Thrombolytic treatment only benefits a limited number of patients with ischemic stroke, and the development of improved therapies for stroke depends upon understanding the unique characteristics of the cerebrovasculature. The limitations of tPA appear to be due in part to unique activities that tPA has in the brain beyond its well established role in fibrinolysis9,17,19,37,38
. While there are clear benefits to patients who receive early thrombolytic treatment3,15
, the nonfibrinolytic effects of tPA suggest that there are unique challenges for the use of thrombolytic therapy in ischemic stroke. To understand these challenges it is necessary to understand the role that endogenous tPA plays in the CNS. It is clear that immediately after stroke onset, ischemia induces changes in cerebrovascular structures. We suggest that these changes include the release of tPA into perivascular tissue17
. This release precedes the loss of BBB integrity and appears to be mechanistically important for the early loss of BBB function since mice lacking tPA are protected from the loss of barrier function17
. The downstream target of tPA in this system is not plasminogen since mice lacking plasminogen exhibit similar levels of cerebrovascular permeability as wild type mice17
. Other potential CNS targets include the NMDA receptor and LRP9,24,39
. Experiments using antagonists of these receptors suggest that the NMDA receptor is not directly involved, whereas antagonists of LRP reduce cerebrovascular permeability after MCAO17,24
. Our studies support this hypothesis, and also identify PDGF-CC as a specific substrate of tPA present within the NVU. PDGF-CC is a member of the PDGF family that binds to the PDGFR-α. Unlike PDGF-A or PDGF-B, PDGF-C contains an amino-terminal CUB domain that renders the dimer inactive. However, removal of the CUB domains by tPA20
generates active PDGF-CC which can then activate the PDGFR-α in the NVU.
The direct injection of active PDGF-CC into the CSF bypasses the requirement of activation and increases cerebrovascular permeability in non-ischemic brain. In contrast to tPA-induced vascular permeability17
, the enhanced permeability of active PDGF-CC is independent of LRP, implying that both tPA and LRP act upstream of active PDGF-CC. Consistent with this, we show that cell-associated LRP enhances tPA-mediated activation of PDGF-CC. In the context of the NVU, LRP may be a necessary cofactor for efficient activation of latent PDGF-CC by tPA. We also show that PDGF-CC and PDGFR-α are expressed in cerebral arterioles, as is tPA40
. The localization of tPA, LRP, PDGF-CC and the PDGFR-α within the NVU of arterioles may facilitate the response to ischemia through activation of the PDGFR-α.
Electron microscopy demonstrates that the development of edema in response to tPA or PDGF-CC is not due to gross destruction of vascular structures or the basement membrane. Instead, these data suggest that the development of edema may be occurring through a regulated process, and thus may represent an exaggeration of a normal vascular response. In pathological conditions such as stroke, neuronal depolarization associated with cerebral ischemia can result in a surge of local tPA activity17,41
, which in turn could lead to continued production of active PDGF-CC, persistent activation of PDGFR-α in the NVU, and ultimately loss of BBB integrity.
The benefit of thrombolytic tPA administered within the first 3 h of stroke is likely dependent upon the early maintenance of BBB integrity. In this situation thrombolysis of the occluded vessel should rescue the affected ischemic zone and improve clinical outcome. However, administering tPA beyond the 3 h window increases the likelihood that changes in the cerebrovasculature impair BBB integrity to the point that thrombolytic tPA crosses into the perivascular tissue16
and interacts with the NVU. We postulate that this tPA may further increase activation of endogenous PDGF-CC, thus prolonging activation of the PDGFR-α. This leads to further deterioration of BBB integrity, and in extreme cases may lead directly to hemorrhagic complications.
Understanding this mechanism offers an opportunity to potentially extend the treatment window of tPA for stroke patients. Since blocking the PDGF-CC/PDGFR-α pathway is unlikely to disrupt tPA’s fibrinolytic function, then strategies specifically targeting the PDGF-CC/PDGFR-α pathway should maintain tPA’s beneficial thrombolytic activity while minimizing BBB dysfunction. Imatinib, also known as Gleevec or STI571, is an FDA-approved drug for the treatment of chronic myelogenous leukemia and other cancers, which binds to and inhibits several tyrosine kinases including PDGFR-α. In our study, Imatinib treatment reduced cerebrovascular permeability and stroke lesion volume as well as hemorrhagic complications associated with late thrombolysis, suggesting the possibility of an off-the-shelf adjunct therapy for use with thrombolytic tPA. Although Imatinib does not efficiently cross the BBB in healthy individuals, Imatinib is present in the CNS after oral administration42
. In addition, dose and time requirements for using Imatinib in stroke should be very different than in anti-tumor applications. A transient high dose of Imatinib may be all that is required to extend the therapeutic window for tPA. However, further studies are needed to explore this possibility since it remains to be determined whether late administration of a combination therapy of Imatinib plus tPA can extend the standard 3 h treatment window of tPA, by reducing hemorrhagic complications and restoring neuroprotection. Toward this goal, a randomized controlled clinical trial is currently planned by the stroke research team at Karolinska University Hospital in Stockholm Sweden (Prof. N. Wahlgren, Department of Neurology, Karolinska University Hospital, personal communication). This trial is designed to evaluate the safety and feasibility of Imatinib, either alone or with tPA, for use within the first few hours after the onset of ischemic stroke.
In summary our data offer a new model for regulation of cerebrovascular permeability by endogenous tPA and PDGF-CC, and describe a potential novel interventional approach that may extend tPA’s therapeutic window in ischemic stroke. Future studies may also examine if other CNS activities affected by tPA, such as seizure spreading, susceptibility to drug addiction, or anxiety-like behaviors43
, also involve tPA-mediated activation of PDGF-CC.