Báo cáo khoa học: Calcium signalling by nicotinic acid adenine dinucleotide phosphate (NAADP)

Calcium Signalling By Nicotinic Acid Adenine Dinucleotide Phosphate (NAADP)

Tác giả: Michiko Yamasaki, Grant C. Churchill, Antony Galione

Lĩnh vực: Department of Pharmacology, University of Oxford, UK

Nội dung tài liệu:
Bài viết này cung cấp một cái nhìn tổng quan về vai trò của nicotinic acid adenine dinucleotide phosphate (NAADP) như một chất truyền tin mới nổi, có khả năng huy động canxi mạnh mẽ. Tài liệu thảo luận về các đặc tính độc đáo của NAADP như một tác nhân huy động canxi, đưa ra các lập luận về việc nó nhắm mục tiêu vào các kho canxi có tính axit thay vì mạng lưới nội chất. Ngoài ra, bài viết còn đề cập đến các con đường sinh tổng hợp khả thi của NAADP trong tế bào và các ứng viên tiềm năng cho kênh giải phóng canxi đích, vốn vẫn chưa được xác định rõ ràng.

Mục lục chi tiết:

• MINIREVIEW
• Calcium signalling by nicotinic acid adenine dinucleotide phosphate (NAADP)
• Keywords
• acidic stores; cADPR; endoplasmic reticulum; InsP3; NAADP
• Correspondence
• A. Galione, Department of Pharmacology, University of Oxford, Mansfield Road, Oxford OX1 3QT, UK
• Fax: +44 1865 271853
• Tel: +44 1865 271633
• E-mail: antony.galione@pharm.ox.ac.uk
• (Received 28 April 2005, accepted 30 June 2005)
• doi:10.1111/j.1742-4658.2005.04860.x
• Nicotinic acid adenine dinucleotide phosphate (NAADP) is a recently described Ca2+ mobilizing messenger, and probably the most potent. We briefly review its unique properties as a Ca2+ mobilizing agent. We present arguments for its action in targeting acidic calcium stores rather than the endoplasmic reticulum. Finally, we discuss possible biosynthetic pathways for NAADP in cells and candidates for its target Ca2+ release channel, which has eluded identification so far.
• Intracellular Ca2+ signals are coordinated to elicit spatiotemporal patterns. These include repetitive Ca2+ transients, which may be localized or propagated as regenerative waves that may also pass into neighbouring cells [1-3]. D-myo-inositol 1,4,5-trisphosphate (InsP3) is a well-established intracellular Ca2+ mobilizing messenger in many cell types [3], and is a paradigm for additional molecules that release Ca2+ from intracellular Ca2+ stores. Cyclic ADP-ribose (cADPR) and nicotinic acid adenine dinucleotide phosphate (NAADP) were first discovered in the sea urchin egg as novel Ca2+ mobilizing agents [4-6]. In this cell, cADPR was shown to target ryanodine receptors (RyRs) to release Ca2+ from the endoplasmic reticulum (ER), and now has been established as an intracellular messenger in several cell types [7,8]. In contrast, NAADP was found to activate a Ca2+ release mechanism distinct from those activated by InsP3 and cADPR, based on pharmacology and self-induced inactivation of the different Ca2+ release mechanisms. It has thus been of great interest to investigate the physiology, enzymology and pharmacology of the NAADP signalling pathway. Recent reports have shown increases in NAADP levels in response to cellular stimuli fulfilling a major criterion for the classification of NAADP as a second messenger not only in sea urchin eggs but also in mammalian cells [9-12]. Here we focus on the Ca2+ mobilizing properties of NAADP and compare them with the actions of InsP3 and cADPR.
• Distinct properties of NAADP
• Since the discovery of NAADP as a Ca2+ mobilizing molecule in sea urchin egg homogenates, the sea urchin egg has remained an important system in which to study the actions of NAADP. NAADP has an ability to release Ca2+ from intracellular Ca2+ stores and is the most potent Ca2+ mobilizing agent described so far. Perhaps the most intriguing property of NAADP is its profound self-desensitization mechanism that is unparalleled by any other intracellular messenger. Sub-threshold concentrations of NAADP inactivate the NAADP evoked Ca2+ release that normally shows a robust Ca2+ release response [13–15]. Although similar effects have been seen in plant cell preparations [16], in intact mammalian cells only high concentrations of NAADP cause such self-desensitization [10,11,17-20], which is also interesting as this occurs in the apparent absence of any Ca2+ release (Fig. 1).
• In sea urchin eggs and egg homogenates, heparin (an InsP3 receptor antagonist), ruthenium red, procaine and 8-NH2-cADPR (ryanodine or cADPR receptor antagonists), inhibit InsP3- and cADPR-induced Ca2+ signals, whilst the NAADP-evoked Ca2+ release persists. In these preparations, thapsigargin, an ER Ca2+-ATPase inhibitor, depletes InsP3 and ryanodine sensitive Ca2+ stores, resulting in the inhibition of InsP3 and cADPR responses. However, NAADP-induced Ca2+ release remains [21]. Similar results were seen in intact sea urchin eggs when photolysing caged derivatives of these messengers. Both photoreleased InsP3 and cADPR failed to evoke Ca2+ release in thapsigargin-treated cells, whilst the response to photoreleased NAADP remained unaffected [22,23] (Fig. 2).
• Pharmacological analyses extended to mammalian preparations have also confirmed the distinct nature of the NAADP-sensitive Ca2+ release mechanism from those regulated by InsP3 or cADPR, particularly in brain [24], and cardiac microsomes [25] as well as in arterial smooth muscle cells [26]. Furthermore, in sea urchin eggs, NAADP-sensitive Ca2+ stores can be separated physically from thapsigargin-sensitive stores sensitive to InsP3 and cADPR by cell fractionation of egg homogenates or intact egg stratification [6,27,28].
• The pharmacology of NAADP-induced Ca2+ release in sea urchin egg homogenates has been found to be different from known Ca2+ release channels. For example, it is sensitive to L-type Ca2+ channel inhibitors, such as dihydropyridines, D600 and diltiazem, and to certain K+ channel blockers, without affecting Ca2+ release via either InsP3 or ryanodine receptors [14,21,24]. Furthermore, NAADP-mediated Ca2+ release is neither potentiated by Ca2+ or Sr2+, nor inhibited by Mg2+ [14,21,29]. Therefore in contrast to
• Ca2+ release channels modulated by either InsP3 or cADPR that participate in Ca2+-induced Ca2+ release mechanism (CICR), the NAADP-sensitive Ca2+ release mechanism is unlikely to do so directly. The apparent inability of NAADP to induce regenerative Ca2+ signals itself implies a role in initiating localized Ca2+ signals, which may then be propagated by recruiting CICR mechanisms. Additional interactions of NAADP signalling pathways with Ca2+ signals may arise since the metabolism of NAADP to inactive NAAD is regulated by a Ca2+-dependent 2′-phosphatase [30].
• Radioligand binding studies employing [32P]NAADP support the idea that NAADP acts on a fundamentally different Ca2+ releasing channel from those gated by InsP3 or cADPR. Binding of radiolabelled NAADP to sea urchin egg homogenate membranes is highly specific [13,31,32] and is unaffected by InsP3 or cADPR [13,31]. Binding studies have revealed another peculiar property of the NAADP receptor where NAADP binds to its receptor in an essentially irreversible manner in the sea urchin egg homogenates [13,31,32]. In mammalian systems, however, [32P]NAADP binding to membrane preparations from rat brain [31], rat heart [25] and MIN-6 cells [10] is reversible. The apparent irreversibility of NAADP binding in sea urchin egg preparations is dependent on high K+ concentrations in the binding medium routinely used [15].
• Ca2+ mobilizing messengers and multiple stores
• Studies of the Ca2+ mobilizing effects of NAADP in intact cells have revealed that this Ca2+ release mechanism rarely operates in isolation. Rather the resultant Ca2+ signals evoked by this molecule are often boosted by Ca2+ release by RyRs, InsP3Rs or both. Interactions between different Ca2+ release mechanisms are critical for shaping Ca2+ signals in response to agonists in many different cell types [33]. The effects of NAADP on Ca2+ release are often abolished or attenuated by both heparin and 8-NH2-cADPR, antagonists for InsP3 and CADPR receptors, indicating that the different Ca2+ release channels are tightly coupled functionally. In sea urchin eggs, ascidian oocytes, and arterial smooth muscle, antagonists of InsP3Rs or RyRs reduce responses to NAADP [26,34,35], whereas in T-lymphocytes, starfish oocytes and pancreatic acinar cells, little effect of NAADP is seen in the presence of these inhibitors [18,36-39]. To explain these phenomena two models have currently been proposed.
• The first describes a single pool, the ER, expressing InsP3Rs and RyRs. Here NAADP interacts either directly with RyRs or via a separate protein that may indirectly activate RyRs [40,41]. This model accounts for the apparent complete abolition of NAADP evoked release by either RyR blockers or thapsigargin. A direct action of NAADP on RyRs is also supported by the findings that NAADP was shown to activate isolated RyRs reconstituted in lipid bilayers from rabbit skeletal muscle (RyR1) [42] and cardiac microsomes (RyR2) [43]. A second model, the two pool or trigger hypothesis, is based on the idea that there is a distinct NAADP-sensitive storage organelle, possibly an thapsigargin-insensitive acidic store [28], that is responsible for a localized signal which is amplified by InsP3Rs and RyRs the on the ER by CICR [22,34,36,38]. This model accounts for the finding in some cells that localized NAADP-induced signals persist in the presence of InsP3Rs and RyR antagonists or thapsigargin, but are abolished by agents that dissipate storage of by acidic organelles, such as the vacuolar H+ pump inhibitor, bafilomycin Al. This has been most clearly demonstrated in the sea urchin egg [28], but also extended to several mammalian cell types [11,44-46]. Two types of pharmacological manipulation of acidic stores have been investigated with regard to NAADP-evoked release. Glycyl-phenylalanyl-naphthylamide (GPN) is an agent that penetrates cellular membranes but is a substrate for the luminal lysosomal enzyme cathepsin C trapping membrane impermeant products within lysosomes resulting in disruption of lysosomal-related organelles by osmotic lysis [47]. The other approach is aimed at collapsing proton gradients thought to power Ca2+ uptake into acidic stores by Ca2+/H+ exchange, such as bafilomycin A1, FCCP and NH3 [48]. These agents selectively inhibit NAADP-induced Ca2+ release, whilst having little effect on the effects of either InsP3 or CADPR [11,28,45,46].
• Changes in endogenous levels of NAADP
• Only recently have NAADP levels been measured directly by using a radioreceptor assay with the NAADP binding protein from sea urchin eggs [9-12,49] and shown to change in response to extracellular stimuli [9-12]. This provided the final piece of evidence required to classify NAADP as a second messenger. NAADP levels have been shown to change in sea urchin sperm during activation before fertilization [9], in pancreatic beta cells in response to glucose [10], in smooth muscle cells in response to endothelin [11], and in pancreatic acinar cells in response to gut-peptide cholecystokinin [12], which has been the most detailed study so far. As outlined above, mouse
• pancreatic acinar cells have been an important system in which investigate mechanisms for the generation of intracellular Ca2+ signals. It has been suggested that interactions between a subset and all three messengers are used to generate specific Ca2+ signatures in response to extracellular agonists such as cholecystokinin and neurotransmitter acetylcholine [20,38,39,50-52]. In this cell type, it has been proposed that an initial increase in NAADP in response to cholecystokinin triggers a primary Ca2+ release, followed by recruitment of InsP3Rs and RyRs by CICR. Although there is much circumstantial evidence from physiological and pharmacological studies that cholecystokinin increases NAADP and CADPR levels, changes in the levels of NAADP or cADPR had not been characterized in response to this agonist until recently. We have recently provided the strong evidence to establish NAADP as a second messenger in pancreatic acinar cells [12]. Significant elevations of both NAADP and CADPR levels in response to a specific agonist, cholecystokinin, in a concentration-dependent manner were reported (Fig. 3). Cholecystokinin A receptors, expressed on mouse pancreatic acinar cells, possess two binding sites for cholecystokinin, high and low-affinity binding sites [53-55]. Concentration-response data suggest that production of NAADP and cADPR can be activated through both high and low-affinity sites on cholecystokinin A receptor. This same study also demonstrated receptor specificity for the production of NAADP and cADPR, whereby increases in CADPR levels via stimulation of acetylcholine muscarinic receptors as well as cholecystokinin A receptors, whereas NAADP increased only through the activation of cholecystokinin A. Intriguingly, the striking difference seen in time courses between NAADP and cADPR production, where the increase in NAADP was rapid and transient, whereas the increase in cADPR was much prolonged, strongly supports the proposed hypothesis that NAADP provides a localized Ca2+ trigger signal at the apical region where InsP3Rs, RyRs and NAADP-sensitive Ca2+ stores coexist [45,56–62] (Fig. 4), and subsequently this localized Ca2+ signal is amplified by a CICR mechanism via sensitization of RyRs throughout of the cell [20,38,39,50-52,60-63]. Cell surface receptors are predominantly located in the basolateral membrane, however, agonist-induced Ca2+ signals initiate at the apical pole before propagating into the basolateral domain by the CICR mechanism. The abundance of RyRs in the basolateral region together with the slow rise in the CADPR levels demonstrated in our recent report [12] may also contribute greatly to such spatiotemporal heterogeneities of Ca2+ signals.
• Although there are few reports of direct NAADP measurements, an interesting correlation is emerging. Inhibition of agonist-evoked signalling by inactivating NAADP concentrations or bafilomycin Al correlates well with receptors whose stimulation leads to elevations in NAADP levels, whereas those that are not sensitive to these pharmacological manipulations are not [11,12,45].
• Outstanding questions in NAADP signalling pathways
• There are still several important aspects of the NAADP signalling pathway that are unclear. Foremost is the nature of the NAADP receptor. Studies from the sea urchin egg system have suggested that NAADP probably acts on a distinct protein that is pharmacologically different from IsnP3Rs of RyRs [64], although direct activation of RyRs has also been proposed [64]. The kinetics of Ca2+ release evoked by NAADP are consistent with the gating of a Ca2+ release channel rather than a transporter protein [65]. Preliminary biochemical characterization of [32P]NAADP binding proteins from sea urchin eggs have shown that such proteins are likely to be integral membrane proteins, and probably smaller than either InsP3Rs or RyRs [30]. However, it is possible that the NAADP binding proteins may not form a pore themselves but rather interact with and modulate other channels. Further, in line with the multiplicity of InsP3Rs and RyR isoforms, it is also possible that multiple isoforms of NAADP ‘receptors’ exist which may go some way in reconciling conflicting pharmacological data from different systems [64]. Perhaps the most detailed study of the functional properties of NAADP receptors, in the absence of their molecular isolation, has come from the study of NAADP signalling in starfish oocytes [37,66]. Here much emphasis has been placed on the ability of NAADP to gate a cation influx in addition to release from internal stores. In contrast to the situation in sea urchin eggs where NAADP induces a brief Ca2+ influx (the ‘cortical flash’), followed by a more substantial mobilization [9], the starfish oocytes exhibits a profound Ca2+ influx of in response to NAADP. It has been proposed that NAADP receptors may be expressed at the plasma membrane of these cells, and thus electrophysiological analyses have been employed to characterize such NAADP-induced currents [67]. An interesting question is whether these currents arise from direct activation of NAADP receptors on the plasma membrane or activation of a plasma membrane channel via calcium released from cortical NAADP-sensitive stores. Taken together these observations imply the widespread distribution of NAADP receptors and multiple roles of NAADP. However, the ultimate
• Putative synthesis pathway for NAADP. In the presence of B-NADP, ADP-ribosyl cyclase catalyses the synthesis of NAADP by a base exchange reaction with an optimum pH of 4 [71]. CD38 and CD157 have been shown to be capable of forming NAADP under the same condition [71-73]. cAMP is a stimulator of NAADP synthesis via ADP-ribosyl cyclase [70].
• resolution of many these questions will require isolation of NAADP-binding proteins.
• It may come as some surprise that the biosynthetic pathway for NA