We have carried out a large-scale identification and characterization of human genes that activate the NF-kappaB and MARK signaling pathways. We constructed full-length cDNA libraries using the oligo-capping method and prepared an arrayed cDNA pool consisting of 150 000 cDNAs randomly isolated from the libraries. For analysis of the NF-kappaB signaling pathway, we introduced each of the cDNAs into human embryonic kidney 293 cells and examined whether it activated the transcription of a luciferase reporter gene driven by a promoter containing the consensus NF-kappaB binding sites. In total, we identified 299 cDNAs that activate the NF-kappaB pathway, and we classified them into 83 genes, including 30 characterized activator genes of the NF-kappaB pathway, 28 genes whose involvement in the NF-kappaB pathways have not been characterized and 25 novel genes. We then carried out a similar analysis for the identification of genes that activate the MARK pathway, utilizing the same cDNA resource. We assayed 145 000 cDNAs and identified 57 genes that activate the MARK pathway. Interestingly, 27 genes were overlapping between the NF-kappaB and the MAPK pathways, which may indicate that these genes play cross-talking roles between these two pathways.

We have deduced the entire 575-amino acid sequence of the human thrombomodulin precursor from cDNA clones. The precursor starts with an 18-residue signal peptide domain, followed by the NH2-terminal domain, a domain with six epidermal growth factor-like structures, an 0-glycosylation site-rich domain, a 24-residue transmembrane domain and a cytoplasmic domain. Simian COS cells transfected with the expression vector pSV2 containing thrombomodulin cDNA synthesized immunoreactive and functionally active thrombomodulin.

Objective— We have recently reported that recombinant human soluble thrombomodulin (rTM) counteracted capillary leakage associated with engraftment, as well as sinusoidal obstructive syndrome after hematopoietic stem cell transplantation. These observations prompted us to explore whether rTM possessed cytoprotective effects on endothelial cells. Methods and Results— Exposure of human umbilical vein endothelial cells to rTM induced expression of antiapoptotic protein myeloid leukemia cell-1 through the activation of extracellular signal–regulated kinase in these cells. Additional studies found that exposure of human umbilical vein endothelial cells to cyclosporine A and FK506, an immunosuppressant used for the individuals receiving hematopoietic stem cell transplantation, induced apoptosis, which was attenuated when human umbilical vein endothelial cells were exposed to these agents in the presence of rTM. Further studies using deletion mutants of thrombomodulin (TM) identified that the epidermal growth factor domain of TM possessed cytoprotective effects. A single nucleotide substitution at codon 376 or 424 of TM, which impairs the ability of TM to produce activated protein C or bind to thrombin, respectively, did not hamper the cytoprotective effects of TM, which suggested that cytoprotective effects of rTM were distinctive from those of activated protein C. Conclusion— TM may be useful for prevention, as well as treatment of endothelial cell damage after hematopoietic stem cell transplantation.

Antithrombotic effect of recombinant human thrombomodulin in mice, both in vitro and in vivo, was studied. The soluble recombinant human thrombomodulin was expressed in Chinese hamster ovary cells and purified from the conditioned medium by a modification of the conventional method. Recombinant thrombomodulin prolonged thrombin clotting time for mouse plasma in a dose-dependent manner. Thrombin was injected into the lateral tail vein of mice and caused acute thromboembolism. All mice injected with thrombin died of thromboembolism; however, preinjection with recombinant human thrombomodulin neutralized the lethal effect of thrombin in a concentration-dependent manner. Histologic examination showed that fibrin deposits were found in all large and small arteries in the lung from mice injected with thrombin; however, fibrin deposits were not detected in any large arteries from the mouse preinjected with thrombomodulin.