Tissue Factor Is Absent from Sites of Cutaneous Wounds: Implications for Bleeding in Hemophilia.

Tissue factor (TF) plays an important role in hemostasis. Its expression is upregulated by a variety of proinflammatory cytokines and thus can link inflammation with thrombosis. We have been studying wound healing in hemophilia B mice. Many of the hemophilic mice do not bleed more than wild type mice at the time of wounding. However, hemophilia B mice show a pattern of bleeding 12 to 24 hours after wounding; in some cases this bleeding is quite severe. We hypothesized that the level and pattern of TF expression might be a major determinant of the bleeding response to cutaneous wounding, since enhanced activity of the TF pathway could overcome the lack of the intrinsic factor X-activating complex (FVIIIa/FIXa) in hemophilia. Briefly, the wound healing model is carried out as follows: a single three mm punch biopsy was placed on the dorsal skin of each C57BL/6 wild-type or FIX knockout mouse. Tissue from around the wound was collected at different days, examined histologically and immunostained with a rabbit polyclonal antibody against the extracellular domain of mouse TF (anti-whole TF, the kind gift of Dr. Mirella Ezban, Novo Nordisk A/S) and a rabbit polyclonal antibody to desmin, a marker for pericytes (Abcam Inc, Cambridge, MA). Under normal conditions, non-capillary blood vessels are surrounded by a prominent TF coat due to expression on pericytes and adjacent adventitial tissues. However, TF surprisingly disappeared from the pericyte coat around blood vessels near the wound by one day after wounding. The TF expression returned around these vessels by six to eight days after wounding. One possible explanation is that the TF-bearing pericytes disappeared, either by migrating away or by apoptosis. We found that desmin-positive cells were present around the vessels in both the absence and presence of TF, indicating that pericytes were present even when TF expression was absent. Furthermore, we found that the highly vascular granulation tissue that grew in to fill the defect at the punch biopsy site was also devoid of TF antigen - through the latest time points sampled at 20 days after wounding - even though the new vessels were surrounded by a sheath of pericytes. We propose two hypotheses that might account for these findings:

1. Pericytes may downregulate TF expression by an as yet undetermined pathway during angiogenesis and wound healing; and/or

2. the inflammatory cell infiltrate migrating through the vessel walls to the wound area may proteolytically cleave TF antigen from the cell membrane.

Thus, tissues in the area of a skin wound have a profound depression of TF expression that may serve to prevent thrombosis of the fragile neovessels as they develop and remodel during healing. Previous studies in this model (Blood 2006; DOI 10.1182/blood-2006-05-020495) have shown that hemophilia B mice have evidence of significant ongoing bleeding around the wound site and in the granulation tissue, while wild type mice have only occasional extravasation of red cells in the granulation tissue. Thus we speculate that absence of TF around new vessels is not a significant liability to hemostatically competent individuals. However, the lack of TF renders hemophilic subjects highly vulnerable to recurrent hemorrhage during this critical period of healing.