Annexin A2: A tPA Amplifier for thrombolytic stroke therapy

Abstract

Hemorrhagic transformation, incomplete reperfusion, neurotoxicity, and the short treatment time window comprise major challenges for thrombolytic therapy. Improving tPA therapy has become one of highest priorities in the stroke field. Recent efforts have been aimed at identifying new strategies that might enhance the thrombolytic efficacy of tPA, while reducing its associated complications related to hemorrhage and neurotoxicity. We believe that the combination of low-dose tPA with recombinant annexin A2 (a tPA and plasminogen co-receptor) might constitute a promising approach. Our pilot study using a focal embolic stroke model in rats supports this hypothesis.

Recent clinical investigations have demonstrated the potential for improving tPA therapy. For instance, PWI-DWI mismatch in MRI studies suggest that some patients may benefit from treatment beyond the conventional 3-hour time window ^{1, 2}. ECASS III, a randomized phase III trial designed to test treatment with tPA at 3 to 4.5 hours, showed that intravenous tPA given during this time window still improved clinical outcomes in patients with somewhat milder stroke scores ³. However, there was still a large difference in odds-ratios between early reperfusion within 90 minutes (approximately 2.8 OR) versus delayed reperfusion (approximately 1.4 OR). Thus, the benefits of thrombolysis are still heavily dependent on time to treatment, and use of tPA may still be associated with intracranial hemorrhage and reperfusion injury. It is, therefore, imperative that we seek combination therapies that will truly broaden the therapeutic window, reduce the risk of tPA-associated hemorrhagic transformation, and enhance thrombolytic efficacy. For this purpose, we believe that the combination of low-dose tPA with rA2 may be promising.

Pleiotropic effects of tPA reperfusion treatment

Ischemic stroke is a cerebrovascular event. In acute ischemic stroke, tPA thrombolytic therapy is intended to reopen occluded vessels by lysis of the thrombus, thereby improving clinical outcome through regional reperfusion and salvage of threatened tissues. tPA acts primarily inside the blood vessel; if blood flow is successfully restored by early thrombolytic reperfusion, then compromised brain tissue can be rescued. However, what happens when tPA thrombolysis is not fully successful as occurs in about 50% of patients, and reperfusion occurs within the context of weakened vessels and perturbed neurovascular homeostasis? Although unequivocal human data are lacking, findings from animal models suggest that tPA reperfusion may have deleterious consequences due to the non-thrombolytic actions of tPA ⁴. Although more investigations are required to dissect the molecular signaling mechanisms initiated by exogenous tPA in the occluded vessel and ischemic brain, emerging data suggest that, besides thrombolysis per se, exogenous tPA may have additional pleiotropic actions within the brain. ⁴. These may include direct vasoactivity, ^{5, 6} enhanced excitotoxicity⁷, and activation of extracellular proteases such as matrix metalloproteinases (MMPs) that may promote neurovascular injury ^7-10. These actions of tPA may increase ischemic neurotoxicity, damage the blood brain barrier, and exacerbate edema and cerebral hemorrhage ⁴, thereby compromising its usefulness as a thrombolytic agent ^{5, 11}.

Re-canalization, moreover, is an important predictor of stroke outcome, regardless of thrombolytic modality employed. Importantly, exogenous tPA may worsen ischemia-induced blood brain-barrier disruption, elevate the risk of symptomatic intracranial hemorrhage, and thus reduce the therapeutic time window. While one strategy to overcome these dose-dependent side effects of tPA might be simply to lower the tPA dose, this step would be likely to lower perfusion efficacy. Clearly, optimization of tPA-induced thrombolysis requires that one balance the potential benefits of reperfusion against the detrimental effects of exogenous tPA. Improving tPA thrombolytic regimens may both lengthen the time-to-treatment window and make reperfusion therapy safer and more efficacious ^{12, 13}. Recent efforts have been aimed at identifying new combination strategies that might increase the thrombolytic efficacy of tPA, while reducing its associated neurotoxicity and hemorrhagic transformation^{9, 14-18}.

tPA receptor annexin A2 and fibrinolytic assembly

In fibrinolysis, tPA serves a key role in regulating the breakdown of fibrin-containing thrombi by enzymatically converting clot-bound plasminogen to active plasmin. Plasmin, in turn, degrades cross-linked fibrin and, possibly, intact fibrinogen, a process called fibrinogenolysis¹⁹. However, recent vascular biology studies have revealed that tPA interacts with cellular receptors that allow it to carry out additional biological functions and to activate specific signal transduction pathways²⁰. A central tenet of cell surface fibrinolysis is the concept of fibrinolytic assembly, in which the tPA-dependent conversion of plasminogen to active plasmin is precisely orchestrated through the formation of a multimolecular complex, consisting of tPA, the annexin A2 heterotetramer, and plasminogen (Figure 1)²¹.

Figure 1.

Working model of fibrinolytic assembly on cell surfaces. In association with its partner protein, p11, annexin A2 is linked to the cell surface via calcium-dependent phospholipid-binding sites located within the four core domain repeats. The A2₂p11₂ heterotetramer binds tPA at the tail domain of annexin A2, while plasminogen appears to bind to residues within the fourth core domain of A2. Fibrinolytic assembly is a precisely orchestrated process that enhances the catalytic efficiency of plasmin generation.

Annexin A2 is a cell-surface protein, which, in complex with its binding partner p11, forms a heterotetrameric (A2₂p11₂) receptor for both plasminogen, the inactive precursor of plasmin, and its activator, tPA. By assembling tPA, annexin A2, and plasminogen, this complex increases the catalytic efficiency of tPA, enabling it to convert plasminogen to plasmin at least 60 times more efficiently than the same amount of tPA alone^{21, 22}. Expressed by endothelial cells, annexin A2 exists in both membrane-bound and soluble forms. ²³and it can be transported to the cell surface in response to cellular stress ²⁴.

Fibrinolytic assembly plays a critical role in maintaining blood and vascular homeostasis²⁵. Interestingly, complete deficiency of annexin A2 in mice leads to a lack of tPA cofactor activity, accumulation of intravascular fibrin, and failure to clear arterial thrombi²⁶. In addition, earlier studies demonstrated that soluble rA2 reduces thrombus formation in rat carotid ²⁷ and middle cerebral arteries in vivo. ²⁸ These studies indicate that the use of rA2 for enhancing tPA thrombolytic efficiency may be feasible.

Combination stroke therapy of recombinant annexin A2 protein plus low-dose tPA

As noted above, it has been demonstrated that tPA converts plasminogen to clot-dissolving plasmin, and that this action is enhanced by complex formation between tPA, annexin A2, and plasminogen. However, clinical experience has shown that treatment with high doses of exogenous tPA alone may be partially responsible for the limitations of tPA reperfusion stroke therapy. Increased risk of hemorrhage, lower reperfusion efficiency, and the abbreviated therapeutic time window associated with higher dose tPA constitute major challenges in this field ^{9, 29}. Our overall hypotheses have been that recombinant annexin A2 protein (rA2) will reduce the dose of tPA required for reperfusion, while enhancing thrombolytic efficacy, and attenuating intracerebral hemorrhagic transformation. We postulate further that the combination of tPA and rA2 will lengthen therapeutic time windows and improve long-term outcomes.

We have tested these hypotheses in a pilot study. For this purpose, we synthesized and purified recombinant human annexin A2 protein (rA2), as previously described²⁷. Consistent with previous reports^{21, 30}, in vitro plasmin activity assays showed that rA2 significantly amplified tPA-mediated plasmin generation.³¹. We also found that equivalent levels of plasmin activity can be reached by using high dose tPA alone or lower-dose tPA in combination with rA2. The simultaneous use of tPA with rA2 at about a 1:2 w/v ratio (or 1:4 molar ratio) increased the plasmin-generating capability of tPA in vitro by almost 4-fold (Figure 2).

Figure 2.

Recombinant annexin A2 accelerates tPA-dependent plasminogen activation in vitro. A range of concentrations of tPA (1, 2.5, 5, 10 μg /ml) with or without the indicated concentrations of rA2 (0, 1, 2.5, 5 μg/ml) were added to wells of a 96-well plate. Plasmin activity was represented as fold plasmin activity relative to 1 μg /ml of tPA alone. The data are expressed as mean + s.e.m., n=4 per group. Reprinted with permission from J Cereb Blood Flow Metab³¹. Copyright 2010, Nature Publishing Group.

We then tested our hypothesis in a focal embolic stroke model in rats. This model was induced by injection of one blood clot (40-mm in length) into middle cerebral artery as we previously described (REF). Because of species-related differences in fibrin specificity, the equivalent effective dose of human recombinant tPA in the rat was about 10 times higher than the dose used in humans, or about 10 mg/kg, ³². In the first set of experiments, animals were treated intravenously 2 hours after initiation of ischemia. Neither intermediate (5 mg/kg) nor low (2.5 mg/kg) doses of tPA alone, nor rA2 (5 mg/kg) alone were effective in improving reperfusion or reducing infarction. However, the combination of low-dose tPA (2.5 mg/kg) plus rA2 (5 mg/kg) was as effective as the standard high dose tPA alone in restoring perfusion and reducing infarct size³¹. These data suggested that rA2 can make low-dose tPA more effective in an embolic stroke animal model.

In the second set of experiments, rats were treated intravenously in a delayed fashion, 4 hours after the onset of stroke, with either saline, high dose tPA (10 mg/kg), or low-dose tPA (2.5 mg/kg) plus rA2 (5 mg/kg). At 24 hours after the initiation of stroke, the combination of low-dose tPA (2.5 mg/kg) plus rA2 (5 mg/kg) significantly reduced infarct volume when compared with either saline or high-dose tPA alone³¹. As expected, high-dose tPA administered at the delayed 4-hour time point induced significant hemorrhagic transformation, whereas hemorrhage was significantly ameliorated by the combination regimen (Figure 3).

Figure 3.

tPA-rA2 combination therapy reduces hemorrhage volume in rats with ischemic stroke. Rats subjected to focal embolic stroke were treated at 4 hours with saline, standard high-dose tPA (10 mg/kg, H-tPA), or low-dose tPA (2.5 mg/kg, L-tPA) plus rA2 (5 mg/kg). The volumes of intracerebral hemorrhage ware quantified by hemoglobin assay at 24 hours after stroke. The data were expressed as mean + s.e.m., n=10, *p<0.05. Reprinted with permission from J Cereb Blood Flow Metab³¹. Copyright 2010, Nature Publishing Group.

To test whether the combination of tPA and rA2 can improve neurological outcomes in longer survival time after stroke, rats were treated in a third experiment in a delayed fashion, 4 hours after stroke, with either saline or a combination of low dose tPA (2.5 mg/kg) plus rA2 (5 mg/kg). At 3 days after stroke, the combination regimen was found to significantly both decrease brain infarct volume (Figure 4A) and neurological deficits (Figure 4B). Overall mortality, furthermore, was significantly reduced in the combination therapy group (21%, 3/14) in comparison to saline-treated animals (42%, 6/14). Taken together, these results demonstrate that addition of the “tPA amplifier” rA2 may reduce the effective thrombolytic dose of tPA, minimize hemorrhage and brain infarction, and prolong the reperfusion time window in stroke. These findings provide a promising new approach for enhancing tPA thrombolytic therapy.

Figure 4.

Neurologic scores and infarction volume are improved in rats receiving combination therapy. Two groups of rats were treated intravenously at 4 hours after stroke with either saline or a combination of low-dose tPA (2.5 mg/kg, L-tPA) plus rA2 (5 mg/kg). A. Three days after the onset of stroke, neurological scores of two group rats were evaluated. B. Ischemic infarction size on hematoxylin and eosin-stained sections were examined in each group, *p<0.05, n=8 for saline and n=11 for the combination. Reprinted with permission from J Cereb Blood Flow Metab³¹. Copyright 2010, Nature Publishing Group.

A number of reports have suggested new strategies for reducing tPA dosage, , thereby improving safety, while enhancing plasmin generation or activity. These approaches are based upon the following findings: (1) local infusion of plasmin directly into the thrombus does not cause excessive bleeding within 6 fold greater than the effective thrombolytic dose of tPA used ³³; (2) antiplasmin neutralizes circulating plasmin within one second of its appearance ³³ whereas tPA has longer half life (4-5 min), and can cross the blood brain barrier (BBB), whether damaged or intact, through LRP-dependent and -independent mechanisms, further compromising BBB integrity and worsening brain damage ^{8, 34, 35}; (3) tPA can induce plasmin-independent MMP-9 up-regulation and MMP-9-mediated microglia activation in stroke animal models ^{8, 36, 37}. (4) initial experimental evidence suggests that administration of rA2 alone in vivo is not associated with organ specific or systemic complications ^{27, 28, 38}. These findings, together with our preliminary data, suggest that optimization of tPA-mediated fibrinolysis may greatly improve the efficacy and safety of this form of stroke therapy.

Because annexin A2 accelerates the activation of plasmin by complexing with tPA and plasminogen, it is tempting to speculate further that the rA2-tPA combination might generate more plasmin locally at the clot site ^{39, 40}. rA2-associated tPA might generate A2 associated plasmin which could be delivered to the thrombus to initiate fibrinolysis. Alternatively, fibrin-associated plasminogen within the thrombus could be activated by A2-associated tPA, resulting in more effective fibrinolysis (Figure 5). rA2-bound tPA and plasmin, moreover, would be relatively protected from their circulating inhibitors, plasminogen activator inhibitor-1 (PAI-1) and alpha₂-antiplasmin (alpha₂-AP), respectively.

Figure 5.

Working models for tPA-related thrombolytic stroke therapies. A. Conventional therapy with tPA alone. When delivered into the circulation, tPA complexes immediately with its high affinity inhibitor, plasminogen activator inhibitor-1 (PAI-1). Free tPA circulates only when the inhibitory capacity of PAI-1 has been exceeded. Free tPA can then bind to fibrin, where plasminogen will already be bound. Fibrin markedly accelerates the catalytic efficiency of tPA-dependent plasminogen activation, generating plasmin, which degrades the thrombus, and generates fibrin degradation products (FDPs). B. Combination therapy with tPA and recombinant annexin A2 (rA2). We postulate that injection of tPA with rA2 leads to the formation of a circulating complex. Two scenarios seem plausible. In the first, the tPA-rA2 complex would also bind plasminogen and activate it to generate rA2-associated plasmin. This complex could then interact with fibrin in a thrombus, and initiate its degradation, forming FDPs. Alternatively, rA2-associated tPA might interact directly with fibrin within the thrombus, and the tPA might then activate fibrin-bound plasminogen, and generate plasmin, which would then dissolve the thrombus, generating FDPs. Both tPA and plasmin would be relatively protected from their circulating inhibitors, plasminogen activator inhibitor-1 (PAI-1) and alpha₂-antiplasmin (alpha₂-AP), respectively. Because, in both scenarios, there would be less circulating free tPA that that seen with conventional therapy, the potential for restoration of blood flow without hemorrhage or neurotoxicity would be much greater.

We speculate that the enhanced thrombolytic efficacy of rA2 plus tPA reflects improved cerebral blood flow. The decrease in side effects associated with this combination therapy may involve a reduction in tPA-associated blood-brain barrier disruption and neurovascular injury. Ultimately, improved blood flow and fewer neurovascular side-effects should translate into smaller infarct volume and improved long-term outcome. It is clear that safety issues and all translational aspects of this potential treatment need to be carefully investigated prior to any future preclinical evaluation.