Abstract
Arango-Lievano M, Boussadia B, De Trieu D Terdonck L, Gault C, Fontanaud
P, Lafont C, Mollard P, Marchi N, Jeanneteau F Cell
Reports 2018;23:1045–1049. Reorganization of the neurovascular unit has been suggested in the
epileptic brain, although the dynamics and functional significance
remain unclear. Here, we tracked the in vivo dynamics of perivascular
mural cells as a function of electroencephalogram (EEG) activity
following status epilepticus. We segmented the cortical vascular bed to
provide a size- and type-specific analysis of mural cell plasticity
topologically. We find that mural cells are added and removed from
veins, arterioles, and capillaries after seizure induction. Loss of
mural cells is proportional to seizure severity and vascular pathology
(e.g., rigidity, perfusion, and permeability). Treatment with
platelet-derived growth factor subunits BB (PDGF-BB) reduced mural cell
loss, vascular pathology, and epileptiform EEG activity. We propose that
perivascular mural cells play a pivotal role in seizures and are
potential targets for reducing pathophysiology.
Commentary
The neurovascular system is a dynamic interface between the brain parenchyma and the circulatory system. One of its key functions is to establish the blood–brain barrier (BBB), which protects against harmful factors in the circulating blood from entering the brain while simultaneously providing controlled, bidirectional exchange of ions, sugars, proteins, and more. Further, the physical properties of the neurovascular system are variable, allowing efficient cerebral blood flow depending on metabolic needs through vasodilatation or vasoconstriction. This dynamically regulated system requires communication between cells of the vasculature and brain parenchyma. This interdependence is reflected in the concept of the neurovascular unit (NVU). Known players in the NVU include neurons, astrocytes, vascular endothelial cells, pericytes, smooth muscle cells, and the extracellular matrix. Over the past decades, a compromised NVU has been implicated in acute seizure genesis, chronic epileptogenesis, and pharmacoresistance to anticonvulsant therapy (1–4).
Recently, Arango-Lievano and colleagues published a sophisticated, longitudinal in vivo imaging study in mice gaining fascinating insights into neurovascular remodeling after kainic acid–induced status epilepticus (SE). Additionally, they tested whether attenuating SE-induced pathologic vascular changes can reduce seizure activity. Using two-photon imaging, the authors were able to follow multiple subclasses of vascular cells, BBB integrity, and vascular function in cortical layer I to III for up to 6 weeks. Because of the cellular complexity of the NVU, the authors use an elegant set of morphologic, genetic, and molecular approaches to track the topography of subclasses of pericytes and vascular smooth muscle cells (VSMCs) following SE. Interestingly, the findings of this study suggest that mural cells (an umbrella term encompassing pericytes and VSMCs) play a central role in seizure-induced neurovascular remodeling and may hold promise for developing vascular-directed antiepileptogenic therapies.
The most impressive aspect of this study is the longitudinal, high-resolution approach to imaging the complexities of the NVU. Their approach is sensitive and specific enough to assess differential cell- and vessel-subtype specific remodeling in vivo. For example, SE caused only small changes in VSMC and pericyte turnover as measured by percentage of cells proliferating or dying. Less than 0.6% of VSMCs or pericytes were found to be KI-67 (proliferation marker) or cleaved-caspase3 (apoptosis marker) positive at any point in time. Also, the net mural vessel coverage seemingly did not change much with SE. However, these rather small changes in “overall numbers” reflect significant loss and addition of mural cells that appear to have severe consequences for vessel physiology. Adding to the heterogeneity of the picture, significant VSMC remodeling was found at a small number of arterioles and small changes of pericytes were found at a large number of capillaries. Importantly, significant functional vessel anomalies (aneurysms, vasospasms) were present in a large proportion of precapillary arterioles, as well as in terminal arterioles and capillaries. These hemodynamically relevant changes correlated with SE severity and were persistent for weeks after SE. These results clearly show the complex effects of SE on the vascular system and support a correlation between the degree of NVU remodeling and the severity of seizures following SE. Assessing the local impact of individual site- or cell-specific postSE alterations on epileptogenesis would be of interest for future projects.
The authors go on to evaluate a rescue strategy to modify mural cell remodeling after SE by promoting the phosphorylation of platelet-derived growth factor receptor β (PDGFRβ) by intravenous application of PDGF-BB (PDGF homodimer consisting of two B subunits). This signaling pathway has been shown to be essential for maintaining cerebrovascular integrity and promotes mural cell growth (5). Interestingly, the authors observed enhanced mural cell growth postSE when animals were treated with PDGP-BB, compared with saline-treated animals. Importantly, on a functional level, cerebrovascular health after SE was improved with PDGF-BB treatment. PDGF-BB partially restored vasoregulatory function and was associated with a reduction of microbleeds, indicating reduced cerebrovascular damage after SE. Microbleeds have been associated with many pathophysiologic processes relevant for seizure disorders as a leaky BBB leads to serum electrolytes, proteins, and immune cell infiltration into brain parenchyma, triggering a plethora of both acute and chronic parenchymal alterations (6,7). Therefore, preventing or attenuating vascular remodeling after SE is a rational strategy to prevent later epileptogenesis.
A key question remains—does PDGF-BB treatment prevent postSE seizures? The kainic acid model of SE, mice generate spontaneous electrical and behavioral seizures after a latent phase of a few weeks (8). Interestingly, the authors found a significant reduction of spontaneous epileptiform EEG activity 2 weeks after SE and PDGF-BB treatment as measured by three different readouts (behavioral seizures, electrographic seizures, spike wave complexes). Epileptiform activity correlated with mural cell loss on arterioles and capillaries in saline-treated animals, but not in PDGF-BB–treated mice. However, caution must be applied because an antiepileptogenic effect could not be found 6 weeks after SE. Further studies are needed to clarify whether there was a transient, but prolonged anticonvulsant effect with PDGF treatment or a antiepileptogenic/disease modifying effect.“ In sum, these encouraging results hint at a disease modifying effect of restoring mural cell integrity on postSE epileptogenesis.
This study literally sheds light onto cell type and location specific differential cerebrovascular modulation following SE. Their findings are of high importance for epilepsy research and all other fields where changes of the NVU have pathophysiologic relevance. From a translational point of view, a more detailed understanding of NVU plasticity has crucial implications for drug development. Several strategies toward a ‘vascular’ antiepileptogenic drug are viable. On one hand, aiming to fully restore physiologic cerebrovascular function after insult is desirable because one could expect efficacy against a broad spectrum of indications. However, this approach more likely involves modulation of upstream cellular signaling pathways, which carries a higher risk for unwanted side effects (like the oncogenic potential of PDGF-BB mentioned by the authors). On the other hand, a strategy aimed at identifying and rectifying only those postinsult alterations relevant for a specific indication might be more feasible. A more selective cellular signaling intervention may be less prone to side effects but may be less effective at reducing seizures. For example, BBB leakage and subsequent brain parenchyma processes have been robustly linked to epileptogenesis: one could work on strategies for maintaining/reestablishing BBB integrity which likely prevents epileptogenesis and other pathophysiologic processes or, for example, target the transforming growth factor-β signaling cascade and identify a downstream effector suitable for pharmacologic manipulation to prevent epileptogenesis (9).
In any case, the elegant methodology used in the study provides important insight into the cerebrovascular changes happening after SE. Combining their approach of longitudinal in vivo studies to identify and locate relevant (patho)physiologic mechanisms with a higher throughput in vitro model of NVU subcompartments could lead to a better understanding of the vasculature's impact on brain function and eventually enable a vascular antiepileptogenic treatment. This study significantly advanced our understanding about mural cell plasticity in healthy and pathologic contexts and emphasizes the importance of non-neuronal cells in the pathophysiology of neurological disease.
So, whom do we call to fix seizures? The plumber or the electrician? For the sake of the patients—let us call both.