| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EDITORIAL FOCUS
Departamento de Ciencias Fisiológicas, Pontificia Universidad Católica de Chile, Santiago, Chile
FROM THE FUNCTIONAL point of view, astrocytes have been long considered as "nursing cells" capable of maintaining the extracellular homeostasis and providing metabolic support for the neuronal functioning. Currently, important roles in spatial buffering mediated by gap junctions as well as in controlling neuronal activity through the glucose metabolism are accepted (1, 4). Under normal conditions, the main energy substrate of the brain is glucose metabolized via glycolysis in astrocytes and oxidative phosphorylation in neurons (2). The latter is favored by a tight metabolic balance between neurons and astrocytes, termed "neurometabolic coupling." This function is accomplished by glucose coming from glycogen stored in astrocytes and lasts for only a few seconds during periods of high demand. Blood-borne glucose is then shuttled over the blood-brain barrier through a mechanism called "neurobarrier coupling" (3). These metabolic responses occur with a close spatial and temporal correlation between high neuronal activity and hyperemia, described more than a century ago (7). To study how these two phenomena are linked, the development of new techniques, including two photon confocal microcopy and fluorescence imaging of Ca2+ indicators, were needed. Only recent in situ studies linked the intracellular astrocyte free [Ca2+] with changes in the brain microvascular tone (13). Consequently, astroglial cells have been proposed as sites for production of vasoactive factors in response to neuronal activity causing rapid and localized changes in cerebral blood flow upon demand (5, 10, 13).
Since the elucidation of a direct cause-effect relationship can be accomplished by eliminating the putative cause, a definitive demonstration of the astrocyte role in the neurovascular coupling had to wait for an ingenious experimental design proposed by Xu et al. (11). It is worth mentioning that they validated their main conclusion by performing numerous positive and negative controls, which I do not comment upon here due to space limitations. They chose the pial arterioles that overlie a thick layer of astrocytes, called the glia limitans. These glial cells isolate pial arterioles from neurons located right below and, thus, leave vascular cells and astrocytes as the potential signaling conduit of the neurovascular unit. They performed in vivo experiments in rats implemented with closed cranial windows. The surface arteriole (pial) response during high-intensity (seizure) and physiological neuronal activation (sciatic nerve stimulation) was recorded. The role of glia limitans and endothelium was assessed after selective chemical ablation of astrocytes and endothelial cells with L-
-aminoadipic acid and light plus dye (fluorescein)-induced injury, respectively. In the absence of astrocytes, the pial arteriolar-dilating responses to neuronal activation were completely blocked. Moreover, neither the bicuculline (the seizure-inducing molecule) nor the sciatic nerve stimulation-associated pial arteriolar dilation was altered in the model lacking the endothelial cells, supporting the role of astrocytes as a conduit of forward signals during neurovascular coupling.
Xu et al. (11) then studied the possible contribution of astrocytic connexin-based channels to the neuronal activation-associated pial arteriolar dilation. In their work, the involvement of intercellular communication mediated by connexin channels (i.e., gap junctions and/or hemichannels) was tested by using specific mimetic peptides. One is a connexin 43/37 (Cx43/37)-selective blocking peptide, gap-27, and the other is a Cx40/37-selective peptide, gap-26. They found attenuation of the bicuculline- or sciatic nerve stimulation-induced pial arteriolar dilation with gap-27. Nevertheless, gap-26 was without effect, supporting the role of Cx43-based gap junction channels and/or hemichannels. Because Cx40 and Cx37 are expressed in cerebral endothelial cells and Cx43 is expressed to a greater extent in astrocytes than in vascular cells, these findings support the role of astrocytes in coordinating the neuronal signals transmitted to pial arterioles. The mimetic peptides are supposed to prevent the formation of new gap junction channels (with half lives of 2 to 3 h) or to directly close hemichannels through binding to extracellular domains. Hence, the gap-26 peptide application (45 min suffusion) was likely to block hemichannels rather than gap junction channels. Nevertheless, an elucidation of this issue requires additional experimental paradigms.
In support of the role of connexin-based channels in the neurovascular coupling, cortical astrocytes are known to present unapposed Cx43 hemichannels on their surface and to communicate to each other via gap junction channels (6). Moreover, the hemichannels open either spontaneously or in response to external stimuli allowing the release of signaling molecules (i.e., ATP) (9). In brain slices, extracellular ATP mediates the propagation of Ca2+ waves (8) that seem to coordinate neurovascular responses (13). Although astrocytes can release other signaling molecules (i.e., glutamate) through hemichannels (12), their putative roles in generating Ca2+ waves (13) require demonstration.
Since astrocytic end feet and arterioles are separated by 
20 nm, vasoactive signals are likely to be transferred via secretion of soluble factors. In support of this notion, Xu et al. (11) demonstrated that an increase in suffusion rate is sufficient to wash out the vasodilating signal induced either by sciatic nerve stimulation or bicuculline treatment. Therefore, with a set of relevant questions and a great deal of imagination, Xu et al. (11) have recently provided solid information indicating that cortical astrocytes act as mediators of the neurovascular unit and release paracrine signals to control vasodilation, both under normal and pathological conditions. Since astrocytes are capable of synthesizing and/or storing numerous bioactive compounds and respond to diverse stimuli, they most likely are crucial in both vasodilation and vasoconstriction responses, and therefore more physiological and pathological studies related to the neurovascular unit could find inspiration in the work of Xu and colleagues (11).
FOOTNOTES
Address for reprint requests and other correspondence: J. C. Sáez, Dept. de Ciencias Fisiológicas, Pontificia Univ. Católica de Chile, Alameda 340, Santiago, Chile (e-mail: jsaez{at}bio.puc.cl)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
