Plasmin and Matrix Metalloproteinase System in Deep Venous Thrombosis Resolution

Clinical Relevance and Background

Deep venous thrombosis (DVT) is a serious health care problem in the United States, with over 250,000 patients affected yearly and at least 200,000 diagnosed yearly with pulmonary embolism. ^1–3 The incidence of total cases of venous thromboembolism exceeds the number of myocardial infarctions and total strokes in this country yearly, whereas the incidence of venous thromboembolism–related deaths exceeds the number of myocardial infarction–related deaths or stroke-related deaths. The incidence of DVT has been increasing with the aging of the population. In those 85 to 89 years old, the incidence is reported as high as 310 per 100,000 population. ⁴ Additionally, treatment costs are in billions of dollars per year. ⁵ The late DVT consequence, chronic venous insufficiency, affects between 400,000 and 500,000 patients with skin ulcerations and 6 to 7 million patients with severe manifestations, including stasis pigmentation and stasis dermatitis.

The standard of care for DVT is rapid anticoagulation with a heparin agent, surgical compression hose, and subsequent institution of a vitamin K antagonist. Although this paradigm is highly effective in preventing recurrent venous thromboembolism and may decrease post-thrombotic syndrome (PTS), the reality of compliance with these therapies is often different. Bleeding complications long term are not uncommon, and despite therapy, development of significant PTS occurs in up to 30% at 8 years. ⁶ Thus, the need for better and safer therapies exists. Similarly, therapies to modulate fibrotic vein wall damage (or PTS), a consequence of DVT lysis, are particularly lacking. Recent promising small series suggest that in cases of iliofemoral DVT, aggressive catheter-directed thrombolysis may result in decreased PTS. This strategy is based on the principle of rapid thrombus removal to decrease vein wall inflammatory damage from the adjacent DVT and the mediators therein. Again, bleeding complications exist, and this aggressive therapy is unlikely to be cost-effective or safe for all patients with DVT. Thus, limited regionally specific rapid thrombus clearance and minimization of vein wall injury are the ultimate goal.

DVT Resolution Is an Inflammatory Process

DVT resolution resembles wound healing and involves both profibrotic growth factors, collagen deposition, and matrix metalloproteinase (MMP) expression and activation. ^7–10 The fact that leukocytes invade the thrombus in a specific sequence suggests their importance in normal thrombus resolution. ¹¹ The first cell type in the thrombus is the neutrophil (polymorphonuclear neutrophil [PMN]), which may contribute to both lysis and vein wall damage. ^12,13 Although PMNs may cause vein wall injury, they are essential for early thrombus resolution by promoting both fibrinolysis and collagenolysis. Specifically, neutropenia in a rat model of stasis DVT was associated with larger thrombi at 2 and 7 days and was correlated with increased thrombus fibrosis and significantly lower thrombus levels of both urokinase-type plasminogen activator (uPA) and MMP-9. ^12,14

The monocyte is likely the most important cell for later DVT resolution, through multiple actions. Monocyte influx into the thrombus peaks at day 8 after thrombogenesis and correlates with elevated monocyte chemotactic protein 1 (MCP-1) levels, one of the primary cysteine-cysteine (CC) chemokines that direct monocyte chemotaxis and activation. ¹⁵ Targeted deletion of CC receptor 2 (CCR-2 knockout) in the mouse model of stasis thrombosis was associated with early and late impairment of thrombus resolution, probably via impaired interferon-γ (IFN-γ)-mediated MMP-2 and MMP-9 activity. Indeed, CCR-2 knockout mice with stasis thrombosis supplemented with exogenous IFN-γ had full restoration of thrombus resolution, in part owing to recovery of MMP-2 and MMP-9 activities, without an increase in thrombus monocyte influx or change in uPA levels. ¹⁶

As the thrombus resolves, numerous proinflammatory factors are released into the local environment, including interleukin (IL)-1β and tumor necrosis factor α. ¹¹ The cellular sources of these different mediators have not been specifically defined but likely include leukocytes, vein wall smooth muscle cells, and fibroblast-like cells within the resolving thrombus. The cellular leukocyte kinetics in the vein wall after DVT are similar to what is observed in the thrombus, with an early influx of PMNs, followed by monocytes. Based on the rat model of stasis DVT, elastinolysis seems to occur early, with partial recovery by 28 days, as measured by vein wall stiffness changes (the inverse of compliance, a property of normal veins). ¹⁷ The increased vein wall stiffness continues through 14 days and is accompanied by elevated MMP-2 and MMP-9 gene expression. However, early vein wall collagenolysis (rather than collagen production) seems to occur within the first 7 days in stasis DVT (P.K. Henke, Thromb Haemost, 2007, in press).

Elevation of profibrotic mediators, including transforming growth factor β (TGF-β), IL-13, RANTES, and MCP-1, accompanies early biomechanical injury from the DVT. These are present within the vein wall and thrombus and may drive the fibrotic response. For example, IL-13 is released from activated T-helper lymphocytes and promotes the expression of MCP-1. Although exogenous MCP-1 may hasten DVT resolution, ¹⁸ it also promotes organ fibrosis in vivo. ¹⁵ The profibrotic growth factor TGF-β is also present in the thrombus and is activated with normal thrombolysis. ¹⁹ This factor may be a key mechanism promoting vein wall fibrosis. Late fibrosis has been observed in the mouse stasis model of DVT with an increase in collagen I and III gene expression, as well as an increase in MMP-2 and MMP-9 gene expression and activity. ²⁰

We recently demonstrated that inhibition of the inflammatory response can decrease vein wall fibrosis, although the exact mechanism is not clear. In a rat study in which animals were treated with either low-molecular-weight heparin or an oral inhibitor to P-selectin 2 days after establishment of thrombosis, P-selectin inhibition significantly decreased vein wall injury, as measured by vein wall stiffness and decreased intimal thickness score, IL-13 levels, MCP-1 levels, and platelet-derived growth factor β levels. ²¹ This effect was independent of thrombus size. Thus, these experimental models suggest that early vein wall injury is associated with collagenolysis and elastinolysis, followed later by collagen and matrix accumulation.

Plasmin Activators and Vein Wall Injury

Plasminogen activation provides localized proteolytic activity to allow fibrinolysis. ^22–24 Plasmin is the primary fibrinolytic protease that directly degrades fibrin, as well as activating other matrix-degrading enzymes, such as pro-MMPs. ^25,26 It has been shown that uPA, rather than tissue plasminogen activator (tPA), is most important in thrombus resolution in the venous system. ²⁷ Specifically, DVT resolution was significantly impaired in uPA knockout mice, with fewer thrombus monocytes, and was not observed in tPA knockout mice. Previous studies in a mouse DVT model have shown a reciprocal activation of the plasmin and MMP systems ¹⁶ and suggest that other mechanisms may be important for established thrombus resolution.

Given the background of uPA and MMP importance in DVT resolution, we asked if reciprocal activation would occur if the plasmin system was inhibited. To do this, we used aprotinin (AP) as a pharmacologic means to inhibit plasmin in the rat stasis DVT model. Not surprisingly, AP increased thrombus size at 7 days, suggesting reduced thrombus resolution (Figure 1A). This was expected because AP inhibits plasmin, which is largely responsible for fibrinolysis in DVT resolution. ^27–30 Beyond the thrombus size measure, other measures of DVT resolution (thrombus perfusion, total collagen, and D-dimer content) ^14,16,17 were not different between groups.

OPEN IN VIEWER

Figure 1.

Aprotinin pretreatment in rats undergoing stasis deep venous thrombosis. A, At 7 days, larger thrombi were observed with aprotinin (AP) treatment. B, Decreased vein wall stiffness was found with AP treatment. C, Upregulation of vein wall matrix metalloproteinase 9 (MMP-9) was associated with AP treatment. *p < .05, all n > 4. Reproduced with permission from Dewyer N, et al. J Surg Res [In press]. IVC = inferior vena cava.

Clinical and experimental studies suggest that the longer a DVT is in contact with the vein wall, the greater the resulting damage (P.R. Henke, Thromb Haemost, 2007, in press). ^31–33 In this same recent study, AP impaired DVT resolution and paradoxically reduced vein wall stiffness (Figure 1B). As this is a full stasis model of DVT, persistent thrombus–vein wall contact occurs, and this mechanism of DVT genesis was the same between both the AP and the control groups. Previous work has also suggested that the thrombus composition is more important than the size in determining vein wall injury, ^17,34 and the resulting damage in the vein wall may be caused by direct activity of the plasmin-activated proteases rather than simply the duration of thrombus–vein wall contact. Similarly, overexpression of uPA has been associated with cardiac injury in a mouse model. ³⁵ Another possible explanation is that attenuation of the inflammatory response by AP caused reduced vein wall injury, or AP inhibited other matrix enzymes, such as cathepsins or serum elastase. ³⁶

Reciprocal upregulation of vein wall MMP-9, but not MMP-2, was also observed (Figure 1C). This is counterintuitive in that plasmin may activate MMPs, specifically MMP-9, in the vascular system. ^37,38 However, multiple pathways provide for plasmin activation and subsequent MMP-activation. ^39,40 Both collagenolysis (type IV) and elastolysis occur with MMP-9 and are elevated in the vein wall during DVT resolution. ^16,20 It is likely that the MMP-9 is derived from influxed monocytes as few PMNs are present after 7 days. ¹⁷ Our data indirectly suggest that MMPs are not the main agents responsible for vein wall fibrotic injury. Interestingly, this has been observed with PMN depletion in experimental stasis DVT, whereby early loss of thrombus MMP-9 was associated with increased vein wall stiffness. ¹⁷ Taken together, these studies suggest that MMP-9 may be more important in juxtaposition to the thrombus for resolution rather than mediating vein wall damage.

Plasminogen Activator Inhibitor and Vein Wall Injury

The other side of the coin is the natural inhibitors of plasmin and MMPs. Several lines of evidence support the role of plasminogen activator inhibitor 1 (PAI-1) in mediating postinjury fibrosis, although this is dependent on the model system, the presence or absence of thrombus, and the analysis of tissue. For example, bleomycin-induced lung injury in PAI-1 knockout mice was lessened compared with controls, as measured by total collagen accumulation and histologic analysis. ⁴¹ Other investigators found that this effect was not dependent on fibrin, suggesting that plasmin's effect may be indirect, through promotion of cellular migration or growth factor activation. ⁴² Similar profibrotic effects of PAI-1 were observed in a model of ureteral obstruction–induced renal fibrosis, possibly by promoting myofibroblast influx. ⁴³ In a mouse model of FeCl3 transendothelium-induced vascular injury, PAI-1 knockout mice were protected from neointimal formation, primarily through decreased vascular smooth muscle cell (VSMC) migration. ⁴⁴ In contrast, in a nonthrombotic model of arterial flow reduction, PAI-1 was shown to be protective of fibrosis by inhibiting thrombin-mediated VSMC proliferation. ⁴⁵ In arterial plaques in apolipoprotein E^−/− mice, PAI-1 was identified in myofibroblasts and associated with increased matrix deposition. ⁴⁶ Taken together, we believe that PAI-1, in the setting of thrombus, promotes matrix deposition, myofibroblast accumulation, and increased TGF-β activity that contributes to fibrotic injury after acute DVT.

Equally important are the natural inhibitors of the MMPs, namely, the tissue inhibitor of metalloproteinases (TIMP). These molecules act to inhibit MMP activity and, in the context of fibrotic diseases, act to promote a net accumulation of collagen and hence fibrotic injury. ^47,48 Their direct role in vein wall remodeling has not yet been fully established.

Summary

The current literature and our own work suggest that a complex and dynamic interplay of fibrinolysis, plasminogen activators and inhibitors, cytokines, and MMPs and possibly TIMPs occurs during DVT resolution (Figure 2). Much work remains to be done to elucidate the relevant mechanisms involved with DVT resolution and concurrent vein wall injury. We are investigating the phenotypes of DVT resolution with uPA knockout and the MMP-2 and -9 knockout mice. Although native plasmin is essential for fibrinolysis, this may paradoxically worsen vein wall injury. By better understanding the process of DVT resolution, adjunctive therapies may be brought to fruition. The ultimate clinical goal is to provide rapid fibrinolysis, decrease vein wall injury, and minimize bleeding risk.

OPEN IN VIEWER

Figure 2.

Proposed hypothetical mechanism (in part) of stasis deep venous thrombosis resolution. Note that the early and late mediators are related to the predominant leukocyte within the thrombus and vein wall. IL = interleukin; MMP = matrix metalloproteinase; PMN = polymorphonuclear; TGFβ = transforming growth factor β; TIMP = tissue inhibitor of metalloproteinases; uPA = urokinase-type plasminogen activator; VSMC = vascular smooth muscle cells.