AJP - Heart Watch the video to see how APS reaches out to developing nations.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 294: H622-H632, 2008. First published November 30, 2007; doi:10.1152/ajpheart.00530.2007
0363-6135/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
294/2/H622    most recent
00530.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, H.-L.
Right arrow Articles by Pelligrino, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, H.-L.
Right arrow Articles by Pelligrino, D. A.

Astrocytes are a key conduit for upstream signaling of vasodilation during cerebral cortical neuronal activation in vivo

Hao-Liang Xu, Lizhen Mao, Shuhua Ye, Chanannait Paisansathan, Francesco Vetri, and Dale A. Pelligrino

Neuroanesthesia Research Laboratory, University of Illinois at Chicago, Chicago, Illinois

Submitted 3 April 2007 ; accepted in final form 26 November 2007


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Astrocytes play an important role in the coupling between neuronal activity and brain blood flow via their capacity to "sense" neuronal activity and transmit that information to parenchymal arterioles. Here we show another role for astrocytes in neurovascular coupling: the ability to act as a signaling conduit for the vitally important process of upstream vasodilation (represented by pial arterioles) during both excessive (seizure) and physiological (sciatic nerve stimulation) increases in cerebral cortical neuronal activity. The predominance of an astrocytic rather than a vascular route was indicated by data showing that pial arteriolar-dilating responses to neuronal activation were completely blocked following selective disruption of the superficial glia limitans, whereas interference with interendothelial signaling was without effect. Results also revealed contributions from connexin 43, implying a role for gap junctions and/or hemichannels in the signaling process and that signaling from the glia limitans to pial arterioles may involve a diffusible mediator.

bicuculline; connexin; gap junction; glia limitans; L-{alpha}-aminoadipic acid; sciatic nerve stimulation

ASTROCYTES EXHIBIT INTIMATE anatomic relationships with cerebral arterioles and neurons. In the brain parenchyma, the extensive ensheathment of cerebral arterioles by astrocytic endfeet far exceeds any direct neural contacts with those perfusion-regulating microvessels. That unique arrangement permits astrocytes to transduce signals arising from activated neurons and to transmit that information to the cerebral microcirculation. These forwarded signals may involve interastrocytic communication and the eventual release of vasoactive factors from the endfeet of astrocytes in contact with resistance vessels, with the result frequently being vasodilation. The above has been labeled "neurovascular coupling," and its components (i.e., neuron, astrocyte, and vascular cells) have been referred to as the "neurovascular unit" (1, 14, 17). It has been envisaged that the components of the neurovascular unit are integrated in such a way as to match local perfusion to local neuronal requirements. A potentially important, yet often overlooked, factor in that integration is multisegmental vasodilation; that is, vasorelaxation of upstream arteriolar segments often need to occur to permit more blood to reach dilated downstream vessels. Without it, insufficient nutrient delivery may result, placing neurons at risk, especially during episodes of intense neuronal activity, such as seizure. The existence of upstream vasodilation in the periphery is well documented (e.g., Ref. 13). Mechanisms include intercellular conduction of vasodilating signals via cells of the vascular wall and flow/shear stress-related mechanisms. However, in the brain, experimental findings have yielded no clear picture regarding the role for such mechanisms in the upstream vasodilation accompanying neural activation [see reviews by Andresen et al. (2) and Drake and Iadecola (8)].

Another unique characteristic of the cerebral circulation relates to the architecture of the arteriolar system. For example, in the cerebral cortex, parenchymal arterioles arise from surface (pial) arterioles, with the two segments linked by penetrating arterioles. Pial arterioles, therefore, represent important upstream vascular segments. However, it is unclear how vasodilating signals, originating in cortical neurons, reach the pial arterioles; that is, the pial arterioles overlie a thick layer of astrocytic processes termed the glia limitans. This essentially isolates pial arterioles, anatomically, from the neurons below. Since pial arterioles are known to dilate in association with a variety of neuronal activation models (17) and in light of the seemingly modest or nonexistent influence from shear-related factors (2), this would seem to leave intercellular conduction as a primary mechanism supporting upstream dilation, with vascular cells and astrocytes as the potential signaling conduits.

In the present study, we addressed the hypothesis that astrocytes, via the glia limitans, as well as vascular endothelium, play a role in the pial arteriolar dilation arising from neural activation. We further hypothesized that connexin-related intercellular communication [i.e., via gap junction and/or hemichannels (5)] contributes to neuronal activation-associated pial arteriolar dilation. Thus, in rats equipped with closed cranial windows, we monitored pial vascular responses during high-intensity (seizure) and physiological [sciatic nerve stimulation (SNS)] neuronal activation. To assess the roles of the glia limitans and endothelium in these responses, we employed two well-established in vivo models: one involving selective chemical ablation of the superficial glia limitans (33, 36) and the other involving light plus dye (L/D)-induced pial arteriolar endothelial injury (36). To address the role of gap junctions/hemichannels, we used topically applied blocking peptides (see Ref. 10, 19, and 35) representing amino acid sequences found in the external loops of connexin (Cx) 43, Cx37, and Cx40.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animal preparation. The study protocol was approved by the Institutional Animal Care and Use Committee. Adult female Sprague-Dawley rats (~200–250 g; and ~2 to 3 mo of age) were used. Most rats were prepared with closed cranial windows 24 h before the study. The procedure for placement of cranial windows was described in a previous article from our laboratory (32). Briefly, the cranial window was placed in the presence of 2% isoflurane/70% N2O-30% O2 anesthesia with the rat stabilized in a head-holder. A 10-mm-diameter craniotomy was performed over the skull midline. The dura was removed carefully to keep the sagittal sinus intact. An 11-mm-diameter glass window outfitted with three ports was glued to the skull using cyanoacrylate. The skin overlying the window was sutured, and the animals were permitted to recover. On the day of study, the rats were anesthetized with isoflurane. Paralysis was then induced with curare (1 mg/kg iv), followed by tracheotomy and mechanical ventilation. During surgery, anesthesia was maintained with 1.2% isoflurane/70% N2O-30% O2. The femoral arteries and veins were cannulated for blood sampling, arterial pressure monitoring, and drug infusions. Rectal temperature was servo-controlled at 37°C with a heating pad. The cranial window was reexposed, and three stainless steel screws were inserted into the skull, along the periphery of the cranial window, for electroencephalogram (EEG) recording. Cannulae were then connected to the three ports. After completion of the surgery, isoflurane was discontinued, and the rat was maintained on fentanyl (10 µg/kg loading dose, and 25 µg·kg–1·h–1 thereafter)/70% N2O-30% O2. Subsequently, a cortical surface suffusion was initiated (at 0.5 ml/min via the cranial window) with a temperature, pH, and PCO2-controlled artificial cerebrospinal fluid (aCSF) solution (32). Rectal temperature, mean arterial blood pressure (MABP), and intracranial pressure were monitored and maintained within normal limits. The pressure under the window was kept at 5–10 mmHg by adjusting the height of the outflow cannula.

Neuronal activation models. The rats were subjected to one of two neuronal activation paradigms: SNS or bicuculline-induced seizure. For the former, the contralateral sciatic nerve was dissected in the region of the sciatic notch. The proximal end was placed on a stimulating silver ring electrode and covered with silicone rubber. The nerve was covered with mineral oil and electrically stimulated for 20 s with square wave pulses (current, 0.7 mA; frequency, 2 Hz; and duration, 0.1 ms) using a Stimuplex Dig RC Nerve stimulator (B. Braun Medical, Bethlehem, PA). Pial arterioles in the contralateral hindlimb projection area of the somatosensory cortex (~2 mm caudal and lateral to the bregma) were selected for study. The same criteria were used for the selection of arterioles in seizure experiments. Pial arterioles (25–45 µm in diameter) were viewed using the x20 objective of a Nikon microscope equipped with an epi-illumination darkfield system (32). Images were captured using a digital video camera (CoolSNAP ES, Fryer, Huntley, IL), projected on a computer monitor, and saved for subsequent diameter measurements, using the MetaMorph software system (Universal Imaging, Downingtown, PA). The total magnification of the image displayed on the computer monitor was approximately x800. Diameter values from at least three arterioles were obtained and averaged. A continuous recording of pial arterioles was obtained, and the peak diameter increase value was used for comparisons. For the second neural activation model, 100 µM bicuculline methiodide was suffused for 20 min via the cranial window, and diameter measurements of pial arterioles were obtained before and at 2, 3, 5, 7, 10, 15, and 20 min of bicuculline suffusion. Bilateral EEG was monitored using an Aspect A-1000 (Framingham, MA). The monitoring system, as employed in previous studies from our laboratory (22, 30), provided EEG total power analyses, based on a fast-Fourier transform of frequency and amplitude values and expressed in decibel units.

In vivo glia limitans and endothelial injury models. Evaluation of the participation of astrocytes in neural activation-induced pial arteriolar dilation used a strategy whereby the superficial portion of the glia limitans was selectively ablated. This involves exposure to the gliotoxin, L-{alpha}-aminoadipic acid (L-AAA, 300 µl of a 2 mM aCSF solution; see Ref. 33), introduced into the space under the cranial window 24 h before study. Control rats received vehicle only. In previous studies, it was shown that topically applied L-AAA neither had any direct effects on pial arteriolar endothelium-dependent vasodilating responses [acetylcholine (ACh)] and pial vascular smooth muscle [S-nitrosoacetyl-penicillamine (SNAP)] reactivity nor altered cortical-evoked electrical responses to somatosensory stimulation (33, 36). The last is indicative of a lack of any effect on neuronal function. Assessments of endothelial contributions were performed using an established L/D endothelial injury model (36). This procedure produces an endothelium-specific injury in illuminated segments of pial arterioles, thus preventing endothelium-dependent dilations and responses conducted via the endothelium. Mercury light was passed first through a filter that allows transmission only at 450–490 nm and then, coaxially, through the microscope objective. The light was focused on an area that included the vessels being studied. Two percent fluorescein dye (0.8 ml/100 g body wt) was given intravenously, and the light was turned off 55 s after the initiation of the dye infusion. Control rats were subjected to the repeat sequence without L/D exposure.

Experimental groups and protocols. The same set of pial arterioles was monitored throughout the course of each experiment. Normal vascular function was established at the start of each experiment via testing CO2 reactivity (i.e., response to ~25 mmHg increase in arterial PCO2; see Refs. 29 and 31). Although rarely observed, any rat not exhibiting a CO2 reactivity of >1.0% diameter increase/mmHg PCO2 increase would have been excluded. The only exception was in studies involving L-AAA treatment, since L-AAA exposure has been associated with an attenuated pial arteriolar CO2 response (33). There were five experimental groups. Each of these was divided into two subgroups: bicuculline-induced seizure or SNS. The first group was designed to test the hypothesis that the pial arteriolar dilation arising from bicuculline suffusion or SNS is completely dependent on neural activation. To that end, we topically applied the Na+ channel/action potential blocker, tetrodotoxin (TTX, 0.1 µM). The experimental sequence was as follows: 40–45 min suffusion with drug-free aCSF->bicuculline suffusion (20 min) or SNS (20 s)->30 min drug-free aCSF->30 min TTX (0.1 µM) suffusion->repeat bicuculline or SNS.

The second experimental group involved L-AAA-treated rats and their vehicle-treated controls (see above). Thus, on the day after window preparation and L-AAA/vehicle application, following 40–45 min suffusion with drug-free aCSF, bicuculline suffusion (20 min) or SNS (20 s) was applied. At the end of some L-AAA (and their vehicle control) experiments, normal reactivity to 1 µM SNAP (>30% diameter increase) was confirmed. Subsequently, brains were perfusion-fixed with 4% paraformaldehyde, in PBS, introduced transcardially. Eight-micrometer coronal sections from paraffin-embedded blocks were prepared using a microtome. The cerebral expression of the astrocytic marker, glial fibrillary acidic protein (GFAP), was examined using immunohistochemistry, as done previously in our laboratory (33), to confirm the presence of injury to the glia limitans. Additional coronal sections from control brains were prepared for immunohistochemical analysis of the neuronal dendritic marker, microtubule-associated protein-2 (MAP2), as well as coexpression of GFAP and Cx43 (29, 31). For GFAP, the primary antibody used was a goat polyclonal (diluted 250:1; Dako). The secondary donkey anti-goat antibody was conjugated to dichlorotriazinylaminofluorescein. For Cx43 detection, a monoclonal primary antibody was used (2,000:1 dilution, BD Transduction; Lexington, KY), along with a goat anti-mouse secondary antibody conjugated to Cy3. The primary MAP2 antibody was a mouse monoclonal diluted 200:1 from Chemicon (Temecula, CA). The secondary antibody was a goat-anti-mouse conjugated to dichlorotriazinylaminofluorescein.

In experiment 3 (L/D endothelial injury), the following sequence was used: 40–45 min suffusion of drug-free aCSF->suffusion of 0.1 and then 1.0 µM SNAP (5 min each dose for determination of vascular smooth muscle reactivity)->drug-free aCSF (20 min)->suffusion of 10 and then 100 µM ACh (5 min each dose for determination of endothelial functional integrity)->drug-free aCSF (20 min)->bicuculline or SNS. The sequence given above was then repeated after 45–60 min. However, just before initiating the repeat sequence, the rats were exposed to the L/D endothelial injury procedure (see above). Repeating the ACh and SNAP suffusions permitted us to confirm a loss of endothelial function in the absence of vascular smooth muscle damage.

In experiment 4, we evaluated the possible role of connexin-related intercellular communication (i.e., via gap junctions and/or hemichannels) in neurally evoked pial arteriolar dilations using the Cx43/37-selective blocking peptide, gap-27, and the Cx40/37-selective peptide, gap-26 (35). To that end, following the initial measurement of pial arteriolar diameter changes during SNS or during bicuculline exposure, baseline conditions were reestablished. After 20 min, a suffusion of gap-27 (SRPTEKTIFII, 300 µM) or gap-26 (VCYDQAFPISHIR, 300 µM) was initiated. Forty-five minutes later, the neural activation was repeated.

The fifth group of experiments involved examining the influence of increased cortical suffusion rate on SNS- and bicuculline-induced pial arteriolar dilations. This was done to detect the presence of diffusible (paracrine)-vasodilating substances that might be generated in the tissue underlying the pial arterioles and thus be susceptible to washout by increased suffusate flow. For these experiments, the aCSF suffusion rate was increased from 0.5 to 1.5 ml/min during successive SNS or bicuculline exposures, separated by 60 min. The height of the aCSF outflow cannula was adjusted to hold the intracranial pressure constant when the suffusion rate was increased.

In each experiment, the MABP was continuously monitored, and arterial blood samples were taken at 30-min intervals for arterial partial pressures of O2 and CO2 and pH analysis using a Radiometer Copenhagen blood gas/pH analyzer (model ABL 520).

All reagents were obtained from Sigma-Aldrich (St. Louis, MO) and dissolved in aCSF. Statistical comparisons of pial arterioles diameter changes and EEG power, during seizure, in groups 1, 3, 4, and 5 were made using a two-way repeated measures analysis of variance (ANOVA), with a post hoc Student-Newman-Keuls test for multiple comparisons (SigmaStat). Data obtained from the same groups in SNS experiments were analyzed using a paired Student's t-test. For group-2 seizure experiments, a one-way ANOVA was used, with a post hoc Tukey analysis applied to compare responses at identical time points in L-AAA-treated versus control rats. In group-2 SNS experiments, an unpaired Student's t-test was used. A level of P < 0.05 was considered significant in all statistical tests. To facilitate presentation, seizure data in the figures are given as the mean values (percent diameter change or EEG power) ± SE for each time point. Data for pial arteriolar responses to SNS are presented as mean of the peak diameter increases ± SE.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Arterial PO2 values were maintained above 100 mmHg in all rats studied, whereas PaCO2, pH, and MABP remained within normal limits (i.e., 32–40 mmHg, 7.35–7.45, and 100–130 mmHg, respectively) over the course of the experiments. Furthermore, no significant differences were observed when comparing PaCO2, pH, and MABP values at equivalent experimental time points in the L-AAA and vehicle groups. Table 1 summarizes the baseline pial arteriolar diameter values (means ± SE) measured just before the initial application of vasodilating stimuli and, except for L-AAA experiments and their vehicle controls, immediately preceding the second round of vasodilating stimuli. With only one exception (a 12% lower diameter following TTX treatment in rats exposed to SNS), no significant changes in baseline diameters were observed.


View this table:
[in this window]
[in a new window]

  		Table 1. Pial arteriolar diameter values measured immediately before onset of initial and 2nd neural activations

 
Neural activation-induced pial arteriolar dilation: reproducibility and effect of TTX. During bicuculline suffusion, neuronal activity was significantly enhanced, as evidenced by the appearance of seizure-like EEG patterns (Fig. 1A) and a substantial increase in total EEG power (Fig. 1C). This neuronal activation was accompanied by a rapid and sustained pial arteriolar dilation (Fig. 1B). SNS was associated with profound pial arteriolar diameter increases in the contralateral hindlimb somatosensory area (Fig. 2). Additionally, repeat stimulations (applied following 60 min of drug-free aCSF suffusion after the previous exposure to bicuculline or SNS) evoked the identical vasodilations (Figs. 1B and 2B) as measured initially in the same pial arterioles. This indicates the reversibility and reproducibility of the response. Topical application of the Na+ channel and nerve action potential blocker, TTX, eliminated bicuculline-induced pial arteriolar diameter increases and EEG changes (Fig. 3, A and B). TTX showed the same effect on the pial arteriolar response to SNS (Fig. 3C). Information regarding the proximity of subpial neural elements to the pial surface was obtained through examining the immunoreactivity of MAP2 [a widely-accepted marker for neuronal dendrites (6)], in the outer ~200 µm of the somatosensory cortex (Fig. 3D). This provided additional clues, beyond vascular and electrical response data, as to the accessibility of suffused bicuculline and TTX to their neuronal targets. Except for the outermost ~20 µm of subpial cortical layer I (where MAP2 immunoreactivity was weak), we did observe a rather extensive dendritic network within layer I. As described elsewhere (7, 38), this general pattern of distribution may apply to both axons and dendrites. This implies that topically applied neuroactive agents, like bicuculline and TTX, do not need to diffuse very far from the pial surface to encounter neural elements.


Figure 1
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 1. A 20-min topical suffusion of bicuculline methiodide (100 µM) is accompanied by changes in the cortical electroencephalogram (EEG) and pial arteriolar dilations. A: representative EEG recordings showing the burst and suppression patterns typical of seizure. Brain electrical activity returns to normal within 1 h of cessation of bicuculline suffusion. B: pial arteriolar diameter increases during seizure (expressed as a percent increase from resting diameter). The data at each time point are presented as means ± SE (n = 4 rats). The data were analyzed using a 2-way repeated-measures ANOVA, with a post hoc Student-Newman-Keuls test. The standard error of the least square mean for treatment (initial vs. repeat) x time = 3.1. Bicuculline suffusion elicits a rapid and sustained pial arteriolar dilation that is reproducible upon repeat bicuculline suffusion, initiated 60 min after cessation of initial exposure. C: EEG total power is expressed in decibel unit increases (22, 30) relative to baseline levels (means ± SE at each time point; n = 4 rats). Data analysis was performed as described in B. The standard error of the least square mean for treatment (initial vs. repeat) x time = 0.65. A nearly identical pattern of EEG power increase is observed upon repeating the bicuculline suffusion 1 h after the initial exposure. The data in B and C represent time controls for subsequent experiments.

 

Figure 2
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 2. Time-related pattern and reproducibility of the pial arteriolar diameter increases accompanying 20 s of sciatic nerve stimulation (SNS). A: raw data obtained in 3 representative rats show the typical pial arteriolar responses characterized by a delayed onset of vasodilation following initiation of SNS, rising to a peak (at 30–40 s), and then falling toward baseline thereafter. B: time controls for subsequent experiments. The peak percent pial arteriolar increase, relative to baseline, as well as the time-to-peak response (means ± SE; n = 4 rats) showed exceptional reproducibility when comparing two SNSs separated by 1 h.

 

Figure 3
View larger version (60K):
[in this window]
[in a new window]

  		Fig. 3. Effects of the Na+ channel/action potential inhibitor, tetrodotoxin (TTX, 0.1 µM; 30 min suffusion initiated 30 min following initial bicuculline exposure or SNS) on pial arteriolar and electrical responses elicited by bicuculline or SNS. A: seizure-induced pial arteriolar dilations were completely blocked in the presence of TTX. The data at each time point are presented as means ± SE (n = 4 rats). A 2-way repeated-measures ANOVA (post hoc Student-Newman-Keuls test) was used for data analysis. The standard error of the least square mean for treatment (initial vs.TTX) x time = 1.8. *P < 0.05 vs. initial. B: TTX was associated with a complete loss of EEG power increases during bicuculline. Power data are expressed in decibel power unit increases relative to baseline levels (means ± SE at each time point; n = 4 rats). Data analysis was performed as in the above. The standard error of the least square mean for treatment (initial vs. TTX) x time = 0.53. C: SNS-induced vasodilation was totally blocked in the presence of TTX (means ± SE; n = 4 rats; *P < 0.05 vs. initial). D: the neuronal dendritic marker, microtubule-associated protein 2, was well expressed within layer I of the somatosensory cortex but lacking in the outermost ~20 µm of cortical tissue. Scale bar = 100 µm.

 
The role of astrocytes. In a recent publication, we reported that 24 h of exposure to the selective gliotoxin, L-AAA, applied to the cortical surface (via a cranial window), elicits a reproducible glia limitans injury, one that is not accompanied by any functional changes in endothelial cells, smooth muscle cells, and neurons (36). A representative example of the pattern of astrocytic damage (GFAP immunoreactivity) elicited by topical L-AAA is given in Fig. 4B (along with a vehicle-treated control; Fig. 4A). Note the substantial disruption of the glia limitans, with normal-appearing astrocytes seen beyond 100 µm below the cortical surface. Despite the complete loss of the bicuculline-associated pial arteriolar dilations following L-AAA exposure, the total EEG power increases associated with bicuculline suffusion were unaffected (Fig. 4, C and D). This is consistent with an absence of any influence of L-AAA on neuronal function, confirming findings from our earlier study (33). In addition, pial arteriolar responses to SNS were completely abolished in the presence of L-AAA (Fig. 4E). Thus astrocytes (especially those comprising the glia limitans) appear to play a vital role in the pial arteriolar dilations associated with neuronal activation.


Figure 4
View larger version (31K):
[in this window]
[in a new window]

  		Fig. 4. Influence of 24 h of exposure to the selective gliotoxin, L-{alpha}-aminoadipic acid (L-AAA; 2 mM, topically applied), on integrity of the glia limitans (GL), as well as pial arteriolar and electrical responses elicited by bicuculline or pial arteriolar reactivity to SNS. A: immunoreactivity of the astrocytic marker, glial fibrillary acidic protein (GFAP), in a control rat (scale bar = 50 µm). B: GFAP immunoreactivity in a rat treated with L-AAA. Note the loss of GL integrity and the diffuse staining pattern of GFAP extending down 50–100 µm below the cortical surface. That pattern probably reflects fragments of the astrocytic intermediate filament protein, GFAP, following cell lysis. C: pial arteriolar responses to bicuculline in L-AAA (n = 5) vs. vehicle-treated (n = 7) rats (data presented as means ± SE at each time point; *P < 0.05 vs. control-data analysis performed via a 1-way repeated-measures ANOVA, with a post hoc Tukey test). Note the complete loss of bicuculline-induced vasodilation in the GL-disrupted animals. D: EEG power changes in control vs. vehicle-treated controls. There was no detectable effect of L-AAA vs. control on seizure-induced increases in electrical activity (data analysis as in Fig. 1C). Contrast this with the complete blockade of increased EEG power seen in the presence of TTX (Fig. 3B). E: SNS-induced vasodilation was almost completely lost in the presence of L-AAA (n = 5 rats) vs. vehicle (n = 5 rats) (means ± SE; *P < 0.05 vs. control).

 
Role of endothelium. L/D endothelium injury selectively damages endothelial cells of pial arterioles, without affecting smooth muscle cells; that is, as shown in previous studies from our laboratory (33, 35), the L/D model we used completely suppressed ACh-induced pial arteriolar dilations but did not affect responses to the direct smooth muscle acting nitric oxide donor, SNAP (Fig. 5, A and B). We found that neither bicuculline nor SNS-associated pial arteriolar dilations were altered by L/D (Fig. 5, C and D), suggesting the lack of involvement of the endothelium in the transmission of neuronally evoked vasodilating signals to the pial arterioles.


Figure 5
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 5. Effects of endothelial injury, using light plus intravascular dye (L/D) exposure, on pial arteriolar reactivity. A: endothelial injury was accompanied by a significant and complete loss of pial arteriolar reactivity to the endothelium-dependent vasodilator, acetylcholine [10 and then 100 µM, topically applied standard error of the least square mean for treatment (initial vs. L/D) x dose = 2.3]. B: no change in the vasodilating response elicited by the nitric oxide donor, S-nitrosoacetyl-penicillamine (SNAP; 0.1 then 1.0 µM, topically applied), was observed when comparing the initial response to the response measured following L/D [standard error of the least square mean for treatment (initial vs. L/D) x dose = 2.7]. This indicates that L/D did not injure pial arteriolar smooth muscle. C: endothelial injury did not affect the pial arteriolar response to bicuculline suffusion. The standard error of the least square mean for initial vs. L/D x time = 1.5. Statistical analyses in A–C were performed using a 2-way repeated-measures ANOVA (post hoc Student-Newman-Keuls test). D: there were no changes in the peak pial arteriolar dilations to SNS when comparing initial to post-SNS responses. Values are means ± SE; n = 5 rats; *P < 0.05 vs. initial.

 
Role of Cx43 gap junctions/hemichannels. Application of gap-27 peptide, which blocks Cx43- and Cx37-containing gap junctions and/or hemichannels (10, 19), attenuated bicuculline-induced pial arteriolar dilation (by ~50%), without altering neuronal activation (Fig. 6, A and B). A similar result was obtained with the SNS-associated pial arteriolar response (Fig. 6D), although the degree of reduction in the vasodilating response (~75%) was somewhat greater. Exposure to the Cx40/Cx37 inhibitory peptide, gap-26 (300 µM), was without effect on bicuculline- or SNS-induced pial arteriolar dilations (Fig. 6, C and E). These findings support a role for Cx43-containing gap junctions/hemichannels in the pial arteriolar responses. Moreover, Cx43 is highly expressed in astrocytes and is especially concentrated in the superficial glia limitans (Fig. 6F) but only weakly expressed in vascular cells (26, 35). On the other hand, Cx40 and Cx37, the expression of which has been identified in cerebral vascular cells (35), are not found in astrocytes. Thus, similar to the results obtained using astrocyte and endothelial injury strategies, the results of experiments targeting gap junctions/hemichannels also support a role for astrocytes and the glia limitans, rather than vascular cells, in the communication of vasodilating signals to pial arterioles.


Figure 6
View larger version (30K):
[in this window]
[in a new window]

  		Fig. 6. Role of connexin (Cx)43-dependent intercellular communication (via gap junctions and/or hemichannels) on pial arteriolar and electrical responses elicited by bicuculline or pial arteriolar reactivity to SNS. A 2-way repeated-measures ANOVA (post hoc Student-Newman-Keuls test) was used for data analysis in A–C. A: the Cx43/Cx37-specific blocking peptide, gap-27 (300 µM topically) was accompanied by a ~50% reduction in bicuculline-induced pial arteriolar dilations [the standard error of the least square mean for treatment (initial vs. gap-27) x time = 2.5]. B: suffusion of gap-27 was not associated with any significant alterations in the increased EEG power elicited by bicuculline [the standard error of the least square mean for treatment (initial vs. gap-27) x time = 1.7]. C: the Cx40/Cx37-specific blocking peptide, gap-26, did not alter the bicuculline response [the standard error of the least square mean for treatment (initial vs. gap-26) x time = 4.1]. D: suffusion of gap-27 was accompanied by a ~75% reduction in the response to SNS, whereas gap-26 (E) had no effect. F: dual immunofluorescence analysis of Cx43 (red) and GFAP (green) revealed that Cx43 is especially concentrated at the GL (narrow arrows), as well as in the astrocytic envelope surrounding blood vessels [examples shown are a penetrating vessel (arrowheads) and parenchymal vessels (wide arrows)]. Scale bar = 50 µm. Values are means ± SE; n = 4 to 5 rats; *P < 0.05 vs. initial.

 
Effect of increasing cortical suffusion rate on neural activation-induced dilation. Increasing the aCSF suffusion rate from 0.5 to 1.5 ml/min during successive 20-min bicuculline exposures, separated by 1 h, was associated with a reduced vasodilating response, especially toward the end of the bicuculline suffusion (Fig. 7A). This occurred despite the absence of any differences in the patterns of EEG power changes when comparing the two suffusion rates (Fig. 7B). In time controls (rats represented in Fig. 1), where the suffusion rate was 0.5 ml/min for both bicuculline exposures, the ratio of the pial arteriolar diameter changes remained close to unity when comparing the initial and repeat exposures (Fig. 7C; see also Fig. 1B), whereas the ratio between the 1.5 and 0.5 ml/min suffusion fell to <50% of the time control ratio by 15 min (Fig. 7C). Similarly, we observed a significant reduction in SNS-induced dilation when comparing the 1.5 to the 0.5 ml/min suffusion rate (Fig. 7D) and a significantly lower ratio (Fig. 7E) compared with the time control (rats represented in Fig. 2B).


Figure 7
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 7. Effects of increasing the rate of cortical artificial cerebrospinal fluid suffusion on the pial arteriolar responses to bicuculline or SNS. A 2-way repeated-measures ANOVA (post hoc Student-Newman-Keuls test) was used for data analysis in A–C. A: some reductions in the pial arteriolar responses to seizure were observed when comparing bicuculline suffusion at 0.5 ml/min to subsequent suffusion at 1.5 ml/min [means ± SE; n = 4 rats; *P < 0.05 vs. initial; standard error of the least square mean for treatment (0.5 ml/min vs. 1.5 ml/min) x time = 2.9]. B: the finding that EEG power increases were not altered by the change in suffusion indicates that increasing the rate does not alter the intensity of the seizure [standard error of the least square mean for treatment (0.5 ml/min vs. 1.5 ml/min) x time = 0.71]. C: expression of pial arteriolar responses to bicuculline suffusion as a ratio between the first and second suffusions in rats where the suffusion rate was increased from 0.5 to 1.5 ml/min and in rats where both suffusions were performed at 0.5 ml/min (taken from the time controls represented in Fig. 1B, 5 to 20 min values only). D: increasing the suffusion rate from 0.5 to 1.5 ml/min was associated with a significant decrease in SNS-induced dilations (means ± SE; n = 5 rats; *P < 0.05 vs. initial). E: pial arteriolar responses were also expressed as a ratio, as in C, where comparisons were made between the time controls depicted in Fig. 2B (2nd and 1st suffusions performed at 0.5 ml/min) and in another group of rats where the 2nd and 1st suffusion rates were 1.5 and 0.5 ml/min, respectively. This analysis also indicated a significant reduction in pial arteriolar reactivity in the presence of an increased rate of suffusion.

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
There are several key indications arising from these experiments. First, both activation models elicit consistent, reversible, and repeatable pial arteriolar dilations. Second, that dilation is completely dependent on increased neuronal activity (blocked by TTX), even though direct neuronal connections are lacking. Third, selective injury to the glia limitans, but not endothelium, prevents neural activation-induced pial arteriolar dilation, suggesting a key role for astrocytes and a lack of any contributions from vascular endothelium. Fourth, the apparent astrocytic signaling pathway is sensitive to Cx43 blockade, suggesting an important contribution from gap junctions and/or hemichannels. Finally, increasing the rate of cortical suffusion during neuronal activation resulted in a reduced pial arteriolar response, implying the participation of a diffusible mediator.

These results seem to be consistent with a model whereby increased activity of cortical neurons is sensed by adjacent astrocytes, resulting in a vasodilating signal being transmitted to not only local parenchymal arterioles but also remote, upstream pial arteriolar segments as well. The route used to achieve the latter would appear to be primarily astrocytic, because disrupting interastrocytic communication at the cortical surface, with L-AAA, completely eliminated neural activation-associated pial arteriolar dilation. On the other hand, L/D endothelial injury, which has been documented to block interendothelial signaling (9), did not alter the pial arteriolar response to neural activation. Thus endothelium does not appear to participate as a signaling conduit in neurally induced pial arteriolar relaxation in vivo. However, this does not eliminate a role for endothelium in remote vasodilation in other upstream segments. For example, conducted upstream vasodilation elicited by local applications of K+ to isolated penetrating arterioles was impaired by endothelial injury (15). In our study, we also cannot completely rule out intercellular signaling via vascular smooth muscle, since this was not directly addressed. Upstream vasodilation via mechanisms that do not necessarily require intercellular communication also merits some consideration. This could include flow and shear-stress-induced vasorelaxation. However, there is no clear consensus regarding the existence of such a mechanism in the cerebral circulation [see review by Andresen et al. (2)]. Thus, in consideration of the fact that L-AAA-induced disruption of the glia limitans completely prevented neural activation-induced pial arteriolar dilation in the present study, in the absence of any detectable injury to the smooth muscle or endothelium, as reflected in normal pial arteriolar responses to SNAP and ACh, respectively [previously reported by us (33, 35)], would seem to leave little or no room for a nonastrocytic component.

One might also consider that, despite the lack of evidence of direct intrinsic neuronal contact with pial arterioles, vasodilating substances arising from neurons could have reached pial arterioles via nonsynaptic (diffusional) neurotransmission (see Ref. 3) and, perhaps, might even account for some of the suffusion rate sensitivity. Although we cannot completely discount this possibility, it does not appear likely in light of the limited presence of neuronal elements (see, for example, Refs. 7, 33, and 38, and Fig. 3D), compared with the abundance of glial processes (Fig. 4A), at the cortical surface, and in consideration of the fact that L-AAA blocks neurally evoked pial vascular responses in the absence of any effects on the neurons themselves. The latter is evidenced by the fact that L-AAA is excluded from neurons (23) and does not affect neuronal appearance (16) or function (see Ref. 33 and Fig. 4D). If there were indeed vasodilating substances diffusing from neurons during their activation, they would still be present following L-AAA exposure and capable of producing some dilation. The observation that L-AAA completely blocked neurally evoked pial arteriolar responses would seem to obviate any direct influences from vasodilators arising from neurons.

Present and earlier findings indicate a relatively low neuronal density in layer I of the rat cerebral cortex, <10% of that seen in subadjacent layers (4), and a paucity of neuronal (somal, dendritic, and axonal) elements in superficial (20–30 µm subpial tissue depth) aspects of layer I (Fig. 3D and Ref. 7). Nevertheless, a high percentage of those neurons (~90%) are GABAergic (4). Thus, the dendritic net depicted in Fig. 3D is likely to be rich in GABA receptors (for example, Ref. 28), which are capable of responding to GABA-releasing local (layer I) interneurons, leading to an inhibition of excitatory pyramidal cells, the somata of which reside in deeper cortical layers (38). In the present study, the fact that, following the initiation of bicuculline suffusion, the initial electrical and pial vascular responses were seen within the first minute but took ~10 min to achieve a plateau may reflect the above anatomical arrangement.

If, as the evidence suggests, cortical astrocytes represent the route of transmission for an upstream-dilating signal to pial arterioles, how might this be accomplished? Intercellular communication among astrocytes, effectively permitting these cells to behave as a syncytium, has been the subject of numerous publications (5, 19, 26, 27, 34). Findings from these studies strongly point to a role for Ca2+ (and Ca2+-mobilizing molecules, such as inositol 1,4,5-trisphosphate) and ATP as key intercellular signaling substances in the process of astrocyte-to-astrocyte communication. Moreover, there is compelling evidence favoring the participation of Cx43-expressing gap junctions (and hemichannels) in this signaling process, presumably via facilitating the passage of the above molecules from cell to cell or into the extracellular compartment (e.g., Refs. 5, 10, 19, 26, and 27). The abundant and selective expression of Cx43 within the glia limitans, coupled with profound reductions in the neurally evoked pial arteriolar responses in the presence of Cx43 blockade that we observed in the present study, strongly supports the above contention. In addition to the role of ATP in intercellular signaling, its presence in the extracellular compartment between the glia limitans and pial arterioles may have a direct role in the activity-evoked pial arteriolar dilation. As discussed in a recent report from our laboratory (34), ectonucleotidases are capable of rapidly hydrolyzing ATP, often resulting in an increased presence of the potent vasodilator adenosine.

Although present findings do not permit us to choose between gap junction and hemichannel contributions, it has been suggested that one may be able (at least in a general sense) to differentiate gap junctional intercellular communication from hemichannel-related communication simply by altering the duration of gap peptide exposure (e.g., Refs. 10 and 19); that is, these peptides target sites on the extracellular loops of connexins. Those loops are docking sites for the formation of a complete gap junction. In a patent gap junction, those sites may be relatively inaccessible and a long exposure (in hours) to the gap peptide may be required to observe a reduction in the vascular response. On the other hand, if short exposure times are effective (e.g., 45 min, as in the present study), then gap peptide influence may be more likely to involve interference with hemichannel-related communication, since the extracellular loops would be more readily accessible.

It is clear from present results that the activation-induced upstream dilation of pial arterioles requires an intact superficial glia limitans. This presumably reflects a key role for interastrocytic, not vascular intercellular, signal transmission and includes contributions from Cx43-containing entities. The lack of vascular endothelial or smooth muscle gap junction/hemichannel participation in that signaling process is supported by 1) the complete loss of activation-evoked dilation in the presence of L-AAA (which does not injure vascular cells) and 2) published findings by us showing that Cx43 expression in the glia limitans greatly exceeds the slight or virtually undetectable expression observed in the overlying pial arteriolar endothelium or smooth muscle, respectively (35). In short, there is fairly strong evidence supporting a nonvascular, major astrocytic conduit in the upstream signaling process. Nevertheless, although the literature is certainly suggestive of a role for Ca2+, as well as ATP, in interastrocytic communication (for example, see Refs. 1, 12, 29, 31, and 34), how these substances specifically contributed to neurally evoked pial arteriolar dilations in the present study remains unresolved, since those issues were not addressed.

Another intriguing scenario that may involve interastrocytic communication and Cx43 relates to K+. Over the past 40 years, the findings of a number of studies (reviewed in Ref. 18) have led to the suggestion that, during brain activation, a spatial redistribution of K+ may occur. Thus it has been proposed that the increases in extracellular [K+] that occur at the sites of increased neuronal activity are removed by nearby astrocytes and transferred, over multiple astrocytes, perhaps via gap junctions, to astrocytic endfeet. From there, the K+ can be released to the extracellular environment of adjacent cerebral microvessels, including pial arterioles. Large-conductance, Ca2+-activated K+ (BK) channels are known to be concentrated in the plasma membranes of astrocyte endfeet facing pial and parenchymal vessels (24). During periods of increased neural activity and increased [Ca2+] in astrocytic endfeet (e.g., Ref. 11), that arrangement could result in increased BK-dependent K+ efflux and substantial increases in [K+] levels in the extracellular fluid bathing pial vascular smooth muscle. Subsequent interactions with extracellular K+-sensitive Kir2.1 channels, known to be widely expressed in cerebrovascular smooth muscle (37), would represent a potent vasodilating stimulus. Evidence of such a BK- and Kir-dependent, neurally evoked vasodilating mechanism in parenchymal arterioles was recently reported by Filosa et al. (11), using a murine brain slice model. On the other hand, a role for glial-derived K+ in neurovascular coupling in the retina was not supported by recent findings from Metea et al. (20). Nevertheless, it remains to be established whether such a process contributes to the pial arteriolar responses seen in the present study.

In conclusion, present and earlier findings have shown that pial arteriolar dilation accompanies increased neuronal activity in the cerebral cortex, despite the absence of direct neuronal connections from the cortex to these vessels. Results from the present study indicate that the vasodilating signal arising in the parenchyma is transmitted to pial arterioles principally via an astrocytic, but not a vascular, route. Thus the expected disruption of the connection between the glia limitans and pial arterioles, via application of the selective gliotoxin, L-AAA, was associated with a complete loss of pial arteriolar dilation, irrespective of whether a physiological (SNS) or supraphysiological (seizure) neural activation stimulus was applied. Moreover, the finding that increasing the rate of cortical suffusion reduced neurally evoked pial arteriolar reactivity in both models also seems to imply the participation of a diffusible paracrine substance, although the nature and source of that suspected substance are unknown. In contrast, Ngai and Winn (21) reported that SNS-induced pial arteriolar dilation was unaffected by increasing the rate of cortical suffusion. We cannot, however, offer any explanation for this lack of agreement, although one possibility relates to the fact that our cranial windows were much larger, permitting a substantially greater cortical area to be influenced by increasing the suffusion rate. It should also be emphasized that, even in the absence of suffusion rate sensitivity, one cannot eliminate a role for diffusion of vasodilating substances from the glia limitans. Astrocyte endfeet, comprising the pial surface of the glia limitans, display unique protein expression patterns on the membranes apposing the vessels. These intramembrane particles, or orthogonal arrays, may represent specialized sites for intercellular communication (25). It is not unreasonable to postulate that such sites may be sufficiently shielded from washout effects during increases in cortical suffusion. Nevertheless, because of the L-AAA sensitivity of the activation-evoked pial arteriolar vasodilating response, the source of the diffusible vasodilators is quite probably the astrocytic processes of the glia limitans.


   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported by a National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-65629 (to D. A. Pelligrino) and an American Heart Association Grant 0635337N (to H.-L. Xu).


   FOOTNOTES
 

Address for reprint requests and other correspondence: D. A. Pelligrino, Neuroanesthesia Research Lab., Univ. of Illinois at Chicago, 835 S. Wolcott Ave., Rm. E-714E, Chicago, IL 60612 (e-mail: dpell{at}uic.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Anderson CM, Nedergaard M. Astrocyte-mediated control of cerebral microcirculation. Trends Neurosci 26: 340–344, 2003.[CrossRef][Web of Science][Medline]
  2. Andresen J, Shafi NI, Bryan RM Jr. Endothelial influences on cerebrovascular tone. J Appl Physiol 100: 318–327, 2006.[Abstract/Free Full Text]
  3. Bach-y-Rita P. Nonsynaptic diffusion neurotransmission in the brain: functional considerations. Neurochem Res 26: 871–873, 2001.[CrossRef][Web of Science][Medline]
  4. Beaulieu C. Numerical data on neocortical neurons in adult rat, with special reference to the GABA population. Brain Res 609: 284–292, 1993.[CrossRef][Web of Science][Medline]
  5. Bennett MV, Contreras JE, Bukauskas FF, Saez JC. New roles for astrocytes: gap junction hemichannels have something to communicate. Trends Neurosci 26: 610–617, 2003.[CrossRef][Web of Science][Medline]
  6. Bernhardt R, Matus A. Light and electron microscopic studies of the distribution of microtubule-associated protein 2 in rat brain: a difference between dendritic and axonal cytoskeletons. J Comp Neurol 226: 203–221, 1984.[CrossRef][Web of Science][Medline]
  7. Chu Z, Galarreta M, Hestrin S. Synaptic interactions of late-spiking neocortical neurons in layer 1. J Neurosci 23: 96–102, 2003.[Abstract/Free Full Text]
  8. Drake CT, Iadecola C. The role of neuronal signaling in controlling cerebral blood flow. Brain Lang 102: 141–152, 2007.[CrossRef][Web of Science][Medline]
  9. Emerson GG, Segal SS. Endothelial cell pathway for conduction of hyperpolarization and vasodilation along hamster feed artery. Circ Res 86: 94–100, 2000.[Abstract/Free Full Text]
  10. Evans WH, De Vuyst E, Leybaert L. The gap junction cellular internet: connexin hemichannels enter the signalling limelight. Biochem J 397: 1–14, 2006.[CrossRef][Web of Science][Medline]
  11. Filosa JA, Bonev AD, Straub SV, Meredith AL, Wilkerson MK, Aldrich RW, Nelson MT. Local potassium signaling couples neuronal activity to vasodilation in the brain. Nat Neurosci 9: 1397–1403, 2006.[CrossRef][Web of Science][Medline]
  12. Grafstein B, Liu S, Cotrina ML, Goldman SA, Nedergaard M. Meningeal cells can communicate with astrocytes by calcium signaling. Ann Neurol 47: 18–25, 2000.[CrossRef][Web of Science][Medline]
  13. Gustafsson F, Holstein-Rathlou N. Conducted vasomotor responses in arterioles: characteristics, mechanisms and physiological significance. Acta Physiol Scand 167: 11–21, 1999.[CrossRef][Web of Science][Medline]
  14. Hamel E. Perivascular nerves and the regulation of cerebrovascular tone. J Appl Physiol 100: 1059–1064, 2006.[Abstract/Free Full Text]
  15. Horiuchi T, Dietrich HH, Hongo K, Dacey RG Jr. Mechanism of extracellular K+-induced local and conducted responses in cerebral penetrating arterioles. Stroke 33: 2692–2699, 2002.[Abstract/Free Full Text]
  16. Khurgel M, Koo AC, Ivy GO. Selective ablation of astrocytes by intracerebral injections of alpha-aminoadipate. Glia 16: 351–358, 1996.[CrossRef][Web of Science][Medline]
  17. Koehler RC, Gebremedhin D, Harder DR. Role of astrocytes in cerebrovascular regulation. J Appl Physiol 100: 307–317, 2006.[Abstract/Free Full Text]
  18. Kofuji P, Newman EA. Potassium buffering in the central nervous system. Neuroscience 129: 1045–1056, 2004.[CrossRef][Web of Science][Medline]
  19. Leybaert L, Braet K, Vandamme W, Cabooter L, Martin PE, Evans WH. Connexin channels, connexin mimetic peptides and ATP release. Cell Commun Adhes 10: 251–257, 2003.[Web of Science][Medline]
  20. Metea MR, Kofuji P, Newman EA. Neurovascular coupling is not mediated by potassium siphoning from glial cells. J Neurosci 27: 2468–2471, 2007.[Abstract/Free Full Text]
  21. Ngai AC, Winn HR. Pial arteriole dilation during somatosensory stimulation is not mediated by an increase in CSF metabolites. Am J Physiol Heart Circ Physiol 282: H902–H907, 2002.[Abstract/Free Full Text]
  22. Pelligrino DA, Gay RL, Baughman VL, Wang Q. Nitric oxide synthase inhibition modulates N-methyl-D-aspartate-induced changes in cerebral blood flow and EEG activity. Am J Physiol Heart Circ Physiol 271: H990–H995, 1996.[Abstract/Free Full Text]
  23. Pow DV. Visualising the activity of the cystine-glutamate antiporter in glial cells using antibodies to aminoadipic acid, a selectively transported substrate. Glia 34: 27–38, 2001.[CrossRef][Web of Science][Medline]
  24. Price DL, Ludwig JW, Mi H, Schwarz TL, Ellisman MH. Distribution of rSlo Ca2+-activated K+ channels in rat astrocyte perivascular endfeet. Brain Res 956: 183–193, 2002.[CrossRef][Web of Science][Medline]
  25. Rash JE, Duffy HS, Dudek FE, Bilhartz BL, Whalen LR, Yasumura T. Grid-mapped freeze-fracture analysis of gap junctions in gray and white matter of adult rat central nervous system, with evidence for a "panglial syncytium" that is not coupled to neurons. J Comp Neurol 388: 265–292, 1997.[CrossRef][Web of Science][Medline]
  26. Saez JC, Contreras JE, Bukauskas FF, Retamal MA, Bennett MV. Gap junction hemichannels in astrocytes of the CNS. Acta Physiol Scand 179: 9–22, 2003.[CrossRef][Web of Science][Medline]
  27. Scemes E, Giaume C. Astrocyte calcium waves: what they are and what they do. Glia 54: 716–725, 2006.[CrossRef][Web of Science][Medline]
  28. Shi L, Pang H, Linville MC, Bartley AN, Argenta AE, Brunso-Bechtold JK. Maintenance of inhibitory interneurons and boutons in sensorimotor cortex between middle and old age in Fischer 344 X Brown Norway rats. J Chem Neuroanat 32: 46–53, 2006.[CrossRef][Web of Science][Medline]
  29. Tian GF, Azmi H, Takano T, Xu Q, Peng W, Lin J, Oberheim N, Lou N, Wang X, Zielke HR, Kang J, Nedergaard M. An astrocytic basis of epilepsy. Nat Med 11: 973–981, 2005.[Web of Science][Medline]
  30. Wang Q, Santizo R, Baughman VL, Pelligrino DA. Estrogen provides neuroprotection in transient forebrain ischemia through perfusion-independent mechanisms in rats. Stroke 30: 630–637, 1999.[Abstract/Free Full Text]
  31. Wang X, Lou N, Xu Q, Tian GF, Peng WG, Han X, Kang J, Takano T, Nedergaard M. Astrocytic Ca2+ signaling evoked by sensory stimulation in vivo. Nat Neurosci 9: 816–823, 2006.[CrossRef][Web of Science][Medline]
  32. Xu HL, Galea E, Santizo RA, Baughman VL, Pelligrino DA. The key role of caveolin-1 in estrogen-mediated regulation of endothelial nitric oxide synthase function in cerebral arterioles in vivo. J Cereb Blood Flow Metab 21: 907–913, 2001.[Web of Science][Medline]
  33. Xu HL, Koenig HM, Ye S, Feinstein DL, Pelligrino DA. Influence of the glia limitans on pial arteriolar relaxation in the rat. Am J Physiol Heart Circ Physiol 287: H331–H339, 2004.[Abstract/Free Full Text]
  34. Xu HL, Pelligrino DA. ATP release and hydrolysis contribute to rat pial arteriolar dilatation elicited by neuronal activation. Exp Physiol 92: 647–651, 2007.[Abstract/Free Full Text]
  35. Xu HL, Santizo RA, Baughman VL, Pelligrino DA. ADP-induced pial arteriolar dilation in ovariectomized rats involves gap junctional communication. Am J Physiol Heart Circ Physiol 283: H1082–H1091, 2002.[Abstract/Free Full Text]
  36. Xu HL, Ye S, Baughman VL, Feinstein DL, Pelligrino DA. The role of the glia limitans in ADP-induced pial arteriolar relaxation in intact and ovariectomized female rats. Am J Physiol Heart Circ Physiol 288: H382–H388, 2005.[Abstract/Free Full Text]
  37. Zaritsky JJ, Eckman DM, Wellman GC, Nelson MT, Schwarz TL. Targeted disruption of Kir2.1 and Kir22 genes reveals the essential role of the inwardly rectifying K+ current in K+-mediated vasodilation. Circ Res 87: 160–166, 2000.[Abstract/Free Full Text]
  38. Zhu Y, Zhu JJ. Rapid arrival and integration of ascending sensory information in layer 1 nonpyramidal neurons and tuft dendrites of layer 5 pyramidal neurons of the neocortex. J Neurosci 24: 1272–1279, 2004.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Physiol.Home page
A. Nimmerjahn
Astrocytes going live: advances and challenges
J. Physiol., April 15, 2009; 587(8): 1639 - 1647.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
294/2/H622    most recent
00530.2007v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Xu, H.-L.
Right arrow Articles by Pelligrino, D. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Xu, H.-L.
Right arrow Articles by Pelligrino, D. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.