Abstract

Tissue factor (TF) exists in an encrypted, inactive form and an active, procoagulant form. The extracellular domain of TF contains two disulfide bonds. The crystal structure of the active form of TF reveals Cys186 and Cys209 in an unusual solvent-exposed disulfide bond. It was recently proposed that encrypted TF contains two free thiols at Cys186 and Cys209, and formation of a disulfide bond between Cys186 and Cys209 is associated with the decryption of TF (Chen et al, Biochemistry 2006). To characterize the structural differences between the putative encrypted TF and active TF, containing one and two disulfide bonds respectively, we developed an assay to quantitate the free thiols and disulfide bonds in recombinant TF and examined the solvent accessibility of the two disulfide bonds under native and denatured conditions. The perturbations of chromophores of TF by UV difference spectroscopy during specific reduction of Cys186–Cys209 were then examined. In order to quantitate the free thiols and disulfide bonds, TF was subjected to Alexa 488-maleimide (green) labeling to identify any free thiols and, after reduction with TCEP, a second labeling with Alexa 647-maleimide (red) to identify newly exposed free thiols formerly involved in disulfide bonds. Alexa-labeled TF was subjected to SDS gel electrophoresis and the gel was developed on a fluorescence imager to provide quantitative analysis of the integrated fluorescence intensity for each Alexa dye separately. The molar concentrations of free thiol in TF before TCEP treatment and thiols exposed after reduction were determined by comparison to calibration curves that were generated using Alexa maleimide-albumin conjugates. Recombinant TF in solution contains 0.2 moles of free thiol. Upon treatment with 10 mM TCEP under native conditions, TF is partially reduced and contains 1.97 moles of half cystine thiols. TF is fully reduced by 10 mM TCEP under denaturing conditions in the presence of 4 M urea (4.29 moles of thiols, theoretical=4.0 moles). This is consistent with the observation that only one disulfide bond is solvent-exposed in native TF and is accessible for reduction. We then performed ultraviolet absorption difference spectroscopy using the partitioned cell technique to monitor any structural transition during partial reduction of TF. The UV difference spectrum comparing the TCEP-treated TF and unreduced TF showed three peaks, one at 278 nm, a major peak at 285 nm and a small shoulder at 292 nm, that is characteristic of red-shifts in absorption of both tyrosine and tryptophan. Under the conditions employed, this conformational transition was completed in 60 minutes. These results comparing fully oxidized, active TF and partially reduced, encrypted TF indicate that tyrosine and tryptophan residues in TF become protected from solvent upon reduction of the Cys186–Cys209 disulfide bond. In conclusion, we have demonstrated by a new assay that only one disulfide bond within active TF is solvent-exposed and accessible for reduction. The partial reduction of TF at this disulfide bond results in a conformational transition that is associated with the protection of tyrosine and tryptophan residues, indicating structural difference between active TF and encrypted, inactive TF.

Author notes

Disclosure: No relevant conflicts of interest to declare.