Bi-Ing Chang

Lectin-like domain of thrombomodulin binds to its specific ligand Lewis Y antigen and neutralizes lipopolysaccharide-induced inflammatory response

Thrombomodulin (TM), a widely expressing glycoprotein originally identified in vascular endothelium, is an important cofactor in the protein C anticoagulant system. TM appears to exhibit anti-inflammatory ability through both protein C-dependent and -independent pathways. We presently have demonstrated that recombinant N-terminal lectinlike domain of TM (rTMD1) functions as a protective agent against sepsis caused by Gram-negative bacterial infections. rTMD1 caused agglutination of Escherichia coli and Klebsiella pneumoniae and enhanced the macrophage phagocytosis of these Gram-negative bacteria. Moreover, rTMD1 bound to the Klebsiella pneumoniae and lipopolysaccharide (LPS) by specifically interacting with Lewis Y antigen. rTMD1 inhibited LPS-induced inflammatory mediator production via interference with CD14 and LPS binding. Furthermore, rTMD1 modulated LPS-induced mitogen-activated protein kinase and nuclear factor-kappaB signaling pathway activations and inducible nitric oxide synthase expression in macrophages. Administration of rTMD1 protected the host by suppressing inflammatory responses induced by LPS and Gram-negative bacteria, and enhanced LPS and bacterial clearance in sepsis. Thus, rTMD1 can be used to defend against bacterial infection and inhibit LPS-induced inflammatory responses, suggesting that rTMD1 may be valuable in the treatment of severe inflammation in sepsis, especially in Gram-negative bacterial infections.

Functional mutation in the promoter region of thrombomodulin gene in relation to carotid atherosclerosis

Thrombomodulin is an important endothelial anticoagulant protein that decreases thrombin activity and activates protein C. Our recent study has shown that the G-33A promoter mutation of thrombomodulin gene is associated with coronary artery disease. This study was conducted to determine whether the G-33A mutation in the promoter region of thrombomodulin gene is a genetic risk factor for ischemic stroke or carotid atherosclerosis. The functional significance of this mutation was also evaluated. We recruited 333 patients (mean age 64 years, 59% male) with ischemic stroke and 257 age- and sex-matched controls. In all study participants, carotid atherosclerosis was assessed by Duplex scanning, and thrombomodulin G-33A promoter mutation was detected by single-strand conformation polymorphism. Luciferase reporter gene assay was used to assess the influence of this mutation on thrombomodulin promoter activity. There was no significant difference in the thrombomodulin G-33A mutation frequency (GA+AA genotypes) between the stroke and the control groups (18.3 vs. 24. 1%, P=0.105). The G-33A mutation frequency was also similar between the study participants with and without carotid atherosclerosis (22.2 vs. 19.8%, P=0.550). When only younger subjects (age </=60 years) were included in the analysis, however, we found the mutation occurred more frequently in participants with carotid atherosclerosis (33.3 vs. 17.3%, odds ratio [OR]=2.38, 95% confidence interval [CI]=1.16-4.90, P=0.027). Multiple logistic regression analyses showed that only diabetes mellitus (OR=3.11, 95% CI=1.33-7.30, P=0.009) and G-33A mutation (OR=2.46, 95% CI=1.14-5.29, P=0.021) were associated independently with carotid atherosclerosis in younger subjects. As assessed by luciferase reporter gene assays, the contructs bearing the G-33A mutation showed a significant decrease (36+/-12%) in transcriptional activity in comparison with the wild type constructs. Our findings suggest that G-33A mutation reduces the thrombomodulin promoter activity and is associated with carotid atherosclerosis in younger subjects.

Antibodies against dengue virus E protein peptide bind to human plasminogen and inhibit plasmin activity

Both mice and rabbits immunized with dengue virus E protein peptide spanning amino acids 100–119 (D4E) produced antibodies that reacted not only with the D4E peptide itself but also with human plasminogen, as shown by ELISA and Western blot. Sera from dengue virus‐hyperimmunized mice and dengue patients also contained antibodies against D4E and plasminogen. Furthermore, such sera all contained plasmin inhibitory activity. Using affinity‐purified anti‐D4E antibodies and free D4E peptide for competitive inhibition, we demonstrated that the inhibition of plasmin activity was due to anti‐D4E antibodies rather than other substances in the sera. Taken together, these results suggest dengue virus E protein amino acids 100–119 are a cross‐reactive immunogenic region, and antibodies against this region may interfere with human fibrinolysis.

Lipocortin 1 co-associates with cytokeratins 8 and 18 in A549 cells via the N-terminal domain

An affinity chromatography strategy was used to search for proteins in A549 cells which interact with the N-terminus of lipocortin 1 (annexin 1). Using the biologically active fragment Lc13-25 as the affinity ligand, two proteins of molecular weight (m.w.) 52 and 48kDa were extracted. Affinity blots of these proteins bound iodinated Lc13-25. Partial tryptic digests of these proteins were analysed by matrix assisted laser desorption mass spectrometry and found to display fragmentation patterns with a strong similarity to those of cytokeratin 8 and 18 respectively. Subsequent blotting with a panel of specific cytokeratin antibodies strongly supported the idea that the two proteins were cytokeratin 8 and cytokeratin 18. Cytokeratin 8 was isolated from A549 cells in intermediate filament (IF) preparations which were also found to contain lipocortin 1 as a potential intermediate filament associated protein (IFAP). This association persisted throughout cycles of IF assembly and disassembly. Dual-labelling immuno-histochemistry in A549 cells showed strong co-localization of lipocortin 1 and cytokeratin 8. The implications of this finding are discussed in the light of the biological activity and possible function of lipocortin 1.

Activation of human and bovine plasminogens by the microplasmin and streptokinase complex

Human microplasmin is a catalytically active fragment of human plasmin. It consists of a 31-residue C-terminal peptide derived from the A chain bound through two disulfide bonds to the intact B chain of plasmin. It has similar amidolytic and proteolytic activities as the native human Lys-plasmin on a molar basis. Human microplasmin can form a complex with streptokinase, in a one to one stoichiometry, like the native human Lys-plasmin. The stoichiometric human microplasmin and streptokinase complex is an efficient activator of bovine plasminogen which can not be activated by streptokinase alone. The formation of human microplasmin.streptokinase complex was also directly demonstrated by a gel filtration column chromatography. Moreover, bovine plasminogen can not be activated by a mixture of bovine or porcine microplasmin and streptokinase. The equimolar complex of human microplasmin.streptokinase, human Lys-plasmin.streptokinase, or streptokinase alone has the same activator activity toward human Lys-plasminogen. The human microplasmin.streptokinase complex, however, has a significantly higher activator activity than human Lys-plasmin.streptokinase complex or streptokinase alone toward human Glu-plasminogen. The direct interaction between streptokinase and light chain domain of human plasmin is demonstrated in the complex formation. The difference in the activator activities of plasmins from various animal sources in complex with streptokinase therefore might be due to the difference in the compositions of light chains of plasmins.

Regulation of plasminogen activator inhibitor activity in endothelial cells by tissue-type plasminogen activator*

Summary Tissue-type plasminogen activator (t-PA) may stimulate the expression of plasminogen activator inhibitor type 1 (PAI-1) mRNA in cultured human umbilical vein endothelial cells. The PAI-1 antigen in the conditioned medium of cells, pretreated with t-PA, was less than the control, probably due to the formation and degradation of the t-PA·PAI-1 complex. However, the PAI activity of the t-PA-pretreated cells reached the same level as control group, 24 h after the residual t-PA activity was rapidly neutralized by the newly synthesized PAI-1. To test the release of PAI-1 from the substratum of the endothelial cells, the cultured cells were metabolically prelabeled with 35S-methionine for 3h and then treated with t-PA. The 35S-PAI-1 of 46 kDa was found in the substratum and culture medium of cultured endothelial cells as analyzed by the SDS-PAGE after immunoprecipitation. During the treatment of endothelial cells with t-PA, the PAI-1 of 46 kDa in the cell substratum disappeared, and the 110kDa t-PA·PAI-1 complex, the 81 kDa degraded t-PA·PAI-1, and the 44 kDa degraded PAI-1 products concomitantly appeared in the conditioned medium instead. In summary, t-PA can regulate the fibrinolytic activity of endothelial cells by enhancing PAI-1 mRNA biosynthesis and release PAI-1 from the substratum to neutralize t-PA activity. The PAI-1 which released into the medium was immediately converted to the inactive latent form in the absence of active t-PA.