Shiga Toxin Upregulates Tissue Factor Gene Expression in and Functional Tissue Factor On Endothelium in the Presence of Complement Activation

Abstract 1075 The pathophysiology of Shiga toxin (Stx)-related hemolytic uremic syndrome remains poorly understood. We hypothesize that tissue factor (TF) expression on vascular endothelium is central to the genesis of this disorder and is driven both by direct effects of Stx on endothelium activated by inflammatory cytokines derived from the action of Stx in the gastrointestinal tract, and by Stx-induced complement activation, which further augments TF expression. We used human umbilical vein endothelial cell monolayers to examine 1) gene expression of TF and TF pathway inhibitor (TFPI) with TNFα (20 ng/ml) ± Stx-1 (10 pM) compared to control, 2) total cellular and cell surface antigenic TF and TFPI, 3) TFPI secretion into supernatant, 4) factor Xa production, and 5) soluble levels of C5b-9 in complete media in response to 3–30 pM Stx-1 in the absence of cells. TF mRNA increased 2.82 ± 0.92-fold (N=13; p <0.0005) with TNFα alone vs. 1.25 ± 0.32-fold (N= 9; p =0.041) for Stx-1 alone (Fig. 1). TNFα plus Stx-1 yielded a 6.51 ± 3.48-fold increase (N=17; p <0.0005; Fig. 1). TFPI mRNA decreased with TNFα (p <0.001) andTNFα plus Stx-1 (p < 0.0005). Total cellular (p = 0.048) and cell surface (p = 0.027) TF antigen increased with TNFα, and no further with TNFα plus Stx-1. Total TFPI cellular (p = 0.0017) and cell surface (p = 0.030) antigen levels, and secretion of TFPI (p = 0.018) decreased with TNFα plus Stx-1. Median factor Xa production for TNFα plus Stx-1 vs. TNFα alone increased (p < 0.001) 3.24-fold. In the absence of cells, M199 with 10% human serum to which was added Stx-1 yielded levels of C5b-9 of 648 and 1102 ng/ml vs. 106 and 313 ng/ml in the presence of 25 mM EDTA (to block the classical and alternative pathways of complement), and intermediate levels in the presence of 15 mM EGTA and 7.5 mM Mg++ (to block the classical pathway). Similar results were found with M199 with 10% human plasma, collected using 0.2 U/ml FC in media of dalteparin. C5b-9 levels increased between 3 and 10 pM Stx-1, but did not increase further at 30 pM Stx-1 (Fig. 2). Finally, by means of immunofluorescence we also detected and measured C5b-9 on fixed cell monolayers, whether treated prior to fixation with TNFα alone or TNFα plus Stx-1. We conclude that Stx-1 augments functional TF on endothelium by increasing TF mRNA, and impairing synthesis, cell-surface association, and secretion of TFPI, while at the same time activating both the alternative and classical complement pathways. These findings support the development of therapeutic approaches to epidemic HUS that incorporate inhibitors of the TF pathway, inhibitors of the alternative and classical complement pathways, or both. Disclosures: No relevant conflicts of interest to declare.