The plasminogen activator/plasmin system is known to initiate a proteolytic cascade resulting in the activation of matrix metalloproteinases in vitro leading to the degradation of extracellular matrix. To investigate whether or not this cascade is present during delayed wound healing and contributes to the pathophysiological basis of impaired healing we examined the temporal expression of urokinase plasminogen activator, plasminogen activator inhibitor-1 and gelatinase-B in fluid collected from chronic venous leg ulcers compared to acute surgical mastectomy wounds. Using a chromogenic substrate assay, levels of active urokinase plasminogen activator in chronic wounds were found to be about five fold higher compared to sera and two fold higher compared to mastectomy wounds. Levels of active plasminogen activator inhibitor-1 in chronic wounds were four times higher than those found in sera and two times higher than those found in mastectomy wound fluid. Using a fibrin overlay system and reverse zymography, we found that when the wound was not healing, the expression of urokinase plasminogen activator in chronic wound fluid was initially detected in the active forms (50 and 33 kDa), but that as the wound healed and decreased in size, was detected as an inhibitor- bound urokinase plasminogen activator–plasminogen activator inhibitor-1 complex (≅ 80–116 kDa). When the expression of active urokinase plasminogen activator was highest, no plasminogen activator inhibitor-1 was detectable. In contrast, urokinase plasminogen activator was always detected in the inhibitor bound form as a urokinase plasminogen activator–plasminogen activator inhibitor-1 complex in blood- and plasma-derived serum and mastectomy wound fluid. Plasminogen activator inhibitor-1 was detected in blood-derived serum and mastectomy wound fluid but not in plasma derived serum. Expression of matrix metalloproteinase-9 in chronic wound fluids, analyzed by gelatin zymography, showed that when urokinase plasminogen activator was detected in the active forms, matrix metalloproteinase-9 was overexpressed by approximately twice that found in mastectomy wounds and approximately 30 times that detected in blood-derived sera. When urokinase plasminogen activator appeared almost entirely as an enzyme- inhibitor complex, the level of expression of matrix metalloproteinase-9 was similar to that seen in mastectomy wound fluid. We conclude that the switch in urokinase plasminogen activator expression from an active to inhibitor bound form correlates with the decrease seen in matrix metalloproteinase-9 expression suggesting the presence of a proteolytic cascade initiated by the plasminogen activator/plasmin system during wound healing leading to the activation of matrix metalloproteinase-9. In addition, expression of urokinase plasminogen activator and matrix metalloproteinase-9 appear to be useful biomarkers to determine clinical wound healing status.