Vasogenic edema is the most common type of brain edema and a result from increased capillary permeability due to breakdown in intact brain-blood barrier (BBB) that is important for the maintenance of brain homeostasis. Vasogenic edema results in an abrupt increase in intracranial pressure and abnormal blood-brain transports of serum-derived molecules, which affect neuronal excitability, properties of regenerating (recovering) vessels, white matter injury and reactive gliosis in brain tumor, exposure of insecticides, traumatic brain injury (TBI), ischemia and brain inflammation (Papadopoulos et al., 2004; Song et al., 2004; Bloch et al., 2005; Kim et al., 2010b; Li and Ehrich, 2013; Kim Y.J. et al., 2015; Ko and Kang, 2015; Bennett et al., 2016; Zhao et al., 2018; Sun et al., 2019). Therefore, the prevention or attenuation of vasogenic edema formation is one of the major therapeutic strategies, which help to alleviate undesirable secondary complications in various neurological diseases.

Status epilepticus (SE) is a medical emergency condition with significant mortality showing continuous seizure activity. SE is considered as one of the precipitating factors for temporal lobe epilepsy (TLE) that is observed in one-third of all cases of epilepsy. Human TLE patients present neuronal loss and reactive astrogliosis in various brain regions including the hippocampus, which are influenced by SE (Fountain, 2000; Koh et al., 2005; Téllez-Zenteno and Hernández-Ronquillo, 2012; Sørensen and Kokaia, 2013). Similar to human patients, SE-induced epilepsy animal models show massive neuronal loss and astroglial dysfunctions in the various brain regions. Furthermore, vasogenic edema is one of the early post-SE events in the hippocampus and extrahippocampal limbic systems including the piriform cortex (PC). Since serum extravasation during vasogenic edema formation leads to spreading depolarizations and epileptiform discharges due to neuronal hyperexcitability, dysfunction of astrocytes and potassium buffering (Sisodiya et al., 2002; Löscher and Potschka, 2005; Kim Y.J. et al., 2015; Lippmann et al., 2017), it is likely that vasogenic edema may be one of the risk factors inducing pharmacoresistant epilepsy and life-threatening complications following SE.

The BBB consists of endothelial cells, astroglial endfeet, pericytes and basement membrane (BM; Banerjee and Bhat, 2007). The BM is vital for endothelial cell adhesion, migration and stabilization under physiological conditions. The BM is a structural and specialized complex of extracellular matrix protein including laminin, fibronectin, vitronectin, collagen, and heparin sulfate proteoglycans (Tilling et al., 2002; Yurchenco and Patton, 2009). Among them, laminin is a major glycoprotein component of BMs in vessels, and is involved in cellular adhesion and migratory processes (Miner, 2008). Thus, laminin is essential for vascular development, vessel dilation and physical integrity (Thyboll et al., 2002). Interestingly, laminin expression is remarkably upregulated in the BM of vessels following SE (Biagini et al., 2008; Gualtieri et al., 2012). Indeed, we have reported that laminin over-expression is closely relevant to the repair of BBB disruption accompanied by the reconstruction of endothelial barrier following SE (Kim et al., 2014). However, the underlying mechanism of up-regulation of laminin expression in vasogenic edema formation has not been fully elucidated.

The laminin functions are mediated by cell-surface receptors of two main classes: the integrins and the non-integrin 67-kDa laminin receptor (67-kDa LR; Lesot et al., 1983; Malinoff and Wicha, 1983; Rao et al., 1983). 67-kDa LR is composed of a 32 ∼ 33 kDa precursor (approximately 37 kDa in sodium dodecyl sulfate polyacrylamide gel electrophoresis). 67-kDa LR plays a role in cell adherence to laminin and stabilizes or modulates the binding of laminin to other receptors (Pellegrini et al., 1994; Ardini et al., 1998). In addition, 67-kDa LR is involved in laminin-mediated extracellular signal-regulated kinase 1/2 (ERK1/2), c-Jun N-terminal kinase (JNK) and p38 mitogen-activated protein kinase (p38 MAPK) signaling pathways (Givant-Horwitz et al., 2004). It also binds to several exogenous agents (including prion proteins, viruses and bacteria) and regulates cellular migration, viability and cytoskeleton reorganization (Wewer et al., 1986; Rao et al., 1989; Huang and Jong, 2009; Vana et al., 2009; Kielian et al., 2010). Furthermore, 67-kDa LR is expressed in the adult rat brain (Baloui et al., 2004). With respect to these profiles of 67-kDa LR, elucidating the correlation between 67-kDa LR expression and vasogenic edema formation may be noteworthy to understand the mechanisms of BBB disruption and laminin over-expression in response to SE. Therefore, during the course of this study, we investigated the alteration in 67-kDa LR expression induced by SE, and the effect of blockade of 67-kDa LR by neutralization on laminin expression or BBB integrity in efforts to understand the role of vascular dysfunction in epileptogenic insult.

Recently, we have reported that p38 MAPK-mediated VEGF over-expression leads to phosphatidylinositol-3-kinase (PI3K)/AKT-mediated eNOS activation during SE-induced vasogenic edema formation, which is also regulated by tumor necrosis factor-α (TNF-α)/nuclear factor-κB (NF-κB)/endothelin B (ET_B) receptor-mediated signaling pathway (Kim et al., 2016; Min and Kang, 2016; Kim and Kang, 2017). However, we could not elucidate the up-stream effectors/signaling pathways concerning p38 MAPK/VEGF-mediated vasogenic edema formation. In the present study, we found that 67-kDa LR might inhibit this signaling pathway by maintaining the integrity of dystrophin-AQP4 complex (Figure 6).

Consistent with previous studies (Castronovo et al., 1991; Guo et al., 1992), the present study shows that 67-kDa LR is identified in astrocytes and vascular endothelial cells. Furthermore, SE diminished its expression level in the PC. The present data also reveal that 67-kDa LR neutralization evoked the up-regulations of p38 MAPK phosphorylation, VEGF induction and eNOS expression in the normal animals, accompanied by serum extravasation and the reduced dystrophin expression, which were abrogated by SB202190. VEGF is a chemokine inducing impairment of BBB integrity, which potentially causes vasogenic edema (Pavlicek et al., 2000; Walter et al., 2001; Argaw et al., 2009; Bates, 2010; Kim et al., 2016; Kim and Kang, 2017). Aforementioned, 67-kDa LR and dystrophin expression levels regulate p38 MAPK activity (Givant-Horwitz et al., 2004; Smythe and Forwood, 2012; Wissing et al., 2014), and activation of p38 MAPK-VEGF axis participates in SE-induced vasogenic edema via the increased PI3K/AKT-mediated eNOS expression (Kim et al., 2016; Kim and Kang, 2017). Thus, our findings indicate that the reduced 67-kDa LR expression may be one of the up-stream molecules to regulate this signaling pathway during vasogenic edema formation. On the other hand, it is plausible that 67-kDa LR neutralization-induced up-regulations of p38 MAPK phosphorylation, VEGF induction, laminin and eNOS expression would be an adaptive reaction to rescue the BBB breakdown rather than a part of further detrimental mechanisms. If this hypothesis were true, the blockade of p38 MAPK signaling pathway would deteriorate vasogenic edema formation induced by 67-kDa LR neutralization. However, the present study reveals that SB202190 co-treatment mitigated serum extravasation induced by 67-kDa LR IgG infusion. Therefore, our findings suggest that 67-kDa LR may negatively regulate p38 MAPK/VEGF axis, which evokes BBB disruption.

In addition to VEGF, p38 MAPK activates cytosolic phospholipase A2 (cPLA2) that mediates arachidonic acid metabolism causing BBB dysfunction and edema. Indeed, p38 MAPK/cPLA2 pathway promotes BBB disruption with secondary vasogenic edema after ischemia-reperfusion injury (Nito et al., 2008). Furthermore, p38 MAPK activates Src family kinase regulates BBB permeability in response to VEGF. p38MAPK/VEGF/Src kinase pathway is also involved in BBB disruption and vasogenic edema formation via inflammation-associated BBB dysfunction in subarchnoid hemorrhage (Han et al., 2018), ischemic stroke (Liang et al., 2009) and cold-induced brain injury, but not experimental pneumococcal meningitis (Paul et al., 2007). Therefore, it is not excluded the possibilities that p38 MAPK-VEGF axis would lead to SE-induced vasogenic edema formation activating the cPLA2 and Src pathways.

Brain-blood barrier breakdown induces the leakage of albumin from blood into brain tissue. Albumin activates microglia and affects its cytokine releases and synthesis of inducible nitric oxide synthase (iNOS) via p38 MAPK pathway (Ralay Ranaivo and Wainwright, 2010; Yang et al., 2016). In addition, albumin results in the productions of inflammatory mediators, matrix metalloproteinase-9 (MMP-9) and myosin light chain kinase (MLCK) in astrocytes through p38 MAPK activation, which lead to disruption of vascular barrier integrity by the reorganization of the cytoskeleton and the dissociation of laminin as well as tight junctions (Rossi et al., 2011; Ralay Ranaivo et al., 2012; Kim Y.J. et al., 2015; Amtul et al., 2018). With respect to these previous reports, it is likely that extravasation of albumin may also reinforce p38 MAPK/VEGF signal pathway, accompanied by down-regulation of 67-kDa LR expression, following vasogenic edema formation.

The interaction between laminin and dystrophin plays an important role in the BBB integrity and AQP4 polarization in brain capillaries and astrocytes (Neely et al., 2001; Nico et al., 2003; Wolburg et al., 2009; Kim et al., 2010b; Sheen et al., 2011). Absence of dystrophin or deletion of base pairs in the dystrophin gene is a cause of Duchenne and Becker muscular dystrophy, which are allelic X-linked disorders with a progressive muscle weakness, a static cognitive impairment, autism and problems of behavior and attention (Young et al., 2008; de Brouwer et al., 2014). In mice models, deletion of dystrophin increases blood-retinal barrier (BRB) permeability by VEGF over-expression, and AQP4 delocalization/down-regulation (El Mathari et al., 2015; Vacca et al., 2016). In the brain, dystrophin deletion also reduces AQP4 expression over perivascular astroglial endfoot membranes (Enger et al., 2012). However, the regulatory signaling pathways for AQP4 expression have been still controversial. For example, p38 MAPK and JNK inhibitors decrease AQP4 protein levels in cultured human primary cortical astrocytes (Salman et al., 2017). In contrast, temozolomide (an effective drug for malignant glioma) decreases AQP4 expression level with increasing p38 MAPK phosphorylation, which is blocked by p38 MAPK inhibitor (Chen et al., 2017). Similarly, ERK1/2 pathway down-regulates AQP4 expression in scratch-injured astrocytes (Shi et al., 2015), while ERK1/2 activation up-regulates AQP4 expression in oxygen-glucose deprivation-recovery-induced injury (Qi et al., 2011). In the present study, 67-kDa LR neutralization diminished expression levels of dystrophin and AQP4 in the PC. These findings indicate that 67-kDa LR-mediated regulation of dystrophin expression may maintain AQP4 expression and BBB integrity under physiological condition. The present study also demonstrates that SB201290 did not affect expression levels of dystrophin and AQP4 in control IgG-infused animals. Furthermore, SB202190 co-treatment could not affect decreases in dystrophin and AQP4 expressions, although it abolished serum extravasation induced by 67-kDa LR neutralization. These findings indicate that impairment of dystrophin-AQP4 complex without activation of p38 MAPK/VEGF signaling pathway may not trigger BBB disruption in the PC, and that p38 MAPK activity may not influence expressions of dystrophin and AQP4. Indeed, AQP4 deletion does not affect the brain morphology and baseline intracranial pressure under physiological condition, while it worsens vasogenic edema in a freeze-injury model and a brain tumor edema model (Papadopoulos et al., 2004). Furthermore, acetazolamide (an AQP4 inhibitor) exacerbates vasogenic edema and astroglial loss in the PC following SE, although it did not induce vasogenic edema under physiological conditions (Kim et al., 2010b). Therefore, our findings suggest that the disruption of dystrophin may trigger BBB breakdown by activating p38 MAPK-mediated VEGF expression, and subsequently may aggravate vasogenic edema due to diminishing AQP4-dependent vasogenic water elimination.

Laminin is a key component of vascular BMs and contributes to vascular permeability by generating the physical barrier of the gliovascular BM and BBB maturation (Menezes et al., 2014). Laminin is also involved in the communication among the cellular components forming the gliovascular units via the integrins and non-integrin receptors (Lesot et al., 1983; Malinoff and Wicha, 1983; Rao et al., 1983; Menezes et al., 2014). In the present study, SE resulted in the down-regulated 67-kDa LR expression accompanied by laminin over-expression. Furthermore, 67-kDa LR neutralization evoked the similar effects in normal animals. Considering the essential roles of laminin in brain vascular development and vessel integrity (Thyboll et al., 2002) and its proteolytic cleavage by 67-kDa LR (Ardini et al., 2002), it is likely that the SE-induced reduction in 67-kDa LR expression and 67-kDa LR neutralization may up-regulate laminin expression to inhibit or recover vasogenic edema formation. Furthermore, SB202190 co-treatment abrogated laminin expression and serum extravasation induced by 67-kDa LR neutralization. Therefore, our findings indicate that 67-kDa LR dysfunction may up-regulate laminin expression via p38 MAPK/VEGF signaling pathway, and that laminin over-expression may be a secondary adaptive response to serum extravasation.

Similar to 67-kDa LR neutralization, recombinant tissue plasminogen activator (r-tPA) induces BBB disruption, and subsequently results in vasogenic edema (Copin et al., 2011), which is mitigated by p38 MAPK inhibitor (Garraud et al., 2016). Rosuvastatin (a lipid-lowering agent) also ameliorates r-tPA-induced BBB damage by reducing the activities of JNK and p38 MAPK (Lu et al., 2018). Indeed, tPA activates ERK1/2, JNK, AKT and p38 MAPK signaling pathways (Pineda et al., 2012). Thus, it is plausible that 67-kDa LR neutralization may lead to vasogenic edema through tPA activation and various signaling pathways as well as p38 MAPK. Further studies are needed to validate the relationship between 67-kDa LR and tPA and the related signaling pathways.

Pilocarpine is a cholinergic agent, which activates muscarinic M1 receptors (Hamilton et al., 1997; Kim and Kang, 2011). Acetylcholine induces vasodilation by eNOS activation (Gericke et al., 2013) and cholinesterase inhibitors such as chlorpyrifos and paraoxon (organophosphorus insecticides) evoke vasogenic edema (Song et al., 2004; Li and Ehrich, 2013). Thus, it is presumable that the direct PILO-mediated eNOS activation would affect seizure activity and vasogenic edema formation, although we blocked its peripheral cholinergic effects by atropine methylbromide in the present study. Recently, our laboratory reveals that N-nitro-L-arginine methyl ester (L-NAME, a non-selective NOS inhibitor) abrogates NO synthesis without altering seizure activity in response to PILO by the real-time simultaneous monitoring of NO and EEG. Thus, we have speculated that NO synthesis may be a consequent response to seizure activity, and that NO itself may not be directly involved in ictogenesis (Lee and Kim, 2018; Jeon and Kim, 2018). Furthermore, ET_B receptor antagonist attenuated vasogenic edema formation via reducing eNOS expression without altering PILO-induced seizure activity (Kim J.Y. et al., 2015). Therefore, it is unlikely that the direct cholinergic effect of PILO on eNOS activity may not influence seizure activity and vasogenic edema formation, at least in our model.

Various brain insults induce an osmotic potential across central necrotic tissue, inflammatory reaction, tight junction degradation and protein extravasation, which are regulated by inflammatory cytokines, MMP, VEGF and leukocyte/neutrophil infiltration (for review; Jha et al., 2019). Similar to other disease models, SE-induced vasogenic edema formation is relevant to neuroinflammation, MMP-9 activation and neutrophil infiltration (Kim et al., 2010a, 2012; Jung et al., 2012; Kim J.Y. et al., 2015). Thus, it is likely that the fundamental mechanisms of vasogenic edema formation induced by SE may be similar to those under various pathophysiological conditions. However, there are many underlying factors including age, sex and genetics to contribute vasogenic edema formation in animal model and human patients. For example, the major population of patients of posterior reversible encephalopathy syndrome (PRES), which is a neurotoxic encephalopathic state showing seizures and SE with white matter vasogenic edema in occipital and parietal lobes, is woman (Kamiya-Matsuoka and Tummala, 2017). Therefore, it seems that vasogenic edema would show gender-dependent characteristics. Indeed, sex steroids influences vascular reactivity in the brain via various signaling molecules including eNOS (Krause et al., 2006; Parikh et al., 2017). However, AQP4 expression is down-regulated in glioma patients without gender-dependent difference in AQP4 expression, although sex hormones such as estrogen, progesterone and testosterone modulate AQP4 expression in astrocytes (Warth et al., 2007). Ischemic stroke also shows the relative protection from brain injury in the female as compared with male, which is conferred by endogenous sex steroids. However, AQP4 is not involved in gender-specific differences in stroke volume and finally, the perivascular pool of AQP4 does not alter after ischemic stroke (Liu et al., 2008). Furthermore, age and genetic polymorphisms (e.g., cytokines, apolipoprotien E, Tau, and mitochondrial genes) affect BBB integrity (Song et al., 2004; Bennett et al., 2016). Therefore, studies concerning gender-, age-, and genetic-based difference in vasogenic edema formation in animal models and human patients are needed to elucidate the underlying mechanisms, the prediction of outcome and clinical management in various neurological diseases relevant to vasogenic edema.

In conclusion, to the best of our knowledge, the present data show novel evidence that 67-kDa LR may be one of the up-stream regulators of dystrophin/p38 MAPK/VEGF/eNOS axis, which plays an important role in the maintenance of BBB integrity (Figure 6). Thus, our findings suggest that 67-kDa LR-mediated astroglio-vascular interface may be a therapeutic target for treatment of vasogenic edema.

All datasets generated for this study are included in the manuscript and/or the Supplementary Files.

All experimental protocols described below were approved by the Institutional Animal Care and Use Committee of Hallym University (Chuncheon, South Korea). Every effort was made to reduce the number of animals employed and to minimize animal’s discomfort.

T-CK designed and supervised the project, analyzed the data, and wrote the manuscript. All authors performed the experiments described in the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.