| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Symposium Reports |
1 Neuroanaesthesia Research Laboratory, University of Illinois at Chicago, IL, USA
Abstract
Owing to their intimate anatomical relationship with cerebral arterioles, astrocytes have been postulated as signal transducers, transferring information from activated neurones to the cerebral microcirculation. These forwarded signals may involve the release of vasoactive factors from the end-feet of astrocytes. This mechanism is termed ‘neurovascular coupling’ and its anatomical components (i.e. neurone, astrocyte and vascular cells) are termed the ‘neurovascular unit’. The process of neurovascular coupling often involves upstream dilatation. This is necessary during periods of increased metabolic demand, in order to permit more blood to reach dilated downstream vessels, thereby improving nutrient supply to the activated neurones. Without it, that downstream dilatation might be ineffective, placing neurones at risk, especially during episodes of intense neuronal activity, such as seizure. In the brain, pial arterioles represent important ‘upstream’ vascular segments. The pial arterioles overlie a thick layer of astrocytic processes, termed the glia limitans. This essentially isolates pial arterioles, anatomically, from the neurones below. Vasodilating signals that originate in the neurones therefore reach the pial arterioles via indirect pathways, primarily involving astrocytes and the glia limitans. Here we discuss a process whereby purinergic mechanisms play a key and neuronal activity-dependent role in astrocyte to astrocyte communication, as well as in glia limitans to pial arteriolar signals leading to vasodilatation.
(Received 21 March 2007;
accepted after revision 24 April 2007; first published online 27 April 2007)
Corresponding author H.-L. Xu: University of Illinois at Chicago, Neuroanesthesia Research Laboratory, 835 South Wolcott Avenue, Room E-714E, Chicago, IL 60612 USA. Email: hlxu{at}uic.edu
Evidence is accumulating to indicate that astrocytes play a vital role in neurovascular coupling (Koehler et al. 2006). Recent findings from our laboratory, and others, indicate that the glia limitans, the layer of astrocytic processes that blankets the cortical surface, has a profound influence on pial arteriolar dilatation in vivo (Leffler et al. 2006; Xu et al. 2004, 2005). Such studies are facilitated by the accessibility of the glia limitans via cranial windows, its close contact with pial arterioles, as well as the availability of the highly effective and selective astrocyte toxin, L-
-aminoadipic acid (L-AAA). The influence of the glia limitans on the overlying pial vasculature may take on particular significance in conditions of increased neuronal activity, for a number of reasons. First, during neuronal activation, whether excessive, such as seizure, or ‘physiological’ (e.g. sciatic nerve stimulation, SNS), pial arterioles dilate despite the lack of any direct contact with the activated neurones. Astrocytes, via the glia limitans, may provide a key link between increased neuronal activity in the brain parenchyma and the remote arteriolar relaxation represented by pial arterioles. Second, although local vasodilatation is important in adjusting nutrient supply to neuronal needs, that response could be ineffective in the absence of dilatation in the pial arterioles, which lie upstream from the arterioles supplying the neurones; this would be particularly true under conditions of high substrate demand, such as seizure.
Communication from cortical neurones to pial vessels via the astrocytic conduit and the glia limitans
Neuronal activity-dependent ATP and purinergic P2 receptor-related interastrocytic signalling. We have found that pial arteriolar dilatation evoked by seizure (topical bicuculline) or SNS is completely blocked in the presence of L-AAA-induced ablation of the glia limitans, but is unaffected following endothelial injury (Xu & Pelligrino, 2005, 2007a). This suggests that astrocytes, but not blood vessels, represent a primary signalling conduit between activated cortical neurones and pial arterioles. That signalling can be divided into three major components: first, neurone to astrocyte; second, interastrocytic; and third, glia limitans to pial arteriole. In recent studies, we have focused on the second and third components. With respect to interastrocytic communication, we have obtained preliminary evidence supporting two key mechanisms. These are connexin 43 gap junction and/or hemichannel-related signalling (Xu & Pelligrino, 2007a) and purinergic processes (Xu & Pelligrino, 2006, 2007b). There is strong evidence to support a mechanism involving ATP release and subsequent binding to purinergic P2 receptors on neighbouring astrocytes (reviewed by Burnstock, 2007). There are two major subdivisions of P2 receptors: P2Y (metabotropic) and P2X (ionotropic). To date, at least three P2Y (1, 2 and 4) and six P2X (1–4, 6 and 7) receptors have been identified in astrocytes. Of these, the P2Y1, P2Y2 and P2X7 may be the most important in interastrocytic signalling (Gallagher & Salter, 2003; Suadicani et al. 2004). While P2Y1 and P2Y2 receptors are activated by physiological increases in ATP, the P2X7 receptor has a generally low affinity for ATP and is unlikely to be active in the physiological range of extracellular ATP concentrations. Accordingly, only under conditions of excessive ATP efflux might this receptor come into play (see Xu & Pelligrino, 2006). Upon activation, the P2X7 receptor can provide a Ca2+ (and Na+) entry channel into astrocytes. Equally intriguing is evidence suggesting that these receptors may mediate ATP release (Suadicani et al. 2006). Thus, the astrocytic P2X7 receptor may play a role in ATP-related astrocyte to astrocyte signalling but only in the presence of neuronal hyperactivation (e.g. seizure).
Purinergic mechanisms in glia limitans to pial arteriolar signalling during neural activation: role of ecto-nucleotidases. While ATP itself is capable of eliciting a P2Y-related (endothelium-dependent) vasodilatation in cerebral arterioles, it can also elicit vasoconstriction via interactions with smooth muscle P2Y4 and/or P2X1 receptors (Horiuchi et al. 2001). With respect to the rise in extracellular ATP occurring during brain activation, it seems more likely that the products of ecto-nucleotidase-mediated ATP hydrolysis will have a greater vasodilating influence than ATP itself. The principal ‘end-product’ of ATP hydrolysis is adenosine (formed from AMP via ecto-5'-nucleotidase action), which itself promotes vasodilatation, to a large extent through interactions with the A2A subtype of P1 receptor. Indeed, there is reasonable evidence that adenosine is an important mediator of the pial arteriolar dilatation accompanying physiological brain activation (Meno et al. 2001).
The relative influence of ecto-nucleotidases may vary according to the level of neuronal activation. Thus, it has been reported (Grobben et al. 1999) that with increases in extracellular ATP levels up to the low micromolar range (as might be observed with SNS), ATP hydrolysis in astrocytes primarily involves direct conversion of ATP to AMP, presumably via ecto-nucleotide pyrophosphatase/phosphodiesterase (E-NPP). Owing to the relatively high levels of ecto-5'-nucleotidase expression in cortical astrocytes (Parkinson et al. 2005; Nedeljkovic et al. 2006), one could expect the AMP to be rapidly converted to adenosine. Conversely, at mid- to high-micromolar levels (as might be seen during seizure), hydrolysis of extracellular ATP by astrocytes may revert to the ecto-ATPase, ecto-nucleoside triphosphate diphosphohydrolase-2 (E-NTPDase-2), allowing for some accumulation of extracellular fluid (ECF) ADP (Grobben et al. 1999), in addition to adenosine (see, however, Wink et al. 2006). Furthermore, recent findings have suggested a possible P2X7-linked translocation of E-NTPDase-2 to the plasma membrane (Vlajkovic et al. 2007). This may have some relevance in our model, owing to the finding of a P2X7 dependency in the pial arteriolar dilatation elicited by seizure, but not by SNS. However, it should be noted that, in the studies cited above, findings were obtained in astrocyte culture systems (primary and glioma) or Chinese hamster ovary (CHO) cells. The specific pathways involved in the hydrolysis of extracellular ATP in vivo have not been established. Furthermore, indirect and inferential experimental approaches need to be used to obtain evidence of neural activity level-related differences in the pathways used to hydrolyse the ATP released into the ECF in vivo. Such evidence may be derived from both published and unpublished data from our laboratory. Thus, we reported (Xu et al. 2001) that in vivo pial arteriolar dilatation elicited by topically applied ADP was unaffected by adenosine receptor blockade. In contrast, in a pilot study (H.-L. Xu & D. A. Pelligrino, unpublished observations), we found that topically applied ATP elicited a pial arteriolar dilatation that was completely blocked in the presence of an adenosine receptor antagonist. Both sets of findings were obtained under conditions of normal (baseline) brain activity, and are more likely to represent conditions present during ‘physiological’ neuronal activation. As illustrated in Fig. 1, and in support of Grobben et al. (1999), the above evidence would tend to diminish a role for enzymes with ecto-ADPase function (e.g. NTPDases 1, 5 and 6; Wink et al. 2006) and favour E-NPP (see Lazarowski et al. 2000) as the major pathway for ATP hydrolysis, when ECF ATP levels are in the physiological range.
Preliminary experimental evidence regarding neural activity-related differences in ecto-nucleotidase function.
In a preliminary study (Xu & Pelligrino, 2007b) in rats equipped with closed cranial windows (Xu et al. 2005), three strategies were employed to detect the presence of activity-related differences in ATP hydrolysis. The first is based upon the postulate that, because of a greater rise in extracellular ATP during seizure compared with SNS, an increased trafficking via E-NTPDase-2 will occur. If this is the case, then one should observe greater signs of ECF ADP accumulation during seizure. An indirect strategy was employed to test this hypothesis and involved topical application of the highly selective ADP-sensitive P2Y1 receptor antagonist, MRS-2179. The selectivity of this antagonist is supported by the literature (Boyer et al. 1998; Baurand & Gachet, 2003) and by published results from our laboratory (Xu et al. 2005), which showed that the pial arteriolar dilatation elicited by topically applied ADP is completely blocked by MRS-2179. The expectation is therefore that if neural activation is accompanied by increased ECF ADP accumulation, this should be revealed as a reduced pial arteriolar dilatation in the presence of MRS-2179. Initial findings yielded some support for the above. Thus, pial arteriolar relaxation during seizure was reduced somewhat (by 
33%, over the 10–20 min interval of seizure) in the presence of the P2Y1 antagonist, whereas SNS-induced dilatation was unaffected (n
= 4, in both cases).
A second strategy used to detect the existence of activity-related differences in ATP hydrolysis employed the semi-selective ecto-nucleotidase blocker, ARL-67156. It was recently reported that ARL-67156 possesses the capability to block the conversion of ATP to AMP, but was without effect on the ATP to ADP converting enzyme, ecto-NTPDase-2 (Muller et al. 2006). As such, ARL-67156 may hinder formation of the adenosine precursor (and ecto-5'-nucleotidase substrate), AMP, while preserving ecto-NTPDase-2-mediated ADP formation. Thus, in the presence of topically applied ARL-67156 (50 µM), we observed a 20–40% reduction in seizure-induced pial arteriolar dilatation over the final 10 min of a 20 min seizure (n = 4). In contrast, ARL-67156 suffusion was accompanied by a substantial attenuation in the SNS-induced dilatation (80% reduction at 50 µM; n = 4). The much greater sensitivity of SNS- versus seizure-induced dilatation to ARL-67156 would seem to be consistent with the ability of ARL-67156 to hinder conversion of ATP to AMP, but not ATP to ADP, coupled with what one might expect based upon a relative increase in the role of E-NTPDase-2 at higher levels of neuronal activity and ECF ATP elevations.
The third strategy involved the application of the ecto-5'-nucleotidase blocker, ,β-methylene ADP (AOPCP). The selectivity of this agent at the dose used in this study (300 µM) has been documented (e.g. Agostinho et al. 2000). In the presence of topically applied AOPCP (n
= 4), as one might predict, no effects on adenosine-induced dilatation was observed. There was a substantial attenuation in the SNS-induced pial arteriolar response, with an 80% reduction (from the initial SNS-induced dilatation) observed at 300 µM AOPCP. With seizure-evoked dilatation, AOPCP application (n
= 5) was accompanied by a 35–40% reduction in responses. The seemingly lesser role of adenosine formation from AMP during seizure versus SNS is also consistent with a more limited generation of AMP and, perhaps, a more robust ADP generation via E-NTPDase-2. These effects of AOPCP are comparable to those observed when adenosine A2A receptor blockade (via 10 µM ZM-241385) was applied (n
= 4), except that, with A2A blockade, pial arteriolar dilatation in the presence of topically applied adenosine was completely suppressed.
Summary and conclusions
Purinergic mechanisms contribute to the propagation of signals arising from increased neuronal activity and ultimately resulting in pial arteriolar dilatation. Thus, the release of ATP from active neurones is ‘sensed’ by astrocytic P2 receptors, which initiates a regenerative interastrocytic communication via the sequential release of ATP and its interaction with P2 receptors on neighbouring astrocytes. The specific P2 receptor subtype participating in this process may depend upon the magnitude of neuronal activity and the quantities of ATP released into the extracellular compartment. For example, ionotropic P2 receptors, which are relatively insensitive to ‘physiological’ increases in extracellular ATP, may come into play during hyperactivation states (e.g. seizure) and excessive ATP release. However, it is important to note that, while purinergic mechanisms seem to play a key role in interastrocytic communication, one cannot ignore contributions from gap junctions and hemichannels (Scemes & Giaume, 2006). Extracellular ATP-linked processes may also play a role in vasodilating signals transmitted from the glia limitans to pial arterioles through the generation of products of ecto-nucleotidase-mediated hydrolysis of extracellular ATP. The principal end-product of this process is adenosine, which can act as a paracrine messenger, interacting with adenosine receptors (mostly A2A) on pial vascular smooth muscle. Evidence obtained in astrocytes (Lazarowski et al. 2000) points to a much greater sensitivity (lower Km) for direct conversion of ATP to AMP (ecto-pyrophosphatase/diphosphohydrolase function, as derived from E-NPP and E-NTPDase-1) than conversion to ADP (ecto-ATPase function, as derived from E-NTPDase-2). Like adenosine, ADP is a potent dilator of pial arterioles, albeit via a different receptor, i.e. the P2Y1 receptor(Xu et al. 2005). However, in contrast to ATP, when ADP is increased in the vicinity of pial arterioles, it is relatively insensitive to hydrolysis (Xu et al. 2001). In preliminary experiments, three separate pharmacological interventions were employed to reveal whether neural activity level-dependent differences in extracellular ATP hydrolysis pathways did indeed exist. These were: first, preferential interference with ATP to AMP, versus ATP to ADP, conversions; second, ADP (P2Y1) receptor blockade; and third, blockade of AMP to adenosine conversion. Thus, with the first and third interventions, SNS-, as opposed to seizure-evoked, dilatation was reduced to a much greater extent, while with the second intervention, the opposite was seen. That particular pattern of effects is completely in accord with the model illustrated in Fig. 1. Thus, adenosine may be partly replaced by ADP in promoting pial arteriolar dilatations in conditions of excessive neural activity. However, it must be emphasized that the above findings do not provide any specific clues as to the mechanisms responsible for the apparent activity-dependent shift in ecto-nucleotidase pathways. Whether, at high levels of neural activity, this is a simple function of ATP saturation of enzymes mediating direct ATP to AMP conversions, resulting in a ‘spillover’ to ADP-generating pathways, or includes more complex mechanisms (e.g. increased expression of E-NTPDase-2 at the astrocyte surface), awaits further study.
|
Agostinho P, Caseiro P, Rego AC, Duarte EP, Cunha RA & Oliveira CR (2000). Adenosine modulation of D-[3H]aspartate release in cultured retina cells exposed to oxidative stress. Neurochem Int 36, 255–265.[CrossRef][Medline]
Baurand A & Gachet C (2003). The P2Y1 receptor as a target for new antithrombotic drugs: a review of the P2Y1 antagonist MRS-2179. Cardiovasc Drug Rev 21, 67–76.[Medline]
Boyer JL, Mohanram A, Camaioni E, Jacobson KA & Harden TK (1998). Competitive and selective antagonism of P2Y1 receptors by N6-methyl 2'-deoxyadenosine 3',5'-bisphosphate. Br J Pharmacol 124, 1–3.[CrossRef][Medline]
Burnstock G (2007). Physiology and pathophysiology of purinergic neurotransmission. Physiol Rev 87, 659–797.
Gallagher CJ & Salter MW (2003). Differential properties of astrocyte calcium waves mediated by P2Y1 and P2Y2 receptors. J Neurosci 23, 6728–6739.
Grobben B, Anciaux K, Roymans D, Stefan C, Bollen M, Esmans EL & Slegers H (1999). An ecto-nucleotide pyrophosphatase is one of the main enzymes involved in the extracellular metabolism of ATP in rat C6 glioma. J Neurochem 72, 826–834.[CrossRef][Medline]
Horiuchi T, Dietrich HH, Tsugane S & Dacey RG (2001). Analysis of purine- and pyrimidine-induced vascular responses in the isolated rat cerebral arteriole. Am J Physiol Heart Circ Physiol 280, H767–H776.
Koehler RC, Gebremedhin D & Harder DR (2006). Role of astrocytes in cerebrovascular regulation. J Appl Physiol 100, 307–317.
Lazarowski ER, Boucher RC & Harden TK (2000). Constitutive release of ATP and evidence for major contribution of ecto-nucleotide pyrophosphatase and nucleoside diphosphokinase to extracellular nucleotide concentrations. J Biol Chem 275, 31061–31068.
Leffler CW, Parfenova H, Fedinec AL, Basuroy S & Tcheranova D (2006). Contributions of astrocytes and CO to pial arteriolar dilation to glutamate in newborn pigs. Am J Physiol Heart Circ Physiol 291, H2897–H2904.
Meno JR, Crum AV & Winn HR (2001). Effect of adenosine receptor blockade on pial arteriolar dilation during sciatic nerve stimulation. Am J Physiol Heart Circ Physiol 281, H2018–H2027.
Muller CE, Iqbal J, Baqi Y, Zimmermann H, Rollich A & Stephan H (2006). Polyoxometalates – a new class of potent ecto-nucleoside triphosphate diphosphohydrolase (NTPDase) inhibitors. Bioorg Med Chem Lett 16, 5943–5947.[CrossRef][Medline]
Nedeljkovic N, Bjelobaba I, Subasic S, Lavrnja I, Pekovic S, Stojkov D, Vjestica A, Rakic L & Stojiljkovic M (2006). Up-regulation of ectonucleotidase activity after cortical stab injury in rats. Cell Biol Int 30, 541–546.[CrossRef][Medline]
Parkinson FE, Xiong W & Zamzow CR (2005). Astrocytes and neurons: different roles in regulating adenosine levels. Neurol Res 27, 153–160.[CrossRef][Medline]
Scemes E & Giaume C (2006). Astrocyte calcium waves: what they are and what they do. Glia 54, 716–725.[CrossRef][Medline]
Suadicani SO, Brosnan CF & Scemes E (2006). P2X7 receptors mediate ATP release and amplification of astrocytic intercellular Ca2+ signaling. J Neurosci 26, 1378–1385.
Suadicani SO, Flores CE, Urban-Maldonado M, Beelitz M & Scemes E (2004). Gap junction channels coordinate the propagation of intercellular Ca2+ signals generated by P2Y receptor activation. GLIA 48, 217–229.[CrossRef][Medline]
Vlajkovic SM, Wang CJ, Soeller C, Zimmermann H, Thorne PR & Housley GD (2007). Activation-dependent trafficking of NTPDase2 in Chinese hamster ovary cells. Int J Biochem Cell Biol 39, 810–817.[CrossRef][Medline]
Wink MR, Braganhol E, Tamajusuku AS, Lenz G, Zerbini LF, Libermann TA, Sevigny J, Battastini AM & Robson SC (2006). Nucleoside triphosphate diphosphohydrolase-2 (NTPDase2/CD39L1) is the dominant ectonucleotidase expressed by rat astrocytes. Neuroscience 138, 421–432.[CrossRef][Medline]
Xu HL, Koenig HM, Ye S, Feinstein DL & Pelligrino DA (2004). Influence of the glia limitans on pial arteriolar relaxation in the rat. Am J Physiol Heart Circ Physiol 281, H331–H339.
Xu HL & Pelligrino DA (2005). Mechanisms of pial arteriolar dilation during seizure. Soc Neurosci Abstract Program no. 10.2.
Xu HL & Pelligrino DA (2006). The role of the ionotropic P2X7 receptor in pial arteriolar dilations elicited by intense (seizure) and physiologic (sciatic nerve stimulation) neuronal activation. Soc Neurosci Abstract Program no. 117.9.
Xu HL & Pelligrino DA (2007a). Astrocytes provide a signaling conduit for upstream (pial arteriolar) dilation during cortical neuronal activation in the rat. 23rd International Symposium on Cerebral Blood Flow, Metabolism and Function & 8th International Conference on Quantification of Brain Function with PET.
Xu HL & Pelligrino DA (2007b). ATP release and hydrolysis contributes to pial arteriolar dilations elicited by neuronal activation. Exp Biol Abstract Program no. 748.1.
Xu HL, Santizo RA, Koenig HM & Pelligrino DA (2001). Chronic estrogen depletion alters adenosine diphosphate induced pial arteriolar dilation in female rats. Am J Physiol Heart Circ Physiol 281, H2105–H2112.
Xu HL, Ye S, Baughman VL, Feinstein DL & Pelligrino DA (2005). 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.
Acknowledgements
This work was supported by grants DK 065629 from the NIH and 065337N from the American Heart Association.
This article has been cited by other articles:
|
|
 |
|
  H.-L. Xu, L. Mao, S. Ye, C. Paisansathan, F. Vetri, and D. A. Pelligrino Astrocytes are a key conduit for upstream signaling of vasodilation during cerebral cortical neuronal activation in vivo Am J Physiol Heart Circ Physiol, February 1, 2008; 294(2): H622 - H632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Filosa Neural glial vascular communication in the brain Exp Physiol, July 1, 2007; 92(4): 633 - 633. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
