The mouse dorsal skinfold chamber as a model for the study of thrombolysis by intravital microscopy

Abstract

Although intravital microscopy models of thrombosis in mice have contributed to dissect the mechanisms of thrombus formation and stability, they have not been well adapted to study long-term evolution of occlusive thrombi. Here, we assessed the suitability of the dorsal skinfold chamber (DSC) for the study of thrombolysis and testing of thrombolytic agents by intravital microscopy. We show that induction of FeCl3-induced occlusive thrombosis is achievable in microvessels of DSCs, and that thrombi formed in DSCs can be visualised by intravital microscopy using brightfield transmitted light, or fluorescent staining of thrombus components such as fibrinogen, platelets, leukocytes, and von Willebrand factor. Direct application of control saline or recombinant tissue-plasminogen activator (rtPA) to FeCl3-produced thrombi in DSCs did not affect thrombus size or induce recanalisation. However, in the presence of hirudin, rtPA treatment caused a rapid dose-dependent lysis of occlusive thrombi, resulting in recanalisation within 1 hour after treatment. Skin haemorrhage originating from vessels located inside and outside the FeCl3-injured area was also observed in DSCs of rtPA-treated mice. We further show that rtPA-induced thrombolysis was enhanced in plasminogen activator inhibitor-1-deficient (PAI-1-/-) mice, and dropped considerably as the time between occlusion and treatment application increased. Together, our results show that by allowing visualization and measurement of thrombus lysis and potential bleeding complications of thrombolytic treatments, the DSC provides a model for studying endogenous fibrinolysis and for first-line screening of thrombolytic agents. Furthermore, using this system, we found that thrombin and clot aging impair the thrombolytic action of rtPA towards FeCl3-produced thrombi.

Figure 1. FeCl₃ vascular injury in dorsal skinfold chambers

A. Representative micrographs of various thrombi under formation in DSC microvessels after injury by FeCl₃. The left panels depict greyscale images taken in brightfield transmitted light, where forming thrombi can be seen as white intravascular aggregates. The right panels show the corresponding fluorescent thrombus staining by rhodamine-6G (red), Alexa 488-conjugated anti-VWF (green), Alexa 594-conjugated fibrinogen (red), or calcein-red platelets (red), as indicated in the figure. The vessel and thrombus edges are highlighted in white. Arrows indicate the direction of blood flow. One can see that VWF tends to accumulate at the rear of the forming thrombus. Bars = 200 μm. B. Five minutes before applying a filter paper saturated with 15% FeCl₃ to microvessels in DSCs, Alexa 594-BSA was injected intravenously to assess vascular permeability. The micrographs show images of the Alexa 594-BSA red fluorescent signal in a FeCl₃-exposed area, immediately, 20 minutes, and 40 minutes after removing the FeCl₃-drenched filter paper. Leakage of fluorescent BSA in the extravascular space indicates that FeCl₃-injured vessels are permeable to proteins of molecular weight up to 66,000 Da. Bar = 200 μm.

Figure 2. Evaluation of thrombolytic therapies by intravital microscopy using the dorsal skinfold chamber

Occlusive thrombosis in DSCs was induced by FeCl₃-induced vascular injury in 1–3 vessels (100–170 μm diameter) per mouse. Twenty minutes after occlusion, various doses of recombinant tissue-plasminogen activator (rtPA) were applied directly in DSCs in the presence or absence of 50 μM hirudin. A. Incidence of vessel recanalization at 1 hour after thrombolytic treatment. Numbers above the bars indicate the number of vessels recanalized/number studied. n = 6–7 mice per group. * indicates a statistically significant difference from the rtPA- and hirudin-untreated control goup. B. Representative intravital microscopy images showing the morphological evolution of rhodamine-6G-labeled thrombi immediately after vessel occlusion and 1 hour after various thrombolytic treatments. Bar = 200 μm. C. Thrombus surface area at 1 hour following various thrombolytic treatments. Results are expressed as percentages of the thrombus surface area just after occlusion. n = 6–7 mice per group.

Figure 3. Visualization of both platelets and fibrin dissolution following thrombolytic treatment

Micrographs of a representative occlusive thrombus stained by Alexa488-fibrinogen and rhodamine-6G before and at 15 and 30 minutes after treatment with 40 μM rtPA and 50 μM hirudin. The vessel edges are highlighted in white. The green fluorescence signal outside the vessel corresponds to leakage of Alexa488-fibrinogen from the FeCl₃-injured vessel in the extravascular space.

Figure 4. Effect of rtPA treatment on tissue hemorrhage

A. Representative images of the skin from control (left panels) and rtPA-treated (right panels) mice. The upper right panel shows bleeding in the vicinity of the FeCl₃-injured area (yellow stain) in the skin of an rtPA-treated mouse. The lower right panel shows bleeding distant from the FeCl₃-injured area in the skin of an rtPA-treated mouse. Bar = 200 μm. B. Proportion of mice presenting skin hemorrhage within 24 hours after treatment with increasing doses of rtPA in the presence or absence of 50 μM hirudin following FeCl₃-induced occlusive thrombosis. Numbers above the bars indicate the number of mice with skin bleeding/number studied in each group.

Figure 5. Comparison of rtPA-induced thrombolysis between wild-type and plasminogen activator inhibitor- 1-deficient mice

After inducing occlusive thrombosis in up to 3 vessels per mouse, the efficiency of a thrombolytic treatment by 10 μM recombinant tissue-plasminogen activator (rtPA) and 50 μM hirudin was compared between wild-type and plasminogen activator inhibitor- 1-deficient (PAI-1 −/−) mice. A. Initial occlusive thrombus size in wild-type and PAI-1 −/− mice before thrombolytic treatment. n = 6 mice per group. B. Incidence of vessel recanalization in wild-type and PAI-1 −/− mice at 1 hour and 2 hours after rtPA treatment. Numbers above the bars indicate the number of vessels recanalized/number studied. n = 6 mice per group. C. Representative intravital micrographs showing the morphological evolution of rhodamine-6G-labeled thrombi immediately after vessel occlusion and 1 hour after thrombolytic treatment. Bar = 200 μm. D. Thrombus surface area in wild-type and PAI-1 −/− mice at 1 hour following rtPA treatment. Results are expressed as percentages of the initial occlusive thrombus size just before treatment. n = 6 mice per group.

Figure 6. The efficacy of rtPA-induced thrombolysis decreases as the interval between occlusion and time of treatment increases

A. Evolution of the thrombus surface at 1 hour after early or late treatment with 40 μM rtPA and 50 μM hirudin. Results are expressed as percentages of the thrombus surface just after occlusion. Each dot represents a thrombus. n = 5–6 mice per group. B. Incidence of recanalization at 30 min and 1 hour after treatment with 40 μM rtPA and 50 μM hirudin, administered either early (within 1 hour) or late (> 4 hours) after occlusion. Numbers above the bars indicate the number of vessels recanalized/number studied. n = 5–6 mice per group.