Improving stroke outcomes in hyperglycemic mice by modulating tPA/NMDAR signaling to reduce inflammation and hemorrhages

ABSTRACT

The pharmacological intervention for ischemic stroke hinges on intravenous administration of the recombinant tissue-type plasminogen activator (rtPA, Alteplase/Actilyse) either as a standalone treatment or in conjunction with thrombectomy. However, despite its clinical significance, broader use of rtPA is constrained because of the risk of hemorrhagic transformations (HTs). Furthermore, the presence of diabetes or chronic hyperglycemia is associated with an elevated risk of HT subsequent to thrombolysis. This detrimental impact of tPA on the neurovascular unit in patients with hyperglycemia has been ascribed to its capacity to induce endothelial N-methyl-D-aspartate receptor (NMDAR) signaling, contributing to compromised blood-brain barrier integrity and neuroinflammatory processes. In a mouse model of thromboembolic stroke with chronic hyperglycemia, we assessed the effectiveness of rtPA and N-acetylcysteine (NAC) as thrombolytic agents. We also tested the effect of blocking tPA/NMDAR signaling using a monoclonal antibody, Glunomab. Magnetic resonance imaging, speckle contrast imaging, flow cytometry, and behavioral tasks were used to evaluate stroke outcomes. In hyperglycemic animals, treatment with rtPA resulted in lower recanalization rates and increased HTs. Conversely, NAC treatment reduced lesion sizes while mitigating HTs. After a single administration, either in standalone or combined with rtPA-induced thrombolysis, Glunomab reduced brain lesion volumes, HTs, and neuroinflammation after stroke, translating into improved neurological outcomes. Additionally, we demonstrated the therapeutic efficacy of Glunomab in combination with NAC or as a standalone strategy in chronic hyperglycemic animals. Counteracting tPA-dependent endothelial NMDAR signaling limits ischemic damages induced by both endogenous and exogenous tPA, including HTs and inflammatory processes after ischemic stroke in hyperglycemic animals.

Parabiosis Discriminates the Circulating, Endothelial, and Parenchymal Contributions of Endogenous Tissue-Type Plasminogen Activator to Stroke

BACKGROUND

Intravenous injection of alteplase, a recombinant tPA (tissue-type plasminogen activator) as a thrombolytic agent has revolutionized ischemic stroke management. However, tPA is a more complex enzyme than expected, being for instance able to promote thrombolysis, but at the same time, also able to influence neuronal survival and to affect the integrity of the blood-brain barrier. Accordingly, the respective impact of endogenous tPA expressed/present in the brain parenchyma versus in the circulation during stroke remains debated.

METHODS

To address this issue, we used mice with constitutive deletion of tPA (tPANull [tPA-deficient mice]) or conditional deletion of endothelial tPA (VECad [vascular endothelial-Cadherin-Cre-recombinase]-Cre∆tPA). We also developed parabioses between tPANull and wild-type mice (tPAWT), anticipating that a tPAWT donor would restore levels of tPA to normal ones, in the circulation but not in the brain parenchyma of a tPANull recipient. Stroke outcomes were investigated by magnetic resonance imaging in a thrombo-embolic or a thrombotic stroke model, induced by local thrombin injection or FeCl3 application on the endothelium, respectively.

RESULTS

First, our data show that endothelial tPA, released into the circulation after stroke onset, plays an overall beneficial role following thrombo-embolic stroke. Accordingly, after 24 hours, tPANull/tPANull parabionts displayed less spontaneous recanalization and reperfusion and larger infarcts compared with tPAWT/tPAWT littermates. However, when associated to tPAWT littermates, tPANull mice had similar perfusion deficits, but less severe brain infarcts. In the thrombotic stroke model, homo- and hetero-typic parabionts did not differ in the extent of brain damages and did not differentially recanalize and reperfuse.

CONCLUSIONS

Together, our data reveal that during thromboembolic stroke, endogenous circulating tPA from endothelial cells sustains a spontaneous recanalization and reperfusion of the tissue, thus, limiting the extension of ischemic lesions. In this context, the impact of endogenous parenchymal tPA is limited.

Blood tissue Plasminogen Activator (tPA) of liver origin contributes to neurovascular coupling involving brain endothelial N-Methyl-D-Aspartate (NMDA) receptors

BACKGROUND

Regulation of cerebral blood flow (CBF) directly influence brain functions and dysfunctions and involves complex mechanisms, including neurovascular coupling (NVC). It was suggested that the serine protease tissue-type plasminogen activator (tPA) could control CNV induced by whisker stimulation in rodents, through its action on N-methyl-D-Aspartate receptors (NMDARs). However, the origin of tPA and the location and mechanism of its action on NMDARs in relation to CNV remained debated.

METHODS

Here, we answered these issues using tPA^Null mice, conditional deletions of either endothelial tPA (VECad-Cre^ΔtPA) or endothelial GluN1 subunit of NMDARs (VECad-Cre^ΔGluN1), parabioses between wild-type and tPA^Null mice, hydrodynamic transfection-induced deletion of liver tPA, hepatectomy and pharmacological approaches.

RESULTS

We thus demonstrate that physiological concentrations of vascular tPA, achieved by the bradykinin type 2 receptors-dependent production and release of tPA from liver endothelial cells, promote NVC, through a mechanism dependent on brain endothelial NMDARs.

CONCLUSIONS

These data highlight a new mechanism of regulation of NVC involving both endothelial tPA and NMDARs.

The choroid plexus: a door between the blood and the brain for tissue-type plasminogen activator

Background

In the vascular compartment, the serine protease tissue-type plasminogen activator (tPA) promotes fibrinolysis, justifying its clinical use against vasculo-occlusive diseases. Accumulating evidence shows that circulating tPA (endogenous or exogenous) also controls brain physiopathological processes, like cerebrovascular reactivity, blood–brain barrier (BBB) homeostasis, inflammation and neuronal fate. Whether this occurs by direct actions on parenchymal cells and/or indirectly via barriers between the blood and the central nervous system (CNS) remains unclear. Here, we postulated that vascular tPA can reach the brain parenchyma via the blood-cerebrospinal fluid barrier (BCSFB), that relies on choroid plexus (CP) epithelial cells (CPECs).

Methods

We produced various reporter fusion proteins to track tPA in primary cultures of CPECs, in CP explants and in vivo in mice. We also investigated the mechanisms underlying tPA transport across the BCSFB, with pharmacological and molecular approaches.

Results

We first demonstrated that tPA can be internalized by CPECs in primary cultures and in ex vivo CPs explants. In vivo, tPA can also be internalized by CPECs both at their basal and apical sides. After intra-vascular administration, tPA can reach the cerebral spinal fluid (CSF) and the brain parenchyma. Further investigation allowed discovering that the transcytosis of tPA is mediated by Low-density-Lipoprotein Related Protein-1 (LRP1) expressed at the surface of CPECs and depends on the finger domain of tPA. Interestingly, albumin, which has a size comparable to that of tPA, does not normally cross the CPs, but switches to a transportable form when grafted to the finger domain of tPA.

Conclusions

These findings provide new insights on how vascular tPA can reach the brain parenchyma, and open therapeutic avenues for CNS disorders.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12987-022-00378-0.

Bumetanide lowers acute hydrocephalus in a rat model of subarachnoid hemorrhage

BACKGROUND

Subarachnoid hemorrhage (SAH) can lead to acute hydrocephalus (AH). AH pathophysiology is classically attributed to an obstruction of the arachnoid granulations by blood. Recent findings in rodents suggest that after intraventricular hemorrhage, AH is related to cerebrospinal fluid (CSF) hypersecretion by the choroid plexus (CP), as it can be reduced by intracerebroventricular (ICV) injection of bumetanide.

OBJECTIVE

Here, we investigated if and how CSF hypersecretion and/or CSF outflow disorders contribute to post-SAH hydrocephalus.

METHODS

Ninety-four Wistar rats were used. SAH was induced by the endovascular perforation technique. The presence of AH was confirmed by magnetic resonance imaging (MRI), and rats with AH were randomly assigned to 4 groups: control group, superior sagittal sinus (SSS) thrombosis to block CSF reabsorption, ICV injection of saline, and ICV injection of bumetanide to decrease CSF secretion. Clinical outcome was evaluated with a neuroscore. A second MRI was performed 24 h later to evaluate the ventricular volume.

RESULTS

Fifty percent of rats that survived SAH induction had AH. Their ventricular volume correlated well to the functional outcome after 24 h (r = 0.803). In rats with AH, 24 h later, ventricular volume remained equally increased in the absence of any further procedure. Similarly, ICV injection of saline or SSS thrombosis had no impact on the ventricular volume. However, ICV injection of bumetanide reduced AH by 35.9% (p = 0.002).

CONCLUSION

In rodents, post-SAH hydrocephalus is may be due to hypersecretion of CSF by the CP, as it is limited by ICV injection of bumetanide. However, we cannot exclude other mechanisms involved in post-SAH acute hydrocephalus.