PerspectiveCalcium signaling

Linking NAADP to Ion Channel Activity: A Unifying Hypothesis

See allHide authors and affiliations

Science Signaling  24 Apr 2012:
Vol. 5, Issue 221, pp. pe18
DOI: 10.1126/scisignal.2002890


Nicotinic acid adenine dinucleotide phosphate (NAADP) is a potent Ca2+-releasing second messenger that might regulate different ion channels, including the ryanodine receptor, two-pore channels, and TRP-ML1 (transient receptor potential channel, subtype mucolipin 1), a Ca2+ channel localized to lysosomes. New evidence suggests that a 22- and 23-kilodalton pair of proteins could be the receptor for NAADP. Labeling of NAADP binding proteins was independent of overexpression or knockout of two-pore channels, indicating that two-pore channels, although regulated by NAADP, are not the NAADP receptors. I propose that NAADP binding proteins could bind to different ion channels and thus may explain how NAADP regulates diverse ion channels.

Evolution has developed various ways to control and adjust the free cytosolic Ca2+ concentration. Changes in the free cytosolic Ca2+ concentration in response to external stimuli are often transduced by small molecules termed second messengers. The most potent Ca2+-releasing second messenger known so far is nicotinic acid adenine dinucleotide phosphate (NAADP) [reviewed in (13)]. Here, I shall discuss recent findings by Lin-Moshier et al. (4) that shed new light on the molecular mechanism of action of NAADP.

NAADP was discovered by Lee and co-workers in the sea urchin egg system in 1995 (5). Initial experiments indicated that Ca2+ release induced by NAADP occurs independently from that triggered by other Ca2+-releasing second messengers, including d-myo-inositol 1,4,5-trisphosphate (IP3) (6) or cyclic adenosine diphosphoribose (cADPR) (7).

Ryanodine receptors type 1 and 2 (RyR1 and RyR2) emerged as the early front-runners as the target channels for NAADP because channel opening was observed upon NAADP addition to RyR1 or RyR2 preparations fused into lipid planar bilayers (8, 9). However, in stratified sea urchin eggs, NAADP released Ca2+ from an internal store that was different from the endoplasmic reticulum (ER) (10). In 2002, this store was identified as the reserve granule of sea urchin eggs, an acidic compartment related to lysosomes (11). Because RyRs are usually localized to the ER or sarcoplasmic reticulum (SR), this led to the proposal of an acidic store equipped with a potentially unidentified Ca2+ channel. Debate over these two models has characterized the scientific discussion on NAADP targets. In fact, evidence for NAADP targeting RyR localized to the ER was obtained in some of the mammalian systems (1218), whereas inhibition of lysosomal Ca2+ uptake or destruction of lysosomes resulted in abrogation of NAADP-mediated Ca2+ signaling in other systems (11, 1922). Then, the discovery of a new class of NAADP-regulated Ca2+ channels, termed two-pore channels (TPCs), in 2009 by Zhu, Galione, and colleagues (23) and Patel and colleagues (24) appeared to be a hallmark and the final step in identification of the NAADP receptor and channel.

Although NAADP effects on TPCs have been demonstrated (2529), a binding site for NAADP has not been reported for TPCs. To directly detect such binding sites, Wal­seth and colleagues developed a photoaffinity probe for the NAADP receptor, 5-N3-NAADP (4, 30, 31), which shows specific and high-affinity binding and the ability to induce Ca2+ release. The radio­actively labeled version of the probe detected three proteins of 30, 40, and 45 kD in sea urchin egg homogenates, which appeared to be different from the putative NAADP target, the sea urchin TPCs (31). In mammalian SKBR3 cells, Ca2+ release upon micro­injection of 5-N3-NAADP was also observed, suggesting functional binding to the mammalian NAADP receptor (4). Radio­actively labeled 5N3-NAADP was then used to identify NAADP binding proteins in SKBR3 cell lysates. Competition experiments with the “cold” ligands NAADP and NADP revealed a 22- and 23-kD doublet as the likely candidates for the NAADP binding proteins. The 22- and 23-kD doublet of putative NAADP binding proteins was more abundant in the cytosolic fraction but was also attached to membranes, and its apparent affinity to NAADP and its selectivity over NADP was higher in the membrane-attached state (4). The same 22- and 23-kD doublet of NAADP binding proteins was detected in human embryonic kidney (HEK) 293 cells and mouse pancreatic acinar cells (4). Because TPC1 and 2 have been discussed as prime candidates for the NAADP receptor (2329), photoaffinity labeling was performed in lysates from SKBR3 cells overexpressing one of the mammalian TPC isoforms. Surprisingly, there were no changes in the labeling pattern as compared to wild-type cells (4), suggesting that TPCs do not directly bind NAADP. Finally, pancreatic samples from mice lacking either TPC1 or TPC2 showed the same labeling pattern of the 22- and 23-kD pair of NAADP binding proteins as wild-type tissue, suggesting that these putative NAADP binding proteins are distinct from TPCs and not just proteolytic fragments (4). These results suggest that either the photoaffinity labeling strategy as such might be unable to detect TPC labeling or that TPCs do not bind NAADP directly, but through NAADP binding proteins (4).

Critical reevaluation of the photoaffinity approach should encompass the following points: (i) The photolabile moiety must not interfere with interaction of the pure ligand with its receptor; (ii) the affinity of the photoaffinity label must not differ substantially from that of the natural ligand; (iii) the photoaffinity label must be chemically stable in light and dark conditions; (iv) the life span of the photoactivated state of the photoaf­finity label must be short enough to prevent dissociation and random labeling of nonrelated proteins; and (v) the process of labeling must not result in damage to the protein (for example, partial proteolysis) to allow for downstream analytical procedures such as SDS–polyacrylamide gel electrophoresis (32). Lin-Moshier et al. (4) have carefully validated their photoprobe by conforming with these guidelines. However, the impact of the photoaffinity labeling approach to identify the NAADP receptor cannot be estimated before its final molecular identification.

Thus, provided that TPCs—and perhaps also the other NAADP receptor candidates, RyR and transient receptor potential chan­nel, subtype mucolipin 1 (TRP-ML1) (33, 34)—do not bind NAADP directly, a unifying hypothesis for the signaling mechanism of NAADP may be the following: The action of NAADP can be divided into several steps (Fig. 1), starting from formation, followed by the interaction of NAADP with its binding protein as the new step discussed here, which then is followed by the process of release of a small portion of Ca2+ in a Ca2+ trigger zone. The latter is subsequently amplified by RyR or the IP3 receptor in the Ca2+ amplification zone (Fig. 1). Rapid local Ca2+ release upon microinjection of NAADP in the subplasmalemmal space that may represent the Ca2+ trigger zone was detected in Jurkat T cells and was rapidly amplified (15). Release of Ca2+ in the trigger zone may occur not because of direct binding of NAADP to an ion channel, but rather may be initiated through NAADP binding to a specific binding protein. The latter, which is the central part of the unifying hypothesis, can interact with different ion channels once it has bound NAADP (Fig. 1).

Fig. 1

A unifying hypothesis for NAADP signaling. Schematic presentation of the four steps of NAADP signaling: (i) formation, (ii) interaction with a NAADP binding protein, (iii) release of trigger Ca2+, and (iv) amplification of the Ca2+ signal. Dashed lines indicate interactions that have not yet been clarified in detail. The direct interaction of NAADP with RyR1 was observed in highly purified RyR1 preparations (9, 17); however, these preparations contain additional proteins that may be involved in the NAADP-RyR1 interaction (17).


Does such a hypothesis fit published data on regulation of ion channels by NAADP? Evans and colleagues tested coupling of NAADP to TPCs or RyRs by overexpression of these channels in HEK293 cells (29). The 22- and 23-kD pair of NAADP binding proteins is present in HEK293 cells (4). Overexpression of TPC1 or TPC2, but not of RyR1 or RyR3, caused the intracellular release of Ca2+ in HEK293 cells stimulated with NAADP (29), indicating that HEK cells provide the correct environment of TPC activation by NAADP bound to its putative endogenous binding partners, the 22- and 23-kD proteins. In T lymphocytes, evidence for NAADP regulating RyR has been obtained (1418). One possible way to reconcile these results is if the 22- and 23-kD NAADP binding proteins would be present in T cells and allow for activation of RyR by the complex of NAADP bound to its endogenous binding proteins. To make the story more complex, another study from Li’s laboratory (33) carried out in fibroblasts confirms earlier data (34) and suggests a role for TRP-ML1 as a lysosomal Ca2+ channel regulated by NAADP. The lack of photoaffinity labeling experiments of NAADP binding proteins in fibroblasts makes a direct comparison of these cells to HEK or SKBR3 cells difficult. Taken together, the data from different cell systems indicate a complex process that enables (or disables) NAADP binding protein interaction with different ion channels. How coupling works precisely is presently unknown, but molecular identification of the NAADP binding protein should hopefully enable such studies in the near future.

References and Notes

Acknowledgments: Research in my laboratory is supported by Deutsche Forschungsgemeinschaft (grants GU 360/10-2, 13-1, and 15-1), the Wellcome Trust (grant 084608/Z/07/Z), and the Deutsche Akademische Austauschdienst. I am grateful to R. Fliegert (Hamburg) and T. Walseth (Minneapolis) for comments on the manuscript. Finally, I would like to acknowledge J. V. and O. V. Gerasimenko and O. H. Petersen and their co-workers because they came up with the idea of a NAADP binding protein regulating ion channels, in particular the RyR, in 2003 (13).
View Abstract

Stay Connected to Science Signaling

Navigate This Article