No abstract

343 Background: Alternatively spliced Tissue Factor (asTF) is a soluble form of Tissue Factor (TF) that lacks the transmembrane domain and exhibits low pro-coagulant activity compared to decrypted full-length TF (Bogdanov et al, Nat Med 2003). Human asTF was recently shown to ligate integrins α6β1 and αVβ3 on endothelial cell surfaces, which triggered neovascularization in vitro, ex vivo, and in vivo in fVIIa- and/or PAR-2 independent fashion (van den Berg et al., PNAS 2009). This principally novel, non-proteolytic biologic activity of asTF was found to activate multiple integrin-linked kinases in human and murine macrovascular endothelial cells (EC), including those comprising the PI3K/Akt pathway. Engagement of PI3K/Akt signaling is known to activate NFkB – a transcription factor involved in upregulation of multiple leukocyte adhesion molecules that play a major role in various disease states, most notably atherogenesis and tumorigenesis. In this study, we sought to determine whether human asTF acts as an agonist on microvascular EC – the endothelial sub-type most relevant to monocyte egress from the circulation. Results: Primary cardiac and retinal human microvascular endothelial cells (micro-EC) expressed α6, β1, and β3 mRNA. Both micro-EC sub-types rapidly adhered to the recombinant N-terminally His-tagged human asTF: a 20+ fold increase in the number of adherent micro-EC was observed as early as 30 min after the cells were added (n=3,100 ng asTF vs BSA, p<0.0001). In both sub-types of micro-EC, anti-β1 integrin antibody completely blocked adhesion to asTF whereas anti-β3 and/or anti-αvβ3 antibody had no effect, likely due to the known scarcity of non-β1 integrins on the surface of human micro-EC (Wilson et al., IOVS 2003). We then examined the changes in gene expression elicited by asTF in micro-EC using Affymetrix Gene 1.0 ST microarrays. A marked upregulation of several cell adhesion molecules (CAMs) was observed in cardiac and retinal micro-EC including VCAM-1 and E-selectin, which was confirmed at the protein level by western blotting. We proceeded to investigate the functional significance of the upregulation of CAMs by performing adhesion assays using pre-labeled THP-1 cells. Under orbital shear stress conditions, a ∼76% increase in THP-1 adhesion was observed for cardiac micro-EC treated with asTF over control (n=5, p<0.001), and a ∼62% for retinal micro-EC (n=5, p<0.005). Addition of polymyxin B and/or non-charged agarose beads elicited no effect, whereas pre-treatment of asTF samples with Ni-charged beads and heat denaturation eliminated the effect, confirming that the observed findings are asTF-specific and not caused by LPS contamination. In the static assay and the laminar flow chamber assay performed under the flow rates found in postcapillary venules, the increase in THP-1 adhesion was 40% (n=3, p<0.001) and 250% (n=3, p<0.05), respectively, for cardiac micro-EC treated with asTF and 83% (n=3, p<0.001) and 290% (n=3, p<0.05), respectively, for retinal micro-EC. While performing laminar flow chamber assays, we observed relocation of some of the adhered THP-1 cells under the EC monolayer, which indicates that asTF may cause monocyte emigration in the absence of a chemokine gradient. To determine whether monocyte egress through asTF-stimulated micro-EC can also occur under a chemokine gradient, we performed transendothelial migration assays using transwell inserts with MCP-1 placed in the lower chamber. Here, we observed that asTF potentiated THP-1 migration through cardiac micro-EC by ∼206% (n=3, p<0.0001), and through retinal micro-EC – by ∼90% (n=3, p<0.005). Conclusions: We show for the first time that (i) human asTF ligates β1 integrin(s) on primary human micro-EC, (ii) this leads to transcriptional upregulation of CAMs, and (iii) this upregulation promotes adhesion and transendothelial migration of monocytic cells under physiologically relevant conditions. We note that the concentrations of asTF used here were in the range found in cervical cancer tissue (van den Berg et al., PNAS 2009). Our results expand the potential scope of asTF's contribution to atherogenesis and tumorigenesis. Evaluation of these findings in vivo is highly warranted as monocyte recruitment triggered by asTF may represent a novel nexus in atherosclerotic progression and/or tumor growth. Disclosures: No relevant conflicts of interest to declare.

1135 Background. Aside from hemostatic maintenance, Tissue Factor (TF) also plays a major role in such pathophysiological processes as thrombogenesis and cancer progression. TF protein occurs naturally in two forms: full length TF (flTF), a well–studied integral membrane glycoprotein that serves as an obligatory enzymatic co-factor of the serine protease FVIIa, and alternatively spliced TF (asTF), which lacks a transmembrane domain and can be secreted. asTF has a unique 40 amino acid C-terminus and can be detected, alongside flTF, in organized arterial thrombi (Bogdanov et al, Nat Med 2003 Apr; 9(4):458-62). Following the discovery of human asTF, the murine form of asTF (masTF) was identified and characterized (Bogdanov et al, J Thromb Haemost. 2006 Jan;4(1):158-67). Like human asTF, masTF lacks a transmembrane domain due to the exclusion of exon 5 from the primary transcript during its splicing, possesses a unique 93 amino acid C-terminus, and exhibits minimal coagulant potential. Most recently, hasTF was discovered to induce cell adhesion and angiogenesis via integrin ligation, independent of FVIIa, PAR-2 cleavage and/or any other proteolytic events (van den Berg et al, Proc Natl Acad Sci U.S.A. 2009 Nov 17;106(46):19497-502). It has yet to be investigated whether masTF exhibits non-proteolytic biologic activity analogous to that of hasTF. As murine models comprise the preferred in vivo platform in preclinical cardiovascular and cancer research, it is highly warranted to ascertain these possible functional properties of masTF. In this study, we performed an initial set of experiments to address this issue. Results. N-terminally His-tagged recombinant masTF mature protein was generated in E. Coli, purified, and assessed by Coomassie staining; masTF's identity was successfully verified by western blotting. Analogously to hasTF, masTF induced adhesion of murine endothelial cells (bEnd.3) in a time-dependent fashion. A 15-fold increase over BSA (n = 3; p < 0.0001) was observed as early as 1 hour after the experiment's onset; at 4 hours, bEnd.3 cells exposed to masTF displayed a 37-fold increase over BSA (p < 0.0005). We noted that masTF also induced bEnd.3 cells to display characteristic endothelial morphology, whereas BSA did not. We subsequently used a wound healing assay to ascertain whether masTF promotes directional migration of bEnd.3 cells, and used VEGF (100 ng/mL) as a positive control to assess the degree of potency. Per wound, the area of complete closure generated by masTF (50 nM) was ∼4.9-fold greater than that of the vehicle (n = 3; p < 0.0001), an effect similar to that elicited by VEGF (area of closure ∼7-fold greater than that of the vehicle; p < 0.0001). Microarray analysis of masTF-treated bEnd.3 cells (Affymetrix Gene 1.0 ST platform) revealed upregulation of the genes encoding multiple CXC chemokines, with CXCL2, CXCL10, and CXCL1 topping the list (R = 1.42, 1.37, and 1.32, respectively). Notably, expression of the major adhesion molecule VCAM-1 was also upregulated (R = 1.3), suggesting that masTF may promote interactions between murine endothelial cells and leukocytes. Indeed, we found that stimulating bEnd.3 cells with 50 nM masTF caused J774A.1 cells (murine monocytes/macrophages) to adhere with a 75% increased affinity (n = 3; p = 0.0001 vs. vehicle) when exposed to orbital shear conditions. Depletion of masTF from the medium by Ni-charged beads and denaturation by heat eliminated increased monocyte adhesion, while addition of polymyxin B and non-charged beads had no effect, confirming that the observed biologic phenomena were elicited by the masTF protein. Conclusions. We report, for the first time, that murine asTF appears to possess non-proteolytic biologic properties analogous to those of human asTF, which indicates that the alternatively spliced TF may be a general cell agonist eliciting changes in gene expression via integrin ligation. We are currently investigating whether the endothelial surface molecules interacting with masTF, as well as the intracellular signaling pathways activated by masTF in endothelial cells, are analogous to those engaged by hasTF. Disclosures: No relevant conflicts of interest to declare.

No abstract