observed by hydrolysis of fibrin. Thus, we suggest that binding of hemorrhagic SVMPs to collagens I and IV occurs through a motif located in the Da subdomain. This allows accumulation of toxin molecules at the site of injection, close to capillary vessels, where their catalytic activity leads to a local hemorrhage. Toxins devoid of this motif would be more available for vascular internalization leading to systemic pro-coagulant effects. This reveals a novel function of the disintegrin domain in hemorrhage formation.

Apoptosis in v-myc-transfected MSU-1.1 fibroblasts is induced by cell-matrix contact and differs from that of normal dermal fibroblasts

In order to characterize a fibroblast cell line representing normal human skin fibroblasts in three-dimensional cultures, we compared the fibroblast line MSU-1.1, derived from human foreskin and immortalized by v-myc, to primary human dermal fibroblasts (NDF). Our results demonstrate that in contrast to NDF, all MSU-1.1 fibroblasts die within 3-4 d when cultured within three-dimensional contractile collagen matrices. Also, in contrast to NDF. MSU-1.1 cells die markedly in anchored collagen gels as well. Death is due to apoptosis and is attenuated by addition of antibodies against collagen-recognizing receptors alpha1beta1 and alpha2beta1. Apoptosis of NDF in collagen lattices was repressed by an inhibitor of caspase-1, which was ineffective on apoptosis of MSU-1.1. Further, apoptosis by MSU-1.l fibroblasts was also observed in anchored, i.e., restrained collagen lattices, an environment that supports proliferation of NDF.

Contraction-Dependent Apoptosis of Normal Dermal Fibroblasts

The mechanisms underlying the contraction-dependent apoptosis of primary fibroblasts are of prime importance in understanding anchorage-dependent survival/apoptosis of dermal fibroblasts. As integrins are essential extracellular matrix receptors in fibroblasts, their role in anchorage-dependent apoptosis/survival of fibroblasts was analyzed. Primary human fibroblasts displayed a marked reduction of apoptosis in mechanically relaxed collagen matrices in the presence of adhesion-blocking antibodies against alpha1beta1 or alpha2beta1. Anti-alphavbeta3 antibodies had a considerably weaker effect. In additional experiments RD cells, which lack alpha2 integrin, displayed no apoptosis in mechanically relaxed collagen matrices. Their susceptibility to apoptosis was restored after transfection with functional alpha2 integrin, and it could be blocked again by adhesion-blocking antibodies against alpha2beta1 integrin. Therefore we conclude that apoptosis of human primary fibroblasts in contractile collagen matrices is - at least in part - inhibited by adhesion-blocking anti-integrin antibodies, suggesting that the mode of apoptosis in this case is different from anoikis. Further, apoptosis in a mechanically relaxed collagen matrix could be abrogated by depolymerization of F-actin using cytochalasin D and also by disturbing actin-myosin interaction using 2,3-butanedione monoxime, indicating a possible dependence of apoptosis on mechanical forces and/or cell shape.

Normal human primary fibroblasts undergo apoptosis in three-dimensional contractile collagen gels

Apoptosis of primary fibroblasts was observed in vivo during wound healing in skin and is expected to occur in other organs as well; however, the environmental signal for induction of apoptosis in fibroblasts and the putative influence of cell-matrix interactions on the regulation of apoptosis remain to be identified. Here we provide evidence for the role of fibrillar collagen in this process, and demonstrate that normal human primary fibroblasts embedded in contractile collagen gels undergo apoptosis as shown by the appearance of cytoplasmatic histone-associated DNA fragments starting at day 1 of culture with a peak around days 2-4. This induction of apoptosis in primary fibroblasts seems to be specific for contractile collagen gels, because apoptosis of primary fibroblasts was neither observed in cells grown on culture dishes or on plastic dishes coated with collagen, nor observed in cells seeded in either anchored collagen gels or contractile fibrin gels. We therefore conclude that a distinct environment such as a contractile collagen matrix determines the susceptibility of normal primary fibroblasts to apoptosis.