Oxidation of a specific methionine in thrombomodulin by activated neutrophil products blocks cofactor activity. A potential rapid mechanism for modulation of coagulation.

Endothelial thrombomodulin (TM) plays a critical role in hemostasis as a cofactor for thrombin-dependent formation of activated protein C, a potent anticoagulant. Chloramine T, H2O2, or hypochlorous acid generated from H2O2 by myeloperoxidase rapidly destroy 75-90% of TM cofactor activity. Activated PMN, the primary in vivo source of biological oxidants, also rapidly inactivate TM. Oxidation of TM by PMN is inhibited by diphenylene iodonium, an inhibitor of NADPH oxidase. Both Met291 and Met388 in the six epidermal growth factor-like repeat domain are oxidized; however, only substitutions of Met388 lead to TM analogues that resist oxidative inactivation. We suggest that in inflamed tissues activated PMN may inactivate TM and demonstrate further evidence of the interaction between the inflammatory process and induction of thrombotic potential.

Human formyl peptide receptor ligand binding domain(s). Studies using an improved mutagenesis/expression vector reveal a novel mechanism for the regulation of receptor occupancy.

Recently, we reported the domain requirements for the binding of formyl peptide to its specific receptor. Based on experiments using receptor chimeras, we also postulated an importance for the amino-terminal domain of the receptor in ligand binding (Perez, H. D., Holmes, R., Vilander, L., Adams, R., Manzana, W., Jolley, D., and Andrews, W. H. (1993) J. Biol. Chem. 268, 2292-2295). We have begun to perform a detailed analysis of the regions within the formyl peptide receptor involved in ligand binding. To address the importance of the receptor amino-terminal domain, we substituted (or inserted) hydrophilic sequences within the amino-terminal domain, expressed the receptors, and determined their ability to bind ligand. A stretch of nine amino acids next to the initial methionine was identified as crucial for receptor occupancy. A peptide containing such a sequence specifically completed binding of the ligand to the receptor. Alanine screen mutagenesis of the second extracellular domain also identified amino acids involved in ligand binding as well as a disulfide bond (Cys98 to Cys176) crucial for maintaining the binding pocket. These studies provide evidence for a novel mechanism involved in regulation of receptor occupancy. Binding of the ligand induces conformational changes in the receptor that result in the apposition of the amino-terminal domain over the ligand, providing a lid to the binding pocket.

A transcriptionally defective long terminal repeat within an endogenous copy of mouse mammary tumor virus proviral DNA.

Mouse mammary tumor virus proviral DNA is endogenous to most inbred strains of mice but in many strains is not transcriptionally active. This inactivity may be due to defects in the proviruses themselves or to position effects mediated by DNA sequences flanking the proviral units. The transcriptional competence of long terminal repeats (LTRs) derived from endogenous proviral DNA at genetic loci Mtv-8, Mtv-9, and Mtv-17 of the C57BL/6 mouse strain was examined with a transient transfection assay in which gene expression was monitored by expression of chloramphenicol acetyltransferase. LTRs from Mtv-8 and Mtv-9 were able to direct glucocorticoid-induced chloramphenicol acetyltransferase expression in this assay, while the LTR from Mtv-17 was only about 5% as effective. Analysis of chimeric LTRs indicated that the glucocorticoid-inducible transcriptional enhancer element within the Mtv-17 LTR is active when linked to a functional promoter from Mtv-8, whereas the promoter from Mtv-17 is defective in directing hormone-induced gene expression, even when linked to the Mtv-8 glucocorticoid-responsive enhancer. The DNA sequence of transcriptional control regions of the LTRs of all three endogenous proviral units was determined; this analysis revealed that the source of the defect in Mtv-17 is a single G-to-A transition at position-75 with respect to the site of transcription initiation that resides within the previously defined binding site for the transcription factor nuclear factor 1. Competition experiments with a gel electrophoresis mobility shift assay indicated that the affinity of nuclear factor 1 for DNA derived from Mtv-17 is significantly less than for comparable sequences derived from Mtv-8.

Formyl peptide receptor chimeras define domains involved in ligand binding.

We have begun to study the structural requirements for the binding of formyl peptides to their specific receptors. As an initial approach, we constructed C5a-formyl peptide receptor chimeras. Unique (and identical) restriction sites were introduced within the transmembrane domains of these receptors that allowed for the exchange of specific areas. Four types of chimeric receptors were generated. 1) The C5a receptor was progressively substituted by the formyl peptide receptor. 2) The formyl peptide receptor was progressively substituted by the C5a receptor. 3) Specific domains of the C5a receptor were substituted by the corresponding domain of the formyl peptide receptor. 4) Specific domains of the formyl peptide receptor were replaced by the same corresponding domain of the C5a receptor. Wild type and chimeric receptors were transfected into COS 7 cells and their ability to bind formyl peptide determined, taking into account efficiency of transfection and expression of chimeric protein. Based on these results, a ligand binding model is presented in which the second, third, and fourth extracellular (and/or their transmembrane) domains together with the first transmembrane domain form a ligand binding pocket for formyl peptides. It is proposed that the amino-terminal domain plays a role by presumably providing a "lid" to the pocket. The carboxyl-terminal cytoplasmic tail appears to modulate ligand binding by regulating receptor affinity.

The short loop between epidermal growth factor-like domains 4 and 5 is critical for human thrombomodulin function.

Thrombomodulin (TM) is a cofactor for activation of protein C by thrombin. We showed that 80-90% of this cofactor activity is lost by oxidation of Met388, located within the short interdomain loop between epidermal growth factor-like domains 4 and 5 (Glaser, C. B., Morser, J., Clarke, J. H., Blasko, E., McLean, K., Kuhn, I., Chang, R.-J., Lin, J.-H., Vilander, L., Andrews, W. H., and Light, D. R. (1992) J. Clin. Invest. 90, 2565-2573). For each of the 3 amino acids of the loop, site-specific mutants are described in which, 1) all possible single amino acid substitutions are made, 2) deletions are made, or 3) alanine is inserted adjacent to each residue of the loop. Most substitutions within the loop (38/57) result in a > 50% decrease in cofactor activity, while changes in the length of this region result in > 90% loss of activity. Only the Met388-->Leu mutant has higher cofactor activity (2-fold) than wild-type TM. A number of soluble and full-length TM analogs with the Met388-->Leu substitution are improved thrombin cofactors, whether produced in bacteria, insect, or mammalian cells. Detailed kinetic analysis of a soluble TM analog consisting of the six EGF-like domains secreted from insect cells shows that the enhanced activity of the Met388-->Leu mutant results from an increased catalytic efficiency (kcat/Km). This enhancement is maximal at physiological concentrations of calcium. The loss of activity following Met388 oxidation in the wild-type protein is the result of both decreased binding to thrombin (Kd effect) and a decreased interaction of the TM.thrombin complex with protein C (Km effect). We demonstrate the critical role of this interdomain loop in the biological anticoagulant properties of TM.